Latest / No Brain Behind / The Myth of the Average Patient
Transcript
- 1 00:00:00,000 --> 00:00:04,640 On behalf of the Society for Equity Neuroscience, welcome to No Brain Behind. 2 00:00:04,640 --> 00:00:13,520 The mission of the Society for Equity Neuroscience or Sequence is to enhance brain health 3 00:00:13,520 --> 00:00:19,600 equity through scientific research. This podcast is dedicated to conversations with equity-focused 4 00:00:19,600 --> 00:00:25,520 neuroscientists and thought leaders. Our guests are working to understand and address barriers 5 00:00:25,520 --> 00:00:32,080 to and facilitate of optimal brain health and nervous system function for everyone, everywhere. 6 00:00:32,080 --> 00:00:37,360 No brain behind explores cutting-edge research methods, amplifies new discoveries, 7 00:00:37,360 --> 00:00:43,040 embossed as collaboration within the scientific community. I'm Dr. Githangeli Doss. 8 00:00:43,040 --> 00:00:49,200 I am Dr. Annalkadeh Nathihada and I'm Dr. Chris Orlando. We are a clinician scientist in the 9 00:00:49,200 --> 00:00:55,760 field of health equity neuroscience who have the distinct pleasure of being your host on No Brain 10 00:00:55,760 --> 00:01:03,680 Behind. Join us as explore the scientific challenges and innovations shaping the future of equity 11 00:01:03,680 --> 00:01:11,920 neuroscience. So welcome everyone to another episode of No Brain Behind. I'm your host, Annalkadehada. 12 00:01:11,920 --> 00:01:16,720 And today we are exploring what really shapes our cardiovascular and brain health from how we 13 00:01:16,720 --> 00:01:22,080 sleep and behave to environmental and social conditions in which we live and how emerging 14 00:01:22,080 --> 00:01:26,960 technologies may help us turn that knowledge into more precise and personal ice care. 15 00:01:26,960 --> 00:01:32,240 I'm honored today to be joined by Dr. Azizi Seixas, professor of psychiatry and behavioral 16 00:01:32,240 --> 00:01:37,280 sciences at the University of Miami, Miller, School of Medicine, director of the Media and 17 00:01:37,280 --> 00:01:42,480 Innovation Lab, co-director of the Center for Translation and Sleep in Surgery Sciences 18 00:01:42,480 --> 00:01:48,320 and share the department of informatics and health data science. Dr. Seyesh was asked a fascinating 19 00:01:48,320 --> 00:01:54,400 question. What really shapes our health over time? His research looks beyond biology alone, 20 00:01:54,400 --> 00:02:00,160 examining how we sleep, behavior, environmental, clinical factors, and other conditions 21 00:02:00,160 --> 00:02:06,240 might influence cardiovascular health and brain aging. He's also working exploring how machine 22 00:02:06,240 --> 00:02:12,240 learning, artificial intelligence, remote monitoring and real world data can help us turn that 23 00:02:12,240 --> 00:02:17,440 knowledge into interventions. Dr. Seyesha, thank you so much for joining us today. It's really a 24 00:02:17,440 --> 00:02:22,880 privilege. Oh, thank you. Thank you so much. I'm here for having me. It's really a pleasure. 25 00:02:22,880 --> 00:02:28,800 What I love about this conversation is that it gets that's our question. I've been asking throughout 26 00:02:28,800 --> 00:02:35,760 my entire career, which is what actually shapes health over time. Because we often think about 27 00:02:35,760 --> 00:02:42,240 health as something that happens inside the body. That's right. But really health is a really 28 00:02:42,240 --> 00:02:50,160 being shaped continuously by a variety of different factors, some that are exogenous, some that 29 00:02:50,160 --> 00:02:58,800 are endogenous. I'm really excited also to discuss some of the work that we're doing, particularly 30 00:02:58,800 --> 00:03:05,840 using the newest technologies, whether it be wearables or digital health or AI. I'll try to capture 31 00:03:05,840 --> 00:03:12,400 the same thing, which is real world data, because many of us believe that some of the data that we 32 00:03:12,400 --> 00:03:20,400 get from clinical sources may not be representative or may not be ecologically valid. And so this is why 33 00:03:20,400 --> 00:03:27,280 we believe that while those wearable technologies are not a panacea that can address and cure all the 34 00:03:27,280 --> 00:03:33,680 eels, particularly the instrumental methodology eels. But I think it gets us closer where we can monitor 35 00:03:33,680 --> 00:03:41,440 people in their real habitats, so to speak, and continuous new measurements. I'm excited about 36 00:03:41,440 --> 00:03:47,200 kind of digging in deeper to some of those ideas. I know you mentioned you're excited, but I even 37 00:03:47,200 --> 00:03:52,800 more excited today because today I'm a student and you are my mentor and professor today. So let's 38 00:03:52,800 --> 00:04:02,720 start with the question. Do we really know about a lot of people to truly personalize their health 39 00:04:02,720 --> 00:04:15,120 or we don't? I think that's an excellent question. I think we know more about disease than we really 40 00:04:15,120 --> 00:04:22,240 know about individuals. To be quite honest, a lot of people would question even that statement. I 41 00:04:22,240 --> 00:04:30,000 think medicine has become extra ordinarily sophisticated at identifying what works primarily on the 42 00:04:30,000 --> 00:04:37,280 average person for large groups of people. And that has saved millions of lives, right? So I'm not one 43 00:04:37,280 --> 00:04:43,200 of those individuals that will say that this one size fits all approach we should throw it out 44 00:04:43,200 --> 00:04:51,520 because it has served us well. But the average patient does not actually exist. You wonder, 45 00:04:51,520 --> 00:05:00,480 why we may share some common ground where both male and the like and we have some here and the like 46 00:05:00,480 --> 00:05:06,880 and some other things, but we're very different, right? And I think two people can have 47 00:05:06,880 --> 00:05:14,640 the same diagnosis for all intents and purposes have similar characteristics in terms of age and 48 00:05:14,640 --> 00:05:22,560 even clinical lab values. Yet they may respond very differently to the same intervention because 49 00:05:22,560 --> 00:05:30,880 the interaction of their biology, behavior, environment, social context, health behaviors like 50 00:05:30,880 --> 00:05:37,760 sleep, physical activity, diet, unstressed management, and your general daily routines are just vastly 51 00:05:37,760 --> 00:05:45,280 different. So when we talk about personalized health, the goal cannot simply be to know someone's 52 00:05:45,280 --> 00:05:53,200 genetics or medical history, which we oftentimes myopically focus on. Instead, we need to understand 53 00:05:53,200 --> 00:06:01,440 what I call this whole person phenotype that means understanding who we are biologically. 