EllipseNet: Anchor-Free Ellipse Detection for Automatic Cardiac Biometrics in Fetal Echocardiography

As an important scan plane, four chamber view is routinely performed in both second trimester perinatal screening and fetal echocardiographic examinations. The biometrics this plane including cardio-thoracic ratio (CTR) cardiac axis are usually measured by sonographers for diagnosing congenital heart disease. However, due to the commonly existing artifacts like acoustic shadowing, traditional manual measurements not only suffer from low efficiency, but also with inconsistent results depending on operators’ skills. In paper, we present anchor-free ellipse detection network, namely EllipseNet, which detects thoracic regions automatically calculates CTR 4-chamber view. particular, formulate network that center of each object as points regresses ellipses’ parameters simultaneously. We define intersection-over-union loss further regulate regression procedure. evaluate EllipseNet clinical echocardiogram dataset more than 2000 subjects. Experimental show proposed framework outperforms several state-of-the-art methods. Source code will be available at https://git.openi.org.cn/capepoint/EllipseNet.