WEBVTT - The Quantum Shift in Biomedical Discovery 0:00:02.000 --> 0:00:07.400 I'm Malcolm Glabwell and you're listening to Smart Talks with IBM. 0:00:07.720 --> 0:00:11.079 When doctor Laura Jahi began treating epilepsy patients in the 0:00:11.119 --> 0:00:16.279 early two thousands, she noticed something unsettling. Different surgeons could 0:00:16.320 --> 0:00:21.080 look at the exact same case and recommend completely different treatments. 0:00:21.440 --> 0:00:24.239 One surgeon might remove one part of the brain, another 0:00:24.440 --> 0:00:27.800 a different part, and a third might not operate at all. 0:00:28.560 --> 0:00:31.120 Doctor j Hi believed there had to be a better way, 0:00:31.520 --> 0:00:35.600 one grounded in data. That conviction set her on a 0:00:35.640 --> 0:00:38.800 path that would lead her to become Chief Research Information 0:00:38.880 --> 0:00:42.800 Officer at Cleveland Clinic and the executive program lead for 0:00:42.840 --> 0:00:47.160 the Discovery Accelerator. The Discovery Accelerator is a ten year 0:00:47.200 --> 0:00:51.720 partnership between Cleveland Clinic and IBM where researchers are using 0:00:51.760 --> 0:00:56.400 AI and quantum computing to make incredible discoveries in healthcare 0:00:56.760 --> 0:01:00.000 and life sciences. I sat down with doctor j High 0:01:00.200 --> 0:01:04.200 to explore what's happening now, what's possible with quantum computing, 0:01:04.800 --> 0:01:08.600 and where this next era of biomedical discovery is headed. 0:01:10.280 --> 0:01:12.959 Epilepsy was your kind of specialty within neurology? 0:01:13.120 --> 0:01:13.880 Correct? Yes? 0:01:13.959 --> 0:01:14.840 What led you to that? 0:01:15.600 --> 0:01:19.000 I was always fascinated by the brain. You know, it's 0:01:19.560 --> 0:01:22.840 the part of our body that leaves still to this day, 0:01:22.880 --> 0:01:27.119 the most to be discovered. So I was always intrigued 0:01:27.280 --> 0:01:33.920 by areas that leave more for discovery, and epilepsy was 0:01:33.959 --> 0:01:38.480 my pragmatic side, wanting to choose a subspecialty where the 0:01:38.560 --> 0:01:43.480 problem can be fixed. In epilepsy, there are many medications 0:01:43.480 --> 0:01:46.399 that are very effective, and there's a brain surgery that 0:01:46.480 --> 0:01:49.400 we can do to stop seizures when medicines don't work. 0:01:50.520 --> 0:01:52.080 That attracted me, you. 0:01:52.000 --> 0:01:56.600 Know, compared to other areas in neurology, like stroke, for example, 0:01:56.800 --> 0:02:02.240 or dementia, where usually the damage is more. I wanted 0:02:02.320 --> 0:02:05.040 to be able to tell my patients that you have 0:02:05.120 --> 0:02:07.200 a big problem, but here's what I can do to 0:02:07.240 --> 0:02:10.079 fix it, and epilepsy offered me that. 0:02:10.960 --> 0:02:14.720 But what would there must have been interesting and intriguing 0:02:15.200 --> 0:02:18.960 unsolved problems in neurology. 0:02:19.680 --> 0:02:22.240 Oh my gosh, it's the whole brain, isn't it. Before 0:02:22.440 --> 0:02:26.560 starting to deal with artificial intelligence and research. Right in 0:02:26.639 --> 0:02:30.720 the neurology, I'm dealing with real intelligence, the human brain 0:02:30.880 --> 0:02:34.160 and how it works and how we think and how 0:02:34.200 --> 0:02:40.560 we make decisions, and how it can grow and evolve, 0:02:40.800 --> 0:02:46.679 and so there is many untaped questions in neurology, and 0:02:47.320 --> 0:02:53.400 that's part of the fascination in it. In epilepsy in particular, 0:02:53.440 --> 0:02:57.880 it's an electrical disease in the brain, it's actually one 0:02:57.960 --> 0:03:04.760 of the conditions in neurology. That's a perfect alignment of 0:03:05.280 --> 0:03:09.720 all of the scientific disciplines. It's biology and physics and 0:03:09.800 --> 0:03:17.720 chemistry all working together to make us who we truly are. Right, 0:03:17.800 --> 0:03:22.560 So every other discipline, if you think of computing, for example, 0:03:22.600 --> 0:03:27.160 it's purely electricity. Or if we do drug development or 0:03:27.240 --> 0:03:32.480 drug discovery, that's mostly chemistry. Experiments in the lab, that's 0:03:32.560 --> 0:03:36.880 mostly biology, But the human brain is all of those 0:03:37.440 --> 0:03:38.320 put together. 