Utilizing Longitudinal DCE-MRI and Tumor Microenvironment Data for Predicting Neoadjuvant Therapy Outcomes in Breast Cancer Through Deep Learning Techniques - Takeaways - MDSpire

Utilizing Longitudinal DCE-MRI and Tumor Microenvironment Data for Predicting Neoadjuvant Therapy Outcomes in Breast Cancer Through Deep Learning Techniques

  • By

  • Lan Yan

  • Xianming Huang

  • Lan Liu

  • Ao Wu

  • Yingyi Luo

  • Hao Li

  • Shaofeng Yi

  • Tenghua Yu

  • Qiao Zeng

  • April 28, 2026

  • 0 min

Share

  • 1

    A multimodal fusion model was developed to predict pathological complete response to neoadjuvant therapy in breast cancer patients.

  • 2

    The model integrates deep learning features from DCE-MRI, peripheral blood inflammatory indices, and tumor-infiltrating lymphocytes.

  • 3

    In the validation cohort, the combined model achieved an AUC of 0.90, outperforming single-modality models significantly.

  • 4

    The Post-2nd-NAT deep learning model showed an AUC of 0.85, indicating the importance of early treatment imaging.

  • 5

    This study highlights the potential of a multimodal approach for personalized treatment planning in breast cancer.

Original Source(s)

Utilizing Longitudinal DCE-MRI and Tumor Microenvironment Data for Predicting Neoadjuvant Therapy Outcomes in Breast Cancer Through Deep Learning Techniques

  • by Lan Yan, Xianming Huang, Lan Liu, Ao Wu, Yingyi Luo, Hao Li, Shaofeng Yi, Tenghua Yu, Qiao Zeng

  • April 28, 2026

Related Content

Comparison of the diagnostic power of serum circulating miR-21, VEGF, and CA15-3 in breast cancer

Dynamic changes and clinical significance of the gut microbiota and serum metabolites in breast cancer onset, progression and chemotherapy intervention

Overcoming immunotherapy resistance in triple-negative breast cancer: a critical review of mast cell plasticity, metabolic reprogramming, and organoid models