Will We See You at This Year's HLTH Conference? Save $150 with code: OUTCOMESROCKETPODCAST150 Use Code Here

 

 

How to Reduce the Burden of Disease with Machine Learning and Mathematics with Dr. Iya Khalil, Co-founder and Chief Commercial Officer at GNS Healthcare

Episode 111

Outcomes Rocket Podcast - Iya Khalil

How to Reduce the Burden of Disease with Machine Learning and Mathematics with Iya Khalil, Co-founder and Chief Commercial Officer at GNS Healthcare

: [00:00:01] Welcome to the Outcomes Rocket podcast where we inspire collaborative thinking, improved outcomes and business success with today’s most successful and inspiring healthcare leaders and influencers. And now your host, Saul Marquez

Saul Marquez: [00:00:19] Welcome back once again to the outcomes rocket podcast where we chat with today’s most successful and inspiring health leaders. I really thank you for tuning in and I welcome you to go to outcomesrocket.health/reviews where you could rate and review today’s outstanding guest because she is a huge contributor to Health Care and an amazing person. Her name is Dr. Iya Khalil. She is the co-founder and chief commercial officer at GNS healthcare. Dr. Khalil is a technology entrepreneur and physicist with a vision of transforming medicine into a discipline that is quantitative predictive and patient centric via big data analytic approaches. She cofounded two big data companies via science and GNS health care and is a Cohen Ventor of the proprietary computational engine that underpins both entities. She trained in theoretical physics at Cornell University and has more than 11 years of experience in big data analytics for healthcare medicine and life sciences. She has led several key foundational collaborations with providers pharmaceutical companies foundations and government agencies. Dr. Kalil’s experience spans applications and drug discovery drug development all the way to treatment algorithms that could be applied at the point of care. She’s a frequent speaker at industry events and conferences which is where I met Aiya and she’s also appeared in several industry journals published several articles in the field and was recognized by President Obama himself at the White House dinner as a leading entrepreneur and genomic medicine. More recently she was named to the farm of voice 100 list of most inspiring people in the life sciences industry. She was recognized for her ability to build bridges across the life science and health care industry which is much needed bringing people together to harness the power of predictive modeling to change the lives of patients and that’s exactly why we have her here today on the outcomes rocket. So I want to welcome you to the podcast Dr. Khalil.

Iya Khalil: [00:02:23] Thank you and thank you for having me here today.

Saul Marquez: [00:02:26] It’s a pleasure. And so I wanted to ask you just to kick things off. Why did you decide to get into the health care sector and why AI.

Iya Khalil: [00:02:34] Yes. So my interest in this field started in 2000 right about the time that I finished my Ph.D. in theoretical physics from Cornell but it was also the time that we had announced the sequencing of the human genome which is a huge monumental feat that involves both private and academic efforts to the NIH to get it done. But it was also the moment at which you know for the first time we could actually measure every base pair and our genes. We weren’t just looking at one gene at a time but we could look at everything and this really paralleled sort of how I thought about systems as a physicist which is you go in and try to measure and get as much data as you can as on that system and then use modeling and algorithms and computation which is really what AI is to help you model it and make predictions. And it became clear that could we use the same powerful approaches that helped us predict the Higgs boson many decades before we discovered it or theorize about gravitational waves which we have that recently measured. Could we use this in healthcare and make it more predictive. You know if we could send a rocket to the moon with technologies around data and computational and I why can’t we do it and use it to make precision medicine of medicine even better.

Saul Marquez: [00:03:52] I think that’s so great. Dr. Khalil it’s one of those things where it’s you ask the question and great leaders like yourself ask the question why not not why. And because of your thinking it’s led you to create this amazing company and you guys are doing some great things today. There’s a lot of buzz in healthcare around artificial intelligence and machine learning. How do you think it’s going to change the way health care is being delivered.

