WEBVTT - Smart Talks with IBM: The Quantum Shift in Biomedical Discovery 0:00:00.160 --> 0:00:02.960 Hey everyone, it's Robert and Joe here. Today we've got 0:00:02.960 --> 0:00:05.240 something a little bit different to share with you. It's 0:00:05.280 --> 0:00:09.400 a new season of the Smart Talks with IBM podcast series. 0:00:09.720 --> 0:00:13.520 This season on Smart Talks with IBM, Malcolm Gladwell is back, 0:00:13.520 --> 0:00:15.760 and this time he's taking the show on the road. 0:00:15.880 --> 0:00:19.160 Malcolm is stepping outside the studio to explore how IBM 0:00:19.239 --> 0:00:22.880 clients are using artificial intelligence to solve real world challenges 0:00:23.079 --> 0:00:25.520 and transform the way they do business. 0:00:25.480 --> 0:00:29.560 From accelerating scientific breakthroughs to reimagining education. It's a fresh 0:00:29.600 --> 0:00:33.040 look at innovation in action, where big ideas meet cutting 0:00:33.120 --> 0:00:33.960 edge solutions. 0:00:34.320 --> 0:00:37.160 You'll hear from industry leaders, creative thinkers, and of course 0:00:37.280 --> 0:00:40.920 Malcolm Gladwell himself as he guides you through each story. 0:00:41.440 --> 0:00:44.680 New episodes of Smart Talks with IBM drop every month 0:00:44.680 --> 0:00:48.279 on the iHeartRadio app, Apple Podcasts, or wherever you get 0:00:48.320 --> 0:00:52.240 your podcasts. Learn more at IBM dot com slash smart Talks. 0:00:52.880 --> 0:00:55.400 This is a paid advertisement from IBM. 0:00:57.200 --> 0:01:02.600 H'm Malcolm Gladwell and you're listening to Smart Talk with IBM. 0:01:02.960 --> 0:01:06.319 When doctor Laura Jahi began treating epilepsy patients in the 0:01:06.360 --> 0:01:11.520 early two thousands, she noticed something unsettling. Different surgeons could 0:01:11.520 --> 0:01:16.319 look at the exact same case and recommend completely different treatments. 0:01:16.680 --> 0:01:19.480 One surgeon might remove one part of the brain, another 0:01:19.680 --> 0:01:23.039 a different part, and a third might not operate at all. 0:01:23.800 --> 0:01:26.320 Doctor j Hi believed there had to be a better way, 0:01:26.760 --> 0:01:30.840 one grounded in data. That conviction set her on a 0:01:30.840 --> 0:01:34.040 path that would lead her to become Chief Research Information 0:01:34.120 --> 0:01:38.000 Officer at Cleveland Clinic and the executive program lead for 0:01:38.080 --> 0:01:42.360 the Discovery Accelerator. The Discovery Accelerator is a ten year 0:01:42.440 --> 0:01:46.960 partnership between Cleveland Clinic and IBM where researchers are using 0:01:47.000 --> 0:01:51.640 AI and quantum computing to make incredible discoveries in healthcare 0:01:52.000 --> 0:01:55.080 and life sciences. I sat down with doctor j Hi 0:01:55.400 --> 0:01:59.440 to explore what's happening now, what's possible with quantum computing, 0:02:00.040 --> 0:02:03.840 and where this next era of biomedical discovery is headed. 0:02:05.520 --> 0:02:09.120 Epilepsy was your kind of specialty within neurology, correct? Yes, 0:02:09.200 --> 0:02:10.080 what led you to that? 0:02:10.840 --> 0:02:14.240 I was always fascinated by the brain. You know, it's 0:02:14.800 --> 0:02:17.760 the part of our body that lives still to this 0:02:17.840 --> 0:02:21.680 day the most to be discovered. So I was always 0:02:21.840 --> 0:02:28.920 intrigued by areas that leave more for discovery. And epilepsy 0:02:29.080 --> 0:02:33.080 was my pragmatic side, wanting to choose a subspecialty where 0:02:33.639 --> 0:02:37.880 the problem can be fixed. In epilepsy, there are many 0:02:37.919 --> 0:02:41.399 medications that are very effective, and there's a brain surgery 0:02:41.520 --> 0:02:44.640 that we can do to stop seizures when medicines don't work. 0:02:45.760 --> 0:02:46.960 That attracted me. 0:02:47.120 --> 0:02:51.000 You know, compared to other areas in neurology, like stroke, 0:02:51.080 --> 0:02:55.560 for example, or dementia, where usually the damage is more definite. 