From India to Innovation (ft. Suchi Saria)

Thanks for tuning to Digital Voices podcast, where we chat digital transformation challenges and opportunities across healthcare and life sciences. And now, your host, Ed Marx. Hey everyone, welcome to another edition of Digital Voices. Thank you for listening. I know you have a lot of choices. There's great, great content out there, and you've chosen to spend a few minutes with us, and we're greatly appreciative, and we'll make it worth your while because today I have Professor and Dr. Suchi Saria. Suchi, welcome to Digital Voices.

Thank you for having me, Ed. Excited.

Yeah, this is great because, you know, you're known as Professor AI. You're a founder. You have this amazing eclectic background, which we're going to dive into, and you're just a good person. That's the best part.

Thank you.

Yeah, thrilled to have you. And we first met, I think we just kind of knew each other via LinkedIn and obviously healthcare tech and things like that. And then we had a chance to get to know one another a lot more this past fall, and we spent some time together. And so I was like, man, I have to have you on Digital Voices. And so, Suchi, you know that the very first question I have for you is, what are the songs on your playlist? What kind of music do you like to listen to?

Oh my gosh. So my— I have a 2-year-old. I am, uh, I didn't know I was gonna enjoy having a kid as much as I'm loving it. And his His favorite new song is this Australian, like, kids' rhyme called Kookaburra. And so we listen to Kookaburra a lot.

Okay, that's a good one because we do have a Spotify playlist also called Digital Voices, and we'll have to add Kookaburra to that playlist. So if you're listening to this playlist in the car or something, your son's with you, and he'll be like, wow, that was— I influenced the world right there.

That's cool.

What about life message or mantra? Are there words that kind of guide you or you live by?

Always give it your best.

And did you sort of grow up with that, or is that something that as an adult you sort of honed in on?

Yeah, I grew up with it. You know, my mother is one of those incredible human beings where, you know, as I've aged, I only keep looking back and thinking, holy majoli, what a superpower and a superwoman she is. She just aces whatever she touches. She grew up in a family where she was the oldest. She took care of all her younger kids. She aced school. She has aced being a mom. It's just incredible. And she teaches meditation. There are lots of people, and she just does it because it's a calling for her, for free. So it's not even—

Wow.

And it's interesting because there are so many people in the community who look to her. You know, they've had difficult life-turning moments where they've come to her. So I'm just so incredibly grateful that I got to learn from her, and she just gives everything her best. Even now, when she's visiting me, With my with my son, you know when she plays with him, she is fully present. Yeah, and the amount of energy she brings to the table—I'm looking at her and thinking like I'm not—I can't match that. So it's just so cool, and it's just such an inspiration.

Yeah, that's very cool. Yeah, tell us a little bit more. Like where did you grow up? Where were you born?

Oh, I was—I grew up in India in this town called—it's the foothills of Darjeeling. So you know you might have. Drank Darjeeling tea. My family has been in tea for a long time. My grandparents, my parents, in particular my dad is a tea fanatic. He thinks, eats, breathes, drinks, exports, manufactures, creates, designs all things tea. And so it was just incredible. It's like the Napa of tea in India, if you will, Darjeeling. And again, growing up, I actually didn't even like tea. 'Cause I was, you know, it's almost like when you have too much of something, you don't fully appreciate it. And I was actually the first person in my family who went into computer science, machine learning, AI, robotics, tech, decided to go do a PhD, became a professor. There were just a lot of firsts, which, you know, they were in business and I was sort of this very odd bird, if you will, doing the things I did.

Oh, that's fun. How did you make your way to the United States?

You know, the way this happened was I got really into AI and robotics quite young, which was also to me surprising in retrospect that it happened. And that took me to this very science tech feeder school in India, which happened to have the only robotics club. I was part of that. And a lot of schools from here would come recruit. And once I started realizing, like, I was very excited about these topics that you just couldn't study in India. And the ability to bring together different disciplines and create really energized me. So I got a full scholarship to come here. And at 17, I broke the news in a letter to my parents. They were shell-shocked, to say the least. They still have the letter in their drawer. Because basically I needed to present them both, not only with the plan that I was— I think I was the first girl from my town to have left. So there was a lot of like, is this safe? Is this reasonable? It wasn't a very common thing to do when I did it back then at my age. And I presented them with a plan that also said how I was going to make sure I covered all the expenses and was only asking them for some minimal money in the beginning to start. So they felt like I had a full plan to, you know, to go along with it. And now they tell me they were very proud to have received that letter. I can tell you at the time, it didn't feel that way.

