How AI Is Slashing Medical Product Development Time by 50% with Megan Rothney

Transcript

[00:00:02] Eoin:  I’m, Eoin. Welcome to Signal to Noise. I’m joined today by Megan Rothney. Megan is, uh, is a PhD in biomedical engineering from Vanderbilt. She’s got her bachelor’s in bioengineering from the University of Pittsburgh, and then since then has had a long and storied career, uh, across diagnostics, precision medicine, and medical imaging. Uh, Megan’s impressive career includes ten years at PathAI, where she served as VP product, Verily Life Sciences, genomic health, GE Global, uh, and the NIH. Uh, Megan’s reputation precedes her in that she’s known for bringing development, clinical, and commercial realms altogether, both the people and the processes, into steering product development and that tying into life cycle strategy, you know, FDA submissions, and then overall, uh, market access for oncology, molecular diagnostics, and, uh, and health technology. So quite a unique blend of technical expertise, uh, and even more unique, I think, strategic lens and vision that Megan brings to the table, and I’m delighted to be, uh, talking to you today about all things, uh, leadership, AI, and and the world of medicine.

[00:01:14] Megan:  So Thanks, Odie. Really excited to chat.

[00:01:16] Eoin:  Excellent. Look. Uh, you know, this being the the signal to noise podcast, let’s start there. What signal are you paying attention to in the market right now, and what noise are you tuning at?

[00:01:27] Megan:  Yeah. So I would say with all of the hype around AI and it being central to my job, I am paying a lot of attention to all the new governance that’s coming out around AI, whether it’s new FDA guidances or the European AI Act. Um, how all these things shake out has a huge impact on how quickly or slowly we’re gonna be able to bring products to the market, so I have to watch it pretty closely. Um, in terms of, you you know, noise, I feel like we’re living in a time where there’s a lot of sort of global economic uncertainty. I’m really trying to tune that down, the phase of life that Pictor Labs is in. We really just need to put our heads down and focus on doing what we do really well and kind of ignore ignore all the the other things that are happening around us.

[00:02:08] Eoin:  Yeah. What’s your what’s your filtering system there? Because like you said, it’s it’s headline after headline. There’s activity that front and center. So how do you filter out the noise? What what what catches your attention?

[00:02:18] Megan:  Yeah. I mean, I think the number one thing is I I try to stay away from the popular media as much as possible because it’s just impossible to avoid it. I think in the scientific news, you know, I really try to focus on what are we hearing about, new technologies that are being developed, new technologies that are being cleared and approved. Um, when I go to trade shows, kind of looking at what are the other companies promoting, and I really just try to pay attention to the science and technology and just stay away from, like I said, the popular media. Um, it’s just too much, I think.

[00:02:47] Eoin:  Yep. That seems, uh, like a sensible move. I wish I could do the same. That’s okay. Well, just because we started to toward the topic, you know, obviously, Megan, you’ve got front row seat to a lot of the significant changes that AI is enabling in health care, whether that’s improving diagnostics to personalizing treatments. How do you personally see AI revolutionizing the health care landscape? I know that’s a huge question, but maybe if you wanna focus on areas like diagnostics or patient outcomes, what’s the what’s the potential? What’s the timeline?

[00:03:17] Megan:  Yeah. It’s it’s a really, really exciting time to be working in AI. I think we’ve been talking about it for a long time, but health care data has been so fragmented that we really couldn’t necessarily action on all of the great ideas that people had. And I think what’s really cool that’s happening right now is the data assets are getting large and a little bit more controlled at the same time that the algorithm technology is kinda catching up. And so we’re really starting to be able to put those two things together and put more products out on the market. Um, I think that, you know, some of the places that I think this is gonna have a huge impact, I think drug discovery is gonna be kind of revolutionized by this, um, you know, getting getting drugs to market much faster. I think in the in terms of of diagnostics, um, really what I’m seeing is it’s just easier right now for us to get to a prototype that we can then get out in the world testing than it’s ever been before. So that’s it’s just incredibly exciting. And I think over the next five to ten years, we’re just gonna see a huge shift in in medical practice towards, you know, physician support tools. So not replacing physicians, which I think is one view of the world that many people had, but really thinking about how can we make them do their jobs better.

[00:04:25] Eoin:  Fantastic. And, yeah, you touched on it, but at PictorLabs, and for the edification of our audience, right, you’re developing AI powered virtual staining platforms. Right? And so how do you envision specifically to that domain AI in histopathology changing some of those standard practices in the labs and the overall in health care delivery?

[00:04:46] Megan:  Yeah. So, you know, histopathology for for people who may not know a lot about it, it’s really the process of taking the the sample that’s removed from your body and coming up with that initial diagnosis, and the diagnosis is done by a pathologist. And they may need now five to seven slides of tissue in order to find the right diagnosis for you depending on, you know, the the things that are in your differential diagnosis. What we’re doing at PictorLabs is we’re trying to reduce that so that you only need one slide to do all of that analysis. And that may not sound like such a big deal, but the reason that it is is twofold. One is you have more tissue available for other genomic analyses. So if you really wanna do precision medicine, you have to have a sample from the patient to really learn about their disease. And so we’re making sure that more patients will be able to benefit from that type of analysis. The second thing we’re doing is saving time. Because we’re able to do these, uh, these virtual stains almost automatically, you know, within a matter of minutes, you know, it really allows a pathologist to make a diagnosis on the same day that the sample arrives in the lab. And any savings that we can get get to the diagnosis can be applied to getting patients on treatment faster. So those are really the two big impacts that I see of the technology that we’re working on.