0:02:56.200 --> 0:02:59.359 I wanted to be able to tell my patients that 0:03:00.080 --> 0:03:02.280 you have a big problem, but here's what I can. 0:03:02.160 --> 0:03:05.280 Do to fix it, and epilepsy offered me that. 0:03:06.200 --> 0:03:09.959 But what would There must have been interesting and intriguing 0:03:10.400 --> 0:03:14.200 unsolved problems in neurology. 0:03:14.919 --> 0:03:17.480 Oh my gosh, it's the whole brain, isn't it. Before 0:03:17.639 --> 0:03:21.799 starting to deal with artificial intelligence and research. Right in 0:03:21.840 --> 0:03:25.959 the neurology, I'm dealing with real intelligence, the human brain 0:03:26.120 --> 0:03:29.400 and how it works and how we think and how 0:03:29.440 --> 0:03:34.000 we make decisions, and how it. 0:03:34.000 --> 0:03:35.800 Can grow and evolve. 0:03:36.040 --> 0:03:42.440 And so there is many untaped questions in neurology. 0:03:41.880 --> 0:03:45.680 And that's part of the fascination in it. 0:03:45.960 --> 0:03:51.400 In epilepsy in particular, it's an electrical disease in the brain. 0:03:51.680 --> 0:03:56.760 It's actually one of the conditions in a neurology that's 0:03:56.800 --> 0:04:02.960 a perfect alignment of all of the scientific disciplines. It's 0:04:03.120 --> 0:04:09.560 biology and physics and chemistry all working together to make 0:04:09.760 --> 0:04:15.920 us who we truly are. Right, So every other discipline, 0:04:15.920 --> 0:04:19.240 if you think of computing, for example, it's purely electricity. 0:04:19.520 --> 0:04:23.560 Or if we do drug development or drug discovery, that's 0:04:23.560 --> 0:04:29.240 mostly chemistry experiments in the lab, that's mostly biology. But 0:04:29.680 --> 0:04:35.279 the human brain is all of those put together. The 0:04:35.400 --> 0:04:39.360 cells in our brain secret these chemical substances that diffuse 0:04:39.520 --> 0:04:43.120 everywhere and hook up where they need to to trigger 0:04:43.200 --> 0:04:48.719 certain circuits and then trigger some effects afterwards. So it 0:04:48.960 --> 0:04:55.200 was just an elegant science that has big impacts. 0:04:55.839 --> 0:04:58.680 We're shortly going to get there and talk about you've 0:04:58.720 --> 0:05:01.880 taken on this kind of very techn aology focused roller 0:05:01.920 --> 0:05:06.440 Cleveland Clinic. I'm curious about if we go back to 0:05:06.520 --> 0:05:09.400 when you were just starting out at Cleveland Clinic. Yeah, 0:05:09.440 --> 0:05:12.560 how much were you thinking about this sort of technology piece, 0:05:13.240 --> 0:05:17.000 about what technology could do for your specialty? About was 0:05:17.000 --> 0:05:19.000 that on your mind or is this something you've come 0:05:19.080 --> 0:05:19.880 to more recently. 0:05:21.560 --> 0:05:24.680 Well, it's been a journey, right, and when I started 0:05:24.720 --> 0:05:28.760 training as a clinician, my priority was just to learn 0:05:28.839 --> 0:05:32.320 how to better care for my patients and how to 0:05:32.360 --> 0:05:34.080 be a better physician. 0:05:34.360 --> 0:05:36.080 Right, And then I. 0:05:36.120 --> 0:05:42.400 Realized that clinical practice provides an immediate reward. I'm interacting 0:05:42.440 --> 0:05:45.080 with a human being and helping them in the moment, 0:05:45.400 --> 0:05:48.719 so there is that immediate reward that comes with that. 0:05:48.839 --> 0:05:51.880 But that wasn't enough. I wanted something more. 0:05:51.960 --> 0:05:58.279 So then I learned biomedical research practices, and research offered 0:05:58.320 --> 0:06:01.400 me this path towards a future. You know, the reward 0:06:01.480 --> 0:06:06.440 there is more long term. I'm studying discovering things that 0:06:06.560 --> 0:06:10.760 could help many people in the future, even though I 0:06:10.800 --> 0:06:13.280 will never get to see them or meet them, or 0:06:13.320 --> 0:06:17.080 you know, have that immediate satisfaction. So that shifted me 0:06:17.200 --> 0:06:20.520 from being a pure clinician to being a clinician scientist. 