[00:00:00] Speaker A: Welcome to the Heart Rate Variability Podcast. Each week we talk about heart rate variability and how it can be used to improve your overall health and wellness.
Please consider the information in this podcast for your informational use and not medical advice. Please see your medical provider to apply any of the strategies outlined in this episode. Heart Rate Variability Podcast is a production of Optimal LLC and Optimal HRV Check. Check us
[email protected] Please enjoy the show.
Welcome friends to the Heart Rate Variability Podcast. I'm Matt Bennett and I'm so excited for this conversation. Today I saw Dr. Seeley's article on my daily strive through HRV research and it was one of those that I believe in my opinion addresses this critical spot where we're at, whether we're in mental health, whether we're in medicine, about using hrv, maybe not as a diagnostic tool, but to get clinical insight. And this is where I see this work evolving. I don't think it's easy work. I don't want to say we're there yet, but Dr. Seeley's article really, I think positioned some ideas where we might use this, some critical and maybe even I like the drama here, Dr. Seeley. So forgive me, maybe some life receiving insight that at the very least is could really lead to better health outcomes for patients, for clients. So I'm really excited to have this conversation with you and one of the things that really stoked my excitement for this conversation was your just diverse background and expertise. So I know we're going to dive into that throughout the episode, but I'd love for you just to start us out with an introduction of yourself and a little bit about your expertise in your work.
[00:01:57] Speaker B: Thank you Matt. And your, your enthusiasm and excitement about HRV is wonderful and, and it's delighted, I'm delighted to meet you and, and congrats on all of your work to date on, on this topic. Well done.
So I'm a clinician scientist in Ottawa, Canada. So Ottawa of course is the capital of and I work at the Ottawa Hospital.
I practice as a chest surgeon, a thoracic surgeon, as well as an ICU doctor.
And my background after, you know, first of my undergraduate degree was in physics and I chose physics cause I enjoyed the topic more than any other.
But I was already interested in going medical school after that. My physics professors were disappointed, of course. They thought I'm going to a lesser science, you know. But I then went to McGill University for medical school. I studied general surgery and did a PhD at McGill and it was really the last paper of my PhD, where I developed a strong interest in variability analysis. And, and I then came to Ottawa, completed my training in thoracic surgery and critical care medicine. I've practiced in Ottawa ever since.
And you know, my interest in variability was really originated from a discovery and an interest in something called complex system science.
And.
I remember seeing a lecture given by a fellow named Peter Macklem and that highlighted how variation in respiratory rate, and so this was in fact respiratory rate variability and tidal volume variability, which is the variation in the amount of the size of each breath that we take, that the patterns of fluctuations of that was the same as the patterns of fluctuations we see in stock market fluctuations or so flares and avalanche, type of power, law behavior.
And I found that to be extraordinarily exciting. And, and.
I was also working in the lab trying to understand how our bodies respond to sepsis, massive infection, severe trauma shock, where the blood pressure is very low.
And so people call that the host response to sepsis, shock and trauma.
And it's what, it's what often will lead to critical illness and multiple organ dysfunction syndrome, which is a form of critical illness which is devastating and in fact how patients pass away in ICUs.
And I became interested in the concept of trying to evaluate the host response to sepsis, shock control, trauma as a complex system, as a whole system in a holistic manner.
And that's what really led me to thinking that variability analysis, heart rate variability, respiratory rate variability, blood pressure variability would reflect the altered system properties. So I approached it as a technology to track the whole system over time.
And that was really where my original interest, you know, was, was sort of born. And, and I, and I wrote my original paper on, on the cons on this topic was I think called Multiple Organ Dysfunction Syndrome, Exploring the Paradigm of complex nonlinear Systems. And that's where I kind of hypothesized that variability monitoring would be helpful and a kind of a new way of monitoring patients in the ICU that I published that paper in July 2000, so 20, 26 years ago. And, and I never realized that it would, it would be, it would take me so long to actually help deliver this technology to the bedside.
But I'm excited to say that that that's what we're doing right now. And, and it's been a fun journey of exploration and, and, and, and, and every step of the way has taught me that there's value in variability analysis. So I know you've seen that in your work and with talking to others, and that's certainly what I've seen as well. So it's cool technology, but it is difficult to apply at the bedside, for sure.
