Q. Hi Yaa, welcome to the Big Unlock podcast. So, I’m Rohit Mahajan. I’m the CEO and managing partner at Big and Demo Consulting. And we’ve had the good fortune of having you on our podcast once before when Paddy was running it. So welcome back. And with that, would you like to introduce yourself to the audience?

Yaa: Absolutely. My name is Yaa Kumah-Crystal. I’m a paediatric endocrinologist and biomedical informaticist, and I serve as a clinical director of health IT at Vanderbilt University Medical Center here in Nashville, Tennessee.

Q. That’s awesome. Tell us a little bit more—how did you start in this field of being a physician, and what attracted you? How did you move from being a practicing physician into health IT, which is a dry field for many people?

Yaa: I think the luck of it all is that one of my features and bugs is that I’m a complainer. I’m always trying to improve systems around me, always finding opportunities. When I came to med school, we were transitioning from paper to the EHR, and as we started using some of the electronic health record systems, I had a bit of a computer science background. I saw that there were lots of ways we could be improving things—the way data flowed, to not have to repeat entry. I think I just complained, complained, complained to enough people that they said, “You should talk to Kevin Johnson.” He was the chair of our biomedical informatics department. He said, “You think you have some good ideas that can improve things?” And I said, “I think I do.” He was like, “You should consider informatics.” And through luck and happenstance, I got involved in a lot of really interesting projects in improving the EHR. Got my master’s in biomedical informatics and was able to get a position serving in health IT to improve the EHR.

Q. That’s awesome. So talk with us about some of these early experiments in improving EHR, because they’ve come a long way now—including, we were just talking about how Epic is integrating generative AI into the product itself and the platform itself. Share with us what you’ve seen so far, and where you think it’s going in terms of all these new developments.

Yaa: Sure thing. It’s funny—I almost want to break up the EHR into three eras: the paper era, the classic data entry era, and now the generative era In the classic era, there are a lot of great concepts and thinking to improve the way we’re. Capturing information because a lot of the problem was we had these paper records and people would just clone a paper form into a digital form and say, have at it. Go ahead and enter all this data in doctors, and please do a good job on the way and don’t miss anything and make sure you can get everything for our quality improvement and billing.” Doctors were getting so frustrated, and we tried different modalities to improve things. We had dictation, voice technology—we even worked on a project that was a voice assistant for the EHR, kind of like Alexa, early on. You could ask it things and get information back. We collaborated with Epic on a tool called “Hey Epic” to talk with the EHR, but architecturally, it was so complicated to plan for every single way someone might ask a question.

Then generative AI—large language models—entered the scene, and suddenly the concept of naturalistic interaction became feasible. You can now ask for things and get responses in the way we think about them as humans. Is just part of what we do now. So a huge aspect that has been an improvement that doctors have seen right away is ambiance workflows. So being able to just have a regular conversation with your patient, have the AI do all the work of the documenting for you. Yeah. And generally, note at the end so you can pay attention. So the person in front of you has been such a game changer. We literally get emails from providers saying, “This is such a game changer. I get my notes done on time. I can pay attention to my patients.” And it’s getting better every day.
So I think this new era, following the standard digital EHR era, is going to bring some of the most exciting advancements in patient care.

Q. Yeah, that’s good to know. One of the things you hit upon—ambient listening and integrating it into workflows—is resonating across the board. It’s quickly becoming a key technology that everyone is going to use. What others do you see on the horizon as you look around and evaluate developments in the generative AI space?

Yaa: What I see—and I’ve always been interested in—is voice technology and being able to converse with technology the way we’re conversing right now. I think everyone listening to this podcast has probably played with ChatGPT, Claude, and other similar tools. One interesting feature I find is voice mode, where you can ask questions and get responses back. Any of us who’ve played with, you know, the older Siri and all that stuff, know that. Sometimes you ask the question, it’s like, um, I’ll search the internet for you. And you’re like, ah, yeah. But with these tools now you ask it and you get something back, and the thing you get back is related in context to the thing you’re asking for, and you can ask follow up questions. My biggest question is when will we be able to integrate a workflow like That into the EHR, because many times when you are trying to put together the story of a patient to figure out what to do next or what’s been going on with them, you have the questions in your mind. You’re like, when did they last get this test? Or how many times have they, um, been to the hospital for this particular issue? These are questions you have in your mind. To find them, you have to navigate the EHR. You have to say, well, if I wanna figure out how many times they have an encounter, I need to go to the encounters tab and I need to filter for ED visits.

And there are these steps that you’re taking to answer this question, but what if you could just like ask the question and then have the EHR understand you because it’s , it has this transformer architecture in the background that. Understands that can inter interpolate what is you’re saying and feed you back the information, and that takes off so much of the cognitive burden of the work to find the things you need to care for your patient.

Q. That’s so interesting you say that because recently, I’ve been talking with several people—including at VIVE —and one of the leaders I speak with often, Dr. Ashish Atreja, described it as a “headless EMR EHR.” The voice interface comes to the front, allowing you to ask questions in natural language, while the EMR EHR itself fades into the background for the purpose of just gives you the answers. It’s fascinating that you’ve been following this for so long.  Let’s flip perspectives now—how does this look from the patient or caregiver perspective?  And we were chatting about patient portals a few minutes ago before we got on the podcast. How are your paediatric patients and their caregivers embracing this technology, and what are they asking for?

Yaa: This has been a very interesting time as a paediatric provider because our patients are young, and their parents are from an early tech-savvy generation. They’re more in tune with technology than, I think, some of the older patient populations in internal medicine, for example. I’ve actually had a patient say to me, Hey, I ran this through chat GPT, and this is what it told me. Do you agree? And what my relief was that they asked. Me for my opinion, yes, to see if I concurred with what chat GPT was giving them course, but I don’t think it’s gonna be a long time until we’re going to start seeing patients discover more and more that these tools are able to give responses. The only concern that I have is how reliable and accurate these responses are, and what responsibility do we have as a medical organization to work, to develop tools where that they can safely ask for questions and can give back more certified, approved answers  to answer some of their questions.

Concern that I have is how reliable and accurate these responses are, and what responsibility do we have as a medical organization to work, to develop tools where that they can safely ask for questions and can give back more certified, approved answers, uh, to answer some of their questions

Also, circling back to voice user interfaces on patient portals—patients are navigating these platforms trying to find information, but sometimes they just have a question. What would it look like if they could ask the portal: “When’s my next appointment with Dr. McChrystal?” or “Are my labs back yet?” or “Hey, I need a refill on my albuterol—can you send that in?” And then those things just happen as part of the process. Right now, the buzzword is “agents.” So we’re talking about smart agents that can make our patients’ lives easier.

Q. That’s awesome that to think that we are, because I would love to do that. Ask my a question and get an answer back instead of us trying to validate my, my user ID and password. 

Yaa:  And that’s the thing because um, there’s so many, there are really interesting ways to authenticate people like it would know your voice. There are other factors, like it knows that, oh, you’re nearby because it detects like your phone’s nearby. There are just. Many ways. It’s such a fun time to be a nerd right now because there’s so many things that are converging with a lot of technologies we’ve been thinking about for a long time. 

Q. That’s great. So that leads me to my next curious question: in a large system like yours, how do you approach innovation and new technologies? Do you measure return on investment? How do you decide your next projects for, say, this year or the next?

Yaa: Great question. Vanderbilt is a large academic organization. We have trainees and a lot of researchers. I think we’re all aware of the changes happening in the broader landscape around funding and priorities, so we have to be thoughtful about how we implement technology—making sure it serves our mission and benefits our providers. At the same time, we want to try new things and partner with vendors to say, “Hey, we’re experimenting with this, and we have opinions and suggestions for improvement. So I think part of our strategy is to do early pilots. So we can help to steer the direction that these tools are going to be materializing when they go to full production. At the same time, we aim to deeply understand what ROI even means in this context. Are we trying to improve turnaround time for notes? Reduce pyjama time for providers Do we want to decrease the number of messages that patients want to send back and forth? Once you can really define what is you’re trying to improve, then you can figure out how to measure it. And once you can measure it. Then you can really know if you’re making an impact. I think a common concern that people raise with looking at ROI is, well, sometimes the ROI is just that your providers are less miserable, that they’re less out, and how do you measure something like that?

There are interesting metrics that can help us assess whether the cognitive and emotional load on providers is being offset by the tools we provide. Some of these tools aren’t inexpensive, but I think we’ll start seeing providers expect a certain level of expectations that their EHR systems are working at a certain state of the art. EHR systems are working at a certain state of the art. I work with trainees and one of my trainees, she’s finishing up her training and she was looking for a new organization and she joked to me, she was like, well, I sure hope they have ambient there because I’m so spoil right now. I don’t know what I’d do without it.

Yeah, that was a joke that she made because it’s so early days, but I think it’s actually gonna become a truth where if your organization is not on the leading edge and not providing these tools to improve the way your providers work. People might decide that they want to go to a different organization. So, I think the people in the C-suite needs to be giving that consideration. 

Q. That’s great to know. So, you touched upon this, topic of one of your trainees. So, talk with us about education, training and learning initiatives in the community or at the. At the health system, how do you go about that? How do you level up and what are some of the things that you can do about it? 

Yaa:  Yeah, this is a really, really fascinating part. A lot of the new tools that we have help people become more proficient and help people be more efficient. You help you see patients faster, help you write notes faster, and that’s great. But on the other side of the coin. Trainees are not just supposed to learn how to be efficient. They’re supposed to learn how to practice medicine and how to balance logistics. So, we need to find a good balance of teaching them how to use the tools they’re gonna be using in the future, but not to the extent that we’re impeding on. Their actual learning process. In medical school, I took a course on how to write a good medical note because you had to think through the diagnosis. You had to figure out of all the things the patient told you, what things do you wanna keep in the note? What things do you leave out of it because it’d just be too much and that’s distracting. You have to do that synthesis and assimilation yourself. Well, we have tools to do that. Now, does that mean that we don’t teach students what a good note is anymore? Do they just solely rely on something that’s generated? These are decisions that we are going to be making and making in real time because no one knows the answer. cause we’re on the pioneering edge of it all.  But I think we need to not forget that trainees being a medical student, a resident, a fellow. It’s not just about learning how to be efficient. You need to reach a level of proficiency before you need to worry about efficiency. So, we need to make sure that we’re teaching them things they need to be good doctors so they can be efficient, good doctors eventually.

Q. Do you give that into your program over there? How does that? 

Yaa:  Absolutely. Yes. We have medical students and we all sorts of trainees that rotate through and I, I love, love working with trainees because they will ask me questions about things I have never thought about. I’m like. Let me go look that up. Yeah, just think of things in very, very new and interesting ways. 

Q. That’s very cool. So just curious that, you know, many times the focus is on the main system, which is the EMR EHR system. Let’s say a pick  Epic in this case, and then there’s patient portals. It could be MyChart or anything else from the same vendor or, or different providers as well. What other systems is a physician touching upon during their work? Which could be transformed digitally. Any thoughts on any other systems which come into play? 

Yaa:  So many, so many. So communication systems in general, a lot of what we do is communicating. Yeah. In the medical system, we’re still using like fax machines and pages. Yeah. Like we’re singly keeping those industries alive. There’s. So many ways, things could be improved about the way we communicate with other medical centers. Interoperability is a huge, huge challenge that we’ve been trying to tackle for years. Where I see a patient, they’ve been referred to me, their paediatrician has all their records and all their growth charts, and I’m like. Boy, I wish I had those. Yes, and they’ll fax me something and I can like pull up a very fuzzy PDF. I’m like, oh, there’s got to be a better way. And we know there is, we have standards like Smart on Fire where yeah, we were integrated appropriately. Everything that they had could magically show up on my growth chart, that so many opportunities there.

Q. I would like to ask you if there’s any other thoughts you have in closing, and if you were to look in the crystal wall, what you’d see, see coming our way in the next, let’s say, one to two years, and any other thoughts that you would like to share with the audience?

Yaa: I think my crystal ball is going to be really interested in how care can be delivered more at home and how we can get more insights from patients based on what they’re doing outside of the medical centre. So I’m a paediatrician, so I care a lot about kids and the fact that they spend most of their time just in school and the school nurse. That’s our main touch point to anything that’s going on with like our patient’s diabetes and what tools do we have that can help us. Better understand what’s going on with our patients when they’re outside of our medical center walls. So with diabetes, they have these things called continuous glucose monitors. That can help. Blood sugars are periodically. What are interesting ways for me to get the information in real time, to get alerts, to get, stratified risks of the patients that have the different technologies so I can know how to act. How useful would it be to have information like that available for. All sorts of patients for anyone on their watch, just so we can get a sense of which patients we need to make sure that we’re really engaging with so we can help improve their health and on the patient’s side, so they can have a sense of autonomy also, to say like, Hey, I really understand these details about my health. I’m getting these insights from my environment, and I can be proactive about how to stay well. 

Q: Great. So with that, yeah, I think thank you for the podcast being a guest on the podcast second time around like we, we talked about, and thank you for your time. 

Yaa: Thank you so much. I look forward to a third, fourth, and fifth time we’ll talk about Yeah, of course. What we got. Implement our, our new robot army to take care. We hope you enjoyed this podcast. 

Subscribe to our podcast series at www.thebigunlock.com and write us at info@thebigunlock.com   

Disclaimer: This Q&A has been derived from the podcast transcript and has been edited for readability and clarity.

Q: Hi Keith. I’m Rohit Mahajan, CEO and Managing Partner of Damo Consulting and BigRio, and also the host of the Big Unlock podcast. Welcome to the Big Unlock podcast. I’d love for you to introduce yourself, Keith—share your background, your role, and what motivates you on a daily basis.  

Keith: That sounds great. I’m Keith Morse. I am a pediatric hospitalist, which means I’m a physician who takes care of children admitted to Lucille Packard, Stanford Medicine’s Children’s Hospital. I’m a clinical informaticist, focused on studying and optimizing the use of technology and data systems for care delivery within a health system, specifically within informatics. My role is Medical Director for Enterprise AI, where I lead the team deploying and evaluating AI and large language models within care operations at the children’s hospital. I’m also an educator. I serve as the Program Director for our Clinical Informatics Fellowship Program, a two-year training program for physicians who have completed their specialty training and want to gain expertise in clinical informatics. 

The program prepares them for roles as CMIOs, Chief AI Officers, or positions in industry. Finally, I’m a researcher. I conduct research in AI and also serve as the Co-Site PI for Stanford’s participation in PEDSnet. PEDSnet is an 11-institution EHR data-sharing consortium that supports large-scale investigations to improve pediatric health.

Q: That’s awesome. I was reading about your background, and as we discussed previously, you did your MBA in healthcare administration and then chose to become a physician.
What motivated you in that direction? Do you still practice and see patients, and how does it work within the Stanford Medicine health system?

Keith: Certainly. I studied Econ and business as an undergrad, and where I was, they offered a combined MBA program. You add on an extra year to your undergrad training and you can start to get a sense of MBA-type training courses. I was doing my pre-med courses then as well, and I had a couple of years between finishing the MBA and starting med school. The job that I had was writing SaaS code for a consulting company, analyzing Medicare and Medicaid data for federal and state government agencies. 

That was essentially a data scientist role before the term “data scientist” was commonly used. And I loved it — understanding what data you have, how you can use it to answer questions, how you summarize it, and how you present it back to the requester. Those are such core things. I was in medical school in Philadelphia at Thomas Jefferson and then did a pediatrics residency in Phoenix, Arizona. During both of those times, I got involved in a couple of research and operational projects where I served as the data scientist. Writing R code, but also getting a sense of, instead of just being a data consumer as a data scientist, I was also a data producer. 

When you start delivering care, you are the person standing at the bedside, talking to the patient, making decisions, and also writing a note, ordering labs, and ordering follow-up. All of that is the beginning of the types of data that eventually get billed and trickle into databases that get used. In residency, I started working with our CMIO to analyze the data that was available through our EHR and really started enjoying it. 

Then I joined Stanford for their Clinical Informatics Fellowship and have been on faculty since then. We are fortunate here at Stanford Children’s to have a forward-looking CMIO, Dr. Natalie Pageler, who supported both myself and a small team to start building out our organization’s capabilities in operational AI back in 2020. We have been working at this for the last five years or so. Obviously, when ChatGPT hit, we got a whole lot busier, but it wasn’t the beginning for us. The processes we have in place now are built on that early investment.

Q: That’s pretty early to start in 2020. That’s a solid five years of experience with various kinds of AI implementations. How do you approach the business problem? Do you have specific use cases, and what do you do with them? How do you decide where to put your energy, time, and money?

Keith: I’ll answer this in three parts. First, I’ll talk about our process for identifying use cases, and then give a brief overview of two use cases that are relatively mature. First off, I love talking about use cases. In some ways, it’s a mythical concept. When operational AI folks get together at conferences or meetings, people sit around and whisper, “What are your use cases?” The reason is it’s such an important topic because we are, in essence, asking: In what ways have you found that AI is valuable? Where is the juice actually worth the squeeze? It’s easy to have research papers or proof-of-concept pilot projects showing that AI is theoretically useful. But when you have to make it work for an enterprise indefinitely, it’s a much different problem to solve.

We think about where AI can bring value without it costing $50 million or taking five years to implement. These are the types of considerations we focus on. This is actually a really hard question because three separate areas of expertise must align to arrive at good use cases.

The first is understanding AI technology in isolation. What is a large language model? What is a deep neural net? What is logistic regression? What can reasonably be expected of that technology? 

The second is understanding what infrastructure is available at my organization to support those tools.
It doesn’t help to have a sophisticated AI tool if you don’t have the data available, the compute power, or if the data isn’t available at the needed cadence for the algorithm. Another tricky part is that AI infrastructure is invisible. You can’t walk into a room and see where the AI lives. It exists in the ether. You have to be plugged into the organization’s IT structure to understand your current infrastructure. And it’s always evolving. We’re growing our infrastructure, making investments. It’s different now than it was two years ago, and it will be very different five years from now. 

The third and by far most important — and the hardest — is that workflow expertise does not solve nebulous, non-specific problems. It must solve a specific problem for a specific human, in a specific job, in a specific part of their workflow. We summarize all of that by saying “workflow,” where in a person’s workflow, AI can potentially be useful. The challenge is that healthcare is a diverse, complicated entity. My health system is different from the health system down the street, different from those in Cincinnati or Texas. 

