Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Clinical Studies

Relation knowledge distillation 3D-ResNet-based deep learning for breast cancer molecular subtypes prediction on ultrasound videos: a multicenter study

British Journal of Cancer volume 133, pages 1178–1188 (2025)Cite this article

Subjects

Abstract

Background

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.

Methods

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.

Results

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).

Conclusions

RKD-R3D when used with US videos becomes a potential supplementary tool to non-invasively assess breast cancer molecular subtypes.

This is a preview of subscription content, access via your institution

Access options

Fig. 1
Fig. 2: Workflow for dynamic breast cancer molecular subtypes prediction.
Fig. 3: ROC curves of R3D and RKD-R3D in the validation, internal test, and external test cohorts.
Fig. 4: Normalized confusion matrices of RKD-R3D method for predicting four-classification breast cancer molecular subtypes.
Fig. 5: The ROC curves of different knowledge distillation methods for predicting breast cancer molecular subtypes.
Fig. 6: Normalized confusion matrices of different knowledge distillation methods for predicting breast cancer molecular subtypes.

Similar content being viewed by others

Data availability

Some or all data, models, or code generated or used during the study are available from the corresponding author by request.

Code availability

Some or all data, models, or code generated or used during the study are available from the corresponding author by request.

References

  1. Zardavas D, Irrthum A, Swanton C, Piccart M. Clinical management of breast cancer heterogeneity. Nat Rev Clin Oncol. 2015;12:381–94.

    Article  CAS  PubMed  Google Scholar 

  2. Goldhirsch A, Wood WC, Coates AS, Gelber RD, Thürlimann B, Senn HJ. Strategies for subtypes-dealing with the diversity of breast cancer: highlights of the St. Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2011. Ann Oncol. 2011;22:1736–47.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Zhou BY, Wang LF, Yin HH, Wu TF, Ren TT, Peng C, et al. Decoding the molecular subtypes of breast cancer seen on multimodal ultrasound images using an assembled convolutional neural network model: a prospective and multicentre study. eBioMedicine. 2021;74:103684.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Yeo SK, Guan JL. Breast cancer: multiple subtypes within a tumor?. Trends Cancer. 2017;3:753–60.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Wang M, He X, Chang Y, Sun G, Thabane L. A sensitivity and specificity comparison of fine needle aspiration cytology and core needle biopsy in evaluation of suspicious breast lesions: a systematic review and meta-analysis. Breast. 2017;31:157–66.

    Article  PubMed  Google Scholar 

  6. Zhu JY, He HL, Jiang XC, Bao HW, Chen F. Multimodal ultrasound features of breast cancers: correlation with molecular subtypes. BMC Med Imaging. 2023;23:57.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Zhang L, Li J, Xiao Y, Cui H, Du G, Wang Y, et al. Identifying ultrasound and clinical features of breast cancer molecular subtypes by ensemble decision. Sci Rep. 2015;5:11085.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Yang S, Gao X, Liu L, Shu R, Yan J, Zhang G, et al. Performance and reading time of automated breast US with or without computer-aided detection. Radiology. 2019;292:540–9.

    Article  PubMed  Google Scholar 

  9. Li X, Zhang S, Zhang Q, Wei X, Pan Y, Zhao J, et al. Diagnosis of thyroid cancer using deep convolutional neural network models applied to sonographic images: a retrospective, multicohort, diagnostic study. Lancet Oncol. 2019;20:193–201.

    Article  PubMed  Google Scholar 

  10. Trepanier C, Huang A, Liu M, Ha R. Emerging uses of artificial intelligence in breast and axillary ultrasound. Clin Imaging. 2023;100:64–68.

    Article  PubMed  Google Scholar 

  11. Cuocolo R, Caruso M, Perillo T, Ugga L, Petretta M. Machine Learning in oncology: a clinical appraisal. Cancer Lett. 2020;481:55–62.

    Article  CAS  PubMed  Google Scholar 

  12. Jiang M, Zhang D, Tang SC, Luo XM, Chuan ZR, Lv WZ, et al. Deep learning with convolutional neural network in the assessment of breast cancer molecular subtypes based on US images: a multicenter retrospective study. Eur Radiol. 2021;31:3673–82.

    Article  PubMed  Google Scholar 

  13. Zhang X, Li H, Wang C, Cheng W, Zhu Y, Li D, et al. Evaluating the accuracy of breast cancer and molecular subtype diagnosis by ultrasound image deep learning model. Front Oncol. 2021;11:623506.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Zhao G, Kong D, Xu X, Hu S, Li Z, Tian J. Deep learning-based classification of breast lesions using dynamic ultrasound video. Eur J Radiol. 2023;165:110885.