[00:05:57] Eoin:  Fantastic. And what are some of the challenges with with implementing that technology so far?

[00:06:03] Megan:  Yeah. I mean, there’s there’s always, to me, two big challenges, and we’re no different from anyone else. The first is that you have to have really compelling evidence. You know, physicians aren’t just gonna try AI because we tell them they should. They wanna make sure that they’re still gonna get the same level of diagnostic quality. And so we have to show it to them in, you know, peer reviewed publications by letting them play around with the technology themselves. So, you know, getting that data is one thing. And then I think we’re always in health care kind of fighting organizational inertia. Right? Change is hard. It can be expensive. And so we have to come up with really compelling value propositions that make people go, yeah. Like, it’s the right time for me to introduce this into practice.

[00:06:41] Eoin:  Yeah. Interesting. And it feels like a lot of kinda hand to hand combat in some ways, right, and and almost at loggerheads with the pace of what’s happening with AI and then the pace of the industry are are kinda quite different. So how do you how do you strike that balance? Because you also then have to layer in FDA approvals and the regulatory process, but AI is moving at a thousand miles an hour, and those things don’t. Well, how are you managing that?

[00:07:03] Megan:  Yeah. I think the first thing is, you know, to your point, like, the algorithms are moving so fast. It’s, like, almost impossible to keep track of them. Right? Like, every day, there’s a new new iteration of a large language model, and you’re like, wow. That could really transform what what we do. I think the key in health care is we’re a long way from being able to, you know, update the model that’s being used in the clinic, you know, every time the development sprint ends on Thursday morning or whatever. Um, what we really are trying to do, though, is to predict a couple generations out based on what do we need the product? How do we need the product to perform? And, you know, are these improvements that we wanna make improving diagnostic quality? Are they improving speed? And then it’s working with regulatory bodies like the FDA to explain what that development path looks like because they do have some roots to allow you to make certain product changes if you let them know that you’re anticipating that those are coming. So I think it’s really incumbent on us to, like, be futuristic in our thinking so that we can make as many of those changes as possible. At the end of the day, the technologies are impacting patients’ lives, so we do have to, you know, take that seriously and do the testing that we need. And it does look a little different from other domains. So I’m I’m optimistic that we’re gonna find that right middle ground between pushing things out quickly, but also making sure that, you know, patient safety continues to be at the heart of everything we do.

[00:08:20] Eoin:  Yeah. But it sounds like you’re getting a receptive partner in the FDA that they’re open to to kinda the advances that are happening here working with you?

[00:08:27] Megan:  Yeah. That’s, I mean, that’s been my experience. I’ve worked with two or three different FDA review committees over the last couple of years, and, you know, I think like everyone else, excuse me, they’re trying to figure out exactly what to do with AI. Um, but they’re not close to it, and they want to learn and understand. Um, that doesn’t always mean that they’re gonna agree with what we propose to them, uh, but that’s their job. Their job is to look at it and say, you know, as a developer of a test, are you actually demonstrating to me that this is safe, effective, and beneficial for patients? Um, but they are they understand that AI is here, um, and they just wanna make sure it’s being used safely and responsibly.

[00:09:04] Eoin:  And do you have you as a product leader, how do you build in some of those safeguards to to your own processes? Just on a on a tangent of the question there?

[00:09:12] Megan:  Yeah. I think, like, the first is really understanding what the clinical performance needs to look like. So, you know, what what is the diagnostic process today? What is the the change that we’re hoping to make in it? Is it really just making something easier for a physician that’s hard today? Is it giving them new information that they might not know how to use? So first, we have to understand that. And then as we go through the development cycle, it’s important to sort of continuously do external validation, what we call external validation studies, studies, which is really having people outside the company evaluate the technology. Um, and that just allows us to believe that we’re doing the right thing and to have evidence all along the way that the changes that we’re making are gonna be positive. Um, again, it takes more time, but I think the end result’s worth it.

[00:09:57] Eoin:  Yeah. And you talked about compressing timelines for, obviously, what’s happening with the patient evaluation, but from your product delivery as well. How how would you quantify, if you can, kinda what that timeline compression looks like? What what did it look like pre AI? What does it look like today?

[00:10:11] Megan:  In terms of, like, patient diagnosis or more in terms

[00:10:14] Eoin:  of Yeah. Internally, I think you had mentioned earlier that, uh, the ability to get something tested and up and running and and in the world, if you like, in in a test environment has accelerated rapidly.

[00:10:24] Megan:  Yeah. So I think, you know, I think that what we’re seeing right now, at least what I’m seeing, is that in in a product that might have taken, let’s say, you know, a year before, I think we can cut out almost half the time, honestly, if we’re using AI and applying, you know, the most current modeling principles. And it’s just because the models you know, twenty years ago, someone would have to code the model manually. We’d have to collect new data. We couldn’t use as much of the data that already exists. And so those two things together, the ability to sort of go back into the past and pull patient data and use that in the models and the actual sort of, um, ways that the models are now much closer to being off the shelf for people to use, I just think they’re creating these massive, massive time line improvements.