[00:07:42] Speaker A: Well, that's where I'd love for you to kind of take us to that bedside as a provider, because I, you know, one of the things that I see in the research, and I think it's sort of an interesting thought experiment, but we need to kind of. I think there could be a lot of benefit into figuring it out is, you know, my, as a layperson walking into an ICU or really having my wife just gone through a surgery not too long ago. She's hooked up to a lot of stuff, and there's a lot of data, you know, coming into the monitor. And, you know, it seems like heart rate variability hasn't been one of the main biometric markers that is taken into account.
And this, again, I'm a layperson kind of trying to get into the literature. So I'm curious, you know, in the icu, do people like the outside yourself, obviously, do they know what heart rate variability is? Are they looking at RMSSD or high frequency? Or is that just so far down the electronic medical record? If it's on there as all that?
It's not even on folks radar.
[00:08:59] Speaker B: It's not on a lot of clinicians radar.
It's increasing now, but it's because they see it on their watch or their garment, their whoop, their oura ring.
So that's how they have knowledge of heart rate variability. Now, of course, obstetrics has used heart rate variability to detect fetal distress for ages and ages.
And the ability to detect a loss of heart rate variability in an unborn new or child is visible just visually.
And so they can. Yeah, although you can quantify it too, but it's also visually evident. So, but, but just a comment about.
So my original Hypothesis back in 2000 was to do kind of what you just suggested, which is that you'd have your vital signs at your bedside and then you would have your vital sign variability next to it, and that you could track that as a marker of how the patient is doing and track it over time.
And I tried to do that. I've tried to do it several times.
And I even did an experiment where I had medical students in a room with patients documenting every single thing that happened to them. And then we tracked their variability over seven to 14 days.
And we try to evaluate the hypothesis that, okay, I can track the variability as a marker of.
Of illness severity and also as a response to treatments.
But unfortunately, there are so many things that happen in an ICU to a patient that cause their heart rate variability to go up and down throughout the day, that you can't really use it as a kind of a vital sign kind of type. Monitor the types of things that happen in an ICU as someone comes in to do suctioning of the patient through an endotracheal tube, or they will be doing a chest X ray, or they'll be doing chest physio, or they will be giving medications that cause sedation for the patient and that goes up and down, you know, and so it's not, you can't. And one of the challenges, but also opportunities reflected by heart rate variability research is that it's a very sensitive marker to a variety of different things that will cause a reduction in hrv.
And so it's not really possible in my experience to use HRV as a marker that you just track continuously over time. And whether you do, you know, the interval width that you measure, it can vary, of course, but what we did find is that if you are measuring it continuously and then you average values over 24 hour periods, well, then you definitely see improvement as the patient improves and deterioration as the patient deteriorates. And it is, it corresponds to the severity of organ dysfunction.
So the sicker a patient is, the worse the variability is.
But that doesn't really help you as a clinician. Like I already know how severe, like how sick a patient is, and some people have published papers on mortality prediction, you know, for example. But that also doesn't really change my management because the patient has a 40% chance of surviving or a 60% chance of surviving or an 80% chance of surviving. It doesn't matter. I'm going to still try the very best possible and, and optimize their chance of survival.
Now, if they have a 99% chance of not making it, then maybe that'll change my, my, my, my, my plan. But usually it's fairly straightforward. In the icu, if there is hope, we pursue all possible care with appropriate limitations if relevant.
And if there is no hope, we change our goals of care to comfort.
So it's more of a binary kind of approach.
But what I have found variability to be very useful is to use it in targeted strategic implementation. And the article you reviewed is an example of that where you restrict other kind of other activities that might be going on. So you have a period of focused evaluation where the patient is stable, they're not being disturbed by a variety of things, they're in one position, and you evaluate their heart rate Variability or I also measure respiratory variability.
So, and just to be clear, heart rate variability, I always get it from an ekg, of course, measuring the inter beat interval time series. And for respiratory rate, we do the inter breath interval time series and we measure that with capnography, where there's, you know, a kind of a step, kind of tracing graph which represents the exhaled CO2 and then it comes back down to zero as you take a breath in.
What we found is that clinicians have a tough time interpreting things like RMSSD or LFHF, or the approximate entropy, or the sample entropy, or the detrended fluctuation analysis of heart rate variability.