0:03:39.960 --> 0:03:41.200 The cells in our brain. 0:03:41.320 --> 0:03:46.000 Secrety is chemical substances that diffuse everywhere and hook up 0:03:46.040 --> 0:03:49.920 where they need to to trigger certain circuits and then 0:03:50.000 --> 0:03:52.280 trigger some effects afterwards. 0:03:52.480 --> 0:03:54.280 So it was just. 0:03:54.200 --> 0:04:00.240 An elegant science that has big impacts. 0:04:00.600 --> 0:04:03.440 We're shortly going to get there and talk about you've 0:04:03.480 --> 0:04:07.320 taken on this kind of very technology focused roller Cleveland Clinic. 0:04:07.640 --> 0:04:11.600 I'm curious about if we go back to when you 0:04:11.640 --> 0:04:14.600 were just starting out at Cleveland Clinic. Yeah, how much 0:04:14.640 --> 0:04:18.279 were you thinking about this sort of technology piece about 0:04:18.279 --> 0:04:21.839 what technology could do for your specialty, about was that 0:04:21.920 --> 0:04:23.919 on your mind or is this something you've come to 0:04:24.000 --> 0:04:24.640 more recently. 0:04:26.320 --> 0:04:27.600 Well, it's been a journey. 0:04:27.839 --> 0:04:32.760 Right when I started training as a clinician, my priority 0:04:32.920 --> 0:04:35.400 was just to learn how to better care for my 0:04:35.480 --> 0:04:38.800 patients and how to be a better physician. 0:04:39.120 --> 0:04:40.840 Right, and then I. 0:04:40.880 --> 0:04:47.160 Realized that clinical practice provides an immediate reward. I'm interacting 0:04:47.200 --> 0:04:49.839 with a human being and helping them in the moment, 0:04:50.160 --> 0:04:53.479 so there is that immediate reward that comes with that. 0:04:53.600 --> 0:04:57.240 But that wasn't enough. I wanted something more. So then 0:04:57.720 --> 0:05:03.000 I learned biomedical research. Practic this is and research offered 0:05:03.080 --> 0:05:06.159 me this path towards a future. You know, the reward 0:05:06.240 --> 0:05:11.200 there is more long term. I'm studying discovering things that 0:05:11.320 --> 0:05:15.520 could help many people in the future, even though I 0:05:15.560 --> 0:05:18.040 will never get to see them or meet them, or 0:05:18.080 --> 0:05:21.840 you know, have that immediate satisfaction. So that shifted me 0:05:21.960 --> 0:05:25.279 from being a pure clinician to being a clinician scientist. 0:05:26.000 --> 0:05:30.080 And as that journey progressed, it became very clear, as 0:05:30.120 --> 0:05:33.160 medicine evolved over the past twenty years, that we cannot 0:05:33.200 --> 0:05:39.520 do any good biomedical research without understanding data and technology. 0:05:39.680 --> 0:05:42.880 You know, the balance is shifting from most of the 0:05:42.960 --> 0:05:45.919 research is happening we call it, you know, on the 0:05:45.920 --> 0:05:51.400 wet bench with actual experiments, physical experiments, to a place 0:05:51.520 --> 0:05:56.360 where most of the work is happening through compute and 0:05:56.480 --> 0:06:01.359 simulations and data. So I became more involved for my 0:06:01.520 --> 0:06:06.920 personal research in building big data models and learning about 0:06:07.000 --> 0:06:11.440 AI and you know, learning about technology in general. And 0:06:11.520 --> 0:06:16.279 then as that journey progressed, I was fortunate enough to 0:06:16.320 --> 0:06:19.320 be an enroll for Cleveland Clinic, where my job is 0:06:19.360 --> 0:06:23.800 to bring that technology and bridge it to research for 0:06:23.960 --> 0:06:27.080 all researchers across our healthcare system. 