[00:11:11] Eoin:  And you touched on there, you know, data quality, obviously, is a paramount importance here because otherwise, you can’t afford much entropy at all in your in your models, if any. Uh, so how have you gone about that? Because you’ve been at varying sizes of organizations and and different start points from where data quality was as you walked in versus, uh, being able to build it from scratch. So talk to me about how you look at that.

[00:11:30] Megan:  It’s gotten a lot easier, so that’s the good news. Um, I think, you know, some companies I’ve been at, we’ve really had to generate the data de novo. Right? We’ve had to do everything ourselves. That’s great because you get exactly the data you’re looking for. But as I’m sure you can imagine, it takes a lot of time, and it’s quite expensive. So, you know, to put that in perspective, when we were developing Cologuard, we, uh, had 72 clinical sites and over 11,000 patients that were prospectively enrolled in the study. Um, that was a huge undertaking. Um, it’s a great data resource now. It was a huge undertaking. Whereas in my last, you know, role at Turing, we were able to use pretty much all open source neuroimaging data and develop the product that way. And both the timeline and the cost are quite different. So, you know, looking at those two models, I think it’s really you you look at the problem that you you need to solve. And in conjunction with the FDA or whoever’s gonna govern it, you kinda look at the different strategies for developing and validating the product. And sometimes you can really like I said, you can do it all on data that already existed in the world, and you don’t have to pay for any of it. Other times, those data don’t exist. So I think, you know, it’s all about finding the data. And, you know, and sometimes, you know, PathAI, I think we’ve we did something in the middle. We got data from a variety of different sources. Um, so it can be really anything. It just depends on the product need and the company’s stage and and budget and all those kinds of things.

[00:12:58] Eoin:  Yep. And it seems like, you know, obviously, you you’re doing what you’re doing, but there’s so many others that are in lots of different directions in health care right now proliferating that dataset. So I think that timeline continues to compress. That cost continues to compress over the coming years, which which is exciting. Yeah. Um, zooming out, obviously, AI just has enormous potential in various disease areas beyond the ones that you’ve specifically worked in. Do you have a a point of view on additional or or diverse applications of AI that could impact other medical fields?

[00:13:29] Megan:  Yeah. And I I think there’s a lot a lot there. So I think it’s not surprising. I think that a lot of the AI started in the imaging disciplines just because it’s a great application and the data largely exists already. I think where we’re gonna see it going is in in in other places. Um, it can be anywhere, honestly, from health care delivery. So how do we more efficiently get patients through a hospital system? How do we make sure they’re getting the right test at the right time? So really operational stuff. Um, as I mentioned earlier, I think drug development is a hugely exciting area. Um, and and I think at the end of the day, you know, any any disease that there’s multiple treatment options and and physicians kinda cycle patients through them, You know, mental health is another area that I think is, like, largely untapped. You know, we know that people cycle through, you know, different antidepressant products. If there was a way to predict who was gonna respond to which one, that would be hugely impactful. So I I just I see it going anywhere where someone has a creative idea. I think there’s an opportunity to at least explore it at this moment.

[00:14:31] Eoin:  Yep. Yeah. Like I said, picture the the tip of the spear is at the point of diagnosis, but there I think there are upstream and downstream opportunities. Right?

[00:14:38] Megan:  And, uh,

[00:14:39] Eoin:  uh, you know, I think we’re we’ll start to see AI play, uh, more of a role in personalized treatment plans as well as we come through, particularly in more complex, you know, disease treatment. So it’ll be an interesting interesting time. You know, on that word time, you know, a lot of conflicting opinions in the world about AGI and where we are on it. It’s a continuum of how advanced this thing is. Like, what’s your own take on where we are?

[00:15:05] Megan:  I honestly have no idea. You know? It’s it’s funny because, like, six months ago, I think my answer would have been totally different than it was today. And so I assume that six months from now, it’ll be even even totally different. Um, it just feels like we’re in this explosive growth phase, uh, to me. Um, what I think is so different is, you know, five years ago or ten years ago, the AI drove the products that that I that I was building, but it didn’t really change the way I worked every day. Now it changes the way I work every day and changes the products. So it feels like a lot of things are coming together. Um, I think right now, to me, what’s really missing is that, you know, you get an answer from AI and most people kinda trust it without maybe even asking a question. But as an expert, you look at it and you go, okay. Like, that’s, like, you know, 80% of the answer. So I think it’s gonna be, you know, how long is it really gonna take to get it from 80% to, you know, 95% accuracy? In product traditional product development, that’s, like, the hardest part is that kinda last mile stuff. Um, I don’t know if we’re quite to that last mile stuff, but I think we are getting to the point where the problems that the AI needs to solve are harder. So I don’t know. It’s, uh, anyone’s guess, I think, at this point.