And instead what they want to see is a prediction of a future event.
Predictive modeling or machine learning, which is at the heart of AI, of course, is basically looking for reliable, reproducible prediction.
As an aside, there's statistical analysis that tells you something is associated with something else, where predictive modeling tells you if something is reliably predictive of something else.
And what is useful clinically is often prediction.
And so we transform variability into prediction. And you do that through large observational studies where you collect hundreds or thousands of patient data.
You see, okay, in the study you reviewed, we found that HRV evaluated just after you present to the emergency room was predictive of future deterioration in patients with infection.
And when we combined that with three laboratory values, we had even better prediction.
And so that's clinically useful because it helps you to determine where should I admit the patient, should I admit the patient to the ICU or should I admit them to the ward?
And so, so the approach that I take with variability is, first of all, we transform the waveforms that we collect at the bedside to variability. We then transform the variability into a predictive model that gives likelihood of something happening in the future.
We then incorporate the predictive model into a clinical decision support tool.
And the clinical decision support tool is what the doctor looks at to influence their decision and to complicate things finally a little. Lastly, when you have a clinical decision support tool that is directly affecting patient care because it's giving the doctor information that is going to influence their decision, then that requires regulatory approval.
So in Canada, by Health Canada, in the United States, it's the fda, Europe, it's CE Mark.
So those are all required.
So that's kind of the roadmap, as it were, to transform monitoring into improved care.
And that's really the work that I've done over the last 15 years or so.
[00:17:46] Speaker A: That's so cool. I'm Just curious as a nerd around AI, whether it saves us all or the outcome's not so good. But like, I, I'm just curious like you, you were thinking in these deep ways about data and health. And I love just the background in physics to, to inform all that as well. And just sort of wondering now with the AI tools, and I know for some people these tools existed before the chat gbt, you know, release, but just sort of what do you see in, in especially like an icu, intense setting, is it really helping the providers synthesize this data to get, to get a depth of information to the condition of the patient that they're giving care to?
[00:18:39] Speaker B: Well, it's an exciting area. I would say there's a variety of ways that AI is potentially helpful to our ability to care for patients. And at least in healthcare, I don't see AI kind of replacing physicians, but definitely I see AI as assisting clinicians to on average make better decisions.
So the tool that we have several tools that we're evaluating, but I'll give you one example of a tool that we're evaluating in a randomized control trial at this very moment. And so every day in the intensive care unit, people unfortunately have to be on breathing machines, ventilators, because they can't breathe on their own. So it might be after a pneumonia, might be after a trauma, might be after a major operation and something like that.
And when we make a decision to take them off the breathing machine and we have to determine what's the optimal timing of that. So we do that decision making every day for multiple patients.
And it's a critical, high stakes life and death kind of decision because if we get it wrong and they fail, what we call extubation, which is taking the endotracheal tube out and they have to go back on the breathing machine within 48 hours. And often it's urgently in the, in the middle of the night, that can be an insult. Just as they were improving from their critical illness, that actually causes irreversible deterioration and can cause death, in fact, and, and on average when you fail extubation, you stay in the ICU nine more days than if you did not fail extubation. So it's an enormous, and it happens about 15% of the time. So it's common, it's harmful, it's costly to hospitals.
So I explored the hypothesis that heart rate variability and, or respiratory variability, respiratory rate variability, would better predict whether someone would come off the breathing machine.
And that was a large study I did back in Just over a decade ago, where we found that there was multiple measures of respiratory variability predicted extubation failure, and two measures of heart rate variability predicted extubation failure. But when you did the.
I'm sorry, they were statistically associated with extubation failure. But when you did the predictive modeling, it was only respiratory rate variability that predicted extubation failure. So I subsequently built a device, a software tool that takes capnography monitoring, turns it into variability, turns it into a prediction of the likelihood of extubation failure, and then I incorporate that into a clinical decision support tool called Extubation Advisor. Okay. And we make that available to doctors whenever they're evaluating a patient's readiness for extubation.
And right now that's being evaluated in a randomized controlled trial. We have 170 patients enrolled, and it's going to take 700 patients to prove or disprove the hypothesis that Extubation Advisor improves care for patients.
