Interview with Iya Khalil of GNS Healthcare

Dr. Iya 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 co-founded two big data companies, Via Science and GNS Healthcare, and is the co-inventor of the proprietary computational engine that underpins both entities. Dr. Khalil’s expertise spans applications in drug discovery, drug development all the way to treatment algorithms that can be applied at the point of care. She is a frequent speaker at industry events and conferences and was recognized by President Obama at a White House dinner as a leading entrepreneur in genomic medicine. Read her full bio.

Interview with Iya Khalil of GNS Healthcare

Q: Artificial intelligence (AI) techniques have sent vast waves across healthcare, even fueling an active discussion of whether AI doctors will eventually replace human physicians in the future. Do you believe that human physicians will be replaced by machines in the foreseeable future? What are your thoughts?

A: I think that there’s a lot of speculation and uncertainty around AI, but I don’t foresee a time when we won’t need physicians. However, I do believe AI will help clinicians make better, faster decisions that will lead to better outcomes.

For example, take the case of newly diagnosed cancer patients. When patients present with their disease, the physician has to make a choice as to which of many treatment options they are going to prescribe and whether to provide a treatment alone or in combination with others. Right now, those decisions are based on the standard of care. But what we know is each of us has different and unique biology and genetics. We now have the data, the technology and the processing speed to build disease models and run computer-based in silico simulations on every possible treatment scenario and inform the physician on the right treatment for each individual based on their biology. That’s the real power of AI.

Q: Can you provide some use cases that have already successfully demonstrate the value of AI/Machine Learning in healthcare?

A: Oncology is certainly a focus area for AI ,machine learning and precision medicine, especially given the launch of the Cancer Moonshot Initiative and the on-going development of innovative therapies like immunotherapy and CAR T-cells.

In terms of use cases, GNS has been partnering with the Alliance for Clinical Trials in Oncology and have discovered genetic mutations in patients with metastatic colorectal cancer, specifically a difference between left sided and right sided tumors. This is a key finding in that it allows clinicians to understand the rate of the disease progression and then base treatment on underlying biology.

Another example would be our joint effort with the Multiple Myeloma Research Foundation, which resulted in the discovery of a biomarker at certain levels that identify which patients are likely to benefit from stem cell transplantation. Stem cell transplantation is a lengthy, costly procedure so understanding who benefits is crucial.

Another area where causal machine learning can impact precision medicine is in Central Nervous System (CNS) disorders, which are difficult to develop therapeutics for because the underlying biology is so complex. GNS leveraged data from the Michael J. Fox Foundation to discover genetic and molecular markers that determine faster motor progression of the disease. This helps accelerate the clinical trial process and the development of effective drugs for Parkinson patients.

Q: What areas in healthcare will benefit the most from AI/Machine Learning applications and when will that be?

A: The simple answer is that we will all benefit. But it’s worth noting that there are many different kinds of AI and applying the right approach to the right problem is key. The AI we developed is called causal machine learning and it is different than others because it discovers new information in the data and allows you to run simulations to discover the outcomes of different treatments. Other types of AI scan available information and make correlations.

In terms of the healthcare industry, we work closely with biopharma companies to help them understand how their drugs work in the real world. As you know clinical trials are conducted in a very controlled environment, but we can run in silico clinical trials, ( clinical trials in the computer) using real world data to understand the mechanisms of action of a particular drug, determine which individual patients are going to benefit from it, how combinations of drugs work and unravel the underlying biology of disease.

We also work with health plans to leverage our AI to identify which care interventions are going to work best for which member and help them understand how their providers and system models are working. This allows health plans to provide personalized care pathways for its members while controlling costs.

Q: What are some of the challenges to realize AI/Machine learning in healthcare?

A: We’re living at a unique time in healthcare. We can now identify every gene variant across an entire genome. We can measure the expression of molecular changes in tens of thousands of cells. We can use advanced imaging technologies to peer into organs and physiology and monitor the state and health of your microbiome across an entire lifetime. But the challenge lies in unraveling the complexity of biology in a way that’s transparent and explainable. In order for healthcare to advance, you need to understand the “why”, why are certain cause and effect relationships happening, and which biomarkers or genetic mutations are driving the progression of disease. So the challenge comes down to leveraging the right technology for the right problem.

Q: How close are we with successfully using AI for the purpose of mining big data?

A: We’re there now. The world is creating two and a half million terabytes of data every day – and nearly 30 percent of that is being generated by the healthcare industry thanks to the explosion of EHR, digital imaging, natural language processing, genetic data and connected medical devices.

The power and potential of AI technology has come a long way, and it is being recognized as a technology that will have real-world impact. The FDA offered its vote of confidence by encouraging the use of AI and other digital tools in medicine and drug development. The 21st Century Cures Act signed into law in 2016, was designed in part to accelerate drug development and includes the expansion of drug labels through the use of analytics and AI to generate real world evidence from observational data, without a new clinical trial. We are using AI and big data today to improve healthcare and it will only get smarter over time.

Q: What is your outlook or vision for use of AI/Machine Learning in healthcare?

A: My outlook is extremely positive. We now have the capability to discover the underlying mechanisms driving disease, so we can optimize treatments and design and develop more effective drugs to battle them. And we can do it in a much shorter amount of time which has real life impacts for patients. Given that we now have the capability to unravel our biology, and predict how to intervene to get to the best possible outcomes, our challenge then becomes how do we apply this to every disease and every aspect of our health, including how we age? I want to figure out a way to get there faster. Not decades from now, but right now. I look forward to a time where we can predict serious illness or life-threatening events well before they happen and actually prevent them.

Q: Is there anything you would like to share with the PMWC audience?

A: We are on the precipice of making precision medicine a reality. We now have all the ingredients necessary to unravel human biology, better understand disease progression and how treatments work based on an individual’s biology. The ability to match the right treatment to the right patient at the right time; that’s a gamechanger and one that will benefit all of us.