← Becker’s Healthcare Podcast
Becker’s Healthcare Podcast
Unlocking AI’s Potential in Healthcare with Shrey Kapoor, CEO of MedSetGo
00:10:41
Loading summary
Transcript19 lines
- [00:00]
A
Hello, everyone. This is Jacob Emerson with the Becker's Healthcare podcast. Today we're on site at Becker's 15th annual hospital meeting in Chicago, where we're going to be discussing how to unlock AI's role in the continued transformation of healthcare. Joining me for today's discussion is Shreya Kapoor, who is the CEO of medset. Go. Shrey. Thank you so much for taking the time to be with me on the podcast today.
- [00:25]
B
Thanks for having me, Jacob. I appreciate you inviting me on.
- [00:28]
A
Yeah, absolutely. So before we dive into everything we want to talk with you about, can you tell us a little bit more about you, yourself, your background in healthcare, and what it is that you all do today at medset? Go.
- [00:37]
B
Yeah, absolutely. So I'm the founder and CEO of medsetgo and we are building the first AI case manager. We've raised around three and a half billion dollars to date for this pretty massive undertaking. But we believe that we can have a really significant impact on how patient care is coordinated when patients are being discharged from the hospital. Me in particular, my background and how I got to this point. I grew up in the San Francisco area, just surrounded by technology my whole life. And then I went to Johns Hopkins where I went for undergrad, medical school and my mba. But along the way, while I was in the hospital, recognized that there are so many different manual processes that happen that really delay patients being placed appropriately after they're being discharged. Patients are sitting in the hospital waiting for things and then they have hospital acquired infections. And ultimately we have no idea what happens when patient leaves the four walls of the hospital. Sure. So I entered medical school during the pandemic. It was raging here in the us. It was worse in India, where my extended family's from. And I lost my grandmother during that time. And it was very, very challenging for me being the only person in my entire family with any medical background. So I was taking a lot of ownership of her career. But what I did after that, I think changed the trajectory of my life. I put together 50 medical schools across the country and we raised over $5 million to send equipment, drugs, et cetera, abroad. We petitioned the Biden government to release vaccines that were stockpiled here in the US not being used the non mRNA. I penned that letter. And lastly, we started a rapid antigen testing program where folks went door to door, tested people and if they tested positive, gave them access to oxygen or steroids, things that were life saving at the time. That's something that I scaled 0 to 1 at 23 years old. Never thought I would do anything better in my life that impacted 30,000 people. That taught me the value of large scale initiatives that I found myself thriving in. And so I realized my place in the world was using technology, specifically automation and AI, to again impact patient care.
- [02:49]
A
Absolutely.
- [02:50]
B
That's what got me here today.
- [02:52]
A
Well, it's an amazing backstory and it's going to be fascinating to hear your take on everything that we're hearing about AI at, at this week's conference Shrey. As you know, we're hearing from health systems this week about their, their rising use of AI. They're continually adopting the, these new technologies into their clinical and administrative workflows. But as you know, it's still early days for many organizations across the country. So what are you hearing at the conference this week? What are you hearing from medset go's perspective the rest of the year? What are the barriers that are really holding health systems back right now in this space? And what's your message to our leaders listening in on how they can begin to really tackle these?
- [03:35]
B
Absolutely. And I think we're at a really exciting time in healthcare where traditionally healthcare as an industry has lagged in innovating and rightfully so, because patient care, data privacy, compliance and security is a very big concern. But what I implore health leaders to understand today is that this is not a question of if it's going to happen, but when. And we need to start thinking about not reasons to say no, but ways to make it work. And I think that's what I'm seeing in the conference today over the past week is everyone is excited about this opportunity and we recognize that if we are all in this together, if all the different stakeholders who are traditionally siloed in healthcare come together and figure out how we can really leverage AI, especially in places that are lowest hanging fruit, we can really 10x or 100x value in healthcare. That's patient quality, that's financial savings, and that's improving efficiency across the landscape.
- [04:39]
A
Yeah, absolutely. But this is, this is a complex issue though. Right? Right, Trey? I mean, training these AI models in the healthcare space, you're deal with sensitive data, you're dealing with often fragmented data. As you said, it's often very siloed. So walk us through how these models are being built and can you give us a practical example? What does this actually look like in action on the ground, in the four walls of the hospital?