Quite some time ago, because I realized that the tool needed regulatory approval, I founded a company called Therapeutic Monitoring Systems in order to help bring the technology to the bedside.
And we had a patent on the concepts as well, and multiple patents, in fact. But the bottom line is that the goal is to improve patient care. And so the company is just an extension of my laboratory at the Ottawa Hospital.
Now, some doctors find that they really like the tool, that it helps them to make a decision.
Some are like, kind of more like, well, I don't need any other tool to help make a decision. So, so it all depends. There's a little bit of, you have to win over kind of the confidence of the physician. And, and how does that happen? They have to experience that the tool is actually helping them out.
And, and, but it's going really well. So, and, and so we, we've done phase one studies and, and, and, and, and, you know, interventional studies. We've done interviews of doctors and respiratory therapists using the tool, and so far it's performed very well. But ultimately it's a randomized controlled trial that determines the true impact of a tool like this, because that's the only way to definitively evaluate it. So. Yeah.
[00:23:59] Speaker A: Cool. Well, I'd love to just dive in a little bit more to your thinking around variability because it seems to have informed, as you, in your introduction and looking at your background and other studies, just a big focus of yours in your work, and I just love to maybe step out of the ICU and just sort of, as you look at yourself and human beings in general, you know that I Love the connection between the breath and the heart rate and the bare reflex and all this amazing stuff going on. And just kind of, how does variability inform you, looking at human beings in general, whether it's in the patient land in the ICU, or somebody just living their life out there, trying to be the best parent, worker, spouse, human being they possibly can?
[00:24:55] Speaker B: Yeah, it's a excellent question and this may be of interest to the subscribers of your podcast because like a lot of people have linked HRV to autonomic nervous modulation of cardiac function. And I think that's a big part of it for sure.
But I also think there's a lot more to variability analysis.
So as you know, there's many, many different techniques that evaluate heart rate variability, time domain, frequency domain, and then there's these complexity domain measures which evaluate the degree of information in heart rate as well as the patterns of the characteristic patterns of hrv. And one of the unique things that blew me away that I told you about originally, that I saw from this lecture, is the concept of power law fluctuations, which is the fact that that variability has its fractal characteristics, which is multiscale self similarity.
So if you have a recording of heart rate variability over 300 minutes and you magnify it to 30 minutes, you see the same pattern, like you can't visually distinguish between 300 minutes to 30 minutes and even three minutes, there's going to be the same pattern, lots of little variations, a few medium sized variations and rare bigger variations.
And that pattern of fractal multiscale self similarity is true in nature everywhere you look. And we have it in anatomy, in terms of our tracheobronchial tree, our vascular networks, they're all fractals. If you look out the window, trees.
And then there's also non biologic structures that demonstrate fractal patterns. So cloud formations, you know, waves on the ocean or a lake or coastlines, solar flares, they're all fractal patterns, bounded fractals.
So I became interested in, well, how could this be? How could you have fractal patterns of heart rate and respiratory rate fluctuations, but yet they're also true in these non biologic systems.
So it became clear that it had to be a physical principle that explained why we have patterns of variability, not a biologic principle.
And I became interested in something called the principle of maximum entropy production, which arose in the science of atmospheric science, but it's now been applied in a whole host of other areas.
But basically what it, what it.
So I'm going to just Jump ahead a little bit and tell you answer your question. Is what you said, what is variability telling you about a human being?
Well, I think it's providing you a measure of your overall health which is reflecting your function at baseline and your ability to augment it. Okay.
So it's basically function and adaptation or adaptability.
And so a high heart rate variability tells you that you're healthy at rest, but you can also run up the stairs if necessary.
And, but, and so, and really the way we can measure that is through resting oxygen consumption.
And people often call that resting energy expenditure or REE and maximum oxygen consumption or VO2 max, which is the. So, so a high resting energy expenditure and a high VO2 max reflects a high degree of health. Okay, and, and, and going back to that also is our, I think it reflects our capacity to produce entropy. So we produce entropy by burning oxygen to carbon dioxide.
And, and by, so a high, high sort of resting metabolism and a high maximal metabolism is a kind of a sign of health. Now if you have a reduction in maximal or an elevation in basal, either of those things will reduce that distance and then you're going to have a reduction in your heart rate variability and respiratory rate variability as well.
