Correctly diagnosing a person with cancer — and identifying the specific type of cancer — makes all the difference in successfully treating a patient.
Today your doctor might draw from a dozen or so similar cases and a big book of guidelines. But what if he or she could instead plug your test results and medical history into a computer program that has crunched millions of pieces of similar data?
That sort of future is looking increasingly possible thanks to CSAIL researchers. Working with a team from Massachusetts General Hospital (MGH), PhD student Yuan Luo and MIT Professor Peter Szolovits have developed a computational model that aims to automatically suggest cancer diagnoses by learning from thousands of data points from past pathology reports. The work has been published this month in the Journal of the American Medical Informatics Association.