0:06:21.200 --> 0:06:25.280 And as that journey progressed, it became very clear, as 0:06:25.360 --> 0:06:28.400 medicine evolved over the past twenty years, that we cannot 0:06:28.440 --> 0:06:34.760 do any good biomedical research without understanding data and technology. 0:06:34.920 --> 0:06:38.120 You know, the balance is shifting from most of the 0:06:38.200 --> 0:06:41.159 research is happening we call it, you know, on the 0:06:41.160 --> 0:06:46.640 wet bench, with actual experiments, physical experiments, to a place 0:06:46.760 --> 0:06:51.600 where most of the work is happening through compute and 0:06:51.680 --> 0:06:56.359 simulations and data. So I became much more involved for 0:06:56.480 --> 0:07:01.799 my personal research in building big data models and learning 0:07:01.839 --> 0:07:05.560 about AI and you know, learning about technology in general. 0:07:06.520 --> 0:07:11.400 And then as that journey progressed, I was fortunate enough 0:07:11.440 --> 0:07:14.320 to be a role for Cleveland Clinic where my job 0:07:14.480 --> 0:07:18.160 is to bring that technology and bridge it to research 0:07:18.720 --> 0:07:22.320 for all researchers across our healthcare system. 0:07:23.280 --> 0:07:25.920 When you were talking about how in your own research 0:07:25.960 --> 0:07:28.360 you were moving in that direction, what was your own 0:07:28.400 --> 0:07:30.240 research focused on. What were you looking at? 0:07:31.920 --> 0:07:38.680 I was looking at brain surgery for epilepsy. It's an 0:07:38.720 --> 0:07:43.240 intervention that's been around for decades actually, but when I 0:07:43.320 --> 0:07:47.160 started practice, I was shocked by, you know, the practice 0:07:47.200 --> 0:07:51.520 that we had were making decisions around surgery, like, you know, 0:07:51.600 --> 0:07:55.920 what patients should get it versus not, how likely is 0:07:55.960 --> 0:07:58.600 it to work? What part of the brain should we remove. 0:07:59.120 --> 0:08:02.880 All of those visions were at the time and the 0:08:03.000 --> 0:08:07.800 early two thousands driven by clinical opinions. Right, you know, 0:08:07.840 --> 0:08:10.160 you have an experienced surgeon, they decide to do this. 0:08:10.320 --> 0:08:13.800 Somebody else might decide to do something completely different. And 0:08:14.680 --> 0:08:16.920 I didn't feel that that was the right way to 0:08:17.000 --> 0:08:17.880 practice medicine. 0:08:17.920 --> 0:08:18.480 You know that we. 0:08:18.520 --> 0:08:22.360 Needed to be more evidence based and data driven. So 0:08:22.720 --> 0:08:26.440 I went in the business in research of building models, 0:08:26.800 --> 0:08:31.960 predictive models that can ingest data from all of the 0:08:32.040 --> 0:08:35.719 tests that we would do about on these patients before to. 0:08:35.720 --> 0:08:36.840 Figure out surgery. 0:08:37.320 --> 0:08:43.360 So I learned how to analyze all types of data 0:08:43.440 --> 0:08:48.840 from genetic data and individuals to pictures to electrical recordings 0:08:49.280 --> 0:08:54.280 and then combine those into these prediction models. 0:08:54.840 --> 0:08:57.840 So basically you're looking at you're taking large numbers of 0:08:58.480 --> 0:09:01.559 surgeries for epilepsy, and you're seeing what kind of connection 0:09:01.640 --> 0:09:05.440 there is between the success of the surgical intervention and 0:09:05.520 --> 0:09:07.240 the underlying presentation of. 0:09:07.200 --> 0:09:08.679 The patient exactly. 0:09:08.800 --> 0:09:12.480 So I would be telling the patient what is your 0:09:12.679 --> 0:09:17.400 specific chance of becoming seizure free with surgery instead of 0:09:17.480 --> 0:09:21.200 giving them statistics about you know, like you know, in general, 0:09:21.440 --> 0:09:24.240 how well is that going to be effective? So that 0:09:24.400 --> 0:09:27.400 piece of individualizing medicine. 