>>Hello everybody. I am Toshihiko Nishimura from Stanford. University is there to TTT out here, super aging, global OMIM global transportation group about infections, uh, or major point of concerns. In addition, this year, we have the COVID-19 pandemic. As you can see here, while the why the new COVID-19 patients are still increasing, meanwhile, case count per day in the United state, uh, beginning to decrease this pandemic has changed our daily life to digital transformation. Even today, the micro segmentation is being conducted online and doctor and the nurse care, uh, now increase to telemedicine. Likewise, the drug development process is in need of major change paradigm shift, especially in vaccine in drug development for COVID-19 is, should be safe, effective, and faster >>In the >>Anastasia department, which is the biggest department in school of medicine. We have Stanford, a love for drug device development, regulatory science. So cold. Say the DDT RDS chairman is Ron Paul and this love leaderships are long mysel and stable shaper. In the drug development. We have three major pains, one exceedingly long duration that just 20 years huge budget, very low success rate general overview in the drug development. There are Discoverly but clinical clinical stage, as you see here, Tang. Yes. In clinical stage where we sit, say, what are the programs in D D D R S in each stages or mix program? Single cell programs, big data machine learning, deep learning, AI mathematics, statistics programs, humanized animal, the program SNS program engineering program. And we have annual symposium. Today's the, my talk, I do like to explain limitation of my science significance of humanized. My science out of separate out a program. I focused on humanized program. I believe this program is potent game changer for drug development mouse. When we think of animal experiment, many people think of immediately mouse. We have more than 30 kinds of inbred while the type such as chief 57, black KK yarrow, barber C white and so on using QA QC defined. Why did the type mice 18 of them gave him only one intervention using mouse, genomics analyzed, computational genetics. And then we succeeded to pick up fish one single gene in a week. >>We have another category of gene manipulated, mice transgenic, no clout, no Kamal's group. So far registered 40,000 kind as over today. Pretty critical requirement. Wrong FDA PMDA negative three sites are based on arteries. Two kinds of animal models, showing safety efficacy, combination of two animals and motel our mouse and the swine mouse and non-human primate. And so on mouse. Oh, Barry popular. Why? Because mouse are small enough, easy to handle big database we had and cost effective. However, it calls that low success rate. Why >>It, this issue speculation, low success rate came from a gap between preclinical the POC and the POC couldn't stay. Father divided into phase one. Phase two has the city FDA unsolved to our question. Speculation in nature biology using 7,372 new submissions, they found a 68 significant cradle out crazy too, to study approved by the process. And in total 90 per cent Radia in the clinical stages. What we can surmise from this study, FDA confirmed is that the big discrepancy between POC and clinical POC in another ward, any amount of data well, Ms. Representative for human, this nature bio report impacted our work significantly. >>What is a solution for this discrepancy? FDA standards require the people data from two species. One species is usually mice, but if the reported 90% in a preclinical data, then huge discrepancy between pretty critical POC in clinical POC. Our interpretation is data from mice, sometime representative, actually mice, and the humor of different especially immune system and the diva mice liver enzyme are missing, which human Liba has. This is one huge issue to be taught to overcome this problem. We started humanized mice program. What kind of human animals? We created one humanized, immune mice. The other is human eyes, DBA, mice. What is the definition of a humanized mice? They should have human gene or human cells or human tissues or human organs. Well, let me share one preclinical stages. Example of a humanized mouse that is polio receptor mice. This problem led by who was my mentor? Polio virus. Well, polio virus vaccine usually required no human primate to test in 13 years, collaboration with the FDA w H O polio eradication program. Finally FDA well as w H O R Purdue due to the place no human primate test to transgenic PVL. This is three. Our principle led by loss around the botch >>To move before this humanized mouse program, we need two other bonds donut outside your science, as well as the CPN mouse science >>human hormone, like GM CSF, Whoah, GCSF producing or human cytokine. those producing emoji mice are required in the long run. Two maintain human cells in their body under generation here, South the generation here, Dr. already created more than 100 kinds based on Z. The 100 kinds of Noe mice, we succeeded to create the human immune mice led the blood. The cell quite about the cell platelets are beautifully constituted in an mice, human and rebar MAs also succeeded to create using deparent human base. We have AGN diva, humanized mouse, American African human nine-thirty by mice co-case kitchen, humanized mice. These are Hennessy humanized, the immune and rebar model. On the other hand, we created disease rebar human either must to one example, congenital Liba disease, our guidance Schindel on patient model. >>The other model, we have infectious DDS and Waddell council Modell and GVH Modell. And so on creature stage or phase can a human itemize apply. Our objective is any stage. Any phase would be to, to propose. We propose experiment, pose a compound, which showed a huge discrepancy between. If Y you show the huge discrepancy, if Y is lucrative analog and the potent anti hepatitis B candidate in that predict clinical stage, it didn't show any toxicity in mice got dark and no human primate. On the other hand, weighing into clinical stage and crazy to October 15, salvage, five of people died and other 10 the show to very severe condition. >>Is that the reason why Nicole traditional the mice model is that throughout this, another mice Modell did not predict this severe side outcome. Why Zack humanized mouse, the Debar Modell demonstrate itself? Yes. Within few days that chemistry data and the puzzle physiology data phase two and phase the city requires huge number of a human subject. For example, COVID-19 vaccine development by Pfizer, AstraZeneca Moderna today, they are sample size are Southeast thousand vaccine development for COVID-19. She Novak UConn in China books for the us Erica Jones on the Johnson in unite United Kingdom. Well, there are now no box us Osaka Osaka, university hundred Japan. They are already in phase two industry discovery and predict clinical and regulatory stage foster in-app. However, clinical stage is a studious role because that phases required hugely number or the human subject 9,000 to 30,000. Even my conclusion, a humanized mouse model shortens the duration of drug development humanize, and most Isabel, uh, can be increase the success rate of drug development. Thank you for Ron Paul and to Steven YALI pelt at Stanford and and his team and or other colleagues. Thank you for listening.

>> Live from the MIT campus in Cambridge, Massachusetts, it's the Cube, covering the 12th Annual MIT Chief Data Officer and Information Quality Symposium. Brought to you by SiliconANGLE Media. >> Welcome back to the Cube's coverage of MITCDOIQ here in Cambridge, Massachusetts. I'm your host, Rebecca Knight, along with my co-host, Peter Burris. We're joined by Dr. Santikary, he is the vice-president and chief data officer at ERT. Thanks so much for coming on the show. >> Thanks for inviting me. >> We're going to call you Santi, that's what you go by. So, start by telling our viewers a little bit about ERT. What you do, and what kind of products you deliver to clients. >> I'll be happy to do that. The ERT is a clinical trial small company and we are a global data and technology company that minimizes risks and uncertainties within clinical trials for our customers. Our customers are top pharma companies, biotechnologic companies, medical device companies and they trust us to run their clinical trials so that they can bring their life-saving drugs to the market on time and every time. So we have a huge responsibility in that regard, because they put their trust in us, so we serve as their custodians of data and the processes, and the therapeutic experience that you bring to the table as well as compliance-related expertise that we have. So not only do we provide data and technology expertise, we also provide science expertise, regulatory expertise, so that's one of the reasons they trust us. And we also have been around since 1977, so it's almost over 50 years, so we have this collective wisdom that we have gathered over the years. And we have really earned trust in this past and because we deal with safety and efficacy of drugs and these are the two big components that help MDA, or any regulatory authority for that matter, to approve the drugs. So we have a huge responsibility in this regard, as well. In terms of product, as I said, we are in the safety and efficacy side of the clinical trial process, and as part of that, we have multiple product lines. We have respiratory product lines, we have cardiac safety product lines, we have imaging. As you know, imaging is becoming more and more so important for every clinical trial and particularly on oncology space for sure. To measure the growth of the tumor and that kind of things. So we have a business that focuses exclusively on the imaging side. And then we have data and analytics side of the house, because we provide real-time information about the trial itself, so that our customers can really measure risks and uncertainties before they become a problem. >> At this symposium, you're going to be giving a talk about clinical trials and the problems of, the missteps that can happen when the data is not accurate. Lay out the problem for our viewers, and then we're going to talk about the best practices that have emerged. >> I think that clinical trial space is very complex by its own nature, and the process itself is very lengthy. If you know one of the statistics, for example, it takes about 10 to 15 years to really develop and commercialize a drug. And it usually costs about $2.5 to 3 billion. Per drug. So think about the enormity of this. So the challenges are too many. One is data collection itself. Your clinical trials are becoming more and more complex. Becoming more and more global. Getting patients to the sites is another problem. Patient selection and retention, another one. Regulatory guidelines is another big issue because not every regulated authority follows the same sets of rules and regulations. And cost. Cost is a big imperative to the whole thing, because the development life-cycle of a drug is so lengthy. And as I said, it takes about $3 billion to commercialize a drug and that cost comes down to the consumers. That means patients. So the cost of the health care is growing, is sky-rocketing. And in terms of data collection, there are lots of devices in the field, as you know. Wearables, mobile helds, so the data volume is a tremendous problem. And the vendors. Each pharmaceutical companies use so many vendors to run their trials. CRO's. The Clinical Research Organizations. They have EDC systems, they can have labs. You name it. So they outsource all these to different vendors. Now, how do you coordinate and how do you make them to collaborate? And that's where the data plays a big role because now the data is everywhere across different systems, and those systems don't talk to each other. So how do you really make real-time decisioning when you don't know where your data is? And data is the primary ingredient that you use to make decisions? So that's where data and analytics, and bringing that data in real-time, is a very, very critical service that we provide to our customers. >> When you look at medicine, obviously, the whole notion of evidence-based medicine has been around for 15 years now, and it's becoming a seminal feature of how we think about the process of delivering medical services and ultimately paying it forward to everything else, and partly that's because doctors are scientists and they have an affinity for data. But if we think about going forward, it seems to me as though learning more about the genome and genomics is catalyzing additional need and additional understanding of the role that drugs play in the human body and it almost becomes an information problem, where the drug, I don't want to say that a drug is software, but a drug is delivering something that, ultimately, is going to get known at a genomic level. So does that catalyze additional need for data? is that changing the way we think about clinical trials? Especially when we think about, as you said, it's getting more complex because we have to make sure that a drug has the desired effect with men and women, with people from here, people from there. Are we going to push the data envelope even harder over the next few years? >> Oh, you bet. And that's where the real world evidence is playing a big role. So, instead of patients coming to the clinical trials, clinical trial is going to the patient. It is becoming more and more patient-centric. >> Interesting. >> And the early part of protocol design, for example, the study design, that is step one. So more and more the real world evidence data is being used to design the protocol. The very first stage of the clinical trial. Another thing that is pushing the envelope is artificial intelligence and other data mining techniques and now people can be used to really mine that data, the MAR data, prescription data, claims data. Those are real evidence data coming from the real patients. So now you can use these artificial intelligence and mission learning techniques to mine that data then to really design the protocol and the study design instead of flipping