54 00:06:01,440 --> 00:06:09,600 How do we behave? What environment do you live in? What are your clinical risks? And how are all of 55 00:06:09,600 --> 00:06:16,880 these things changing over time? And the last part is critical because I think traditional medicine 56 00:06:16,880 --> 00:06:24,080 gives us snapshots, right? Of someone's clinical state in terms of their persona, their profile. 57 00:06:24,080 --> 00:06:32,000 However, personalized medicine requires trajectories. So do we know enough today completely about 58 00:06:32,000 --> 00:06:38,880 personalized health? Flatly I will say no. But for the first time, I think we are developing the tools 59 00:06:38,880 --> 00:06:45,920 to move in that direction of having more precise and personalized approach. And I think the big 60 00:06:45,920 --> 00:06:55,280 shift is not asking what works for people like you and I, but really eventually asking what is most 61 00:06:55,280 --> 00:07:02,560 likely to work for you right now in the context of your life. And that's what we have created in the 62 00:07:02,560 --> 00:07:08,720 precision and personalized population health framework. And in the same interest in terms that you 63 00:07:08,720 --> 00:07:14,480 mentioned, the whole body phenotype, I like that. So we should start using that more because you know, 64 00:07:14,480 --> 00:07:23,040 some people, some providers from different disciplines, my belief that they know very well that the 65 00:07:23,040 --> 00:07:28,640 human being, the person and therefore they give all these recommendations and they get surprised 66 00:07:28,640 --> 00:07:33,280 when the person comes back and magically the patient haven't done any of that because obviously 67 00:07:33,280 --> 00:07:38,000 what affects the other patient didn't fix this one in particular. They have different jobs, they have 68 00:07:38,000 --> 00:07:45,760 different environmental factors, different things that obviously might not be compliant with what 69 00:07:45,760 --> 00:07:50,480 the physician or any other provider told them. So that's a way to... You are spot on. Yeah, you are 70 00:07:50,480 --> 00:07:56,400 spot on, but I just want to, you know, since you know, I know that your platform is very far reaching, 71 00:07:56,400 --> 00:08:02,400 I am not again just an underscore. I'm not saying how we've practiced medicine is completely wrong. 72 00:08:02,400 --> 00:08:10,560 Instead, what I'm suggesting is that, you know, you know, that method and that approach has taken us 73 00:08:10,560 --> 00:08:20,160 so far. But given the idiosyncrasies of individuals, maybe we treat this one-size-fits-all approach 74 00:08:20,160 --> 00:08:26,400 as a starting point, knowing very well as I mentioned previously that we know have the tools to monitor 75 00:08:26,400 --> 00:08:34,800 people continuously and longitudinally so that we can create more adaptive types of interventions 76 00:08:34,800 --> 00:08:41,200 where the sequence could begin with this one-size-fits-all. If it's a statin, right? Instead of just 77 00:08:41,200 --> 00:08:47,920 providing a statin that's a one-size-fits-all, maybe we use pharmacogenetics to be able to determine 78 00:08:47,920 --> 00:08:53,840 what class of statin actually works with a particular individual. When I was at NYU School of Medicine, 79 00:08:53,840 --> 00:09:00,880 one of my mentors, Benga Ogidegby, did a great classical work looking at ACE inhibitors 80 00:09:00,880 --> 00:09:06,160 and to show how it may work differently across racial ethnic groups, particularly people 81 00:09:06,160 --> 00:09:12,800 of different genetic ancestries. And so that could be a starting point because that would then 82 00:09:12,800 --> 00:09:20,880 lower the level of one side effects and generally side effects has a negative feedback loop because 83 00:09:20,880 --> 00:09:28,320 when people have or experience adverse side effects, they're less likely to comply with medication 84 00:09:28,320 --> 00:09:33,520 adherence, right? People don't just say I'm not going to take the medicine per se. If you make the 85 00:09:33,520 --> 00:09:39,600 case that it is going to be of benefit to them, they will likely take it. But if it gives them brain fog 86 00:09:39,600 --> 00:09:45,680 or if it makes them feel lethargic and tired or they don't see a just noticeable difference over a 87 00:09:45,680 --> 00:09:52,000 period of time, then they're less likely to take it. So I think if we can use precision at the outset in 88 00:09:52,000 --> 00:09:59,680 the beginning to then lead to greater outcomes sooner, then it can essentially portend long-term 89 00:09:59,680 --> 00:10:06,560 adherence. I totally agree. That's something we all should start working harder in every case. 90 00:10:06,560 --> 00:10:14,800 As I have followed your work recently, I know your career has brought together so many different 91 00:10:14,800 --> 00:10:19,760 areas and you have mentioned sleep medicine, psychology, cardiovascular and brain health, 92 00:10:19,760 --> 00:10:27,040 population now, AI that you have into reading your work as well. So what first do you do to this work 93 00:10:27,040 --> 00:10:30,480 and how did your research able to bring all these areas together? 94 00:10:30,480 --> 00:10:37,520 Wow, that's an excellent question and you know, I don't know if we've had enough time to be quite 95 00:10:37,520 --> 00:10:47,680 honest, but I'll try my best. I have a very eclectic training background, and I'm going to start from 96 00:10:47,680 --> 00:10:54,000 undergrad. I went to college other holy cross in Worcester, Massachusetts as philosophy and 97 00:10:54,000 --> 00:11:03,200 pre-med and philosophy really allowed me to think deeply and critically about unanswered questions, 98 00:11:03,200 --> 00:11:10,640 but more importantly, it really helped me to create and to develop the right question, not necessarily 99 00:11:10,640 --> 00:11:17,120 the right answers, the right questions. And then from there, it evolved into psychology and then 100 00:11:17,120 --> 00:11:23,120 it involved into medicine and then it evolved into engineering and it evolved into statistics and 101 00:11:23,120 --> 00:11:31,440 machine learning and AI and digital technology. And essentially, you know, that evolution really 102 00:11:31,440 --> 00:11:38,480 was not serendipitous although, you know, kind of living it, it felt serendipitous, but I had an 103 00:11:38,480 --> 00:11:47,200 unquenchable first to seek after, you know, a prevailing, you know, research question and it's more 104 00:11:47,200 --> 00:11:55,440 of a human question. I want to find the best ideas and solutions to improve the health and well-being 105 00:11:55,440 --> 00:12:02,560 enough others. And I don't believe that one particular discipline i.e. medicine should have a 106 00:12:02,560 --> 00:12:09,600 monopoly over that, that I believe in this concept of arbitrage where that ideas in another discipline 107 00:12:09,600 --> 00:12:18,640 when transposed to another discipline may even have greater value and utility. And we see this in AI 108 00:12:18,640 --> 00:12:24,880 today, particularly neural networks, you know, neural networks have been eroned for decades, 109 00:12:24,880 --> 00:12:32,240 but once they were brought into financial services and health care, we started to see greater value 110 00:12:32,240 --> 00:12:39,280 and utility of them. So personally, my work really evolved realizing that the problems I cured 111 00:12:39,280 --> 00:12:47,360 of both did not respect these disciplinary boundaries, right? And particularly why I centered 112 00:12:47,360 --> 00:12:55,680 around sleep is because I felt that sleep was a foundational phenomenon. Is it biological? 