Relation knowledge distillation 3D-ResNet-based deep learning for breast cancer molecular subtypes prediction on ultrasound videos: a multicenter study.
Authors
Affiliations (9)
Affiliations (9)
- Cancer Center, Department of Ultrasound Medicine, Zhejiang Provincial People's Hospital (Affiliated People's Hospital), Hangzhou Medical College, Hangzhou, Zhejiang, China.
- Key Discipline of Zhejiang Province in Public Health and Preventive Medicine (First Class, Category A), Hangzhou Medical College, Hangzhou, Zhejiang, China.
- School of Mathematical Sciences, Zhejiang University, Zijingang Campus, Hangzhou, Zhejiang, China.
- Department of Ultrasound Medicine, the Second Affiliated Hospital of Harbin Medical University, Harbin, Heilongjiang, China.
- Department of Ultrasound Medicine, Sichuan Provincial Maternity and Child Health Care Hospital, Chengdu, Sichuan, China.
- Cancer Center, Department of Clinical Engineering, Zhejiang Provincial People's Hospital (Affiliated People's Hospital), Hangzhou Medical College, Hangzhou, Zhejiang, China.
- Department of General Surgery, the Second Hospital of Hebei Medical University, Shijiazhuang, Hebei, China. [email protected].
- Department of Ultrasound Medicine, the Second Affiliated Hospital of Harbin Medical University, Harbin, Heilongjiang, China. [email protected].
- Cancer Center, Department of Ultrasound Medicine, Zhejiang Provincial People's Hospital (Affiliated People's Hospital), Hangzhou Medical College, Hangzhou, Zhejiang, China. [email protected].
Abstract
To develop and test a relation knowledge distillation three-dimensional residual network (RKD-R3D) model for predicting breast cancer molecular subtypes using ultrasound (US) videos to aid clinical personalized management. This multicentre study retrospectively included 882 breast cancer patients (2375 US videos and 9499 images) between January 2017 and December 2021, which was divided into training, validation, and internal test cohorts. Additionally, 86 patients was collected between May 2023 and November 2023 as the external test cohort. St. Gallen molecular subtypes (luminal A, luminal B, HER2-positive, and triple-negative) were confirmed via postoperative immunohistochemistry. The RKD-R3D based on US videos was developed and validated to predict four-classification molecular subtypes of breast cancer. The predictive performance of RKD-R3D was compared with RKD-R2D, traditional R3D, and preoperative core needle biopsy (CNB). The area under the receiver operating characteristic curve (AUC), sensitivity, specificity, accuracy, balanced accuracy, precision, recall, and F1-score were analyzed. RKD-R3D (AUC: 0.88, 0.95) outperformed RKD-R2D (AUC: 0.72, 0.85) and traditional R3D (AUC: 0.65, 0.79) in predicting four-classification breast cancer molecular subtypes in the internal and external test cohorts. RKD-R3D outperformed CNB (Accuracy: 0.87 vs. 0.79) in the external test cohort, achieved good performance in predicting triple negative from non-triple negative breast cancers (AUC: 0.98), and obtained satisfactory prediction performance for both T1 and non-T1 lesions (AUC: 0.96, 0.90). RKD-R3D when used with US videos becomes a potential supplementary tool to non-invasively assess breast cancer molecular subtypes.