Now it’s time to introduce you to Iya Khalil, who co-founded GNS Healthcare with Colin Hill and serves as the company’s Chief Commercial Officer. When I sat down to talk with Iya, I wanted to get her insights into the company’s work and how it was going to bring its’ industry leading capabilities to more businesses and situations. I also wanted to know more about what the company is doing to identify and fight disease. She was able to help me understand how GNS is working with partners to improve the application of GNS’ solutions and the important work that remains to be accomplished. Now the work is to find the right partners to move the work GNS is doing to the end user – the patient.
When Iya co-founded GNS Healthcare, it was with the goal and vision of moving medicine and healthcare from an empirical practice to one that is quantitative, predictive, and patient-centric. Iya’s background is in theoretical physics and she believes that the mapping of the human genome was a game changer that opened the doors for GNS to use causal machine learning to bridge that gap. “When the human genome was mapped, all of a sudden we had the capability to use vast new data sets to unravel complicated human biology,” said Iya. “We knew that with powerful mathematical approaches and high-performance computing power, we could derive amazing new insights to unravel the black box of medicine and this is what I have been focused on for nearly 20 years.”
Feedback Loops are the Real Magic
Iya knew early on, through her work as a theoretical physicist, the power of mathematics and computational modeling to generate new knowledge of biology that can be tested in the real world. This new knowledge, put into practice, generates new data that can be used to further improve those models and derive ultimate causal power. It is this paradigm and the feedback loop of rigorous mathematical modeling with data and experiments that can generate predictions beyond what we can intuitively or empirically derive, that has led to some of the most monumental discoveries of our time from the discovery of the Higgs boson to gravitational waves. “My vision was to implement and enable that feedback loop, which had been used in other scientific disciplines, in medicine and healthcare so that we could predict and then understand the causes of a patient’s health and disease trajectory, then identify the optimal treatment that will lead to vastly improved health outcomes for every individual,” she explained.
Key to enabling this feedback loop is the data that we are now able to generate from being able to sequence entire genomes, to measuring whole proteomes, lipidomes, and metabolomes along with longitudinal digital health records and other real-time measurements of our health (device data, imaging, mobile health aps, etc.). “For the first time, we can use these rich longitudinal patient datasets to reconstruct the underlying causal mechanisms that drive our health and disease trajectory. At GNS we have built a learning engine that is able to reconstruct or reverse engineer models that learn those complex mechanisms and then in silicoidentify which treatment or intervention will work for an individual patient. This is the paradigm of precision medicine that we are committed to enabling,” said Iya. “Our job today is to bring that capability together with the right partners in order to bring the benefits back to the patient and continuously learn what works for whom so we can get better at treating and even some day curing disease and significantly extending human life,” said Iya.
Deliver Better Outcomes Requires Engaged Upstream and Downstream Partners
GNS has built the causal machine learning platform necessary to analyze massive data sets at lightning speed. It has proven it can deliver never-before-dreamed-of insights into how best to match patients with treatments that work best for them to insights on mechanisms driving a number of complex diseases that will can some day enable development of new treatments. Making that knowledge beneficial to individuals and populations means delivering it to the right end users. That requires partners upstream and downstream from the process.
“Our upstream partners are those who can give us patient-centric data sets for our work,” said Iya. “That data is the nuts and bolts that we need to get the insights that our customers and partners can use to take action that benefits their patients.”
The biggest challenge is getting data out of the disparate system and into usable formats so models can then be built around them. Currently important genetic and longitudinal data is held in “silos within silos within silos” according to Iya. The systems that hold that data are not interoperable. They are also preventing patients from receiving the benefits that could be obtained by a streamlined data retrieval and machine learning process.
When I asked Iya what databases would look like if they were no longer siloed, she didn’t hesitate to detail their components and how they would enhance the work of GNS and healthcare overall. She said that if databases were compiled and shared in a way to be most useful to population health and disease management, they would include the following: baseline genetic markers, large scale molecular profiling measures in tissue and blood, longitudinal health records that are available to both patients and service providers, learning systems that support education of patients about treatments and innovations to fight their disease. The more data available, the greater the opportunities for breakthroughs, and the better the results.
GNS’ downstream partnerships are their customers. The organizations, patient foundations, and healthcare providers that will apply this new knowledge and provide feedback, help improve the algorithms and take the work to the next level. This is the space where pharmaceutical companies, providers, and health plans live. When they use GNS’ capabilities to fight disease, and return information on successes and failures, it makes the algorithms smarter and more valuable.
“Throughout our history, we’ve refined the pathways to get to the answers that will change population health for the better,” said Iya. “The challenge is finding the right partners to sit at the table with us to move the work forward and enable a learning healthcare system that is empowered by machine learning.”
What’s the most important work of GNS Healthcare today?
It’s clear that Iya thinks big. When I asked her about GNS’ most important work, she responded, “It’s turning massive amounts of data into the information that supports precision medicine. That’s the work that GNS does every day – working out the algorithms necessary to tailor disease-fighting treatments to the individual patient.” Once tailored treatments can be identified, costs can be reduced, not only for the patient, but for the healthcare system as a whole.
Iya is driven by a fierce belief that there is much more work to be done to solve the problem of healthcare today and GNS is perfectly positioned to take a leading role in this effort. She talks in terms of “flipping the paradigm of healthcare” and creating “big shifts” instead of incremental change. “I rarely thought about how hard it was going to be,” she said. “But the world is ready for this. We are on the precipice of moving from a world where medicine is empirical and people are taking educated ‘best guesses’, to a data driven world in which decisions are based on algorithms that accurately predict outcomes and identify the best set of interventions for every individual.”
And that’s when we’re going to see the real action.