[00:16:19] Eoin:  Yep. Look. As a as a product leader, I feel like maybe more than most functions, we’re at a step change moment in what that function looks like because of AI. And I think I’m, you know, talking to a lot of product leaders from a lot of different domains, uh, not just in health care and bio, but in, you know, software and hardware hard tech. There are a number of people starting to feel like very quickly they’re getting left behind, and they’re trying to future proof their skill set. So how have you managed to stay current with what’s relevant, what’s not, uh, and actually gain the skills. Because it’s like, it’s happening in real time. You know?

[00:16:51] Megan:  It’s happening in real time. And, um, for me, you know, I I look at it in a couple of ways. One, as as a product leader, at least in the way I like to implement my role, it’s inherently very cross functional. And so, um, the reason that I think I can bring value to an organization is not just because I understand product development. It’s because I understand regulatory, and I understand quality, and I understand clinical operations and and all of these kinds of things. And, you know, at least right now, I think AI still can’t quite put all those pieces together. Um, and in some cases, you know, I’m still faster than the AI, although I know that’s probably not always gonna be true. Um, I think for me, you know, there’s been some great ways that I can use AI to sort of just make me better at what I do. So I’ll, you know, write a set of requirements and ask the AI, like, what have I missed? And it’s great at finding things that I’ve missed. So for me, I’m really kind of leveraging it for rapid prototyping, um, to make sure that the direction that I’m giving to various teams, like, makes sense and is complete. Um, and I think it really has kind of transformed the way I do my day to day work. Um, I hope I’m not replaceable yet. Um, let’s see.

[00:18:00] Eoin:  No. No. Your reputation precedes you as being phenomenal at what you just described. The cross functional work is required. The human element of this is Yeah. Is a huge component still, and I think that, uh, the human in the loop is not something we can replace if we’re gonna continue to have trust if if nothing else, I think, you know, is at this entry. So yeah. Fantastic. Um, to change gears a little bit, uh, talking about a little bit more on this tension between innovation and FDA compliance, medical device regulatory areas. You know, do you think there’s a world where the approval process will also become somewhat AI automated, uh, and it could start to make those decisions, or am I living in fantasy land here?

[00:18:44] Megan:  I think we’re we’re a ways off of that, but I think that there are gonna be parts of the documentation review process that can be done by AI, and I think it can be tremendously accelerated. So, you know, if you imagine a world where you take your FDA submission package, which, you know, I don’t know how many you’ve seen, but, you know, they’re they used to be physical binders of papers and hundreds of thousands of documents. If you imagine a world where all that’s going through AI as an initial filter and it’s raising questions or concerns that then go to the human reviewer, you can imagine that that process would go radically faster than the current process where they have to have, you know, an expert enough a reviewer who’s expert

[00:19:21] Eoin:  enough in each document, review it, and

[00:19:22] Megan:  identify issues. Um, I think going fully, Um, I think going fully towards AI approval, you know, I think we need a lot of data to show that the AI is finding the right things. And by right things, I mean, the things that most likely impact patient safety because those are really the things that we have to be most concerned about. Um, so I I think there’s a world in which AI becomes integral to the the approval process and allows the approval process to occur much more quickly. Um, similar to what we just talked about, I think human human out of the loop is a little bit a little bit, uh, little bit out there, but, uh, never say never.

[00:19:59] Eoin:  Yeah. Fair enough. No. Fantastic. Like, in terms of the the long term vision, I mean, it used to be because of the reasonable reliability of the timing of an FDA approval process, how long things took, you’re to your point in a year, I know I could expect to go in testing and I can get to here. All of those compressing timelines mean it must be harder as a product leader to build a multiyear strategy or as any company to build build a multiyear strategy. So how far out can you look and how do you how do you mitigate for that?

[00:20:30] Megan:  You know, in terms of a a product strategy that I think is likely to be executed, I don’t really look beyond twelve months anymore. Part of that’s being in a startup environment where things are just inherently changing as you learn more about your business. Part of it is to your point with the technology, you know, exploding the way it is. Ideas for, you know, two, three years from now, they’re likely not gonna be great ones. Um, I think it’s still important to set a bold vision as a product leader, and I think the vision can still still have a longer life than that. Right? And, um, so I think the the vision doesn’t necessarily change. I think the level of tactical detail in a road map that’s, like I said, for me, more than twelve months out, it’s not a great use of my time to build it because it’s not gonna it really probably isn’t gonna be relevant.

[00:21:20] Eoin:  Yep. What what is the the vision for Pictor?

[00:21:24] Megan:  Yeah. So I think our our vision is for, you know, broad clinical adoption of virtual staining, um, and that that is a multiyear process. Um, and so I think beyond just the virtual stains that we produce today, there’s a number of different exciting avenues we could could go in. You know, there’s number of different things that we could use our virtual stains to screen for. So, for example, identifying patients who are most likely to have findings in downstream genomic sequencing, which is still quite expensive. Um, there there’s other things that we can predict from those stains, all kinds of quantification that we can add to them. So right now, we just give the stain, but we don’t necessarily say if a biomarker is positive or negative. That’s something that we could do in the future. So there’s really a world of possibilities. For me, one of the things that’s exciting is, you know, we’re really we’re really early in sort of the entry point of the workflow. So we have an opportunity if people are using our technology, I think, to really easily add on more and more and more and more, um, because we’re kinda getting the sample first.