113 00:12:55,680 --> 00:13:02,000 Is it behavioral? Is it something more esoteric, right? Where it's this unconscious experience? 114 00:13:02,000 --> 00:13:09,520 Despite that, we won't get into that philosophical debate per se because I think many people, 115 00:13:09,520 --> 00:13:15,920 you know, depending on what group they belong to, we'll have a valid view as to whether or not if 116 00:13:15,920 --> 00:13:22,880 it's biology or behavior or something more mystical. But what was regardless of those disciplines, 117 00:13:22,880 --> 00:13:30,560 what became evident to me was that sleep was primordial, that without it, we could not function. 118 00:13:30,560 --> 00:13:38,080 Why? Because all of our biology depended on that. And if we want to extend sleep beyond sleep in 119 00:13:38,080 --> 00:13:45,360 terms of that seven to nine, or was that we typically, you know, get on a given day, if we brought 120 00:13:45,360 --> 00:13:52,720 me to someone circadian rhythm under biology that everything works on clock work in terms of how 121 00:13:52,720 --> 00:14:00,160 your cells operate and your tissue and your organs on the like, that to me made me realize that 122 00:14:00,160 --> 00:14:12,080 a human being that really is just a speck of dust in this vast place that we call this world, 123 00:14:12,880 --> 00:14:20,320 that sleep was one of the conditions and states that we don't have a lot of control over, but yet 124 00:14:20,320 --> 00:14:27,680 still we need it, right? And then I realized that sleep in many ways seems very counter culture, 125 00:14:27,680 --> 00:14:35,040 very antithetical to Western civilization, right? Because Western civilization reveres being 126 00:14:35,040 --> 00:14:42,320 active and being productive and sleep was that one state where the body said, no, I am in 127 00:14:42,320 --> 00:14:50,160 control, you need to rest, you need to restore, you need to rejuvenate, you need to consolidate 128 00:14:50,160 --> 00:14:57,360 brain function and all those other key components that our bodies can't do and it's learned some 129 00:14:57,360 --> 00:15:04,720 our consciousness can't. And so for me sleep became this fulcrum, this pivot, this axis by which 130 00:15:04,720 --> 00:15:12,240 everything kind of emanated, right? And so for me, I then looked at sleep in its multifaceted 131 00:15:12,560 --> 00:15:18,240 way because stress affects it, work affects it, neighborhood affects it, environmental factors 132 00:15:18,240 --> 00:15:24,720 affected, you know, and then I realized how so many different cardiovascular, you know, so many 133 00:15:24,720 --> 00:15:29,840 different health outcomes are cardiovascular disease, brain health outcomes, autoimmune disease, 134 00:15:29,840 --> 00:15:37,360 cognition and just general function and all of that really is significantly affected by sleep, 135 00:15:37,360 --> 00:15:43,440 which is why I have said sleep is an omnifinaminone because it could be a risk factor, it could be a 136 00:15:43,440 --> 00:15:49,360 correlate, it could be a predictor, it could be a correlate, it could be a consequence of so many 137 00:15:49,360 --> 00:15:58,160 things that we do. And so for me, you know, I think it's important to then say start from there 138 00:15:58,160 --> 00:16:05,120 and then find ways in which we can understand the human being. So very early on it became, 139 00:16:05,840 --> 00:16:11,600 you know, difficult for me to study sleep with though also studying cardiovascular disease, 140 00:16:11,600 --> 00:16:16,480 population on health and environmental factors on social conditions. And then when I realized 141 00:16:16,480 --> 00:16:20,800 and said, hey, we needed the best technologies and the best solutions to measure these people in 142 00:16:20,800 --> 00:16:27,920 real world, you know, settings and to capture it in a more ecologically valid way. And so, you know, 143 00:16:27,920 --> 00:16:33,280 question and emerge, if health is being shaped by all of these interacting factors, how do we 144 00:16:33,280 --> 00:16:39,920 actually measure them? And that's what led me to informatics and wearable technology and 145 00:16:39,920 --> 00:16:47,280 remote money, turn a machine learning and AI. So, you know, I don't really see my career as moving 146 00:16:47,280 --> 00:16:55,040 from one field to another. I see it as following the problem, the problem kept telling me you cannot 147 00:16:55,040 --> 00:17:01,360 understand the person by studying one organ, one behavior, or one data point in isolation. And 148 00:17:01,360 --> 00:17:05,840 that's really what brought all of these seemingly disparate areas together. 149 00:17:05,840 --> 00:17:10,960 No, and that's interesting that you mentioned because as a neurologist, we always 150 00:17:10,960 --> 00:17:17,920 told our patients or family members how important our sleep is. You know, most of the people 151 00:17:17,920 --> 00:17:22,800 believe sleep is very simple. When we are in training, even though we made those recommendations 152 00:17:22,800 --> 00:17:27,440 to our patients, we don't comply with them either because of how the system is built. As you 153 00:17:27,440 --> 00:17:33,600 mentioned, we just want to be active, active, productive either personally, either as academic, 154 00:17:33,600 --> 00:17:39,920 environments. So, it's a simple turn on the surface, right? But it's a very complex 155 00:17:39,920 --> 00:17:48,560 health issue that we have nowadays. And I know you mentioned that sleep can be either a 156 00:17:48,560 --> 00:17:56,160 cause can be an effect, it can be just a vice-standard, it can be anything in the process. And, you know, 157 00:17:56,160 --> 00:18:03,840 what it seems so simple, okay, I fed the heart, I fed the brain, I fed the body, I fed our 158 00:18:03,840 --> 00:18:10,400 side. So, so many important aspects, sometimes we might just ignore either because we believe 159 00:18:10,400 --> 00:18:14,400 that other things are more important. We pay so much attention to blood pressure, which is 160 00:18:14,400 --> 00:18:19,840 totally correct. So, you know, a person that's been on the sleep can have like high 161 00:18:19,840 --> 00:18:26,880 protection and that's probably the cause, right? So, there are so many abstract issues that sleep 162 00:18:26,880 --> 00:18:33,920 because we can entirely measure it correctly, but we so much important, so any part of our health, 163 00:18:33,920 --> 00:18:38,960 and that we don't pay attention enough. And, you know, your research really looks at the whole 164 00:18:38,960 --> 00:18:45,200 picture as the answer you just gave me, and not only biology and behaviors of conditions. So, 165 00:18:45,200 --> 00:18:51,520 do you mention health, brain health? So, when it comes to dementia and cognitive aging, 166 00:18:51,520 --> 00:18:55,920 how do all these factors come together to influence our brain health? 