The sense of purpose and focus that led Klaudyne Hong to do her PhD at MIT in two years helped her turn a chance comment during a fly-fishing trip into a groundbreaking startup. Peach IntelliHealth is building AI for acute medical care. Her project was seeded by the Government of Singapore where she found great help in building her data set. She’s now seeking FDA approval.
- Sal Daher Introduces Klaudyne Hong, PhD, Founder of Peach IntelliHealth
- What Peach IntelliHealth Does
- Peach IntelliHealth’s AI Can Predict a Patient’s Response to Treatment
- The Founding Story of Peach IntelliHealth
- “So, if you don’t remember anything about what I say today, remember the next point. It’s the highlight of the show, okay? I actually caught two striped bass that morning.”
- “And we started then doing a historical analysis, and we realized that our algorithms actually fit that data very nicely.”
- Peach’s AI Got Regulatory Approval in Singapore – Now Replicating Pivotal Study in the U.S.
- “Sepsis, by the way, is the most expensive health care condition, hospitalized condition, in the U.S.”
- “If someone’s going to go into cardiac arrest, and you know it several hours in advance, you can make a huge difference.”
- Current Status of Peach IntelliHealth
- Funding for Peach IntelliHealth – the Government of Singapore Came in as a Seed Investor
- “I grew up in Singapore, and I left when I was relatively young. So, I was as an undergrad in California.”
- Her Favorite Show “Cheers” Showed Boston in Such a Good Light that Klaudyne Chose to Apply to MIT
- “…my thesis…was about controlling gene expression for gene therapy applications.”
- Klaudyne’s Two-Year PhD
- “I drew him a picture of this DNA that was wearing cowboy boots, and somehow told him it’s all about the cowboy boot being a gene therapy vector. And somehow, he understood what I was saying. He said, “Yeah, come join me in the lab.””
- Klaudyne Hong’s Entrepreneurial Journey
- After J&J Klaudyne Consulted but Found It Unrewarding, Lacking Ownership – Enter Entrepreneurship
- Peach Benefitted from the Entrepreneurial Ecosystem in Singapore that Allowed the Rapid Acquisition of Patient Data
- “But I find that the larger the company, sometimes the slower things can be, because I think they call way too many meetings that you have to attend.”
- “They see that you have this passion. And they see that look, if it’s that crazy, why do you have so many good people embarking on this journey with you?”
- “It is the most robustly supported result in the study of entrepreneurship that more co-founders, better chance of success.”
- Where the AI Will Fit in the Clinic
- “I also see AI in a sense of helping make informed choices about the kinds of medicine we take.”
- Sal Talks About Portfolio Company Leuko
- Klaudyne Hong’s Closing Comments
Transcript of, “AI Improving Clinical Options”
GUEST: CEO KLAUDYNE HONG, PHD
Sal Daher Introduces Klaudyne Hong, PhD, Founder of Peach IntelliHealth
Sal Daher: Welcome to Angel Invest Boston, conversations with Boston’s most interesting angels and founders. Our guest today is Klaudyne Hong. Hi, Klaudyne.
Klaudyne Hong: Hi, Sal.
Sal Daher: Great to have you on. Now Klaudyne in the New York, so we’re bending the rules a little bit, but we connected by the graces of my brother-in-law, Martin Aboitiz, and she’s in the business that touches the business he’s in… Anyway it’s very interesting business that she has. Very strong Boston connection. She did her PhD at MIT.
So, Klaudyne, we’re going to talk today about Peach IntelliHealth, which is her company. And so, kick it off, Klaudyne. Tell us what Peach IntelliHealth does… what problem it’s solving, why it’s important.
Klaudyne Hong: Well, first of all, Sal, thank you for inviting me on your show and shout out also to Martin, who introduced me to your show. I’ve been enjoying them, learning a great deal from them, so I’m honored to be here today.
What Peach IntelliHealth Does
So, Peach IntelliHealth, as our name suggests, uses artificial intelligence for the purpose of health care. Our mission at Peach is to improve patient outcomes, while reducing the cost of health care. And the way we do this is by creating new types of AI-based algorithms that can predict what would happen to a patient.
Peach IntelliHealth’s AI Can Predict a Patient’s Response to Treatment
So, I’ll give you an example. When you have a patient, who is in the hospital, for example, in the ICU, intensive care, we can analyze that data in real time, and then let the care team know how this patient will progress in the next one or two days. If they see that the patient will have a positive response to the existing intervention, then you know everything will be okay. You don’t need to make any changes. However, if they see that the patient will have a negative response in 24 hours or 12 hours, then they will have time to modify the existing treatment. So the analogy is like driving on a dark highway, and we don’t know what lies ahead. AI is that tool that tells us if there’s going to be a big pothole that we should avoid or there’s going to be smooth sailing. Similarly, in medicine, if we know something bad would happen to a patient and we have this extended window of knowledge, one or two days before that something bad would happen we can do a lot to prevent it from happening.
Sal Daher: Okay. Why don’t we get things a little bit more concrete, and maybe we can have the founding story of Peach IntelliHealth. I understand that you trained your AI in Singapore. I lived in Singapore for a year and have very-
Klaudyne Hong: Oh wonderful.
Sal Daher: … fond memories of the place, a wonderful place. Tell us how the company came about.