[00:22:25] Eoin:  Mhmm. Yeah. That brings up a question as well. Obviously, uh, a lot of the health care system as we know it today is very compartmentalized, and there are very discreet touch points across the board. Do you think that is important to have those gates in order to kinda think through you know, whether it’s privacy or just, you know, stopping monopolies in terms of, you know, people who are able to manipulate a system? Or do we see that, you know, like you said, you guys being tip of the spirit diagnostics, should all of their down through treatment, etcetera, live under one roof, do you think?

[00:22:57] Megan:  You know, it’s it’s interesting because there’s advantages and disadvantages to both. Right? If you’re purely thinking about efficiency of diagnosis, you really wanna break down those silos. Right? Because that’s what’s gonna allow the data to flow the as quickly as possible and, you know, get the treating physician all the information that they need to make the decision. And so I think, you know, that being said to your point, if you have one person that’s kind of monopolizing the entire value chain, do you worry about their incentives? And are they really doing what’s right for patients? And are they really being good stewards of health care dollars? Um, I think that largely right now, the the divide that that we live in is there’s sort of lab services, whether it’s pathology or blood testing or whatever else it is, And then there’s sort of clinical services like the oncologist that’s actually sitting in front of the patient, you know, discussing their treatment options. Um, I think keeping that a little bit siloed is is not a terrible thing. Um, I’ve worked on some products where kind of the physician product by name, that’s what they receive even if that may not be the best product for the patient. So I think, you know, this this tension, I think, is gonna continue to exist. I think there’s probably a few too many barriers right now, but I don’t think we need to kind of smash the system and get rid of everything.

[00:24:13] Eoin:  Yeah. Ultimately, if we can somewhat democratize access to the data, then each of the the people servicing different components of that data would be great, but that’s that’s the big. Thing. That’d be utopia.

[00:24:25] Megan:  Right? Exactly. And I think that your point on democratizing is so important because that is one of the huge values to me of AI and health care is now somebody who, you know, is in a small town, Their physician’s gonna have access to more information on a population level than they’ve ever had before. So patients don’t necessarily need to travel, you know, thousands of miles to go to MD Anderson or Stanford or wherever wherever their cancer center of choice is. You know, they can receive that same high quality care where they are. Um, that’s gonna be huge.

[00:24:57] Eoin:  Yep. Yeah. That’s, uh, as they say, it’s, uh, it’s not equally distributed by a long shot right now. And and I think all of this, we’re kinda skirting around as well, but cost is in implicit in all of this, particularly in The United States, but, you know, generally across the globe that the ideally, we can relieve some of that burden, you know, through all of that, and that’ll be a great effect for, ultimately, the the people. So Yeah. Patience. Fantastic. Quite a little bit of a tongue in cheek question because we just basically said we can’t possibly look too far into the future. But, uh, if I was to ask you to kinda hypothesize of where we will be with AI in health care over the next five to ten years, what do you think? Where where are we gonna see the greatest transformation in that realistic timeline?

[00:25:43] Megan:  Yeah. I mean, I think right now, AI is kind of the exception. Right? There’s a number of products out there, and there certainly are physicians and practices that have adopted them, but I don’t think it’s really front and center in the day to day practice of medicine. I think by the time we get to ten years, it really will be. And I think it comes in the form of, you know, AI diagnostics. It comes in the form of you know, what we what we were just kinda mentioning, being able to aggregate data across, you know, all patients that have the same diagnosis, really using that to help physicians drive treatment decisions. So when that’ll happen, you know, whether it’s five years or ten years, hard to say, but I I think that it it’s gonna move from being this thing that doctors hear a lot about and don’t really do much with to something that’s front and center about the way that they practice medicine.

[00:26:31] Eoin:  What do you think is the biggest barrier to that happening?

[00:26:35] Megan:  Um, I think one is is regulation. Right? These products are gonna have to go through some amount of regulatory scrutiny. Uh, I think the second is there are still barriers around data that we have to keep breaking down. Um, health care data’s are datasets are not always interoperable, and that’s stopping some of this. Um, I think that we have to also continue to encourage physicians to understand that there’s no AI is not coming for their job. Right? The goal is to supplement them, to give them more tools to make the decision. I think we all understand there’s nothing that replaces the the physician sitting in front of the patient and, you know, hearing what they care about, learning about their life, and then understanding how to guide them through this process. So I think there’s, you know, data barriers, the regulatory barriers, and then the just physician behavior barriers that we have to kinda push through.

[00:27:26] Eoin:  Yep. Yeah. A lot of physicians are still in pen and paper. Right? And to and barely at that, but I’ve seen some of the scribbles. So Yeah. Yeah. Fantastic. Uh, and I guess then the last question on on this area, and we’ll move into maybe some stuff on leadership in a moment. Uh, but for you and, obviously, you’ve had a hugely impactful career in health care and technology. What keeps you motivated every day, and how do you stay grounded amidst all of this, uh, the the tempo of the the world you’re living in right now?