Even within my organization, there are so many different workflows that no one person understands all of them. Our process for learning about these workflows — and this is something my team spends a lot of time doing — involves talking to people within our organization to help them tell us what problems exist and how AI could potentially help.
You would think that’s easy. You might think you can just talk to somebody and figure it out.
It turns out to be surprisingly difficult. The reason is this: if you imagine a standard organizational hierarchy — with director, supervisor, or executive oversight at the top, then managers, and then boots-on-the-ground staff — you find some patterns. This could be in Revenue Cycle, Supply Chain, Sanitation, or any other department. 

Usually, when you talk to the more senior folks, they are very excited about what AI can provide: speed, efficiency, safety, uniformity. They are generally onboard. Being a manager is different than being a boots-on-the-ground employee. Leadership often doesn’t understand what happens on a day-to-day basis in their teams. That’s not a dig on leadership; it’s just a fundamentally different job. People who oversee big teams — it’s too much to know. You can’t look to leadership to tell you where the problems are. Also, leadership is not the group whose jobs will be directly supported by AI. It’s going to be the people working on the ground. You have to talk to the folks whose workflows are potentially going to be impacted by AI to understand exactly what their workflows are and where AI could be helpful. Usually, the way we do this is to start at the top and work our way down. We get buy-in from leadership, then get passed to middle managers. They might suggest three or four areas that are worth exploring. We then meet with each of those teams and ask specifically what it is they do, what data they look at, what problems they encounter, and how AI could potentially support their work. Our hit rate in those meetings is relatively low. Most problems are not solved by sophisticated AI. There could be other, simpler solutions that work just as well. Often, our biggest takeaway is directing teams to existing data and reporting tools that can solve their problems without the need for advanced AI. But through this process, we do identify good use cases, and that informs our future efforts. 

The main takeaway is that no one outside your organization can credibly provide a guaranteed use case because they don’t know your AI infrastructure or your specific workflows. I don’t even know all the specific workflows within my own organization, so how could a third party or a hospital elsewhere know? There is a long-term role for internally identifying use cases, because they aren’t easily transferable from other institutions. One area where this is starting to shift is when AI is embedded within your electronic health record (EHR).

For example, our EHR is Epic. We use some tools they provide, like drafting responses to patient messages. This is a use case gaining traction nationally because Epic controls or manages all three areas I mentioned earlier: First, Epic understands that drafting patient messages is within the capabilities of a large language model. Second, all the infrastructure needed to use this drafting tool exists within Epic — no extra resources are needed. Third, the workflow for responding to patient messages is entirely within Epic, meaning they have a good understanding of the process. If the workflow required multiple steps outside the system — reading something in Epic, going to another machine, speaking to a patient, and coming back — it would introduce complexity and variability. But responding to a patient message is done entirely within Epic, so the workflow remains visible and consistent. That’s why use cases like this are more broadly transferable — because all the necessary components are self-contained.

Q:  Could you please share with the audience your experience with some of the large language models and where you have been successful in implementing them?

Keith: We have a couple of implementations, and we tend to publish most of the work we do, so a lot of it is available in the literature. One of our major use cases is using machine learning to help identify confidential content in teen notes. There are laws in the state of California and in most other states that protect teens seeking care for certain types of sensitive issues like mental health, sexual health, and substance use. In California, there are explicit laws that prohibit providers who care for adolescents from informing the teens’ parents that they are receiving this care. The reason for these laws is that there is strong evidence showing teens are more likely to seek care for sensitive issues if they are assured confidentiality. The challenge is that federal rules also require health systems to make patient records available to teens and their parents.

The 21st Century Cures Act requires that health information be available without undue effort. Basically, this means parents must be able to access and review records, usually through a patient portal within the health system. Anyone who takes care of adolescents faces a challenge: the federal government says you have to share all of the patient’s notes with the patient and their parents, but state laws say you can’t share portions related to three sensitive topics — mental health, sexual health, and substance use. We address this with what we call a “confidential note type,” where providers can document sensitive information. These confidential notes can be excluded from what’s shared with parents, and that system works well. Confidential information in these notes that we are theoretically sharing with the parents and families. This is a great application for large language models because it essentially processes large volumes of text to identify certain definable concepts. This is one of the projects we worked on before large language models came out. We developed a bespoke NLP model to help identify these concepts. We have replicated that analysis using large language models, and the way that we use it now is as a retrospective audit and feedback mechanism. We process the notes of different divisions, and then we can see who in the division is documenting in a way that’s potentially problematic. We can bring that back to the individual and also the leadership and say, “Hey, just FYI, we noticed that your notes contain a lot of information that would run afoul of the California state privacy laws.” 

We can then help providers improve their documentation. Often, what it is there is some sort of smart phrase or automatic pulling in of patient information that’s considered confidential. We just sort of edit or help update somebody’s note template, and the problem goes away . It’s those types of nudges that we think are helpful and really get at what we want, which is to be able to share patient information with the teens and their parents. We want this stuff to get out, we just need to do it in a way that is thoughtful and responsible.

Q: How about any other use cases?

Keith: Hospital operations. We have one that is not pediatric-specific, and it’s around tracking and quality metrics. This is work that we haven’t published quite yet, but health systems, as you well know, are on the hook to report on various quality metrics. One of the quality metrics is around surgical site infections—identifying what percentage of patients have an infection at the site of surgery within 30 or 90 days after the procedure. That’s a marker of surgical quality, infection control, and lots of really important things for health systems. We have teams within our hospital whose job it is to both track these metrics and then identify ways that we can intervene and identify bundles. That we can introduce to help improve these metrics. But there is a core function of tracking and identifying whether a surgical site infection occurred. The way that works is that there is a team of folks who essentially read the patient’s chart after a procedure to identify whether or not a surgical site infection occurred. Reading patient notes about how the patient came back to the ER with what looks like a surgical site infection, where they were prescribed antibiotics by their PCP—those types of things would flag as an identified infection. 

It’s a ton of work to have a human read every chart following every surgery for 30 to 90 days. So, we developed a process by which a large language model is used to review the notes in that relevant time window. We ask the model a relatively straightforward question: “Do you think this note contains evidence of a surgical site infection?” Very simply, that. We provide a few examples to help it understand what are some key indicators. We do that for every single note the patient has in that post-surgical time window and come up with a score, basically, what percentage of the notes the large language model thinks refer to a surgical site infection. When we run this on our historical data, we see that the large language model thinks somewhere between zero and 60% of the notes refer to a surgical site infection. What we can do then is draw a threshold, say 5%. What we do is have the human review everything above 5%. We still need a human to understand the nuances of what counts as a surgical site infection—reading the lab results, reading the imaging results. 

We need a human to identify the true positives, but there are many true negatives. Somebody who never has an infection, does great, doesn’t come back to the health system—those probably aren’t worth a person’s time to review. If we can identify the true negatives, the reviewer could spend more time on the true positives. What’s fun about this is that the numbers are exceptional. For our preliminary data, looking at 2023 and 2024, if we set the threshold at 5%, the reviewer would be reviewing 70 to 80% fewer cases. We would miss somewhere between two and five cases. It’s not perfect, but we are nudging toward a world where we can spend our time reviewing the cases that are problematic and positive, and less time filtering through the vast majority of surgeries that don’t have an infection or an issue, moving those to the back of the queue. We certainly aren’t replacing a person. We still need the person to be involved, but we are helping that person focus their energies on the cases where their expertise and their interpretation of what is actually happening is maximized.

Q: Yes, that’s great to know. As we all know, Gen AI and LLMs are becoming all-pervasive. You have a large workforce at your health system, people at various skill levels, and now they have to either use some of these systems that are going to be deployed or are already participating in what you described as a use case process. If there is more appreciation of Gen AI, LLMs, and AI, they would be in a better position to do their job and embrace it. So how do you go about learning and development and upskilling the workforce? What are your thoughts about that?

Keith: That’s a great question, and I think it is foremost on the minds of health system leaders who are hoping to use AI in any meaningful way. The reason it’s so important is that it is totally unrealistic to expect somebody to use a tool that they are unfamiliar with, particularly when that tool is something as amorphous and multifaceted as a large language model. If you think about it, if we use the analogy, large language models distilled down are basically like office productivity tools. Think Microsoft Excel—Microsoft Excel has been around for 40 years. I used it for middle school projects. My mom used it when she was working at a bank in the eighties. 

Most people, by the time they enter the workforce, have experience with using Excel. It’s not unreasonable to have that be a requirement of the job, or when Excel gets augmented in new ways, for people to be able to jump on it. It’s relatively straightforward. We have none of that with large language models. Expecting people to use these things without a ton of training is really unfair and unrealistic. That applies up and down the organizational chain. What I mean is that leaders, just because you are an executive VP, you have exactly the same two and a half years of awareness of these large language models as everyone else. Historically, people could turn to coworkers or get extra training, but that doesn’t exist here because it is so new throughout the organization. You make an excellent point about needing to upskill our workforce and attaching an experiential component to that training. 

Sometimes we hear about organizations using online modules for training, but I don’t think this is the sort of thing where watching some videos will really give you the insight you need to understand how these tools work. The reason it’s trickier in healthcare is that it’s one thing to tell somebody, “Hey, go play around with ChatGPT. See what you learn.” But we don’t want people to use ChatGPT to come up with recipes or poems in pirate. We want people to use large language models for their job because we suspect there are major productivity benefits that can come out of this. In a health system, most people’s jobs involve PHI—patient health information. You can’t put PHI into public models. So we need to make things available. We need to make a PHI-compliant large language model available to our employees. 

At Stanford Children’s and our adult hospital, we have made a large language model that is PHI-compliant available to our entire workforce. We’ve had it available for about a year now. Our IT department is called Digital Information Solutions, DigiIS, and our chatbot is called Ask Digi. The idea is to encourage people to start experimenting with an appropriately provisioned large language model to figure out how it can make their life better. I have no idea how somebody in a revenue cycle role could use large language models to make their life better. Most folks in that role don’t know yet either. We’re going to figure it out in a couple of years, but we have to let people experiment to learn that. 

We have three broad approaches for training: One is online modules and training. We have a couple of training tools, both generally about what large language models are and about Ask Digi and the local tools we have available. The second is prompt engineering workshops. Two of our clinical informatics fellows have developed a two-hour hands-on, in-person workshop on how to develop prompts and how to know that your prompts are doing what you want. Engaging with a large language model through a prompt is a totally new skill. Starting to get people to understand what is required in a prompt, lowering the fear factor, and giving people the confidence that you don’t have to be a computer scientist to prompt these models—it’s relatively accessible—and that comes through education sessions. The third is having a local champion or expert who gets involved in a pilot project and then brings that insight back to their teams. For example, in our work with surgical site infections, we are working with the folks in infection prevention and control. 

They are seeing the ways that large language models are helpful and not helpful. They are learning alongside us. Those folks will now become the experts in their team for how these tools can be used. Hopefully, that will propagate into more training. Engaging folks in these pilot use cases is helpful not only because you learn about the use case, but also because you train the person in that area, creating ripple effects across the organization.

Q: That’s great to know that there are so many initiatives for upskilling and learning, and development in the organization. One other topic that I thought I would touch upon—we talked about it a little—is PEDSnet. You mentioned it early on in the conversation today. I would love for you to share with the audience how this helps in terms of interoperability and data sharing, and any new aspects it might bring as well.

Keith: Definitely. Happy to. Just for a quick background, PEDSnet is a pediatric EHR data-sharing consortium that has been around since 2014–2015. It’s primarily run out of CHOP—Children’s Hospital of Philadelphia—which serves as the coordinating center.We have a relationship with them, as all members of the network do, and we send them a harmonized version of our EHR data four times a year. The harmonization is based on an OMOP model. The Odyssey community is an open-source EHR standardization initiative, and PEDSnet has adopted their common data model as the bedrock of what we use. 

We have some minor modifications that are specific for pediatrics and for care in the U.S., but it enable us to share data to a central location and conduct studies with a volume of patients that is unmatched elsewhere. About 15% of the children in this country are represented in our PEDSnet database. It’s a chance for us to do large-scale studies at a fraction of the cost it would otherwise take to develop these types of tools. As we move into the world of large language models, it’s not hard to envision a future where we are able to have a large language model help us process unstructured information from these different sites, extract relevant insights from notes, and conduct large-scale studies using unstructured data. That’s not here yet, but it’s in the near future. It’s really exciting to think about the potential.

Q: That’s awesome. Talking about the near-term future and potential, if you were to look in the crystal ball, what are some of the things that you see coming down the pipe? Agentic AI that people are talking about, virtual nursing, and ambient listening. There’s so much going on. What do you think are some of the big things coming our way?

Keith: Definitely. We are rolling out a pilot of ambient listening and have a similar cue as many organizations for things that we’re going to be adopting in the near future. Taking a slight step back from a regulatory and oversight perspective, it’s important to remember that those types of issues aren’t going away, regardless of how excited we are about the technology. Any investment in technology requires an understanding of the upsides and the downsides. I think we’re currently a little over-indexed on the upsides of this technology. We’re very excited. 

From a regulatory and oversight perspective, it’s important to remember that those types of issues aren’t going away, regardless of how excited we are about the technology. Any investment in technology requires an understanding of the upsides and the downsides. I think we are a little over-indexed right now in terms of what the upsides of this technology are. We’re very excited about what it can do now and what it can do in five years. We’re starting to see some efficiency gains. The feedback, particularly around DAX and other ambient listening, is generally very positive. We are less concrete about the downsides, and what I mean by that is there’s lots of talk about potential issues that come with AI. 

In the past, we have relied on federal or state agencies to provide oversight to make sure that those downsides aren’t present or are appropriately mitigated and recognized. AI, and particularly large language models, is proving very difficult to regulate because it’s such an amorphous entity. I think it is unrealistic that we’re going to see a robust regulatory system in the near future. What that means is that the burden for making sure this technology works is falling on the provider, and that’s great—until it’s not. What I mean by that is the burden is on providers to use these tools appropriately. At some point, we’re going to see a lawsuit from someone who claims they were harmed because of a large language model’s involvement in their care. Once that happens, we are going to get a very concrete piece of evidence about the types of downsides inherent in using these tools. We haven’t reached that point yet. 

Right now, the downsides are so amorphous that they’re easy to ignore. Once there is a price tag attached to the cost of mistakes, then things become different. If that price tag of a mistake is enormous, the overall value of these tools could change substantially. We know that, particularly early on, we are potentially introducing risks and mistakes with the use of these tools. Even if the output of a large language model is 99% accurate and you have a human in the loop who is reviewing it, and their review is 99% accurate, there are still errors that are present. Part of the reason I bring that up is that at Stanford, we take it very seriously. We are ultimately responsible for the use of these tools, how they impact our providers, and how they impact our patients. Nobody is going to take that responsibility from us. That is appropriate, but we are building systems and processes with that worst-case scenario in mind in order to prevent it from happening.

Q: That is very cautious. So with that, I think this is a great session. Any closing remarks?

Keith: Thank you so much for providing this opportunity. It’s exciting to be able to talk about the work that we’re doing here. I think there’s lot to discuss regarding pediatric implications, so maybe we’ll find some time in the future to talk again.

We hope you enjoyed this podcast. Subscribe to our podcast series at www.thebigunlock.com and write to us at info@thebigunlock.com   

Disclaimer: This Q&A has been derived from the podcast transcript and has been edited for readability and clarity.

Rohit: Hi Angelo, welcome to The Big Unlock podcast. We’re so happy to host you, and thank you for taking the time to join us today.

Angelo: Thank you, Rohit and Ritu. It’s an absolute pleasure. I was delighted to be invited. I’ve listened to some of your episodes—you have a fantastic program. It’s an honor to be part of what you’re doing here.

Rohit: Thank you! That’s awesome. I’m Rohit Mahajan, Managing Partner and CEO at Damo Consulting and BigRio, and also the co-host of The Big Unlock podcast.

Ritu: Angelo, welcome. Absolute pleasure having you here. I’m Ritu Roy, also a Managing Partner at BigRio and Damo Consulting, and co-host of the podcast. We’re looking forward to an invigorating discussion today.

Angelo:  Thank you again both for the invitation and the opportunity to speak with you today. I really like what you’re doing with this show. I think it is speaking to issues that are so important right now in healthcare, in innovation, in healthcare management.

So, it’s a tremendous pleasure. I can tell you a little bit about who I am and what I do, and then we can jump deeper into the conversation. But my name is Angelo Zo. I am trained as a pediatric cardiologist and still spend a little less than half of my time in clinical practice. And my clinical practice includes patients from prenatal life all the way through to adulthood who have primarily congenital forms of cardiac disease, but also acquired forms of cardiac disease.

And my clinical practice keeps me very busy. It’s a very interesting field. It’s a field where we have many applications of technology, which I think are really interesting. It’s a—pardon me—a field that includes some procedural medicine, some diagnostic medicine, and we work very closely with other specialists, including surgeons and anesthesiologists and clinical care physicians.

So, it really is a fantastic type of clinical practice, and it’s one that many people may not be familiar with, but is a really important part of what we do in the realm of children’s healthcare.

The other part of my work is as an administrator. So I am the Chief Medical Officer for what we call the Duke Health Integrated Practice. So I am here at Duke University Health System, which is based in Durham, North Carolina.

The Duke Health Integrated Practice—you can think of it as the faculty practice of our health system. It’s actually a bit more than that. We have about 4,200 physicians and advanced practice providers—so our nurse practitioners and physician assistants are counted within the faculty practice as an important part of what we do.

And we really are the specialty practice of the health system. We have another organization that manages primary care as their primary book of business—no pun intended. Primary care, primary book of business. So that is not part of my purview, but I work very closely with that organization. That organization has its own Chief Medical Officer, but I’m the Chief Medical Officer that interacts with our 18 clinical departments—everything from pediatrics to internal medicine to OB-GYN to radiology to surgery to dermatology, et cetera.

And I help to develop strategy—primarily for our ambulatory platform—and also implement and operationalize clinical innovation procedures that help us improve access, that help us open up our door, so to speak, to new patient populations, to new areas of practice—whether it’s medical or surgical, or diagnostic or procedural. So I have a broad responsibility here and a broad engagement.