How Peach IntelliHealth Was Founded
Klaudyne Hong: Sure. So let’s start “once upon a time.” So this would take us back a few years to the summer of 2013. I was in Boston on a business trip and usually when I come by Boston, I’ll try to meet up with one of my favorite professors at MIT, and that happens to be Gerry Wogan. He was the head of toxicology, which became the Bioengineering Department. And we’ve known each other since I was a grad student in his department. So we kept in touch after I graduated, and each time I returned, we would usually end up at Toscanini’s ice cream. Big fans of Tosci’s.
So, this time, however… This time, in the summer of 2013, we decided we’re going to do something really different, instead, we’re going to fly fish instead. So it was 4:30 AM. He picked me up from the Tech Hotel, drive about an hour north east to the Essex River, and that’s where we fished the rest of the morning. So it was a beautiful summer day. You could see the sky reflected in the water. The water was so flat that, if a small insect landed on the water, there would be this huge ripple effect of concentric circles going a long way from the little insect.
Sal Daher: Wow.
Klaudyne Hong: Yeah, it was gorgeous. So there’s nobody there, except for Gerry, me and the fly fishing guide and his boat. So I looked at this-
Sal Daher: What fun.
Klaudyne Hong: So, you have the picture this, right? Gerry’s a world-class, top-notch, fly-fishing expert. In total contrast, I never fly fished before. So beautiful day, flat water, and I’m looking at Jerry I said, “Okay. I’m just going to apologize in advance…” It’s like 6:00, 6:30 AM by then. “I know that my clumsy cast is going to scare your fish way.” So Gerry, it’s typical of him, he’d probably just laugh and say something like, “Don’t worry. We’re going to get them. And then, I just started thinking out loud, “I wish that I had an AI to predict where the fish is going to be.” And he said, “Oh Klaudyne, I didn’t know you were interested in AI.” So that one hour drive from the Tech Hotel from the Essex River, we were talking about health care, especially oncology, because that’s his specialty. He’s amazing when it comes to cancer research.
“So, if you don’t remember anything about what I say today, remember the next point. It’s the highlight of the show, okay? I actually caught two striped bass that morning.”
So, we start talking about AI, and I told him what I wanted to do. And then he said, “Oh, do you remember Professor so-and-so in computer science and bioengineering at MIT?” And that’s how Jerry became our Chief Strategy Officer. Two other MIT professors, Bruce Tidor and Collin Stultz, became the CTO and CMO. And we hired another MIT alum, Luigi Vacca, he became Chief Architect in Machine Learning. And Peach was born. So if you don’t remember anything about what I say today, remember the next point. It’s the highlight of the show, okay? I actually caught two striped bass that morning.
Sal Daher: Oh, wow. Oh, stripers are delicious. I love stripers. Stripers fly casting, that’s quite an achievement.
Klaudyne Hong: Beginner’s luck. Just to let you know, Gerry caught over 30, and that’s why he’s the MIT professor.
Sal Daher: My gosh. You have to throw them all back in.
Klaudyne Hong: That was a great time.
Sal Daher: You have to throw them all back in, because there’s a limit. I don’t know. In a river, I don’t know what it’s like, but in the sea around here, you can take one striper. Catch anymore, you have to throw them back in.
Klaudyne Hong: We caught and release. Yeah, Yeah. That was the only way to do it. And everything I’ve done, since then, is caught and release.
Sal Daher: Yeah. When I was told, `”catch and release,” I was, “How about catch and eat? What happened to catch and eat. I want to catch and eat.” “Oh, yeah, you can catch and eat all the bluefish you want.”
Klaudyne Hong: Yeah.
Sal Daher: So, yeah. Why are these fishing trips… Why do you have to always get up at like 5:30 in the morning to go on a fishing trip?
Klaudyne Hong: Yeah. That’s it. That’s it.
Sal Daher: So, anyway-
Klaudyne Hong: It’s always very early. Yeah.
Sal Daher: You got this group together. The idea was to look at a lot of patient records and to try to see if you could predict outcomes based on bunch of variables. And so, how did the connection with Singapore emerge?
Klaudyne Hong: Well, so the story is this. We spent a lot of time, roughly about four plus years, we look at millions and millions of data points. It’s not just this Singapore patient database, we also look at U.S. hospital database. And we develop, refine different types of algorithms using these millions of data points on a supercomputer.
“And we started then doing a historical analysis, and we realized that our algorithms actually fit that data very nicely.”
And the link to Singapore came about in 2015-2016, when we just started doing what we call “initial prototype, initial AI algorithms.” One of their large tertiary hospitals, a 1,250 bed hospital, at National University of Singapore, top-notch place… We were introduced to them, and they asked us what we were doing. We showed them what we’re doing, and we told them about our interest in the intensive care arena, being able to get AI out there, so you can use it and try to reduce the number of people who need to stay there that long and then, improve their outcomes. It’s also the most expensive part where you can think about for hospitalization. It’s just a horrible place to be, and you try to make your stay there as short as possible. So we looked at that. We started thinking about how we can collect all these different data, all these different algorithms. And we started then doing a historical analysis, and we realized that our algorithms actually fit that data very nicely. And then we further improve on it. So, in other words, we customize our algorithms to what’s available over there, and then life implemented it.