through the year MAR data manually. So patient collection, for example, is no patients, no trials, right? So gathering patients, and the right set of patients, is one of the big problems. It takes a lot of that time to bring those patients and even more troublesome is to retain those patients over time. These, too, are big, big things that take a long time and site selection, as well. Which site is going to really be able to bring the right patients for the right trials? >> So, two quick comments on that. One of the things, when you say the patients, when someone has a chronic problem, a chronic disease, when they start to feel better as a consequence of taking the drug, they tend to not take the drug anymore. And that creates this ongoing cycle. But going back to what you're saying, does it also mean that clinical trial processes, because we can gather data more successfully over time, it used to be really segmented. We did the clinical trial and it stopped. Then the drug went into production and maybe we caught some data. But now because we can do a better job with data, the clinical trial concept can be sustained a little bit more. That data becomes even more valuable over time and we can add additional volumes of data back in, to improve the process. >> Is that shortening clinical trials? Tell us a little bit about that. >> Yes, as I said, it takes 10 to 15 years if we follow the current process, like Phase One, Phase Two, Phase Three. And then post-marketing, that is Phase Four. I'm not taking the pre-clinical side of these trials in the the picture. That's about 10 to 15 years, about $3 billion kind of thing. So when you use these kind of AI techniques and the real world evidence data and all this, the projection is that it will reduce the cycle by 60 to 70%. >> Wow. >> The whole study, beginning to end time. >> So from 15 down to four or five? >> Exactly. So think about, there are two advantages. One is obviously, you are creating efficiency within the system, and this drug industry and drug discovery industry is rife for disruption. Because it has been using that same process over and over for a long time. It's like, it is working, so why fix it? But unfortunately, it's not working. Because the health care cost has sky-rocketed. So these inefficiencies are going to get solved when we employ real world evidencing into the mixture. Real-time decision making. Risks analysis before they become risks. Instead of spending one year to recruit patients, you use AI techniques to get to the right patients in minutes, so think about the efficiency again. And also, the home monitoring, or mHealth type of program, where the patients don't need to come to the sites, the clinical sites, for check-up anymore. You can wear wearables that are MDA regulated and approved and then, they're going to do all the work from within the comfort of their home. So think about that. And the other thing is, very, terminally sick patients, for example. They don't have time, nor do they have the energy, to come to the clinical site for check-up. Because every day is important to them. So, this is the paradigm shift that is going on. Instead of patients coming to the clinical trials, clinical trials are coming to the patients. And that shift, that's a paradigm shift and that is happening because of these AI techniques. Blockchain. Precision Medicine is another one. You don't run a big clinical trial anymore. You just go micro-trial, you just group small number of patients. You don't run a trial on breast cancer anymore, you just say, breast cancer for these patients, so it's micro-trials. And that needs -- >> Well that can still be aggregated. >> Exactly. It still needs to be aggregated, but you can get the RTD's quickly, so that you can decide whether you need to keep investing in that trial, or not. Instead of waiting 10 years, only to find out that your trial is going to fail. So you are wasting not only your time, but also preventing patients from getting the right medicine on time. So you have that responsibility as a pharmaceutical company, as well. So yes, it is a paradigm shift and this whole industry is rife for disruption and ERT is right at the center. We have not only data and technology experience, but as I said, we have deep domain experience within the clinical domain as well as regulatory and compliance experience. You need all these to navigate through this turbulent water of clinical research. >> Revolutionary changes taking place. >> It is and the satisfaction is, you are really helping the patients. You know? >> And helping the doctor. >> Helping the doctors. >> At the end of the day, the drug company does not supply the drug. >> Exactly. >> The doctor is prescribing, based on knowledge that she has about that patient and that drug and how they're going to work together. >> And out of the good statistics, in 2017, just last year, 60% of the MDA approved drugs were supported through our platform. 60 percent. So there were, I think, 60 drugs got approved? I think 30 or 35 of them used our platform to run their clinical trial, so think about the satisfaction that we have. >> A job well done. >> Exactly. >> Well, thank you for coming on the show Santi, it's been really great having you on. >> Thank you very much. >> Yes. >> Thank you. >> I'm Rebecca Knight. For Peter Burris, we will have more from MITCDOIQ, and the Cube's coverage of it. just after this. (techno music)

>> Live from San Francisco, it's the Cube, covering Google Cloud Next 2018. Brought to you by Google Cloud and its ecosystem partners. >> Hello and welcome back to The Cube coverage here live in San Francisco for Google Cloud's conference Next 2018, #GoogleNext18. I'm John Furrier with Jeff Frick, my cohost all week. Third day of three days of wall to wall live coverage. Our next guest, Carol Carpenter, Vice President of Product Marketing for Google Cloud. And Ayin Vala, Chief Data Science Foundation for Precision Medicine. Welcome to The Cube, thanks for joining us. >> Thank you for having us. >> So congratulations, VP of Product Marketing. Great job getting all these announcements out, all these different products. Open source, big query machine learning, Istio, One dot, I mean, all this, tons of products, congratulations. >> Thank you, thank you. It was a tremendous amount of work. Great team. >> So you guys are starting to show real progress in customer traction, customer scale. Google's always had great technology. Consumption side of it, you guys have made progress. Diane Green mentioned on stage, on day one, she mentioned health care. She mentioned how you guys are organizing around these verticals. Health care is one of the big areas. Precision Medicine, AI usage, tell us about your story. >> Yes, so we are a very small non-profit. And we are at the intersection of data science and medical science and we work on projects that have non-profits impact and social impact. And we work on driving and developing projects that have social impact and in personalized medicine. >> So I think it's amazing. I always think with medicine, right, you look back five years wherever you are and you look back five years and think, oh my god, that was completely barbaric, right. They used to bleed people out and here, today, we still help cancer patients by basically poisoning them until they almost die and hopefully it kills the cancer first. You guys are looking at medicine in a very different way and the future medicine is so different than what it is today. And talk about, what is Presicion Medicine? Just the descriptor, it's a very different approach to kind of some of the treatments that we still use today in 2018. It's crazy. >> Yes, so Presicion Medicine has the meaning of personalized medicine. Meaning that we hone it into smaller population of people to trying to see what is the driving factors, individually customized to those populations and find out the different variables that are important for that population of people for detection of the disease, you know, cancer, Alzheimer's, those things. >> Okay, talk about the news. Okay, go ahead. >> Oh, oh, I was just going to say. And to be able to do what he's doing requires a lot of computational power to be able to actually get that precise. >> Right. Talk about the relationship and the news you guys have here. Some interesting stuff. Non-profits, they need compute power, they need, just like an eneterprise. You guys are bringing some change. What's the relationship between you guys? How are you working together? >> So one of our key messages here at this event is really around making computing available for everyone. Making data and analytics and machine learning available for everyone. This whole idea of human-centered AI. And what we've realized is, you know, data is the new natural resource. >> Yeah. >> In the world these days. And companies that know how to take advantage and actually mine insights from the data to solve problems like what they're solving at Precision Medicine. That is really where the new breakthroughs are going to come. So we announced a program here at the event, It's called Data Solutions for Change. It's from Google Cloud and it's a program in addition to our other non-profit programs. So we actually have other programs like Google Earth for non-profits. G Suite for non-profits. This one is very much focused on harnessing and helping non-profits extract insights from data. >> And is it a funding program, is it technology transfer Can you talk about, just a little detail on how it actually works. >> It's actually a combination of three things. One is funding, it's credits for up to $5,000 a month for up to six months. As well as customer support. One thing we've all talked about is the technology is amazing. You often also need to be able to apply some business logic around it and data scientists are somewhat of a challenge to hire these days. >> Yeah. >> So we're also proving free customer support, as well as online learning. >> Talk about an impact of the Cloud technology for the non-proit because6 I, you know, I'm seeing so much activity, certainly in Washington D.C. and around the world, where, you know, since the Jobs Act, fundings have changed. You got great things happening. You can have funding on mission-based funding. And also, the legacy of brand's are changing and open source changes So faster time to value. (laughs) >> Right. >> And without all the, you know, expertise it's an issue. How is Cloud helping you be better at what you do? Can you give some examples? >> Yes, so we had two different problems early on, as a small non-profit. First of all, we needed to scale up computationally. We had in-house servers. We needed a HIPAA complaint way to put our data up. So that's one of the reasons we were able to even use Google Cloud in the beginning. And now, we are able to run our models or entire data sets. Before that, we were only using a small population. And in Presicion Medicine, that's very important 'cause you want to get% entire population. That makes your models much more accurate. The second things was, we wanted to collaborate with people with clinical research backgrounds. And we need to provide a platform for them to be able to use, have the data on there, visualize, do computations, anything they want to do. And being on a Cloud really helped us to collaborate much more smoothly and you know, we only need their Gmail access, you know to Gmail to give them access and things. >> Yeah. >> And we could do it very, very quickly. Whereas before, it would take us months to transfer data. >> Yeah, it's a huge savings. Talk about the machine learning, AutoML's hot at the show, obviously, hot trend. You start to see AI ops coming in and disrupt more of the enterprise side but as data scientists, as you look at some of these machine learnings, I mean, you must get pretty excited. What are you thinking? What's your vision and how you going to use, like BigQuery's got ML built in now. This is like not new, it's Google's been using it for awhile. Are you tapping some of that? And what's your team doing with ML? >> Absolutely. We use BigQuery ML. We were able to use a few months in advance. It's great 'cause our data scientists like to work in BigQuery. They used to see, you know, you query the data right there. You can actually do the machine learning on there too. And you don't have to send it to different part of the platform for that. And it gives you sort of a proof of concept right away. For doing deep learning and those things, we use Cloud ML still, but for early on, you want to see if there is potential in a data. And you're able to do that very quickly with BigQuery ML right there. We also use AutoML Vision. We had access to about a thousand patients for MRI images and we wanted to see if we can detect Alzheimer's based on those. And we used AutoML for that. Actually works well. >> Some of the relationships with doctors, they're not always seen as the most tech savvy. So now they are getting more. As you do all this high-end, geeky stuff, you got to push it out to an interface. Google's really user-centric philosophy with user interfaces has always been kind of known for. Is that in Sheets, is that G Suite? How will you extend out the analysis and the interactions. How do you integrate into the edge work flow? You know? (laughs) >> So one thing I really appreciated for Google Cloud was that it was, seems to me it's built from the ground up for everyone to use. And it was the ease of access was very, was very important to us, like I said. We have data scientisits and statisticians and computer scientists onboard. But we needed a method and a platform that everybody can use. And through this program, they actually.. You guys provide what's called Qwiklab, which is, you know, screenshot of how to spin up a virtual machine and things like that. That, you know, a couple of years ago you have to run, you know, few command lines, too many command lines, to get that. Now it's just a push of a button. So that's just... Makes it much easier to work with people with background and domain knowledge and take away that 80% of the work, that's just a data engineering work that they don't want to do. >> That's awesome stuff. Well congratulations. Carol, a question to you is How does someone get involved in the Data Solutions for Change? An application? Online? Referral? I mean, how do these work? >> All of the above. (John laughs) We do have an online application and we welcome all non-profits to apply if they have a clear objective data problem that they want to solve. We would love to be able to help them. >> Does scope matter, big size, is it more mission? What's the mission criteria? Is there a certain bar to reach, so to speak, or-- >> Yeah, I mean we're most focused on... there really is not size, in terms of size of the non-profit or the breadth. It's much more around, do you have a problem that data and analytics can actually address. >> Yeah. >> So really working on problems that matter. And in addition, we actually announced this week that we are partnering with United Nations on a contest. It's called Sustainable.. It's for Visualize 2030 >> Yeah. >> So there are 17 sustainable development goals. >> Right, righr. >> And so, that's aimed at college students and storytelling to actually address one of these 17 areas. >> We'd love to follow up after the show, talk about some of the projects. since you have a lot of things going on. >> Yeah. >> Use of technology for good really is important right now, that people see that. People want to work for mission-driven organizations. >> Absolutely >> This becomes a clear citeria. Thanks for coming on. Appreciate it. Thanks for coming on today. Acute coverage here at Google Could Next 18 I'm John Furrier with Jeff Fricks. Stay with us. More coverage after this short break. (upbeat music)

