SPEAKER_00 0:00

You hear so much today about longevity, about AI. What I get to do in my work is that I pull it all together. Would you trust AI to help you live longer? I hope listeners will get out of today is new, something different than what you’re already hearing on all the podcasts. Not just what to do, but how to think about what to do. So how to decide what your own personal goals are, how to get there using the amazing tools that are available to us.

SPEAKER_01 0:27

Welcome back to the 925 Wellness Podcast. It’s the podcast where your career and your health grow together. I’m your host, Aisha Tahir. Would you trust AI to help you live longer? I’m sure you’re asking yourself this question. It’s a question that would have sounded like science fiction not too long ago. But today it’s here. And that’s why, on this episode, I’m exploring how AI could change the way we understand our health, the treatments medicine can offer, and the choices worth our time and money. Dr. Lin, welcome to the 925 Wellness Podcast. I’m so glad you’re here.

SPEAKER_00 1:07

Thank you so much for having me, Aisha. This has been an amazing uh experience, just getting to know you and getting on the show.

SPEAKER_01 1:14

So Oh, thank you so much. Thanks for joining us today. And, you know, I want to start for something that our listeners and viewers are probably already thinking. Someone who’s just hitting play on this episode, what can they expect to take away from our conversation today? Yeah.

SPEAKER_00 1:35

You hear so much today about longevity, about AI. And what I get to do in my work is that I pull it all together. So I hope what listeners will get out of today is new, something different than what you’re already hearing on all the podcasts, not just what to do, but how to think about what to do. So how to decide what your own personal goals are, how to get there using the amazing tools that are available to us and towards the topics of longevity and using AI.

SPEAKER_01 2:05

That’s great. That’s great. And I can tell you that a lot of my viewers and listeners are looking forward to that. So let’s start where we are at. Today, our smartwatch, it already tells us like how well we slept. Apps can already tell us like when to rest. And now AI is reading our labs, our genes, our habits, telling us how to live longer, right? That’s either the best health tool a busy professional has ever had, or it’s a lot of trust to hand over to a program or an algorithm. Dr. Lin, would you trust AI to help you live longer? And if yes, why?

SPEAKER_00 2:54

Yes, I absolutely would. And you can, you’ve heard it here. I, a physician, I do trust AI to help me live longer, but only if it’s put in the right hands. So something that people always debate about is can AI replace us doctors? And and truth is, I think you can, as long as you’re wielding this powerful tool correctly. Of course, if you put AI in the wrong hands, you can manipulate it to say whatever you want it to do. But if you go in with the right mindset and the clear experience that fortunately a clinician has, but you know, any person could develop that level of discernment and experience, then yes, I use AI all the time to help interpret the example you told me about um we’re all wearing wearables. I’ve got my ring on right now. Yeah, I see that. Yeah. So um, and you know, I wear so many gadgets at home, like a watch and I even have a sleep monitor that’s like separate from this because I’m testing it out. And uh they all tell me different things, is the truth. So uh a lot of people are a little bit wary. They’re like, these products, they’re not yet so accurate or reliable. Like, what do you do with all that info? And I’m like, well, the key is to get a lot of information and synthesize it using AI tools or, you know, sometimes when you don’t have the AI, like we have to use our just all, we have to understand it and uh, you know, put it all together. But it is still better to get information even if it’s not perfect right now. That’s awesome.