0:09:27.520 --> 0:09:30.800 So Cleveland Clinic decides to create a post called Chief 0:09:30.800 --> 0:09:31.840 Information Officers. 0:09:31.880 --> 0:09:35.760 If research information we have research, yes, always, chief information 0:09:35.880 --> 0:09:37.000 officer runs it. 0:09:37.800 --> 0:09:40.080 Oh yes, chief research information. 0:09:39.880 --> 0:09:41.440 Yeah, so it for research. 0:09:41.640 --> 0:09:47.040 This is parenthetically a huge job. Yes, So you apply 0:09:47.160 --> 0:09:49.560 for this, do you know that you're going to be 0:09:49.640 --> 0:09:52.559 thinking and talking and dealing with quantum. 0:09:52.880 --> 0:09:58.760 I started in January twenty twenty, and like every leader 0:09:58.760 --> 0:10:01.640 who's put in a position, remember everybody tells you should 0:10:01.679 --> 0:10:05.920 read that the first ninety days, Oh great, how to plan. 0:10:06.480 --> 0:10:10.040 So I was reading that and doing my listening tours 0:10:10.360 --> 0:10:15.199 to understand, and then COVID hits and I had a 0:10:15.240 --> 0:10:20.559 call from our executive suite about all, there is this 0:10:20.640 --> 0:10:24.000 thing called COVID that's coming. We will start testing people 0:10:24.080 --> 0:10:28.160 in four days. People will want to do research with COVID. 0:10:28.600 --> 0:10:29.360 Make it happen. 0:10:29.800 --> 0:10:33.839 So it was there's no chapter in the book about that, 0:10:33.960 --> 0:10:40.320 you know, So so I then it really it was 0:10:40.400 --> 0:10:42.760 like a pressure cooker, you know. 0:10:42.920 --> 0:10:46.079 Test with Yeah, I have to create. 0:10:45.840 --> 0:10:51.400 This access to data, structured you know, resources so that 0:10:51.440 --> 0:10:55.720 we can learn from it as quickly as possible. And 0:10:56.240 --> 0:10:59.840 so that was a catalyst. So then comes twenty twenty one. 0:11:00.200 --> 0:11:03.360 That was the year of our centennial one hundred years 0:11:03.360 --> 0:11:06.520 for Cleveland Clinic. So everybody, not just me, we were 0:11:06.559 --> 0:11:10.880 in a mindset where we were thinking long, long term, 0:11:10.960 --> 0:11:14.160 you know, like what made us specialized till now? How 0:11:14.200 --> 0:11:17.320 do we stay relevant? Where is the world going to 0:11:17.320 --> 0:11:21.160 be ten years from now? And what should I get 0:11:21.240 --> 0:11:25.360 going right this moment to shape that and be ready 0:11:25.440 --> 0:11:27.959 for it. And that's when quantum came in my mind, 0:11:28.800 --> 0:11:34.000 where unless we invest in it now twenty twenty one, 0:11:34.640 --> 0:11:38.199 we will not be ready for this next computer revolution 0:11:38.360 --> 0:11:39.840 that's coming after AI. 0:11:40.440 --> 0:11:42.920 Everybody in the world right now is doing nothing but 0:11:43.040 --> 0:11:47.600 talking about AI, and you are already thinking one step 0:11:47.640 --> 0:11:52.080 beyond the quantum? What's different about the opportunity that quantum 0:11:52.640 --> 0:11:56.439 creates for American research than AI. 0:11:57.720 --> 0:11:58.400 Biology? 0:11:58.880 --> 0:12:07.000 The human body, by definition, is much closer to fundamentals 0:12:07.000 --> 0:12:11.040 of quantum, you know, to quantum mechanics and quantum physics 0:12:11.080 --> 0:12:16.040 than it is to AI. AI is a classical computing 0:12:16.160 --> 0:12:21.440 approach that in essence reduces every piece of data to 0:12:21.600 --> 0:12:26.199 a black or white binary categorization of a. 0:12:26.200 --> 0:12:27.160 One or a zero. 0:12:28.280 --> 0:12:33.480 At its core nature around us, the human body, there 0:12:33.559 --> 0:12:39.560 is nothing categorical about it. It's that whole, you know, 0:12:39.679 --> 0:12:48.680 continuum of colors of life. Quantum its principles are that, 0:12:49.120 --> 0:12:53.440 