A shortage of about 100,000 physicians is expected in the US by the year 2025, which will force many people into long waiting periods and delays in the detection and treatment of potentially life-threatening conditions. In terms of providing health care in a timely manner to the growing population, we are reaching the limits of human capacity. So, where do we turn? According to the team at Geisinger, the answer is machines. Brandon K. Fornwalt, MD, PhD, and associate professor and director of the Geisinger Department of Imaging Science and Innovation, and Aalpen Patel, MD and Chair of the Geisinger System of Radiology join the podcast to discuss the problems with the current system of medicine and diagnostic imaging, how the implementation of machines and machine learning can help, and the specific projects they’re working on to improve patient outcomes and help physicians to be more efficient. In a recent project, the team at Geisinger trained a computer model to predict death in patients by using over 700,000 videos of the heart. The result? The machine made predictions with an accuracy level far superior to that of clinicians, allowing for early intervention for heart attack and stroke. Tune in for a compelling discussion, detailed description of other projects in the works, and Geisinger’s long-term goals. To learn more, check out geisinger.org.