>> Announcer: Live, from Las Vegas, it's the Cube, covering HPE Discover 2017. Brought to you by Hewlett Packard Enterprise. >> Hey, welcome back everyone. We are here live in Las Vegas for day two of three days of exclusive coverage from the Cube here at HPE Discover 2017. Our two next guests is Bill Mannel, VP and General Manager of HPC and AI for HPE. Bill, great to see you. And Dr. Nick Nystrom, senior of research at Pittsburgh's Supercomputer Center. Welcome to The Cube, thanks for coming on, appreciate it. >> My pleasure >> Thanks for having us. >> As we wrap up day two, first of all before we get started, love the AI, love the high performance computing. We're seeing great applications for compute. Everyone now sees that a lot of compute actually is good. That's awesome. What is the Pittsburgh Supercomputer Center? Give a quick update and describe what that is. >> Sure. The quick update is we're operating a system called Bridges. Bridges is operating for the National Science Foundation. It democratizes HPC. It brings people who have never used high performance computing before to be able to use HPC seamlessly, almost as a cloud. It unifies HPC big data and artificial intelligence. >> So who are some of the users that are getting access that they didn't have before? Could you just kind of talk about some of the use cases of the organizations or people that you guys are opening this up to? >> Sure. I think one of the newest communities that's very significant is deep learning. So we have collaborations between the University of Pittsburgh life sciences and the medical center with Carnegie Mellon, the machine learning researchers. We're looking to apply AI machine learning to problems in breast and lung cancer. >> Yeah, we're seeing the data. Talk about some of the innovations that HPE's bringing with you guys in the partnership, because we're seeing, people are seeing the results of using big data and deep learning and breakthroughs that weren't possible before. So not only do you have the democratization cool element happening, you have a tsunami of awesome open source code coming in from big places. You see Google donating a bunch of machine learning libraries. Everyone's donating code. It's like open bar and open source, as I say, and the young kids that are new are the innovators as well, so not just us systems guys, but a lot of young developers are coming in. What's the innovation? Why is this happening? What's the ah-ha moment? Is it just cloud, is it a combination of things, talk about it. >> It's a combination of all the big data coming in, and then new techniques that allow us to analyze and get value from it and from that standpoint. So the traditional HPC world, typically we built equations which then generated data. Now we're actually kind of doing the reverse, which is we take the data and then build equations to understand the data. So it's a different paradigm. And so there's more and more energy understanding those two different techniques of kind of getting two of the same answers, but in a different way. >> So Bill, you and I talked in London last year. >> Yes. With Dr. Gho. And we talked a lot about SGI and what that acquisition meant to you guys. So I wonder if you could give us a quick update on the business? I mean it's doing very well, Meg talked about it on the conference call this last quarter. Really high point and growing. What's driving the growth, and give us an update on the business. >> Sure. And I think the thing that's driving the growth is all this data and the fact that customers want to get value from it. So we're seeing a lot of growth in industries like financial services, like in manufacturing, where folks are moving to digitization, which means that in the past they might have done a lot of their work through experimentation. Now they're moving it to a digital format, and they're simulating everything. So that's driven a lot more HPC over time. As far as the SGI, integration is concern. We've integrated about halfway, so we're at about the halfway point. And now we've got the engineering teams together and we're driving a road map and a new set of products that are coming out. Our Gen 10-based products are on target, and they're going to be releasing here over the next few months. >> So Nick, from your standpoint, when you look at, there's been an ebb and flow in the supercomputer landscape for decades. All the way back to the 70s and the 80s. So from a customer perspective, what do you see now? Obviously China's much more prominent in the game. There's sort of an arms race, if you will, in computing power. From a customer's perspective, what are you seeing, what are you looking for in a supplier? >> Well, so I agree with you, there is this arms race for exaflops. Where we are really focused right now is enabling data-intensive applications, looking at big data service, HPC is a service, really making things available to users to be able to draw on the large data sets you mentioned, to be able to put the capability class computing, which will go to exascale, together with AI, and data and Linux under one platform, under one integrated fabric. That's what we did with HPE for Bridges. And looking to build on that in the future, to be able to do the exascale applications that you're referring to, but also to couple on data, and to be able to use AI with classic simulation to make those simulations better. >> So it's always good to have a true practitioner on The Cube. But when you talk about AI and machine learning and deep learning, John and I sometimes joke, is it same wine, new bottle, or is there really some fundamental shift going on that just sort of happened to emerge in the last six to nine months? >> I think there is a fundamental shift. And the shift is due to what Bill mentioned. It's the availability of data. So we have that. We have more and more communities who are building on that. You mentioned the open source frameworks. So yes, they're building on the TensorFlows, on the Cafes, and we have people who have not been programmers. They're using these frameworks though, and using that to drive insights from data they did not have access to. >> These are flipped upside down, I mean this is your point, I mean, Bill pointed it out, it's like the models are upside down. This is the new world. I mean, it's crazy, I don't believe it. >> So if that's the case, and I believe it, it feels like we're entering this new wave of innovation which for decades we talked about how we march to the cadence of Moore's Law. That's been the innovation. You think back, you know, your five megabyte disk drive, then it went to 10, then 20, 30, now it's four terabytes. Okay, wow. Compared to what we're about to see, I mean it pales in comparison. So help us envision what the world is going to look like in 10 or 20 years. And I know it's hard to do that, but can you help us get our minds around the potential that this industry is going to tap? >> So I think, first of all, I think the potential of AI is very hard to predict. We see that. What we demonstrated in Pittsburgh with the victory of Libratus, the poker-playing bot, over the world's best humans, is the ability of an AI to beat humans in a situation where they have incomplete information, where you have an antagonist, an adversary who is bluffing, who is reacting to you, and who you have to deal with. And I think that's a real breakthrough. We're going to see that move into other aspects of life. It will be buried in apps. It will be transparent to a lot of us, but those sorts of AI's are going to influence a lot. That's going to take a lot of IT on the back end for the infrastructure, because these will continue to be compute-hungry. >> So I always use the example of Kasperov and he got beaten by the machine, and then he started a competition to team up with a supercomputer and beat the machine. Yeah, humans and machines beat machines. Do you expect that's going to continue? Maybe both your opinions. I mean, we're just sort of spitballing here. But will that augmentation continue for an indefinite period of time, or are we going to see the day that it doesn't happen? >> I think over time you'll continue to see progress, and you'll continue to see more and more regular type of symmetric type workloads being done by machines, and that allows us to do the really complicated things that the human brain is able to better process than perhaps a machine brain, if you will. So I think it's exciting from the standpoint of being able to take some of those other roles and so forth, and be able to get those done in perhaps a more efficient manner than we're able to do. >> Bill, talk about, I want to get your reaction to the concept of data. As data evolves, you brought up the model, I like the way you're going with that, because things are being flipped around. In the old days, I want to monetize my data. I have data sets, people are looking at their data. I'm going to make money from my data. So people would talk about how we monetizing the data. >> Dave: Old days, like two years ago. >> Well and people actually try to solve and monetize their data, and this could be use case for one piece of it. Other people are saying no, I'm going to open, make people own their own data, make it shareable, make it more of an enabling opportunity, or creating opportunities to monetize differently. In a different shift. That really comes down to the insights question. What's your, what trends do you guys see emerging where data is much more of a fabric, it's less of a discreet, monetizable asset, but more of an enabling asset. What's your vision on the role of data? As developers start weaving in some of these insights. You mentioned the AI, I think that's right on. What's your reaction to the role of data, the value of the data? >> Well, I think one thing that we're seeing in some of our, especially our big industrial customers is the fact that they really want to be able to share that data together and collect it in one place, and then have that regularly updated. So if you look at a big aircraft manufacturer, for example, they actually are putting sensors all over their aircraft, and in realtime, bringing data down and putting it into a place where now as they're doing new designs, they can access that data, and use that data as a way of making design trade-offs and design decision. So a lot of customers that I talk to in the industrial area are really trying to capitalize on all the data possible to allow them to bring new insights in, to predict things like future failures, to figure out how they need to maintain whatever they have in the field and those sorts of things at all. So it's just kind of keeping it within the enterprise itself. I mean, that's a challenge, a really big challenge, just to get data collected in one place and be able to efficiently use it just within an enterprise. We're not even talking about sort of pan-enterprise, but just within the enterprise. That is a significant change that we're seeing. Actually an effort to do that and see the value in that. >> And the high performance computing really highlights some of these nuggets that are coming out. If you just throw compute at something, if you set it up and wrangle it, you're going to get these insights. I mean, new opportunities. >> Bill: Yeah, absolutely. >> What's your vision, Nick? How do you see the data, how do you talk to your peers and people who are generally curious on how to approach it? How to architect data modeling and how to think about it? >> I think one of the clearest examples on managing that sort of data comes from the life sciences. So we're working with researchers at University of Pittsburgh Medical Center, and the Institute for Precision Medicine at Pitt Cancer Center. And there it's bringing together the large data as Bill alluded to. But there it's very disparate data. It is genomic data. It is individual tumor data from individual patients across their lifetime. It is imaging data. It's the electronic health records. And trying to be able to do this sort of AI on that to be able to deliver true precision medicine, to be able to say that for a given tumor type, we can look into that and give you the right therapy, or even more interestingly, how can we prevent some of these issues proactively? >> Dr. Nystrom, it's expensive doing what you do. Is there a commercial opportunity at the end of the rainbow here for you or is that taboo, I mean, is that a good thing? >> No, thank you, it's both. So as a national