SPEAKER_01 4:23

I want to highlight what you just said that it’s important to understand that you are getting information from these tools. And, you know, having information is always valuable. The interpretation of it is where we are probably gonna say, okay, just use it as a tool, not exactly as not all data is exact as far as all of these tools are concerned. And we’re gonna get into it further uh down the line. But that’s a great way to put it. That’s gonna stick with me, actually. That, you know, it’s okay to use tools. It’s okay to use different tools to understand what’s happening within your body, too. But whether you trust it or not, it’s, you know, really gonna depend on uh where you’re at in your health and what your healthcare practitioner or your physician is telling you to do. Whether we’ll use it or not, but how can we use it wisely? So I want to dig into something that I think a lot of people wonder about, but don’t always say it out loud. When you hear how tools like Claude or ChatGPT, which are LLM models, and they are, you know, just trained on text, right? Huge amount of text and data, and they predict what should come next. If we are just looking at that, a really, really advanced version of you know, autocomplete on your phone. Now I want to just translate that to healthcare. We see our watch, right? It’s a tool and it’s just giving us a bunch of data about our body as compared to an LLM, which is now able to pass medical licensing exam, summarize research papers in seconds, and catch a pattern in someone’s lab results that might have been like totally overlooked. That’s where it starts to feel like there’s something more going on with AI. So, how does guessing the next word like turning to solving problems that takes physicians years of training to handle? And for those of us thinking about using AI for health, this really matters. And I’m asking as a patient myself, because I’m like, should I actually trust this to like I don’t know if I wanna tell Claude my blood work? Like, I I don’t know if I want to upload my lab results to onto Claude and my family history because I want to know whether it actually understands what it’s telling me. That being said, how can a model that predicts text solve heart problems? Does it actually predict or know anything about our health?

SPEAKER_00 7:20

This is such a fundamental question that I don’t think we fully understand overall. Like even the AI researchers out there who are working on this do not completely understand every last detail as to how the LLMs work. But I will, I will kind of use analogies to help us get to a comfort level that I feel comfortable with. So uh imagine the analogy that some people have mentioned. If you give a monkey a keyboard, it pounds away enough times, it will eventually create Shakespeare, right? That’s just a matter of random chance. However, an LLM is a lot smarter than that because it has the history of all written language in its experience, it’s training uh uh data. So it has read the internet, it can predict, it’s a prediction model, it knows what next token to put in. And so it will complete sentences the way humans have done so for you know all of the internet, which is mainly its training data. So your point is if we’re just predicting the next letter, the next word, how is that real reasoning? So I like to use the example of math. So you may have seen all of these headlines lately about how AI has solved some of the most difficult mathematical problems out there. It basically has won several fields medals worth, you know? And how is it doing that if it’s just predicting tokens? It’s because it’s trained on a lot of information and questions. And it’s not just learning, like, oh, this is how a mathematical proof should end. And so it like makes something up. That’s maybe like a very rudimentary early version of an LM. That’s what it would do. But uh it through the use of uh educating or training it with how math works, then it knows or it understands how to prove certain things, how certain algorithms play out, how to calculate certain things. Not to mention, we give tools to the LLMs. So an AI LLM program, so to speak, is not just predicting anymore. It’s also using tools to help predict. And so there’s a lot to it, but that goes to show that it’s not as simplistic as we’ve all been led to believe. It’s very, very complicated. That’s why it’s costing billions and billions of dollars every time uh these frontier labs are training the next new version. And so for medicine, I don’t even think it’s that hard, is the truth. We’re not uh coming up, it’s not scientific discovery. Actually, LLMs are being used for scientific discovery today. Like there’s projects at, you know, OpenAI and Anthropic and so forth working on biotech. And this is true scientific discovery. But when you’re practicing medicine or trying to come up with a clinical plan, it’s actually not as hard of a problem. And I know that feels sad because we’re clinicians. We want to feel like we have this special talent to impart upon our patients. But I think the intellect part, the knowledge part of it is not that difficult. The hard part is actually getting to know a person and understanding their personal goals, their context, and then fitting that in with what the so-called best plan is. So uh there’s a long-winded way of saying it, I have a lot of trust actually in the knowledge, the factual result of an LLM, uh, fitting it to your personal context is the trick. Yeah.

SPEAKER_01 10:51

And I think that’s where, you know, healthcare professionals come in because we can help the patient understand, okay, you know, this is where it actually applies to you. Exactly. Exactly. And this is where I wouldn’t think, give it a second thought, or I would take it with a grain of salt, like what it’s saying, right? Because it’s not really applicable to you. I love that. Thanks for breaking that down so wonderfully, because I don’t think so. I was aware that how much work it takes on the back end to train that model to get in onto next generation or next version of it.

SPEAKER_00 11:30

And like I I believe anyone who’s following along is probably as astounded as I am. But every new model that’s coming out, it’s shocking to me because we’re making leaps and bounds. We’ve certainly crossed some sort of Rubicon where uh it feels super intelligent compared to humans in many ways. It’s still spiky, you’ll hear that, but uh it’s already at a level that’s scary good.