Peach’s AI Got Regulatory Approval in Singapore – Replicating Pivotal Study in the U.S.
So, we ran a live clinical study, and this became a pivotal study, for one year, in Singapore. This was all conditions for about 5,000 adults in the ICU… so it’s a long study and lots of patients. The results were great. It was a very successful study. And we got regulatory approval from the equivalent of the FDA there. So it’s a medical device, and what we’re doing next is essentially replicating the study in the U.S. to get 510(k) FDA approval.
Sal Daher: Awesome. Awesome. Awesome. So basically, from 2013 to 2017, was gearing up.
Klaudyne Hong: That’s right.
Sal Daher: And then, you connected with this opportunity in Singapore. And then, you trained your AI, and, within two years, you have enough data and approval to start using it. So clinically, it’s in a clinic in Singapore right now?
Klaudyne Hong: Yeah. So it can be implemented in the clinics or commercialization. That is possible, because we do have that regulatory approval and what they call “ISO 13485 certification.” So we can do that. What we’re interested in is also making sure that, as we build up that platform… So in Singapore, we were looking at a couple of different things to predict. So one of them is the inflammatory response of the patient. And the other thing is the health of the organ system. So the six major organ systems, we predict what’s called a “SOFA score,” so “sofa” like the furniture that we’re sitting on, “sofa.” And it’s short for “Sequential Organ Failure Assessment” score. It’s a mouthful. Basically, these two responses, if you will, the patient’s response, inflammatory and the organ dysfunction, they are scores that have been created by clinicians originally to see if someone has sepsis.
“Sepsis, by the way, is the most expensive health care condition, hospitalized condition, in the U.S.”
And sepsis, as you probably know, is the body’s systemic and huge inflammatory response to fighting an infection. So body goes haywire. It starts attacking itself, instead of just the invading organism, usually bacteria. And this leads to organ failure, and 20% of sepsis patients will die. Sepsis, by the way, is the most expensive health care condition, hospitalized condition, in the U.S. We’re talking about $27 billion a year spent on it. It affects about one and a half million patients. That was what we tackle in Singapore, whether or not we can predict inflammatory response, whether or not we can predict organ failure, as well. So if you predict in advance, you can try to prevent that.
And since then, what we wanted to do was grow our portfolio, and that portfolio means being able to predict other things. So now, we can predict vital signs. So, a patient who’s in the hospital in ICU, or a patient who’s got a medical device, a wearable that they have at home with them, we can actually predict the vital signs. We can predict if someone’s going to have a cardiovascular condition, so if someone’s going to go into cardiac arrest. And that’s pretty helpful. If someone’s going to go cardiac arrest, and you know it several hours in advance, you can make a huge difference.
“If someone’s going to go cardiac arrest, and you know it several hours in advance, you can make a huge difference.”
Sal Daher: There’s something you can do it.
Klaudyne Hong: Oh yes.
Sal Daher: Yeah. There’s something you can do. I don’t know, to what extent, you can do something about sepsis. There are many causes for sepsis. I mean, it can just be an organ failing and bleeding inside. If their kidney function is degraded, you begin to have breakdown of tissue, and then you have stuff going places it’s not supposed to. So there’s nothing you can do about that. But in a case where you can do something about it, I can see the value,
Klaudyne Hong: You know, interestingly enough, sepsis, it starts of an infection that becomes systemic. So the body tries to fight it. Then it starts to attack itself. But that’s the whole point, if you can diagnose or you can detect… or, in this case, predict early enough, before any of this serious bad stuff happens to the body, you can prevent it from happening. Because it’s all about the different kinds of signaling molecules that come about, you can then slow them down, if you know they’re coming up. So sepsis is a very interesting condition. And when we think about inflammation, the systemic inflammation that is causing or associated with sepsis, there is this likeness, if you will, to what you hear about in the news every day is COVID-19.
So, if you think two or three months ago, when we were told about coronavirus infection, it was all about, “Okay. It can infect you. It can cause pneumonia. In some patients, that it can cause lung damage.” And there was a mad scramble over the world for ventilators. The lung was a primary focus. And then, COVID-19 hit us big, here in the U.S., and we have since had first-hand experience of thousands and thousands of patients. And we started discovering and realizing that a significant number of them, who are hospitalized for a long time, they actually also have developed this inflammatory response, very much like sepsis, in the sense that you have what you read in the papers and they call this “cytokine storm.” The cytokines are released. They attack the body and can lead to organ failure. It can lead to a heart attack. It can lead to the kidneys, as you point out earlier with sepsis.
So, all of this, if know in advance if a patient is going to have sepsis or COVID-19 and how they’re going to progress down a different cascade pathway, then you know how to treat effectively, and you can treat early. That’s the important part, being able to know something in advance and then knowing the proper treatment and whether or not your patient is responding well to that treatment. These are the things that make a huge difference in healthcare.
Sal Daher: Are you looking at COVID-19 data sets to try to-
Klaudyne Hong: We would love to work with COVID-19 data sets. What’s available now are not the kind of data sets that we can use, In other words, we need the kind of data sets that come from the hospitals, directly from individual patients. So these are your electronic health records. These are things like what Martin does really well.
Sal Daher: Yes. This is what Martin is working on the COVID-19 Research Database.
