The Future of Medicine Is Already Here w/ Bertalan Mesko, The Medical Futurist Institute

For more than a decade, Bertalan Meskó has been writing about a future in which technology fundamentally reshapes medicine. Now, with agentic systems, consumer wearables, and ChatGPT style tools landing in clinical workflows, the future he sketched out years ago is arriving. We are now asking whether the digital transformation will make people healthier or just add new layers between doctor and patient along with new costs. Hi, everyone.

I'm David Williams, president of strategy consulting firm Health Business Group, and host of the Health Biz podcast, where I interview top healthcare leaders about their lives and careers. My guest today, Dr. Bertalan Meskó, physician, geneticist, and founder of the Medical Futurist Institute. He's known to his global audience as the medical futurist, and he helps make sense of the technologies reshaping medicine.

Bertalan, welcome to the Health Biz podcast. Hi, David. Thank you so much, first of all, for having me. It's a wonderful podcast, and I was very much looking forward to it.

Outstanding. So you've been writing about the future of medicine since before most people had even heard the term digital health or, or, or, you know, even since before we were using that term. So I wonder, how did you become the medical futurist, and how have some of your early predictions panned out? Especially because when you're talking about the future, if you live long enough, eventually you get to see it.

As a professional futurist, I always feel like I'm being poked with the expression prediction, but I, I will get into that. Very briefly, I, I knew exactly from the age of, like, five or six that I wanted to become a scientist, and my, my first decision was about going into medicine so then I can give - I can become a clinical scientist. And genetics was my first love, uh, during my high school years. You know, we are talking about the, uh, late 1990s, early 2000 years when genomics had a huge boom.

The whole genome sequencing projects came out. The Human Genome Project was finished, so, um, it was very exciting to get into that field. So even before I even finished medical school, one and a half before that, one and a half years before that, I already started my PhD, uh, program, and the whole PhD was about genomics. Uh, our major aim was to find out if it's possible to predict the responsiveness to biological therapies in certain autoimmune conditions from a simple blood test using gene expression profiles.

So it was very much life sciences and, and a big part of genomics. But as soon as I finished my PhD, I felt that something was missing from my research life, and that missing part was, was my geek self, which I've been, you know, since my - since being a kid. I've always worked and, and lived with technologies and devices. I had Amiga computers and all the personal computers I had, I all built them myself.

So technology always played an important role in my life, and I felt that in my genomic research, and I was trying to extrapolate for the next, like, five, six decades working as a scientist, being optimistic, of course, here, I felt that that part would be missing. So what can you do if you cannot fill into any known categories of life science scientists? You start finding out what else can you do that would merge these, like, different personalities, would merge my, my clinician personality, my scientist personality, and my geek self.

And as it turned out, future studies, so the, the science and research of, uh, potential plausible possible futures, that's the one where you can merge these different fields. And actually, future studies seem to be that sub part of scientific research where I can, on one day, I can zoom into a very specific field, like analyzing the impact of artificial intelligence on a certain medical specialty, but on the same day, I can zoom out and look at healthcare from a bird's eye perspective.

I- in what else fields or research focuses or interests can you do the same? So I thought, that's exactly my field, and I will try to bring in clinical life science knowledge into future studies, but nobody seemed to be doing that before. So I had to brand it to make it easier for people to understand what I am doing, what we are doing with my colleagues now and, and PhD students and, and team members around the world. So I branded it The Medical Futurist, simply implying that we are merging two fields, the evolution of technological and cultural changes in medicine and healthcare with the amazing methods of future studies.

And I'm sorry for being so long about this re- response, but about the second part of your question about the predictions, I'm, I, I have been fighting so hard, you know, against the idea that it, it doesn't make any sense to make predictions because making predictions implies that there is one singular future. There's one timeline, and myself as a professional researcher, a futurist, I should have a better knowledge than, than you do about what's gonna happen next. But that's, that couldn't be further from the case.

In fact, multiple futures exist, and what we futurists can do is that we can use a bunch of established systematic methods coming from future studies for observing, discovering, and analyzing these futures. So I'm much more interested in highlighting trends for business healthcare policymakers than making predictions for them. I'm much more interested in helping those decision and policymakers finding out the desired future for a field of interest or hospital or medical specialty than telling them what I think is going to happen.

So still answering your question, there are a few things that I, I think are clearly happening now, which, for which when I used to mention them in my keynotes more than a decade ago at medical events, um, I used to spark some debates, such as patient empowerment, the idea that patients are becoming members of their medical team, the idea that digital - the end of the digital health story is how patients are becoming the point of care. And now I think part of these, again, predictions is that artificial intelligence as a technological entity will join that medical team now consisting of patients and all sorts of medical professionals Well, I think you did a great job of answering the question because I, I, it was a difficult question, and sometimes when you ask someone, like, to summarize their PhD thesis, you know, it's like you're really asking for trouble.