And I think the final thing I would say is that our health system is associated with—obviously—Duke School of Medicine, which is a fantastic medical school, always considered a top-tier medical school. So we have a very significant mission around education, around training, around research, and around advocacy. And all those functions are directly connected to the clinical mission. So everywhere we practice medicine, we also want to innovate. We want to do research. We want to teach. We want to educate. And we want to advocate as well. So even though my role is primarily a clinical leadership role, I get to work with other leaders in all those other areas, which makes my job really multifaceted and really exciting.

Rohit:  That’s awesome. Thank you for that introduction, Angelo. You know, it’s been quite a journey for you. You’ve been with this health system for a long time, so would you like to share your journey with the audience? Like how did you get started? What interested you, and what are some of the new things that you are working on? And also, talk to us a little bit about the patient population in the local area.

Angelo:  Sure. No, I’d be happy to. So in terms of my journey into healthcare and healthcare administration, it actually started quite early when I was finishing up my fellowship training. I was offered the opportunity to build a branch of our academic pediatric cardiology practice away from the mothership, so to speak—to build essentially a large satellite office.

So very soon in my career, I got involved in practice building, practice management, and healthcare administration. Within about a year, I took over the medical directorship of this practice, and over the course of about 10 years, grew the practice from a very small operation with just a couple of physicians seeing a few dozen patients a week, to what is now a very large, multi-specialty practice that has about three dozen medical and surgical specialists coming in and out of our doors, seeing thousands of patients a week.

Very proud to have been part of this project from its initiation, from its conception, and through its phases of growth. Along the way, I had very significant hands-on training with executive leadership, with management, and also with day-to-day clinical operations, which became an area of interest to me.

I was eventually offered the opportunity to take on a new role, and that role was as the Vice Chairman for Clinical Practice for the Department of Pediatrics. And in that role, I took what I had learned at the local practice level and brought it to an entire department, and became the operational and strategic leader for everything we do in our Department of Pediatrics, which includes about 250 or so clinical faculty, another 50 or so research faculty—primary care, specialty care, diagnostic care, urgent care, emergent care, inpatient, outpatient—all different aspects of our care.

I did that work for a little less than a decade, and I’m actually moving away from that role now, because last year I stepped into yet another role as the Chief Medical Officer for the Duke Health Integrated Practice. So I’ve been in that role now for a few months, and I hit the ground running when I took on the role—immediately, as in my first day on the job—had to help with some challenges we were going through as an organization.

It’s been very exciting, in just these first few months, to begin to see the opportunities that I will have. In terms of—you know, you asked about some of the things that we’re working on—and we’re going to talk today about some clinical innovation, and that’s been extremely exciting to work on.

Rohit:  That’s awesome. Yeah, so that leads us right into our next topic. What are some of the clinical innovations or challenges you’ve faced that you were able to overcome? And what are some of the digital health programs you’re getting involved with? Anything you’d like to dive into and share more about?

Angelo: Yeah, fantastic question. So, my first day on the job, I had to address the audience of our entire practice—all the clinical leaders from across the health system—and I told them that I saw my role as Chief Medical Officer as one who is there to support clinicians in their day-to-day work.

My platform rested on the idea that I want to bring aid, tools, and tactics into the practice that actually help people do their work. In the last few years, we’ve been facing workforce challenges—people feeling burned out, disillusioned, like cogs in a machine. I want to bring the joy back into the practice. And I think the way to do that is to allow physicians and other clinical providers to spend as much time treating patients as they can.

So we’ve identified areas where we can innovate—where we can bring technology to bear to improve the experience of delivering care for our doctors, nurse practitioners, physician assistants, and other clinical personnel.

One thing we’ve done—something very concrete and in partnership with a very innovative healthcare technology organization—is apply what’s called ambient listening technology. And we’ve done this in a way that’s been thoughtful, in response to a specific problem: how can we alleviate our clinical providers of some of the burden? In this case, the burden of documentation.

We know that physicians and other clinical providers today spend a lot of time in front of a computer—often while they’re in the room with a patient. This can make people feel like their job is data entry. We don’t want the job to be data entry—we want the job to be managing patients. We want it to be engaging. Spending that face-to-face time matters, because we know so much of diagnosis comes from the history, from having an in-depth conversation and hearing the patient’s story.

Yes, we can do a lot with diagnostic testing, but at the end of the day, it’s about engagement with the patient. That’s how we build a differential diagnosis. That’s how we begin to form our assessment and plan.

So what we’ve done is bring this ambient listening technology into the clinical encounter. The tool records the conversation between the clinical provider and the patient, and from that back-and-forth, it creates the encounter documentation.

It may seem like a simple evolution of the standard tape recorder-to-human-scribe model, or even having a human scribe in the room. But this is truly the modern version. The conversation is filtered through a natural language processing algorithm, and the finished product is something that’s usable as clinical documentation.

This has been a very exciting opportunity for us to leverage technology at the point of care and remove some of that documentation burden.

Rohit:  That’s a great use case—one of the hottest we’re seeing in many health systems, Angelo. Tell us a bit more.
You mentioned some metrics before we started. How do you measure the success of this wide-scale implementation? And you also said it’s integrated with the EMR/EHR system. These are critical for those still thinking about this.
Many have already taken the plunge with varying degrees of success, but you’ve really embraced it and rolled it out broadly. Any key takeaways from your experience that were instrumental?

Angelo:  Yeah, fantastic questions—thank you.
I’d start by saying our pre-implementation data was both qualitative and quantitative.

From a qualitative perspective, we were literally hearing from our doctors, nurse practitioners, and clinical providers that they were losing the joy of practice. These are people who’ve invested so much time and energy into becoming clinical providers. They want to engage with patients—that interaction is a special opportunity.

But over the last several years, we’ve created systems that make them feel disconnected. These electronic health records require tons of digital input—typing, checking boxes, clicking through screens.
At the same time, we’ve opened the door to patient interactions outside the visit. They’re now answering after-hours messages, doing telemedicine and telehealth visits, which are great but can also add stress.

So it starts with qualitative data. Are our people enjoying their practice? If not, we have a problem. We saw workforce challenges—many physicians and nurses left during the pandemic. We need countermeasures.

On the quantitative side, we look at “signal” data—engagement with the EHR: login times, time spent in the system, responsiveness to messages, how quickly notes are closed, and whether they meet patient expectations.

We also track provider ratings. We ask questions like: Did your provider listen? Know your history? Speak clearly? Explain things well? These bridge qualitative and quantitative—they’re numeric but reflect the experience.

If a provider has low ratings, we drill down. Often they say, “I just don’t have enough time with my patients.” And it’s not just visit length—it’s quality.
Those were some of our pre-implementation metrics.

Another key part of our approach: bringing our performance engineers to the practices. We call them performance excellence engineers. They’re incredibly skilled at turning data into actionable information.

But in recent years, they’ve been working remotely or from centralized offices—not their fault, it’s how we structured things.
We’ve found that when you put them at the point of care—whether it’s the OR, ED, primary care, or endoscopy suite—they really begin to understand the work. That’s when they can help translate practice data into something actionable.

This is classic business school 101—Toyota production system. Bring the people doing the work into the improvement process. We had drifted from that. Now we’re returning to it and seeing the benefits

Ritu:  That’s awesome. Dr. Angelo, I was at a conference recently where Abridge—one of the main players in the ambient space—shared that 81% of notes didn’t need to be edited by doctors. That led to a huge jump in satisfaction. Are you seeing similar numbers?

Angelo: Ritu, this is a very interesting question, and it’s interesting for a number of reasons. First of all, full disclosure — we are using Abridge, so that is our partner in this space. One thing that’s been wonderful about that relationship is they’ve craved our feedback about the product.

They want to iterate, and that’s why they were such a natural partner for us — we share that DNA around innovation. They want the product to keep getting better and better as we use it.

To your comment about 80% of the notes not requiring editing — this is a really interesting data point that we are looking at. We’re actually tracking the number of times notes do or do not get edited, and there’s a signal in both directions that we need to be careful about.

We need to be careful about the people who never edit their notes — because that’s one kind of story — and also ask about the people who are always editing their notes.

I can tell you that in my practice — and I have more than 20 years of clinical experience — and as someone who likes his notes done a very specific way, I’ve actually been quite happy with the quality of the content.

And when I have edited, it’s been primarily not to correct factual errors, but to change the language to better match my personal style. So it’s really been more about the aesthetic of the note rather than the content itself.

I’ve found the content to be quite good. And when it hasn’t been good — this is the other nice thing about it, Ritu, you mentioned earlier that it’s embedded in our EHR — we can provide feedback immediately.

So if I see a note that didn’t come out the way I expected, we have a chat box — we can go right into it and talk directly to the development team and say, “This is why I don’t like this note.”

We can also see the stems — the original stems of our conversation — and how the natural language processing model created the note from those stems. And again, having that direct pipeline to the developer has been critical.

Just a few weeks ago, the Abridge team came to us. They did an extensive two-day site visit — and again, they wanted to hear directly from us, from our doctors, from our clinical providers — how they were engaging with the technology, what was working, what wasn’t.

With very rare exception, most of our people have said — and I hesitate to use the term “game changer” because everyone uses it and it’s almost lost meaning — but this has been disruptive in the best possible way.

This has enhanced our practice. I’m able to close my notes. I’m able to leave the office at the end of the day not having to go home and work after dinner, or work in my pajamas, or while I’m trying to do something else — wanting to spend time with my kids, or being taken away from my hobbies and interests.

This is allowing me to close the books at the end of the day, feeling good about the quality of the content.

And again, so much of the upside is that ability to spend time — and we’re really interested to see how we improve on that one key metric: Did your provider spend enough time with you? I’m going to be really interested to see how that improves with the implementation of this technology.

Ritu: Yeah, so in the same conversation they also mentioned that some of their physicians are seeing Russia traffic for the first time, which can be, you know, good or bad depending on how you look at it.
So to your point about being able to close the notes and go home — that’s really good to know.

Angelo: I think it is. One thing we really like about this technology is the modularity. You can use it for certain elements of your note and not others. You could use it for the entire note if you like, and that flexibility has helped adoption.

If you’re really good at documentation and only want to use it for your history of present illness, you can do that. We’ve encouraged everyone to try it, and almost universally, people have at least given it a shot.
Most who’ve tried it have stuck with it — and we’re tracking that with metrics. We know what percentage of users only tried it once, how often it’s being used across different note types, etc. We’re seeing very strong onboarding of the technology. We’re not at the plateau yet — we’re still ramping up — but it’s exciting to see adoption growing.

Rohit: Yeah, shifting gears — I know we haven’t touched on this yet, but Angelo, what are your thoughts on value-based care? Is that something your practice is actively thinking about?

Angelo:  It’s very important—and it’s part of what we’re trying to do. I’ll tell you, in this part of the country, the Southeast United States, I think the move toward value-based models—bundled care and similar approaches—has been slower.

When we look at our colleagues in the Northeast or on the West Coast, we see folks who’ve been very innovative and progressive in this space. We’re certainly trying, and I think there are a number of reasons why adoption has been slower here—probably more than we can cover in this conversation.

Nevertheless, we’re very excited about the opportunities to improve care efficiency and reduce costs. We’re looking at how to bundle services in a way that clearly shows value—taking that classic formula of quality divided by cost. We believe our quality—the numerator—is very, very high. Now it’s about shrinking the denominator and reducing costs.

There are ways to do this that may involve technology, but also some analog solutions too.

We work very closely with payers and are really interested in what they have to say. Over the past few years, we’ve built strong, bilateral relationships with payers to understand what’s on their radar in terms of quality and performance metrics. At the same time, we want to hear from our own clinicians—what’s important to them?

Sometimes that aligns with what payers are looking for, and sometimes it doesn’t. But when it doesn’t, we have strong enough relationships that we can go to payers and say, “Hey, this matters to us.”

For example, our pediatricians have raised concerns about adolescent immunization rates. A payer might be more focused on vaccines in the first year of life or among older adults—like meningococcal, pneumococcal, or RSV vaccines for seniors. But we’re hearing from our team that teens aren’t coming in for pre-college immunizations.

In most cases, payers are at least willing to have that conversation.

We’re also thinking about how to clearly demonstrate our value proposition—especially with so many large employers moving into Central North Carolina. With the research triangle and three major universities here, this region is growing fast.

We want to make sure that when employers include us in their health plans, we can go to them and show the value—whether it’s for hip replacements, organ transplants, or more common services like primary or preventive care.

Historically, large academic medical centers haven’t always been great at cost containment. But we’re learning. We’re identifying opportunities, especially around clinical operations. There’s a lot of potential for efficiency there—and when you focus on that, you can truly begin to lower the overall cost of care.

Ritu:  That’s awesome. I would like to ask you about AI education and the trustworthiness of AI. Did you face any barriers from your physicians on those regards? Sometimes we see that physicians can be reluctant to try new solutions. There was a New York Times article that showed when ChatGPT was diagnosing on its own, it was much better. But when the physician came in, they brought in their own biases, and it actually did worse because they didn’t trust the AI and took it in a different direction. How have you made sure everyone is educated and had the right context within your organization for such a wide rollout?

Angelo:  Again, fascinating question. We could spend a whole podcast on just this. We have certainly needed to address this issue, but I’ve been pleasantly surprised that most of our clinical providers—physicians, advanced practice providers, physician assistants—have asked questions in a very thoughtful way. They’re not necessarily showing hesitancy, but they are trying to keep pace with developments in this area. They want to ensure we’re good stewards of the technology.

They’ve asked questions like, “What are we doing with these recordings? Are we using patient data to improve the system? Are we training the language models with our information?” They’ve also wanted to know about the process for obtaining informed consent. And yes, we are certainly asking for permission every time we use this technology, in every instance.

One thing I’ll mention is that we have a very robust structure here for digital innovation. We have several teams, and one that focuses specifically on AI. They’ve done great work ensuring our faculty are educated on basic terminology around AI, natural language models, and generative AI. Most people feel they have amazing colleagues providing these educational materials.

Another key step was the pilot program. Before rolling out anything, we did a well-thought-out pilot. We tested a few technologies and opened it up to a large number of clinicians across the entire system. This was incredibly beneficial because during the pilot phase, which lasted several weeks, clinicians provided us with feedback. We conducted pre-, intra-, and post-pilot questionnaires to evaluate all the technologies and assess what worked and what didn’t.

During the pilot, we also did a lot of education. We had review modules, town halls, and went to all 18 of our clinical departments to answer questions. Our data and IT teams, including many doctors, were also part of those conversations. That was a huge benefit because when you have IT leaders who are also physicians, it builds a lot of credibility with the clinical audience. I think that was key to winning people over.

Rohit:  Yeah, I’ll chime in with another one, Angelo. We’ve all been hearing about agentic AI, and how agents are the next big wave in generative AI. We’re currently working with several clients to define use cases for implementing agents in the workplace, as they’re talking about creating a new digital workforce that includes both humans and agents. Any thoughts on that?

Angelo:  It’s an area that’s going to grow exponentially. Right now, we’re testing the ability to use AI to screen inbound messages. As I mentioned earlier, one of the outcomes of opening up digital access to our patients is that we’re receiving thousands of messages per day. That’s a good thing—we want our patients to communicate with us—but it’s a lot of work, real work.

We have an AI innovation research team working on models that can review these messages and sort them into categories. For example, distinguishing between a message that just says “thank you for the great care” and one that says “I need a prescription refill” or “I need an urgent appointment with my orthopedic surgeon.”

The next stage is not only sorting these messages but starting to build at least the skeleton of a response. This is a really complicated problem, even for humans to do. But what’s exciting is that we have people who recognize that this is another way to alleviate some of the clerical burden on our clinical teams. If we can reduce the number of messages that require human attention, we can focus on the urgent ones. The others are still important, but they’re not as urgent. If we can triage this, it will be incredibly helpful.

And to your point, Rohit, I think the future will involve a synergy between human and technological capital. If we can create that synergy, perhaps people won’t lose their jobs but will take on new kinds of work. That’s the future I see.

I’m old enough to remember the early days of the internet, and the doom-and-gloom stories that came with it. What we’ve learned is that we need to approach these tools with an open mind and recognize that we can use them responsibly if we choose to. It’s about figuring out the most responsible and applicable use cases, and being good stewards of the technology.

I’m encouraged because many people in the space, especially those applying these solutions to healthcare, really understand the practice of healthcare. That’s why partnerships with institutions like Bridge, or technologies like Copilot, are so important. When you have a strong clinical-technological partnership, amazing things can happen.

Rohit: Angelo, would you like to share some closing remarks?

Ritu:  I think we need another podcast for more questions! But Angelo, thank you for being such a great guest and providing really insightful answers. We’d love to hear your closing thoughts.

Angelo: This was a fascinating conversation, and again, I love what you guys are doing with this show. I hope we can convene again at some point. I’d love to come back.

In closing, one thing I’d leave the audience with is that these feel like incredibly challenging times, but we can do incredibly challenging things in response. Sometimes, the most amazing road to success is when you’re pressed and facing obstacles. The tools we have today have amazing potential. We just need to be good stewards of that potential—thoughtful stewards of it.

We shouldn’t be afraid. Instead, we should continue to ask the fundamental questions and think of the future as a synergistic one, where we combine the best of our human capital with technological tools and innovation, figuring out the next step forward.

I believe in the idea of abundance. I think we have a future full of great potential, but it’s a matter of deciding how we engage with that abundance. How do we take the tremendous potential of natural language models, generative AI, and other technologies, and apply them thoughtfully, rationally, and in a way that speaks to the needs of those moving these things forward?

In our case, we listened to what our people were telling us about their work and realized that the only way to really move forward is to change the work itself. We can provide support, offer a shoulder to cry on, or an arm to hold in times of distress. But when we change the fundamental nature of the work, that’s when we begin to truly change people’s engagement with it and make them feel supported, with a clear way forward.

Doctors are resilient. We don’t lack resilience. What we need are tools to help us engage with the work by changing the terms of the engagement.