    Article  PubMed  Google Scholar 

  15. Qiu S, Zhuang S, Li B, Wang J, Zhuang Z. Prospective assessment of breast lesions AI classification model based on ultrasound dynamic videos and ACR BI-RADS characteristics. Front Oncol. 2023;13:1274557.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Meng H, Lin Z, Yang F, Xu Y, Cui L. Knowledge distillation in medical data mining: a survey. 5th international conference on crowd science and engineering. ACM Digital Library. 2021:175–82.

  17. Xie X, Niu J, Liu X, Chen Z, Tang S, Yu S. A survey on incorporating domain knowledge into deep learning for medical image analysis. Med Image Anal. 2021;69:101985.

    Article  PubMed  Google Scholar 

  18. Canny J. A computational approach to edge detection. IEEE Trans Pattern Anal Mach Intell. 1986;8:679–98.

    Article  CAS  PubMed  Google Scholar 

  19. Kataoka H, Wakamiya T, Hara K, Satoh Y. Would mega-scale datasets further enhance spatiotemporal 3D CNNs? arXiv: 2004.04968 [Preprint] 2020. Available from https://arxiv.org/abs/2004.04968.

  20. Park W, Kim D, Lu Y, Cho M. Relational knowledge distillation. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE. 2019: 3962–71.

  21. Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-cam: visual explanations from deep networks via gradient-based localization. 2017 IEEE International Conference on Computer Vision (ICCV). IEEE. 2017:618–26.

  22. Tian Y, Krishnan D, Isola P. Contrastive representation distillation. arXiv: 1910.10699 [Preprint] 2019. Available from: https://arxiv.org/abs/1910.10699.

  23. Passalis N, Tefas A. Learning deep representations with probabilistic knowledge transfer. European Conference on Computer Vision. Cham: Springer International Publishing. 2018:283–99.

  24. He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE. 2016: 770–8.

  25. Luo WQ, Huang QX, Huang XW, Hu HT, Zeng FQ, Wang W. Predicting breast cancer in breast imaging reporting and data system (BI-RADS) ultrasound category 4 or 5 lesions: a nomogram combining radiomics and BI-RADS. Sci Rep. 2019;9:11921.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Zhang Y, Chen JH, Lin Y, Chan S, Zhou J, Chow D, et al. Prediction of breast cancer molecular subtypes on DCE-MRI using convolutional neural network with transfer learning between two centers. Eur Radiol. 2021;31:2559–67.

    Article  PubMed  Google Scholar 

  27. Mao N, Zhang H, Dai Y, Li Q, Lin F, Gao J, et al. Attention-based deep learning for breast lesions classification on contrast enhanced spectral mammography: a multicentre study. Br J Cancer. 2023;128:793–804.

    Article  PubMed  Google Scholar 

  28. de Margerie-Mellon C, Chassagnon G. Artificial intelligence: a critical review of applications for lung nodule and lung cancer. Diagn Interv Imaging. 2023;104:11–17.

    Article  PubMed  Google Scholar 

  29. Raja H, Akram MU, Shaukat A, Khan SA, Alghamdi N, Khawaja SG, et al. Extraction of retinal layers through convolution neural network (CNN) in an OCT image for glaucoma diagnosis. J Digital Imaging. 2020;33:1428–42.

    Article  Google Scholar 

  30. Musthafa MM, RM T, VK V, Guluwadi S. Enhanced skin cancer diagnosis using optimized CNN architecture and checkpoints for automated dermatological lesion classification. BMC Med Imaging. 2024;24:201.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Xu C, Coen-Pirani P, Jiang X. Empirical study of overfitting in deep learning for predicting breast cancer metastasis. Cancers. 2023;15:1969.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Zhang T, Tan T, Han L, Appelman L, Veltman J, Wessels R, et al. Predicting breast cancer types on and beyond molecular level in a multi-modal fashion. NPJ Breast Cancer. 2023;9:16.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Wang F, Casalino LP, Khullar D. Deep learning in medicine-promise, progress, and challenges. JAMA Intern Med. 2019;179:293–4.

    Article  PubMed  Google Scholar 

  34. Tan PH, Ellis I, Allison K, Brogi E, Fox SB, Lakhani S, et al. The 2019 World Health Organization classification of tumours of the breast. Histopathology. 2020;77:181–5.

    Article  PubMed  Google Scholar 

  35. Pölcher M, Braun M, Tischitz M, Hamann M, Szeterlak N, Kriegmair A, et al. Concordance of the molecular subtype classification between core needle biopsy and surgical specimen in primary breast cancer. Arch Gynecol Obstet. 2021;304:783–90.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

We thank the Harbin Medical University Cancer Hospital for the data support. We thank all those who helped us during the writing of this research.