And if I mischaracterize what a futurist is, that makes it even worse, and you brought it back to the predictions because also, uh, although the, uh, being a futurist is not about predictions, it doesn't mean you're not allowed to ever make a prediction, uh, so that you don't go off-brand. So, uh, thank you for that. I will consider it a very concise answer to the question, and apologize for my role in perpetuating any misconceptions about, uh, what your role is. And David, please, um, I didn't want to, you know, say these things about that, but it's very important that we just put this discussion on the table that future studies are not about predicting the future, but about analyzing those potential futures.

And you might think it's a, you know, very simple thing to say out loud, but I, I come across this discussion so often talking with business leaders, government officials, even patient and physician leaders, that they think, "Well, I should be able to tell them what's gonna happen next." And I'm happy to tell them what I think could happen next, but I also tell them that it doesn't matter. Because what if, uh, nine out of 10 of the predictions I make would turn out to be right, but then if one is wrong, and, you know, by statistics I will have wrong predictions, then what do you do with my next prediction?

Well, what kind of role will it play in your decision-making process? And hopefully you will have doubts. And that's why in the last, like, two years or so, one of the major efforts or, or focuses in my whole research and science communication efforts have been about focusing on how to bring futures methods for the masses, so then anyone can start using these futures, at least some of the basic, of course, futures methods. So then we could talk about the future in a different sense, almost with the same, like, um, um, consistency and self-confidence as we have when we talk about the past.

Because we all study history lessons and we all talk about legends and stories and, and from, from our childhood. So we have quite an experience talking about the past, but not so much about the future. So I've been trying to change that while also focusing on, of course, the future of medicine and healthcare. So it's absolutely not your fault, it's … I think it's a population-level issue that we have almost zero relationship with the future.

So let's talk about some of the digital health trends and technologies that are exciting today. Of course, AI has gotta be one of them, but sometimes it feels like, like it's really the only thing. Are there some other non-AI technologies that are important today? Absolutely.

I think some trends are not even technological ones. Just the idea that, that now patients and physicians and technological entities will work in one team, I think it's quite unprecedented in the history of medicine. But if I have to pinpoint one particular technological trend, then recently I've been quite excited about, uh, PPG, photoplethysmography. You know, the technology that our smartwatches use to measure heart rate and blood oxygen levels, but the same technology could be used to measure a few other quite important health parameters, vital signs, or even biomarkers from blood glucose levels to blood pressure.

And just the idea that we could shape blood pressure management by not giving patients digital blood, uh, pressure monitors, but they could wear chest patches or they could wear their smartwatches without even, you know, realizing that they are wearing it, and the smartwatch or the chest patch could measure blood pressure continuously. And there are now existing examples for each. Even some of them are being proven through peer-reviewed studies. Now, that would be quite, um, a, a, a life-changer for actually hundreds of millions of patients dealing with high blood pressure.

So PPG, I think, shows more promise than, than, um, we normally think. Great. Okay, that sounds, that sounds interesting. Let's come back a little bit to the discussion about how technology is changing the relationships between physicians and patients, and it comes back to some of your earlier, if not predictions, discussions, uh, about the empowered patient and so on.

Now, a lot of what's happened in medicine can have some unforeseen circumstances, at least in the US healthcare system, where what seems to be, wow, that could, you know, that could reduce cost or improve quality, may just be a business opportunity for someone to insert themselves into the system to add cost and not necessarily, uh, be looking after either the patient or the physician. So how do you see, you know, the, uh, the role of, of technology in the evolving doctor-patient relationship, and are we headed more in the good direction or a bad one?

I think it's necessarily a good direction because it's, it always - So we al- when we talk about the role of technologies in healthcare in general, I always feel like the discussion is about whether we choose to reach out to this influx or range of advanced technologies because we have a choice. We can decide not to reach out to them, but that's not the case. So, healthcare today faces a very basic mathematical problem. About 6 million healthcare workers are missing today.

The World Health Organization shared that data a few years ago. And by 2030, 10 million healthcare workers will be missing worldwide. So, we will simply, we, we can conclude that we will never train as many healthcare professionals as we need, while the number of patients requiring our constant medical help will keep on rising because, not because we are getting sicker, but because we're getting better at diagnosing patients and monitoring them and treating them on the long term, especially in the case of chronic conditions.