you know, so there's all the scientific principles about quantum 0:12:53.640 --> 0:12:58.200 physics and superposition and in tanker, all of these complex 0:12:58.240 --> 0:13:00.760 things that people have a hard time with. But for 0:13:01.600 --> 0:13:07.080 in essence, it really is much more aligned, Like if 0:13:07.160 --> 0:13:10.640 I want to draw a colored picture, I will not 0:13:10.760 --> 0:13:14.199 go and pick up charcoal. You know, It's much easier 0:13:14.240 --> 0:13:17.880 for me to draw it if I had a colored 0:13:17.920 --> 0:13:18.640 palette with me. 0:13:18.800 --> 0:13:20.760 And quantum offers that. 0:13:20.840 --> 0:13:26.240 There is plenty of situations in medicine that are just intractable, 0:13:26.679 --> 0:13:29.960 you know, meaning it's not an issue just of it 0:13:30.000 --> 0:13:34.000 being AI being slow, or it doesn't have enough data, 0:13:34.480 --> 0:13:37.400 or if only we got more GPUs, you know, we 0:13:37.440 --> 0:13:42.479 can answer those questions. There are some problems in medicine 0:13:42.760 --> 0:13:46.600 that are fundamentally such that even if you give me 0:13:46.679 --> 0:13:49.680 all the GPUs in the world, there is no way 0:13:50.400 --> 0:13:55.800 that AI can model accurately how these molecules in the 0:13:55.880 --> 0:14:01.120 body are interacting among each other, or how comp electrons 0:14:01.160 --> 0:14:05.880 are moving within the mitochondria. These are the engines within 0:14:06.040 --> 0:14:11.920 our cells. There's these fundamental things in biology that AI 0:14:12.120 --> 0:14:16.320 and classical computers are just not built to be able 0:14:16.360 --> 0:14:17.040 to simulate. 0:14:17.720 --> 0:14:21.240 So you must go to dinner parties, doctor j. High. 0:14:21.600 --> 0:14:25.000 People must ask you what is quantum computing? What do 0:14:25.040 --> 0:14:25.760 you tell them? 0:14:26.280 --> 0:14:29.520 Yes, although I often, you know, we talk about other 0:14:29.560 --> 0:14:30.760 things at dinner parties. 0:14:32.760 --> 0:14:35.480 Eventually it comes down to you. If I was at 0:14:35.480 --> 0:14:38.160 that dinner party with you, I would ask you what 0:14:38.280 --> 0:14:39.040 is quantic computer? 0:14:39.440 --> 0:14:41.840 I would say, depends on how much time we have 0:14:42.000 --> 0:14:46.000 at the party to explain it. The short answer would be, 0:14:46.280 --> 0:14:51.640 it's a completely different way of working with computers than 0:14:51.680 --> 0:14:55.360 what we're used to right now. You can imagine that 0:14:55.520 --> 0:14:58.360 AI as being like a car that you take to 0:14:58.400 --> 0:15:02.200 go from one place to another. No matter how fast 0:15:02.480 --> 0:15:06.120 that Ferrari can get and how much fewre you put 0:15:06.160 --> 0:15:09.120 in it, it's never going to be a fighter jet. 0:15:09.440 --> 0:15:10.920 It's never going to be a plane. 0:15:11.840 --> 0:15:15.240 AI is the car, the plane is quant They are 0:15:15.680 --> 0:15:18.640 ways to get from point A to point B, but 0:15:18.880 --> 0:15:24.920 they work very differently, and we always use them in together. 0:15:25.400 --> 0:15:30.200 If I'm flying from Shaker Heights to Yorktown Heights, I 0:15:30.480 --> 0:15:33.240 drive to the airport, get on the plane, then take 0:15:33.280 --> 0:15:37.120 an uber to get to Yorktown Heights. In research, we 0:15:37.160 --> 0:15:39.640 will do the same. We do some piece of it 0:15:39.720 --> 0:15:42.200 in AI, some piece of it in Quantum, and then 0:15:42.240 --> 0:15:43.120 go back and forth. 