We hope you enjoyed this podcast. Subscribe to our podcast series at www.thebigunlock.com and write to us at info@thebigunlock.com   

Disclaimer: This Q&A has been derived from the podcast transcript and has been edited for readability and clarity.

 Q. Hi Dwight, welcome to The Big Unlock podcast. My name is Ritu Uberoy, and I’m one of the co-hosts. We’re very happy — this is season six, and we’re glad to have you back. Delighted to hear new perspectives, especially on AI and your initiatives at your new organization. Let’s get started with your introduction and then dive into a few topics.

Dwight: Great. Really great to meet you and be here, Ritu. I’m Dwight Raum, Chief Digital Information Officer for Rochester Regional Health. We are a health system with about nine hospitals in Western and Northern New York. We span the spectrum from large, urban acute care hospitals to small, rural regional hospitals.

We also have a large ambulatory practice throughout the region, and we really take care of patients from birth to death — and every aspect of healthcare in between. I’ve been with the organization just over a year.

Before joining Rochester Regional Health, I was in the startup world for a short time. Prior to that, I served at Johns Hopkins Medicine as the Chief Technology Officer and Interim CIO. I’ve been in healthcare for quite a long time. I really came up as a technologist, which is increasingly unusual in senior IT leadership roles. But I’ve truly enjoyed my time in healthcare — the mission really calls to me.

 Q. That’s a very interesting introduction, Dwight. It seems like you’ve seen all three aspects of technology and implementation — being at Johns Hopkins, which is a world-renowned teaching hospital, then moving to a startup, and now at Rochester. What was your role at the startup, and what did the startup do, before we get into your current role?

Dwight: Sure. It was technically a joint venture between Independence Blue Cross and Comcast. The startup was called Quil Health at the time. We were trying to build an in-home monitoring system that would allow seniors to age in place.

That consisted of various IoT sensors placed throughout the seniors’ homes. From those sensors, we ran AI models to anticipate and predict certain behaviors and patterns. Insights were drawn from the data and shared with caregivers.

It was a really interesting idea — just struggled to get it across the line, if you know what I mean. But I think it had tremendous potential. It actually speaks to the opportunities we have with signals and AI, and how that can really transform the way we provide care.

Q. Yeah, that’s very interesting. We had a webinar yesterday where we were presenting some case studies. One of the case studies we were going to present — but didn’t — was for Cherish Health. They do something very similar, radar-based. They have a device used in senior homes that can detect falls, go through walls, and anticipate events using AI.

Dwight: Yeah, I mean, the technology has evolved certainly over the last several years, and radar fall detection has really improved. But I still think there are a lot of challenges — in terms of privacy and getting over that technical hurdle for seniors to adopt and use the technology with trust.

Those are still barriers that aren’t easily overcome, even with the technology in place.

Q. Absolutely. Okay, great. So tell us a little more about your role at Rochester Regional Health — what the digital landscape is like there, your experience with generative AI, and some of the initiatives happening at your organization.

Dwight: Absolutely. As I mentioned, we’re a nine-hospital system. We’re mostly on one EMR — we’re an Epic shop, single instance. We do have some smaller EMRs, but we’re slowly but surely moving everything into a single instance.

In terms of initiatives, I’d group them into a couple of categories. The first is optimizing the system and the engine of care delivery we already have. That means ensuring throughput, doing so in a safe and high-quality manner, and providing access to our patients.

The second part is really transforming how we deliver care — shifting from a traditional fee-for-service model to a much more risk-based, value-based model. We’re focusing on treating patients for their outcomes instead of on a transactional basis.

That shift brings a lot of changes — not just in how technology works, but also in how our operators and clinicians provide care. It’s a transformational change for the system.

A lot of credit goes to our CEO, Dr. Chip Davis, who’s really leading that charge and planning for the future. When 2030 comes around and CMS starts to pay more earnestly based on value, I think we’re going to be very ready here in Northern New York to take on that challenge.

Q. In terms of the Clinical Access Center — is it a unified center? We recently worked with Sentara on a huge multi-year project where they brought everything into one place. Now they’ve implemented AI on top of that to handle call routing.

We did a webinar on their case study, and they shared something interesting — because of the AI they implemented in the call center, they can now predict when call volumes will be lower. That allows nurses to take flexible time off, which has been a big plus for them.

Dwight: I think that’s a great comparison for us. We do have a unified call center, and we are absolutely using AI for call routing and predicting high call volumes, etc. We’ve really been able to drive efficiency using AI in the call center.

Our call center is located right next to our command center, so we’re constantly looking at overall system operations — understanding volume, where to shift resources, where to move patients, and so on. Combining that with the call center is incredibly powerful.

Our call center isn’t just receiving calls — we’re also driving outbound traffic. And the third leg of the stool is that we’ve just rolled out a digital front door on our website — a robust digital experience for patients. It offers capabilities for patients to serve themselves, schedule appointments, and find the closest care options. We’re really trying to cover the full experience and serve patients as consumers as best we can.

Q. For the digital front door, do you have AI as well? Like virtual chatbots? How does that work?

Dwight: We’re not deploying chatbots in a very robust way at this point. Honestly, I’m not sure I see them as a strong tool for care delivery. There’s probably some limited front-door use for chatbots, but I don’t think they’re a great solution overall.

That said, our digital front door does provide access to direct virtual care. Patients can usually get an appointment within an hour or less — they’re able to see a provider almost right away. So it’s really about combining access with technology to deliver care where patients are, in almost real time.

Q. So. Sentara mentioned they use something called Edia for the AI part of their call center. Do you know which AI implementation you’re using?

Dwight: Yeah, we’re a customer, and there are embedded AI tools within Genesys — that’s what we’re using.

Q. So, how big is the technology component at Rochester Regional? And do you see a role for a Chief AI Officer? That’s a hot topic these days — ethics, governance committees, Chief AI Officer roles. Would love to hear your thoughts.

Dwight: That’s a broad question — let me take a shot at it. In total, we’re about 400 people in our IT organization. I might have had a bit of a Freudian slip earlier — I’m very much trying to reorient the organization to think about AI, but more broadly, innovation.

So, how do we bring technology and new ways of thinking to transform the tools and experiences we deliver?

To your specific question — do I anticipate us having a Chief AI Officer? The simple answer is no. I don’t think that makes a lot of sense, quite frankly.

What we have done is create a Center of Excellence for AI. It’s a group of the right people who need to be at the table to do two things. First, there’s the “pull” — a lot of demand from our operators who see new tools and ask, “Hey, can I get this?” We need to be thoughtful about privacy, contracts, ethics — all the right teams are part of the COE to address those questions and respond responsibly to that pull.

But there’s also the “push.” AI has been evolving at an incredible pace — even in just the last 6 to 18 months. It’s impossible for any one person, or even our providers, to keep up with it all. So the second part of our AI COE’s responsibility is to articulate what’s possible — the art of the possible.

We’re maintaining an initiative list of opportunities we think are ripe for AI — places where we can deploy capabilities, whether it’s LLMs or machine learning, to have an immediate or significant impact.

Q. And for all your AI implementations, are you planning to stay within Epic, since they’re building their own AI tools? Or are you looking to integrate outside solutions too?

Dwight: I think we’ll always look at what we already own first. Epic is a chassis in many ways — we can plug in different Epic modules and activate them. So yes, we’ll definitely start there.

But that doesn’t mean we’ll only look there. As I mentioned earlier, Genesys offers incredible AI capabilities — that’s an example of something outside of Epic but still integrated into our environment.

So no, this isn’t going to be an Epic-only play. But there are definitely some low-hanging opportunities within Epic that we’ll evaluate, ensure they provide value, and then activate them for our providers.

Q. Okay. I’m just mulling over your answer because other CIOs and CTOs we’ve been talking to feel the need for a Chief AI Officer, for the same reasons you mentioned—trust, governance, and the pace of change. It’s difficult for one person to keep up. Some believe you need a dedicated person whose job is to look at that landscape, understand what’s happening, and decide what’s best for the company. Just playing devil’s advocate here, trying to look at both sides.

Dwight: My reaction is, if we were in the business of developing AI ourselves, I would agree. But we’re not. We’re a system focused on serving our patients. Our priority is patient care, and we’ll use AI to support that. I do think it requires a collection of experts. From legal, privacy, and algorithmic safety standpoints—maybe a Chief AI Officer could be the lightning rod that brings that all together.

But honestly, we’ve come up with a methodology that works well for us in the COE. It allows us to bring together those experts and, more importantly, align their interests. Even if you have a Chief AI Officer, they still have to orchestrate all the soft power and influence that drive real change. In my view, the AI COE approach forces that to happen from the start.

Q. Yeah, that’s a very astute observation. That’s exactly what we’ve learned in our webinars—it can’t be mandated; everyone has to buy into it. It’s a behavioral change, especially with these new tools coming in.

The other point of view I have, which came up in a webinar yesterday, is that the EU has passed a law mandating AI training for any company using AI. What do you think about the training landscape? Are all your employees getting AI training?

Dwight: No, we haven’t mandated broad-based AI training yet. That’s likely to happen depending on the tools we’re using. We talk about it a lot as leadership. As we approach deployment of smart ML models for diagnosis or therapies, there will be very targeted and intensive training around each one.

Q. So as of now, people just learn on their own? Both—like LLMs or ML tools.

Dwight: So, as I said, I’ve personally hosted several webinars across the organization to provide baseline information about how tools work. But we’re not yet offering formalized training. The challenge is that healthcare is very busy and, in many ways, understaffed. Capacity is a real concern. We have to be cautious and ensure training is impactful.

My sense is that as opportunities become more acute, and we identify specific use cases, we’ll do more. We’ve also deployed an AI policy that specifies what is considered safe use for generative AI.

Q. In your hospital environment, do you have people using ChatGPT and possibly uploading data without realizing it could become public or used to train the model?

Dwight: That’s been part of our internal communications campaign—clarifying what’s appropriate. It’s hard to police, and I can’t say for certain it hasn’t happened. But we’ve provided guidance and education. A lot of people don’t realize that unless you opt out, anything you type can be used to train the model.

Q. Exactly. Okay, great. So, Dwight, talking to multiple CIOs and CTOs, the main generative AI success stories we’re hearing fall into two categories: scribing and ambient. What’s your experience with either, and do you have a success story?

Dwight: Honestly, we’re a little late to the game. We’ve just started our pilot for Ambient Digital Scribe. We’ve done early testing, and it’s clearly a win for providers. It helps reduce pajama time and after-hours documentation. But these tools are also incredibly expensive, so we’re evaluating ROI—how to unburden providers while sustaining the investment financially.

Q. Yeah, very interesting. At HIMSS last year, Nuance shared that ambient tools save about six minutes per patient. That doesn’t sound like a lot, but across multiple physicians and patients, it adds up quickly.

Dwight: Absolutely. Over the last 15 years, digitizing healthcare hasn’t been seen as a net plus by most providers. It’s negatively impacted their quality of work. COVID contributed to burnout, but so has technology fatigue. Any opportunity to unburden providers using tech—we should seize it.

Q. Good to know you’re in the early stages and testing. Any other digital programs you’d like to highlight?

Dwight: Yeah, I think, you know, there are several that I would highlight, and I would kind of back it up to a higher level for us. As I mentioned earlier, Chip, our CEO, has really been very much driving transformation. But I think a supporting function there is really innovation writ large, and AI fits into that innovation model as well.

So we’ve really begun to create an entire innovation program for all of RH, where we have two parts to that program. The first is focused on internal performance improvement—transformation and innovation. And then we’re coupling that with an accelerator program to actually take some of those innovations to market, and also to partner with other startups to bring them back into RH.

We’re doing this in a very prescriptive and deliberate way, ensuring it’s aligned with our strategy. So, where we have opportunities or challenges, we’re trying to line up those companies that may actually fit and help us solve some of those niche problems. I’ll give you a really good example of this.

Healthcare is currently experiencing a drop-off in employees, and there’s certainly a challenge in maintaining staffing. Nurses, I think, are the most acute area where we see that. But we’re also in this perfect storm where demand is increasing at a remarkable rate. So that gap between supply and demand—the only reasonable way to think about closing that gap is through technologies.

As you mentioned, AI—and we talked about Abridge as a digital scribe—as one way of doing that. But there are certainly other opportunities in that space. Just scheduling nurses, for instance, is incredibly burdensome right now and leads to tremendous inefficiency. Every little point of improvement we get there pays multiples in dividends.

So as we look at opportunities, a good example is nurse scheduling—can we apply AI to actually improve the speed and efficiency of that process? And the answer is yes. That’s a great example of a strategically aligned innovation opportunity that we’re actively pursuing at RH.

And I can stack up numerous examples like that. We talked about the digital front door and trying to improve access by casting a broad digital net, but also empowering our patients with tools to self-serve and gain access to care as quickly as possible.

It’s those types of activities that we see as the through line from the innovation continuum I was talking about earlier. So while an innovation may start by solving a very acute problem within RH, we believe we can develop those ideas into something that scales across the entire organization—and potentially even goes to market more broadly.

Q. That’s really interesting. The accelerator program—does it include in-house participation too?

Dwight: It does. A great example is a physician in emergency medicine who developed an app called COIST. It’s a heads-up display on an iPad that guides cardiac resuscitation based on ACLS guidelines and documents the process in real-time. We helped develop it, and it’s a great example of internal innovation with life-saving potential. It also eases the provider’s burden in emergency care—and it’s not a problem unique to RH. We believe it has wider potential.

Q. Yeah. So they’ve seen really good success with longitudinal care plans and generating care plans—still with the human in the loop to make the final call.

Dwight: I think especially with LLMs, that’s critical. I don’t see that changing anytime soon. One of my personal hobbies is coding, and I use LLMs all the time. It’s remarkable how good they are, but also remarkable how they hallucinate—just make things up. So I think there’s a definite need for caution and keeping a human in the loop.

That’s why I think RAG implementations have seen greater success—because you can bound them, put guardrails in place, and ground them in truth.

Q. Any challenges you’d like to talk about? What do you feel is your biggest challenge with technology?

Dwight: Yeah, I mean, I think, you know, the financial pressures in healthcare are certainly still very acute. And I think, in a lot of ways, that’s the oxygen you need to really drive wholesale innovation and transformation. So that’s certainly a challenge.

I think cybersecurity is also something that continues to be a top-of-mind concern for any CIO these days. And, for that matter, the accompanying investments you have to make to really tighten things up, so to speak. Mm-hmm.

Those are the challenges I see. And there are challenges around workforce, too—maintaining a team that can really drive and advance the ball is very critical as well.

By and large, I’m an optimist by nature—and a technologist by nature. I actually think there is real opportunity. If you take a step back and look at that gap I was talking about earlier—around supply and demand—and, for that matter, how much we spend as a country on healthcare, I do think there’s good reason to be optimistic about the future and how technology can play a really important role in helping us, as a society, meet the demand of healthcare and actually improve people’s lives.

We hope you enjoyed this podcast. Subscribe to our podcast series at www.thebigunlock.com and write to us at info@thebigunlock.com   

Disclaimer: This Q&A has been derived from the podcast transcript and has been edited for readability and clarity.

Q. Thank you for joining the podcast, Sophy. It’s great to have you here.​ Likewise. I’m Rohit Mahajan, managing partner and CEO of BigRio and Damo Consulting, and the host of the Big Unlock podcast. As you know, it was started by Paddy a long time ago. We’re honored to continue this podcast. We’re now in the 150s episodes, so it has come a long way. Sophie, could you please introduce yourself?

Sophy:  Thank you. My name is Sophy Lu. I am currently the Senior Vice President of Digital Integration and Business Relations at Northwell Health. Previously, I’ve held multiple executive roles at Northwell, including System CIO, and I’ve been with Northwell for approximately 15 years. I grew up in the application space. Prior to that, I was trained as a chemical engineer. I enjoy continuous learning, integration, connecting people, and solving problems by bringing common to chaos. I live in Brooklyn, happily married with two grown kids. I have four Labradors and two Norwegian cats.

Q. That’s awesome. So, how does a chemical engineer like you, Sophy, wind up in one of the largest healthcare systems in the country? Please tell us a bit about your journey and about Northwell as well.

Sophy: Absolutely. I often say it’s a mix of luck, timing, and a lot of hard work, along with the networks you’re exposed to. Initially, I wanted to be a kindergarten teacher, but my love for science and math led me to chemical engineering. I worked in that field for about seven years after graduation, focusing on process and industrial design for chemical plants. This role took me across Europe and Indonesia.​

During that time, I was drawn to sustainability, working on clean energy projects. About 20 years ago, I was involved in expanding liquefied natural gas and combined natural gas. While these concepts are well-established today, back then, scaling them in the transportation and energy industries was quite ambitious.​

As the engineering sector shifted more towards the central United States, I wanted to stay close to my family here. I explored other opportunities, sending out numerous resumes at job fairs. A consulting company noticed my background and brought me on board to work on technology integration and implementation, leveraging my industrial gas experience.​

This consulting role involved extensive project management, business integration, and participation in mergers and acquisitions. Notably, I worked with St. Vincent’s Hospital, which expanded from a single institution to about seven or eight. Unfortunately, it eventually faced bankruptcy. This experience provided deep insights into management during financial crises and organizational turnarounds.​

This was my first exposure to healthcare—a sector dedicated to care and wellness but often challenged by reimbursement issues and technological limitations. After St. Vincent’s closure, I was recruited by Northwell Health to work in their application space. I joined during the inception of their revenue cycle implementation, marking the beginning of their patient journey and EHR initiatives. Since then, I’ve been with Northwell and have never looked back.​

Q. Oh yeah, yeah. So you spent many years at Northwell and that’s a fabulous, you know, segue from your industrial engineering and engineering background into healthcare. You know, and you got a glimpse up close of a, of an institution that, that went through that phase as well of M& A and, you know, turn around.

So Sophy, tell us about your new role. I understand that you have a new role here at Northwell and that you are looking at, uh, some very different initiatives. So how’s that coming along and, and. What can you share with the audience? 

Sophy: Yes. So as we started the digital transformation journey at Northwell, laying out the foundation of simplifying the ecosystem and selecting strategic functions that we wanted to implement relative to sort of the cloud foundation, the data platform and evolving our EHR for the entire organization, there was a need that we looked into that we wanted to focus on for digital transformation integration and business relations. And what does that mean? If you look at the heart of where technology and a seat at the table, along with operations and clinical care and patient experience has married, the new discipline of digital relationship management really is trying to make sure that we are bridging operations to technology to patient at the center.