167 00:18:55,920 --> 00:19:02,960 The easiest way to understand brain, you know, brain health is to start thinking of dementia, 168 00:19:02,960 --> 00:19:09,280 something that certainly begins late in life. You know, brain health in my home breast 169 00:19:09,280 --> 00:19:16,080 animation really accumulates across decades, and I think we're seeing, you know, studies, you know, 170 00:19:16,080 --> 00:19:23,840 provide evidence of that. And, you know, one, you know, you know, published a literature 171 00:19:24,560 --> 00:19:35,040 that really has, I want to say, this in perspective, I almost use the word canonized, but I think there's 172 00:19:35,040 --> 00:19:41,920 more to, there's more evolution in this line of work that that that I think it's not canonized, 173 00:19:41,920 --> 00:19:48,240 but it's quite a canonical type of research. It's the 2024 Lancet Commission on Demential, 174 00:19:48,240 --> 00:19:53,920 where they identified 14 potentially modifiable risk factors across the life course. 175 00:19:54,880 --> 00:20:01,200 Like the hypertension that you just mentioned, diabetes, you know, or cardiometabolic health, 176 00:20:01,200 --> 00:20:09,280 risk or metabolic syndrome, physical inactivity and smoking and obesity, depression, hearing loss, 177 00:20:09,280 --> 00:20:19,600 and, you know, untreated vision loss and hearing loss and social isolation and, you know, history of 178 00:20:19,600 --> 00:20:26,720 brain injury. And some other, you know, health outcomes as well too, you know, like, you know, 179 00:20:26,720 --> 00:20:32,560 people who consume alcohol. And I believe your pollution is an environmental factor. 180 00:20:32,560 --> 00:20:39,200 And then people focus on cognitive reserve, meaning people with lower education, 181 00:20:39,200 --> 00:20:46,160 you know, attainment rather are more likely to have increased risk for dementia. 182 00:20:46,800 --> 00:20:53,040 And one of the things that I believe the Lancet Commission did was that, and I think this is why I felt 183 00:20:53,040 --> 00:21:03,280 it was quite, I mean, quite a kind of a formative line of research, you know, so I work very closely with, 184 00:21:03,280 --> 00:21:09,360 you know, one of the best, you know, human genetic groups, you know, in the world, 185 00:21:10,800 --> 00:21:19,360 led by Dr Margaret Peric-Vance. And even her, she was the one who identified in this public APU, 186 00:21:19,360 --> 00:21:31,120 and even her, she believes that there is ex-posomic types of mechanisms that interact with genetic 187 00:21:31,120 --> 00:21:40,080 risk factors. Because the Lancet even also said that about 45% of dementia cases could be 188 00:21:40,080 --> 00:21:47,920 attributed to some of these modifiable factors, right? And so modifiable doesn't necessarily mean 189 00:21:47,920 --> 00:21:56,160 preventable, but if you were to kind of do that, you know, kind of epistemological logic to say, 190 00:21:56,160 --> 00:22:01,120 well, if it's modifiable, then it could potentially be prevented. And I think that has really given 191 00:22:01,120 --> 00:22:09,200 lots of new energy in dementia research to be able to say, hey, you know, what if we can better 192 00:22:09,200 --> 00:22:19,440 understand early stage dementia and brain health, vis-a-vis, you know, these modifiable risk factors, 193 00:22:19,440 --> 00:22:27,920 maybe it can lower the prevalence on incidence of neurodgenerative conditions, like dementia, 194 00:22:27,920 --> 00:22:34,880 specifically Alzheimer's disease, unrelated dementia, Parkinson's, and the like. And I think that's 195 00:22:34,880 --> 00:22:40,400 key. So I think, you know, hypertension, nothing's the brain, physical activity, 196 00:22:40,400 --> 00:22:45,760 effects, the brain's sleep, effects, metabolism, and the like environment, influence, 197 00:22:45,760 --> 00:22:50,880 these opportunities for physical activity under some, you know, research that has shown that, 198 00:22:50,880 --> 00:23:01,840 you know, you know, consumption of of of neurotoxins, as well as some other environmental factors 199 00:23:01,840 --> 00:23:12,000 through ear pollution and plastics may have some deleterious effects on general brain health, 200 00:23:12,000 --> 00:23:18,880 and may increase people's risk. And I think, you know, for me, I think of brain health as cumulative 201 00:23:18,880 --> 00:23:26,400 result of biology interacting with behavior, with environment over time. And that is why a purely 202 00:23:26,400 --> 00:23:33,120 neurological approach is incomplete, you know, and I think, you know, speaking to a neurologist, 203 00:23:33,120 --> 00:23:40,880 you know, I'm sure you would certainly, you know, you know, acknowledge that that that we need to 204 00:23:40,880 --> 00:23:46,240 take into consideration all these other factors because the brain doesn't live separately from the body, 205 00:23:46,240 --> 00:23:54,000 and the body does not live separately from society. And that's this whole person framework that I 206 00:23:54,000 --> 00:24:00,720 have been pushing is needed, but we have to figure out ways in which to architect how do we capture 207 00:24:00,720 --> 00:24:06,240 this whole person? It's not a theoretical thing, and I think that's what separates our research 208 00:24:06,240 --> 00:24:15,520 program from others, right? It's the fact that we believe that all of these multi-level, multifaceted 209 00:24:15,520 --> 00:24:22,800 factors are important, but how do we measure it? How do we collate those data? How do we create them? 210 00:24:22,800 --> 00:24:29,200 How do we process them? How do we analyze them? And how do we interpret them? Right? And when we 211 00:24:29,200 --> 00:24:34,880 interpret, how do we then disseminate it to the public? That is the line of our work, and that's 212 00:24:34,880 --> 00:24:43,920 what we think is needed in medicine today, especially trying to address this growing and vexing 213 00:24:43,920 --> 00:24:48,800 prevalence of poor brain health that we see globally. 