I can imagine. That's a great, great story. And I want to talk a little bit about your career. Then we'll dive into what you're doing now with Bayesian Health. So what was the catalyst to go into STEM? Was this just a natural thing? Obviously, it didn't, you know, your parents, your dad in particular was into tea and things like that. Was there something that happened in your youth or how did it come about?

So I grew up in a very massive joint family. I'm very close to all my extended cousins, etc. I was also one of the younger ones in the family. So what that meant was I definitely grew up in an environment where there was girls can't do this, girls can't do that. You're a little one, you can't do X and you can't do Y. And I was a bit of a rebel. And so to me, it started by first, I loved creating things. I loved building things. I loved tinkering with those. So those attributes were always there as a kid. But to me, there was also an element of like, this is hard. And I was very motivated by proving to them that I could just do about anything that needed to happen and doing hard things. And so that's really how I got into it. I thought it was fun and exciting, but I was also determined because I was gonna prove to them how brilliant I was and how I was very good at doing the hard things that they thought I couldn't do. Well ahead of schedule. Yeah, that's fascinating.

That's a story in and of itself. So you went from jumping into STEM, being sort of counter your local culture and being one of the first to do these sort of things. And then you became a researcher, a professor, an investor, and today a founder. So how do you split your time with all these different roles that you have? And mom, of course.

You know, what's been actually fascinating is I used to really believe in this idea of like what I call work-life harmony. Like if all the things just go together and they make sense, it's actually like work doesn't feel like work to me, it's play. It brings me so much joy and flow. So for instance, the way I ended up, you know, as I did AI and robotics research over the years and at Hopkins where a lot of my group's work was focused on advancing the technology to get it to work for the kinds of difficult data we're seeing in healthcare, which literally, quite literally, when I first started working in this area back in 2008, I think there was like crickets. Like, you know, I remember I would be invited to give talks and big meetings in AI and people loved the idea. They thought it was so cool that like as machine learning AI researchers and technologists, we could make an impact on human health. But in terms of listening to the idea versus doing something about it, right? There's a big gap because we needed to understand health systems deeply, the data deeply. There's just so much that needs to be unpacked.

Yeah.

So a lot of my research at Hopkins started focusing on discovering and closing those gaps. And it was just amazing to be working with some of the best multidisciplinary group of researchers from machine learning AI to human factors engineering and implementation science and clinical medicine, regulatory, et cetera. And then naturally that led to, because of my own personal journey too and interests, and I unfortunately lost my nephew to sepsis. Sepsis happens to be an area where I had been doing a ton of research, wrote probably one of the first papers in the field back in 2013, 2014, '15, saying how AI could help you with early detection in sepsis. And most people, I think there was zero people who understood both the word sepsis and AI at the time. And then I lost my nephew to sepsis, which was very sad for me, for my family. And it sort of made me realize like we had done a huge amount of the work in the field, which now writ large we recognize, right? Like the field's come a long way. There's been a lot of progress. There's a lot we can do with it. And it naturally felt like there was an opportunity to scale and translate this, not just in sepsis, but hundreds of other areas. And that's kind of what led to Bayesian. I've always been very motivated by bringing research to life at scale. So not research for the sake of research, but finding really, really hard human problems that we can solve by using advancing technology to close gaps, but then also figuring out not just solving it, but the scaling of it. And I've experienced that from a very young age. I've been part of other startups as I was doing research even while back at Stanford. And so it all felt very natural. It didn't feel like I had so many different lives. It felt like all these lives go together. The research helps advance the translational work we're doing. The translational work informs some of the policy work we're doing. And, you know, and the investor is just natural in this. I mean, that I have to say, there are just so many opportunities where AI can make such a big difference.