SPEAKER_01 11:58

So true. And the pace is fast, right? Just as you said, that you know, we kind of like can’t keep up with it. No, certainly.

SPEAKER_00 12:06

Yeah. So and I encourage uh everyone who has not felt that yet. So first make sure that you’re playing with the actual frontier models, the actual most advanced LLMs out there. Yes, it’s worth the five to even $500 a month, I guarantee it, uh, because it will change your life. Yeah. And other than using that, make sure that you’re really pushing it. A lot of people don’t, they’re like afraid for some reasons. And I totally understand you want to get comfortable with security permissions and so forth. But the more permissions you give, like an LLM to access your information, your data, the more context you give it, the more valuable the results. So I I tell everyone the first step is to tinker yourself. And, you know, we have all become builders in this age of AI. True, so true.

SPEAKER_01 12:59

As much as it’s helping us do a lot of things, like, you know, different LLM models and AI just in general, wherever it’s used, it’s also giving us a lot of agency, a lot of tools to build what we want. So that’s such a great point that you made. And I wanted to underscore it. That being said, once we understand how it works, we can see how it both has strengths and some limitations. And that’s okay. That actually helps us understand. Okay, where do we come in then? As you said. So let’s bring this back to everyday life of a busy professional, which is my primary patient. So, Dr. Lynn, what would a useful AI-assisted healthcare model look like, or the future of healthcare with AI-assisted models looks like for a single patient?

SPEAKER_00 13:55

So I think you’re referring to how can a person who isn’t, you know, building an entire software necessarily for themselves, how can they utilize uh AI to, you know, function for their day-to-day health? So I have seen many startups try to create personalized health records. And that’s the idea of like having your own, it’s like a binder, you know. Back in the day, people used to bring binders full of their records, their labs and everything to the doctor and be like, help me. And that was like the trainee’s job to go through like the medical student or the resident. And uh, the personal health record is the idea that we could turn that into software. And we kind of skipped past the point where that was actually useful. And I’ll explain why. It’s because now that we have AI, everybody can create their own PHR. We don’t need a third-party company to create a PHR software anymore. And so And no more EMRs. Yes. That’s I actually think that is true. The literal only reason why some of these incumbents are still alive, these EHR companies, we all know who it’s which they are, is uh two reasons. One is they have license licensure sort of relationships with big, big health systems. Yeah. And two is the insurance reimbursement system.

SPEAKER_01 15:19

Yeah, that’s still dependent on EHR or EMRs. Yes, exactly.

SPEAKER_00 15:24

But everything else, everyone thinks like an EHR or EMR so complicated. And uh it’s really just the integration of a bunch of tools. And a lot of those tools, once you start building yourself, you’re like, oh, like, did I really need about 50 tools? Like, no, in most cases, uh, now we’re speaking from the clinician’s perspective, but I encourage clinicians too. I’m like, if you have any bandwidth at all to do so, I encourage you to build for yourself. You can build your own documentation system, your scribing system, your uh e-prescription system. And then you don’t need to rely on a standard solution, a templated solution that an EHR company is going to give you. So uh, but that aside, I think from the patient’s perspective, you can also do the same thing. You can create something, and I’ve done this for myself as a little pilot for my own platform, because I I build a platform as well. And uh you can even grab your Apple Health data. So I have everything funneling into Apple Health, and then now I’ve got a uh I do it daily, a daily sync where the Apple Health export will go into my personal health record essentially. And I have an AI coach that tells me how I’m doing with that information from my workout information. I’m also telling it all day long, like what I’m eating. So it’s tracking my macros, my nutrients. Uh, it has my sleep info, obviously. And so it has a great synthesis. Uh I could do it every day. I I kind of get overwhelmed. So I just tell it, like, just coach me once a week. Yeah. Yeah. But this is the kind of thing anybody can do with no extra skills. I’m not an engineer by training. It’s uh you could just set this up on your own.

SPEAKER_01 17:07

Yeah, no, that’s wonderful. That’s a wonderful way to use AI for your daily or weekly, you know, uh routine and to understand, okay, what does it look like for me today?

