Data-Driven Inference Group (DIG)

The Data-Driven Inference Group uses machine learning and computer vision to improve outcomes in medicine, finance, and sports. Our group focuses on the application of advanced computational techniques to medicine. Current projects include prediction of adverse medical events, prediction of response to therapies, non-invasive monitoring and diagnostic tools, and tele-medicine.

Researchers from MIT CSAIL, the University of Michigan, Brigham and Women’s Hospital in Boston and Harvard Medical School have developed a new tool that can more accurately determine risk of death in patients who have suffered a heart attack. Results of the study could prove life saving for the millions of Americans who suffer heart attacks every year.

The new technique involves searching for subtle indicators of risk hidden in a patient’s electrocardiogram (EKG or ECG) history. The electrocardiogram measures and displays the electrical activity of the heart. Current techniques for determining risk of death in patients who have suffered a heart attack tend to only identify a small portion of resulting fatalities.

The study was conducted by MIT EECS Professors and CSAIL Principal Investigators John Guttag and Collin Stultz (who is also an investigator in MIT RLE), University of Michigan Professor (and MIT EECS and CSAIL alumnus) Zeeshan Syed and Brigham and Women’s Hospital cardiologist Benjamin M. Scirica. Results of the study are published in a new paper in the Sept. 28 edition of Science Transitional Medicine.