Right. And some of those objectives, no matter whether you’re looking at an adoption of technology that you just are implementing, like an EHR or your new emerges and acquisitions of how are you going to integrate the business  looking at building out a new facility? And what are you thinking about relative to innovative workflows and engagement and experience for, for the patient and the families, right?

And so you look to bridge in that relationship, strategic alignment, you want to make sure that the stakeholders are engaged in the decision making and the journey. And you want to facilitate effective communications because everybody doesn’t always speak the same language at the same time. It is what’s important to me and what’s relative to my role at that time, right?

You want to get synergy across that and you promote the continuous improvement and selection of what works best to meet the mission and the vision of whatever that initiative that you’re working on. Always focusing on process improvement.  Efficiencies and helping out the users, whether the users persona is the clinician or the operations or, of course, the patient and the family as they’re going through the continuum and last, at least you want to ensure you’re translating all that into something that is executable, um, have a translation, understanding that the implementation is not technology only right that are addressing the problem you’re trying to solve, you’re addressing the vision that you have sort of enabled across the institution, and then you want to measure and look at how you’re doing in that implementation real  often and repeat. Basically, uh, there is constant tweaking and recalibrating because that changes life changes and roles change.

Q. Yeah, of course. And you mentioned Sophy previously, before we started the podcast, about change management and cultural alignment, right? Would you illustrate that with an example from your new role? What works, how does it work, and what works best? You also mentioned something interesting about measuring it—any such measures you actually use in the business enterprise?

Sophy: Yeah, I’ll use a couple of examples. Right now, our main focus is preparing for our EHR implementation. As part of this readiness, while we ensure best practices and enterprise standards in our processes, we also prioritize operational engagement. We work closely with operations to understand their needs, ensuring workflows complement and meet their requirements.

We focus on operational input regarding readiness and translating that understanding across departments. Another key aspect of cultural success is our partnership with perioperative teams. Since our organization has grown through mergers and acquisitions, we don’t have a standardized system across our 21 institutions. This presents challenges, especially in OR throughput.

Think about OR throughput—not just scheduling surgeries, but also coordinating surgeons, care teams, room logistics, pre-op, and post-op processes. We collaborated with operations to understand these challenges and explored technology solutions to optimize data and improve real-time decision-making. This turned out to be a great success. We now have significant measurable outcomes in optimizing OR scheduling and real-time adjustments.

Planning on paper is easy, but managing the complexity of multiple ORs, surgeons, and a triple-digit workforce in real-time is challenging. We successfully balanced workloads, improving clinician and patient experiences while optimizing throughput. While efficiency was a key outcome, we also increased volume and revenue. Most importantly, we brought joy back to clinicians and improved patient scheduling.

Q. That’s a great example, Sophy. Thank you for sharing that. You mentioned that Northwell has grown significantly through mergers and acquisitions, now comprising 21 organizations. Given your involvement in this process, could you tell us more about Northwell’s approach to M&As?

Sophy: Absolutely. Northwell Health is the largest healthcare provider in New York State and one of the largest employers, with nearly 90,000 employees. We have 21 hospitals—11 of which are Magnet-designated—and almost 1,000 outpatient facilities. Our workforce includes about 12,000 physicians and over 19,000 nurses. We continue to grow while investing in community education, research, and outreach.

One of our CEO’s key visions is intentional growth. We don’t expand randomly; we grow in a way that ensures integrated, continuous care for the communities we serve. This approach also applies to our M&A strategy. We prioritize synergies, ensuring that mergers align with our mission and values. We focus on contiguous growth, enhancing referral networks, and leveraging our size and capabilities to improve healthcare access and quality.

Most recently, we’ve been working on a major merger with a health system about a third of our size. While it’s not yet approved, it would be one of our largest acquisitions.

Q. That’s awesome. So, Sophy, you mentioned earlier in our conversation that there is no standard playbook for M&As and that success comes from finding complementary technology strategies. Could you share your thoughts on how mergers and acquisitions are approached at Northwell and what makes for some success stories? 

Sophy:  Yeah. I hear this a lot from our partners and customers—Northwell’s approach to mergers and acquisitions isn’t a forced takeover. It’s a very collaborative, synergetic process.

The key to our approach is thoughtfulness—whether it’s a partnership or an M&A, we prioritize cultural and value alignment. If those two things are in place, everything else becomes much easier. When you have similar mindsets and ethics, you’re already aligned on what you’re trying to achieve.

From there, we move into the business aspects. Every M&A is different because it depends on the type of deal, the initial objectives, and the long-term goals. Ultimately, the goal is to optimize and integrate the best of both worlds while learning from each other.

Our playbook starts with integrating people first—creating a sense of unity, where we operate as one team with one purpose. Next, we align on strategic initiatives, assess key risks, and map out long-term goals. We ask: What are your priorities? What challenges need to be addressed? How can we help each other move forward? These discussions help us build a roadmap for business integration.

From there, technology plays a critical role in accelerating and supporting these objectives. If there’s a low-complexity, high-impact opportunity that addresses a pressing challenge, we double down on it. That way, we can create immediate value while laying the foundation for future innovation and growth.

At the same time, risk mitigation is a top priority. We need to ensure that our infrastructure is solid—cybersecurity, resiliency, regulatory compliance—these are non-negotiables. They form the foundation that allows us to focus on investments that drive meaningful impact for both organizations.

Q: That’s great to know. And that leads us to the innovation and AI aspect of things. Sophy, no podcast would be complete without touching on artificial intelligence, augmented intelligence—any type of intelligence in today’s conversations! Looking into the future, especially regarding EHR adoption and the broader integration of AI in business, what are some things you see coming our way, and what might you be working on?

Sophy: Yeah, I want to emphasize again—and maybe it’s my bias now—that any successful AI initiative should always start with defining the problem you’re trying to solve and involving the operators in that process. We can easily get caught up in the tech itself, especially with the fast pace of innovation, and try to find sponsors and problems to fit the tools.

For AI to be successful and scalable, you really need to start with the operators in mind. And by operators, I mean the people who will be using it day to day. AI isn’t just about generative or machine learning; there’s a broad spectrum of AI that includes RPA and other technologies that we can leverage. But to truly scale it in clinical, administrative, or experience spaces, you have to go to the source.

When I talk about AI, I always remind people that AI has been around for decades. In the last five years, generative AI has taken center stage and is often viewed as the “magic” solution for everything. But we should remember that AI is much broader than just generative AI.

At Northwell, we’ve been exploring AI across multiple areas, including imaging and precision medicine, for years, and we’re now diving into generative AI. We’ve had an internal AI hub in place for about five years, focused on security and data protection, which are our most important assets. This hub allows our workforce to securely access generative AI and leverage it for a variety of use cases, all while protecting sensitive information.

We’ve also put in place a governance structure to better understand the needs of our team, identify high-value use cases, and drive innovation. One of the areas we’ve already made an impact is in administrative tasks, like using generative AI to search for information, saving time when employees need to sift through policies and documents. These small efficiencies can add up significantly across a large organization.

Q: That’s amazing! Giving time back to 90,000 people adds up to a huge impact. There are so many use cases that people don’t even realize. They might not be flashy, but they solve real problems that make a big difference in people’s daily work. As we come toward the end of the podcast, Sophy, I’d like to touch on something you mentioned before we started—the celebration of International Women’s Month.

Sophy: Oh, yes! Thank you, Rohan, for reminding me. I’d love to wrap up today by recognizing and celebrating International Women’s Month. Some people focus on just a day, but I say we should make it a whole month! It’s important to recognize the incredible women and all the support behind the agendas advocating for women. Let’s make sure we support each other, raise the bar together, and make a real impact in the world.

Lastly, thank you all for the work you do, and remember: self-care is just as important. It’s okay to take a compliment and take care of yourself, so you can give your best every day.

We hope you enjoyed this podcast. Subscribe to our podcast series at www.thebigunlock.com and write to us at info@thebigunlock.com   

Disclaimer: This Q&A has been derived from the podcast transcript and has been edited for readability and clarity.

 Q. Hi Soumya, welcome to the Big Unlock Podcast. It is a pleasure to have you here. Thank you very much, Rohit. Appreciate it that you are able to spare time for this. So Soumya, we would like to start with some quick introductions. So I’m Rohit Mahajan, I’m the CEO and Managing Partner at BigView and Demo Consulting.

We’ve been doing this podcast for a while, so it’s pretty popular with our listeners. So I’m sure they’re looking forward to understanding your thought process and, and what advice you have for them. So could you please start with your introduction as well?

Soumya: Thanks Rohit. Very nice to meet you again. I am Dr. Soumya Viswanathan. And I’m the Chief Physician Executive at BayCare Health System in Tampa, Florida.

That’s amazing. So what motivated you, Soumya, into this direction? That is my, one of my questions. And second is, what are the differences that you find between the two, let’s say, geographies or the roles that you had here previously in Massachusetts and now what you’re doing over there?

Soumya: Absolutely. So I heavy hitter on the clinical side. You know, I love clinical. My patients love me the typical clinician, but I started getting roles and opportunities to take on projects on the management side or so called the leadership side. When I took the projects on, I felt very comfortable. Then I was doing them and I started getting more projects because as you deliver on the outcomes and the results for each of the projects, people say, Oh, she’s done this.

So let’s give her more. So it ended up being that I was, uh, first I raised my hand for projects and then I just started getting them and I had some really good mentors. And the mentors trusted me, which is also important. So I circled back to them and said, you know, I like doing this, but I’m not sure do I want to give up clinical or not.

So that was the time when I decided to do MBA and MHCM just to make sure that I do want to go down this path of leadership. And it turned out that I started liking it a lot. I liked it as much as I liked clinical. So it came to a point where I couldn’t do both equally well if I spent 100 percent of my time on both of them.

So it’s 200 percent of my time. So I had to carve out and figure out which direction I wanted to go in. And finally, once I did my MHCM, I realized I really like leadership and I want to give it my best. Put my best foot forward and give it my best shot and see how well I do. And that’s what led me to taking on leadership role.

And Baycare has been pivotal because the health system is on a tremendous growth mode. We are expanding like crazy. So what also helped us in that journey is a lot of the health systems in Northeast North. I should say all the northern states through COVID  they allow their people to work remotely. And as you may have seen in the news, a lot of people gravitated or they move migrated down to southern states for warm weather.

And they could work remotely. So the fastest growing cities ended up being, at that time it was Austin, Texas and Tampa, Florida. So, the population here has grown so much that the system, and BayCare is the largest health system in the West Central Florida. So, and it is rated as one of the best in quality outcomes, patient safety, people provide clinically excellent care.

So, it became imperative that, you know, we are on a growth mode, we need to support the population that has migrated here. And I jokingly say, when people move down and move, migrate to other areas, they don’t bring their doctors with them. So we had to support, you know, we need more doctors here. That’s what helped me make the decision to make this my, my stop where I can contribute to leadership team and administration.

Big differences that I see is at least in Bay Care, I should say that, you know, Bay Care has traditionally come together as a health system with the acquisition of community hospitals. And then we set up a medical group. to move forward in expansion of the medical group. So there hasn’t been a lot of focus on academics, which is heavy in Massachusetts, as you know.

So for me, that was a big difference is okay. We need to support, we need to support our physician pipeline for the next generation with a pipeline for recruitment. So what we have done now is moving forward. I’m doing a huge. focus on building the academic mission for Bay Care Health System. We have not delved into academics, and right now we are on the run for being one of the fastest growing GME programs in the country, almost, with 650 residents who will be graduating by 2029, which means in three years we went from 24 residents to, we’ll be looking at 500 plus residents who are approved by ACGME next year, in the next year.

So it’s a very exciting time for us. I want to say academics is one big piece culturally or from a practice pattern perspective. Massachusetts and Northeast is heavily steeped in value based care and, um, the fee for service mentality is still a little bit present in some of the southern states. So we’re moving towards value and taking on more risk.

So that’s a difference I do see that the risk taking capability of physicians is definitely ramped up significantly over the past few years and at least in the region where I am.

That’s pretty cool. And, and I was noticing, uh, Soumya that, uh, you recently celebrated your 10th anniversary and, uh,

Sowmya: One of our hospitals, one, one of our 16 hospitals. So it’s St. Joseph South celebrated our 10th anniversary. And, uh, that’s one of the newer facilities we had built, which has been really, growth has been tremendous in that region, but Baycare by itself is 25 years old. As a health system.

That’s cool. And also curious, Soumya, that you are the chief physician.

Executive. Executive. What does that mean as in relation to, let’s say, a chief information officer, chief medical information officer, or a CTO? So what kind of roles do you play and where is the overlap or where is the synergy with your other?  Colleagues.

Sowmya: Yeah. So from the chief physician executive role has been evolving through many health systems where traditionally the chief medical officers would lead certain aspects of the health system such as quality, patient safety, you know, dealing with physician enterprise only in the form of peer review and clinical care.

The chief physician executive officer pretty much bridges a little bit more than that So what it does is it it requires us to have a little bit more input into operations A little bit more input into looking at financials and cost accounting and gap, you know So it’s kind of a well rounded Cmo role where you pull into tap into all the other aspects that are required for a health system To be very successful, I want to say.

So we form a very close relationship with every other aspect of our health system. Whether it’s the technology officer, informatics officer, CMIO, or the chief financial officer, the COO, we form a very close dyadic partnership with almost everybody in the organization. The CMIO and the CIO, so we do have an informatics officer and we do have chief medical informatics officer.

Work very closely with them because now that there is so much of enhancement and technology input into almost everything we do, especially for doctors to  raising the electronic health record and everything, you know, we cannot do without that kind of partnership. 

That’s awesome. That was my segway. I was going to ask you that you mentioned new care delivery models and also, you know, about patient engagement and technology now.

So how does it all come together for you? Some of your what are some of your digital, you know, initiatives? that you are, you are, you know, thinking about or you’re happy about that you have implemented? What are some of the things in the future that you’re looking for?

Sowmya: We have quite a few things, you know, electronic health record optimization obviously is one of the key areas we want to continue to focus on.

We have Cerner as our EMR, so we definitely want to make sure there is always room for improvement and how the clinicians are taking on the EMR. positives and the pros and cons within an electronic health record. So optimizing that flow. The second piece that we are working actively on is tele visits and tele health in general.

Tele health and remote patient monitoring is a big focus for us just because I already mentioned the population has grown and the access to care has been amplified several folds. So we need more visits. to be delivered to this community. The third one is AI innovation in all forms. So artificial intelligence, which is the buzzword these days and data driven, you know, a lot of data driven platforms that we strive to make sure that when we make changes that impact any workflow, that it’s backed up and validated by data.

So data driven platforms are very near and dear to us. The other most important one, I think, is the patient engagement. Bayfield is very focused on providing the right care at the right time in the right place. So we follow that mission to the T, where we want to make sure that patients Providers. So physicians allied health professionals as well as nurses are all taken care off in a similar manner.

So patients should have ease of access to care. So all that becomes very critical for us. And so patient engagement tools is another big portion that we’re looking at. And of course, you know, interoperability and cyber security come hand in hand with everything that we just discussed.

Yeah, yeah. So, Swami, you touched on a lot of things. Could you help us understand more in detail or with any examples that you might have of any digital health initiatives that you found, you know, success in the past with, you know, or that you’re moving in the direction of?

Sowmya: So, in terms of just the tools of engagement, we have engaged. So, we do a lot of  We have done a lot of pilots.

We have done a lot of phased approaches to engaging in digital health platforms.  So let me just share a few of them without mentioning specific names. So for example, we have engaged in a telehealth platform that improves the care coordination of patients. So we are working with a platform right now that actually engages patients.

And then they also have the care management or the care coordination support on the back end. So with the tools first, we risk stratify the patients on who is high risk for certain diseases and disease conditions. And then we have the care coordination team actually work with us on identifying the disease conditions.

So there is a patient engagement. There is a chronic disease management component to it. And then there is the engaging the patient through the care coordination team that we have. And then looking at, you know, how the outcomes measure out. So that is one we are actively working on. Then we also have like rehab potential, right?

So when a patient has acute hospitalization and they are discharged to home, for example, let’s say they have a heart attack, post heart attack. Oftentimes, the past would be that cardiac rehab has to, the patient has to come in. To the hospital to have the rehab done. Now we are looking at platforms that actually will engage with the patient at home so they are able to do rehab at home without having to drive around and having to take that long distance trial time for commute to come in and have the rehab option available to them.

The other piece that we are working on is early detection. So, early detection is critical. The speed to detecting certain conditions can actually change the outcome of that patient’s condition. So, how quickly do we identify a stroke in a patient? And is it a stroke mimic or is it a real stroke? We are working with a company that allows us for Accurate and early stroke detection.

That’s one of them We also have a couple of others where we have an eicu  platform Where it’s an electronic icu because we need to monitor the icu in patients, but we cannot have that many intensivists rounding 24 seven. So we have an electronic methodology to identify that. We’re looking at platforms that look at sepsis bundles, you know, so there is a whole host of conditions that they’re looking at, which are digitally oriented, that allows us to look at them.

You said what has gone well and what hasn’t, I will share with you that some of the exciting ones are. The radiology ones, the radiology digital analytics, because the radiology ones help to identify the accuracy is outstanding. It’s like they have done studies on how many AI driven platforms can actually identify incidental findings  versus a radiologist reading it or, you know, how many radiologists do you need to read X number of films? And so we are looking at platforms that look at incidental findings in the lungs, as well as lesions on the heart through echocardiogram.  So there’s a whole host of digital platforms that allow us to do this. One thing that we have to watch out for is the fatigue that can set alarm fatigue. So sometimes you have these platforms and it’s pinging away nonstop because it depends on how you set the platform. If you say that you need to alarm me when this happens and you have the metrics not correct, it will be blowing out of control and the alarm fatigue sets in amongst the end user.

So I want to say that we have had our mixed bag where we tested a few and we had to give up on a few and some of them we are many of them we are moving forward with.

That’s great to know Soumya. So is there a way that you think about innovation in certain areas and how about working with startups? Any thoughts or suggestions around those?

