Review: mpMRI plus AI can non-invasively assess HER2 status in breast cancer

Multiparametric MRI radiomics and AI models predicted HER2 status in breast cancer with internal AUCs exceeding 0.90 (up to 0.923 for three-level classification), but external validation AUCs were lower (0.66-0.84), showing need for prospective multicenter studies.

• Multiparametric radiomics incorporating peritumoral features achieved AUCs of 0.84–0.923 for three-level HER2 classification.
• Radiomics and deep learning distinguished HER2-low from HER2-zero tumors with AUC up to 0.892.
• Major limitations: retrospective design, variable reference standards, and limited external validation (external AUCs only 0.66–0.84).

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.