Klaudyne Hong: Exactly. Exactly. It’s going to be very interesting to tap into that. And, of course, we’re always open to also collaborating with health care partners here in the U.S. I think when things get to be a calmer stage, where we have a chance to breathe a little bit, especially we’re talking about our health care providers, who’ve been working so hard… Yeah, it would be a really good time to try to make more sense of that data set. We’ve just heard too many unexpected occurrences of events, if you will. And it’s going to be a big mess to try figure it out. And I think AI will be very helpful in that regard.
Sal Daher: Yeah. COVID-19 has thrown us for a loop in so many ways, that is, just a lot of unexpected turns. Excellent. So, what’s the current status of Peach IntelliHealth?
Current Status of Peach IntelliHealth
Klaudyne Hong: Well, we are going to be starting a U.S. study sometime this year, after the whole COVID-19 thing, again, calms down a little bit. And then, it’s a replicate of what we did in Singapore, except it’s going to be a much shorter study, with a few hospitals. Get FDA approval. And then, it’s ready for commercialization. A lot of what we have been doing is focus on the hospital, but there’s also that bridge to before patients come to the hospital or after a patient leaves the hospital, how do you prevent them from returning, because they don’t want to return, right? So if you can prevent them from deteriorating while they’re at home or they’re at work… Are there AI algorithms that will allow us to predict someone outside the hospital.
And this is where we’re looking to partner with physical medical device companies, so those who make, for example, vital sign equipments. If we can get vital signs of the patients, who already have algorithms that have been customized, can plug in play and basically say, “Okay, if you’re worried about a patient with a heart condition, that they may have cardiac arrest after you release them from the hospital for the first X number of days, 30 days, this is what we can predict 12 hours 24, hours in advance.” That prediction shows up the hospital database. It shows up with the primary care whoever’s the designated carer for this patient. They can come back, and then this can be prevented.
Sal Daher: Huh. This gets me thinking about one of my portfolio companies, a company called Senscio. And they are doing exactly that. They have devices for patients at home to monitor and to try to learn their behavior and to try to prevent them from having to be hospitalized. These are patients dealing with chronic conditions, serious chronic conditions, and they want to prevent it from becoming acute, from having to be hospitalized. So they’re monitoring them and looking at using AI to just see what’s happening with them and to keep track. Have you run across them at all? Are they on your radar screen?
Klaudyne Hong: Oh, no. I can’t say that I’m aware of them, but it sounds wonderful what they’re trying to do. If there’s a space or opportunity for collaboration, I would love to work with them.
Sal Daher: Actually, talking to you, I was saying, “Oh jeez. I’d like to get the co-founder and CEO, Piali De, on my podcast. She should be on my podcast.” I think it there’s a lot of work being done on this area with chronic illnesses. There’s a lot of room for development.
Funding for Peach IntelliHealth – the Government of Singapore Came in as a Seed Investor
Great. Basically you started back in 2013. What kind of backers do you have for your venture?
Klaudyne Hong: So we’ve been very fortunate. From the science and technology side, we’ve got clinicians in various hospitals, U.S., Singapore, Japan, elsewhere. They’re very interested and supportive of what we do, key opinion leaders. And from a financial perspective, we’ve been super fortunate. We have had the Singapore government come in as our seed investor, as well as other angels, and also, an only VC… so very, very lucky in that regard. We’re always open to creating new alliances and partnerships with others, just broadening our network. So we’re always open to that.
Sal Daher: Excellent. Excellent. Were you living in Singapore?
“I grew up in Singapore, and I left when I was relatively young. So, I was as an undergrad in California.”
Klaudyne Hong: I grew up in Singapore, and I left when I was relatively young. So I was as an undergrad in California. And I started out there’s a pre-veterinarian major. So Singapore is an all urban, all city. And the first time I saw cattle was in California.
Sal Daher: We might as well get to… We talked about the company, and so, maybe we can do a little bit of biography and entrepreneurial journey here. You were born in Singapore.
Klaudyne Hong: So I grew up in Singapore. Came here as an undergrad to California. And, because I grew up with seven dogs, and I read it every single book from James Herriot and he’s-
Sal Daher: How Green Was My Valley [All Creatures Great and Small, actually]
Klaudyne Hong: Ah, there you go. Yes. So I thought I was going to go up, take care of giraffes, dogs, cats… So I actually took animal husbandry, worked at three or four vet clinics as an undergrad and then, realized I actually wanted something different. This came about the time when molecular biology was having its huge, early hay days. It was bordering on when Eric Lander and others were about to start the Human Genome Project. And I’ve always loved DNA. I love DNA sciences. And I was teaching microbiology to undergrad, so transitioning from animal to bioscience turned out to be, actually, quite easy. It turned out to be a really good fit.
So, at that time, then I was doing my bio sciences in the evenings, I would watch, when I was having dinner the sitcom, Cheers. Do you remember Cheers, the theme song where everybody knows your name?
Her Favorite Show “Cheers” Showed Boston in Such a Good Light that Klaudyne Chose to Apply to MIT
Sal Daher: Oh, Cheers. Oh, yeah, yeah, yeah. It was supposed to be here in Boston. I remember the Cheers bar.