0:06:28.040 --> 0:06:30.679 When you were talking about how in your own research 0:06:30.720 --> 0:06:33.120 you were moving in that direction, what was your own 0:06:33.160 --> 0:06:35.000 research focused on. What were you looking at? 0:06:36.680 --> 0:06:43.440 I was looking at brain surgery for epilepsy. It's an 0:06:43.480 --> 0:06:48.000 intervention that's been around for decades actually, but when I 0:06:48.120 --> 0:06:51.920 started practice, I was shocked by, you know, the practice 0:06:51.960 --> 0:06:56.280 that we had where making decisions around surgery like, you know, 0:06:56.360 --> 0:07:00.719 what patients should get it versus not, how likely is 0:07:00.760 --> 0:07:03.359 it to work, what part of the brain should we remove. 0:07:03.880 --> 0:07:07.640 All of those decisions were at the time and the 0:07:07.760 --> 0:07:12.560 early two thousands driven by clinical opinions right. You know, 0:07:12.640 --> 0:07:14.920 you have an experienced surgeon, they decide to do this. 0:07:15.080 --> 0:07:18.600 Somebody else might decide to do something completely different, and 0:07:19.440 --> 0:07:21.520 I didn't feel that that was the right. 0:07:21.400 --> 0:07:22.640 Way to practice medicine. 0:07:22.680 --> 0:07:25.239 You know, that we needed to be more evidence based 0:07:25.280 --> 0:07:28.920 and data driven. So I went in the business and 0:07:29.040 --> 0:07:35.520 research of building models, predictive models that can ingest data 0:07:35.760 --> 0:07:38.600 from all of the tests that we would do about 0:07:38.840 --> 0:07:43.640 on these patients before to figure out surgery. So I 0:07:44.240 --> 0:07:49.000 learned how to analyze all types of data, from genetic 0:07:49.080 --> 0:07:54.520 data and individuals to pictures to electrical recordings, and then 0:07:54.840 --> 0:07:58.680 combine those into these prediction. 0:07:58.400 --> 0:08:03.760 Models you're looking at You're taking large numbers of surgeries 0:08:03.760 --> 0:08:06.520 for epilepsy, and you're seeing what kind of connection there 0:08:06.640 --> 0:08:10.360 is between the success of the surgical intervention and the 0:08:10.440 --> 0:08:12.360 underlying presentation. 0:08:11.880 --> 0:08:13.440 Of the patient exactly. 0:08:13.560 --> 0:08:17.240 So I would be telling the patient, what is your 0:08:17.440 --> 0:08:22.160 specific chance of becoming seizure free with surgery instead of 0:08:22.240 --> 0:08:25.960 giving them statistics about you know, like you know, in general, 0:08:26.200 --> 0:08:29.000 how well is that going to be effective? So that 0:08:29.160 --> 0:08:32.160 piece of individualizing medicine. 0:08:32.280 --> 0:08:35.560 So Cleveland Clinic decides to create a post called Chief 0:08:35.559 --> 0:08:36.600 Information Officers. 0:08:36.640 --> 0:08:40.520 If research information we have research. Yes, always chief information 0:08:40.640 --> 0:08:41.760 officer runs it. 0:08:42.559 --> 0:08:44.440 Oh, yes, chief research information. 0:08:44.640 --> 0:08:46.200 Yeah, so it for research. 0:08:46.400 --> 0:08:51.840 This is parenthetically a huge job. Yes, So you apply 0:08:51.920 --> 0:08:54.320 for this, do you know that you're going to be 0:08:54.400 --> 0:08:57.319 thinking and talking and dealing with quantum. 0:08:57.640 --> 0:09:03.520 I started in January twenty twenty, and like every leader 0:09:03.559 --> 0:09:06.439 who's put in a position, remember, everybody tells you should 0:09:06.480 --> 0:09:10.680 read that the first ninety days great how to plan. 0:09:11.280 --> 0:09:14.800 So I was reading that and doing my listening tours 0:09:15.120 --> 0:09:19.959 to understand, and then COVID hits and I got a 0:09:20.000 --> 0:09:25.360 call from our executive suite about all, there is this 0:09:25.400 --> 0:09:28.760 thing called COVID that's coming. We will start testing people 0:09:28.840 --> 0:09:32.920 in four days. People will want to do research with COVID. 