Scholarlink

Objective <p>Develop a multimodal fusion model combining MRI radiomics and deep learning (DL) to predict pathologic complete response (pCR) in breast cancer patients post-neoadjuvant chemotherapy (NACT).</p> Methods <p>Patients with locally advanced breast cancer from two centers (Center 1: 404 training, Center 2: 174 testing) were enrolled. Pretreatment MRI radiomic features and DL-derived features were extracted, selected via Minimum Redundancy Maximum Relevance and Least Absolute Shrinkage and Selection Operator, and integrated with clinical risk factors into a joint multimodal fusion model using XGBoost. The model was validated internally on an independent test cohort and externally on the Duke Breast Imaging Dataset (284 patients) to assess generalizability across diverse populations and imaging protocols.</p> Results <p>The fusion model achieved AUCs of 0.935 (training), 0.916 (testing), and 0.83 (external validation), surpassing standalone radiomics and DL models. External validation confirmed robustness to institutional variability, with strong calibration and clinical utility.</p> Conclusion <p>The XGBoost-integrated radiomics-DL-clinical model accurately predicts pCR, validated across independent cohorts, and may optimize personalized treatment strategies.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Integrating MRI radiomics and deep learning for predicting pathologic complete response to neoadjuvant chemotherapy in breast cancer: a multicenter study

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月14日

查看全文
  • Min Wei,
  • Zhenyu Shu,
  • Maohua Lyu,
  • Yanting Liang,
  • Yuguo Wei,
  • Ying Wang,
  • Guangying Zheng

摘要

Objective

Develop a multimodal fusion model combining MRI radiomics and deep learning (DL) to predict pathologic complete response (pCR) in breast cancer patients post-neoadjuvant chemotherapy (NACT).

Methods

Patients with locally advanced breast cancer from two centers (Center 1: 404 training, Center 2: 174 testing) were enrolled. Pretreatment MRI radiomic features and DL-derived features were extracted, selected via Minimum Redundancy Maximum Relevance and Least Absolute Shrinkage and Selection Operator, and integrated with clinical risk factors into a joint multimodal fusion model using XGBoost. The model was validated internally on an independent test cohort and externally on the Duke Breast Imaging Dataset (284 patients) to assess generalizability across diverse populations and imaging protocols.

Results

The fusion model achieved AUCs of 0.935 (training), 0.916 (testing), and 0.83 (external validation), surpassing standalone radiomics and DL models. External validation confirmed robustness to institutional variability, with strong calibration and clinical utility.

Conclusion

The XGBoost-integrated radiomics-DL-clinical model accurately predicts pCR, validated across independent cohorts, and may optimize personalized treatment strategies.