Klaudyne Hong: So I loved the theme song. And it was a program, it came on while I was having my dinner, and it got to me. And I figured Boston might be really nice, so I applied to MIT for grad school, and, strangely enough-
Sal Daher: Because of Cheers.
Klaudyne Hong: … they let me in.
Sal Daher: Who would have though?
Klaudyne Hong: Because of Cheers. So, thankfully, Boston, also turned out to be totally amazing. I think if you are a student in Boston, it’s one of the best places, if not the best place to be at, especially if you love what you’re doing. Places like MIT, they’re very rare. I mean, you walk along infinite corridor any other day and night. If you want something to not be bored, you’re going to find it.
Sal Daher: There’s always something happening. The hardest thing at MIT is getting to sleep.
Klaudyne Hong: I agree. I agree.
Sal Daher: I think there are lots of very hard things, but the hardest thing is getting enough sleep, because there’s so much going on.
Klaudyne Hong: It’s a chronic thing. It’s a chronic condition there, and sometimes in a good way.
“…my thesis…was about controlling gene expression for gene therapy applications.
Sal Daher: So, I joined MIT. I studied toxicology in the beginning, and then eventually joined the lab, a chemical engineering, bioengineering lab. And the professor who became thesis advisor was Professor Doug Lauffenburger. I did my thesis, and it was about controlling gene expression for gene therapy applications. And pretty soon after the thesis… It was a short thesis. It was about two years. It went by really quickly, and I was offered a job at Schering AG. And they were based in the Bay area, as well as [inaudible 00:22:48], so I went to the Bay area office.
Klaudyne Hong: So this is what the T says there was a short he says it was about two years went by really quickly and I was offered a job at sharing are gay and they were based out in the bay area’s was bullying. So I went to the Bay Area Office.
Sal Daher: So, this is what Martin referred to as a “two-year PhD.”
Klaudyne Hong: Oh, yes. I think so.
Sal Daher: How did you get a two-year PhD?
Klaudyne’s Two-Year PhD
Klaudyne Hong: It was all those things. You know, part of it was just Paulo Coelho thing, when you really, really want something bad enough, you got to believe the whole world conspires to help you get it. And I think that’s partially what happened. I had already been doing another these at a different lab. And there were some things that didn’t work out. I couldn’t wait around for more samples to come out. They just weren’t happening. We weren’t in control of the receipt of these samples.
“I drew him a picture of this DNA that was wearing cowboy boots, and somehow told him it’s all about the cowboy boot being a gene therapy vector. And somehow, he understood what I was saying. He said, “Yeah, come join me in the lab.””
And I figured, I’ve always loved DNA sciences. I wanted to work in gene therapy. And so, I found a way. I interviewed with two other professors and Doug Lauffenburger. I drew him a picture of this DNA that was wearing cowboy boots, and somehow told him it’s all about the cowboy boot being a gene therapy vector. And somehow, he understood what I was saying. He said, “Yeah, come join me in the lab.” So I try to be very specific about my time, and I said, “Hey, I’m going to do X, Y and Z and would that be enough to basically submit a thesis?” He said, “Well, if you can do X, Y and Z, yeah, sure, of course.” “And if I do this in two years, which is my timing, would you be okay letting me go?” And he says, “If you can do it in two years, sure.” Two years later, I reminded him. I showed him X, Y and Z, and he’s like, “Okay. Prepare for your thesis defense,” which is good, because I had been going around interviewing for jobs in pharma.
And Schering AG, not long then, basically said, “Okay, come on over to California.” Gave me really great projects to work on. Start out with cancer vaccine. Then I did gene therapy, stem cells. You know, work on things like Parkinson’s, multiple sclerosis, heart disease. Fantastic, really, really interesting projects. Learned a lot.
And then, a few years later, Johnson & Johnson are in New Jersey. They were looking for someone to build a team for their sensor platform. And they heard about me and the things I was doing at Schering. And one thing led to another, made me an offer. I returned to the East Coast and built out what became J&J’s largest regen med [regenerative medicine] team and portfolio.
Sal Daher: Oh, wow.
Klaudyne Hong: So, yeah, it was quite amazing, the kind of opportunities, the kind of resources a company like J&J associated with. Good times.
Sal Daher: Johnson & Johnson are a huge company with-
Klaudyne Hong: Deep pockets.
Sal Daher: … very deep pockets. Very deep pockets.
Klaudyne Hong’s Entrepreneurial Journey
Klaudyne Hong: So, I guess, you’re asking about my entrepreneurial journey. It’s basically following and trying to do what I enjoy. It’s a simple logic, if we are doing something that we enjoy we’re going to be better than it we’re doing something we don’t enjoy or are not interested in.
Sal Daher: So Klaudyne, let’s go back to that fishing trip. Okay.
Klaudyne Hong: Okay.
Sal Daher: So, at the time of the fishing trip with your PhD… It was your PhD advisor, right?
Klaudyne Hong: He was not my advisor, but he was one of the professors.
Sal Daher: So, what were you doing then? Were you working for Johnson & Johnson?