0:09:33.360 --> 0:09:34.120 Make it happen. 0:09:34.600 --> 0:09:38.600 So it was there's no chapter in the book about that, 0:09:38.760 --> 0:09:45.080 you know. So so I then it really it was 0:09:45.160 --> 0:09:47.520 like a pressure cooker, you know. 0:09:47.679 --> 0:09:48.280 Test with. 0:09:49.040 --> 0:09:53.359 Yeah, I have to create this access to data, structured 0:09:54.320 --> 0:09:57.800 you know, resources, so that we can learn from it 0:09:57.840 --> 0:10:02.320 as quickly as possible. And so that was a catalyst. 0:10:02.679 --> 0:10:05.880 So then comes twenty twenty one. That was the year 0:10:06.000 --> 0:10:10.120 of our centennial one hundred years for Cleveland Clinic. So everybody, 0:10:10.200 --> 0:10:12.640 not just me, we were in a mindset where we 0:10:12.640 --> 0:10:16.840 were thinking long, long term, you know, like what made 0:10:16.880 --> 0:10:21.080 us specialized? Now, how do we stay relevant? Where is 0:10:21.120 --> 0:10:24.719 the world going to be ten years from now? And 0:10:24.760 --> 0:10:28.920 what should I get going right this moment to shape 0:10:28.960 --> 0:10:31.880 that and be ready for it. And that's when quantum 0:10:31.920 --> 0:10:36.720 came in my mind, where unless we invest in it 0:10:37.160 --> 0:10:41.120 now twenty twenty one, we will not be ready for 0:10:41.200 --> 0:10:44.600 this next computer revolution that's coming after AI. 0:10:45.200 --> 0:10:47.720 Everybody in the world right now is doing nothing but 0:10:47.800 --> 0:10:52.360 talking about AI, and you are already thinking one step 0:10:52.400 --> 0:10:56.840 beyond the quantum. What's different about the opportunity that quantum 0:10:57.440 --> 0:11:00.880 creates for a medical research than AI? 0:11:02.480 --> 0:11:03.160 Biology? 0:11:03.640 --> 0:11:11.760 The human body, by definition, is much closer to fundamentals 0:11:11.760 --> 0:11:15.800 of quantum, you know, to quantum mechanics and quantum physics 0:11:15.840 --> 0:11:16.400 than it. 0:11:16.320 --> 0:11:16.959 Is to AI. 0:11:17.840 --> 0:11:24.480 AI is a classical computing approach that in essence, reduces 0:11:24.679 --> 0:11:29.360 every piece of data to a black or white binary 0:11:29.559 --> 0:11:34.360 categorization of a one or a zero. At its core 0:11:35.080 --> 0:11:40.120 nature around us, the human body, there is nothing categorical 0:11:40.400 --> 0:11:46.719 about it. It's that whole, you know, continuum of colors 0:11:46.760 --> 0:11:54.360 of life. Quantum its principles are that, you know, so 0:11:54.520 --> 0:11:59.880 there's all the scientific principles about quantum physics and superb 0:12:00.040 --> 0:12:02.960 position and in tankerment, you know, all of these complex 0:12:03.000 --> 0:12:05.520 things that people have a hard time with, but for 0:12:06.360 --> 0:12:11.880 in essence, it really is much more aligned. Like if 0:12:11.920 --> 0:12:15.400 I want to draw a colored picture, I will not 0:12:15.559 --> 0:12:17.080 go and pick up charcoal. 0:12:17.160 --> 0:12:19.240 You know, it's much easier for. 0:12:19.160 --> 0:12:23.160 Me to draw it if I had a colored palette 0:12:23.160 --> 0:12:27.040 with me, and quantum offers that. There is plenty of 0:12:27.200 --> 0:12:32.160 situations in medicine that are just intractable, you know, meaning 0:12:33.120 --> 0:12:36.600 it's not an issue just of it being AI being slow, 0:12:36.920 --> 0:12:40.480 or it doesn't have enough data, or if only we 0:12:40.600 --> 0:12:43.880 got more GPUs, you know, we can answer those questions. 0:12:43.960 --> 0:12:49.800 There are some problems in medicine that are fundamentally such 0:12:49.880 --> 0:12:52.360 that even if you give me all the GPUs in 0:12:52.400 --> 0:12:56.760 the world, there is no way that AI can model 0:12:56.880 --> 0:13:03.000 accurately how these mollyuels in the body are interacting among 0:13:03.040 --> 0:13:08.560 each other, or how compounds electrons are moving within the mitochondria. 