- [05:04]
B
Great question. My motto at this point is you can't train smart algorithms on dumb infrastructure. And the first thing we need to do is make sure that we are connecting all of the pieces of the puzzle together from a data infrastructure layer standpoint. First of all, the hospital EMR has certain data and then the patient leaves and the outpatients have a different EMR or they're on the same EMR but on a different instance. The entire patient's story needs to be connected. And that's what we do at medseco is we first come into the hospital system and say, let's analyze all the different stakeholders in your specific local region and connect all those data pieces together first. Once you have the patient story connected, then you can build smarter AI algorithms because you can search, summarize, exchange clinical information at lightning speed. Now with AI between different vendors in the community who traditionally are phone calling, emailing, faxing. And so that's really the value of AI. But I think the first piece, and I think this is something that all need to work together on, is how can we connect the EMRs together? And that's the value proposition of a medset. Go.
- [06:19]
A
Yeah, no, it makes complete sense and it really aligns with what we're hearing from leadership this week in terms of they're all saying that AI is going to be a key driver of their organization's success certainly very soon, probably in the next five years or if not sooner. So what are you seeing outside the conference? What are you seeing across the industry right now? What are the trends you're watching most closely? And then bring that back to what our health system leaders, what should they be exploring now to unlock the long term value of this tech?
- [06:52]
B
I'd say treat AI like a clinical trial, pilot it out, see what works, see what doesn't. Because there is so much opportunity here that I think that is the practical way to go about this. The second piece is I think there's so many different AI companies today. I was walking along the exhibit hall and every company is AI. You know, it's like this, this, this phone, this recording that we're doing is AI. Like this, everything's AI. And so health system leaders, if they think about what are the lowest hanging fruit, I'd say scribing, triaging and discharge planning are probably the easiest way. Where there are so many manual entry points for physicians, for case managers, for nursing, for staff, that this is a place where they can easily change operational workflows and drive significant efficiency in everything that they do in the hospital. And I'll give them a specific example where, and this is around throughput, where there are so many hospitals across this country who have a throughput problem where they can't place a patient, they can't discharge a patient and there's so many more people waiting in the emergency department waiting to get care. And so if you can use AI to predict discharge, you can use AI to connect with the home healths, the SNFs, the post acute partners of the world and track their availability data and be smarter about when patients are being discharged in the post acute space. Then you can figure out the bed availability and you can smartly place patients from your hospital to the post acute. Then more patients can get care. Right. And so that is something that we've already seen as medset go, we've implemented in a hospital system in San Diego. And what we're seeing is not only are we replacing 1 to 1.5 FTEs worth of time at the hospital setting, but also patients are being discharged four hours faster.
- [08:55]
A
Wow.
- [08:56]
B
And so can you can imagine the savings from a cost standpoint. But more importantly, there are patients waiting in the ED who don't get treatment. And now they are. And I think every health system needs to understand how valuable the solution can be if they have that forward thinking how can we make it work Mindset.
- [09:14]
A
Certainly, certainly. Before we go Shrey, I know there's so much we could continue to talk about on this topic, but what else are we missing? Like we've mentioned, we have so many other leaders who are not here in Chicago with us this week. They're listening in from all over the country. What do you want to tell them, what bits of advice you want to offer them as they continue this journey along with a lot of new technology.
- [09:37]
B
I'd say give it a chance. You know, I speak selfishly for, for from an AI case manager standpoint. There are so many hospitals with 33% vacancy rates or for they can't find case managers. And so supplement that with AI. Started as a copilot, it's an assistant. Right. See what it does for your hospital system and you are going to alleviate so much of the efficiency burden on your hospital that it's worth the shot. And so I mean reach out to me. But not even if it isn't medsego that solves this problem, somebody else will. And that's the exciting part of today. So I'm excited about where healthcare is headed.
- [10:16]
A
Absolutely. Well, it's great last bits of advice. I think a great place to leave things. So Shreya, thank you so much for taking the time to be with us and for sharing your insights with our audience. We really appreciate it.
- [10:26]
B
Thank you, Jacob.
- [10:27]
A
I'd also like to thank our podcast sponsor, medset. Go. You can tune into more podcasts from Becker's Healthcare by visiting our podcast page@beckershospitalreview.com.