So, while healthcare is improving, that number is going to get bigger. There is a huge niche between how many physicians and, and medical professionals we can train and how many patients need their help. That gap cannot be filled in with simple HR tricks. That gap can only be filled in with advanced technologies.

So just, I wanted to make sure that we talk about this issue from that perspective that, of course, it's always better for a patient to be able to meet a physician in person, and they can build a relationship, you know, using empathy and compassion and trust But that is becoming, gradually becoming a luxury for most of us worldwide. So instead of that, technology could be the lifesaver, but it's extremely hard to include that in that process because we are talking about, you know, patients' lives are at stake, first of all, and then we are talking about people having relationships with other people, and technology always has a very complicated role in that.

So that's why, uh, for about one and a half decades, we have been publishing papers, and I've been trying to, you know, use all my channels in science communication to talk as much about it as possible that why we are witnessing an influx of advanced technologies, at its core, digital health is a, is a cultural con- is a cultural transformation of healthcare. Because the way the traditional hierarchy of the doctor-patient relationship has been transforming into an equal-level partnership, which is, which is coming with its own rules and, you know, new ways of forming relationships and building trust.

So all this is happening because of technologies, but the cultural component of the change is much more impactful than which variable sensor, smartwatch or AI algorithm comes out this year. And the, the sooner I think healthcare systems, governments, business leaders acknowledge the power of this cultural transformation, I think the better decisions they can make. Just to give you a real-life example, I've, I've, I've given countless, countless, uh, keynotes at medical events to my peers, and around 15, 10 years ago, I had so many fights, uh, with them and debates about why they thought that I was talking about AI and digital technologies taking over their roles and their responsibilities, while it couldn't be further from the case.

The fact was that they, that they had fears, and these physicians rejected the use of advanced technologies, not because they thought that based on peer-reviewed studies they chose not to use them, but because they had primal, uh, feelings about them, emotions like having anxiety about using those technologies or fears about privacy or even fears about being replaced in their very important doctor-patient relationship. So in these cases, we cannot just focus on pushing the technologies on them, but we have to focus on finding out why they reject the technology, why they have anxiety about that, and then working on those parts before even discussing any specific technologies in that relationship.

And one of my PhD students, uh, published a study about, uh, all these emotions from fear through anxiety, uh, of… and even resistance and rejection, and how the use of digital health technologies and AI can create these emotions in physicians and what we can do about these. So I think if, if we can focus on the cultural components first, then we get into a better position of finding out which technologies to include in the doctor-patient relationship that will hopefully improve it and not diminish it.

It's very interesting if you go beyond even what you're describing to the, to the broader perspective and think about how there's this shortage that we always hear about, uh, in medicine of physicians and, and others, whereas in other fields there's a concern about, you know, unemployment, AI's gonna take everybody's job, where it's really, "Hey, actually we're looking to fill some jobs here." And one way that, that the jobs have been filled in the rich countries, in particular the US but also in, in Europe, is through immigration, which on the one hand is great, but on the other hand some people are moving from, you know, middle income or lower income countries to the US and it's exacerbating the shortages that they may have somewhere else.

And so the kind of the whole tie with immigration policy, technology, culture, you know, patient expectations, I think is, uh, is very, it's very broad, and you filled in an important part, uh, related to culture there as well. You know, almost a decade ago, there were investors and, and, um, technology company leaders, and I, I don't want to name names for a reason. Y- I think you guess who I'm talking about. And they had predictions, very, very much predictions about how radiology would get obsolete in a few years, how AI would replace that - every physician in that medical specialty.

And then, um, there is a meme, you know, about how self-driving cars will become a thing and how AI will replace radiologists. And in the newest picture, it's 2026, and radiologists are going to work in their self-driving cars. Because that shows exactly what has been happening to that specific medical specialty, which actually has many tasks that will be prone to automation, because some of those tasks are either data-based, image-based, or repetitive. So d- those tasks are prone to automation, but it doesn't mean that the medical specialty will get vanished from the palette of medical specialties.

It, it only means that radiologists will have to do fewer and fewer tasks which are repetitive and data-based, and some of their precious time will be freed up to work on even more AI-based technologies that can help them do even more jobs. So they t- these predictions described how the specialty would get obsolete while it's in the, I think, top three, uh, the most wanted residents of all the medical specialties, uh, in the Uni- United States, and one of the best paid medical specialties, too.

So obviously, it's not going away. If not, it's becoming one of the top, um, like, um, success rates among all the medical specialties. So it's, it's very easy as an investor, as a financial or technological person, to come up with predictions so easily because indeed, if you check the FDA's list of all the FDA-approved and cleared AI-based medical technologies, you will find that 81 - 85% of all of them are focusing on radiology, which might make you think that, "Oh, so then that specialty will get replaced."