0:13:08.679 --> 0:13:13.599 These are the engines within our cells. There's these fundamental 0:13:13.679 --> 0:13:19.200 things in biology that AI and classical computers are just 0:13:19.360 --> 0:13:21.800 not built to be able to simulate. 0:13:22.480 --> 0:13:26.040 So you must go to genner parties. Doctor j. High 0:13:26.400 --> 0:13:29.760 people must ask you what is quantum computing? What do 0:13:29.800 --> 0:13:30.520 you tell them? 0:13:31.040 --> 0:13:34.280 Yes, Although I often, you know, we talk about other 0:13:34.320 --> 0:13:35.360 things at dinner. 0:13:35.080 --> 0:13:40.080 Parties, eventually it comes down to you. If I was 0:13:40.120 --> 0:13:42.800 at that dinner party with you, I would ask you 0:13:42.840 --> 0:13:43.800 what is quantic computer? 0:13:44.200 --> 0:13:46.640 I would say, depends on how much time we have 0:13:46.760 --> 0:13:48.480 at the party to explain it. 0:13:48.920 --> 0:13:50.720 The short answer would. 0:13:50.559 --> 0:13:56.160 Be, it's a completely different way of working with computers 0:13:56.200 --> 0:13:59.760 than what we're used to. Right now, you can imagine 0:13:59.840 --> 0:14:02.920 that AI as being like a car that you take 0:14:03.040 --> 0:14:06.520 to go from one place to another. No matter how 0:14:06.600 --> 0:14:10.600 fast that Ferrari can get and how much fewer you 0:14:10.679 --> 0:14:13.880 put in it, it's never going to be a fighter jet. 0:14:14.200 --> 0:14:15.760 It's never going to be a plane. 0:14:16.640 --> 0:14:20.000 AI is the car, the plane is quant They are 0:14:20.440 --> 0:14:23.400 ways to get from point A to point B, but 0:14:23.640 --> 0:14:29.720 they work very differently, and we always use them in together. 0:14:30.160 --> 0:14:34.960 If I'm flying from Shaker Heights to Yorktown Heights, I 0:14:35.240 --> 0:14:38.000 drive to the airport, get on the plane, then take 0:14:38.040 --> 0:14:41.880 an uber to get to Yorktown Heights. In research, we 0:14:41.920 --> 0:14:44.440 will do the same. We do some piece of it 0:14:44.480 --> 0:14:46.960 in AI, some piece of it in quantum, and then 0:14:47.040 --> 0:14:48.680 go back and forth, back. 0:14:48.560 --> 0:14:52.000 In twenty twenty one, Cleveland Clinic and IBM announced that 0:14:52.040 --> 0:14:56.040 they were starting something called the Discovery Accelerator. What is that. 0:14:56.920 --> 0:15:01.400 It's an initiative, a program, a partnership really that is 0:15:02.080 --> 0:15:10.000 designed to bridge advanced computational tools and technology through the 0:15:10.120 --> 0:15:16.760 leader in that space, IBM, with biomedical science and research 0:15:17.040 --> 0:15:22.800 and life sciences problems, and that is a Cleveland Clinic. 0:15:23.280 --> 0:15:26.840 We called it a Discovery Accelerator because that was our goal. 0:15:27.200 --> 0:15:31.320 We were both on both sides challenged to the fact 0:15:31.320 --> 0:15:34.160 that discovery in medicine was just taking too long. The 0:15:34.320 --> 0:15:39.240 classical example that really brought it to life to us 0:15:39.280 --> 0:15:43.360 then was drug discovery that it took over a decade 0:15:43.560 --> 0:15:47.720 and it still does actually over a decade. From the 0:15:47.800 --> 0:15:51.480 moment that there is a compound that someone in a 0:15:51.800 --> 0:15:56.720 lab biomedical lab thinks it would be effective to treat 0:15:56.760 --> 0:16:00.480 a certain disease, it takes about ten to thirteen from 0:16:00.520 --> 0:16:03.400 that moment to when that drug is on a shelf 0:16:03.520 --> 0:16:06.160 for a patient to get to from a pharmacy. And 0:16:06.200 --> 0:16:09.840 that was just too much of a gap to allow 0:16:10.480 --> 0:16:13.920 when we have so many health conditions that we needed 0:16:13.960 --> 0:16:16.840 to address, and a big part of that gap could 0:16:16.880 --> 0:16:22.680 be computationally solved better simulation of compounds, designing drug trails 0:16:22.760 --> 0:16:25.720 that are more efficient, you know, that would finish faster. 