Funding

Key Research and Development Project of Vanguard and Leading Goose in Zhejiang Province grant Nos. 2024C03069 (SLT) Medical Science and Technology Project of Zhejiang Province grant Nos. 2025KY564 (WYN) National Natural Science Foundation of China grant Nos. 82371984 (TJW).

Author information

Author notes

  1. These authors contributed equally: Yingnan Wu, Lei Zhou.

Authors and Affiliations

  1. Cancer Center, Department of Ultrasound Medicine, Zhejiang Provincial People’s Hospital (Affiliated People’s Hospital), Hangzhou Medical College, Hangzhou, Zhejiang, China

    Yingnan Wu, Yanqing Peng, Xiaoying Li, Yaoting Wang, Sutian Zhu, Chunjie Hou, Pei Du & Litao Sun

  2. Key Discipline of Zhejiang Province in Public Health and Preventive Medicine (First Class, Category A), Hangzhou Medical College, Hangzhou, Zhejiang, China

    Yingnan Wu

  3. School of Mathematical Sciences, Zhejiang University, Zijingang Campus, Hangzhou, Zhejiang, China

    Yingnan Wu

  4. Department of Ultrasound Medicine, the Second Affiliated Hospital of Harbin Medical University, Harbin, Heilongjiang, China

    Lei Zhou & Jiawei Tian

  5. Department of Ultrasound Medicine, Sichuan Provincial Maternity and Child Health Care Hospital, Chengdu, Sichuan, China

    Jing Zhao

  6. Cancer Center, Department of Clinical Engineering, Zhejiang Provincial People’s Hospital (Affiliated People’s Hospital), Hangzhou Medical College, Hangzhou, Zhejiang, China

    Lei Ling

  7. Department of General Surgery, the Second Hospital of Hebei Medical University, Shijiazhuang, Hebei, China

    Ying Wang

Authors
  1. Yingnan Wu
    View author publications

    Search author on:PubMed Google Scholar

  2. Lei Zhou
    View author publications

    Search author on:PubMed Google Scholar

  3. Jing Zhao
    View author publications

    Search author on:PubMed Google Scholar

  4. Yanqing Peng
    View author publications

    Search author on:PubMed Google Scholar

  5. Xiaoying Li
    View author publications

    Search author on:PubMed Google Scholar

  6. Yaoting Wang
    View author publications

    Search author on:PubMed Google Scholar

  7. Sutian Zhu
    View author publications

    Search author on:PubMed Google Scholar

  8. Chunjie Hou
    View author publications

    Search author on:PubMed Google Scholar

  9. Pei Du
    View author publications

    Search author on:PubMed Google Scholar

  10. Lei Ling
    View author publications

    Search author on:PubMed Google Scholar

  11. Ying Wang
    View author publications

    Search author on:PubMed Google Scholar

  12. Jiawei Tian
    View author publications

    Search author on:PubMed Google Scholar

  13. Litao Sun
    View author publications

    Search author on:PubMed Google Scholar

Contributions

YNW: Conceptualization, Data curation, Formal analysis, Investigation, Writing - original draft, Supervision, Writing - review & editing. LZ: Data curation, Investigation, Writing- original draft. JZ: Data curation, Resources. YQP: Data curation, Software, Validation, Visualization; XYL: Formal analysis. YTW: Data curation, Software, Validation, Visualization. STZ: Software, Validation, Visualization. CJH: Formal analysis. PD: Formal analysis. LL: Data curation, Resources. YW: Data curation, Resources; JWT: Funding acquisition, Supervision, Writing - review & editing. LTS: Methodology, Project administration, Supervision, Writing - review & editing.

Corresponding authors

Correspondence to Ying Wang, Jiawei Tian or Litao Sun.

Ethics declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

This study protocol was centrally approved by the institutional Clinical Research Ethics Committee of the Second Affiliated Hospital of Harbin Medical University (approval number: KY2016-127), which conforms to the ethical guidelines of the 1975 Declaration of Helsinki. As this was a retrospective cohort study and individually identifiable information was removed during retrospective collection, informed consent was waived.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wu, Y., Zhou, L., Zhao, J. et al. Relation knowledge distillation 3D-ResNet-based deep learning for breast cancer molecular subtypes prediction on ultrasound videos: a multicenter study. Br J Cancer 133, 1178–1188 (2025). https://doi.org/10.1038/s41416-025-03146-7

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Version of record:

  • Issue date:

  • DOI: https://doi.org/10.1038/s41416-025-03146-7

This article is cited by

Search

Advanced search

Quick links