by Anna Fiorentino
Rai Winslow, director for the Roux Institute speaks at a panel about AI in the Life Sciences. (Photo by Heratch Ekemekian)
If anyone understands the potential impact of artificial intelligence (AI) on health care, it’s Rai Winslow. In 2005 the world-renowned leader in computational medicine founded and then for 15 years directed the legendary Institute for Computational Medicine at Johns Hopkins University, where he set out to prove that modeling is vital to the future of medicine. Now, as director of life science and medicine research at Northeastern University’s Roux Institute, he’s paving the way for caregivers to translate those theoretical results into medical applications that run on algorithms.
“AI and machine learning are poised to play a huge role not only in life sciences, but in medical research,” says Winslow. “Clinical trials on how an algorithm can function as part of the health care team will enable better medicine, better therapy, and better diagnoses for patients.”
Last month, at the Institute for Experiential AI (EAI) launch event, Winslow took his vision to a panel discussion still resonating with many of the 400 attendees on how modeling-based approaches are becoming increasingly important to the field of medicine.
“The medical literature is full of fantastic work with enormous potential when it comes to predictive analytics, but oftentimes those papers aren’t further developed,” he says.
That’s where the “experiential” piece of the Institute for Experiential AI comes into play, he says.
“Going that last mile from idea to proof of principle in the clinic is a really hard thing to do. I think we need to tackle that hard task, and that’s my mission in the Roux and the Institute for Experiential AI,” says Winslow.
Five years ago there wasn’t a single hospital in the country capable of collecting continuous and rapidly evolving data in real time. Today, many health care settings are processing high-frequency data from wearable patient monitors that are constantly tracking vital signs and behavior changes. The goal is to land on actionable data based on the moment-to-moment physiological state of patients so clinicians can make real-time predictions about their health trajectories. Medicine, says Winslow, is becoming a computational science.
“AI recommendations are like taking a blood sample and feeding its laboratory analysis back to a doctor so they evaluate what to do with it,” says Winslow. “Only now the process is augmented with this new kind of insight.”
If we can give caregivers advanced warning about patients headed towards a negative outcome, then we enable caregivers to intervene as early as possible. On the road to the future of medical AI though, we must acknowledge that methods are only as good as the data from which they learn — and that data can have inherent biases. So as computational medicine evolves over the next decade, part of the challenge will be educating caregivers on how to shift their thinking to interpret data.
“Ultimately, it’s the caregiver that needs to decide on those recommendations and put them into action, but we need to present the results of complex algorithms to caregivers in the most supportive and insightful ways. We need to enable them to understand the strengths and weaknesses of the predictions we make,” Winslow says.
Because after all, as he likes to say, AI is just another member of the health care team.
“Human-in-the-loop is a phrase I first heard from [Inaugural Executive Director for the Institute of Experiential AI] Usama Fayyad in the early days of the institute,” says Winslow. “I’ve never envisioned AI being anything but human-in-the-loop. It is invariably human-in-the-loop.
Watch lively multidisciplinary presentations and panel discussions from the April 6 Discover ExperientialAI event, which celebrated the launch of this foundational center for research and applied AI solutions. Contact us now to join forces with EAI in research or business.