Yeah, no, that's a great natural segue because I want to hear more about the company that you have founded. And you shared the origin story, you know, about the tragedy with your nephew and with sepsis. And so you've taken that on as a mission. Can you share some examples? I know that you work with some of my former colleagues at Cleveland Clinic, at Nova, other places. Can you share some examples about, for those who haven't heard of Bayesian, exactly what it does?

Yeah. So today, health systems are, you know, they are sitting on all these gobs of data, right? They're like data-rich, but insight-poor. And when I say insight-poor, I mean, at what level? I mean, from a frontline clinician perspective, right? Like, Our patients are more complex. They're burnt out because, you know, like our nursing colleagues will say, we are given like 24 hours of work to do in a 12-hour shift.

Yeah.

Right? There's too much work to do. This is true across the board. You know, we need to set up systems to augment them in a way that they can be maximally effective to get done what they need to get done, to make the right thing the easy thing to do. And we're already seeing some early results from AI, right? Like it started with something as simple as scribing. where you're listening, you're transcribing, you're summarizing a note. But that's only the beginning, right? So to me, think of that as the sensing layer. The sensing layer is listening, biobuttons, where we've now implemented ways to collect vitals or cameras that are collecting information. So there's, in the last 3 to 4 years, health systems have started implementing and making a lot of investments in what I call the sense layer. But then there's the think and act layer.

Yeah.

Ultimately, all this data is coming together. That's the data problem. How is this getting converted, this multimodal data getting converted into something that you can, in real time, reason, augment clinicians with these insights, tie them to streamlined workflows that help them get done what they need to get done and complete what they need to do? So just bringing it back to my nephew, sepsis, leading cause of hospital deaths, impacts their finances in a very meaningful way. You know, it's interesting where there's a system on the West Coast, very large system. They said they have 20, like huge numbers, 20 million in denials tied to sepsis claims getting rejected because of documentation. It's one lever. Care variation in terms of length of stay. Of course, not the most important reason why I got into it, right? The fact that like, if the patient, if only we had recognized it earlier, you could have given the right treatments. Yeah. This patient would've lived. this person would've lived. And I get stories like that all the time. Like, I literally got a story a few days ago on LinkedIn where someone read about our work and wrote to me saying they took their mother to this hospital in Florida. She was an ED walk-in. She wasn't feeling well. They kept her in the ED for several hours. Ultimately, what ended up happening is she ended up getting septic shock. This was just like a casual walk-in. She was not recognized early enough. She sat in the ED for way too long. And a week after she passed away. And she and her father were devastated and she had learned and she wrote to me saying, can we bring this work to this hospital? So as a technologist, just sort of hearing these stories of how much impact we can be having, and Sepsis is only one such example. Since I started doing this work back in 2015, what has been incredible is learning with our customers, right? Cleveland Clinic, with— Rochester, with Inova, they're saying there are 150 such problems we could tackle that every day they wrestle with, where if we could figure out a way to put these new streamlined workflows in place that allows them to both catch, act, treat, document, report, comply, boom, you know, it would save them a huge amount of time and make doing the right thing easy.

No, I love it. And that's a perfect value proposition because the heart of it for you, and you can hear it in your voice and the heart for me too, is, you know, saving people's lives. So definitely save people's lives. But for, for the realities that some have to deal with in terms of cost, it actually pays. These systems pay for themselves. Are they pretty examples?

Wait, actually, I want to say a little bit about the way more than pay for themselves. Like, yeah, there's a 2,000-bed, like, collection of hospitals put together, 2,000-bed system where we're going live based on the early data we're seeing. We will have saved 130 lives this year. Isn't that nuts? Yeah. This kind of thing, it never gets old. It makes me so grateful that I get to do the work I do.

