Two dozen scientists from around the world and a variety of research backgrounds converged at Carnegie Mellon University in April 2019 to discuss how to push the field of cancer research forward using artificial intelligence.
The workshop was organized in conjunction with the Mark Foundation for Cancer Research, an organization that funds groundbreaking research worldwide that was created by Carnegie Mellon alumnus and Pamplona Capital Management founder Alex Knaster (E 1980).
“Computation and data-driven approaches are changing the landscape [of cancer research],” Carnegie Mellon President Farnam Jahanian said in opening the two-day workshop. “The possibilities of tomorrow are reflected here today.”
Kicking off the discussion at the workshop was a keystone address by Stephen Friend, a longtime innovator in the cancer biology research world. Friend founded Sage BioNetworks, an organization that provides tools for collaborative research looking at large amounts of disease and health data. He now serves as a visiting professor of connected medicine at Oxford University. Friend gave attendees an overview of his career studying gene expression and diseases and the lessons he’s learned along the way about working efficiently but intelligently with data. His most recent focus is on zooming into individual patients to look at how their symptoms vary from day-to-day and how these fluctuations could be tracked and studied to figure out how and why diseases manifest differently in different people.
“The end goal is we’re interested in whether you could build a personal health assistant,” Friend said of his work.
Following the keynote talk, Glen de Vries Dean Rebecca W. Doerge led a panel of four participating researchers to discuss the difficulties and possibilities of taking AI tools from the lab into the clinic to use in treating patients. The panelists noted the difficulties in explaining AI to clinicians and patients who usually don’t have backgrounds in computer science but also said that they’d found many doctors were still eager to embrace this data and tools if it meant helping to find cures and treat patients better.
For the rest of the conference, participants presented their own research in short talks to attendees and then broke off into small groups to discuss current challenges facing AI research on cancer and their ideas to get past those challenges. A common theme expressed by nearly all the researchers was the difficulties they encountered in accessing datasets, many of which are held by different organizations with different procedures and rules for using them. Participants also pitched ideas for various collaborations among themselves, leveraging their own specialties and resources to open up new areas of research.
Seeing these fruitful relationships form among attendees was very encouraging, Mark Foundation CEO Michele Cleary said in closing the workshop. “Please keep these conversations going.”