Deep Learning in Breast Cancer Molecular Subtyping: Recent Advances and Clinical Translation

Authors

  • Yiying Cao
  • Yanshan Ouyang
  • Yingning Wu
  • Junneng Yang

DOI:

https://doi.org/10.54097/rn3wbj06

Keywords:

Breast Neoplasms, Radiomics, Deep Learning, Molecular Subtyping

Abstract

Breast cancer, as the second most prevalent malignancy among women in China, necessitates molecular subtyping for precision treatment decisions. While traditional biopsies face limitations due to invasiveness and spatial heterogeneity, deep learning demonstrates significant potential in enhancing non-invasive subtype prediction through automated feature extraction and multimodal data fusion. This review comprehensively summarizes recent advances in radiomics research and DL applications across mammography, MRI, and ultrasound imaging for predicting breast cancer molecular subtypes. It critically evaluates the strengths and limitations of existing methodologies, providing a reference framework for future DL-based research and clinical implementation of non-invasive subtyping approaches.

Downloads

Download data is not yet available.

References

[1] ZHENG R S,CHEN R,HAN B F, et al.Cancer incidence and mortality in China, 2022[J].Chinese Journal of Oncology, 2024, 46 (03): 221-231. DOI:10.3760/cma.j.cn112152-2024 0119- 00035

[2] Li Y, Mo H, Guan X, et al. Real world initial palliative treatment patterns and clinical outcomes in premenopausal patients with hormone receptor-positive, HER2-negative metastatic breast cancer: A study of the National Cancer Center, China[J]. Breast, 2022,61:129-135.DOI:10. 1016/j. breast. 2021.12.017.

[3] Perou C M, Sorlie T, Eisen M B, et al. Molecular portraits of human breast tumours[J]. Nature, 2000,406(6797):747-752. DOI:10.1038/35021093.

[4] Sorlie T, Tibshirani R, Parker J, et al. Repeated observation of breast tumor subtypes in independent gene expression data sets [J]. Proc Natl Acad Sci U S A, 2003,100(14):8418-8423. DOI:10.1073/pnas.0932692100.

[5] Wahba H A, El-Hadaad H A. Current approaches in treatment of triple-negative breast cancer[J]. Cancer Biol Med, 2015,12(2): 106-116.DOI:10.7497/j.issn.2095-3941. 2015. 0030.

[6] Weiss A, King T A, Hunt K K, et al. Incorporating Biologic Factors into the American Joint Committee on Cancer Breast Cancer Staging System: Review of the Supporting Evidence[J]. Surg Clin North Am, 2018,98(4):687-702.DOI: 10. 1016/j.suc. 2018.03.005.

[7] Li X, Zhang Y, Gong J, et al. Development of a breast cancer invasion score to predict tumor aggressiveness and prognosis via PI3K/AKT/mTOR pathway analysis[J]. Cell Death Discov, 2025,11(1): 157. DOI:10.1038/s41420-025-02422-y.

[8] Litjens G, Kooi T, Bejnordi B E, et al. A survey on deep learning in medical image analysis[J]. Med Image Anal, 2017,42: 60-88.DOI:10.1016/j.media.2017.07.005.

[9] Garberis I, Gaury V, Saillard C, et al. Deep learning assessment of metastatic relapse risk from digitized breast cancer histological slides[J]. Nat Commun, 2025,16(1):5876. DOI:10. 1038/ s41467-025-60824-z.

[10] 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 [J]. Br J Cancer, 2025.DOI:10.1038/s41416-025-03146-7.

[11] Morano J, Aresta G, Grechenig C, et al. Deep Multimodal Fusion of Data With Heterogeneous Dimensionality via Projective Networks[J]. IEEE J Biomed Health Inform, 2024, 28 (4): 2235-2246.DOI:10.1109/JBHI.2024.3352970.

[12] Qi Y J, Su G H, You C, et al. Radiomics in breast cancer: Current advances and future directions[J]. Cell Rep Med, 2024,5(9): 101719.DOI:10.1016/j.xcrm.2024.101719.

[13] Hu Y, Xie C, Yang H, et al. Computed tomography-based deep-learning prediction of neoadjuvant chemoradiotherapy treatment response in esophageal squamous cell carcinoma[J]. Radiother Oncol, 2021,154:6-13.DOI:10.1016/j. radonc. 2020. 09. 014.

[14] Si K, Xue Y, Yu X, et al. Fully end-to-end deep-learning-based diagnosis of pancreatic tumors[J]. Theranostics, 2021,11(4): 1982-1990.DOI:10.7150/thno.52508.

[15] Lotter W, Diab A R, Haslam B, et al. Robust breast cancer detection in mammography and digital breast tomosynthesis using an annotation-efficient deep learning approach[J]. Nat Med, 2021,27(2):244-249. DOI:10.1038/s41591-020-01174-9.