And going back to the concept of what is healthy, well healthy in terms of variability is having a high degree of variability and a high degree of complexity of variability. So you have those multi scale self similar patterns of breathing as well as your heart rate.
And I think that the reason why we have that is again a physical principle, because again this is a Hypothesis, it's not 100% proven.
Is that the reason why we have this fractal heart rate variability and fractal structures in our lungs, tracheobronchial tree is that that's when you deliver a pulsatile force into a fractal. So a fractal pulsatile force into a fractal, an anatomic network.
You're going to maximize your capacity to deliver oxygen to the tissues and clear carbon dioxide from the tissues as well.
And by the way, I'm on call today and that was my pager going off, so I'm on call for virtual critical care. But, but I'll tell you the other thing that I think is also cool about envisioning human health as human entropy production and I have several papers written or written about this, but is that it's not just our metabolism that reflects our health. The other thing that we do of course is our cognition, our consciousness and it's the amount of Information that we take in to our brain and converting that to memories, thoughts and ideas, theories like all of that represents a reduction in the quantity of information and reduction of information is also entropy production.
We're getting a little bit off topic from variability.
So I think that variability gives you a measure of overall human health which I think can be quantified as human entropy production.
And you can measure human entropy production by the way, by our heat production divided by our temperature. So delta S equals delta Q over T and you can measure that in a lab.
And it supports a lot of the concepts that we have in physiology and neuroscience.
But it's a kind of a unifying kind of approach and again it's a holistic approach as well based on complex systems research paradigms, time.
So I think, I mean and as we know with our like I have a whoop monitor as well that there's a variety of things that we can do to improve our hrv. You know, and these are things that are well known. You know, good sleep, you know, getting good exercise, eating well, hydrating well, you know, those are the critical things. And I think interval style type training can really help with hrv.
So having maximal expense for short periods of time where you, you know, increase your heart rate and respiratory rate and get completely out of breath with exercise, that's really helpful. I often suggest to my patients to do interval walking training in preparation for surgery where instead of going out for a slow walk for 30 minutes, they just, just push themselves for three minutes to go as fast as they can.
And so bringing variability into your walking is helpful to bring to enhancing your variability of your heart rate and respiratory rate. So.
But that was a little bit of a long answer to your excellent question. But again, I have different papers on the topic if you're interested in reading more. But it's an area of interesting exploration at this time.
[00:34:56] Speaker A: Yeah, and I'm curious, I mean just mind blowing stuff there.
And we'll have to. Please send me the links to those papers because I'm sure the audience would love to explore those.
Is there anything in this way of thinking which I just the nerd side of me, my head's just blowing up right now that that may make you look at human health wellness. I mean you talked about adding variability into like the walking piece but like with bringing in consciousness there, which I probably don't need to tell you is a fascinating, controversial, a little bit like be just because we can't really man, you know, consciousness is one of those elusive things that won't show up on an FMRI for us in a perfect way. But I just kind of wonder as you think deeply about this and boy, the. I know I. You didn't want to be called a physicist, but yeah, I could see a young budding physics major in undergrad in what you're saying. Just any unique insight that this gives you into human health and something maybe beyond the sleep, movement, nutrition, anything around there that may bring some insight into your thinking?
[00:36:27] Speaker B: Well, if human health is about having a high resting metabolism that has a capacity to go to a maximal metabolism, it means that pushing yourself and your capacity to augment your metabolism is a good thing.
So we already talked about that.
And so if you live on the 10th floor, walk up the stairs once a week and get completely out of breath as you do.
But the concept about the fact that we are taking in information, what I mean by that is that you're taking in sensory information all the time. So auditory information, olfactory information you're reading, you're digesting and there's a lot of sensory information that we're experiencing every single moment.
And so I think there's value in like we know, for example, taking walks in nature is really therapeutic for mind and body. And I think that one of the reasons for that is because you're exposed to such a robust, complex informational kind of scene.
When you're seeing the beautiful vista of a, you know, on the top of a mountain or a beautiful ocean view, that act of just seeing that whole expanse of information I think is therapeutic. And I was told that taking walks in nature is absolutely proven a methodology to improve mental health.
And so I think that this provides a scientific kind of background for that because you're optimizing your entropy production by taking in more information and converting and reducing it.