But the fact, it's, it's, uh, it is - the reality is more about how their individual tasks will change over time. So then maybe we have to start teaching prompt engineering to radiologists very specifically. I think to every other physician, too, but specifically radiologists because they will start working with more and more generative AI-based, uh, interfaces because they will keep on developing AI that can do, that can take care of even more tasks and scans and will flag those that should chi- still be checked by, by human physicians.

So it's called partial automation, and radiology is the prime example for that. So as long as we're picking on radiologists, let's continue because I think there's a lot to say culturally there. So when, um… And I'll, I'll pick on radiologists even though some of my friends are radiologists. So, so one thing I always joked about is when you meet a radiologist, they usually tell you how much money they're making and when they're gonna retire.

And then there's also, uh, quick to, uh, look at the threats. Uh, before AI it was about, you know, other specialties doing imaging and why they shouldn't, uh, do that. And the radiologist generally doesn't have a relationship with the, with the patient so much, so it's more, uh, you know, if you want doctor-patient relationship, people are rarely talking about their relationship with their radiologist. But then, and you know, you talk about it being an attractive specialty, i- we talk about the ROAD specialties, I think radiology, orthopedics, anesthesiology, and dermatology, which are high-paid and you don't have to work that hard.

So those are s- those continue to be sought after. But The, the, and then this whole element as it ties into job loss and the type of analysis you're talking about. So you can - you start with these predictions and say, "Well, 70% of the job can be done by technology and, and therefore 70% of the jobs are gonna be gone." Not necessarily, right?

It's, it's, it may be that the job changes and maybe even you need, need more people. So I'm glad you brought up radiology. Absolutely. You, you mentioned that you don't have to work that hard.

I have a, a friend who is an anesthesiologist and he works very - He just came into my mind. Of course, I know what you meant by that. Um, absolutely, that's the point, that even though let's say that actually 70% of their tasks could be prone to automation, it doesn't mean that what's left is nothing and they will lose their jobs. But it means, well, they can use their absolute medical ex-expertise and experience to work on something.

So maybe we can do three times more radiology scans in the next five years because AI will take care of so many relatively easier tasks that are, you know, low risk decisions. And of course, AI will keep on getting better at that without taking breaks or without having bad days. Plus, they will learn from every single mistake they make, and no human brain can physically compete with that power and efficiency. But in the meantime, radiologists will work on other AI-based technologies.

Plus, I think it opens so many new doors. Like in, just in the case of the MRI machine, it would be possible to just a patient go through the machine in five minutes instead of spending 25 for like a heart MRI, and the machine could analyze the scan, but there would be no image to be analyzed. Because the - actually, AI doesn't need the image to be seen, it just needs the data and the spin directions coming out from the, from the machine. We people need the scan so we can look at something that we understand.

And for me, far the most exciting aspect of using AI in these specialties is not just how it changes certain tasks and, and how it replaces repetitive and database components of their jobs, but how AI will find unusual associations we, you know, clinicians, physicians could have never found in other ways. Like, there was a study not so long ago showing that an AI could f- could detect a skin color from chest X-rays, and there were no data, no text, no des- details about the patient on the chest X-rays, just the X-rays themselves.

We physicians think it's impossible to be able to do that, but AI finds something that we don't understand or how. There was another other study showing that just looking at or listening into vocal recordings, phone recordings of patients, it could detect the risk of, uh, Alzheimer's and, and Parkinson's disease We thought maybe scientifically it's possible, but we couldn't, you know, find out the way of doing that. But AI finds something that we don't, and that's why I always say that, and even published a paper about why the real era of the art of medicine will come with AI.

That it's not that AI is taking away that art of medicine from us now, but when AI finds treatments, cures, you know, diagnostic procedures that we couldn't find for centuries in, you know, very smart people looking for those potential solutions in research and, and clinics and all these. And if we find something, then we will have to understand how it came to that conclusion that, that we couldn't find any other ways. And, and we will need the real art of medicine from knowledge in, you know, pathology, all the medical specialties, how AI works, to try to reason with the AI machine, finding out how it could find something that we couldn't.

That's, I think, the most exciting part of AI these days. That's interesting. I, I'm gonna, I'm gonna suggest something and you can say if it's, it's nonsense. So, you know, when you hear some of the, uh, like, old-school physicians talk about what things used to be like.