Iya Khalil: [00:04:16] There is a lot of buzz and it’s causing a lot of confusion and now that doesn’t mean that because there’s confusion there are real things happening. You know a lot of progress is being made on being able to leverage data of all kinds in our space whether it’s data coming from a text from physicians notes to being able to measure and sequence your genome. But there’s a lot of definitions out there and it’s good to clarify and so really what I’m focused on is using AI and its most powerful form and I’m using a form of ADD that I like to call causal machine learning and it’s really transforming millions of data points that we can now mashers from the data that we can collect which we have to recognize. We weren’t able to collect this data really well two decades ago. We can now we can go into a patient and measure their genomes and look at long attitudinal detail the electronic medical records labs claims and then leverage that using causal machine learning to learn the causal relationships underlying that data and really learn what’s behind that data. Weekend In Silico start to run simulations and ask very specific what if questions down to that patient down to what’s going to enable us to treat that patient accurately and precision what the right treatment is. And that opens up all sorts of doors in the industry and in places where they’re currently sort of bottlenecks you know how do we better stratify patients for clinical trials so that we’re giving the right drug to the right patient and we can better and more quickly get drugs through the approval process. How do we help health plans do a better job of making sure that the right interventions are getting to their members at not just the best quality but also you know optimizing across cost and optimizing dags with that individual helping providers do a better job of delivering care. So this ability to predict and learn from data it’s really really powerful. This is what is at the heart of AI and machine learning is our ability to learn. And now we want to take that and apply it down to the individual patient to get better at optimizing their treatments and their health.

Saul Marquez: [00:06:20] That’s fascinating Dr. Khalil and this causal artificial intelligence that’s really helping your company deliver value not only to pharma but it sounds like also payers as well as providers in a big way.

Iya Khalil: [00:06:35] Yes. It cuts across everything that we do there and much in the same way that algorithms machine learning AI cuts across everything that an engineer might be doing right. This might be doing it at the heart of how you are able to come up with insights and help her. Our goal is to take those insights now and have it impact the patient.

Saul Marquez: [00:06:56] That’s wonderful. So there’s no doubt you’re adding tons of value today in five to 10 years. Let’s get into the future what do you think health care looks like and how does artificial intelligence change how medicine is practiced.

Iya Khalil: [00:07:10] Yeah really great question. In five to 10 years I’m hoping that we’re at a place where in terms of the kinds of interventions and treatments we can get to patients we now have just a much bigger plethora of options right gene therapy is coming on the right stem cell therapy so we’re going to have all sorts of ways of treating patients that go beyond just traditional drugs. And then the question’s going to be what is the right treatment for that individual. And right now today you know for me as an AI machine learning expert to get a hold of that data that allows me to create the algorithms and learn and make that prediction. There’s just a limitation. Sometimes I can’t get all the data that I need. Griz also limitations around sharing of data and the data is siloed. So I’m hoping that five to 10 years our ability to access data and that scale on many individuals is just will no longer be a bottleneck and we’ll have now systems and processes in place for layering the machine learning and AI on top of the data so that we can get to accurate and more better predictions for individuals.

Saul Marquez: [00:08:09] That’s inspiring to think of the possibilities. Once those obstacles are lifted and as health leaders strive to give the best to their communities as pharma companies look to put together the best of what they can to deliver new new medicines to patients. What do you think is at the forefront of their minds a hot topic that you should be on every health leaders agenda and how are you and your organization approaching it.

Iya Khalil: [00:08:34] Yeah you hear a lot about rising costs in healthcare and that the numbers just don’t add up right. We don’t necessarily see better outcomes with just more spend and it’s an interesting statement for people to be making these days especially when we are literally living through one of the most transformative times in healthcare. So much innovation is happening right. And there is a bit about sort of how are we going to pay for this innovation. How are we going to afford it. And in my mind that the real sort of power that AI has is to really transform healthcare into a precision medicine model so that the innovation and the new treatments that we’re bringing to patients we’re spending it on the right patients. We’re actually figuring out ahead of time who’s going to benefit who isn’t going about if it had a better benefit patients. And when we can be precise about matching interventions and treatments and drugs to patients and know ahead of time what the outcome is and actually get that right then I really believe we can start to move the needle towards you know lowering our total costs of care are making this innovation affordable for everybody.

Saul Marquez: [00:09:38] I think that’s such a great call out. Dr. Khalil and a lot of folks look at things like evidence based medicine. It’s really kind of just average based medicine right at the end of the day. Right. And so what you’re saying just resonates with me because it just provides an opportunity to get a pinpointed precision strategy for each patient.

Iya Khalil: [00:10:01] Exactly. Exactly right.