SPEAKER_00 17:20

Like, I can tell you really interesting findings. So I always tell folks, we are N of one. Like we are different. You and I are very different. Like, and everybody on social media tries to generalize, which is such a problem, as you know. Yes. And uh I have people arguing in the comments, like, oh, you should be on this dose of vitamin D, or you know, make sure you get this much REM sleep. I’m like, you know, this is besides the point. Like, every human body is very different. So true. But with AI helping you synthesize your own data, you can learn certain things. Like the one random fact about me that is probably common to a lot of listeners, actually, is that I actually sleep on average a full half an hour less during my luteal phase. And like for the mental cycle, right? That’s for a lot of women. Yeah, a lot of people are pretty common. Yeah. Yes, yes. And then I was like, oh, it’s it’s so interesting that you can see that. And um, and also, you know, I know exactly where it’s coming from, right? For me, it’s the end of sleep. You some people have trouble falling asleep. I have no problem with that. I always have trouble with the end of sleep. So yeah, so little, little thing, details like that. Yes, you could uh you could gain inspiration and generate hypotheses from uh social media if you want, like go on Reddit and find out. Of course. But then you can do the experiment on yourself with your own data.

SPEAKER_01 18:40

Yeah, which is pretty neat, right? With AI, now we can all be the researchers of our own body. So that’s great. That’s great. So what I’m hearing is that AI assisted healthcare isn’t just about like one big breakthrough. It’s about actually giving you the agency as a patient that, you know what, now you have this tool. You can just like you said, you can sync your smartwatch or whatever, like the ring or whoop if you wear it, with your model, LLM model and train it to guide you.

SPEAKER_00 19:17

And I’ll even take it one step further because in longevity clinics like mine, we use a lot of funny tests. So, like uh, you know, like uh some that many listeners will have heard of, some that you haven’t, like biological aging tests, true diagnostics, systems age, like all of these tests that are honestly, they’re still in the research world. And I say that because I know for a fact that it’s always better to verify with direct blood testing, with the standard blood test that, you know, most clinicians are using. Sure. But of course, in longevity, we’re excited. We want to kind of know what’s possible. So we we use these research grade sort of tests. And it’s not wrong, like just because they’re not quite uh as reliable or like we don’t know the meaning quite as surely as we do for like these more traditional tests. You just have to take it in mind, like, okay, this is this is a research project. So what does this mean? So for example, I just got a recent, I do these serially for myself. I I got a recent true diagnostics result back. Yeah, yeah. Yeah. And I was like comparing it to a past one, and these are epigenetic tests. So they’re gonna tell you so many things. And and they will, the company even, you know, is very straightforward about it. It’s like you can’t assume that what something says actually means something because there’s a lag, it’s epigenetic, there’s a lot of different factors at play. So AI is really good though at telling you what is possibly meaningful and what is less meaningful because it totally contradicts something else that you already tested. So I find AI to be incredible in situations like that. Cause I I frankly, um, just to tell a true story, like I had COVID uh in in early 2025, so a little over a year ago, and I had post-COVID, like medium bad. I know it gets very bad for some patients. Yeah. But I had like medium bad. And uh I actually, in my epigenetic testing six months later, I could see the results of that. I could see some immune results, and I was like, oh man, like will I ever recover from this? And more recently, when I rechecked it, I was like, oh, everything’s like completely back to normal. So, and I feel it too. Like, obviously, in this case, like you can tell. Yeah, you can tell. Yeah, but it’s just interesting stuff like that that we never used to be able to understand because there’s way too much noise, or it’s like it’s way too early in the research kind of realm. Now we can utilize that kind of data because of AI. That’s wonderful.

SPEAKER_01 21:50

I just want to make sure that our listeners and viewers are aware of what we are talking about. True diagnostics has an epigenetic test. You can just, you know, it’s like A blood draw. Yeah. You just drop it off at the lab, and then you know it it’s sent to their specialized lab.