So before they had, you know, the CT scans and ultrasound and, and all sorts of just imaging and, and advanced tests that you could do, and sort of the art of laying on hands and being able to listen and hear things. And sometimes you hear the older physicians say, "H- today's physicians, they don't know anything except how to order a test. You know, they get the result." Are we talking about potentially getting to the next era?

So instead of the laying on hands or just saying, "There's a, there's an output and I'll just take it," to actually come back to kind of like reasoning, but, uh, like maybe a more advanced form of, of laying on hands and really having to engage with it in a more sophisticated way? I, I don't think so, and I don't think that should be the case. I mean, it's gonna happen, but I see it from a different perspective, not such a, an observation being applied to every medical professional.

But I think primary… And it might be a long shot, and that's now a prediction. I cannot, I cannot st- I yet, I cannot yet back this up with medical studies. I think primary care will be entirely AI-based. Simply not, again, not because we choose that it works this way, but because that's the only chance we have to keep on providing or receiving care.

So primary care will be a line of AI agents, chatbots, conversational agents, large language models, generative AI in general, working as the primary care physicians that we have today. And in the primary care line, everyone will have access to this because it's the easiest, cheapest, much easier to give access to a large language model than to, to help someone get access to a medical professional, either remotely or in person. And the secondary line will be what we call now primary care, and the third line will be what we call now secondary care.

Because everyone, a- as, as soon as a patient realizes that i- in every single parts of their lives, from banking to transportation, from commerce to, you know, ordering books online, they always reach out to digital technologies and everything is simple through that way. But once they realize that actually they can do the same about their own health. Now, ChatGPT has ChatGPT Health, Claude has Claude Health, so obviously these large language models started launching their own health-related versions Which cannot replace a physician, but let's be honest, when hundreds of millions of patients ask questions to these large language models about their disease, treatments, diagnosis, and health in general, then it makes more sense for those companies to develop health-related versions too.

Medical ones will be coming too, only trained on medical databases and, and medical textbooks. So it's understandable. And as patients, we reach out to information. This is the evolution we are looking at.

And as a futurist, I'm always interested in, in analyzing the evolution of technology. So very briefly, 20-something years ago, it used to be what I call doctor neighbor, that patients went home after a doctor-patient visit, and they discussed the, the medications they got, the prescriptions, the diagnostic procedures with their neighbors, their friends, their families. In the early 21st century, we got Dr. Google.

Everyone started doing searches on Google about diseases and treatments, and I'm sure you can remember how many debates and discussions there, there were about from physicians and professors around the world about why patients shouldn't do that, even though it was understandable that they reached out to the simplest information resource out there. And now it's the era of Dr. AI, when we cannot just search for information, we can have discussions about the information. We can bring in medic- the test results and, and discharge summaries and ask AI to help analyze it and, and help us digest the information in it.

We can ask questions. We can ask what questions we should ask at the next doctor-patient meeting. So now that's the evolution we are looking through, and it's inevitable that patients will keep on reach out to these. The question is how we physicians can deal with that, whether we understand our new role, which is not being a key holder to the ivory tower of medicine, but being a guide in this jungle of AI and digital information coming from wearables and websites and social media and search engines and all these.

And we have to act as the guide because patients will keep reaching out to us for help, for guidance, and we need to be able to provide them, uh, with that Can you talk about proactive medicine and the role of agentic AI in that? Absolutely. It's - I think most of the things I learned about being proactive in health, uh, is coming from, uh, e-patient Dave de Bronckart. He's the most famous e-patient in the world, a very good friend of mine.

He was just in Budapest, uh, at us, visiting us, giving a talk at the first scientific symposium about medical future studies, and he was my international top speaker at the end of a line. So we were ha-very happy to have as a patient giving a scientific speech at the Hungarian Academy of Sciences. And he always talks about proactivity in a way that, that physicians usually imply that if we let patients, then they can finally access information resources only we physicians could access for centuries within the ivory tower of medicine.

And e-patient Dave always tells me that they don't need our permission to have this autonomy over the information they can finally access. So I think it-it - from that sense, proactivity for me just means that, of course I reach out to information resources, to channels, to whatever data I can or insights I can obtain if it leads to me trying to be in a better position to improve my health or disease management. And proactivity from that sense is that I'm not waiting for a symptom to appear, and only after that go to ask for medical help at the physical point of care.

But I am proactive already. I want to live a long and healthy life. I want to dedicate efforts, time, privacy, money, uh, anything needed to that. But then I will need a medical professional to get sense of the data to help me find out how to use these AI chatbots To learn prompt engineering.