Sowmya: We are very open to the areas of focus are definitely driven by, like I said, the three pronged approach. You know, is this going to help our patients? Is this going to help our physicians in any way? Is this going to help our nursing team in any way? We are working with technology tool that’s actually looking at You probably have heard of this, the doctor’s dictations.

Yeah, it used to be transcription in the past. And now we are looking at a dictation modality that allows them to just have a conversation with a patient  and not have to worry about the dictation. And there is AI methodology that works on it. We have looked at a similar, we are working with another company that’s doing a similar modality for nurse driven, nurse voice is being recognized, that a nurse can walk into a patient’s room and have a conversation and not have to worry about, have I captured everything that the patient told me? So, you know, every aspect of clinical care is being looked at very closely to see, does it make sense for us to invest in this or not?

In terms of, you know, so we definitely look at the feasibility of. The patient, the physicians, the nursing, and if it helps improve the way we deliver care, we look at it more closely. The second thing that we do look at, you know, obviously there is going to be all kinds of macroeconomics that jumps into all this with, you know, funding and what is the cost and all that, but we, we weigh it very carefully.

As far as startups go, the one thing that Baycare is very open to is if there are programs that are going to deliver on improving care and help our trio team, as I mentioned, and they are willing to co develop a product with us. We are willing to look at it. So Baycare is not just going to say, well, I’m going to open up my purse strings and just, you know, invest in all this technology because there is so much out there.

So we want to make it appropriate, available for our team. But at the same time, if people are willing to co develop, we are willing to look into it. That’s fantastic approach.

Yeah, so as we’re getting close to the end of the podcast and this session here, Soumya, I would like to touch upon the subject of chat GPT, Gen AI, and all the new things which are coming our way in terms of agentic AI.

So any thoughts there or anything that you are, you know, kind of. Heading in the direction of.

Sowmya: I think it’s very fascinating. I’m waiting to see what the next generation of even chat GPT and Jenny do  because it’s mind blowing how data can be aggregated. Data can be utilized and how it can effectively help us.

But we also have to be cognizant that, you know, there is always, there could be a downside that we have to watch out for. And unless. It is fine tuned to the point where we feel comfortable, you know, all of us in the digital world feel comfortable that there is no harm, no threat that’s coming out of this.

I think we should forge ahead and move forward because that is the next generation that’s going to come our way, whether we like it or not. And, uh, it’s going to be in every industry, you know, healthcare has traditionally been. I want to say probably at the bottom of the totem pole trying to adapt and adopt technology because we always felt as a doctor, we always felt, well, technology can never overtake doctors and take care of patients, but.

When we started understanding how technology can help us drive better patient care, the adoption became easier. So we have to get to that stage where the chat GPTs of the world have to help us drive better patient care and better outcomes for our patients. Then I think the comfort level will set in.

Rohit: Absolutely. So on that beautiful note, Soumya, thank you so much for this interaction. Really enjoyed it. And, uh, if you would like to add anything, I really appreciate this opportunity. And, uh, so, so much appreciate your time too today.

We hope you enjoyed this podcast. Subscribe to our podcast series at www.thebigunlock.com and write to us at info@thebigunlock.com   

Disclaimer: This Q&A has been derived from the podcast transcript and has been edited for readability and clarity.

 Q. Welcome to the Big Unlock podcast, Mark. As you might be aware, Paddy used to do these, and I’m continuing after that. We’re up to the 150 to 160 range in terms of the number of episodes now. So it’s going pretty well. I’ll do a quick intro for myself, then I’ll ask you to introduce yourself as well. I’m Rohit Mahajan, the Managing Partner and CEO at Damo Consulting and BigRio. Mark, please go ahead and introduce yourself to the audience. 

Mark: Yeah, thank you. I’ve been on once before, probably about three years ago with Paddy. It was a great time, so I appreciate you guys having me back. I’m Mark Weisman. I am the Chief Information Officer and Chief Medical Information Officer at Tidal Health. We’re on the eastern shore of Maryland. We are a two, soon-to-be-three-hospital system, and we take care of a large chunk of the eastern shore. It’s a great place to be. Ocean City, Maryland, is probably the most familiar area in this region. Thanks for having me on. I look forward to talking, particularly about digital engagement, which is one of the areas I’d love to bring forth to our organization and help our patients and clinicians interact.

Q. That’s awesome, Mark. So could you please share with us how you got started? What attracted you to the healthcare industry segment and your role? How has it changed over the years? Please share the story. 

Mark: Yeah, sure. So I started as a doctor. Uh, if you go back to, to, to that beginning, and actually I started as a paramedic and a firefighter.

That was how, okay. When I was much younger and much stronger, I was involved with volunteer fire and rescue, and really loved it. I really loved the medical side of it and then went on to medical school and at some point got involved in in primary care and getting into data and analytics and quality measures.

It’s really where I got my start. I started moving over once I saw how dirty the data was and how much. I got more into the informatics side and eventually led to my role as a CMIO and then the CIO. I love being the CIO and the CMIO. I get to use my medical knowledge and bring the technology knowledge and say, All right, how do we make the technology better?

blend into the background for the clinicians and for the patients as much as possible so that let the people interact. That’s what we want. We want those interactions. We want people who are spending time on the harder level things, let machines do easy, repetitive, boring things so that, that the clinicians can really focus on the, on the important stuff and the patients as well. That’s how I got my start. It was really around data analytics and then more used the technology. 

Q. That is so interesting that you have both perspectives and you’re able to wear both hats in the same organization and add so much value to patient care. So, Mark, what do you think about—like you said—dirty data? That is something that caught my attention. How do you get from dirty data to clean data?

Mark: Yeah, good luck. But mostly, I think the most effective way has been getting operational leaders to look at the data. Until they can touch and feel, see it, and go, “Oh, that particular data point is wrong because we don’t do it that way. We do this other thing on the floor,” and then they realize the measurement is just totally wrong.

Without that operational engagement, I can’t know everything about how every different unit does it differently. And neither can Epic, Cerner, Meditech, or whichever. The data will always be dirty in healthcare until the operational leaders sit down, look at their data, engage with it, use it to make operational decisions, and then challenge it. They need to say, “This just doesn’t feel right. There’s no way our time to seeing the patients is that wrong.”

Well, because of the workflow, it’s not capturing something. And until you do the workflow the way the designers made the program to capture it, the data will be dirty. So healthcare is loaded with dirty data—it’s just the nature of the beast.

Q. Yeah, and I quite agree with you. This is a very interesting perspective. You said that operational people need to look at it. How do you do that at an enterprise scale, Mark, both from the people perspective and the technology perspective? If you can give us some insights into that. 

Mark: Yeah, our HR department authorized me to use tasers because it’s really effective on operational leaders. It’s the only thing that gets them to look at their data.

It is a challenge. One of the things we use is Epic, and the Epic Slicer Dicer is their self-service analytics tool. It’s really easy to use. Now, people are terrified of it at first. They’re intimidated. But once you get them to sit down and play with it a little bit, a portion—maybe 5, 10 percent—of the operational leaders will say, “Okay, I can use this tool and I can explore the data, start to ask questions, and then sit with an analyst and say, ‘Show me where this came from.'”

Then we start to realize, “Oh, how we’re capturing it—that little flow sheet row isn’t doing what we thought it should do.” The only way it works is if the senior leaders ask for data. If they ask data-driven questions, their directors and managers will go out and try to look at the data.

Otherwise, they do what they do today and don’t make data-driven decisions. It has to come from the top, with interested people who are willing to explore and experiment with data, and then you’ll start to see the ball move slowly. It’s a journey. There’s nothing quick about this.

Q: Absolutely. And what are your thoughts, Mark, on data governance? How did you put a structure to it, even though you mentioned Epic? How do you put a structure on the data governance piece?

Mark: I don’t want to mislead anyone into thinking that I’ve mastered this by any means. Let’s be realistic here. Yes, we’ve been trying for years to put in place the basics of data governance, particularly operational data governance.

I happen to have an excellent partner in finance. Her name’s Kathy, and she’s wonderful because she’s very data-driven and knows the value of data governance and how the “wild west” happens with people showing data that’s not finance-approved—particularly in finance. It’s all about the data, the numbers, and they’re data-driven. They deal with numbers all day. She’s really been a strong proponent for pushing it forward, and I partner with her.

We’re trying to get definitions for the major projects going on in the organization today. What are you measuring? How do you know if the project is successful? That’s where we start. We ask, “Okay, you’ve got this in your head about what makes it successful, but what’s the definition of that?” And we’re starting to collect these definitions. It’s just one of those things where you just get started—that’s the key piece. Start somewhere.

We then started to figure out how we’re collecting that information. It’s taking too long, so we need data stewards who are closer to the data. The world will start to develop, and it will begin to snowball. We’re still pretty raw at this, but I can see it starting. At least we’re making baby steps.

For so long, we’ve just said anyone who wants data gets data and can present it however they want. We’ve occasionally been bitten by that, and we’re finding that putting some structure in place and finding the champions—those who are passionate about it—is where it starts.

Our population health team is also very good, so we’ve started with some initiatives with them. There are other areas in the organization that aren’t as good, so I’m not starting there. I’ll go back and prove how this works in certain silos, and then we’ll work back.

Q: And Mark, we were talking before about some new Epic tools that are very easy to use and simple in terms of standing them up for patient engagement. Would you like to share some of those experiences?

Mark: There’s a new tool that’s out. Again, we’re an Epic shop, and so anytime I start to play with a new tool, I like to talk about it—especially if my colleagues get interested in these things. This new tool is called MyChartBuilder.

What it enables us to do is build a very simple one-page microsite. It’s a very static page—it might have a link or a phone number in it for people to take action. For instance, we have a new rheumatologist who just joined our organization. Well, maybe people who have rheumatologic diseases would want to know about that. Rheumatology is severely underserved in our area, as in most areas. So we’re lucky to have a new one.

I can now put in MyChart a link that says we have a new rheumatologist, but only present it to those where it makes sense using the medical knowledge we have in Epic. I can say, “Look, knowing that this doctor only sees people over 18, don’t show it to a 17-year-old. Show it to those who are eligible to see this kind of doctor.”

It’s that kind of tool where we can start to use this vast repository of data we have and actually use it to bring good education to patients. I should be showing diabetic education to diabetics, not those with some other disease. And so now we can do that.

What’s nice is that we can do this in IT. We took some of the templates to marketing and branding and said, “Look, I know your logo, we need to use it. This is what it’s going to look like,” and we take that. So when we’re building our microsite, it’s just these templates that the branding center gives us—colors we use. But once that’s set up, now IT can spin up a page in a matter of 20 minutes.

It’s a drag-and-drop website designer. I don’t know if anyone’s ever built a Wix website or one of those, but it’s that simple. These are drag-and-drop and repetitive templates you can reuse. And all of a sudden, you’ve got quick and easy tools to deploy that patients are now engaging with. You get analytics on that using Slicer Dicer—you can see patients are clicking on it when you set it up this way, they’re not clicking on it when you did it another way.

It’s wonderful to see patients engage with some of the tools that were early. This has been out for probably six months now, and we just picked it up in the last month or so. We’re starting to put these in place, and we’re just experimenting with a handful of them. But I can see this being really valuable in terms of things like lung cancer screening. We’ll just put that in front of those who qualify for the screening and not put the information in front of those who don’t. Stuff like that is going to be really powerful.

Q: That’s good to know. So, which are some of the other areas of either patient care or technology at the intersection, Mark, that you’re super excited about? Are there any new digital initiatives that you’re thinking about or seeing coming down the pipeline?

Mark: I think most CIOs are now focusing on virtual nursing and how to scale that. Most of us have done the pilots now. It’s a good thing. The nurses do seem to be responding well to it.

The technology’s in the room. They’re not bothered by it—the “big brother’s watching me” thing. No, it’s here. These are tools that are here to help you. And then the AI we can start to bring in.

Now that we have cameras and microphones going into the rooms, we can bring AI to detect when a patient’s moving out of bed. Right now, we have humans staring at screens, like little Hollywood Square screens—18 screens that they’re just staring at, saying, “Okay, is this patient moving?” It’s a very repetitive, boring job. Computers do that well.

Let’s let the computer do that work and then set off an alert. Let someone know this patient has moved out of a threshold of what we said was safe, and bring the human into the loop to correct the situation or go to the bedside.

This kind of technology is possible now because we’re really investing in virtual nursing. And, oh, by the way, there’s this AI that’s going to come with it, which I’m looking forward to working with these AI models and starting to deploy. So, I think virtual nursing, and all the benefits that come from that, is a great area that most CIOs are now starting to play in.

Q: So, any other aspects of Epic or any other tools that you’re using, Mark, in your health system that are driven by new technologies, like generative AI? Any thoughts on that?

Mark: We all talk about generative AI and some of the use cases—some are panning out, some maybe not so much. But what we’ve been discussing in our organization is the education needed about large language models—how they work, where they can be good, and where they may not be great. We also need to be careful because they could lead a doctor astray.

We have some doctors who are very seasoned and experienced, and I think they’re wonderful candidates to use a large language model because they’ll know when it’s wrong. Then, we have some new trainees who are willing and eager to use AI, but I’m not sure we trust them yet to know enough about medicine to spot potential issues. It takes intuition and experience to know when something doesn’t feel right.

So, the sweet spot for using AI seems to be with doctors who are willing, eager, and engaged to use it—but usually not the most seasoned physicians. My junior physicians are ready to try it, but they might not be quite ready yet. We’re watching and learning as we go, but we’re definitely exposing them to it and teaching them about it.

I think that’s essential for where we are now. As we start putting microphones in the exam rooms and recording conversations with patients, we can input that data into large language models to help with things like differential diagnosis or determining the next best cost-effective test to order. That’s coming—but not quite yet. I could take a transcript today, copy and paste it into a Microsoft Copilot (that’s what we have available), and ask questions about the data. That’s interesting, and in the future, it’ll happen automatically.

I could ask the large language model to develop a Python program for me, but I’m not that good at it myself. I don’t have the subject matter expertise to know if what it’s producing will launch a rocket or serve me breakfast. So, you should have some domain expertise when you’re working with a large language model. It doesn’t replace that, not yet.

Q: That’s true. Which other areas of interest, Mark, on the innovation side? How do you innovate in a health system like yours? How do you encourage innovation with physicians, nurses, or other staff? How do you bring in third-party startups into your ecosystem? Any thoughts on that or advice for fostering innovation?

Mark: It’s challenging. I’d consider us an academic community hospital system. We have residents, medical students, and fellows, but we’re not typically working with biotech startups. If we’re bringing in new tech, it’s because we think it solves a problem, and we see value in it. Sometimes, it’s from a startup, and there’s definitely room for that.

I surveyed my IT team—CIOs, I encourage you to do this. Ask your teams, “How many failures are you allowed to have this year?” Some will say zero (and those are my network guys!). I love them, but yes, correct. There should be zero failures on the network. But there should also be areas where we’re okay with experimenting—even in a network. What if we try something new, like a software-defined wide-area network instead of traditional MPLS? What if we try it in a small clinic? There won’t be any major fallout if it fails, but we get some experience.

Risk tolerance varies across teams. Most IT people are on the conservative side, favoring no failures, because they see a failure as a waste of time and money. But I try to push the team to say, “Let’s try this. It might not work, and I’ve had failures working with startups, but that’s okay.” One tool we tried didn’t work for us, but it will be successful someday. We had invested time and energy into it, but finance wasn’t comfortable with it. We moved on, and that’s the job of a CIO—taking losses and moving forward.

Advice: For other CIOs, assess your team’s risk tolerance. If your team says, “We can’t fail,” you’ll have trouble being innovative. On the other hand, if you have people willing to try, innovating while handling daily operations is hard. You must carve out time, protect it, and give it a safe space. In a community health setting, even with an academic background, it’s still tough to get daily operations to accept new pilots.

Q: I understand. So, Mark, as we’re coming towards the end of this podcast, what do you see in the next 1-3 years for your healthcare setting?

Mark: I see movement toward the cloud, and it should accelerate. Our data center has some weaknesses that would be expensive to fix. If I were to build another data center, AWS or Azure might do it better than we could, actually. So, we’re exploring that. I think we’re middle of the pack in this regard, but there are cutting-edge examples like Dr. Shafiq Robinson, who has gone all-in on AWS and done great things with Epic, or Sentara with all-Azure. Our teams are starting to get on board mentally—they see the benefits of cloud adoption, though it’s a new skill set for us.

That’s exciting. The computing power we gain will let us use data in new ways. And if we wanted to use our own large language model, I don’t know if we will in three years, but we should at least be positioning ourselves to participate in the game. Otherwise, we’ll be left behind.

Q: That makes sense. Thank you, Mark. This was wonderful. I really appreciate your thoughts, advice, and suggestions. We’ll have another session.

Mark: It’s always great to be on the show. I enjoyed the conversation—it was easy-flowing, and time flew by. If you ever need me back, just let me know. People can reach me on LinkedIn, that’s the best way.

We hope you enjoyed this podcast. Subscribe to our podcast series at www.thebigunlock.com and write to us at info@thebigunlock.com   

Disclaimer: This Q&A has been derived from the podcast transcript and has been edited for readability and clarity.

Q: Welcome to The Big Unlock podcast. It is great to have you as our guest. We are looking forward to an exciting discussion to share with our audience.

I’m Rohit Mahajan, CEO and Managing Partner for Damo Consulting and BigRio. As you may know, we’ve done over 150 of these episodes, and I believe this is our 152nd episode of The Big Unlock podcast. With that said, would you like to introduce yourself?

Jeremy: Absolutely. First of all, Rohit, thank you very much for the opportunity to speak on The Big Unlock podcast and to all the listeners out there—I appreciate it.

My name is Jeremy Cauwels, and I serve as the Chief Medical Officer for Sanford Health, based in the Upper Midwest. I have been in this role for roughly the last five or six years. Before that, I worked in direct patient care as a hospitalist at the Sanford USD Medical Center in Sioux Falls, South Dakota, where I still currently live.