0:16:26.320 --> 0:16:33.520 So that was the motivation to bring computational tools and 0:16:33.600 --> 0:16:36.560 technology closer to biomedical researchers. 0:16:36.760 --> 0:16:40.400 Cleveland Click and IBM team up. And I'm assuming the 0:16:40.440 --> 0:16:44.280 IBM quantum guys and other people descend on Cleveland and 0:16:44.320 --> 0:16:47.480 you have your first meeting with them. Are they telling 0:16:47.480 --> 0:16:50.680 you things you would never thought of? Or I mean, 0:16:50.840 --> 0:16:53.160 I'm just curious about what's the difference between what you 0:16:54.920 --> 0:16:58.920 thought was the potential was and what you discovered the 0:16:58.920 --> 0:16:59.680 potential was. 0:17:01.040 --> 0:17:05.040 That is an excellent question. If I had to prioritize. 0:17:05.119 --> 0:17:08.920 One lesson that I learned over the past few years 0:17:09.000 --> 0:17:13.119 of doing this, it is that you can never know 0:17:14.359 --> 0:17:17.480 what your you know where your brain is going to 0:17:17.560 --> 0:17:23.879 go and discovery until you talk. You know, you have 0:17:23.920 --> 0:17:27.040 to open it up and really listen to try to 0:17:27.119 --> 0:17:28.679 learn what the other people are saying. 0:17:29.080 --> 0:17:31.280 It's it went both ways. 0:17:31.440 --> 0:17:32.480 Can you give you an example. 0:17:32.640 --> 0:17:35.120 So, okay, we built a quantum right, so it took 0:17:35.480 --> 0:17:38.240 it took like eight months to get this machine put together, 0:17:38.720 --> 0:17:40.440 and we put it in our cafeteria. 0:17:41.080 --> 0:17:43.680 That's the whole other story where you. 0:17:43.600 --> 0:17:48.240 Put the quantum machine computer in your cafeteria. Yes, yes, yes, 0:17:50.080 --> 0:17:51.679 can you see it when you're eating lunch? 0:17:52.119 --> 0:17:52.399 Yeah? 0:17:52.480 --> 0:17:55.280 Yeah, we have people eating lunch around it all the time. 0:17:55.760 --> 0:17:58.440 We wanted to have a machine that people can see. 0:17:58.680 --> 0:18:02.720 Otherwise it it the program wouldn't launch properly, so I 0:18:02.800 --> 0:18:03.359 needed to. 0:18:03.280 --> 0:18:04.919 Have it physically. 0:18:05.480 --> 0:18:09.000 So we had to look at retrofitted in existing space, 0:18:09.040 --> 0:18:12.560 and the cafeteria space worked out. It was far enough 0:18:12.600 --> 0:18:15.879 from the street, you know, there was no vibration that 0:18:16.560 --> 0:18:19.320 it was a double floor, you know, so the ceiling 0:18:19.520 --> 0:18:22.719 was high enough. It just like technically fit all of 0:18:22.760 --> 0:18:28.720 those requirements. And it was either there or we put 0:18:28.760 --> 0:18:31.800 it in our data center, which is a building and 0:18:31.960 --> 0:18:36.920 like another city close to Cleveland, and we picked the cafeteria. 0:18:37.200 --> 0:18:39.600 Yeah. Now, why I know this is sort of this 0:18:39.680 --> 0:18:41.600 is kind of hilarious, But there's a serious point I 0:18:41.880 --> 0:18:45.480 I want to touch on, which is why do you 0:18:45.520 --> 0:18:46.600 need it on premises? 0:18:46.800 --> 0:18:49.080 On the premises Cleveland Clinic research? 0:18:49.160 --> 0:18:53.600 We needed to change how we think about research and 0:18:53.800 --> 0:19:00.760 shift the mindset of all of our researchers, and all 0:19:00.800 --> 0:19:05.440 of our researchers I'm talking about three thousand individuals who 0:19:05.440 --> 0:19:09.359 are one hundred percent doing biomedical research in Cleveland Clinic, 0:19:10.040 --> 0:19:15.679 two hundred and thirty labs individual pis. So that's the 0:19:15.720 --> 0:19:19.440 scale that I'm talking about that we had to create 0:19:19.560 --> 0:19:20.680 an impact on. 0:19:21.440 --> 0:19:24.560 So having it, having it. 0:19:24.920 --> 0:19:31.040 Be there was as much for inspiration and to trigger 0:19:31.040 --> 0:19:34.840 our motivation to change as it was a you know, 0:19:34.880 --> 0:19:39.040 a practical solution, say, because we had the space and 0:19:39.080 --> 0:19:41.440 the connections and all of that. 0:19:41.680 --> 0:19:44.720 I saw in it. IBM headquarters. They're beautiful. 