Performance of clinical breast cancer risk prediction models versus a mammography-based artificial intelligence risk model

Key Points

• The study aims to evaluate the accuracy of artificial intelligence models compared to clinical breast cancer risk prediction models.
• Compared the MIRAI AI model with clinical breast cancer risk models.
• Assessed the discriminatory accuracy for predicting five-year overall and invasive breast cancer risk.
• Evaluated calibration of the models for accurate risk assessment.
• MIRAI showed higher discriminatory accuracy than clinical models for predicting breast cancer risk.
• AI models overestimated risk for overall and invasive breast cancer endpoints.