More than 1,800 health care leaders gathered in Boston recently at the World Medical Innovation Forum, hosted by Partners HealthCare, for focused discussions on the use of artificial intelligence (AI) and machine learning in clinical care. Below are some highlights from the forum’s three dynamic days.
Day One: Unleashing AI’s Potential
In rapid-fire presentations, 18 Harvard Medical School investigators highlighted promising AI applications. Henry Chueh, MD, Director of the Massachusetts General Hospital (MGH) Laboratory of Computer Science, presented DXplain, a clinical tool that expands the list of conditions a clinician should consider based on a patient’s symptoms, as a way to reduce missed diagnoses. Bharti Khurana, MD, Director of Emergency Musculoskeletal Radiology at Brigham Health, presented a new algorithm designed to assess a patient's imaging history to enhance early detection of injury patterns consistent with intimate partner violence (IPV).
Massachusetts Governor Charlie Baker outlined opportunities to use AI to develop new treatments, streamline decision-making and lower costs. "AI has tremendous potential to…help practitioners make really difficult decisions in the increasingly sophisticated, complicated, and powerful world of health care," he said, while also encouraging the audience to be stewards of patient privacy and trust.
Gregg Meyer, MD, Partners Chief Clinical Officer, highlighted AI’s potential to rehumanize health care. "AI could actually give me the opportunity to really be there with the patient… talking about their illness and working with them," he said. AI’s use in improving the drug development process was highlighted by industry speakers such as Bayer CMO Michael Devoy, MD, though they acknowledged that more work needs to be done to create systems that support these efforts.
Administrator for the Centers of Medicare and Medicaid Services Seema Verma summarized the opportunity ahead. "Interoperability means collecting all data, from the time a baby is born, in one place. If you combine that with AI, it'll give incredible insight into where the individual is and where they need to be."
Day Two: A Focus on the Future
The second day challenged participants to envision health care in 2054, with AI leading the way toward early detection and prevention. "I think the answer will be a mantra of early prediction—meaning knowing your family history—and your genetic risk you face just coming into this world, for a neurodegenerative disease like Alzheimer's, and that will guide at what point you do early detection," noted Rudy Tanzi, PhD, Vice Chair of Neurology and Director of the Genetics and Aging Research Unit at MGH.
Marcela Maus, MD, PhD, Director of the Cellular Immunotherapy Program at the MGH Cancer Center, discussed the potential to extend cell-based therapies into treatments for heart disease and neuroinflammation. In the context of these diseases, cells could be used as regenerative therapeutics, to restore organ function or to modulate the immune system to reduce chronic inflammation, she suggested.
Other discussions focused on making health care more transparent and equitable. "We need a system that allows the free flow of data and price information, that pays for providers to discover and implement treatments based on all that data, and that keeps the patient at the center of that process,” said the Honorable Alex M. Azar II, United States Secretary of Health and Human Services. Regina Barzilay, PhD, of MIT, speaking on the AI in Cancer Care panel, addressed disparities specifically in the context of clinical trials. "One statistic that really shocked me…is that today all clinical decisions are based on the 3% of the population which is participating in clinical trials," Barzilay said. "As a result, our treatments are often not applicable to whole demographics."
Final Day: Pioneering AI, from Practical to Aspirational
Presentations by the 2019 Partners Innovation Discovery Grants awardees showcased technologies designed to enhance cancer detection, improve mental health treatment, and increase visual guidance for brain surgery, among others. (Innovation Discovery Grant Program Book)
Stefan Oelrich, board member and President of Pharmaceuticals at Bayer AG, highlighted ways the company is using AI and machine learning technologies across the entire production cycle—from adapting molecules to better interact with targets to identifying the best patients for trials. "All of a sudden, thanks to machine learning, we're capable of not just building on a hypothesis and improving on a hypothesis, but to go the other way around and come to a different way of achieving an endpoint," he said.
The final panelists examined targeted treatment and prevention strategies in neurodegenerative disease. "We are at an unprecedented time in the neurosciences—we have a much better understanding of the biology behind some illnesses…but at the same time we have challenges with an aging population and the huge number of people affected by neurological illnesses," said Merit Cudkowicz, MD, Chief of Neurology at MGH. "It's an incredible opportunity to see how we can harness some of the things we’ve heard over these last two days."
The forum closed by unveiling the “Disruptive Dozen”—read more about these 12 breakthrough innovations expected to reinvent AI here.
According to Chris Coburn, Partners HealthCare chief innovation officer, the forum encourages critical debate and idea sharing among multiple audiences, from clinical experts to venture capital and industry. “Convening in-depth discussions of how AI plays a role in advancing medical innovation reflects the visionary, entrepreneurial spirit that’s found throughout Partners HealthCare,” he said. “The Forum creates a platform for collaborative dialogue to advance patient care.”
Read more about the World Medical Innovation Forum here.
Image above: the "Disruptive Dozen" panel speakers at the 2019 World Medical Innovation Forum