The competitors will be evaluated on their ability to develop AI software to aid in finding breast cancer on screening mammography images from a dataset contributed by mammography screening programs in Australia and the U.S.
The accuracy of machine-learning models developed by the contestants will be evaluated against this ground truth dataset, which includes detailed labels, with radiologists' evaluations, and follow-up pathology results for suspected malignancies.
The dataset will remain available for use in further research at the conclusion of the challenge, scheduled to end March 31, 2023.
This challenge is part of a broader research project that will examine how models generated in the competition perform against previously unseen data and compare their performance to that of expert human observers. These questions are critical in determining how AI tools will perform in clinical settings, notes the RSNA.
Copyright © 2022 AuntMinnie.com