Dissertation Talk: Expert-Level Detection of Acute Intracranial Hemorrhage on Head Computed Tomography using Deep Learning
Presentation | May 10 | 1-2 p.m. | Sutardja Dai Hall
Computed tomography (CT) of the head is the workhorse medical imaging modality used worldwide to diagnose neurologic emergencies. However, these grayscale images are limited by low signal-to-noise, poor contrast, and a high incidence of image artifacts. A unique challenge is to identify, with perfect or near-perfect sensitivity and high specificity, often tiny subtle abnormalities occupying 106 pixels. We used a single-stage, end-to-end, fully convolutional neural network to produce state-of-the-art exam-level classification performance on an independent test set, comparable to that of U.S. board-certified radiologists, in addition to robust localization of abnormalities including some that are missed by radiologists, both of which are important to this application.