After J&J Klaudyne Consulted But Found It Unrewarding, Lacking Ownership – Enter Entrepreneurship
Klaudyne Hong: No. No. So I had just come out and started my consulting firm at that time. And so, I was trying to figure out, other than consulting, what I could do, what else I would be interesting to me. And one of the things… Consulting was very interesting. You obviously get to meet CEOs, CSOs, CTOs of companies, big and small. And what I was helping them with was, because I was on the buy side of the table in pharma, and they were trying to sell to pharma… They were trying to come out with, “What is that magic formula that will make their product more desirable?” Was it certain more tests that were needed? Does the test need to have X, Y and Z components to it? And what the competition will look like, how would they place a position of their product? So it was really interesting. It was a good feeling to be able to help. The problem was that you never had ownership of the projects.
Sal Daher: Yes. Yes. Yeah. Well, because consultant, you’re literally a hired gun.
Klaudyne Hong: You’re a hired gun. You’re a hired gun. So, it doesn’t really appeal to that creative innovation side of me. That side just felt really starved. So I decided I was going to look for some kind of startup opportunity, potentially, but this fishing trip, as I mentioned earlier, I think I had a consulting meeting in Boston, and then, everything else just came up from that.
Sal Daher: Fantastic. Fantastic. How did you get Singapore as a backer in this? Which arm? Is this the Sovereign Wealth Fund?
Klaudyne Hong: Yeah, so the Singapore government has several different arms to invest in different kinds of technologies, companies or platforms. The Sovereign Wealth Funds tend to be very late-stage investments. So, in the startup finance world, it would be in a Series C, D. Check sizes are really big. So Singapore government has other arms that deal with the earlier stage of the startup world. And this is one of the government arms called “NRF.” And there have been also other types of support.
Peach Benefitted form the Entrepreneurial Ecosystem in Singapore that Allowed the Rapid Acquisition of Patient Data
The ecosystem there is quite remarkable. It’s a very small country, but when they’re committed to supporting innovation, they go all in. In fact, a couple of the professors on the Peach team are also professors in Singapore, as part of the Singapore-MIT Alliance program. So we’re all very familiar with how they look at innovation, how to support innovation.
And money is one thing, but it’s also something else when you think about the ecosystem for the kind of data that you can get your hands on. For startups here in the U.S. when it comes to health care data, often times, you have a hard time getting high quality and massive amounts of high-quality data. And our partners there have been very generous. It’s just been a really good relationship, a great deal of trust and been very transparent. So that’s also an environment that fosters, I think, good collaboration and innovation together.
Sal Daher: Yes, yes, yes. Singapore doesn’t do anything halfway. I was at a business here in the U.S., and I set up a company in Singapore and was involved, was there for a year. And then we shut it down, because our business here in Boston just exploded, and I was needed back here. And it was also the wrong time. Singapore just wasn’t quite ready for the stuff that we were doing. Ten years later, it would have worked, but not at the time that we’re doing… This is back in ’93, ’94.
But the thing that struck me was just how incredibly easy it was to do business in Singapore. Everything. It was just like very easy to get paperwork done with the government. Very easy to get our lease. Very easy to find office space, to find housing, to find people, to hire people, to fire people if you needed to. I think it really surprised me. Everything about the place is incredible.
We lived in this place that was like a 17th floor apartment, and we had this enormous piece of furniture, that every time that was moved, got broken. Okay. When it got moved in Singapore, I said, “These guys are going to carry this massive piece of furniture down 17 flights of stairs. It’s going to be splintered by the time they got it.” Now every time that I moved before that, that thing had gotten broken. We had to have it repaired. When those guys move the up the stairs and down the stairs, I couldn’t believe it. Nothing. Not a scratch.
Klaudyne Hong: It was too big for the elevator, that’s why they had to take it by stairs.
Sal Daher: It was too big for the elevator. We had to go down 17… It was an enormous thing. And I thought, “Oh gosh.” American movers are much bigger than Singaporean movers. “These guys are going to carry this?” It was incredible. I mean, it was just amazing.
And then, I came back here. We had this apartment over there. We head an office and all that stuff. I had my cousin, who was there for an internship in Singapore, just close up shop, and I couldn’t believe it. I mean, we got all our furniture back. The deposit on the apartment, I got it. I thought, “I’ll never see that money again.” Got it exactly on time. The legal system works. It’s amazing. The place is just incredible.
Klaudyne Hong: Yeah, it’s very well run. It’s very structured and it’s orderly, so you know what to expect. Everything is written. And they try to be fair about things. It’s also about relationship building there. It’s a small place. If you do something, if you do it well, your reputation grows. And it’s the opposite, if you don’t do it well [crosstalk 00:32:04]-
Sal Daher: Everybody knows about you. Everybody knows about you. It is a small place. Everybody’s like, “Ah, yeah, my cousin told me this guy, forget about it.” Well, you’re going to find that the world is very small. With all these online resources that exist, the world has gotten very small for nasty people, let me tell you. Bad bosses, for example, get outed very quickly. I remember in the ’70s and the ’80s, you could be a very terrible boss, and there was no way for people to know. If you were tapped into the grapevine of the corporation, you’d know, “Oh, that guy. Forget about him. Nobody wants to work with him.” But if you were some luckless person coming from the outside, you’d have no way of knowing. And now we do. So information travels. That’s great.
Klaudyne Hong: Definitely. Yeah it’s changing more and more. Yeah, even in health care.
Sal Daher: Even in health care, accelerated by COVID-19, of course.