214 00:24:49,840 --> 00:24:56,640 No, I agree with you completely, and part of the, what I got from you, what you just said, 215 00:24:56,640 --> 00:25:02,320 and what we know from before is the research you do in those areas, especially in bringing 216 00:25:02,320 --> 00:25:08,560 aging and the mentioned sleep is very complex, and it's difficult to measure. That's why sometimes 217 00:25:08,560 --> 00:25:14,160 we have different studies, they have different results. And it came because they measure things 218 00:25:14,160 --> 00:25:22,000 differently, and that's why it's difficult to put it all together. And you know, going back to 219 00:25:22,000 --> 00:25:31,120 the initial point that I asked you regarding the personalized medicine, and we, as I mentioned earlier, 220 00:25:31,120 --> 00:25:38,480 we give recommendations to many different people, family, patients, colleagues about exercise more, 221 00:25:38,480 --> 00:25:47,360 it's better, it healthier, it's stress, it doesn't work for them, maybe it works for me. So for you, 222 00:25:47,360 --> 00:25:53,680 since you've been working on these for quite a while, how do you describe a personalized 223 00:25:53,680 --> 00:26:02,000 behavioral intervention in clinical practice? Yeah, excellent. I think this is where I think medicine, 224 00:26:02,000 --> 00:26:07,120 you know, as I mentioned earlier, can move away from generic recommendations. 225 00:26:07,840 --> 00:26:14,560 So when someone goes to, you know, a provider, what do they generally tell them? Exercise more, 226 00:26:14,560 --> 00:26:21,600 sleep better, eat better, reduce stress. And I think there's nothing incorrect with it, those 227 00:26:21,600 --> 00:26:29,520 recommendations, but sometimes they're not very actionable. A personalized intervention asks a 228 00:26:29,520 --> 00:26:36,240 different set of questions, right? So in our approach, we will ask what is preventing this particular 229 00:26:36,240 --> 00:26:44,480 person from doing or engaging or adhering to the healthy behavior? When are they most vulnerable? 230 00:26:44,480 --> 00:26:50,640 What intervention is most likely to work for them? And how should that intervention change when 231 00:26:50,640 --> 00:26:57,440 their circumstances change? And I think this is where we have led, as well as others, you know, in this 232 00:26:57,440 --> 00:27:02,160 scientific framework, called just in time adaptive interventions, or what called jitties, right? And we've 233 00:27:02,160 --> 00:27:08,880 done some other things in terms of the development of radiographic profiling and the light to be able to 234 00:27:08,880 --> 00:27:16,400 really phenotype someone psychographically and then be able to create a whole new type of sequence 235 00:27:16,400 --> 00:27:24,000 of interventions. And I think to be quite honest, this is why, you know, you know, many of the 236 00:27:24,000 --> 00:27:31,840 kind of lifestyle, you know, healthy lifestyle apps have gone, like the nooms and some of the others. 237 00:27:32,320 --> 00:27:38,160 Where, you know, they feel as if that we need to provide something a little bit more precise. 238 00:27:38,160 --> 00:27:47,280 Even in the proliferation of real new classes of drugs like GLP ones, where they seem like they 239 00:27:47,280 --> 00:27:53,920 are the miracle drugs, right? Because, you know, people can actually, you know, see significant 240 00:27:53,920 --> 00:28:02,560 reductions in weight, you know, increase kind of heart health outcomes, cholesterol levels, 241 00:28:02,560 --> 00:28:11,360 and other areas. Regardless of that, you know, GLP ones are just a bridge. What happens 242 00:28:11,360 --> 00:28:20,880 after the person essentially kind of recalibrates, they amount to GLP one receptors in their bodies, 243 00:28:20,880 --> 00:28:28,000 right? You know, at the end of this is a long term plan. It's still unknown whether or not 244 00:28:28,000 --> 00:28:36,560 if people can text GLP ones, you know, long term. What does that then portend for brain health? 245 00:28:36,560 --> 00:28:44,880 You know, there's been some evidence, you know, you know, not, you know, final, but there are some 246 00:28:44,880 --> 00:28:52,800 folks who are even shown in potentially that taken, taken, you know, kind of cholesterol medication 247 00:28:52,800 --> 00:28:59,280 may have a deleterious effect on, you know, brain health in terms of, you know, granite matter. 248 00:28:59,280 --> 00:29:05,120 So, so those are some of the things where we said that even though we have some of the best 249 00:29:05,120 --> 00:29:12,880 pharmacological treatments that these treatments for some, we have to figure out, are they a bridge 250 00:29:12,880 --> 00:29:19,120 for the individual to kind of write the ship, right? So to speak. But then at the end, what's a long 251 00:29:19,120 --> 00:29:25,200 term solution? And so this is why we've really invested a lot in a lot of healthy lifestyle, 252 00:29:25,200 --> 00:29:32,160 behavior modification, but one that is based on precision approaches. So the idea is to provide 253 00:29:32,160 --> 00:29:37,200 the right type of support at the right time based on an individual's changing state on context. 254 00:29:37,200 --> 00:29:42,800 So imagine, you know, someone trying to become more physically active, a generic app might send 255 00:29:42,800 --> 00:29:49,760 the same reminder every day at 6 p.m., right? A personalized system might learn that this person is 256 00:29:49,760 --> 00:29:58,720 most likely to walk after lunch, that rain reduces their activity, that poor sleep predicts 257 00:29:58,720 --> 00:30:06,080 lower motivation, and that encouragement from a friend increases adherence. So the intervention can 258 00:30:06,080 --> 00:30:13,040 adapt. And I think there are four things, you know, a truly personalized intervention has to get right. 259 00:30:13,040 --> 00:30:20,160 The right person, the right intervention, the right time, and the right context. And then that 260 00:30:20,160 --> 00:30:26,480 will then help with sequencing, endorsing off the intervention. That's what, you know, I call 261 00:30:26,480 --> 00:30:31,040 precision on personalized population health, others have called it precision behavioral health. 