Now again, that's very hard to prove that, but that's so I think expo, like that expression, everything in moderation.
I think that there's some value to having a diversity of exposures physically and from an information processing. Going to see a concert where the music is complex or having, you know, we all, you know, like, like enjoy eating or smelling food. That's, that's interesting and complex. Like I think all of those things really do contribute to, to human health and well being.
But, but it's. And, and I think that variability is just reflective of, of that the complexity of your, of your internal system. Like in a way, I think of heart rate variability and respiratory variability as reflecting all the complex feedback loops that are within our body that are healthy.
And so it's kind of a biomarker that's kind of indicating are things working well.
And I think, I think yeah, that's the way I would answer your question on that.
[00:40:08] Speaker A: Matt, I love this and I just have to ask this question because of where this conversation's gone. But if you haven't kind of thought about from this perspective, that's okay too. But I just gotta ask while I have you, I'm curious if you have any thoughts on HRV biofeedback residents frequency breathing, coherence breathing?
Because when we also have these systems that we can, while they're mostly autonomic, we can take over the wheel and synchronize like resonance frequency breathing, the, our heart rate, our breath rate, our bare reflex.
I'm really nerding out about piezo 2 receptors now. I think we're sinking those as well though. We need to, need a lot more research for me to make any statement with confidence there. It's speculation only. But like these, getting these in sync and I just kind of with, with your thinking very deeply about this, I, I just love if. Yeah, I'll just kind of open it up. Do you have any, you know, thoughts or perspective on the benefits of HRV biofeedback in the models that inform your thinking?
[00:41:23] Speaker B: Well, first let me say that I have no experience in any of that kind of work, but I do think it could be really beneficial. And so you've alluded to the fact that the breath is the mechanism, it's really the connection between mind and body.
And so I think that, and for thousands of years, and truly thousands of years, people have practiced breathing techniques in order to improve mental health and well being, improve performance.
And so I do see a value in actually using a biofeedback mechanism to kind of evaluate the response to a variety of breathing techniques.
And one of the principles, of course, complex system is every individual system is its own unique system.
So you may be, you know, you may respond better to certain types of breathing techniques compared to others. So most, many breathing techniques are about slowing the out breath.
And of course slowing the out breath will activate the parasympathetic nervous system and enhance heart rate variability. And so, but, yeah, but, but slowing the out breath could be four, you know, box breathing, you know, do the four count in, hold your breath for four, out for four and then hold for four, et cetera.
But there's also four, seven, eight breathing, which is, you know, you hold for seven and breathe out for eight after inspiring for four.
And then there's like the WIM HOF breathing, of course, that is, there's hyperventilation followed by breath holding and there the physiology is interesting. It's, you have a slight elevation in your CO2 if you are breathing less and that is beneficial for the lungs, for oxygen delivery, for vasodilation.
So I think that, but I think that HRV would be a great way to evaluate whether those things are working for you.
I think that for people that suffer a bit from anxiety or fear or even other emotions that are a little bit overly strong, I think HRV and breathing techniques can really help with those.
So, yeah, I think it would be great.
And the key to a biomarker, by the way, I think is that it helps you accomplish something. Like if it's just telling you that, oh, you had a bad sleep last night, but there's no, you know, you need a mechanism to improve.
[00:44:20] Speaker A: You're going to feel terrible today whether you want to or not. Yeah,
[00:44:25] Speaker B: yeah.
But it, but yeah, I think there's a really exciting sort of future in that. And I see it like a lot of friends or just people on the street have HRV monitors and they're tracking it. It I know one person who had a marked drop in their HRV and they were subsequently diagnosed with an infection.
So that definitely happens as well.
But it's remarkable how despite many years of research, it's still not applied in everyday healthcare. It's just not on the radar for most doctors.
[00:45:08] Speaker A: Yeah, excellent. Well, I love to ask this as sort of a wrap up question.
There are plenty of rabbit holes we could explore and the invitation to come back is wide open.
But I would love to see, I'd love to just ask, I love asking thinkers like you this question. Let's say you walk in, the studies you're doing all turn out to be positive.
You walk into an ICU 15 years from now. Let's say AI continues to make progress the way it has, it gets well integrated, let's just say for sake of argument, into the medical environment.
