An explainable predictive machine learning model for axillary lymph node metastasis in breast cancer based on...

Area of Science:

• Oncology
• Medical Imaging
• Artificial Intelligence

Background:

• Accurate prediction of axillary lymph node metastasis (ALNM) is crucial for breast cancer staging and treatment.
• Integrating multimodal data (imaging and pathology) can improve ALNM prediction accuracy.

Purpose of the Study:

• To develop and assess explainable machine learning models for predicting ALNM in breast cancer.
• To evaluate the clinical utility of these models for risk stratification and diagnosis.

Main Methods:

• Retrospective study of 401 breast cancer patients.
• Implementation of ten machine learning algorithms, including Naïve Bayes, Random Forest, Logistic Regression, and Support Vector Machines.
• Application of SHapley Additive exPlanations (SHAP) for model interpretability.

Main Results:

• The Naïve Bayes classifier achieved the highest performance.
• Validation cohort achieved an Area Under the Receiver Operating Characteristic Curve (AUC) of 0.902.
• Key predictive features included lymph node parenchymal thickness, lymph node enlargement, and tumor width.

Conclusions:

• The Naïve Bayes model offers robust and interpretable prediction of ALNM.
• This explainable tool has clinical value for risk stratification and personalized treatment strategies.