SPEAKER_00 22:06

Um, if you want to like comment more on it, please tell out our viewers because everyone might not know. Yeah. So I mean, I’ll I’m happy to mention any or all of the types of tests that I’m always trying out. And I frame them all as experimental because as I said before, in the longevity world, we love data, but we understand that a lot of data is still research realm. So true diagnostics is one of these such tests, very popular because you can kind of measure a biological age. It’s actually a finger prick, which is not the most fun for a lot of people because you have to get a lot of blood from your finger. So just imagine like dropping blood onto like a blood spot. It’s like a lot of blood. So it’s not it’s a petri dish. It’s a petri dish at home. It’s not it’s not the most fun, but I I actually prefer it. Some people they hate that um kind of method of delivery. There’s these new tassos, which are great. Um, some people may have seen the the big red button that you can put on your arm, and then it uh it it’s an easier way to collect blood as well.

SPEAKER_01 23:10

Yeah, I I just recently used one and I was like, that was pretty neat. It was just like, you know, slapping a CGM on you. It was so painless. Yeah, so painless. Yeah, and then it it’s just you know, you can tape it off, then you send it over. Yeah. So yeah, go ahead.

SPEAKER_00 23:25

So it’s a finger prick. Finger prick, uh dry blood spot. And the reason why, by the way, in case you’re wondering, why all these different testing companies use different methods is because once you pick a method, you have to get approval for that particular method. You can’t switch now that they’ve decided they’re gonna do dry blood samples, like the blood spots, they can’t go ahead and be like, now we’re gonna do tasso, or now we’re gonna do like a venous straw. So that’s that’s why you see so many different kinds, and that’s why they don’t change them. So true diagnostics, they uh they use this epigenetics, they are looking at markers on your DNA essentially that show uh signals, correlations in these big biobank studies that they performed earlier, like a decade ago. And they correlate it with other people who had the direct blood draws. So uh a random one, like uh DHEA, it’s a hormone-related number. They will tell you, like, okay, you’re in the top 99th percentile. I actually, this, this again, because I’m comfortable using myself as an example. I went from being like so low to like being uh exceptionally high. Like it went to 99th percentile, which is absurd. Like I was like, I’m not, I’m not supplementing, I’m not doing anything. So that’s something that I’m like puzzling over, but it’s not a direct blood test. So it’s just a correlation. So always, if you want to be sure, you then go to the regular lab, whether it’s you know, bioreference quests, lab core, whatever it is, and you get a direct blood draw. So then you can actually check what your real testosterone and estradiol levels are. So um, but they can do all of those uh based on a single drop, a large drop, but it you can imagine you don’t have to draw 100 test tubes. You can just check the ones that uh looked kind of interesting based on that first sample.

SPEAKER_01 25:19

Yeah, thanks for explaining that. I I was like, okay, let me just, you know, pause here and expand on this. But thank thank you so much for doing that. Yeah. And something you you just said hit home for me because a lot of times we, even as clinicians and practitioners, we’re like, oh, okay, we are just gonna take this test as it is. But what you said was, oh, but no, this is an epigenetic test. Just remember that. It’s it’s just understanding, uh understanding the uh DNA, right? Yeah. You have to use it in conjunction with your blood test, right? Yeah. So that’s where it comes in. Like a lot of times we are like then so stuck. Yeah.

SPEAKER_00 26:00

Or or a patient might be like, oh no, what is this? Right. Yeah. I think um this is going to sound a little bit uh like a disconnect, so but I’m gonna try and explain that AI enables a person to think for themselves what something means. And what I mean is in medicine, especially a very busy clinic, often we have to go by guidelines. We’re like, okay, if your LDL is over 100 and you have a family member who has had a heart attack before age 50, then you are super high risk and we’ll put you on a drug. Yeah. But if you’re not, then you don’t qualify and you know, whatever. But that is so overly simplistic, obviously. And we should do more. Like you and I know, like ideally, we should be treating people based on their actual desire to be aggressive about treatment, risk, and benefit. So all of these details that take way too much time, sadly, in a traditional primary care clinic, let’s just say. But now, because you have AI that can so quickly synthesize everything and also understands the flaws of testing. So, for example, if you go in one day versus another, you’re gonna have different triglycerides and LDL levels. And depending on what you ate last night, if that’s gonna make a difference. And the trends of whether something is lower or higher, that’s going to make a difference as well. So knowing all of that in context and then being able to tell you, okay, this might mean something a little bit more than you thought, or even it it doesn’t mean that much, is that those are valuable nuances that are currently they sound too complicated for many listeners. I get comments all the time in my YouTube channel being like, Can you just tell me what to do? And I’m like, Well, that that wasn’t the point of this whole video, was to help you learn how to think, actually. And and that is harder in the beginning, but once you understand like these are the things to consider, to think for yourself, then going back to your point, then you have true autonomy, true agency over your own health. You don’t have to rely on other people to just tell you what to do.