0:19:44.720 --> 0:19:48.080 Their works of art they are They're gorgeous, and the 0:19:48.080 --> 0:19:53.600 one that we have is the most gorgeous. 0:19:53.200 --> 0:19:56.120 One of all. This is not me, you know, the mom. 0:19:57.400 --> 0:19:59.440 Bias talking about it. 0:20:00.560 --> 0:20:03.479 You know, they got an award, the Red Dot Award 0:20:03.640 --> 0:20:08.520 for Design went to IBM and Cleveland Clinic for our 0:20:08.640 --> 0:20:16.919 quantum and the quantum that IBM built for RPI a 0:20:16.960 --> 0:20:21.120 couple of years after hours. They modeled it after hours, 0:20:21.160 --> 0:20:23.800 not after the you know the ones from before. 0:20:24.040 --> 0:20:27.560 Just just so people know, we're talking about basically a 0:20:27.600 --> 0:20:31.080 small garage. 0:20:30.680 --> 0:20:34.640 But that size eleven foot by eleven eleven feet eleven feet, you. 0:20:34.560 --> 0:20:36.720 Know, the cube small, So. 0:20:36.720 --> 0:20:40.879 It's a it's a glass cube, and the glass comes 0:20:40.960 --> 0:20:45.520 all the way from Italy. It's the same glass that 0:20:45.840 --> 0:20:49.920 protects the crown jewels and you know the Mona Lisa 0:20:50.000 --> 0:20:54.320 and all that. So so there's a glass all around, 0:20:54.760 --> 0:20:58.040 and then there is the tube that you see, the 0:20:58.359 --> 0:21:00.080 stainless tube. 0:20:59.800 --> 0:21:01.560 That's shiny, you know, and clean. 0:21:01.640 --> 0:21:04.520 But the technologies are inside of it, you know, the 0:21:04.600 --> 0:21:08.560 chandelier and then the processor and the bottom and it's 0:21:08.640 --> 0:21:14.240 just a fascinating thing to watch and it hums. It 0:21:14.359 --> 0:21:17.439 makes the sound so even it sounds alive. 0:21:19.760 --> 0:21:22.800 You have a great deal of affection for your computer. 0:21:22.560 --> 0:21:24.359 And we love our machine. 0:21:24.960 --> 0:21:27.200 I want to go back to something you said before, 0:21:27.280 --> 0:21:30.840 which is when you had your initial conversations with IBM, 0:21:31.359 --> 0:21:35.280 both sides learned things that they hadn't previously thought of. 0:21:35.520 --> 0:21:39.520 Give me an example of something that either side hadn't 0:21:39.600 --> 0:21:41.960 realized could be done with this new technology. 0:21:42.040 --> 0:21:46.240 Sure, I mean on the equivalent clinic side, we thought 0:21:46.600 --> 0:21:50.720 that what quantum should be good for is that it 0:21:50.760 --> 0:21:53.840 would be a faster computer, right, so that we have 0:21:54.000 --> 0:21:58.920 these big data sets that are requiring much more compute 0:21:58.960 --> 0:22:01.000 power and you know, GIP used than what we have 0:22:01.080 --> 0:22:04.520 and we should just run them on the quantum. And 0:22:04.920 --> 0:22:09.240 what we came to realize is that Quantum actually does 0:22:09.280 --> 0:22:12.280 not do well with these large data sets. What it 0:22:12.359 --> 0:22:16.879 does well with our smaller, better defined data sets, but 0:22:17.040 --> 0:22:21.879 ones where simulation is more important. You know, we have 0:22:22.000 --> 0:22:27.400 to run them through multiple models, multiple simulations of how 0:22:27.440 --> 0:22:30.880 they interact. As an example, one of the very first 0:22:30.960 --> 0:22:35.800 projects we throw at Quantum was wanting to predict complications 0:22:35.800 --> 0:22:40.960 cardiac complications from general surgery using electronic health record data. 0:22:41.040 --> 0:22:45.720 So data from our electronic health records are by definition 0:22:45.840 --> 0:22:47.679 these really big data sets. 0:22:47.800 --> 0:22:50.000 You have in them every single thing. 