0:15:43.600 --> 0:15:47.200 Back in twenty twenty one, Cleveland Clinic and IBM announced 0:15:47.200 --> 0:15:50.800 it they were starting something called the Discovery Accelerator. What 0:15:50.920 --> 0:15:51.280 is that? 0:15:52.160 --> 0:15:56.600 It's an initiative, a program, a partnership really that is 0:15:57.320 --> 0:16:03.600 designed to a bridge at advanced computational tools and technology 0:16:04.800 --> 0:16:11.160 through the leader in that space, IBM with biomedical science 0:16:11.240 --> 0:16:17.120 and research and life sciences problems, and that is a 0:16:17.160 --> 0:16:21.240 Cleveland clinic. We called it a discovery accelerator because that 0:16:21.400 --> 0:16:24.120 was our goal. You know, we were both on both 0:16:24.160 --> 0:16:28.120 sides challenged to the fact that discovery in medicine was 0:16:28.160 --> 0:16:32.800 just taking too long. The classical example that really brought 0:16:32.800 --> 0:16:37.040 it to life to us then was drug discovery that 0:16:37.120 --> 0:16:40.160 it took over a decade and it still does actually 0:16:40.640 --> 0:16:44.360 over a decade. From the moment that there is a 0:16:44.480 --> 0:16:50.320 compound that someone in a lab biomedical lab thinks it 0:16:50.360 --> 0:16:53.880 would be effective to treat a certain disease, it takes 0:16:53.920 --> 0:16:57.240 about ten to thirteen years from that moment to when 0:16:57.280 --> 0:16:59.760 that drug is on a shelf for a patient to 0:16:59.760 --> 0:17:02.680 get tool from a pharmacy. And that was just too 0:17:02.720 --> 0:17:06.960 much of a gap to allow when we have so 0:17:07.119 --> 0:17:10.359 many health conditions that we needed to address, and a 0:17:10.400 --> 0:17:14.720 big part of that gap could be computationally solved. Better 0:17:14.800 --> 0:17:19.280 simulation of compounds, designing drug trails that are more efficient, 0:17:19.520 --> 0:17:23.679 you know, that would finish faster. So that was the 0:17:23.920 --> 0:17:31.800 motivation to bring computational tools and technology closer to biomedical researchers. 0:17:32.000 --> 0:17:35.639 Cleveland Click and IBM team up, and I'm assuming the 0:17:35.680 --> 0:17:39.520 IBM Quantum guys and other people descend on Cleveland and 0:17:39.560 --> 0:17:42.680 you have your first meeting with them. Are they telling 0:17:42.720 --> 0:17:45.920 you things you would never thought of? Or I mean, 0:17:46.080 --> 0:17:48.399 I'm just curious about what's the difference between what you 0:17:50.160 --> 0:17:54.159 thought was the potential was and what you discovered the 0:17:54.160 --> 0:17:54.919 potential was. 0:17:56.240 --> 0:18:00.000 That is an excellent question if I had to prioritize. 0:18:00.200 --> 0:18:03.840 Is one lesson that I learned over the past few 0:18:03.920 --> 0:18:08.000 years of doing this. It is that you can never 0:18:08.280 --> 0:18:12.560 know what your you know where your brain is going 0:18:12.640 --> 0:18:19.000 to go and discovery until you talk. You know, you 0:18:19.040 --> 0:18:22.200 have to open it up and really listen to try 0:18:22.240 --> 0:18:23.879 to learn what the other people are saying. 0:18:24.320 --> 0:18:26.520 It's it went both ways. 0:18:26.680 --> 0:18:27.720 Can you give you an example. 0:18:27.880 --> 0:18:30.320 So, okay, we built the quantum right, so it took 0:18:30.680 --> 0:18:34.200 it took like eight months to get this machine put together, and. 0:18:34.119 --> 0:18:38.720 We put it in our cafeteria. That's the whole other story. 0:18:38.440 --> 0:18:43.480 Where you put the quantum machine computer in your cafeteria? Yes, yes, yes, 0:18:45.320 --> 0:18:46.919 Can you see it when you're eating lunch? 0:18:47.359 --> 0:18:47.639 Yeah? 0:18:47.720 --> 0:18:50.520 Yeah, we have people eating lunch around it all the time. 0:18:51.000 --> 0:18:53.680 We wanted to have a machine that people can see. 0:18:53.920 --> 0:18:57.520 Otherwise it the program wouldn't launch properly. 0:18:57.600 --> 0:18:59.960 So I needed to have it physics. 