supercomputing center, our resources are absolutely free for open research. That's a good use of our taxpayer dollars. They've funded these, we've worked with HP, we've designed the system that's great for everybody. We also can make this available to industry at an extremely low rate because it is a federal resource. We do not make a profit on that. But looking forward, we are working with local industry to let them test things, to try out ideas, especially in AI. A lot of people want to do AI, they don't know what to do. And so we can help them. We can help them architect solutions, put things on hardware, and when they determine what works, then they can scale that up, either locally on prem, or with us. >> This is a great digital resource. You talk about federally funded. I mean, you can look at Yosemite, it's a state park, you know, Yellowstone, these are natural resources, but now when you start thinking about the goodness that's being funded. You want to talk about democratization, medicine is just the tip of the iceberg. This is an interesting model as we move forward. We see what's going on in government, and see how things are instrumented, some things not, delivery of drugs and medical care, all these things are coalescing. How do you see this digital age extending? Because if this continues, we should be doing more of these, right? >> We should be. We need to be. >> It makes sense. So is there, I mean I just not up to speed on what's going on with federally funded-- >> Yeah, I think one thing that Pittsburgh has done with the Bridges machine, is really try to bring in data and compute and all the different types of disciplines in there, and provide a place where a lot of people can learn, they can build applications and things like that. That's really unusual in HPC. A lot of times HPC is around big iron. People want to have the biggest iron basically on the top 500 list. This is where the focus hasn't been on that. This is where the focus has been on really creating value through the data, and getting people to utilize it, and then build more applications. >> You know, I'll make an observation. When we first started doing The Cube, we observed that, we talked about big data, and we said that the practitioners of big data, are where the guys are going to make all the money. And so far that's proven true. You look at the public big data companies, none of them are making any money. And maybe this was sort of true with ERP, but not like it is with big data. It feels like AI is going to be similar, that the consumers of AI, those people that can find insights from that data are really where the big money is going to be made here. I don't know, it just feels like-- >> You mean a long tail of value creation? >> Yeah, in other words, you used to see in the computing industry, it was Microsoft and Intel became, you know, trillion dollar value companies, and maybe there's a couple of others. But it really seems to be the folks that are absorbing those technologies, applying them, solving problems, whether it's health care, or logistics, transportation, etc., looks to where the huge economic opportunities may be. I don't know if you guys have thought about that. >> Well I think that's happened a little bit in big data. So if you look at what the financial services market has done, they've probably benefited far more than the companies that make the solutions, because now they understand what their consumers want, they can better predict their life insurance, how they should-- >> Dave: You could make that argument for Facebook, for sure. >> Absolutely, from that perspective. So I expect it to get to your point around AI as well, so the folks that really use it, use it well, will probably be the ones that benefit it. >> Because the tooling is very important. You've got to make the application. That's the end state in all this That's the rubber meets the road. >> Bill: Exactly. >> Nick: Absolutely. >> All right, so final question. What're you guys showing here at Discover? What's the big HPC? What's the story for you guys? >> So we're actually showing our Gen 10 product. So this is with the latest microprocessors in all of our Apollo lines. So these are specifically optimized platforms for HPC and now also artificial intelligence. We have a platform called the Apollo 6500, which is used by a lot of companies to do AI work, so it's a very dense GPU platform, and does a lot of processing and things in terms of video, audio, these types of things that are used a lot in some of the workflows around AI. >> Nick, anything spectacular for you here that you're interested in? >> So we did show here. We had video in Meg's opening session. And that was showing the poker result, and I think that was really significant, because it was actually a great amount of computing. It was 19 million core hours. So was an HPC AI application, and I think that was a really interesting success. >> The unperfect information really, we picked up this earlier in our last segment with your colleagues. It really amplifies the unstructured data world, right? People trying to solve the streaming problem. With all this velocity, you can't get everything, so you need to use machines, too. Otherwise you have a haystack of needles. Instead of trying to find the needles in the haystack, as they was saying. Okay, final question, just curious on this natural, not natural, federal resource. Natural resource, feels like it. Is there like a line to get in? Like I go to the park, like this camp waiting list, I got to get in there early. How do you guys handle the flow for access to the supercomputer center? Is it, my uncle works there, I know a friend of a friend? Is it a reservation system? I mean, who gets access to this awesomeness? >> So there's a peer reviewed system, it's fair. People apply for large allocations four times a year. This goes to a national committee. They met this past Sunday and Monday for the most recent. They evaluate the proposals based on merit, and they make awards accordingly. We make 90% of the system available through that means. We have 10% discretionary that we can make available to the corporate sector and to others who are doing proprietary research in data-intensive computing. >> Is there a duration, when you go through the application process, minimums and kind of like commitments that they get involved, for the folks who might be interested in hitting you up? >> For academic research, the normal award is one year. These are renewable, people can extend these and they do. What we see now of course is for large data resources. People keep those going. The AI knowledge base is 2.6 petabytes. That's a lot. For industrial engagements, those could be any length. >> John: Any startup action coming in, or more bigger, more-- >> Absolutely. A coworker of mine has been very active in life sciences startups in Pittsburgh, and engaging many of these. We have meetings every week with them now, it seems. And with other sectors, because that is such a great opportunity. >> Well congratulations. It's fantastic work, and we're happy to promote it and get the word out. Good to see HP involved as well. Thanks for sharing and congratulations. >> Absolutely. >> Good to see your work, guys. Okay, great way to end the day here. Democratizing supercomputing, bringing high performance computing. That's what the cloud's all about. That's what great software's out there with AI. I'm John Furrier, Dave Vellante bringing you all the data here from HPE Discover 2017. Stay tuned for more live action after this short break.

>> Welcome to the Intel AI Lounge. Today, we're very excited to share with you the Precision Medicine panel discussion. I'll be moderating the session. My name is Kay Erin. I'm the general manager of Health and Life Sciences at Intel. And I'm excited to share with you these three panelists that we have here. First is John Madison. He is a chief information medical officer and he is part of Kaiser Permanente. We're very excited to have you here. Thank you, John. >> Thank you. >> We also have Naveen Rao. He is the VP and general manager for the Artificial Intelligence Solutions at Intel. He's also the former CEO of Nervana, which was acquired by Intel. And we also have Bob Rogers, who's the chief data scientist at our AI solutions group. So, why don't we get started with our questions. I'm going to ask each of the panelists to talk, introduce themselves, as well as talk about how they got started with AI. So why don't we start with John? >> Sure, so can you hear me okay in the back? Can you hear? Okay, cool. So, I am a recovering evolutionary biologist and a recovering physician and a recovering geek. And I implemented the health record system for the first and largest region of Kaiser Permanente. And it's pretty obvious that most of the useful data in a health record, in lies in free text. So I started up a natural language processing team to be able to mine free text about a dozen years ago. So we can do things with that that you can't otherwise get out of health information. I'll give you an example. I read an article online from the New England Journal of Medicine about four years ago that said over half of all people who have had their spleen taken out were not properly vaccinated for a common form of pneumonia, and when your spleen's missing, you must have that vaccine or you die a very sudden death with sepsis. In fact, our medical director in Northern California's father died of that exact same scenario. So, when I read the article, I went to my structured data analytics team and to my natural language processing team and said please show me everybody who has had their spleen taken out and hasn't been appropriately vaccinated and we ran through about 20 million records in about three hours with the NLP team, and it took about three weeks with a structured data analytics team. That sounds counterintuitive but it actually happened that way. And it's not a competition for time only. It's a competition for quality and sensitivity and specificity. So we were able to indentify all of our members who had their spleen taken out, who should've had a pneumococcal vaccine. We vaccinated them and there are a number of people alive today who otherwise would've died absent that capability. So people don't really commonly associate natural language processing with machine learning, but in fact, natural language processing relies heavily and is the first really, highly successful example of machine learning. So we've done dozens of similar projects, mining free text data in millions of records very efficiently, very effectively. But it really helped advance the quality of care and reduce the cost of care. It's a natural step forward to go into the world of personalized medicine with the arrival of a 100-dollar genome, which is actually what it costs today to do a full genome sequence. Microbiomics, that is the ecosystem of bacteria that are in every organ of the body actually. And we know now that there is a profound influence of what's in our gut and how we metabolize drugs, what diseases we get. You can tell in a five year old, whether or not they were born by a vaginal delivery or a C-section delivery by virtue of the bacteria in the gut five years later. So if you look at the complexity of the data that exists in the genome, in the microbiome, in the health record with free text and you look at all the other sources of data like this streaming data from my wearable monitor that I'm part of a research study on Precision Medicine out of Stanford, there is a vast amount of disparate data, not to mention all the imaging, that really can collectively produce much more useful information to advance our understanding of science, and to advance our understanding of every individual. And then we can do the mash up of a much broader range of science in health care with a much deeper sense of data from an individual and to do that with structured questions and structured data is very yesterday. The only way we're going to be able to disambiguate those data and be able to operate on those data in concert and generate real useful answers from the broad array of data types and the massive quantity of data, is to let loose machine learning on all of those data substrates. So my team is moving down that pathway and we're very excited about the future prospects for doing that. >> Yeah, great. I think that's actually some of the things I'm very excited about in the future with some of the technologies we're developing. My background, I started actually being fascinated with computation in biological forms when I was nine. Reading and watching sci-fi, I was kind of a big dork which I pretty much still am. I haven't really changed a whole lot. Just basically seeing that machines really aren't all that different from biological entities, right? We are biological machines and kind of understanding how a computer works and how we engineer those things and trying to pull together concepts that learn from biology into that has always been a fascination of mine. As an undergrad, I was in the EE, CS world. Even then, I did some research projects around that. I worked in the industry for about 10 years designing chips, microprocessors, various kinds of ASICs, and then actually went back to school, quit my job, got a Ph.D. in neuroscience, computational neuroscience, to specifically understand what's the state of the art. What do we really understand about the brain? And are there concepts that we can take and bring back? Inspiration's always been we want to... We watch birds fly around. We want to figure out how to make something that flies. We extract those principles, and then build a plane. Don't necessarily want to build a bird. And so Nervana's really was the combination of all those experiences, bringing it together. Trying to push computation in a new a direction. Now, as part of Intel, we can really add a lot of fuel to that fire. I'm super excited to be part of Intel in that the technologies that we were developing can really proliferate and be applied to health care, can be applied to Internet, can be applied to every facet of our lives. And some of the examples