262 00:30:32,160 --> 00:30:38,160 And that's important to listen and understand, I remember when I was a stroke fellow and I was 263 00:30:38,160 --> 00:30:45,760 having my clinics every week and when I used to give all this general, generalize recommendations, 264 00:30:45,760 --> 00:30:53,200 generalistic, just exercise again, sleep better, use your CIPA machine at least for hours. And I'll, 265 00:30:53,200 --> 00:30:59,280 you know, speaking with you, I look back and say, wow, maybe this approach was all the time wrong, 266 00:30:59,280 --> 00:31:05,760 and we all do that. And maybe all the samples that you just mentioned, I will, the compliance, 267 00:31:05,760 --> 00:31:11,840 if a friend's workout review, if you exercise after lunch is better, I think those practical 268 00:31:11,840 --> 00:31:18,080 advice as soon as they're like, look simple again, probably that made that person a stick to the 269 00:31:18,080 --> 00:31:23,440 to the plan. And, and all the same as we always find in clinic, I don't do clinics anymore. I'm 270 00:31:23,440 --> 00:31:29,040 mainly in patient, I see you, but I'm just remembering all these clinics I have when I was a resident, 271 00:31:29,040 --> 00:31:35,360 and a fellow. And, and if I have to chant things, now I will incorporate those into my, into my 272 00:31:35,360 --> 00:31:41,840 clinic, back then, and for the audience, maybe you have, like, some insightful tips today to apply 273 00:31:41,840 --> 00:31:49,680 moving forward. So one thing that I'm also interested to know about you is you're working AI. And 274 00:31:49,680 --> 00:31:56,720 one thing that we need to understand is it's tremendous excitement about artificial intelligence, 275 00:31:56,720 --> 00:32:02,960 imaging, but there are also important concerns about bias, privacy, access, and the possibility 276 00:32:02,960 --> 00:32:09,440 that these technologies will actually widen existing health disparities. How do we make AI 277 00:32:09,440 --> 00:32:16,240 tool for health equity rather than another source of inequity? There's a great question. 278 00:32:16,240 --> 00:32:24,640 I want to broaden the conversation if that's okay. And, and, and, and it's really an apolitical 279 00:32:24,640 --> 00:32:32,320 statement, but I can understand how some people may consider what I'm about to say as, as, as political. 280 00:32:32,320 --> 00:32:40,640 So I think, you know, for quite some time, I actually have not used word equity. Because I think 281 00:32:40,640 --> 00:32:50,240 some will hijack it to make it mean what it's not intended. And for me, I've always said health for all. 282 00:32:50,800 --> 00:32:57,040 I want to treat everyone, right? And, and I think the reason why I am trying to 283 00:32:57,040 --> 00:33:08,640 underline that paradigmatic shift is because when you say health for all, right, you want to make 284 00:33:08,640 --> 00:33:16,240 sure that the personalization is for all, right? No. One of the things that we have done is that we 285 00:33:16,240 --> 00:33:23,760 believe that, you know, our best solution should work for everyone. But I also think some of our best 286 00:33:23,760 --> 00:33:30,480 and most sophisticated tools can actually address some of our more vexing challenges in health care, 287 00:33:30,480 --> 00:33:38,640 right? So this could cut across different demographic lines and particularly, you know, place-based 288 00:33:38,640 --> 00:33:45,600 health in terms of urban and rural, you know, differences that we see, right? And so for me, 289 00:33:45,600 --> 00:33:55,440 you know, AI must be flexible enough to represent all because once you, you know, kind of, you know, 290 00:33:55,440 --> 00:34:05,360 kind of create a laser focused vision for AI, then the question then is always in, you know, 291 00:34:05,360 --> 00:34:14,080 presented in this dichotomy of is AI good or bad, right? And I think it sometimes us human beings, 292 00:34:14,080 --> 00:34:23,680 we, we, it's easier for us to see it in that way. But, but the work that we do, we like to say that AI, 293 00:34:23,680 --> 00:34:32,960 and the different types of technologies that are aligned or similar or different, is really a 294 00:34:32,960 --> 00:34:40,800 tool, right? And so it's a tool that could be used by the entire healthcare ecosystem, 295 00:34:40,800 --> 00:34:51,280 patients, providers, payers, whomever, right? Policy makers, it can be used as, but the goal is how do 296 00:34:51,280 --> 00:34:59,120 we use it for good, right? Because at the end of the day, if we use a tool and it makes a situation 297 00:34:59,120 --> 00:35:07,760 worse, then it's no longer a tool. It's a weapon, right? And so this is why I want to broaden the 298 00:35:07,760 --> 00:35:13,760 conversation because it's, it's really a new ones to one. So, you know, for me, you know, I think this 299 00:35:13,760 --> 00:35:21,120 is one of the most important questions in, I think the entire AI conversation, AI will not automatically 300 00:35:21,120 --> 00:35:28,880 make healthcare, fair or more fear. In fact, if we are not intentional, I believe AI can automate 301 00:35:28,880 --> 00:35:36,720 on scale some, you know, disparities that already exist. And there are several ways that can happen. 302 00:35:36,720 --> 00:35:43,120 So, you know, it's well documented that if the training data doesn't represent the populations 303 00:35:43,120 --> 00:35:52,560 where the model will be used, performance can vary. And sometimes, you know, performance, you know, 304 00:35:52,560 --> 00:36:01,440 varying can be an okay thing because there are unique differences. But when it comes to diagnosing, 305 00:36:01,440 --> 00:36:09,840 when you have an algorithm to detect and to determine kidney functioning, that you don't want it 306 00:36:09,840 --> 00:36:19,280 to fairly or unfairly diagnose one group of individuals better than the other, because that's not 307 00:36:19,280 --> 00:36:24,800 healthcare for all of, right? So, it's not an equity issue. It's not healthcare for all of. We're not 308 00:36:24,800 --> 00:36:32,000 fulfilling the mission of medicine and healthcare, right? When, when someone, you know, makes that 309 00:36:32,000 --> 00:36:39,760 hypocritic oath, they're not saying, you know, I'm only going to treat this person who I identify with 310 00:36:39,760 --> 00:36:46,880 no, it's anyone who is ill. I want to take care of it. So, if technology requires expensive 311 00:36:46,880 --> 00:36:53,600 devices or broadband access, some populations also can be excluded, unfortunately. And that's why 312 00:36:53,600 --> 00:37:01,840 we've partnered with some, you know, big tech giants, we've partnered with, you know, the Amazon, 313 00:37:01,840 --> 00:37:08,400 the Microsoft, the Googles, and even T-Mortals as well. You know, we led, we created our own remote 314 00:37:08,400 --> 00:37:15,440 health monitoring solution called the mailbox. And we ran this across three NIH studies. And we had, 315 00:37:16,480 --> 00:37:23,040 we have, you know, six or seven, you know, wearable devices. And one of the things that we realized 316 00:37:23,040 --> 00:37:29,760 that broadband access was a huge deal. So, we worked with T-Mobile. We provided people during the 317 00:37:29,760 --> 00:37:37,680 duration of their remote health monitoring an iPhone, an Ami-Fi device so that in areas where 318 00:37:37,680 --> 00:37:45,520 connectivity was on issue, that they would not be excluded. And so also, if an application is 319 00:37:45,520 --> 00:37:51,680 available only in English or assumes a high level of digital literacy, access does not translate 320 00:37:51,680 --> 00:37:59,280 into meaningful use. We have, you know, you know, purposefully addressed issues around digital literacy. 