SPEAKER_01 28:05

Yes, and that’s a great power to have. It’s like empowering for the patient. Yes, right. For every person. I want to tie this into biological age because you mentioned that with true diagnostic test and um or any epigenetic test, because that’s what they give you. So my Garmin watch tells me one biological age for me, right? And then I wore whoop for a very long time. I’m not wearing it anymore, but it used to tell me another age. My epigenetic test tells me something else, although they are kind of like off by just a little bit, not too much. But a lot of times, like people have come to me and they’re like, we don’t know which one to trust. What does this score, does this score actually reflect what is happening inside? So, Dr. Lynn, does a younger biological age score mean like better health? What should a patient trust? What most of our viewers and listeners are like thinking about that already.

SPEAKER_00 29:05

And they are all all different. You know, you mentioned some of the wearables, the common one that’s done with blood testing, like function will report a pheno age. Yeah. And there’s also, I mentioned true diagnostics, there’s also glycan age, which is becoming quite popular, which is measuring the glycans on your blood levels. They all need to be taken into context. And I tell people all the time, don’t use it as an actual age. It’s kind of a score. And scores are imprecise and imperfect. So don’t be sad just because it’s like older than your age or whatever it might be. It means it’s a signal or a synthesis of the thousands of data points and that the company that created that score is trying to make sense of. We don’t have any truly reliable biological age that performs better than chronological age for really big questions. Like, are you going to like be more likely to die in like the next year or something like that? Yeah. So I tell people, don’t freak out. Like, this is just a signal of some kind. Yeah. And depending on what the algorithm is, that can clue you into what to dig into. So, for example, I mentioned glycan age. So, glycan age is based on these like little sugar glycan things on your blood levels, and they actually go down in age if you have a good hormone optimization. So, I was actually talking to a fellow physician about this, and he’s like, um, he was like, uh, we can drive a person’s age down by like 20, 30 years, uh, but only really in men, not women. And it’s not that they’re improving men’s health that much. And by the way, this is related to estrogen in both men and women. Yeah, yeah. Uh, but it’s more that there’s this like lever that’s bigger for the men. For men’s screen. Yeah, for that particular score. And so that’s why I encourage people to get to understand when you get a score, like why is that? Like your whoop or whatever is on your wrist or your ring, that’s most likely uh driving a score based on your heartbeat and your HRV. So it’s probably signaling that there’s something about your heartbeat, your resting heart rate, your HRV, uh, maybe your sleep score or your sleep quality that is uh a better or worse than someone else’s, if your quote unquote biological age is better or worse. So long story short, I treat all calculations and in fact all measurements very loosely. Like I don’t, I don’t trust any one thing. And uh that actually encourages me to get even more data than less, if that makes sense.

SPEAKER_01 31:53

Yeah, no, it does. Yeah, uh, absolutely. And thanks for expanding on that. Yeah. Because honestly, that’s so good to hear. Because I think a lot of people treat that number like a final grade. Yes. But that’s not what it is. And what you are just saying is it’s more like a snapshot. Yes. And, you know, just keep that in mind and see where that information is coming from and just treat it like that because it’s not the whole story. It definitely shouldn’t replace the basics like your blood work and blood pressure and how you are actually feeling day to day. No, no wearable will be able to tell you that. Yeah. Yeah, right. So those are a few things that I think a lot of people, you know, should just keep them front and center. Okay, so in the age of AI, what would they need from a physician? What does what do you think physician-patient relationship would look like? I know you have talked about this. I was uh listening to your podcast that now a patient can come in. Mostly they are actually already coming in with a lot of information already, you know, through these wearables and or other blood tests and/or tests, just tests like longevity tests. Like they’re going for whole body MRIs already, and they’re coming and bringing all of that information to you. So, what do you think? Where are we headed as far as physician-patient interaction is concerned?