that John mentioned are extremely exciting right now and these are things we can do today. And the generality of these solutions are just really going to hit every part of health care. I mean from a personal viewpoint, my whole family are MDs. I'm sort of the black sheep of the family. I don't have an MD. And it's always been kind of funny to me that knowledge is concentrated in a few individuals. Like you have a rare tumor or something like that, you need the guy who knows how to read this MRI. Why? Why is it like that? Can't we encapsulate that knowledge into a computer or into an algorithm, and democratize it. And the reason we couldn't do it is we just didn't know how. And now we're really getting to a point where we know how to do that. And so I want that capability to go to everybody. It'll bring the cost of healthcare down. It'll make all of us healthier. That affects everything about our society. So that's really what's exciting about it to me. >> That's great. So, as you heard, I'm Bob Rogers. I'm chief data scientist for analytics and artificial intelligence solutions at Intel. My mission is to put powerful analytics in the hands of every decision maker and when I think about Precision Medicine, decision makers are not just doctors and surgeons and nurses, but they're also case managers and care coordinators and probably most of all, patients. So the mission is really to put powerful analytics and AI capabilities in the hands of everyone in health care. It's a very complex world and we need tools to help us navigate it. So my background, I started with a Ph.D. in physics and I was computer modeling stuff, falling into super massive black holes. And there's a lot of applications for that in the real world. No, I'm kidding. (laughter) >> John: There will be, I'm sure. Yeah, one of these days. Soon as we have time travel. Okay so, I actually, about 1991, I was working on my post doctoral research, and I heard about neural networks, these things that could compute the way the brain computes. And so, I started doing some research on that. I wrote some papers and actually, it was an interesting story. The problem that we solved that got me really excited about neural networks, which have become deep learning, my office mate would come in. He was this young guy who was about to go off to grad school. He'd come in every morning. "I hate my project." Finally, after two weeks, what's your project? What's the problem? It turns out he had to circle these little fuzzy spots on these images from a telescope. So they were looking for the interesting things in a sky survey, and he had to circle them and write down their coordinates all summer. Anyone want to volunteer to do that? No? Yeah, he was very unhappy. So we took the first two weeks of data that he created doing his work by hand, and we trained an artificial neural network to do his summer project and finished it in about eight hours of computing. (crowd laughs) And so he was like yeah, this is amazing. I'm so happy. And we wrote a paper. I was the first author of course, because I was the senior guy at age 24. And he was second author. His first paper ever. He was very, very excited. So we have to fast forward about 20 years. His name popped up on the Internet. And so it caught my attention. He had just won the Nobel Prize in physics. (laughter) So that's where artificial intelligence will get you. (laughter) So thanks Naveen. Fast forwarding, I also developed some time series forecasting capabilities that allowed me to create a hedge fund that I ran for 12 years. After that, I got into health care, which really is the center of my passion. Applying health care to figuring out how to get all the data from all those siloed sources, put it into the cloud in a secure way, and analyze it so you can actually understand those cases that John was just talking about. How do you know that that person had had a splenectomy and that they needed to get that pneumovax? You need to be able to search all the data, so we used AI, natural language processing, machine learning, to do that and then two years ago, I was lucky enough to join Intel and, in the intervening time, people like Naveen actually thawed the AI winter and we're really in a spring of amazing opportunities with AI, not just in health care but everywhere, but of course, the health care applications are incredibly life saving and empowering so, excited to be here on this stage with you guys. >> I just want to cue off of your comment about the role of physics in AI and health care. So the field of microbiomics that I referred to earlier, bacteria in our gut. There's more bacteria in our gut than there are cells in our body. There's 100 times more DNA in that bacteria than there is in the human genome. And we're now discovering a couple hundred species of bacteria a year that have never been identified under a microscope just by their DNA. So it turns out the person who really catapulted the study and the science of microbiomics forward was an astrophysicist who did his Ph.D. in Steven Hawking's lab on the collision of black holes and then subsequently, put the other team in a virtual reality, and he developed the first super computing center and so how did he get an interest in microbiomics? He has the capacity to do high performance computing and the kind of advanced analytics that are required to look at a 100 times the volume of 3.2 billion base pairs of the human genome that are represented in the bacteria in our gut, and that has unleashed the whole science of microbiomics, which is going to really turn a lot of our assumptions of health and health care upside down. >> That's great, I mean, that's really transformational. So a lot of data. So I just wanted to let the audience know that we want to make this an interactive session, so I'll be asking for questions in a little bit, but I will start off with one question so that you can think about it. So I wanted to ask you, it looks like you've been thinking a lot about AI over the years. And I wanted to understand, even though AI's just really starting in health care, what are some of the new trends or the changes that you've seen in the last few years that'll impact how AI's being used going forward? >> So I'll start off. There was a paper published by a guy by the name of Tegmark at Harvard last summer that, for the first time, explained why neural networks are efficient beyond any mathematical model we predict. And the title of the paper's fun. It's called Deep Learning Versus Cheap Learning. So there were two sort of punchlines of the paper. One is is that the reason that mathematics doesn't explain the efficiency of neural networks is because there's a higher order of mathematics called physics. And the physics of the underlying data structures determined how efficient you could mine those data using machine learning tools. Much more so than any mathematical modeling. And so the second thing that was a reel from that paper is that the substrate of the data that you're operating on and the natural physics of those data have inherent levels of complexity that determine whether or not a 12th layer of neural net will get you where you want to go really fast, because when you do the modeling, for those math geeks in the audience, a factorial. So if there's 12 layers, there's 12 factorial permutations of different ways you could sequence the learning through those data. When you have 140 layers of a neural net, it's a much, much, much bigger number of permutations and so you end up being hardware-bound. And so, what Max Tegmark basically said is you can determine whether to do deep learning or cheap learning based upon the underlying physics of the data substrates you're operating on and have a good insight into how to optimize your hardware and software approach to that problem. >> So another way to put that is that neural networks represent the world in the way the world is sort of built. >> Exactly. >> It's kind of hierarchical. It's funny because, sort of in retrospect, like oh yeah, that kind of makes sense. But when you're thinking about it mathematically, we're like well, anything... The way a neural can represent any mathematical function, therfore, it's fully general. And that's the way we used to look at it, right? So now we're saying, well actually decomposing the world into different types of features that are layered upon each other is actually a much more efficient, compact representation of the world, right? I think this is actually, precisely the point of kind of what you're getting at. What's really exciting now is that what we were doing before was sort of building these bespoke solutions for different kinds of data. NLP, natural language processing. There's a whole field, 25 plus years of people devoted to figuring out features, figuring out what structures make sense in this particular context. Those didn't carry over at all to computer vision. Didn't carry over at all to time series analysis. Now, with neural networks, we've seen it at Nervana, and now part of Intel, solving customers' problems. We apply a very similar set of techniques across all these different types of data domains and solve them. All data in the real world seems to be hierarchical. You can decompose it into this hierarchy. And it works really well. Our brains are actually general structures. As a neuroscientist, you can look at different parts of your brain and there are differences. Something that takes in visual information, versus auditory information is slightly different but they're much more similar than they are different. So there is something invariant, something very common between all of these different modalities and we're starting to learn that. And this is extremely exciting to me trying to understand the biological machine that is a computer, right? We're figurig it out, right? >> One of the really fun things that Ray Chrisfall likes to talk about is, and it falls in the genre of biomimmicry, and how we actually replicate biologic evolution in our technical solutions so if you look at, and we're beginning to understand more and more how real neural nets work in our cerebral cortex. And it's sort of a pyramid structure so that the first pass of a broad base of analytics, it gets constrained to the next pass, gets constrained to the next pass, which is how information is processed in the brain. So we're discovering increasingly that what we've been evolving towards, in term of architectures of neural nets, is approximating the architecture of the human cortex and the more we understand the human cortex, the more insight we get to how to optimize neural nets, so when you think about it, with millions of years of evolution of how the cortex is structured, it shouldn't be a surprise that the optimization protocols, if you will, in our genetic code are profoundly efficient in how they operate. So there's a real role for looking at biologic evolutionary solutions, vis a vis technical solutions, and there's a friend of mine who worked with who worked with George Church at Harvard and actually published a book on biomimmicry and they wrote the book completely in DNA so if all of you have your home DNA decoder, you can actually read the book on your DNA reader, just kidding. >> There's actually a start up I just saw in the-- >> Read-Write DNA, yeah. >> Actually it's a... He writes something. What was it? (response from crowd member) Yeah, they're basically encoding information in DNA as a storage medium. (laughter) The company, right? >> Yeah, that same friend of mine who coauthored that biomimmicry book in DNA also did the estimate of the density of information storage. So a cubic centimeter of DNA can store an hexabyte of data. I mean that's mind blowing. >> Naveen: Highly done soon. >> Yeah that's amazing. Also you hit upon a really important point there, that one of the things that's changed is... Well, there are two major things that have changed in my perception from let's say five to 10 years ago, when we were using machine learning. You could use data to train models and make predictions to understand complex phenomena. But they had limited utility and the challenge was that if I'm trying to build on these things, I had to do a lot of work up front. It was called feature engineering. I had to do a lot of work to figure out what are the key attributes of that data? What are the 10 or 20 or 100 pieces of information that I should pull out of the data to feed to the model, and then the model can turn it into a predictive machine. And so, what's really exciting about the new generation of machine learning technology, and particularly deep learning, is that it can actually learn from example data those features without you having to do any preprogramming. That's why Naveen is saying you can take the same sort of overall approach and apply it to a bunch of different problems. Because you're not having to fine tune those features. So at the end of the day, the two things that have changed to really enable this evolution is access to more data, and I'd be curious to hear from you where you're seeing data come from, what are the strategies around that. So access to data, and I'm talking millions of examples. So 10,000 examples most times isn't going to cut it. But millions of examples will do it. And then, the other piece is the computing capability to actually take millions of examples and optimize this algorithm in a single lifetime. I mean, back in '91, when I started, we literally would have thousands of examples and it would take overnight to run the thing. So now in the world of millions, and you're putting together all of these combinations, the computing has changed a lot. I know you've made some revolutionary advances in that. But I'm curious about the data. Where are you seeing interesting sources of data for analytics? >> So I do some work in the genomics space and there are more viable permutations of the human genome than there are people who have ever walked the face of the earth. And the polygenic determination of a phenotypic expression translation, what are genome does to us in our physical experience in health and disease is determined by many, many genes