321 00:37:59,280 --> 00:38:08,080 You know, we're, we don't believe that, you know, wearables or digital solutions should only be for 322 00:38:08,080 --> 00:38:13,440 a certain class of individuals. What we've gone out and done is that we've created our own, 323 00:38:14,240 --> 00:38:22,320 so to speak educational, you know, material to get someone prepared so that they can participate 324 00:38:22,320 --> 00:38:29,040 in remote health monitoring or decentralized trials. And if AI identifies risk, but there is no 325 00:38:29,040 --> 00:38:34,800 pathway to care, then we have simply created a more sophisticated way to tell someone that they 326 00:38:34,800 --> 00:38:41,520 have a problem that does not serve anyone at all. Which is why we're very bullish that in the mailbox 327 00:38:41,520 --> 00:38:47,440 that we've created, we want to make it a physical in the box. And so people who may not have access 328 00:38:47,440 --> 00:38:55,120 to a provider. And I'm not just talking people at a social economics status. I'm talking about 329 00:38:55,120 --> 00:39:03,760 someone like myself where, you know, I had to wait six months to see a specialist, right? And I have 330 00:39:03,760 --> 00:39:11,840 some influence, right? I have, you know, some privilege because our system is just overwhelmed 331 00:39:11,840 --> 00:39:22,320 with demand, right? And not enough access to enough providers, right? And when, when we do see someone, 332 00:39:22,320 --> 00:39:30,640 it's generally a quick in and out. And it doesn't give us the investigative process that is needed 333 00:39:30,640 --> 00:39:36,800 to find unique challenges, right? And, you know, I'm working with some companies doing some great work, 334 00:39:36,800 --> 00:39:44,080 you know, to be able to do this. So for me, the goal is not simply a health for all, you know, 335 00:39:44,080 --> 00:39:49,920 equity and technology. The goal is health for all, you know, and I think that, you know, through, 336 00:39:49,920 --> 00:39:56,480 you know, the five is framework. We call it the RISOMatic approach that we have developed. It was 337 00:39:56,480 --> 00:40:02,320 published in frontiers where we believe that, you know, it's not just about access, but just about 338 00:40:02,320 --> 00:40:10,480 availability. It's about adequacy. It's about acceptability and affordability. You know, where people can 339 00:40:10,480 --> 00:40:15,280 access a technology. Is it actually available or they live? Is it adequate for the clinical and 340 00:40:15,280 --> 00:40:21,920 cultural needs? Is it acceptable and trustworthy and can they afford it? So in my principle, you know, 341 00:40:21,920 --> 00:40:31,680 AI can democratize expertise only if we democratize the infrastructure around AI. That's what we are 342 00:40:31,680 --> 00:40:38,640 trying to do. You know, otherwise, we risk creating extra ordinarily sophisticated tools for the 343 00:40:38,640 --> 00:40:44,640 people who already have the greatest access to care. That's not what we're trying to do. That's why 344 00:40:44,640 --> 00:40:53,520 we say whatever we create, it must serve on. Well, first of all, I want to say that I'm learning some 345 00:40:53,520 --> 00:41:00,880 terminology today again. So first, the old person body phenotype and now health for all, I like, 346 00:41:00,880 --> 00:41:07,040 I like those better. So I started incorporating those into my daily, daily practice. 347 00:41:07,040 --> 00:41:13,760 You're very different. Yeah. Thank you for those for those terms. So, you know, you mentioned great 348 00:41:13,760 --> 00:41:22,240 things about the AI, how can we can make them useful in our society? And we have a new generation 349 00:41:22,240 --> 00:41:28,640 of clinicians, scientists, and they're beginning their careers. How much do they need to understand 350 00:41:28,640 --> 00:41:36,080 AI and that a science to remain relevant and where should they start? Oh, this is great. So, so one 351 00:41:36,080 --> 00:41:41,200 of my roles and one of the things that I've been working on over the last four years being 352 00:41:41,200 --> 00:41:45,600 cheer of the Department of Informatics and Health, Data Science at University of Miami, 353 00:41:45,600 --> 00:41:52,000 Miller School of Medicine is we run ourselves very differently. We run ourselves almost like a 354 00:41:52,000 --> 00:41:58,720 start up or a scale. And I'm not just saying this hyperbolicly, it's not for hype. We really are 355 00:41:58,720 --> 00:42:04,080 organized differently and I'll and you'll see why I'm bringing this in bringing this up. So, 356 00:42:04,640 --> 00:42:12,160 so we have five verticals, right? And five verticals are critical in any modern day, 357 00:42:12,160 --> 00:42:23,760 condemning setting, particularly in, in medicine. And essentially, the value proposition that we 358 00:42:23,760 --> 00:42:29,280 have created is that we want to train the next generation of clinician scientists and practitioners. 359 00:42:29,280 --> 00:42:36,560 They're going to be steeped in innovation. And innovation for us is we want them to be not just 360 00:42:36,560 --> 00:42:44,000 literate, not just proficient, but also dexterous in new technologies like artificial intelligence, 361 00:42:44,000 --> 00:42:51,440 but also to understand the the foundations of those like ontologies, right? Which are so critical 362 00:42:51,440 --> 00:42:58,080 for people because we want to create the next phenotype of clinician, clinicians who we call 363 00:42:58,080 --> 00:43:02,640 clinician innovators. We want to train the next generation of scientists, which we call scientists 364 00:43:02,640 --> 00:43:06,480 innovators. And we want to create the next generation of practitioners, which we call practitioners 365 00:43:06,480 --> 00:43:12,640 innovators. That is what medicine needs, right? No, we don't just need people who are just following 366 00:43:12,640 --> 00:43:18,880 this kind of assembly line, we're host type of medicine. We're going to need to be able to create 367 00:43:18,880 --> 00:43:24,800 internal entrepreneurs in medicine because if people can create and think innovatively, 368 00:43:24,800 --> 00:43:29,360 then they are going to be able to solve some of the most challenging needs in health care. 369 00:43:29,360 --> 00:43:36,880 So whether or not if you're in rural Kansas or if you're in, you know, Boston, you can find some 370 00:43:36,880 --> 00:43:42,320 innovative ways to be able to address the needs of your patients and your health care system. 