SPEAKER_00 33:22

Yeah, I always tell people that you can get to the level of a clinician in terms of understanding, but the reason why it’s not as easy as just play putting it into Chat GPT and then getting out an answer is because you the it’s honestly the the, if you want to think it in AI terms, the context. So if you had infinite context, then you would understand everything that any clinician could tell you. But a given person mostly has experienced not being a doctor. So you don’t see patterns as clearly, and you don’t know, you don’t have that instinct or that gut feeling yet of what’s important and what’s not important. And so that’s still something that I see over and over again. People come in, they’re so worried about everything that they see because like it’s so much information now. Like so uh whole body MRIs, whole body MRIs are amongst the most controversial thing that I talk about. And I’ve gotten a couple, I’ve had patients who’ve had them. And even though they are still not at the level where I’m encouraging them for everyone because there’s a lot of factors, like, you know, if you find a finding, what are you gonna do? Get biopsies on everything, you know, there’s a lot of questions still. I I still am for the concept of a whole body scan somehow because I think we need to get there. Uh like everything else I’ve spoken about, we’re still in this research phase. We need to make it easier for people to get some kind of full body scan. So we understand what’s in our body. We have a baseline, we see how things change. Uh, and it may not be a whole body MRI, actually. Uh, it may be something else. Uh, there’s even like some companies thinking about doing a spa ultrasound, which is very out there, but I I kind of understand the idea because it makes it easy for you to get very frequently. So spas are more popular now. Spa’s so much more popular. It’s easier. It looks like the concept I saw. This is the mid-journey concept where you know you just go on a big bath. And yeah, that sounds way more fun than an MRI. So, like, you know, I would do it every month. So absolutely. Absolutely.

SPEAKER_01 35:35

So, Dr. Lynn, to kind of like debunk or get your viewpoint on a few things, because I think right now a lot of people, like, as soon as they go on TikTok or social media, they’re like bombarded with so many different things as far as supplements go, as far as testing is concerned. And a lot of them are like coming to me, like, what do you think of this? What do you think of that? What do you think of it? Right. So I want to do like this lightning round with you. Sure. Where I have like a few quick fire questions, and in a sentence or two, you can just tell me if they’re yay, nay, or it depends, or maybe later. Yeah. Right. Okay. Okay. So the first category is supplements. Creatine.

SPEAKER_00 36:22

Creatine. Yes, for strength and training, not there in evidence for longevity in general. Nice. Vitamin D. Yes, if you have got a deficiency, like me, a very deep deficiency, but test your levels and treat to your levels. Don’t just take someone’s word for what dose to get. I like that. Yes, so true. Yeah. But most of us are deficient.

SPEAKER_01 36:47

So just make sure you get tested. We’re all indoors. Oh, yeah, absolutely. Especially now that winter season, right here in the northern hemisphere. Okay, protein powders.

SPEAKER_00 36:58

Protein powders are great if you don’t get protein through your whole food. I love whole food for protein better, but protein powders are very high quality these days. So yay. Awesome. NAD boosters. NAD boosters, I’m kind of on a meh for because there’s really not sufficient evidence for the longevity. But I have patients who insist they help so much for energy. So I’m not a nay nay. I just think it’s a it depends.

SPEAKER_01 37:29

Yeah, it does depend. I don’t find them as helpful for energy, but I do notice. I I notice like uh my hair and nail um grow much longer. So there’s something about the nutrients being more available. I that’s what I read.

SPEAKER_00 37:46

But don’t know. The jury’s still out, and generally I’m open-minded. If someone wants to try, I’m like, go for it. Make sure you’re up to date with all your cancer screening is like the tiny little voice in the back of my head I always tell people because there’s a small, small risk that NAD can also help cancer cells grow more. I’m so glad that you mentioned that. Okay, gray market peptides. Definitely a nay for me. I know there’s so much excitement. I don’t want to get tons of hate online, but it’s because the supply chain is just not very trustworthy. Even if you do get it from a compounding pharmacy, they aren’t able to get it from a GMP certified manufacturer. It’s just impossible because of the regulatory situation. I am totally for it once we have much more safe regulated supply chains. I like that answer. Okay, now tests. Apo B and LP little A. Yes, definitely APOB whenever you can, every six months if it’s a concern for you. LP little A at least once in your lifetime. And now we’re having drugs come to market in the next few years. You might be able to actually get it regularly and see your LP little A come down, hopefully. Oh, yeah, wonderful. Oh the next two to three years, it’ll be real. Oh, nice.