At Sanford, we provide care for roughly 425,000 lives from a health plan standpoint and 2.4 million lives from a broader patient population perspective. We cover a large geographical area, stretching from Wyoming in the west to Wisconsin and even the Upper Peninsula of Michigan in the east. Our largest medical centers are primarily located in South Dakota and North Dakota, both affiliated with the universities in those states.

Recently, we also expanded through the acquisition of the Marshfield Clinic, adding additional clinic sites in and around Marshfield, Wisconsin. This has been an excellent addition to our system.

Sanford Health operates 56 hospitals, over 270 clinics, and 144 senior care communities. Currently, we have 4,500 physicians and advanced practice providers.

For perspective, I used to say that we covered a landmass the size of Texas. Now, we are even larger than Texas.

Q: Wow, that’s amazing. And in the rural setting too, Jeremy, so that makes it even more challenging in terms of being able to reach the patients and serve them in such a setting, right? Where you’re covering such rural areas.

Jeremy: Absolutely. If you think about the Dakotas, western Minnesota, and northern Wisconsin, all of those areas are known for relatively small towns and communities—lots of open space between people, generally farming or ranching communities.

So, it is not unusual for us to deal every day with patients who may have to drive an hour or two hours to see their doctor, and potentially even farther to see a specialist.

Q: That’s amazing. So, Jeremy, please tell us more about how you ended up on the technology side of things. You started, as you said, as a physician and were practicing. What attracted or motivated you to this side of the house?

Jeremy: For me, it was really about how we reach patients. When I stepped into this role, I wanted to quickly understand what technology could do to help us overcome the longstanding challenges of healthcare access in the Midwest.

What we know is that many of the counties we live and work in are considered healthcare deserts. Even more of them are mental healthcare deserts. It’s impossible for many families to take a full day off work to drive to the doctor’s office, have a 15-minute appointment, and then drive back home—especially for single-parent families or those where making ends meet every day is the primary concern.

What got me into this was the idea that we could provide virtual care to patients and meet them where they live. Rather than the old phrase, “Your doctor will see you now,” we like to think of it as, “The patient will see us now.”

Q: That’s beautiful. That’s a great flip. So, talking about all things digital—you mentioned telehealth and telemedicine as well. What are some of the digital initiatives you’ve worked on in the past, and what are you looking at for the future? Jeremy, can you share how this digital front door or digital approach is likely to increase access for patients?

Jeremy: Absolutely. Thankfully, Sanford has been blessed with philanthropic support dollars that have been earmarked for improving our digital health footprint. With that, we’ve been able to expand. We actually just opened our virtual care center, a standalone building outside of our hospitals focused on using technology to better reach our patients.

Inside, you’ll find everything from 3D-printed models that physicians can use to understand anatomy before they have to work with it firsthand, to virtual connections that allow physicians to see patients remotely. More importantly, as an educational organization, we have nearly 350 residents and fellows at any given time. So, having the ability to teach them “webside manner” instead of bedside manner—getting them comfortable interacting through a screen, just like you and I are talking today—is critically important.

We also have an innovation hub, a safe space where we can test digital tools in ways we wouldn’t be able to during an actual patient encounter. Alongside that, we house a significant portion of our clinical operations for virtual care in the same building. That way, if something isn’t working as it should, IT resources are right there to ensure smoother operations. They’ve done a wonderful job of putting all of these resources at our fingertips in one location.

Q: That’s awesome. And, Jeremy, in any podcast these days, it’s impossible to avoid the topic of AI and GenAI. So, of course, I have to bring it up—what are your thoughts on AI? Have you been using any traditional AI, and now more generative AI, in your innovation initiatives?

Jeremy: Absolutely. One of the things we’ve focused on with AI is making physicians more efficient in finding the information they already have access to—if only they had unlimited time to pore over the electronic medical record.

For example, we’ve published research with the American College of Gastroenterology showing that our AI can identify 85 different factors that affect an individual’s risk for colon cancer. It can literally score you and explain why my risk for colon cancer might be different from yours.

The upper Midwest has a higher risk for colon cancer, and it’s also an area where screening rates are lower than they should be. So, we’ve been pushing forward, gathering the evidence base we need to roll out a prospective study that proves this AI model works. The goal is to help physicians by giving them real-time insights—when they look at a patient, the screen next to them might say, “Jeremy is at higher risk, and we may need to be more aggressive with screening.”

Another challenge we’ve tackled is the shift in screening guidelines. A couple of years ago, the U.S. Preventive Services Task Force lowered the recommended screening age for colon cancer from 50 to 45. In the Dakotas and the upper Midwest, that change meant that overnight, 100,000 people who didn’t need screening yesterday suddenly did today.

So, we had to think beyond just colonoscopies. How do we reach people who may not be able to come in? One way is through stool-based testing that can be done at home. And as you can imagine, whether someone lives in Howard, South Dakota—a town of 848 people—or anywhere else, they’re used to getting Amazon deliveries. They know how that works. So, if we can put a box at their door that helps them stay on top of their healthcare, we’re moving them forward.

We’ve worked hard to bring all these programs together—not just for patients at normal risk but also to identify who’s at slightly higher risk versus significantly higher risk. Because of who we are and the extensive data we have on our patients, we can personalize care in a way that wasn’t possible before.

Q: I see. Just a curious question coming to mind, Jeremy—do you see a very diverse patient population speaking many different languages, or is it pretty homogenous? How does it look when you do population health analytics?

Jeremy:  Interestingly enough, we have more diversity than people expect in the Upper Midwest.

The largest ethnically diverse group is Native Americans. In some of our regions, the three counties surrounding some of our hospitals are entirely Native American land. In South Dakota alone, three of the poorest counties in the United States are on Native American land. So, those areas present unique challenges that we have to work through.

I’ll also say that one of the communities we regularly serve speaks 57 different languages because a large portion of that town and county is supported by the meatpacking industry. We have several areas like that, where different ethnic pockets of people absolutely need to be served and cared for—regardless of their ethnic background or the language they speak.

The key question is: What is the best way to reach them? How do we find the approach that works for them?

Q: And again, a curious question—one of our clients is one of the largest telemedicine companies. We saw a huge spike during COVID in telemedicine visits, up to 8,000%. But now, people seem to be going back to normal, and COVID is starting to be forgotten.

How has your experience been with telemedicine visits, given the geography and rural nature of your area?

Jeremy: So far, we’ve saved our patients roughly 33 million miles of driving. To put that in perspective, that’s about the closest distance Mars gets to Earth during its rotation.

In addition to that, we’ve now done over 800,000 virtual visits. Twenty percent of our mental health visits every month are still virtual—one in five patients still choose to see their mental health provider virtually.

Some of that may be because they simply can’t find a mental health provider in their area. But for others, it’s about privacy. I drive a blue Ford pickup, and in some small towns, there are folks who wouldn’t want their blue Ford pickup seen parked outside a psychologist or psychiatrist’s office. Stigma still exists, and giving people the option to see their doctor in a way that makes the most sense for them is extraordinarily helpful.

Q: That’s great to know. One other thought I wanted to discuss with you, Jeremy—there are three aspects I’m trying to think about together: What are patients looking for? What are caregivers looking for? And what are clinicians looking for in some of these digital approaches you’re taking?

Jeremy: Absolutely. So, from the patient’s perspective—you can see over my right shoulder, I have three children. They’re much younger, and I’ll tell you, my kids haven’t gone to an in-person medical visit in the last year, except when they needed a sports physical.

If they feel ill, they set up a virtual visit. If they have a follow-up with a specialist, they do it virtually. What’s great is that it allows flexibility for both the specialist and the patient. My kids have even had medical visits while eating breakfast at the kitchen table. That concept would be unheard of to my parents, who would much rather drive to the doctor’s office than try to interact over the phone.

So, from a patient’s standpoint, there’s definitely a shift happening.

From a clinician’s perspective, we’ve been lucky enough to implement AI-driven automated dictation technology with ambient listening. One of our busiest family practice doctors recently switched to it—he simply sets his phone between himself and the patient, and he no longer has to look away at the computer screen.

Not only are his notes completed by the time he finishes his scheduled day at 5 PM, but he told me just the other day: “I finished my day at five, my notes were done by 5:15, I made it home for dinner with my family—which almost never happens—and I even had enough extra time to take my daughter to buy her first used car.”

That’s amazing. Ambient listening is definitely proving to be one of the top use cases for generative AI. I completely agree—it’s only going to grow from here.

People always talk about the “good old days,” whenever that was for them. But many doctors remember a time when they didn’t have to be attached to a keyboard, constantly documenting. This brings us back to that—where the most important thing is the interaction with the patient.

Whether it’s through a screen like this or one-on-one in a room, it’s a much more satisfying discussion for both the doctor and the patient.

Q: Yeah, that’s amazing, Jeremy. What are some of the digital tools or new approaches you’re using for taking care of patients at home, including remote patient monitoring and home-based care?

Jeremy: Yeah, so one of the things we rolled out in 2023 was a text-based patient monitoring tool.

We can use it for common diagnoses like diabetes, heart failure, and depression. We also use it for patients who have been discharged from the hospital to ensure they receive rapid follow-up and scheduled interventions. The technology allows patients to provide feedback without having to call, wait in line, or book an appointment.

I’ll share one example. We had a patient who came into one of our emergency rooms for depression. It’s always heartbreaking when someone is in such distress that the ER becomes their only option, but unfortunately, it happens every day across the country.

After discharge, we sent her home with this monitoring tool, and about a week or two later, she responded to a mood scale with a score of one out of ten. That immediately set off an alarm for us. We called her back and found out she was a single mother, overwhelmed by life’s challenges. She had already written a suicide note and was planning to end her life.

Because of this tool, we were able to intervene, get her the help she needed, and months later, she and her young family are continuing to grow and thrive.

Q: That’s an amazing story. As we come to the end of our conversation, Jeremy, any other thoughts you’d like to share with the audience? Any news or new initiatives you’re focusing on this year?

Jeremy: Sure. I’d like to expand on what I just mentioned about mental health. Many places across the country are what I’d call “mental health deserts.”

In our region, about 95% of the Dakotas—95% of our footprint—is in a mental health desert. That means there are limited providers and significant challenges in getting timely psychiatric care.

To address this, in addition to the virtual visits I mentioned earlier, we’ve implemented virtual psychiatry consults in our emergency rooms. If someone needs to see a psychiatrist and they’re already in an ER—which is never an ideal situation—they often have to wait until morning, or worse, wait for an open psychiatric bed, which could take days.

Now, instead of waiting, we can connect them with a licensed psychiatrist virtually, right there in the ER, even if the psychiatrist isn’t in the same city or state. That immediate access makes a huge difference in getting people the right care when they need it most.

Q: That’s amazing, Jeremy. It’s incredible that technology enables us to do this now. Any closing remarks?

Jeremy: I just want to say thank you. Thank you for allowing me to share a little bit about Sanford and how we continue to reach out to our patients. Despite the distances, we’re doing everything we can to bring them closer to the care they need.

We hope you enjoyed this podcast. Subscribe to our podcast series at www.thebigunlock.com and write to us at info@thebigunlock.com   

Disclaimer: This Q&A has been derived from the podcast transcript and has been edited for readability and clarity.

Q: Hi, Ashish, welcome to the Big Unlock podcast. I’m Rohit Mahajan, CEO and Managing Partner at Damo Consulting. It’s fantastic to have you back on the podcast after a long while. I’m sure things have come a long way. Would you like to start with your introduction?

Ashish: Happy to, Rohit. It’s always a pleasure to talk to you and be on the podcast. I’m Ashish Atreja here at UC Davis Health, and I’m also a venture partner at Global Ventures. I have a mission to create clinicians as co-creators and co-founders for mission-driven initiatives in that regard. I’ve been in the role of Chief Innovation Officer in Medicine at Mount Sinai. Before that, I was an informaticist at Cleveland Clinic. I still practice gastroenterology one day a week at UC Davis Health. I worked as a CIO and Chief Digital Health Officer at UC Davis Health, supporting transformations at scale.

Q: That’s awesome, Ashish, that you’re able to pack so much into one day. Could you please share with us what motivated you on this journey? How did it all start, and how have you been navigating so far? Where are you headed? I think you have a very interesting group which is Valid AI, so perhaps talk about that as well.

Ashish: Absolutely, Rohit. I consider this my fifth career life, and I feel very lucky. I started off as a physician and a public health scholar. My first day of residency at Cleveland Clinic, I saw a lung transplant patient. It was the time when lung transplant was just getting into practice, and the books I used to read didn’t mention how to take care of a lung transplant patient.

There was no knowledge in the books because it took about five years for new knowledge to be printed in books. This was around 1999-2000. Someone told me to look at UpToDate, which was the first good online textbook. I managed my first patient learning from that online textbook. After successfully managing the patient, a spark idea came to me: Wow, it was just a database, but technology made it so easy for knowledge to flow. This was something that was never taught to us in medical school.

I could see, even in 1999-2000 convergence, how much technology could enable us. That led me to pursue informatics and IT. I did an informatics fellowship at Cleveland Clinic, supported inpatient EPIC implementation, and then began my journey in innovation. That was my second life, moving beyond informatics to innovation at Mount Sinai. Then, it was transformation at scale at UC Davis Health. Over the last 20-25 years, I’ve been initially 50 percent and now close to 100 percent in IT, digital health, and AI, with the same mission: How can we leverage technology to make us better physicians, better clinicians, and better scientists?

Q: That’s awesome. You did mention AI, Ashish, and you have a very big initiative around AI with Valid AI. Could you explain what motivated you, what it’s all about, and where it is heading?

Ashish: Absolutely, I’d love to. Earlier, when I was working in digital health, we helped establish one of the first digital health labs in the country at Mount Sinai. We felt the need to have an evidence-based way of looking at digital health applications. At that time, I started Node Health, a network of digital evidence in health—a nonprofit that brought health systems together to independently evaluate digital health applications. I felt the same need in the age of AI, especially generative AI. There are so many solutions coming down the pike—so many algorithms, so many products. How do we even evaluate them? How do we integrate them into our systems? How do we create value from them? Within our health system, we can still work on these solutions, but we’re basically repeating each other’s efforts. If I’m evaluating a solution, another UC is likely doing the same thing somewhere else. We simply don’t have the bandwidth in our health system to duplicate all this work.

Valid AI was created with the goal of bringing people together—to learn together, work together, and create efficiency. This program was endorsed by the UC Office of the President, and we officially launched it at UC Health about a year and two months ago.

We initially had most states in the U.S. represented, with around 38 to 40 health systems and health plans joining the network. We meet biweekly to discuss a range of topics—from organizational maturity and governance to identifying high-impact use cases and necessary monitoring. The goal is to help systems learn from each other and create an efficient approach to transformation.

As we enter year two, VALID stands for Vision, Alignment, and Learning—those were our focus areas in the first year. This year, we’re adding Implementation and Dissemination of Science as our next set of goals. As we receive feedback and grow, the network has expanded to 54 organizations. We’re now thinking seriously about how to implement AI at scale—not just for VALID AI founding partners, but beyond, to extend its benefits to a broader audience. We also want to make AI more affordable and accessible to everyone. To support that, we’re building shared teams designed to empower the entire healthcare ecosystem. It’s a very exciting and important next step for us.

Q: That’s awesome to know. I never knew the full form of VALID like you just described—thanks for explaining it so clearly, Ashish. It’s truly an ocean out there, as you mentioned. Tens, if not hundreds, of health systems are on this AI journey. With such a large landscape, how do you drive meaningful transformation? And what are some of the capabilities you’re working to build around that?

Ashish: Great question. I look at the external landscape and then the internal landscape within organization. I believe there’s enormous potential to create efficiency across the healthcare stream. To achieve that, we need more than isolated efforts—we need a network of organizations working together. So, I look very closely, I’m a part of the working group for Gen AI for CHAI, Coalition of Healthcare AI. So, we need to look at FDA and the governance from the federal government, but then we need something like CHAI to kind of create frameworks. And, and, you know, guidance back in that regard, but then we also need something like train, which I’m part of as well from UC Davis Health, which looks at the train network, looks at how do we kind of really distill things to more infrastructure level and kind of monitor that in a safe manner where VALID AI positions is more on impact.

How do we follow the guidelines from CHAI and the federal government, but how do we actually bring it to reduce to practice is what we call rubber meets the road and really create an impact. So, it really has a return on investment for us in that regard. So, that’s the learning part. But when we look at how do we really create an impact within our organization, moving from, and I had a great discussion with some of the folks and I was saying, all these external network organizations actually are interdependent and support each other.

For example, if we implement something and give feedback to CHAI to, Hey, your open model card really is working and this is how we’re using it. That supports CHAI vision as well.  we came with the concept of this Olympic rings, you know, seven Olympic rings or five Olympic rings, which are interconnected.

Yeah. So, we need interconnected different organizations outside to support us from that. But when we look at the landscape for internal to an organization, let’s say UC Davis Health as example, where I’ve been leading the charge on transformation, digital health and AI for the last four years, we need to look first, look at our AI strategy or our overall health system strategy.

Is our focus being in a quaternary hospital? Do we want to expand geographically through care at home? What are the major things we as an organization want to achieve? Because that major strategy then governs how we put AI into it, right? And so, it starts with a strategy-first approach, and then we distill from strategy to governance, which means.

How do we bring AI, which all approvals we have to need, we need cybersecurity team to look at it. We need approvals from ethics and equity compliance kind of committees, whatever structures are there. We need to look at someone from procurement, how they procure in that regard, whether it’s in UCS. We need to have some governance structure, which not only helps us move this along, but also the governance structure also helps us prioritize. Because there are so many possibilities right now, we can and we should not in any way bring all the AI technologies. We cannot and should not do that. We need to look at, out of all those technologies, which is completely aligned with our strategy, that meets our gap. So strategic lens, and then using the strategy lens, prioritize the top five or ten. Then those top five or ten then go through the governance and the implementation and the procurement process.

Once the procurement process happens, then we call it AI adoption roadmap needs to come in, where we have project manager, project coordinators, integration experts, after the procurement, actually putting it together. Then the go-live happens. And then we also need to do evaluation at scale and monitoring at scale.

We need to measure that the algorithm doesn’t degrade. We also need to know that we are creating an outcome which we want. Many times we are guilty of taking a new technology, just implementing it and not measuring its impact and not monitoring it. So I think there’s a new—we call it implementation science—layer that has to be built in.