So, Klaudyne, I just wanted to get the sense. So, basically, your move into entrepreneurship was, you’d been in a large corporation. You saw the resources they had. And then you went out and you tried this thing, I’m going to be a consultant.” And the consulting was interesting, because you got to connect with people of all sorts of levels and so forth, but you didn’t have ownership of your project. So then, when did it dawn on you that the ultimate thing for you to do is to have your own venture? So how did that thing come about in your head? Did you have… Were your parents entrepreneurs?
Klaudyne Hong: Yeah. I never really thought about when it actually clicked in my head. I think it must have been from I was a grad student. You know, you were doing your own thesis… And I had to switch thesis midstream, so…
Sal Daher: Oh, okay.
“But I find that the larger the company, sometimes the slower things can be, because I think they call way too many meetings that you have to attend.”
Klaudyne Hong: So being able to start and end a thesis within two years… And some things you have to do for that, is be able to strategize and plan and execute as flawlessly as you can. And so, that means you spend a lot of time up front thinking about what you’re going to do before you do it. And when you’re doing it, it’s a very conscientious effort of getting it right the first time. And I think part of that, when then you then join pharma, and this is not what happens in real world in pharma. You’ve got great people, and everyone teaming up. Things get done. But I find that the larger the company, sometimes the slower things can be, because I think they call way too many meetings that you have to attend. You know, when I became a team lead and the project got so big, I spent, at least, about 70-80% of my time during some weeks, just sitting at meetings and half of the meetings weren’t necessary.
So, I’ve always known that I wanted to work with like-minded people, who really cared about what they were doing, and that they wanted to do it well the first time around, or as close to being the first time around. So this idea seeded early on. And I looked at AI. I looked at health care. And we were talking about 2013, 2012, the Health Care Act, the access to EHR data… All of that was just coming about. Prior to that, it’s a mess. Now you have all this great data sitting around and is latent. It’s basically gold that’s sitting around. No one knows how to mine it, mine it well.
And I thought if you had the right AI people, who understood both AI and how to change it and how to manipulate it, in order to customize it to health care data… And you almost never get the two types of expertise in one person, so you really need a very niche team, each of whom has been doing what they’re good at for quite a while. And then, they have to bridge together with each other to be able to do that well collectively. Then you can have something really powerful, and that’s really what that thought process was. And when Gerry and I were fly fishing, it just completely makes sense. He mentioned some professors. I knew some of the ones he was talking about. I think one of them was a postdoctoral fellow at Whitehead when I was a grad student, so I knew him a little bit. And it’s like, “Okay. It’s time to get back in touch.” And that was Bruce Tidor, so he became my co-founder and CTO.
Sal Daher: Oh. Okay.
“They see that you have this passion. And they see that look, if it’s that crazy, why do you have so many good people embarking on this journey with you?”
Klaudyne Hong: It’s always been there. It’s about how quickly you want to do something, how well you want to do something and what that purpose is. If you believe in it, then don’t wait around. Obviously, you need the funding. You need like-minded investors, who are supportive of this. And, especially, in the early stage, angels are just this incredible group of investors. Ideas might seem a little bit flimsy. Might seem, often times, absolutely crazy. And this is when you know you’re meeting a like-minded angel. They see right through you. They see that you have this passion. And they see that look, if it’s that crazy, why do you have so many good people embarking on this journey with you?” And then, they look at the data and make up their own minds. And this one of the pure joy for me, being in this company at Peach, is when I see it click in someone’s eyes that, “Oh, this is not crazy. This is actually very innovative, and I can’t believe they’ve done this.”
Sal Daher: It must be rewarding. I like what you’re saying about bringing people in. Sometimes that you, jokingly I refer to a startup as a “folly.” It’s a crazy thing, because you’re doing something nobody’s ever done before.
Klaudyne Hong: Oh, it absolutely is.
Sal Daher: If you look at it rationally, you shouldn’t be doing it, because it’s not a rational thing. It’s a folly. But then if you bring another capable person in, who’s complementary to you in skills, and that person is not insane, and they believe in your folly, it becomes less of a folly. Are you familiar with the term “folie à deux?”
Klaudyne Hong: No, I am not.
“It is the most robustly supported result in the study of entrepreneurship that more co-founders, better chance of success.”
Sal Daher: It’s a delusion where two people participate in the delusion, and the term is “folie à deux,” meaning a madness of two people. Sometimes there’s “folie à plusieurs,” which means, “madness of many,” but those are very rare. So if you can bring in two or three people into your folly, all of a sudden. It’s become serious business. It’s stuff that… People believe in it and stuff gets done. It is the most robustly supported result in the study of entrepreneurship that more co-founders, better chance of success.
Klaudyne Hong: Oh, absolutely. Absolutely. You got to just surround yourself with people better than you in other areas. It’s that synergy that when bunched together… Like that old saying or that old story about a bunch of sticks, it’s really hard to break them-
Sal Daher: Oh, the fasces. Yeah. Yes.
Klaudyne Hong: … as opposed to one. And this is what building up a startup team is. You want people who can think differently, who can look at a problem from a different angle, because, ultimately, when it comes to bioscience, it’s a four-dimensional puzzle. The x-axis is the genetics. The y-axis are the environmental, external influences, when this being or this person was growing up. The z-axis is, basically, very specific incidents. It could be the softball hitting my head. It could be cancer. It could be something. And then your fourth axis is, obviously, time. So being able to place all four of them together, using AI to do that, it is a little bit insane, but it’s possible.