So it's a very complicated ecosystem, which used to be about, very briefly, of course, I'm trying to summarize, you know, 30 years of research in a few lines, but there was an ivory tower of medicine. Every information, data, peer support, second opinion was inside of it with medical professionals too, and the patients were outside, and they were waiting for a symptom. And when they needed medical help, when something was wrong already, already, no prevention here, then they had to enter the ivory tower.

They were told what to do, and then they left. And about half of them kept on complying with the treatment. You cannot show me an industry with the lowest, lower efficiency rate than what healthcare has in general. But now, because of proactivity, it's different.

It's patients who are leading the narrative. They are the ones reaching out to wearables and digital health tools and apps and AI chatbots, whatever information channels and resources they can have access to, and then they need our help in guiding through the process, the so-called jungle of healthcare, which is full of decisions, big ones about, you know, medicine, smaller ones, lifestyle choices day by day, and they need this guidance. So the medical team is still there, but, but the, the cultural components behind it have changed dramatically because now it's not, as a physician, I'm telling patients what to do, and then we don't meet again.

But it's we are sitting at the same table, we are looking at the same interfaces, a- and that's new to e- each of these stakeholders, and we have to find our new ways of building relationships and, and creating, you know, trust along the process You mentioned, uh, Dave deBronkart, e-patient Dave, and I remember the first time that I met him, he was just newly into it, and it was in the era of the first, uh, Google Health, and I remember he stood up at a, at a small conference where we were and he says, "I'm your worst nightmare, a patient with a blog."

And he started to describe how, I think it was the, Beth Israel, uh, had, had allowed his data to be exported to Google Health, and it analyzed it, and it saw all these different things he was being tested for, and which were ruled out. But it, it concluded that he had all these things. Um, so that was kind of the, the, the early days. But I, I do remember that and it, and it's good to know that he's, uh, he's continuing to, uh, walk this earth and to do well, 'cause that wasn't the path he was on, uh, originally.

So good for him. So let's go back and talk a little bit more about wearables. You already talked about, uh, you know, the potential, uh, for them for, for blood pressure. You know, early, when I'm thinking about like, uh, when you had, uh, blood glucose monitoring and patient might bring some data, and the physician didn't know what to, what to do with it even though they saw it was relevant, and they were also afraid to get more of a real time feed because well now, "Gee, if something goes wrong and it's out of range, I'm responsible for it."

Now there's much more data, um, and potentially even much, much more on top of that. How, how … Bring me up to speed on what's, what's changed since those early days of kind of the, the printouts of your blood pressure or your, or, or your blood glucose to now all the data we're dealing with. There are several things that are extremely different today than it was not 10 or 15, but even five years ago. First of all, sleep tracking used to be something for the geek and the nerdy people like myself.

While I think smart sleep alarms are living through a, like a golden age, and I still think that the smart sleep alarm is the holy grail of health tracking. If I, if anyone asks me, because I've been going through quite a longevity journey myself, having, um, I think more than 300 different medical tests and examinations and multiple blood markers and all these genomic tests, microbiomes and, and hardware-based devices too. If any - if someone wants to do something great about their health, start with a smart sl- a smart sleep alarm, because that will change their life.

They will wake up feeling energized because a smart sleep alarm will wake them up every morning from a light sleep and not from deep sleep. And there might be five minutes in between the two stages, but you cannot tell because you are asleep. Your smartwatch can because it analyzes data, and it can make that very simple decision. And then there is one gentle vibration from the smartwatch, and you're awake, feeling energized, no snooze is needed.

And so that's far the best option for everyone out there, and that has changed a lot, and I think it's, it's mu- it's much more widespread than it used to be. Second, about five, 10 years ago, uh, using wearables for health purposes was still, like, for a niche population. And now at The Medical Futurist, we actually checked, uh, based on data statistics worldwide, more than one billion people worldwide have at least one device that can measure a vital sign or a health parameter A third thing I think is about HRV, heart rate variability.

Even now when I wake up and I check my sleep quality, besides of course how I'm feeling in the morning, I always check my HRV because that's the best indicator of not j- not only that night's sleep quality, but how I'm doing on the long term too. The lo- in general or in very briefly, the lower your resting heart rate is while sleeping, the higher your HRV can be, and it, it's, it, um, gives a good picture about your fitness stage or level in general. It shows… It, I think it's the best number today about sleep quality, so HRV has become like the, the, of, of utmost importance these days.

I als- I also tell people that find a hobby or, you know, something you like to do during which you have a low resting heart rate. Uh, for example, for me, it's watching baseball. If I watch baseball, it's gonna, it's gonna have an amazing impact on my mental health and on my health in general because my resting heart rate will go down and I will just immerse myself in watching the Angels lose again, uh, day by day because, um, I, I was silly enough to choose that team some few years ago as my favorite team.

