The American Society of Neuroradiology (ASNR) and the Radiological Society of North America (RSNA) have teamed up for a unique health care data science challenge to detect intracranial hemorrhage from brain CT images. Four research institutions from North and South America provided large volumes of de-identified CT studies that were assembled to create the challenge dataset: Stanford University, Thomas Jefferson University, Unity Health Toronto and Universidade Federal de São Paulo (UNIFESP). The ASNR organized and recruited a cadre of more than 60 neuroradiologist volunteers to label over 25,000 exams for the challenge datasets. The ASNR volunteer members toiled throughout the summer months to get all of the data labeled in preparation for the official release of the competition datasets and machine learning challenge both of which will be hosted on Kaggle (www.kaggle.com).
The release of the datasets and start of the challenge is scheduled for September 2019. Contestants worldwide will compete to generate the most accurate machine learning algorithm to detect intracranial hemorrhage and classify hemorrhage subtypes (e.g. subarachnoid, intraventricular, intraparenchymal, subdural and epidural). The competition will conclude in mid-November and the winners will be announced at the RSNA Annual Meeting in Chicago on December 2nd. Stay tuned for more announcements as the competition begins!