Researchers from Stanford University have released a dataset of 10,000 labeled echocardiogram videos to advance the development of machine learning systems that can analyze videos in clinical settings. The videos are from patients who underwent echocardiography between 2016 and 2018 at Stanford University Hospital and include labels such as the volume of blood in a ventricle after the heart contracts. AI systems that can analyze echocardiogram videos could help diagnose conditions that make it more difficult for the heart to pump blood.
Image: Daisuke Koya, Kazuyuki Shibuya, Ryuichi Kikkawa and Masakazu Haneda