What do you think looks different if you were to look forward into a future that you're actively day to day, through your research, through your work, trying to bring to reality, what's the ICU look like 15 years from now if you and folks like you are successful in and proving this to be an invaluable tool.
[00:46:16] Speaker B: So in healthcare, we are evaluating patients, identifying problems that merit investigations and we make decisions regarding the need for testing or the need for interventions that we believe to be therapeutic for patients.
What I Would see is that for major decisions that we are making for a patient, like taking them off a breathing machine, do we do that today or tomorrow? Or do I put them on a type of dialysis machine? There's different types of dialysis.
Or does in fact the patient need to go on a breathing machine?
Or, you know, do they.
Are they responding favorably to my antibiotic therapy that I've just started them on?
I see all of those decisions as being assisted by a form of clinical decision support which is offering added value to the physician, saying, oh, the risk of this event is X.
And you know that the risk of that event was 5% less yesterday. So I've improved things today.
So it's giving me sort of probabilistic prediction of the future that is enhancing my own clinical judgment, my own clinical.
I think that that's where AI can help is by enhancing my capacity to predict the future, knowing that the future is always uncertain. There's no certainty.
But ultimately, and I think AI could help in offering explanations to patients as well of various conditions. I think there's a patient. There's an education aspect that can be helpful. I mean, patients already do that by searching for information.
But having regulated or approved tools that are really optimized for that, I think could be helpful.
But ultimately, I think that people.
So it'll make physicians better, but it's not going to replace them. If we're having a discussion with a family about the fact that there is unfortunately no hope for survival for a particular patient, they want to hear that from a fellow human being. And I think that that's not going to change.
If a surgeon is doing surgery, can we be assisted by tools in the operating room? Absolutely. But is it going to replace the surgeon? No, I don't see that.
So I think that others have said that it's Dr. Plus. AI is going to be better than Dr. Alone, but it's not going to replace the doctor. There are unfortunately a variety of areas where AI really does threaten to eliminate jobs.
But I think it may help help make doctors more efficient and hospitals more efficient.
But I don't think it'll fundamentally, you know, change the patient, the doctor patient relationship. And that's fundamental to patient care.
[00:50:10] Speaker A: I love that. What a beautiful way to end it. I love the work that you're doing because as somebody who just. I love heart rate variability, I found this science. You know, as we shared, we were talking before from the mental health perspective, but obviously recognizing that, can you really separate mental and physical health? I.
[00:50:28] Speaker B: No, no, no. Descartes.
[00:50:29] Speaker A: We can't. Right. You know, it's, it's all so inner. It's connected and we're, we're a holistic system. And you know, when I, when I talked, I mean there's a lot of great HRV research being done in emergency rooms and ICUs maybe the most studied place of medicine because we got people hooked up to machines, pretty much the whole experience.
And where your work, I believe informs the future too is. Yeah, HRV might not be one of the top five things for any reason you're in the ICU that you'd want a human alone to focus on. But what a complementary thing that it could be if you provide it to the, you know, the, the physician, the surgeon in a meaningful way that, that you get. And I'm not the hugest fan on readiness scores sometimes because it's. That allows maybe for too loosey goosey with the research, but knowing your approach that hey, yeah, here's their rmssd, here's their high frequency. But this is maybe where you need to look out for with the breath rate variability.
That's where I think the work you're doing is going to take all this science of HRV and give the provider a screwdriver level thing to say, okay, I've got all this other data for whatever disease state the person's in.
This is a real supplementary of the trending or like you said, when to take out the breathing assistant and all that stuff. So I just want to thank you for your work and sharing it with the audience.
[00:52:03] Speaker B: It's my pleasure. And Matt, thank you so much for your interest and your enthusiasm, your curiosity and your excitement with hrv.
And it's been a pleasure having this discussion with you and good luck to you and I look forward to hearing more on your podcast in the future.
[00:52:24] Speaker A: Thanks so much, Dr. Seeley. And I'll bug you for those links to your past articles because I know the request I will get.
So we'll put, as always, information links in the show notes. And Dr. Seeley, thank you so much for your time, your work with this and as always, we'll see you soon on the heart rate Variability podcast. Thank you everybody.
[00:52:46] Speaker B: Fantastic.