0:22:49.840 --> 0:22:54.280 That you know about the patient that you've collected, and 0:22:54.400 --> 0:22:58.440 we thought that Quantum would help us build models, better 0:22:58.600 --> 0:23:00.600 models with this data. 0:23:00.720 --> 0:23:04.320 And it failed miserably. It wasn't good, you know, at 0:23:04.359 --> 0:23:05.480 doing that thing. 0:23:05.560 --> 0:23:07.879 And why didn't it like that problem because it's too 0:23:08.240 --> 0:23:09.399 it's too bigger than wieldy. 0:23:10.000 --> 0:23:13.480 Well, because the hardware isn't tready right, So the hardware 0:23:13.640 --> 0:23:17.760 with Quantum, that issue was back then. You know, now 0:23:17.800 --> 0:23:22.160 we've upgraded our processor. But still Quantum now is limited 0:23:22.200 --> 0:23:28.120 by noise and by its ability to correct for errors 0:23:28.240 --> 0:23:33.960 when it's doing computation. And the more data that you 0:23:34.119 --> 0:23:36.600 throw at it that you require it, you know, to 0:23:36.800 --> 0:23:41.200 put in its system, you know to model interactions, the 0:23:41.240 --> 0:23:44.080 more errors it's likely to make, so then the harder 0:23:45.080 --> 0:23:47.440 it is for it to get to an answer that 0:23:48.000 --> 0:23:51.359 you can trust. Now, we made a lot of progress sense, 0:23:52.359 --> 0:23:55.960 which I'm sure we'll get to I hope we'll get 0:23:55.960 --> 0:23:59.119 to with some recent breakthroughs that we made in that 0:23:59.200 --> 0:24:05.320 space with modeling large compounds and interactions. But the way 0:24:05.480 --> 0:24:07.560 that got us to where we are now, where we 0:24:07.600 --> 0:24:11.400 could model large interactions, it took us figuring out that 0:24:11.480 --> 0:24:15.200 we shouldn't be doing everything on quantum. 0:24:15.480 --> 0:24:18.880 So give me an example of a problem that quantum 0:24:19.080 --> 0:24:22.840 is ideally suited for that the quantum really. 0:24:22.720 --> 0:24:27.320 Likes in drug discovery, for examine chemistry. It likes chemistry 0:24:27.400 --> 0:24:31.480 a lot because in chemistry what it has to what 0:24:31.520 --> 0:24:35.600 we wanted to model is how a drug that we 0:24:35.840 --> 0:24:39.080 put in our body is going to interact with We 0:24:39.119 --> 0:24:43.320 say the protein you know, the target, the ligand that 0:24:43.400 --> 0:24:46.000 it needs to buind too. Like you know, the drug 0:24:46.080 --> 0:24:47.840 is a key and it needs to fit in a 0:24:47.920 --> 0:24:50.880 lock in certain parts of the body to open it, 0:24:51.320 --> 0:24:55.800 get in, do it stay. And there are many keys, 0:24:55.960 --> 0:25:01.280 many potential compounds that we could test for many parts 0:25:01.359 --> 0:25:03.640 of our body. We don't really know how they're going 0:25:03.680 --> 0:25:05.840 to interact. Traditionally, what we do is we have to 0:25:05.960 --> 0:25:08.280 build all the keys. We have to manufacture all these 0:25:08.280 --> 0:25:12.200 compounds and then do actual trial clinical trials. Put them 0:25:12.200 --> 0:25:14.560 in people, put them in animals, and see what happens. 0:25:14.800 --> 0:25:15.840 See which one is best? 0:25:15.920 --> 0:25:19.639 See yeah, which like two fit best together. So we 0:25:19.760 --> 0:25:22.520 use AI to try to help us with that. But 0:25:22.800 --> 0:25:26.240 AI can only model what it had learned, right, So 0:25:26.400 --> 0:25:33.120 for rare diseases conditions that there isn't enough information out 0:25:33.160 --> 0:25:36.480 there on what the you know, the locks look like, 0:25:37.000 --> 0:25:40.040 it's really hard to then model it with AI and 0:25:40.119 --> 0:25:43.199