SPEAKER_01 39:08

I like that. And you know what? That would really help people a lot. So much. That would be nice. Okay, hemoglobin A1C and glucose levels.

SPEAKER_00 39:16

Absolutely. Yay. For hemoglobin A1C, fasting glucose, also I’ll add fasting insulin. It’s the fastest to change. So you can see if something’s kind of going the wrong direction, even before you see any change with the other two. Love that. Dexascan. Yay, for sure. I think we first of all, we talk about muscle mass, visceral fat all the time. Those are absolutely important to track. But I’ve actually caught so many people with osteopenia or early osteoporosis, like you would not even expect. And I suspect it’s that we our guidelines didn’t allow us to check before a certain age or in the broad population. And now we’re catching it all over the place.

SPEAKER_01 39:58

Yes. Yes. And you kind of already answered the whole body MRI. So I’m just gonna fast forward from there. If someone has limited time and money and is listening to our conversation, what is the most useful next step after this conversation for them?

SPEAKER_00 40:15

Get to know your own body and use AI to help you synthesize all that data. And by getting to know your own body, I mean try to collect all the information you can. Even if you have not a single wearable, just journal, like how are you feeling? How much do you sleep? How do you feel each day? What are you eating? These are all things that are basically free to do. If you have a wearable, dive in. If you are able to get labs checked, dive into that. All the other things are icing on top.

SPEAKER_01 40:43

I love that. I just love all the takeaways you gave us. Real quick, what are you excited about as far as AI and medicine are concerned?

SPEAKER_00 40:52

Oh my gosh. I think this next year is going to be mind-boggling in terms of innovations. The AI prescribing platforms are actually top of my list. So I think this is controversial, perhaps, but I think that protecting patients by uh gating prescriptions behind uh clinicians is actually harming more people than it’s good. First of all, for the doctors, it makes you a rubber stamp. So it turns us into a transaction piece. We’re part of an assembly line. We don’t need that to be valuable to patients, truly. And also, it means that people just need to check the box, like go to your doctor, get whatever medication. It’s very adversarial. And AI is so much better at doing this broad data collection and synthesis, like we mentioned. Is this person, you know, abusing a certain drug? Are they, you know, are they ready for a refill? Do they need one right now? Or like should they do something with a dose? AI is perfect for that. You don’t necessarily need a clinician to sit there and like think through all these transactional pieces. So I’m very excited and I actually think they will succeed. And it’s a matter of maybe five years, maybe, maybe 10, if we have very antagonistic regulation before AI is prescribing all of our medication.

SPEAKER_01 42:14

Wow, that does sound exciting. Yeah, because it’s going to be automated. Like just like you just said, like it’s gonna be able to recognize the patterns better than we do or the time we take.

SPEAKER_00 42:23

Yeah, and it’s so silly. You know, in other countries, I’m from Taiwan. Uh prescriptions are quite a bit looser. Yes, they’re written by clinicians, but a lot of times you go to the pharmacy and you’re like, you explain your symptoms and they’ll give you whatever medication it is. Yeah. So I I remember just buying like a banana bag like from a pharmacy. I was like, my grandmother needs IV flu. I’ve been always so I mean, some of these things are perfectly safe. So we it’s all about we need to increase access to good health care. And sometimes that means getting rid of silly things. Yeah. I agree. Thank you so much for your time, Dr.

SPEAKER_01 43:01

This has been so much fun. Oh, this was so much fun. Thank you so much. And I would love to have you back so that we can talk a little bit more about curator. Yes when you know it’s a good time. Of course. Thank you so much, everyone, for listening and watching. Thanks for joining us on the Nine to Five Wellness podcast. Bye for now. Thank you.