No, I love it. And for those of our colleagues, you know, and I understand the struggle, so there's no judgment, but they're like, I want to do stuff with AI. I want to leverage modern technology, but, you know, we don't have any money. We're downsizing. But in fact, investments like these not only do the right thing, save people's lives, improve quality of care, make the clinician's experience much better as well, patient, family, hits all, everything, including the financials. So these systems like this pay for themselves. So that it's just like the perfect, you know, it hits the quintuple aim, if you will. So that's great. What's one thing that you're most proud of to date? I know there's many things and you already cited some amazing statistics. But is there any other one thing that stands out, like, for your team or for the impact that you're all having?

I mean, obviously, like, the lives we're able to save, that is just, you know, it's just the most sacred thing we can do. Incredibly, incredibly proud of it. The second thing I feel very proud of is, like, you know, every time I get to work with great people, I feel very grateful that they're choosing to work on this and choosing to work with me. Because the only way to make very, very hard things happen is not alone. It's with other great people who are also committing their time, their energy, their life to doing it with you. And, you know, you know, nothing in healthcare is easy. So first of all, AI is not easy. Provider systems, care team-facing software, none of this stuff is easy. So the fact that people are spending time to do this thing because it matters is just like, I always get so much joy and pride when I see them succeed and feel privileged that they're working with me.

Yeah. No, that's well said. And it's not surprising after hearing about your mom in particular, you can tell they're great people. And so the heart in which you live, you know, is a reflection of your parents and, you know, the missional type work that you're doing. So I'm thankful for you and everything that That, that you do. So, Suchi, we'll have listeners right now that are like, wow, this person's amazing. So let's jump a little bit into the leadership side. But before we get there, you're also involved with the World Economic Forum. I found that kind of interesting. So tell us about that and then we'll unpack leadership a little bit.

So at the World Economic Forum, I mean, I got started getting involved with them or got roped into getting involved with them almost 10 years ago as a professor, as a researcher. They had a whole council focused, like a task force focused on AI that started almost about 10 years ago. So I was a subject matter expert that was part of that alongside 10-ish other experts or 12 other experts from around the world. Several of them people now recognize and know who are more in core tech and less in healthcare specifically. And that was so fascinating to hear. And it was a lot about sort of where is AI going and policies and so on and so forth. Of course, extremely relevant now. It felt even extremely relevant back then, where we were thinking about it very broadly. And then, I've been involved in a number of efforts through that, and they recognized me as a— I don't know how, but got nominated and got recognized as a Young Global Leader, which is a 40s Under 40s program they have, through which you get a lot of opportunities and exposure.

Yeah, that's fabulous and a great testimony to your— again, back to your heart and your capabilities, your skills. Did anything early in your life sort of prepare you? I mean, you're an amazing leader. Was there anything unintentional or intentional that happened that helped make you the leader you are today?

Oh my gosh. I used to never think of myself as a leader. I think I just really love doing things, and I love solving hard things, and I like partnering with people to make it happen. So, I think my walk into leadership was a very incidental sort of thing. You know, when I became a professor at Hopkins, I found myself with a lab where I needed to lead lots of people. And I often saw myself as just working with them to make new inventions.

Yeah.

And then as I've run Bayesian, it's just been such a humbling experience to realize like, oh my gosh, this is a craft in its own right. It is actually feels, you know, physics, you like, there were lots of hard science fields, but this feels I'd say even harder, nailing what it means to be a great leader.

Yeah. Yeah. There's no simple formula and you sort of have to find your way, but there's definitely a few fundamentals like you've been talking about. So as professors, what would you say to the graduating class? So let's just say you were selected to give one of the speeches at the next graduation. What are maybe one or two things along the lines of leadership that you might share?

Funny you ask this. I feel like I want to give this one more thought. I wish I had done some prep. It just so happens that I gave a speech, convocation speech, last year. And so the thing I told that class was, whenever you feel you've tried really, really hard and you're about to give up, take a deep breath, go drink a drink, do what you need to do to make yourself feel better, and then try again 3 more times.

Yeah.

Because in my experience, what has been incredible, learning from my parents, watching them. You can do really, really hard things. The main thing where things fall apart is like you almost always, that happens because you gave up just a tiny bit too early.

Yeah. Yeah, I like that rule of 3. So at least try it a few times. Don't give up too early. But you also have a limit, like, okay, I've tried 3 times.