0:19:00.720 --> 0:19:04.240 So we had to look at retrofitted in existing space, 0:19:04.280 --> 0:19:06.560 and the cafeteria space worked out. 0:19:06.720 --> 0:19:09.439 It was far enough from the street, you know, there was. 0:19:09.320 --> 0:19:13.720 No vibration that it was a double floor, you know, 0:19:13.760 --> 0:19:17.240 so the ceiling was high enough. It just like technically 0:19:17.320 --> 0:19:22.040 fit all of those requirements. And it was either there 0:19:22.200 --> 0:19:26.080 or we put it in our data center, which is 0:19:26.119 --> 0:19:30.800 a building and like another city close to Cleveland, and 0:19:30.920 --> 0:19:32.160 we picked the cafeteria. 0:19:32.440 --> 0:19:34.840 Yeah, now why I know this is sort of this 0:19:34.880 --> 0:19:37.119 is kind of hilarious, but there's a serious point I 0:19:37.119 --> 0:19:40.679 I want to touch on, which is why do you 0:19:40.760 --> 0:19:41.879 need it on premises? 0:19:42.000 --> 0:19:45.640 On the premises Cleveland Clinic research, We needed to change 0:19:46.240 --> 0:19:51.800 how we think about research and shift the mindset of 0:19:52.480 --> 0:19:56.159 all of our researchers and all. 0:19:56.040 --> 0:19:57.000 Of our researchers. 0:19:57.119 --> 0:20:01.320 I'm talking about three thousand individuals who are one hundred 0:20:01.359 --> 0:20:05.919 percent doing biomedical research and Cleveland Clinic two one hundred 0:20:05.920 --> 0:20:11.960 and thirty labs individual pis. So that's the scale that 0:20:12.440 --> 0:20:15.600 I'm talking about that we had to create an impact 0:20:15.760 --> 0:20:17.080 on so. 0:20:17.200 --> 0:20:22.040 Having it having it be there was. 0:20:22.400 --> 0:20:27.159 As much for inspiration and to trigger our motivation to 0:20:27.320 --> 0:20:31.800 change as it was a you know, a practical solution, say, 0:20:31.800 --> 0:20:36.359 because we had the space and the connections and all of. 0:20:36.280 --> 0:20:39.960 That I saw in it. IBM headquarters. They're beautiful. 0:20:39.960 --> 0:20:43.320 Their works of art they are They're gorgeous, and the 0:20:43.320 --> 0:20:49.080 one that we have is the most gorgeous one of all. 0:20:49.240 --> 0:20:53.560 This is not me, you know, the mom bias talking 0:20:53.600 --> 0:20:54.600 about it. 0:20:54.600 --> 0:20:57.920 It is you know, they got an award, the Red 0:20:57.960 --> 0:21:03.159 Dot Award for Design went to IBM and Cleveland Clinic 0:21:03.280 --> 0:21:10.800 for our quantum and the quantum that IBM built for 0:21:11.080 --> 0:21:15.439 RPI a couple of years after hours. They modeled it 0:21:15.560 --> 0:21:19.040 after hours, not after the you know the ones from before. 0:21:19.280 --> 0:21:22.800 Just just so people know, we're talking about basically a 0:21:22.840 --> 0:21:26.320 small garage. 0:21:26.280 --> 0:21:29.879 Size eleven foot by eleven eleven feet eleven feet. 0:21:29.720 --> 0:21:31.040 You know, the cube small. 0:21:31.720 --> 0:21:35.760 So it's a it's a glass cube, and the glass 0:21:35.800 --> 0:21:40.080 comes all the way from Italy. It's the same glass 0:21:40.560 --> 0:21:44.720 that protects the Crown jewels and you know the Mona 0:21:44.760 --> 0:21:49.080 Lisa and all that. So so there's a glass all 0:21:49.160 --> 0:21:52.920 around and then there is the tube that you see, 0:21:53.080 --> 0:21:56.919 the stainless tube that's shiny, you know, and clean, but 0:21:57.400 --> 0:22:00.159 the technologies all inside of it, you know, the the 0:22:00.240 --> 0:22:02.960 leer and then the processor and the bottom. 0:22:02.560 --> 0:22:09.320 And it's just a fascinating thing to watch. And it hums. 0:22:09.400 --> 0:22:12.679 It makes the sound so even it sounds alive. 0:22:15.000 --> 0:22:17.679 You have a great deal of affection for your computer, 0:22:17.800 --> 0:22:18.000 and we. 0:22:18.080 --> 0:22:19.560 Love our machine. 0:22:20.200 --> 0:22:22.440 I want to go back to something you said before, 0:22:22.520 --> 0:22:26.080 which is when you had your initial conversations with IBM, 0:22:26.600 --> 0:22:30.520 both sides learned things that they hadn't previously thought of. 0:22:30.760 --> 0:22:34.800 Give me an example of something that either side hadn't 0:22:34.840 --> 0:22:37.160 realized could be done with this new technology. 