and the interaction of many, many genes and how they are up and down regulated. And the complexity of disambiguating which 27 genes are affecting your diabetes and how are they up and down regulated by different interventions is going to be different than his. It's going to be different than his. And we already know that there's four or five distinct genetic subtypes of type II diabetes. So physicians still think there's one disease called type II diabetes. There's actually at least four or five genetic variants that have been identified. And so, when you start thinking about disambiguating, particularly when we don't know what 95 percent of DNA does still, what actually is the underlining cause, it will require this massive capability of developing these feature vectors, sometimes intuiting it, if you will, from the data itself. And other times, taking what's known knowledge to develop some of those feature vectors, and be able to really understand the interaction of the genome and the microbiome and the phenotypic data. So the complexity is high and because the variation complexity is high, you do need these massive members. Now I'm going to make a very personal pitch here. So forgive me, but if any of you have any role in policy at all, let me tell you what's happening right now. The Genomic Information Nondiscrimination Act, so called GINA, written by a friend of mine, passed a number of years ago, says that no one can be discriminated against for health insurance based upon their genomic information. That's cool. That should allow all of you to feel comfortable donating your DNA to science right? Wrong. You are 100% unprotected from discrimination for life insurance, long term care and disability. And it's being practiced legally today and there's legislation in the House, in mark up right now to completely undermine the existing GINA legislation and say that whenever there's another applicable statute like HIPAA, that the GINA is irrelevant, that none of the fines and penalties are applicable at all. So we need a ton of data to be able to operate on. We will not be getting a ton of data to operate on until we have the kind of protection we need to tell people, you can trust us. You can give us your data, you will not be subject to discrimination. And that is not the case today. And it's being further undermined. So I want to make a plea to any of you that have any policy influence to go after that because we need this data to help the understanding of human health and disease and we're not going to get it when people look behind the curtain and see that discrimination is occurring today based upon genetic information. >> Well, I don't like the idea of being discriminated against based on my DNA. Especially given how little we actually know. There's so much complexity in how these things unfold in our own bodies, that I think anything that's being done is probably childishly immature and oversimplifying. So it's pretty rough. >> I guess the translation here is that we're all unique. It's not just a Disney movie. (laughter) We really are. And I think one of the strengths that I'm seeing, kind of going back to the original point, of these new techniques is it's going across different data types. It will actually allow us to learn more about the uniqueness of the individual. It's not going to be just from one data source. They were collecting data from many different modalities. We're collecting behavioral data from wearables. We're collecting things from scans, from blood tests, from genome, from many different sources. The ability to integrate those into a unified picture, that's the important thing that we're getting toward now. That's what I think is going to be super exciting here. Think about it, right. I can tell you to visual a coin, right? You can visualize a coin. Not only do you visualize it. You also know what it feels like. You know how heavy it is. You have a mental model of that from many different perspectives. And if I take away one of those senses, you can still identify the coin, right? If I tell you to put your hand in your pocket, and pick out a coin, you probably can do that with 100% reliability. And that's because we have this generalized capability to build a model of something in the world. And that's what we need to do for individuals is actually take all these different data sources and come up with a model for an individual and you can actually then say what drug works best on this. What treatment works best on this? It's going to get better with time. It's not going to be perfect, because this is what a doctor does, right? A doctor who's very experienced, you're a practicing physician right? Back me up here. That's what you're doing. You basically have some categories. You're taking information from the patient when you talk with them, and you're building a mental model. And you apply what you know can work on that patient, right? >> I don't have clinic hours anymore, but I do take care of many friends and family. (laughter) >> You used to, you used to. >> I practiced for many years before I became a full-time geek. >> I thought you were a recovering geek. >> I am. (laughter) I do more policy now. >> He's off the wagon. >> I just want to take a moment and see if there's anyone from the audience who would like to ask, oh. Go ahead. >> We've got a mic here, hang on one second. >> I have tons and tons of questions. (crosstalk) Yes, so first of all, the microbiome and the genome are really complex. You already hit about that. Yet most of the studies we do are small scale and we have difficulty repeating them from study to study. How are we going to reconcile all that and what are some of the technical hurdles to get to the vision that you want? >> So primarily, it's been the cost of sequencing. Up until a year ago, it's $1000, true cost. Now it's $100, true cost. And so that barrier is going to enable fairly pervasive testing. It's not a real competitive market becaue there's one sequencer that is way ahead of everybody else. So the price is not $100 yet. The cost is below $100. So as soon as there's competition to drive the cost down, and hopefully, as soon as we all have the protection we need against discrimination, as I mentioned earlier, then we will have large enough sample sizes. And so, it is our expectation that we will be able to pool data from local sources. I chair the e-health work group at the Global Alliance for Genomics and Health which is working on this very issue. And rather than pooling all the data into a single, common repository, the strategy, and we're developing our five-year plan in a month in London, but the goal is to have a federation of essentially credentialed data enclaves. That's a formal method. HHS already does that so you can get credentialed to search all the data that Medicare has on people that's been deidentified according to HIPPA. So we want to provide the same kind of service with appropriate consent, at an international scale. And there's a lot of nations that are talking very much about data nationality so that you can't export data. So this approach of a federated model to get at data from all the countries is important. The other thing is a block-chain technology is going to be very profoundly useful in this context. So David Haussler of UC Santa Cruz is right now working on a protocol using an open block-chain, public ledger, where you can put out. So for any typical cancer, you may have a half dozen, what are called sematic variance. Cancer is a genetic disease so what has mutated to cause it to behave like a cancer? And if we look at those biologically active sematic variants, publish them on a block chain that's public, so there's not enough data there to reidentify the patient. But if I'm a physician treating a woman with breast cancer, rather than say what's the protocol for treating a 50-year-old woman with this cell type of cancer, I can say show me all the people in the world who have had this cancer at the age of 50, wit these exact six sematic variants. Find the 200 people worldwide with that. Ask them for consent through a secondary mechanism to donate everything about their medical record, pool that information of the core of 200 that exactly resembles the one sitting in front of me, and find out, of the 200 ways they were treated, what got the best results. And so, that's the kind of future where a distributed, federated architecture will allow us to query and obtain a very, very relevant cohort, so we can basically be treating patients like mine, sitting right in front of me. Same thing applies for establishing research cohorts. There's some very exciting stuff at the convergence of big data analytics, machine learning, and block chaining. >> And this is an area that I'm really excited about and I think we're excited about generally at Intel. They actually have something called the Collaborative Cancer Cloud, which is this kind of federated model. We have three different academic research centers. Each of them has a very sizable and valuable collection of genomic data with phenotypic annotations. So you know, pancreatic cancer, colon cancer, et cetera, and we've actually built a secure computing architecture that can allow a person who's given the right permissions by those organizations to ask a specific question of specific data without ever sharing the data. So the idea is my data's really important to me. It's valuable. I want us to be able to do a study that gets the number from the 20 pancreatic cancer patients in my cohort, up to the 80 that we have in the whole group. But I can't do that if I'm going to just spill my data all over the world. And there are HIPAA and compliance reasons for that. There are business reasons for that. So what we've built at Intel is this platform that allows you to do different kinds of queries on this genetic data. And reach out to these different sources without sharing it. And then, the work that I'm really involved in right now and that I'm extremely excited about... This also touches on something that both of you said is it's not sufficient to just get the genome sequences. You also have to have the phenotypic data. You have to know what cancer they've had. You have to know that they've been treated with this drug and they've survived for three months or that they had this side effect. That clinical data also needs to be put together. It's owned by other organizations, right? Other hospitals. So the broader generalization of the Collaborative Cancer Cloud is something we call the data exchange. And it's a misnomer in a sense that we're not actually exchanging data. We're doing analytics on aggregated data sets without sharing it. But it really opens up a world where we can have huge populations and big enough amounts of data to actually train these models and draw the thread in. Of course, that really then hits home for the techniques that Nervana is bringing to the table, and of course-- >> Stanford's one of your academic medical centers? >> Not for that Collaborative Cancer Cloud. >> The reason I mentioned Standford is because the reason I'm wearing this FitBit is because I'm a research subject at Mike Snyder's, the chair of genetics at Stanford, IPOP, intrapersonal omics profile. So I was fully sequenced five years ago and I get four full microbiomes. My gut, my mouth, my nose, my ears. Every three months and I've done that for four years now. And about a pint of blood. And so, to your question of the density of data, so a lot of the problem with applying these techniques to health care data is that it's basically a sparse matrix and there's a lot of discontinuities in what you can find and operate on. So what Mike is doing with the IPOP study is much the same as you described. Creating a highly dense longitudinal set of data that will help us mitigate the sparse matrix problem. (low volume response from audience member) Pardon me. >> What's that? (low volume response) (laughter) >> Right, okay. >> John: Lost the school sample. That's got to be a new one I've heard now. >> Okay, well, thank you so much. That was a great question. So I'm going to repeat this and ask if there's another question. You want to go ahead? >> Hi, thanks. So I'm a journalist and I report a lot on these neural networks, a system that's beter at reading mammograms than your human radiologists. Or a system that's better at predicting which patients in the ICU will get sepsis. These sort of fascinating academic studies that I don't really see being translated very quickly into actual hospitals or clinical practice. Seems like a lot of the problems are regulatory, or liability, or human factors, but how do you get past that and really make this stuff practical? >> I think there's a few things that we can do there and I think the proof points of the technology are really important to start with in this specific space. In other places, sometimes, you can start with other things. But here, there's a real confidence problem when it comes to health care, and for good reason. We have doctors trained for many, many years. School and then residencies and other kinds of training. Because we are really, really conservative with health care. So we need to make sure that technology's well beyond just the paper, right? These papers are proof points. They get people interested. They even fuel entire grant cycles sometimes. And that's what we need to happen. It's just an inherent problem, its' going to take a while. To get those things to a point where it's like well, I really do trust what this is saying. And I really think it's okay to now start integrating that into our standard of care. I think that's where you're seeing it. It's frustrating for all of us, believe me. I mean, like I said, I think personally one of the biggest things, I want to have an impact. Like when I go to my grave, is that we used machine learning to improve health care. We really do feel that way. But it's just not something we can do very quickly and as a business person, I don't actually look at those use cases right away because I know the cycle is just going to be longer. >> So to your point, the FDA, for about four years now, has understood that the