Well, I think the question is, okay, so then if I had a little bit more time, you want to understand why you you're failing. The why is very important because when you understand the why, that helps you understand what do you need to close the gap. Now, if the why is because the idea makes no sense, then you shouldn't do it.

Right.

If the why is because there was something you didn't know how to do, then you need to figure out, can you learn it? Can you find somebody else who could do it? Right? So the why is so key, but sometimes people fail, give up because they're frustrated. And that frustration, like, clouds judgment in terms of not wanting to understand the why. You're not going deep enough to understand why. I kid you not, Ed. I remember back in 2008, 2009, when I first started doing this kind of work while I was back at Stanford, when I was like, this makes total sense. The future is, you know, they were just implementing Epic and Cerner. So the idea of implementing these kinds of systems off top was completely bananas. And so when I, when we were discussing ideas like this, which we happened to get funded by really smart people. I remember at the time, Stanford has this great, wonderful biodesign program, and there were colleagues and peers who, as part of the feedback, would say, this idea is never gonna work. There's no reimbursement for it. Well, okay, maybe there isn't reimbursement for it today, but there could be reimbursement for it tomorrow. Then they're like, the idea won't work. The data is so dirty. Oh, I see, because you're coming from medicine, but AI is good at working with this kind of data. This is exactly the kind of— things that AI systems are very good at versus rule-based systems. So I felt like each time I understood the why, that is what allowed me to keep going and for us to get here. And there were so many nos along the way, and each time it was understanding why. So another one was hospitals would never pay for it. Yes, but the reality is in many of these problems we're solving, if you go down to the underpinnings for it, there is a very sound financial reason In addition to the mission. So if you can make that case, people love it. So, so I think it's always, let's get to the bottom of the why.

Yeah. And that's a super great insight. What about recharging your batteries or remaining fresh? So obviously you do a lot of different things, plus, you know, your personal side of your persona, who you are. How do you take time for yourself and sort of refresh and recharge?

I love— my husband is an introvert, which is really funny. So for him, his downtime is hanging out with himself. For me, my downtime is hanging out with him. So I love just finding a glass of wine and hanging out with him. That's probably my number one way to spend time. And then of course, now that my kid Sammy's been born, it's really funny because he's such a ball of energy and I just love hearing him babble. You know, I was saying all these whys and people tell me, you know, your kid is about to enter that age where they're going to be asking you why for everything. So, we'll see how that goes. But probably those two are my favorite ways to spend time.

Yeah, no, that's great. This has been an amazing conversation. Like I said from the outset, I've known you for some time now and you're an amazing individual. I knew that we'd get a lot of great insights learning about some of your favorite music, which comes from your 2-year-old. And then the mantra that you grew up with, you know, always give it your your best, which really correlated nicely also with sort of the words of advice you give to aspiring leaders in terms of, hey, when you feel things are too hard, don't give up. Try at least 3 times. And then each iteration, ask why you failed. We talked about how you grew up, where you grew up in India, how you came to the United States, how you started your career as a pioneer in STEM and tech, and then transitioned into healthcare. Out of that, you created this amazing company, Bayesian Health and helping to save lives around the world. Just doing great things. It's just amazing as I glance at my notes. And then, yeah, we talk a lot about just leadership as well and how you grew into who you are today. What did I miss, or is there anything you want to double down on? I'll give you the last word.

Love. Spread love. It's just such an— it's been such a tricky last several years.

Yes.

Everybody, everybody is overworked, tired, exhausted. So just want to, you know, feel like it's so nice when people come give you a hug, you can get a hug, smile. So last word is be happy, give love, be kind, be generous. And there has never been a better time to be doing what we're doing because there are so many opportunities ahead. If we can only hold hands, think together straight, put one foot in front of the other, and do the right thing.

Suchi, excellent words to leave this podcast with. Thank you so much for being my guest.

Thanks so much for having me, Ed.

Thank you for listening to Digital Voices Podcast with Ed Marx. If you enjoyed this episode, subscribe on your The Verge streaming service and leave a rating and review. And most importantly, thanks again for listening. Thanks for listening.