Comparison of mammography AI algorithms with a clinical risk model for 5-year breast cancer risk prediction: An observational study

Although several clinical breast cancer risk models are used to guide screening and prevention, they have only moderate discrimination. The aim of this research study is to compare selected existing mammography artificial intelligence (AI) algorithms, including Mirai, a tool developed at the MIT Jameel Clinic by a team of researchers lead by Jameel Clinic AI Faculty Lead, Regina Barzilay, and the Breast Cancer Surveillance Consortium (BCSC) risk model for prediction of 5-year risk. Negative screening mammographic examinations were analysed with five artificial intelligence (AI) algorithms; all predicted breast cancer risk to 5 years better than the BCSC clinical risk model, and combining AI and BCSC models further improved prediction.