process that has been given to them by their board of directors, otherwise known as Congress, is broken. And so they've been very actively seeking new models of regulation and what's really forcing their hand is regulation of devices and software because, in many cases, there are black box aspects of that and there's a black box aspect to machine learning. Historically, Intel and others are making inroads into providing some sort of traceability and transparency into what happens in that black box rather than say, overall we get better results but once in a while we kill somebody. Right? So there is progress being made on that front. And there's a concept that I like to use. Everyone knows Ray Kurzweil's book The Singularity Is Near? Well, I like to think that diadarity is near. And the diadarity is where you have human transparency into what goes on in the black box and so maybe Bob, you want to speak a little bit about... You mentioned that, in a prior discussion, that there's some work going on at Intel there. >> Yeah, absolutely. So we're working with a number of groups to really build tools that allow us... In fact Naveen probably can talk in even more detail than I can, but there are tools that allow us to actually interrogate machine learning and deep learning systems to understand, not only how they respond to a wide variety of situations but also where are there biases? I mean, one of the things that's shocking is that if you look at the clinical studies that our drug safety rules are based on, 50 year old white guys are the peak of that distribution, which I don't see any problem with that, but some of you out there might not like that if you're taking a drug. So yeah, we want to understand what are the biases in the data, right? And so, there's some new technologies. There's actually some very interesting data-generative technologies. And this is something I'm also curious what Naveen has to say about, that you can generate from small sets of observed data, much broader sets of varied data that help probe and fill in your training for some of these systems that are very data dependent. So that takes us to a place where we're going to start to see deep learning systems generating data to train other deep learning systems. And they start to sort of go back and forth and you start to have some very nice ways to, at least, expose the weakness of these underlying technologies. >> And that feeds back to your question about regulatory oversight of this. And there's the fascinating, but little known origin of why very few women are in clinical studies. Thalidomide causes birth defects. So rather than say pregnant women can't be enrolled in drug trials, they said any woman who is at risk of getting pregnant cannot be enrolled. So there was actually a scientific meritorious argument back in the day when they really didn't know what was going to happen post-thalidomide. So it turns out that the adverse, unintended consequence of that decision was we don't have data on women and we know in certain drugs, like Xanax, that the metabolism is so much slower, that the typical dosing of Xanax is women should be less than half of that for men. And a lot of women have had very serious adverse effects by virtue of the fact that they weren't studied. So the point I want to illustrate with that is that regulatory cycles... So people have known for a long time that was like a bad way of doing regulations. It should be changed. It's only recently getting changed in any meaningful way. So regulatory cycles and legislative cycles are incredibly slow. The rate of exponential growth in technology is exponential. And so there's impedance mismatch between the cycle time for regulation cycle time for innovation. And what we need to do... I'm working with the FDA. I've done four workshops with them on this very issue. Is that they recognize that they need to completely revitalize their process. They're very interested in doing it. They're not resisting it. People think, oh, they're bad, the FDA, they're resisting. Trust me, there's nobody on the planet who wants to revise these review processes more than the FDA itself. And so they're looking at models and what I recommended is global cloud sourcing and the FDA could shift from a regulatory role to one of doing two things, assuring the people who do their reviews are competent, and assuring that their conflicts of interest are managed, because if you don't have a conflict of interest in this very interconnected space, you probably don't know enough to be a reviewer. So there has to be a way to manage the conflict of interest and I think those are some of the keypoints that the FDA is wrestling with because there's type one and type two errors. If you underregulate, you end up with another thalidomide and people born without fingers. If you overregulate, you prevent life saving drugs from coming to market. So striking that balance across all these different technologies is extraordinarily difficult. If it were easy, the FDA would've done it four years ago. It's very complicated. >> Jumping on that question, so all three of you are in some ways entrepreneurs, right? Within your organization or started companies. And I think it would be good to talk a little bit about the business opportunity here, where there's a huge ecosystem in health care, different segments, biotech, pharma, insurance payers, etc. Where do you see is the ripe opportunity or industry, ready to really take this on and to make AI the competitive advantage. >> Well, the last question also included why aren't you using the result of the sepsis detection? We do. There were six or seven published ways of doing it. We did our own data, looked at it, we found a way that was superior to all the published methods and we apply that today, so we are actually using that technology to change clinical outcomes. As far as where the opportunities are... So it's interesting. Because if you look at what's going to be here in three years, we're not going to be using those big data analytics models for sepsis that we are deploying today, because we're just going to be getting a tiny aliquot of blood, looking for the DNA or RNA of any potential infection and we won't have to infer that there's a bacterial infection from all these other ancillary, secondary phenomenon. We'll see if the DNA's in the blood. So things are changing so fast that the opportunities that people need to look for are what are generalizable and sustainable kind of wins that are going to lead to a revenue cycle that are justified, a venture capital world investing. So there's a lot of interesting opportunities in the space. But I think some of the biggest opportunities relate to what Bob has talked about in bringing many different disparate data sources together and really looking for things that are not comprehensible in the human brain or in traditional analytic models. >> I think we also got to look a little bit beyond direct care. We're talking about policy and how we set up standards, these kinds of things. That's one area. That's going to drive innovation forward. I completely agree with that. Direct care is one piece. How do we scale out many of the knowledge kinds of things that are embedded into one person's head and get them out to the world, democratize that. Then there's also development. The underlying technology's of medicine, right? Pharmaceuticals. The traditional way that pharmaceuticals is developed is actually kind of funny, right? A lot of it was started just by chance. Penicillin, a very famous story right? It's not that different today unfortunately, right? It's conceptually very similar. Now we've got more science behind it. We talk about domains and interactions, these kinds of things but fundamentally, the problem is what we in computer science called NP hard, it's too difficult to model. You can't solve it analytically. And this is true for all these kinds of natural sorts of problems by the way. And so there's a whole field around this, molecular dynamics and modeling these sorts of things, that are actually being driven forward by these AI techniques. Because it turns out, our brain doesn't do magic. It actually doesn't solve these problems. It approximates them very well. And experience allows you to approximate them better and better. Actually, it goes a little bit to what you were saying before. It's like simulations and forming your own networks and training off each other. There are these emerging dynamics. You can simulate steps of physics. And you come up with a system that's much too complicated to ever solve. Three pool balls on a table is one such system. It seems pretty simple. You know how to model that, but it actual turns out you can't predict where a balls going to be once you inject some energy into that table. So something that simple is already too complex. So neural network techniques actually allow us to start making those tractable. These NP hard problems. And things like molecular dynamics and actually understanding how different medications and genetics will interact with each other is something we're seeing today. And so I think there's a huge opportunity there. We've actually worked with customers in this space. And I'm seeing it. Like Rosch is acquiring a few different companies in space. They really want to drive it forward, using big data to drive drug development. It's kind of counterintuitive. I never would've thought it had I not seen it myself. >> And there's a big related challenge. Because in personalized medicine, there's smaller and smaller cohorts of people who will benefit from a drug that still takes two billion dollars on average to develop. That is unsustainable. So there's an economic imperative of overcoming the cost and the cycle time for drug development. >> I want to take a go at this question a little bit differently, thinking about not so much where are the industry segments that can benefit from AI, but what are the kinds of applications that I think are most impactful. So if this is what a skilled surgeon needs to know at a particular time to care properly for a patient, this is where most, this area here, is where most surgeons are. They are close to the maximum knowledge and ability to assimilate as they can be. So it's possible to build complex AI that can pick up on that one little thing and move them up to here. But it's not a gigantic accelerator, amplifier of their capability. But think about other actors in health care. I mentioned a couple of them earlier. Who do you think the least trained actor in health care is? >> John: Patients. >> Yes, the patients. The patients are really very poorly trained, including me. I'm abysmal at figuring out who to call and where to go. >> Naveen: You know as much the doctor right? (laughing) >> Yeah, that's right. >> My doctor friends always hate that. Know your diagnosis, right? >> Yeah, Dr. Google knows. So the opportunities that I see that are really, really exciting are when you take an AI agent, like sometimes I like to call it contextually intelligent agent, or a CIA, and apply it to a problem where a patient has a complex future ahead of them that they need help navigating. And you use the AI to help them work through. Post operative. You've got PT. You've got drugs. You've got to be looking for side effects. An agent can actually help you navigate. It's like your own personal GPS for health care. So it's giving you the inforamation that you need about you for your care. That's my definition of Precision Medicine. And it can include genomics, of course. But it's much bigger. It's that broader picture and I think that a sort of agent way of thinking about things and filling in the gaps where there's less training and more opportunity, is very exciting. >> Great start up idea right there by the way. >> Oh yes, right. We'll meet you all out back for the next start up. >> I had a conversation with the head of the American Association of Medical Specialties just a couple of days ago. And what she was saying, and I'm aware of this phenomenon, but all of the medical specialists are saying, you're killing us with these stupid board recertification trivia tests that you're giving us. So if you're a cardiologist, you have to remember something that happens in one in 10 million people, right? And they're saying that irrelevant anymore, because we've got advanced decision support coming. We have these kinds of analytics coming. Precisely what you're saying. So it's human augmentation of decision support that is coming at blazing speed towards health care. So in that context, it's much more important that you have a basic foundation, you know how to think, you know how to learn, and you know where to look. So we're going to be human-augmented learning systems much more so than in the past. And so the whole recertification process is being revised right now. (inaudible audience member speaking) Speak up, yeah. (person speaking) >> What makes it fathomable is that you can-- (audience member interjects inaudibly) >> Sure. She was saying that our brain is really complex and large and even our brains don't know how our brains work, so... are there ways to-- >> What hope do we have kind of thing? (laughter) >> It's a metaphysical question. >> It circles all the way down, exactly. It's a great quote. I mean basically, you can decompose every system. Every complicated system can be decomposed into simpler, emergent properties. You lose something perhaps with each of those, but you get enough to actually understand most of the behavior. And that's really how we understand the world. And that's what we've learned in the last few years what neural network techniques can allow us to do. And that's why our brain can understand our brain. (laughing) >> Yeah, I'd recommend reading Chris Farley's last book because he addresses that issue in there very elegantly. >> Yeah we're seeing some really interesting technologies emerging right now where neural network systems are actually connecting other neural network systems in networks. You can see some very compelling behavior because one of the things I like to distinguish AI versus traditional analytics is we used to have