Back to all papers

A multimodal cross-temporal fusion system for complementary evaluation of neoadjuvant chemotherapy in breast cancer.

September 9, 2026pubmed logopapers

Authors

Sun X,Wang D,Zhou Q,Liu W,Gao X,Shen X,Yang X,Ma T,Ma T,Liu C,He Y,Li Y,Wang W,Song L,Mao Y,Wang H

Affiliations (11)

  • Breast Disease Diagnosis and Treatment Center, The Affiliated Hospital of Qingdao University, Qingdao 266000, Shandong Province, China.
  • School of Computer Science and Technology, Tongji University, Shanghai 201840, China.
  • Department of Interventional Radiology, The Affiliated Hospital of Qingdao University, Qingdao 266000, Shandong Province, China.
  • Department of Radiology, Qingdao Women and Children's Hospital, Qingdao 266000, Shandong Province, China.
  • Department of Radiology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430000, Hubei Province, China.
  • Department of Radiology, The Affiliated Hospital of Qingdao University, Qingdao 266000, Shandong Province, China.
  • Department of Breast Surgery, The Affiliated Hospital of Jining Medical University, Jining 272029, Shandong Province, China.
  • Department of Medical Imaging, The Affiliated Hospital of Jining Medical University, Jining 272029, Shandong Province, China.
  • First Department of General Surgery, Weihaiwei People's Hospital, Weihai 264200, Shandong Province, China.
  • Breast Disease Diagnosis and Treatment Center, The Affiliated Hospital of Qingdao University, Qingdao 266000, Shandong Province, China. Electronic address: [email protected].
  • Breast Disease Diagnosis and Treatment Center, The Affiliated Hospital of Qingdao University, Qingdao 266000, Shandong Province, China. Electronic address: [email protected].

Abstract

Neoadjuvant chemotherapy (NAC) has been established as a standard treatment for breast cancer. We aimed to develop a deep-learning multimodal system using longitudinal cross-temporal dynamic contrast-enhanced MRI (DCE-MRI) and clinical data to provide complementary evaluation of NAC. Here, we developed a multimodal automatic NAC assistance system (MANAS), which leverages a dual-pathway attention-based multi-scale feature fusion framework and a multicenter dataset. This dataset comprises pre- and post-NAC DCE-MRI and associated clinical information of 1,515 patients with locally advanced breast cancer. The performance of MANAS was evaluated using various metrics. Confidence-based decision criteria were employed to defer low-confidence cases to clinicians. MANAS consisted of three modules from breast cancer diagnosis to follow-up treatment, designed for predicting molecular subtype, NAC response, and residual tumor burden, respectively. MANAS achieved area under the curves (AUCs) of 0.8066 (95 % CI: 0.7552-0.8599), 0.7748 (95 % CI: 0.7210-0.8275) and 0.7661 (95 % CI: 0.6980-0.8252) on the internal, pooled external and I-SPY2 test sets for molecular subtype prediction; AUCs of 0.8400 (95 % CI: 0.7903-0.8896) and 0.8299 (95 % CI: 0.7886-0.8687) on the internal and pooled external test sets for NAC response prediction; and AUCs of 0.8885 (95 % CI: 0.7891-0.9657), 0.8628 (95 % CI: 0.8047-0.9132) and 0.8695 (95 % CI: 0.7856-0.9432) on the internal, pooled external and I-SPY2 test sets for residual tumor burden prediction. Confidence-based decision criteria could enhance their predictive performance. Our MANAS serves as a valuable complementary tool, providing longitudinal supportive evidence throughout the NAC process, from initial diagnosis to surgical intervention.

Topics

Journal Article

Ready to Sharpen Your Edge?

Subscribe to join 11k+ peers who rely on RadAISlice. Get the essential weekly briefing that empowers you to navigate the future of radiology.

We respect your privacy. Unsubscribe at any time.