Saul Marquez: [00:10:03] It’s amazing. Can you give an example to the listeners of how you and your organization have created results by thinking and doing things differently.

Iya Khalil: [00:10:11] Yeah I first want to call out that it’s not just about the algorithms you know what allows us to do these things really is the abberation aspect and we fundamentally believe in taking now. You know that smart brains are physicists and mathematicians and engineers and marrying it with clinicians in the clinical community and the data we’re so fortunate we have the option to do this now with many great institutions. One of them is the Multiple Myeloma Research Foundation which is committed to a finding and bringing better new treatments for most myeloma patients. So we set up a collaboration with them they have an initiative called Compass that people volunteer to submit their samples and data and Amamoor often takes on the responsibility of measuring now in their tumour cells what are the genetic and genome and drivers of the cancer what are the molecular changes that happened in the cancer. How does the outcome of that patient change over time as they’re given different treatments and we have Machine learning and company are you going to access the data and we generated algorithms that are trying to learn the response of patients to that treatment based on that data. So we looked at a specific question which is a really important one which is a decision around stem cell therapy and who is benefiting from something that could be great benefits to it and there might be some patients where they may not be getting downshifted and we’d love to know who they are and what we can do to change their course of treatment ahead of time so they do get better fit. So we Legro that data to learn use causal machine learning Taggett learn algorithms from that data that would make that prediction and then took in those predictions and went to did the Dana Farber Clinical Center here in Boston to validate those patients and we’re starting to see some really great validation results. And this work this ability to learn the markers that were potentially predict benefit was all done once the data was collected computationally and within months less than three months. And then by one of my engineers and then him taking those results and working with clinicians to validate them and seeing how we can then get to the next step up eventually and hopefully delivering that insight to the patient.

Saul Marquez: [00:12:15] Such an insightful project Dr. Khalil and it’s awesome to hear you talk about the importance of those partnerships at the core. I feel like it’s easy for founders to get seduced by their technology and it’s so great to hear you just talk about that with as much passion as you do about the algorithms. What would you say was one of your proudest leadership experiences in medicine to date.

Iya Khalil: [00:12:39] So I mean when you’re an entrepreneur and you’re out there itching and trying to recruit people to come to your company and also join your cause your leader is often tested and you have to figure out more and more ways to motivate people and get them on board and I’ve always been very passionate about that especially in machine learning and I’m helping bridge understanding and get people on board but I think one of them may have come up more recently because I’m learning more and more I can only interact with people there’s a sort of a limit to how many people I can speak to and it becomes sort of how you’re able to represent. So I’ve been given the opportunity to serve Barnum Charlie Baker is one of his counsels for Chiha which is a hub around all of the information and data collected in the state of Massachusetts on our health care system and with the goal of being able to see that data to measure it and create metrics that help us make health care affordable for everybody here. So to be able to be part of that and to be at that intersection of how I can serve my government and health care right and reach.

Saul Marquez: [00:13:40] Pretty awesome.

Iya Khalil: [00:13:40] Yeah. It’s really amazing and I think it’s also sort of that to represent. Right. So still not many women and minorities and whatnot serving on some of these things and to get her to represent and hopefully provide that leadership through example and reach more people.

Saul Marquez: [00:13:56] I think that’s so wonderful. Congratulations on that. And just a trailblazing for the ladies in health care. In fact I really look at this Dr. Khalil as an opportunity. I mean I feel like in health care a lot of women are starting to come out like yourself to really make a difference and put not only the technology but also the heart into improving outcomes. Can you tell us a little bit more about an exciting project that you’re working on today.

Iya Khalil: [00:14:22] Yeah so one of the things that one of the trends that have been happening is more and more pharmaceutical companies are collecting data early in their trials on their patients and data. Beyond the detailed clinical record data on the genomic level molecular level. So we can see the molecular and granular changes that are happening in the individual patient as they’re given these new treatments and new drugs and we’re taking in that data and using causal machine learning to learn about who really truly is benefiting. Where there might be still some patients that don’t benefit. And with that learning using that to inform the next trial from a face to face 3. And the goal is to hopefully enable you know faster better approval of new treatments and where we’re targeting the right treatment the right patient. You know I’d love us to live in a paradigm where we are using these trials to learn really what’s working and for who and where machine learning is enabling that faster and better so that we can get better treatments to patients much sooner than we are today with much higher success rates.