[00:27:58] Megan:  Yeah. I mean, I think for me, I’m a very mission driven person. I’m I’m in health care because I want the work that I do to positively benefit people. And, um, I’ve been really fortunate in my career to to work on a couple of products that have had huge impacts on patients’ lives. So when things are really tough, that’s what I try to remember, that somewhere out there, there’s a patient that’s gonna benefit from this technology. And, um, that’s really, really important to me. I I remember I was at home visiting my parents, you know, a few years ago, and my mom’s Cologuard collection kit arrived at at the house. And it was a real moment where I got said, gosh. Like, that’s something I worked on, and now it’s here, and it’s, you know, helping someone that I love. Um, so it’s moments like that that I think also kinda keep me keep me grounded and keep me motivated because there really is a a human on the other end of all this work.

[00:28:47] Eoin:  Fantastic. Now what a great moment to to see that arrive, I’m sure. Oh, fantastic. Changing gears to just talking about you as a as a leader, less specific to to AI and and the medical space, but have you had or do you do you pinpoint anywhere, like, a pivotal moment in your career or your leadership journey? Uh, what changed? How did you navigate it?

[00:29:09] Megan:  Yeah. I I got this feedback maybe seven or eight years ago from from a manager who’s still my friend. Um, but he told me that I had single-minded focus. I didn’t care who I left behind, that I was gonna achieve my goal no matter what. I said, gosh. That doesn’t sound so great. Like, I’m gonna leave people behind. Like, that’s that’s not not terrific. And so I really, like, use that as a moment of self reflection to say, like, yeah. The goal is important, but how you get there is also important. Mhmm. Um, and so I think, subsequently, I’ve focused a lot more on how do I give everyone on the team time and space to express their opinions and also to just kind of keep up and make sure that they understand what we’re doing, why we’re doing it. They’re on board with it. And so I think that was just a moment a moment in time where I said, you know, maybe I’m not behaving as the leader I wanna be, and how how do I do that more? And I think it’s been really, really helpful. I feel like people, uh, respond more positively when they feel more more connected, and they don’t feel like they’re gonna get left behind. So that that was kind of a huge huge turning point for me.

[00:30:16] Eoin:  Well, well, big reflection on it. I guess, do you have systems or frameworks you use to help you be consistent in that manner? Or has it become has it become a habit by now, I guess?

[00:30:25] Megan:  It’s become a habit by now, and and I think I didn’t even necessarily realize how much it had changed until, you know, a couple years later, someone gave the feedback. They’re like, wow. You’re really patient with your teams. And I’m like, patient? That doesn’t really sound like me. Um, so I realized that, like, somehow I’d internalize this notion of, like, okay. We’re gonna gonna have these discussion meetings. I’m gonna give people time and space, then we’re gonna make a decision. So it really just became a habit.

[00:30:51] Eoin:  Yep. Fantastic. Uh, that type of change is not easy, particularly in like, we talked about the pressures of your field, feeling that pressure on behalf of the patients that you’re servicing as well as on the business, then moving a thousand miles an hour to be able to keep bringing everybody on that journey with you is, uh, it’s important, but easier said than done. Yes. Yeah. Oh, um, on the subject of teams and and feedback, do you have a particular belief or strategy in how you build or scale teams? I’m particularly interested if you feel like you have a an unconventional, uh, an unconventional, uh, take on this.

[00:31:25] Megan:  Yeah. I don’t know if this is unconventional or not, but I’m I’m sure it’s controversial to some people. I really try to focus less on pedigree and more on, you know, people’s core skills and values. I’ve been in some organizations where there’s a huge pressure to sort of hire people out of feeder schools, and I’ve never really felt like that was a successful strategy for me. I think it’s really, for me, about finding people who have you know, who at their core are aligned with the mission are, you know, people who are problem solvers, people who accept feedback well, and really targeting those behavioral attributes, um, rather than than focusing on, you know, who their mentor was or where they went to college.

[00:32:07] Eoin:  Yeah. I’m completely aligned with you on that one. I I’ve hired for skill in the past and whipped on values and been sorely disappointed and gone the other way around and hired on values and trained on skill and never been disappointed. You know? It’s a you gotta you gotta all be rolling in the same direction at a fundamental level, and and that that you can’t fake that. You may be able to fake it through an interview process, but you can’t fake that in the job for Agreed. For tariff at all. And I think you touched in there on, you know, taking feedback well and and really what you’re alluding to a growth mindset as well, right, which I think is a a crucial part in the here and now as we touch on AI and everything else that you have to feel like your, uh, incompetence in any any given field is a is a blessing, and it’s an opportunity to learn rather than a than a failing, especially right now.

[00:32:49] Megan:  Absolutely. You can’t be scared of change. Right? Especially the smaller the company, the more true that is. Right? Because the company is gonna change. And, you know, I’ve had this conversation sometimes with, you know, early career employees who go, but my job description says this, and I’m like, that’s awesome. But at the moment, like, we need something a little bit different. And so I’m asking you to grow and respond to this change. And some people really rise to that, and other people, you know, really struggle with it. Um, that’s okay. That just means a different environment will be a better fit for them.

[00:33:20] Eoin:  Yeah. I can’t remember whose quote it is, but I I love the line of, I look forward to looking back in a year at how stupid I am today. Right? It’s like I feel like that. Constant mindset of of growth and learning, and it makes you enjoy the journey. But, no. Fantastic. Uh, when you’re looking to recruit, then, I’m hearing you on lining up on values, a growth mindset, and problem solving. But how do you identify the great from the good? How do you make that distinction as you’re going through there?