So that's it. So i, i- I think in general what has changed is the amount of insights we can get from these wearables, not just the pure data, because what do you do with the data? And even what can your primary care physician do with the data? I've, I've tried mine many times, and sometimes she told - my primary care physician told me that, "Look, I, I don't know the answer to your question about these wearables, but let's come and sit next to me.

Let's do a search online together. Let's find out the answer together." And I think that's what wearables can really do, that it can bring together physicians and patients and, and help strengthen their relationship, their bond, while the physician being able to say out loud, "Look, I don't know, but we can find out together." And I think the, the most, uh, successful wearables today provide not just pure data, but insights.

And even some of these have, um, uh, large language models built in, so I can even have discussions about my insights and data with those generative AI interfaces. So that's, that's the direction I think we are moving towards. You mentioned briefly genomics and the microbiome. What's the current status and, and how are those evolving in terms of the information you can get, the insight, and the actual difference it could make to a patient?

The reason why I'm smiling is that the field is, has evolved amazingly, and from a g- whole genome sequencing service, from a microbiome service, you can learn a lot. I learned what medications would cause me serious side effects, so I, I have to avoid them. I learned what, uh, medical conditions or cancer types I have a risk for, so I can take preventive measures. It's extremely useful.

But I'm smiling because when anyone is asking me, "Well, which service I should reach out to? Which genetic testing service I should use?" I cannot suggest them even one. I, I've had many.

I've had, I think, eight or nine genetic tests with different companies. Some of these since then went bankrupt or, um, uh, stopped providing health insights. Um, you know, 23andMe is a good example about having all these changes around their company structure. One UK-based company which I loved because the insights were very useful, turned out that their leaders, the two, uh, CEOs, left the country and brought all the data, all the genomic data with them, so we don't even have access to our own genomic reports anymore.

So there is not one service I can wholeheartedly suggest or recommend to anyone. Even though the field itself is … It's blooming and there is so much information to exploit from our genomic data, yet it's extremely hard for an average individual to get access to that, those interpretations Let's go back to something you, you talked about earlier about how the access to physicians, uh, just to paraphrase, is becoming kind of a luxury good. It's not generally available. And then I've heard you talk about, you know, 300 tests you had and eight or nine genetic, uh, tests.

And I - And so on the one hand, what I'm hearing is that, you know, those who are very sophisticated, have, uh, financial access, and know how to, to work with the system, it may be kind of an ultra-luxury, um, kind of approach. And those are maybe the same people who can access the, the physicians and leaders that have the new art of medicine. On the other hand, you're also talking about the sort of technologies that can democratize this by bringing it out of the ivory tower and giving an average person incredible access to information.

Um, how do we reconcile those, uh, two things? Are they both happening at the same time? Does one overwhelm the other? Absolutely.

It's quite the elephant in the room. But I look at digital health, AI, and longevity services from, from different perspectives. Digital health, of course, is the oldest out of these three, and in that, I think the, in the digital health revolution, the biggest issue first was not even about money, but about the ability to analyze the data. So the, the geekier y- you were, the higher a chance you had to analyze the data.

And if the system wasn't even working for geek people like me, who were willing to, you know, sacrifice privacy, money, efforts, dedication, and time into using these wearables, and even I couldn't benefit from using them in my general healthcare system, then how could I expect the same to happen to hundreds of millions of patients out there? But now it has matured a lot, and regarding a lot of smartphone apps and, and wearable sensors, at-home lab tests, I think many of these have become accessible to the masses.

Even some of these being reimbursed in the US, in Germany. So there are amazing examples for that, at least because that's where digital health is in the, in its own maturity process. With AI, it's a bit different. It's younger, but, uh, general- generative AI agents - not agents, the large language models are freely accessible and at, at about one billion people, I over- almost said patients, one billion people use them on a weekly basis But to access the, the paid versions or to ex- have access to very complicated versions, I don't see the benefit that much.

So it doesn't mean that if I can buy the biggest tools out there, AI tools out there, then I will benefit three, four, five times more in my health. That's not the case for now. So AI has bec- generative AI has become, since the ChatGPT, the launch of ChatGPT in 2022, quite a big driver of making this information source accessible to the masses. But longevity is different.

And I'm talking about not the kind of longevity that has been around for decades, you know, waiting for a, a miracle drug that would, that could either, either reverse aging or ma- make age, a- or make aging into a chronic condition we can treat. But I'm talking about practical longevity. The, all these longevity packages that you can buy at companies, and then they will measure everything about you and do all the medical tests. I've - I had such one of these tests, uh, last year with a company.