371 00:43:42,320 --> 00:43:47,840 Then the second vertical is research, right? And we say that we want to be the best in real world 372 00:43:47,840 --> 00:43:54,880 data, digital biomarkers and the like, you know, we are also focused on computational and digital 373 00:43:54,880 --> 00:44:00,240 biology. So do clinicians need to be aware of this? Absolutely, because one of the things that 374 00:44:00,240 --> 00:44:05,760 we have built and we're continuing to build is a digital twin or the digital patient that yes, 375 00:44:05,760 --> 00:44:09,760 you might say, I don't need to know anything about protein binding or anything like that, 376 00:44:09,760 --> 00:44:16,000 but when we create a digital twin, your digital twin off that patient would allow you to actually 377 00:44:16,000 --> 00:44:22,560 lead in silico types of evaluations and experiments to see what drug will work for your patient. 378 00:44:22,560 --> 00:44:28,800 So you're no longer doing trial on error of medicine. So that's why I believe that clinicians 379 00:44:28,800 --> 00:44:33,680 and students need to be aware of that. And then our third vertical is clinical. 380 00:44:33,680 --> 00:44:40,400 Need to be a depth in, you know, what we call decentralized care or what we call health wherever 381 00:44:40,400 --> 00:44:45,040 you are, right? That, you know, I know you said that you're in the impatient setting, but what if 382 00:44:45,040 --> 00:44:51,520 we can create a healthcare program whereby we can offset some of the burdening of reconmore to 383 00:44:51,520 --> 00:44:57,280 health systems and then put some of that where people actually live and have some bit of task 384 00:44:57,280 --> 00:45:03,680 shifting, but all of that can be controlled through this very elaborate control systems management 385 00:45:03,680 --> 00:45:11,200 ecosystem, built on, you know, cloud or built on technology. That's what, you know, people need to be 386 00:45:11,200 --> 00:45:15,920 able to do that and they need to be a depth in implementation science and how AI can help to 387 00:45:15,920 --> 00:45:24,000 transform that. And then our fourth vertical is our own venture that many of your trainees, 388 00:45:24,000 --> 00:45:34,160 right now fellow trainees, you know, they will say, do I want to subject myself to the harsh reality 389 00:45:34,160 --> 00:45:40,240 that there's going to be an AI tool that will do some off the word that I do. And, and that's a 390 00:45:40,240 --> 00:45:47,920 reality. And I don't think that AI is going to take over per se from, you know, what physicians do. 391 00:45:47,920 --> 00:45:55,360 It will kind of offload some of the mundane tasks that they normally do, but it allows clinicians 392 00:45:55,360 --> 00:46:03,680 to be more creative. And I can see the next generation of clinicians and trainees creating companies 393 00:46:03,680 --> 00:46:09,680 that would allow us to elevate the level and the type and the quality of care. So it's not just 394 00:46:09,680 --> 00:46:16,560 about fee for service anymore or treating symptoms, but now we can probably create the next generation 395 00:46:16,560 --> 00:46:24,160 of companies that really gets at the core cause and etiology of disease. And perhaps we can actually 396 00:46:24,160 --> 00:46:32,480 anticipate when something may go wrong. And therefore you may stave off or arrest disease trajectories. 397 00:46:32,480 --> 00:46:39,360 And then the last fifth vertical is what we call service outreach policy and ethics. And not saying 398 00:46:39,360 --> 00:46:47,360 that, you know, providers need to be advocates. But in order for you to have a business, you must have 399 00:46:47,360 --> 00:46:54,960 a customer. And the patients are the customers and the consumers. But our most sacrosanct virtue is 400 00:46:54,960 --> 00:47:03,680 to do no harm, but also to ensure that is also patient centric. And so ethics and policy and service 401 00:47:03,680 --> 00:47:11,440 are core values that must be restored in medicine as opposed to treating them as just symptoms or just 402 00:47:11,440 --> 00:47:19,440 de-identified data. And I think that is where, you know, AI can help young clinicians and scientists 403 00:47:19,440 --> 00:47:26,080 and practitioners to elevate the quality, elevate the quality of medicine and research and science 404 00:47:26,080 --> 00:47:32,320 that we currently are doing. And that's the vision, our own precision education that we are leading, 405 00:47:32,320 --> 00:47:36,640 particularly our department at the University of Miami's meeting currently. 406 00:47:36,640 --> 00:47:41,440 Well, that's great. I know for sure we can apply all those tips that you mentioned, 407 00:47:41,440 --> 00:47:48,240 especially those fight columns that you describe and that we can apply to research for communities, 408 00:47:48,240 --> 00:47:54,000 our daily work. So that's fascinating. And Dr. Sace's, again, this has been an 409 00:47:54,000 --> 00:47:59,760 immensibly fascinating conversation with you. I have learned a ton today. I have a new vocabulary 410 00:47:59,760 --> 00:48:07,760 to apply every day. And we appreciate the work you do, you and your team to make the change 411 00:48:07,760 --> 00:48:13,920 and the difference in our daily lives. And I know you're making a great and positive impact 412 00:48:13,920 --> 00:48:19,760 and the health of many people, not only in Florida but around the world. And thank you for joining us 413 00:48:19,760 --> 00:48:24,880 today in Open Behind, yeah. Thank you, it's my honor and keep up the good work, my friend. 414 00:48:25,440 --> 00:48:31,280 And to our listeners, thank you for joining us on No Brain Behind, the official podcast of sequence 415 00:48:31,280 --> 00:48:36,640 until the next time. Remember to stay connected, stay inspired and keep building the future of Brain Health. 416 00:48:36,640 --> 00:48:45,280 Thank you for listening to No Brain Behind. No Brain Behind is the official podcast of the 417 00:48:45,280 --> 00:48:50,240 Society for Equity and Neuroscience. Opinions shared in this episode are those of the individuals 418 00:48:50,240 --> 00:48:55,280 being interviewed. To hear more from sequence, subscribe to the podcast, follow us on social media 419 00:48:55,280 --> 00:48:59,440 and give us feedback. And what we would like to hear in future episodes. No Brain Behind is 420 00:48:59,440 --> 00:49:05,200 brought to you by the Society for Equity and Neuroscience led by Dr. Bruce Obiagli and Liana Mesa. 421 00:49:05,200 --> 00:49:10,720 This podcast is produced by Dr. Vanessa Cooper, Rachel Foreman and Parnit Graywall. And our 422 00:49:10,720 --> 00:49:17,280 production team includes Sonia Glitcher, Tino Delamer said, Joshua Homs, Euchane Hwang, Dr. Angel 423 00:49:17,280 --> 00:49:27,840 Cadena Tejada, Gatangeli Doss and Chris Orlando are hosts.