We published a paper in New England Journal of Medicine AI about the need for creating AI implementation science centers in health systems. Very interesting and right, and I’m a gastroenterologist, so I often crack the joke: you know, the food is there, but our stomach has to be empty enough and there has to be hunger so we can, ingest the food.  Once we ingest the food, then we can digest the food. Right. So we need to increase our capacity as health systems to be able to ingest the right kind of AI solutions, and then to be able to digest and create the maximum value from the solutions.

Q: That’s awesome. So, Ashish, since you have been involved with AI and now GenAI for some time, could you narrate any use cases that you might have—being involved with yourselves or people that you know have done successfully? What are some of the things on the landscape that are making it into the health system? And what are some of the future directions that you see over there as well?

Ashish: Yeah, happy to. So, when we started doing research work with Valid AI, we are looking at every single publication, preprint or major journal publication that’s coming for used case. So, we’re setting up a use case library of 165 used cases. And the domains we are finding are either on the employee and physician productivity. So, we are seeing like AI scribe is perfect in that space on physician productivity, but we also seeing emergence of tools which will cater to all employees of the healthcare worker, not just the clinical workforce. And we are kind of looking at those possibilities.

Then we’re also looking at things at the patient automation side. Can we bring generative AI solutions which can guide the patients wherever they are? Because we have an access problem. And it’s not easy just to pick up a phone and talk to anyone in the health system. But can some of the routine work be done by generative AI solutions and guide the patients automated or semi-automated with humans in the loop sometime. Then we also seeing operational efficiency solutions like RCM and those which can allow better enhanced capture. So, we’re seeing kind of a three-legged kind of approach to that.

And we are using a PRISM kind of a method from Gartner, which looks at feasibility as well as time. And we are putting a next layer—is impact. So, you look at the lens from whatever is feasible today and the time to onboard that is not much, and it creates a maximum impact that can help you prioritize the most relevant used cases. In fact, because it’s not about bringing AI, it’s about solving what problem. Yes. And then you find the relevant things, AI or non-AI, to be able to do that. So, part of us is also educating our executives and leaders to start with a design thinking approach or problem-first approach, a strategy-first approach.

And that’s something our phase two of Valid AI looks like, where we create kind of a way to align everyone around strategy and problem-first approach and the most meaningful, impactful use cases that come. And then we are able to measure the impact. So, few cases that we already have implemented. One is AI scribe, and I was talking two years ago, it’s going to be like water. Everyone is going to be using AI scribe. Yes. Because—and this is the first time, Rohit, I have seen the chairman kind of arguing with each other who should go first. Well, technology typically has a productivity paradox when it comes to healthcare. Every industry has benefited with productivity except us in healthcare. But I do feel finally the time is here when we unlock the productivity paradox with technology for healthcare. We generate AI, and AI scribe is a perfect example of that.

So, we implemented in pilot phase and we just got ravishing reviews, including physicians just writing that this is the best technology they have seen till date. And this is a physician who’s about to retire.  We didn’t have to train them.

We’ve also brought technologies in vision AI. Something like Aidoc, where we were able to find that more PEs can be detected at an earlier time. So, we fast-tracked that implementation as a platform. So, we have seven sectors in AI from a strategic lens.

So we’ve been able to bring applications from risk prediction, like sepsis algorithms built in LA, then we fine-tuned that to things which can predict how the person is going to go worse based on alerts coming from wearable sensors, to vision AI—computer vision technologies, to technologies like generative AI—from AI scribe, but also then looking at and understanding generative AI. Can we unleash it to all the employees, bring the policies and other together to make every employee more powerful and Impactful.

Q: That’s awesome. So, Ashish, it was really nice meeting you at JPM, a few days ago, actually in sunny San Francisco this time. And, I was at your launch, you know, would love to learn more about what you talked about at the launch on the venture partnership and the kind of startups that are coming through the Venture Studio. Could you share some more information about that?

Ashish: Yes, Rohit. Happy to. I think throughout this informatics and innovation journey, I was lucky to be part of a spin-out from Mount Sinai, which took our app prescribing platform. We were the first ones where doctors could finally prescribe apps, in addition to medicine and surgery, and that got patented and led to a new company launching a startup. I was very grateful to be on the board and see the VC-backed company finally get acquired last year after touching 30 million lives directly and 60 million through UnitedHealthcare. So, in total, 90 million lives touched.

As I looked back on the journey, especially in the age of AI, I feel we can have a better way of partnering or creating startups in partnership with health systems and health plans, maybe embedding our soul. I look at startups as raising a kid. You have to put your heart and soul into it. Initially, their capacity is they need to learn how to poop, they need to wear diapers. But then, they build their capabilities, and that takes time. You need to be patient during that process, nurture them.

So part of it was, especially with generative AI, how can we create an ecosystem where we learn as healthcare organizations how to better partner with startups? Maybe those startups are co-created by different health systems and health plans, that led to the concept of GVX Global Venture Accelerator, where we are creating soulful startups, but in partnership with healthcare organizations. So they are not created outside by someone and then trying to disrupt you inside. They are actually from the very beginning supporting inside-out transformation, right? Because their soul is actually from clinicians as co-founders—a health system-led initiative.

I was very happy to see that launched at JPM. We also launched, along with that, our first startup, called Fusion Care, which is planning to support health systems, health plans, and individual practices in obesity centers for excellence. With the GLP-1, I do feel the biggest trends we have seen in the last few years for us have been, one is obesity and metabolic syndrome, which I, as a gastroenterologist, never felt could be treated by a physician, except for pediatric surgery. Now, this entire, the biggest disease we have of modern civilization, can now be treated with medicine.

But there are many more things that have to come together to treat it, including a whole care team of nutritionists, psychologists, a whole digital plan, which automates the patient follow-up, and putting them into the right reimbursement model. So, Fusion Care is launching and enabling a kind of practice enablement for obesity centers of excellence and metabolic centers of excellence.

And then we are working from a valid AI perspective, how do we bring other things together to fruition of shared services in general in that regard.

Q: That’s awesome. I was there at the panel discussions, and listening to the FusionCare CEO was fantastic. How they’re intending to make this change from inside out, as you’re describing, is the right way to do it, Ashish.

So, Ashish, as we come up on the end of the podcast, are there any other plans for the future or any new things that you are working on that you would like to share with the audience?

Ashish: This is good. Thank you. I have devoted, I think, for the next phase to, as I was supporting transformation at scale at UC Davis, really taking some time off to support the same journey for multiple health systems, health plans, and pharma together. Bringing things together, kind of expanding from the Valid AI experience into that.

I feel this will be fascinating because, can we, through a Venture Studio or Venture Accelerator, over the next 10 years, touch 1 billion lives? That’s the goal which I have put. We touched 30 million, 60 million lives in the past one. But how great would that be? As a physician, I would not be able to meaningfully touch more than 10,000 patients in the traditional care I can deliver. But through technology as an enabler, we have the potential to touch masses if we do the things right. Very  large footprint.

So, I’m looking for very meaningful partnerships with health plans, health systems, and life sciences to co-create entities together. So we don’t have to follow someone else leading and disrupting us and creating all those barriers. We can evolve ourselves, kind of a true metamorphosis that happens from inside out. So that’s what we are on the journey for, and I’d love to hear from others, get them as partners, and together we can shape this next phase of our exciting journey.

We hope you enjoyed this podcast. Subscribe to our podcast series at www.thebigunlock.com and write to us at info@thebigunlock.com   

Disclaimer: This Q&A has been derived from the podcast transcript and has been edited for readability and clarity.

Q: David, welcome to the 150th episode of The Big Unlock Podcast. Padmanabhan had done a lot of the previous episodes of this podcast and I’m carrying on his legacy.

So, David, would you take a few minutes and like to introduce yourself to the audience and talk about yourself and the health system as well where you have been working? 

David: Yeah, sure. Thanks a lot for hosting me. It’s an honor to be here with you this afternoon.

I’m David Berger. I started out as an academic surgeon and surgical oncologist. I did my general surgery training at SUNY Downstate and a fellowship in surgical oncology at MD Anderson Cancer Center in Houston, Texas. I initially thought I would become an academic surgeon—maybe a surgical oncology division chief, chair of surgery, or even a dean someday—but my career trajectory shifted due to unique opportunities.

Early in my career, I undertook a management role as the Chief of the Operative Care Line at the Houston VA, the Michael E. DeBakey Medical Center. It’s the largest, most complex, and busiest VA medical system in the country. Around that time, as part of the sandwich generation, I was also taking care of both my children and my parents. I realized how difficult it was to navigate the healthcare system. If it was challenging for me—someone who knows the system and has contacts—I couldn’t imagine how hard it must be for those without the same resources.

This experience shaped my career goal: to make healthcare simpler. The healthcare system is extremely complex, and I saw the potential of digital tools to transform care delivery, making it more accessible and effective. While at the VA, I implemented sophisticated digital health tools. One notable project involved real-time location systems (RTLS) for managing our operating rooms. We reduced turnover time between cases by nearly 50% and achieved a 95% on-time first case start rate—remarkable for any healthcare system, not just the VA.

Building on this success, I moved to the private sector as Chief Operating Officer at Baylor St. Luke’s Medical Center. There, we implemented machine learning tools to predict which patients were at risk of sepsis-related decompensation. By intervening early, we reduced sepsis mortality by almost 50%. This was back in 2014, well before the recent wave of AI publicity, but it reinforced my belief that digital health, AI, and machine learning are poised to revolutionize healthcare.

After Baylor St. Luke’s, I returned to SUNY Downstate, where I had completed my training, to serve as the Chief Executive Officer of University Hospital at Downstate. I started in September 2020 and recently left to focus on my passion: digital health and artificial intelligence in healthcare.

Now, I advise early-stage startups, participate in advisory boards and venture studios, and am about to become an LP in early-stage venture funds supporting digital health companies. That’s a little about me and my journey.

Q: That’s really great to know, David. It’s fantastic that physicians like yourself are taking on this journey because your deep domain knowledge can drive meaningful change in digital health.

I also learned that you’ve been an entrepreneur. Could you share a bit about your entrepreneurial journey and the kinds of software or digital solutions you worked with?

David:  So, about 10–12 years ago, I was approached by an early-stage startup company based in Canada. They had come across an article we published on early warning signs for readmission after colon cancer surgery.

We had identified at least 10 early warning signs and developed a machine learning algorithm to predict who was at risk of being readmitted. This software company wanted to integrate our findings into their tool. We conducted a couple of successful pilots and started discussing a long-term relationship. The idea was that we, as clinicians and researchers, would provide the intellectual property to power their software. 

But we thought they were undervaluing what we were bringing to the table. So we actually went to Baylor College of Medicine, and they spun us off as a company, as our own company, from Baylor Technologies.

I learned a lot from trying to start my first company. Not all the learnings were good, but they were important.

What I learned is, one, if you’re going to start a startup, you have to devote all your time to it. You can’t do it on the side. I was heavily involved as a clinician and an administrator, and I was trying to do this on the side.

Second, you must have alignment of your board. We had five board members—me, two from Baylor, and two from the company that was helping us on the IT side. Well, the people on the IT side and the people at Baylor College of Medicine were diametrically opposed on the direction, and we got stuck and couldn’t move forward.

So after about running through a million bucks, the company died. But what’s interesting is that there are a couple of startups just getting going with the same idea we proposed and were trying to move forward 10 years ago.

So, the idea was the right one, but the alignment was poor, and I wasn’t committing full-time to it. Those are really important lessons for anyone trying to start. 

Q: Great lessons learned there. Yeah. And then, David, you were talking about how AI is, you know, going to be changing the landscape, and you have a certain way of looking at it, especially for healthcare systems. So we’d love to learn your thoughts about that—about AI, GenAI, and where you think it is going, and how one should, you know, kind of embrace it and adopt it to get more value. 

David:  So I am extremely excited about the potential for artificial intelligence to help us provide healthcare because there are so many repetitive processes in healthcare that are being done by people that probably can be done by technology.

When I think about a health system—and, you know, I was running a hospital—about the different places where artificial intelligence could have a major impact, I break it down really into four different areas. This is general; there’s overlap.

So the first one is obviously back-office functions—things like revenue cycle, coding, billing, documentation, pre-authorization. All those things that have to do with revenue cycle.

Then I look at the front end or access—giving patients the opportunity to interact with a provider, giving them the opportunity to schedule, how they’re going to pay for their healthcare, et cetera.

And then on the care delivery side, there are things that help make care easier to deliver, and then there are things that actually help provide care.

So let’s break down those two things. When I say make it easier to deliver care, I’m talking about things like artificial intelligence scribes or ambient listening, ambient dictation.

When I talk about things that help actually deliver care, I’m looking at adjuncts to help read X-rays. So one use case that’s fairly well-developed is the issue around mammography. An important fact and something that radiologists look at in mammograms is breast density.

Breast density sometimes predicts further cancer down the road. Radiologists are really bad at quantitating breast density, and there’s a total lack of correlation from one radiologist to another. But using artificial intelligence, you can improve the ability to determine breast density to almost 95 percent accuracy.

So that is a place where artificial intelligence could augment care and improve how care is delivered and the diagnostic accuracy. That’s what I mean when I say help facilitate care versus actually help deliver care.

Q: Understood. And David, in particular, just for a moment to segue into all the new generative AI technologies that are coming our way. It’s not just about text; it’s also about so many different forms of care and media that it can transform. So any thoughts on where you think generative AI might play a larger role or be more conducive in the healthcare setting?

David: Yeah, so in terms of generative AI, one of the solutions we implemented is something called AVO. And what AVO does is ambient listening—it helps physicians to write the note. But that’s just part of it; that’s not really the full generative AI component.

What it then does is pull relevant information from the electronic medical record to create a real functioning note. It puts in front of the clinician a differential diagnosis, as well as the clinical pathways related to that diagnosis. So yes, it’s ambient listening, which is important, but it’s also natural language processing. It helps to generate feedback, a differential diagnosis, and then the proper treatment pathway for that patient.

It also presents the note in a way that facilitates coding, billing, etc., for the hospital or system to get reimbursed. That has been a game changer for our patients.

Thinking about implementation, though, the implementation of any new tool is really, really challenging. What we found with the ambient listening and ambient dictation is that there was a different level of adoption between specialists like surgeons and primary care doctors. Why do you think that is? I’ll throw it back at you.

Q: I couldn’t even guess that. First of all, I’m thinking to myself, is it the primary care physicians who adopt it more, or is it the—yeah?

David: It’s the specialists and the surgeons who adopt it more readily. For them, a note is something they have to do; it’s not really what they want to do. They want to focus on their operation, their procedure, etc. So, for them, the note is more of a barrier to doing what they really want to do.

For a primary care doctor, one of their primary goals is to create a detailed note. So, for them, the nuances around creating the note are really important. Whereas for a surgeon, they don’t care as much if it’s perfect—they just want to get it done. So, the adoption was much quicker with surgeons than with primary care physicians.

We also implemented it in the emergency room, which is a tight space with a lot going on. Some of the ambient listening got distracted because of other noises in the environment. We had to figure out ways to ensure that, during the interaction between the clinician and the patient, there was as little ambient noise as possible. But we worked through that, and the adoption was incredible.

People have talked about AI replacing physicians. Five years ago, someone came out and said that within five years, there wouldn’t be any need for radiologists. Well, clearly, that hasn’t happened. But I see AI being used as an adjunct to clinicians—helping them to be better at what they do, not just in terms of efficiency, but also in improving accuracy in care and diagnoses.

I think clinicians are going to become partners with artificial intelligence. Another area where I think AI, generative AI, augmented reality, and virtual reality are going to have a huge role is in education.

Even in teaching within the operating room. Right now, the way you teach in the operating room is by working with a resident across from you. Over time, you give them more and more experience and freedom to conduct the procedure. But think about this: if you have really good augmented reality, you could have them work with a headset in an environment outside of the operating room to hone their skills in a space where there’s no risk to patients.

So, there are amazing opportunities for augmented reality and virtual reality, not just in education, but also in surgery. For instance, I used to do a lot of liver surgery. The liver is a three-dimensional structure, but the inputs—like where the vessels are or where the tumor is—come from 2D images. These are from CT scans, MRIs, or ultrasounds.

Then, once you’re in the operating room, you’re trying to reconstruct that 2D image in your mind into a 3D picture, relating it to the organ in front of you. It’s challenging, but you get better at it with experience. Now, with augmented reality, you can actually see, right in front of you, where the tumor is in relation to other structures within the organ.

This has been really well-developed in neurosurgery, but it’s an exciting opportunity for AI, virtual reality, and augmented reality in other types of surgery as well.

Q: Yeah, great example, David. So, my last question is about your new venture or the new role you’re going to be taking on soon, as you described earlier in the podcast. Which aspect are you more excited about? Because you’ve seen the startup side of things, and now you’ve been in a large health system. Where do you think you’ll be spending more of your time and adding value?

David: So, I think healthcare is slow to change from the inside. A number of the bigger systems have developed innovation labs and are trying to move the field forward, but I think a lot of the change is going to come from outside—from companies outside of healthcare.

What I’ve decided to do is move away from running a healthcare system or hospital to being more involved in the digital health and AI ecosystems. I think the way I can best contribute is as an advisor to early-stage companies within this space, but also as an advisor to investors and venture companies looking at which technologies are actually going to make a difference.

Since I have the perspective of both a clinician and an operator within the health system, I think I bring a unique lens to identifying which digital health solutions are really going to make an impact.

Q: Awesome opportunity, David. I’m wishing you all the best in this role? And then would you have any other closing thoughts for our audience? 

David: I think we are at an inflection point, between technology and the desperate needs within the health system. The costs have risen dramatically, people do not have adequate access and there are huge problems within our health system. I think we have opportunity right now to make a difference. 

To move healthcare forward and to improve the healthcare system within this country, I encourage people who are thinking about careers within healthcare to think about the innovation side and the digital health side, because that is really going to have a huge effect and move the needle. 

We hope you enjoyed this podcast. Subscribe to our podcast series at www.thebigunlock.com and write to us at info@thebigunlock.com   

Disclaimer: This Q&A has been derived from the podcast transcript and has been edited for readability and clarity.