Federated machine learning to take on brain-tumor diagnosis
The University of Pennsylvania is teaming with Intel to lead a federation of more than two dozen medical and research centers on an AI system for identifying brain tumors.
The system will train on tumor data from Penn Medicine and 28 institutions around the world using federated learning, which anonymizes patients while gleaning their applicable clinical characteristics.
Lead investigator Spyridon Bakas, PhD, of UPenn’s Center for Biomedical Image Computing and Analytics suggests the globally distributed data sweep will ward off AI bias by tapping “ample and diverse data that no single institution can hold.”
In making the announcement, Intel cites statistics from the American Brain Tumor Association showing almost 80,000 people, including 4,600 children, will be diagnosed with a brain tumor this year.
Along with UPenn in the U.S., institutions participating in the project represent Canada, the U.K., Germany, the Netherlands, Switzerland and India.
“In order to train and build a model to detect a brain tumor that could aid in early detection and better outcomes, researchers need access to large amounts of relevant medical data,” the company notes. “By utilizing this [federated learning] approach, researchers from all partner organizations will be able to work together on building and training an algorithm to detect a brain tumor while protecting sensitive medical data.”
Dave P. has worked in journalism, marketing and public relations for more than 30 years, frequently concentrating on hospitals, healthcare technology and Catholic communications. He has also specialized in fundraising communications, ghostwriting for CEOs of local, national and global charities, nonprofits and foundations.