0:22:37.280 --> 0:22:41.480 Sure, I mean on the equivalent clinic side, we thought 0:22:41.840 --> 0:22:45.960 that what quantum should be good for is that it 0:22:46.000 --> 0:22:49.040 would be a faster computer, right, so that we have 0:22:49.200 --> 0:22:54.160 these big data sets that are requiring much more compute 0:22:54.200 --> 0:22:56.399 power and you know GPUs than what we have and 0:22:56.440 --> 0:22:57.160 we should. 0:22:56.880 --> 0:22:58.959 Just run them on the quantum. 0:22:59.560 --> 0:23:04.280 And what we came to realize is that quantum actually 0:23:04.280 --> 0:23:07.320 does not do well with these large data sets. What 0:23:07.400 --> 0:23:11.960 it does well with our smaller, better defined data sets, 0:23:12.000 --> 0:23:16.960 but ones where simulation is more important, you know, we 0:23:17.040 --> 0:23:22.280 have to run them through multiple models, you multiple simulations 0:23:22.280 --> 0:23:25.560 of how they interact. As an example, one of the 0:23:25.680 --> 0:23:29.560 very first projects we threw at Quantum was wanting to 0:23:29.640 --> 0:23:35.320 predict complications cardiac complications from general surgery using electronic health 0:23:35.359 --> 0:23:40.080 record data. So data from our electronic health records are 0:23:40.240 --> 0:23:43.639 by definition these really big data sets. You have in 0:23:43.680 --> 0:23:47.320 them every single thing that you know about the patient 0:23:47.520 --> 0:23:51.480 that you've collected. And we thought that Quantum would help 0:23:51.560 --> 0:23:56.600 us build models, better models with this data. And it 0:23:56.760 --> 0:24:00.720 failed miserably. It wasn't good, you know, at doing that thing. 0:24:00.800 --> 0:24:03.119 And why didn't it? Like that problem because it's too 0:24:03.480 --> 0:24:04.640 it's too bigger than wieldy. 0:24:05.240 --> 0:24:08.720 Well, because the hardware isn't tready right, So the hardware 0:24:08.880 --> 0:24:13.000 with Quantum, that issue was back then. You know, now 0:24:13.040 --> 0:24:17.400 we've upgraded our processor. But still Quantum now is limited 0:24:17.440 --> 0:24:23.360 by noise and by its ability to correct for errors 0:24:23.440 --> 0:24:29.159 when it's doing computation. And the more data that you 0:24:29.359 --> 0:24:31.840 throw at it that you require it, you know, to 0:24:32.040 --> 0:24:36.440 put in its system, you know, to model interactions, the 0:24:36.480 --> 0:24:39.320 more errors it's likely to make. So then the harder 0:24:40.320 --> 0:24:42.960 it is for it to get to an answer. 0:24:42.640 --> 0:24:44.040 That you can trust. 0:24:44.600 --> 0:24:48.520 Now, we made a lot of progress sense, which I'm 0:24:48.560 --> 0:24:50.240 sure we'll get to. 0:24:50.440 --> 0:24:51.320 I hope we'll get. 0:24:51.200 --> 0:24:54.359 To with some recent breakthroughs that we made in that 0:24:54.440 --> 0:25:00.560 space with modeling large compounds and interactions. But the way 0:25:00.720 --> 0:25:02.800 that got us to where we are now where we 0:25:02.800 --> 0:25:06.600 could model large interactions, it took us figuring out that 0:25:06.720 --> 0:25:10.360 we shouldn't be doing everything on quantum. 0:25:10.720 --> 0:25:14.120 So give me an example of a problem that quantum 0:25:14.320 --> 0:25:18.080 is ideally suited for that the quantum really. 0:25:17.960 --> 0:25:22.560 Likes in drug discovery, for examine chemistry, it likes chemistry 0:25:22.640 --> 0:25:26.680 a lot because in chemistry what it has to What 0:25:26.760 --> 0:25:30.840 we wanted to model is how a drug that we 0:25:31.080 --> 0:25:34.320 put in our body is going to interact with We 0:25:34.359 --> 0:25:38.560 say the protein you know, the target, the ligand that 0:25:38.640 --> 0:25:41.240 it needs to bind too, like you know, the drug 0:25:41.320 --> 0:25:43.080