[00:33:48] Megan:  Yeah. I think there’s there’s a couple things that I think people who are really great, at least in my field, do better than other people. Obviously, there’s this core ability to sort of break down a problem and identify multiple possible solutions to it. I I think there’s also the ability to sort of act without complete information is hugely important, especially as we’ve been talking about how fast things are changing. You know, you have to you have to be able to act without complete information. And then I also think it’s it is sort of the rapid integration of information. So as things change around you, how do you kind of update your framework? How do you update the way you’re approaching problems and the solutions you’re proposing? So I think it’s it’s, you know, a mix of things, and and it’s hard because it’s kind of an intangible. Right? It’s very difficult to measure. And for me, you kind of I kinda have an inkling when I hire someone that that’s special. But until you kinda see them in action, it’s hard to know if they really, really have, you know, the the motivation and the intention to pair with those, like, core capabilities to really, you know, be a transformative person within an organization.

[00:34:57] Eoin:  Yep. I’m sure we’ve got, uh, many failure and success stories in in all of the above. Right? Yeah.

[00:35:03] Megan:  Everyone does.

[00:35:04] Eoin:  Yeah. I I wouldn’t have a career in in executive search if it was a perfect science. You know? So, yeah, it’s, uh, it’s a different thing. But, uh, fantastic. Um, with another change in in topic here slightly, but with all the noise, no more we touched on AI a little bit, but whether AI or otherwise, what’s a trend or a technology that you think right now is overhyped and, conversely, one that you think is underhyped or underappreciated?

[00:35:31] Megan:  Yeah. I think something that I think is a little bit kinda overhyped right now, at least if the advertisements on my television are any indication, are these at home testing services. Um, I love the idea of people being able to to bring more of their their health care journey to their home, um, but I just haven’t seen them work yet. And it’s both the reliability of the result, but also the process of collecting a sample. I actually sat down with a colleague, you know, in a conference room to try to collect one of these at home blood samples. We got the blood everywhere but in the collection tube. I mean, we made a huge mess. And I just look at it and goes, gosh. This is just, like, not ready for prime time yet. And you can imagine even if I got that sample in the tube, um, it would probably be a very low quality. So I’d say I’m very kinda dubious about this particular portion of the health care industry right now. Um, I think in terms of something I think is a little bit underappreciated still, I think we’re really not yet exploring and exploiting the full power of wearable devices. So there’s certainly a lot of companies that are producing data and interesting algorithms on wearables, and I think, you know, people are using them. I think the problem is there hasn’t been an effective way yet to integrate most of that data into medical practice. Um, it’s just it’s just noise for the physician. And so I think there’s this huge opportunity to figure out how do we make those data that people are collecting at home just as part of wearing their Apple Watch or, you know, monitoring their exercise on their their Garmin device or whatever it turns out to be, packaging it up for clinicians. Because I think right now, we’re missing so much because of the episodic nature of of going to the doctor versus the continuous stream of data that’s part of our real life.

[00:37:16] Eoin:  That’s episodic, but also people are prone to tell stories. You know? And they’ll make up their own health data. That data won’t lie on. But I think you talked touched on this already when you talked about the physician in front of the patient. A lot of what’s happening in that moment is data capture. It’s lifestyle. It’s diet. It’s a lot of stuff that ideally could be automated, like you said, not spun in in such a way as to to suit the teller.

[00:37:37] Megan:  So Exactly.

[00:37:38] Eoin:  Yeah. Yeah. Fantastic. Uh, do do you have a wearable or wearables yourself? Like, what’s your go to?

[00:37:44] Megan:  So I am a terrible example because I tell you how important this is, and I don’t do it. Um, I can’t stand wearing watches. Like, I, um, half the time, I don’t even know where my phone is. But I know so many people that are devoted to these devices, and I’m like, well, what do you do with all this data? Like, you’re super excited about completing your rings or getting your 10,000 steps a day, but, like, what actually happens with that?

[00:38:08] Eoin:  Yeah. Fair. Yeah. I was the same. I had the Apple Watch for a while, and then I just found it quite a distraction. And so I got rid of it. But I think we’re due, I think, uh, and again, a step change and potentially with AI, We’re seeing miniaturization. We’re seeing lots of different applications of what a wearable means anymore. It could be clothing outright at this point. So I think those opportunities should should improve. We’ll see. Excellent. Uh, couple other quick fire questions here for you. And if it’s the same answer as earlier, no problem. We can admit this piece. But what is the most impactful lesson you’ve learned in your career that you wish more leaders understood?

[00:38:45] Megan:  Yeah. I think for me, it’s that leadership is not something you do just because it’s the next step or it pays more money. Like, leadership is something that you do because you know that you can inspire others to make a bigger impact. And so I I see so many people that embark on a leadership role without that mindset, and I feel like it usually doesn’t go great.

[00:39:08] Eoin:  And how do you distinguish leadership versus management, or do you?

