The RSNA partnered with the Society of Thoracic Radiology to create the massive dataset, which includes CT scans from five international research centers and detailed clinical annotations by more than 80 expert thoracic radiologists. The collection is the largest publicly available, expert-annotated pulmonary embolism CT data for AI, according to the RSNA.

The challenge is being run on a platform provided by Kaggle, and winners will receive $30,000, according to the society. Submissions are due October 26. The RSNA will announce the winners on November 23, with the top submissions receiving recognition during the virtual RSNA meeting the following week.