question-answering systems. I used to query a database and create a report to find out how many widgets I sold. Then I started using regression or machine learning to classify complex situations from this is one of these and that's one of those. And then as we've moved more recently, we've got these AI-like capabilities like being able to recognize that there's a kitty in the photograph. But if you think about it, if I were to show you a photograph that happened to have a cat in it, and I said, what's the answer, you'd look at me like, what are you talking about? I have to know the question. So where we're cresting with these connected sets of neural systems, and with AI in general, is that the systems are starting to be able to, from the context, understand what the question is. Why would I be asking about this picture? I'm a marketing guy, and I'm curious about what Legos are in the thing or what kind of cat it is. So it's being able to ask a question, and then take these question-answering systems, and actually apply them so that's this ability to understand context and ask questions that we're starting to see emerge from these more complex hierarchical neural systems. >> There's a person dying to ask a question. >> Sorry. You have hit on several different topics that all coalesce together. You mentioned personalized models. You mentioned AI agents that could help you as you're going through a transitionary period. You mentioned data sources, especially across long time periods. Who today has access to enough data to make meaningful progress on that, not just when you're dealing with an issue, but day-to-day improvement of your life and your health? >> Go ahead, great question. >> That was a great question. And I don't think we have a good answer to it. (laughter) I'm sure John does. Well, I think every large healthcare organization and various healthcare consortiums are working very hard to achieve that goal. The problem remains in creating semantic interoperatability. So I spent a lot of my career working on semantic interoperatability. And the problem is that if you don't have well-defined, or self-defined data, and if you don't have well-defined and documented metadata, and you start operating on it, it's real easy to reach false conclusions and I can give you a classic example. It's well known, with hundreds of studies looking at when you give an antibiotic before surgery and how effective it is in preventing a post-op infection. Simple question, right? So most of the literature done prosectively was done in institutions where they had small sample sizes. So if you pool that, you get a little bit more noise, but you get a more confirming answer. What was done at a very large, not my own, but a very large institution... I won't name them for obvious reasons, but they pooled lots of data from lots of different hospitals, where the data definitions and the metadata were different. Two examples. When did they indicate the antibiotic was given? Was it when it was ordered, dispensed from the pharmacy, delivered to the floor, brought to the bedside, put in the IV, or the IV starts flowing? Different hospitals used a different metric of when it started. When did surgery occur? When they were wheeled into the OR, when they were prepped and drapped, when the first incision occurred? All different. And they concluded quite dramatically that it didn't matter when you gave the pre-op antibiotic and whether or not you get a post-op infection. And everybody who was intimate with the prior studies just completely ignored and discounted that study. It was wrong. And it was wrong because of the lack of commonality and the normalization of data definitions and metadata definitions. So because of that, this problem is much more challenging than you would think. If it were so easy as to put all these data together and operate on it, normalize and operate on it, we would've done that a long time ago. It's... Semantic interoperatability remains a big problem and we have a lot of heavy lifting ahead of us. I'm working with the Global Alliance, for example, of Genomics and Health. There's like 30 different major ontologies for how you represent genetic information. And different institutions are using different ones in different ways in different versions over different periods of time. That's a mess. >> Our all those issues applicable when you're talking about a personalized data set versus a population? >> Well, so N of 1 studies and single-subject research is an emerging field of statistics. So there's some really interesting new models like step wedge analytics for doing that on small sample sizes, recruiting people asynchronously. There's single-subject research statistics. You compare yourself with yourself at a different point in time, in a different context. So there are emerging statistics to do that and as long as you use the same sensor, you won't have a problem. But people are changing their remote sensors and you're getting different data. It's measured in different ways with different sensors at different normalization and different calibration. So yes. It even persists in the N of 1 environment. >> Yeah, you have to get started with a large N that you can apply to the N of 1. I'm actually going to attack your question from a different perspective. So who has the data? The millions of examples to train a deep learning system from scratch. It's a very limited set right now. Technology such as the Collaborative Cancer Cloud and The Data Exchange are definitely impacting that and creating larger and larger sets of critical mass. And again, not withstanding the very challenging semantic interoperability questions. But there's another opportunity Kay asked about what's changed recently. One of the things that's changed in deep learning is that we now have modules that have been trained on massive data sets that are actually very smart as certain kinds of problems. So, for instance, you can go online and find deep learning systems that actually can recognize, better than humans, whether there's a cat, dog, motorcycle, house, in a photograph. >> From Intel, open source. >> Yes, from Intel, open source. So here's what happens next. Because most of that deep learning system is very expressive. That combinatorial mixture of features that Naveen was talking about, when you have all these layers, there's a lot of features there. They're actually very general to images, not just finding cats, dogs, trees. So what happens is you can do something called transfer learning, where you take a small or modest data set and actually reoptimize it for your specific problem very, very quickly. And so we're starting to see a place where you can... On one end of the spectrum, we're getting access to the computing capabilities and the data to build these incredibly expressive deep learning systems. And over here on the right, we're able to start using those deep learning systems to solve custom versions of problems. Just last weekend or two weekends ago, in 20 minutes, I was able to take one of those general systems and create one that could recognize all different kinds of flowers. Very subtle distinctions, that I would never be able to know on my own. But I happen to be able to get the data set and literally, it took 20 minutes and I have this vision system that I could now use for a specific problem. I think that's incredibly profound and I think we're going to see this spectrum of wherever you are in your ability to get data and to define problems and to put hardware in place to see really neat customizations and a proliferation of applications of this kind of technology. >> So one other trend I think, I'm very hopeful about it... So this is a hard problem clearly, right? I mean, getting data together, formatting it from many different sources, it's one of these things that's probably never going to happen perfectly. But one trend I think that is extremely hopeful to me is the fact that the cost of gathering data has precipitously dropped. Building that thing is almost free these days. I can write software and put it on 100 million cell phones in an instance. You couldn't do that five years ago even right? And so, the amount of information we can gain from a cell phone today has gone up. We have more sensors. We're bringing online more sensors. People have Apple Watches and they're sending blood data back to the phone, so once we can actually start gathering more data and do it cheaper and cheaper, it actually doesn't matter where the data is. I can write my own app. I can gather that data and I can start driving the correct inferences or useful inferences back to you. So that is a positive trend I think here and personally, I think that's how we're going to solve it, is by gathering from that many different sources cheaply. >> Hi, my name is Pete. I've very much enjoyed the conversation so far but I was hoping perhaps to bring a little bit more focus into Precision Medicine and ask two questions. Number one, how have you applied the AI technologies as you're emerging so rapidly to your natural language processing? I'm particularly interested in, if you look at things like Amazon Echo or Siri, or the other voice recognition systems that are based on AI, they've just become incredibly accurate and I'm interested in specifics about how I might use technology like that in medicine. So where would I find a medical nomenclature and perhaps some reference to a back end that works that way? And the second thing is, what specifically is Intel doing, or making available? You mentioned some open source stuff on cats and dogs and stuff but I'm the doc, so I'm looking at the medical side of that. What are you guys providing that would allow us who are kind of geeks on the software side, as well as being docs, to experiment a little bit more thoroughly with AI technology? Google has a free AI toolkit. Several other people have come out with free AI toolkits in order to accelerate that. There's special hardware now with graphics, and different processors, hitting amazing speeds. And so I was wondering, where do I go in Intel to find some of those tools and perhaps learn a bit about the fantastic work that you guys are already doing at Kaiser? >> Let me take that first part and then we'll be able to talk about the MD part. So in terms of technology, this is what's extremely exciting now about what Intel is focusing on. We're providing those pieces. So you can actually assemble and build the application. How you build that application specific for MDs and the use cases is up to you or the one who's filling out the application. But we're going to power that technology for multiple perspectives. So Intel is already the main force behind The Data Center, right? Cloud computing, all this is already Intel. We're making that extremely amenable to AI and setting the standard for AI in the future, so we can do that from a number of different mechanisms. For somebody who wants to develop an application quickly, we have hosted solutions. Intel Nervana is kind of the brand for these kinds of things. Hosted solutions will get you going very quickly. Once you get to a certain level of scale, where costs start making more sense, things can be bought on premise. We're supplying that. We're also supplying software that makes that transition essentially free. Then taking those solutions that you develop in the cloud, or develop in The Data Center, and actually deploying them on device. You want to write something on your smartphone or PC or whatever. We're actually providing those hooks as well, so we want to make it very easy for developers to take these pieces and actually build solutions out of them quickly so you probably don't even care what hardware it's running on. You're like here's my data set, this is what I want to do. Train it, make it work. Go fast. Make my developers efficient. That's all you care about, right? And that's what we're doing. We're taking it from that point at how do we best do that? We're going to provide those technologies. In the next couple of years, there's going to be a lot of new stuff coming from Intel. >> Do you want to talk about AI Academy as well? >> Yeah, that's a great segway there. In addition to this, we have an entire set of tutorials and other online resources and things we're going to be bringing into the academic world for people to get going quickly. So that's not just enabling them on our tools, but also just general concepts. What is a neural network? How does it work? How does it train? All of these things are available now and we've made a nice, digestible class format that you can actually go and play with. >> Let me give a couple of quick answers in addition to the great answers already. So you're asking why can't we use medical terminology and do what Alexa does? Well, no, you may not be aware of this, but Andrew Ian, who was the AI guy at Google, who was recruited by Google, they have a medical chat bot in China today. I don't speak Chinese. I haven't been able to use it yet. There are two similar initiatives in this country that I know of. There's probably a dozen more in stealth mode. But Lumiata and Health Cap are doing chat bots for health care today, using medical terminology. You have the compound problem of semantic normalization within language, compounded by a cross language. I've done a lot of work with an international organization called Snowmed, which translates medical terminology. So you're aware of that. We can talk offline if you want, because I'm pretty deep into the semantic space. >> Go google Intel Nervana and you'll see all the websites there. It's intel.com/ai or nervanasys.com. >> Okay, great. Well this has been fantastic. I want to, first of all, thank all the people here for coming and asking great questions. I also want to thank our fantastic panelists today. (applause) >> Thanks, everyone. >> Thank you. >> And lastly, I just want to share one bit of information. We will have more discussions on AI next Tuesday at 9:30 AM. Diane Bryant, who is our general manager of Data Centers Group will be here to do a keynote. So I hope you all get to join that. Thanks for coming. (applause) (light electronic music)