You can’t do everything. We haven’t done everything, but we are on a path. And I think, as you said, you’re other startup is also doing some other things, but for chronic care, where, as we specialize more in the acute, subacute care space, it can be done.
Sal Daher: Mm-hmm (affirmative). I’ve interviewed Gauss Pharmaceutical, the machine vision company that counts stuff that’s going into a patient, out of a patient. They have camera in the operating room. They can tell a piece of cloth if it has blood on it, if it doesn’t have blood on it, and that kind of stuff. There is AI that identifies cancer cells, various types. I’m an investor in a company that’s AI for dermatology. And so, where do you see the AI coming in? Is the AI going to be an aid to clinicians? Where do you think it will fit?
Where the AI Will Fit in the Clinic
Klaudyne Hong: In the process of care, I really see AI as the ultimate in clinical decision support. So it’s not about replacing a clinician or specialist. It’s about helping them look at things they might not have had the chance to look at, not have the time to look at, because they’re taking care of so many different patients, or they may not have access to. So we are looking at millions and millions of data points. So for any one patient, any one time, we can look at multiple inputs from that patient at that time point, at multiple time points before. So if you think about it graphically, it’s many different graphs converging together, to give you a pinpoint result of what would happen in the future. So this is where I see predictive AI being very helpful is giving them, the clinic, the hospitals, that precision in medicine of what would happen. And if they were to do something, then what that different outcome, hopefully a better one, could look like.
“I also see AI in a sense of helping make informed choices about the kinds of medicine we take.”
I also see AI in a sense of helping make informed choices about the kinds of medicine we take. So I’ll talk about oncology for just a split second. I’m very interested in cancer research. I did a lot of cancer modeling at MIT. One of the things we did very well, and I think that the team starts to do an amazing job, is understanding the types of mutations that will lead to certain kinds of cancers and how different patients with the same kinds of cancers will behave differently to different drugs. So, if you have AI that can help predict why someone or that someone is going to have an adverse event to a new class of immuno-oncology product, and you can help taper down that dosage, so they don’t have such bad adverse event. We’re talking about many, many thousands more patients, per drug class, being able to take the same drug that they otherwise would give up on, because adverse events are so harsh. So these are a couple of different, how shall we say, impactful ways that AI can help.
Sal Talks About Portfolio Company Leuko
Sal Daher: Yeah, yeah. Yeah. That brings to mind another company, talking about people taking cancer therapy and becoming immunocompromised. I’m an investor in a company called Leuko. And, basically, they are building something that ultimately is going to be instantiated something that looks like a pulse oximeter that goes on your finger. Right now I think it’s like it’s much bigger than that, but the idea is that they will be able to count what blood cells going through the capillaries, right next to your fingernail. And so, they’ll be able to tell if someone’s white cell count is dropping, and, therefore, their immune system is compromised. There could be tremendous, huge savings in terms of lives and in terms of costs of treating people who get these opportunistic infections, because they’ve become immunocompromised when they’re undergoing chemotherapy or treatment for Parkinson’s, for example, I think it can also result in that. Instead of somebody having blood drawn every two weeks or something, they can just check all the time.
Klaudyne Hong: I love that example, because this is one of the things that we predict. We can predict the hematological lab value. So for example, the exact white blood cell counts in a patient’s blood 12 hours 24 hours from now. So if someone is taking a new drug that’s going to deplete the white blood cell counts or their platelets or their hemoglobin, we can predict that. So before the levels that down too much, you supplement them or you give them something else to prevent that. So absolutely, I think these are all really great use cases and in technologies that can help enhance that experience is what we should be thinking about creating.
Sal Daher: Awesome. Awesome. Klaudyne, we’re reaching the end of our interview. And so, I’d like to open up the forum to you and to just give you a chance to address our audience about anything that you would like to say. You can enthuse about fly casting, or you can talk about something else.
Klaudyne Hong’s Closing Comments
Klaudyne Hong: Yeah. No, I think I’ll leave so fly casting for another time, but, first, Sal, thank you. It’s been wonderful to be here. I think if any of your listeners are interested in learning more about how AI can benefit them or their hospitals, or they want to collaborate, or they want to apply for a job, they should feel free, they can find me on LinkedIn.
Sal Daher: Should I give out your name, as well, so they can find you on LinkedIn?
Klaudyne Hong: Oh, yes. Go ahead. Yes, it’s Klaudyne Hong.
Sal Daher: Yeah. Klaudyne Hong spelled K-L-A-U-D-Y-N-E. Hong, H-O-N-G. That’s tremendous. Klaudyne Hong, I’m very grateful to you for making time talk to me here at Angel Invest Boston. I think this has been a very informative interview. I’m very grateful that you made time to be on.
Klaudyne Hong: It’s a pleasure. Thank you.
Sal Daher: This is Angel Invest Boston. I’m Sal Daher. I’m glad you were able to join us. Our engineer is Raul Rosa. Our theme was composed by John McKusick. Our graphic design is by Katharine Woodman-Maynard. Our host is coached by Grace Daher.