MIT Researchers Develop New Way to Crowd-Source Drug Data
Researchers at MIT have developed a new way for drug sponsors and researchers to share information without compromising patient privacy or intellectual property. There have been several efforts to create open-source programs and speed up clinical trials, but programs sophisticated enough to do the work couldn’t be safely encrypted.
MIT researchers Brian Hie, Hyanghoon Cho and Bonnie Berger solved the riddle by building a neural network that relies on a program cryptographers call “secret sharer.” Rather than trying to run all the data through a large algorithm — which can get clogged up by non-numeric fields that ordinarily would be marked as “confidential” or “X” — the system runs several simple, blinded operations at once and then recombines them.
