Novel deep learning-based prediction of HER2 expression in breast cancer using multimodal MRI, nomogram, and decision curve analysis

This study demonstrates that integrating deep learning with multi-sequence breast MRI and clinical data provides a highly effective and reliable tool for predicting HER2 expression in breast cancer. The model's performance, validated through rigorous evaluation, offers significant potential for clinical implementation in personalized oncology, improving decision-making and treatment planning for breast cancer patients.