Advances in DCE-MRI Radiomics for Non-Invasive Prediction of Breast Cancer Molecular Subtypes: Research Progress and...

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Dynamic contrast-enhanced MRI (DCE-MRI) is a key imaging technique.
• Radiomics extracts quantitative features from medical images.
• Predicting breast cancer molecular subtypes non-invasively is crucial for treatment.

Purpose of the Study:

• To review methodological innovations in DCE-MRI and radiomics for breast cancer.
• To evaluate clinical validation and translational applications of this integrated approach.
• To identify challenges and future directions in radiomic biomarker development.

Main Methods:

• Systematic review of studies integrating DCE-MRI and radiomics.
• Analysis of standardized DCE-MRI protocols and multidimensional radiomic features.
• Evaluation of deep learning models for prediction and risk stratification.

Main Results:

• Standardized DCE-MRI and radiomics improved subtype discrimination (ER, HER2, triple-negative).
• Deep learning models enabled early treatment response assessment and recurrence risk stratification.
• Radiomic models showed concordance with gene assays and potential to reduce biopsies.

Conclusions:

• DCE-MRI and radiomics provide a powerful non-invasive tool for breast cancer subtyping.
• Challenges include data heterogeneity and mechanistic interpretation.
• Radiogenomic analysis and organoid platforms are emerging for precision therapy.