[00:39:13] Megan:  So, you know, to me, they’re they’re interrelated concepts. But, you know, to me, leadership is a lot is a lot about vision, um, but it’s also a lot about, like, the care and concern that you have for the business. Right? You become a steward of the business, of the investor’s money, of, you know, all of the time and effort of the company. Um, and so, you know, to me, management is, you know, largely functional and tactical, um, whereas leadership is a lot more about, you know, care, concern about the business, strategic direction, and vision.

[00:39:47] Eoin:  I like that. The word stewardship in there as well. That’s yeah. Great way to great way to frame it. Excellent. Uh, what’s one tool or book or framework that you think every leader should know?

[00:39:58] Megan:  Well, I I grew up corporately at GE, so I, uh, I I’ve spent a lot of time reading all the Jack Welch books, and I think I’ve been heavily influenced by a lot of the principles in the Jack Welch books. Um, and I think, you know, regardless of how GE has ebbed and flowed as a company, I feel like there was something really magical about that period of leadership. Um, I think there was a lot of focus on having a vision, but really holding yourself accountable to execution, um, really focusing on creating a meritocracy even if it was a little bit brutal at times, but really judging people based on the quality of their work, um, you know, focusing on simple messaging that everyone’s gonna understand, uh, being comfortable with change and pushing for innovation. And I think those are really, like, timeless messages and tools. Um, so that’s kind of where where my leadership inspiration has often come from.

[00:40:50] Eoin:  Fantastic. Well, everything you just described, I think, comes through in the way you’ve talked today. So security’s stuck and Yeah. And it’s something you say you get it’s a habit for you. It’s it’s and it’s embedded. So, yeah, fantastic. Uh, I understand as well you’re an avid traveler, so not a question I ask of everyone. But, uh, I I would hate to make you pick a favorite, but if you were recommending a a destination, maybe one that’s less known, where would you recommend for us?

[00:41:15] Megan:  Gosh. That is a really tough question. I I’m I’m a big adventure traveler, and I love wildlife. So for me, it’s all about going to to the best places to see animals. Um, and so probably one of my favorites was I went to Borneo to go trekking with orangutans in the in the jungle of Malaysia. And, um, there’s huge ecological devastation in the area, which is very sad, but the moment that we we’ve been trekking for probably four or five hours in the rainforest, and all of a sudden, this mom and baby orangutan just, like, appeared out of nowhere. And it was just such a magical moment because you know how rare they are, um, and to just get to have that experience and have it be, um, something that wasn’t manufactured. Right? There was no guarantee we’d see Uh, and that’s what I love is I love going places where there’s honestly no guarantee, um, where you have to you have to work for it, and you have to be patient, and you have to really want it. Um, so that was that was a really fun a really fun adventure.

[00:42:14] Eoin:  Phenomenal. No no Apple watches or cell phones there either for that one.

[00:42:19] Megan:  That’s also great. The boss can’t find you.

[00:42:22] Eoin:  Oh, that’s fantastic. Like I said, life is so devoid of genuine surprises anymore. That type of experience is is gotta feel so much more satisfying.

[00:42:29] Megan:  Yeah. I

[00:42:29] Eoin:  love it. Fantastic. Alright. Um, well, great. I think, uh, actually, one more question. Uh, in a word, if you can, what does great leadership mean to you?

[00:42:41] Megan:  I think it comes down to a word that we talked about a few minutes ago, which is stewardship. Right? That for me is what it’s all about. It’s about that sort of servant leadership mindset, the notion that you’re preserving and protecting something that isn’t really yours, and you’re just there to make it better in the time that you have. And so that’s really that’s my my leadership word.

[00:43:02] Eoin:  I love that. Yeah. I’m a long time rugby player, coach, etcetera, and one of the philosophies of the All Blacks rugby team is leave the jersey in a better place. Right? It’s that concept of, you know, you’re just you’re taking it for a time, and you’ve gotta pass it on somewhere better to to the next person.

[00:43:17] Megan:  Exactly. Same same exact idea.

[00:43:19] Eoin:  Yeah. I love that. Fantastic. Uh, well, is there anything else, Megan, that I haven’t touched on that you wanted to make sure we talked about today that you’d like to expose whether for picture or otherwise or we could bear?

[00:43:30] Megan:  Just it’s it’s been fun chatting with you and, uh, it’s always fun to speculate about the future.

[00:43:37] Eoin:  Oh, yeah. Like I said, we’ll we should write these down and put them somewhere so we can look back and see who was right, but I guarantee none of us will be. So not on this AI case. So, uh, yeah.

[00:43:46] Megan:  Things are things are just changing so fast right now.

[00:43:48] Eoin:  Yeah. Well, it’s fantastic, especially for your field. I think it’s it the promise is immense. And I think it’s the area I’m most excited about where AI is being applied is is there more than anywhere. You know?

[00:43:57] Megan:  So Yeah. It’s it’s gonna be really exciting.

Outro: Signal to Noise is brought to you by Riviera Partners, leaders in executive search and the premier choice for tech talent. To learn more about how Riviera helps people and companies reach their full potential, visit rivierapartners.com. And don’t forget to search for Signal to Noise by Riviera Partners on Apple Podcasts, Spotify, or anywhere you listen to podcasts.