Saul Marquez: [00:15:26] Well there’s no doubt that your efforts are going to be creating some major ripple effects for everybody listening to this. So listeners it’s one of those opportunities for you to just understand and learn more. We’re going to share at the end of this episode how you get in touch and research the company. But before we do I just want to walk through a little program here with you close to the end of the podcast Dr. Khalil you and I are going to build a leadership course on what it takes to be successful in healthcare today. It’s the 101 of Dr. Iya Khalil and so we’re going to write out a syllabus a lightning round with four questions followed by your all time favorite book for the listeners. You ready.

Iya Khalil: [00:16:06] Yes.

Saul Marquez: [00:16:07] Awesome. What’s the best way to improve health care outcomes.

Iya Khalil: [00:16:11] Precision medicine.

Saul Marquez: [00:16:12] What is one of the biggest mistakes or pitfalls to avoid.

Iya Khalil: [00:16:16] Understand the problem that you’re trying to solve. At every level clinically biologically and then really come up with the right solution. Now whether it’s infrastructure people but also the technology.

Saul Marquez: [00:16:27] Love it how do you stay relevant as an organization. Despite constant change.

Iya Khalil: [00:16:32] One of our models is we’re going to use the most powerful mathematics possible and do the most rigorous science.

Saul Marquez: [00:16:37] Beautiful. What’s one area of focus that drives all else in your company.

Iya Khalil: [00:16:42] Solving what we call the matching problem just for every patient being able to predict ahead of time what the right treatment what the right intervention is at the right time.

Saul Marquez: [00:16:52] That is a moonshot and I love that you guys are thinking big like that. What book would you recommend to the listeners as part of the syllabus.

Iya Khalil: [00:17:01] I’m going to give an unconventional sort of advice. All right let’s hear my people go what are you doing. So it’s called Causal Inference and statistics a primer.

Saul Marquez: [00:17:12] Right. There we go

Iya Khalil: [00:17:14] And you don’t have to read the whole thing. If all you do is read the first chapter and just get through the first round I guarantee it will just transform how you think about data statistics and what it can do for medicine.

Saul Marquez: [00:17:27] Amazing a great recommendation. And listeners don’t worry about writing that down. We’re going to provide all of that to you on the website. Just go to outcomesrocket.health/GNS and you’re going to find all of the show notes a transcript of our conversation today as well as links to the things that we’ve discussed including the GNS health care website. Before we conclude I love if you could just share a closing thought. And then the best place where the listeners can’t get in touch with you or follow you.

Iya Khalil: [00:17:59] Thank you. So my closing part is that a lot of noise in our marketplace right now with all the buzzwords where there’s machine learning and data. But the buzz is real and there is real steps behind it. And really for those who want us to get to that next level in medicine health care you know it’s upon all of us to really learn about what’s happening and to figure out how we can contribute to making medicine and healthcare better for all of us.

Saul Marquez: [00:18:23] Beautiful and what would you say the best place for the listeners to get in touch with or follow you.

Iya Khalil: [00:18:28] Right. So two ways. I’m from Linkedin so I love to connect to be that way. And then GNS health care we put out a blog that to release content and hopefully also you know help really people understand the space better and get them involved better so we’d love it if you would join our blog as well.

Saul Marquez: [00:18:46] Is that on the GNS Healthcare.com website?

Iya Khalil: [00:18:50] It should be.

Saul Marquez: [00:18:51] Outstanding. So listeners will go ahead and include that link to the blog. So you could stay tuned in to Dr. Khalil and what her team are doing because today has just been the tip of the iceberg. And so I want to just give you a huge thanks from everybody listening Dr. Khalil and we’re looking forward to staying in touch.

Iya Khalil: [00:19:10] Thank you. My pleasure, this was a really great session. I really enjoyed it.

: [00:19:17] Thanks for listening to the outcomes rocket podcast. Be sure to visit us on the web at www.outcomesrocket.health for the show notes, resources, inspiration and so much more.

Recommended Book/s:

Causal Inference in Statistics: A Primer

The Best Way To Contact Iya:

Mentioned Link/s:

http://www.gnshealthcare.com/

Episode Sponsors:

Healthcare Podcast

Outcomes Rocket Podcast - Iya Khalil