Offer them… Company offered me a package worth more than $20,000, $25,000. Of course, I said yes. I wanted to see what it looks like in practice and, you know, from the inside. I had 50-plus medical examinations, genome sequencing, microbiome profiling, dietary tests, uh, more than 300 blood biomarkers.

Anything you can imagine, they measure it about me. And then I got - I, uh, it took them eight hours to tell me on four, in four settings to tell me what's wrong. And they told me if I'm in excellent health, health, but if they, if they wanted to give me all the positive information, it would have taken three or four days. So we will only focus on what might be wrong or might go wrong later.

So imagine someone getting access to that test for $20,000, $25,000. It's, it's not a, you know, not the kind of choice whether I buy a wearable sensor, so I might benefit from it for my cardiac health. But if I buy a car or a longevity package for the next few years, it's a vastly different decision. So yes, I think I'm afraid that practical longevity is very much accessible now for the wealthiest groups or individuals.

But I'm - I remain positive that as it has changed with the majority of digital health, and it has changed with how AI has become more accessible to the masses, hopefully prices will go down in a similar way. I- if, if we are trying to remain positive here, whole genome sequencing service used to cost about $70,000 15, 16 years ago. And now I think for a few hundred, huh, maybe $200, $300, uh, $400, it's possible to have a whole genome sequencing service. So I expect the same to happen with, with all these three fields.

So here's my last question. You're an advisor to, you know, hospitals, governments, large companies on how to interpret the future, what to expect, where medicine is headed. And so if a health system CEO is listening in today, as they, as they tend to do, um, and they're trying to figure out, "Okay, what should I focus on, not in five years, but, you know, this year? Where should I be, uh, focusing?"

What do you suggest? What, what a fair question to ask a futurist to focus on one single thing. Uh, because now I have an internal debate in my mind between two things. I, I, I thought you would say, "But I cannot say anything about AI," because, you know, we talked about AI so much.

Because prompt engineering is the first thing that comes to mind. I think that the reason why, and I'm, I'm always interested as a, as the medical futurist in the reason why, not just highlighting a trend, but why do we have to acknowledge that trend? But it's not, it's, it - in this discussion, AI is not important. What's important is that there will be, you know, thousands more, uh, times more AI-based technologies in medic- for medical and health purposes.

We cannot expect all the stakeholders of healthcare to learn to use all of them. It's physically impossible. Not even for me, and I'm an AI researcher, I cannot learn to use all the AI tools in the technical details, but I can very much learn to use one interface with which I can have human-like conversations, and that interface will translate my ideas, my thoughts, my questions into the language of those different types of AI-based medical technologies. I think even for hospitals, physicians will keep on disc- having discussions with one large language model, and that model will help them have discussions with the AI being used by the radiology department, the AI being used by the electronic medical records, the one that the patient brings in that comes with their EC- smartphone-connected ECG.

So, all these range of AI-based technologies, the interface between them and between medical professionals and patients will be a large language model. So, the better we can use them, the better we can exploit all of these AI-based devices. And, and prompt engineering is the ability to, to ask the right questions, to share the right, um, uh, conditions and circumstances with these large language models to get the best outcomes. But the internal debate I had in my mind is that this is very AI-specific, and if I could advise a CEO or a hospital leader to do something that's not technology-related, then please start using futures methods.

So, you won't need futurists like me to tell you which trends to be excited about or which trends to fear, because you don't need people. You have been using futures methods like the futures wheel, scenario analysis, vision writing, horizon scanning yourself within your team, within your organization. So, of course, you know which futures might be possible for your team or your company. Of course, you know which one is the desired future.

You even backcasted towards today, finding out which steps, legal, scientific, technological, financial steps we have to start making from today to have a higher chance to reach the desired future, not the other ones around. So just by using a few basic futures methods, I think every hospital, organization, healthcare institution would be in a much better position to anticipate, not predict, to anticipate what could happen next, and they could design strategies, they could take preventive measures about preparing for most of these.

And then I think it takes a lot of anxiety and fear out of the equation when you know what kind of futures you have to be prepared for. Well, that's it for yet another episode of the Health Biz Podcast. I'm David Williams, president of Health Business Group. My guest today has been Dr.

Bertalan Meskó. He is founder and director of the Medical Futurist Institute. If you like what you heard and y- you don't mind that I picked on the radiologist and the anesthesiologist, please subscribe on your favorite podcast platform. And Bertalan, thank you so much for joining me today.

The pleasure was all mine. Thank you so much, David.