Interpretable machine learning for personalized breast cancer screening recommendations.

Area of Science:

• Oncology
• Medical Informatics
• Machine Learning

Background:

• Breast cancer is a leading cause of cancer death in U.S. women, making early detection crucial.
• Current personalized mammography screening models are often computationally complex, hindering practical application.
• There is a need for efficient and accurate methods to guide individualized breast cancer screening decisions.

Purpose of the Study:

• To develop and evaluate a machine learning-based approach for personalized breast cancer screening recommendations.
• To address the computational challenges associated with traditional decision-process models.
• To generate explainable insights and actionable rules for healthcare providers.

Main Methods:

• Utilized a machine learning framework to analyze patient medical history and risk factors.
• Developed models to predict optimal screening intervals and recommendations.
• Incorporated explainability techniques to interpret model decisions.

Main Results:

• Machine learning models achieved high accuracy in personalized screening recommendations.
• The proposed approach significantly reduced computational complexity compared to existing methods.
• Actionable decision rules were derived from model insights, aiding clinical decision-making.

Conclusions:

• Machine learning provides an accurate and computationally efficient alternative for personalized breast cancer screening.
• Explainable AI insights can translate complex models into practical clinical guidelines.
• This approach has the potential to improve breast cancer early detection and reduce mortality.