[16] Teichgraeber D C, Guirguis M S, Whitman G J. Breast Cancer Staging: Updates in the AJCC Cancer Staging Manual, 8th Edition, and Current Challenges for Radiologists, From the AJR Special Series on Cancer Staging[J]. AJR Am J Roentgenol, 2021,217(2):278-290. DOI:10. 2214/ AJR. 20. 25 223.

[17] Cozzi S, Finocchi G S, Tava F, et al. Radiation-Associated Angiosarcoma of the Breast: The State of the Art of a Rare and Aggressive Disease[J]. J Pers Med, 2024,14(8). DOI: 10. 3390/ jpm14080859.

[18] Liu Z, Yao B, Wen J, et al. Voxel-wise mapping of DCE-MRI time-intensity-curve profiles enables visualizing and quantifying hemodynamic heterogeneity in breast lesions[J]. Eur Radiol, 2024,34(1): 182-192.DOI:10.1007/s00330-023-10102- 7.

[19] Zhang Y, Zhang H, Zhang H, et al. Glioblastoma and Solitary Brain Metastasis: Differentiation by Integrating Demographic-MRI and Deep-Learning Radiomics Signatures [J]. J Magn Reson Imaging, 2024,60(3):909-920.DOI: 10. 1002/ jmri.29123.

[20] Lv J, Wu L, Hong C, et al. Multi-class brain malignant tumor diagnosis in magnetic resonance imaging using convolutional neural networks[J]. Brain Res Bull, 2025,225: 111329. DOI: 10. 1016/j.brainresbull.2025.111329.

[21] CHENG C,LI H E,GU Y, et al. The establishment and verification of a model for identifying TNBC based on the deep learning radiomics nomogram by combining ultrasound with MRI[J]. Radiologic Practice,2025,40(03):339-348.DOI:10. 13609/ j.cnki.1000-0313.2025.03.009.

[22] Yoon J, Lee H S, Kim M J, et al. AI-CAD for differentiating lesions presenting as calcifications only on mammography: outcome analysis incorporating the ACR BI-RADS descriptors for calcifications[J]. Eur Radiol, 2022,32(10):6565-6574. DOI: 10.1007/s00330-022-08961-7.

[23] MA M M,CUI Y P,LI X, et al.Mammography-based radiomic analysis for predicting benign and malignant of breast mass and calcification[J]. Radiologic Practice,2022,37(03):325-330.DOI: 10. 13609/j.cnki.1000-0313.2022.03.008.

[24] Mota A M, Mendes J, Matela N. Breast Cancer Molecular Subtype Prediction: A Mammography-Based AI Approach[J]. Biomedicines, 2024,12(6): 1371.DOI: 10.3390/ biomedicines 12061371.

[25] LIU S T,YU Z,LI Z R, et al.Deep learning of mammograms combined with nomogram of clinical characteristics for predicting human epidermal growth factor receptor-2(HER-2) expression status of breast cancer[J]. Chinese Journal of Medical Imaging Technology,2023,39(11):1659-1664. DOI:10. 13929/j. issn.1003-3289.2023.11.014.

[26] GONG Y,JIANG Y N,LIU T T, et al.The Predictive Value of Deep Learning Models Based on FFDM and DBT Features for HER2 Expression Status in Invasive Breast Cancer[J].Journal of Clinical Radiology,2025,44(07):1203-1210.DOI:10.13437/j. cnki.jcr.2025.07.025.

[27] ZHANG Y X.The study of radiomics and deep learning based contrast-enhanced spectral mammography for prediction of triple-negative breast cancer and pathological features[D]. Shandong University,2022. DOI:10. 27272/d. cnki. gshdu. 2022. 006787.

[28] Cheng X, Chen C, Xia H, et al. 3.0 T Magnetic Resonance Functional Imaging Quantitative Parameters for Differential Diagnosis of Benign and Malignant Lesions of the Breast[J]. Cancer Biother Radiopharm, 2021,36(6):448-455.DOI:10. 1089/ cbr.2019.3040.

[29] Ma W, Yang L, Zhang Y, et al. Heterogeneity Assessment of Breast Cancer Tumor Microenvironment: Multiparametric Quantitative Analysis with DCE-MRI and Discovery of Radiomics Biomarkers[J]. Breast Cancer (Dove Med Press), 2025,17: 573-581.DOI:10.2147/BCTT.S530834.

[30] Chan S W, Lin C A, Ouyang Y C, et al. Characterizing Breast Tumor Heterogeneity Through IVIM-DWI Parameters and Signal Decay Analysis[J]. Diagnostics (Basel), 2025,15 (12). DOI: 10.3390/diagnostics15121499.

[31] Zhang J, Huang Y, Chen J, et al. Potential of combination of DCE-MRI and DWI with serum CA125 and CA199 in evaluating effectiveness of neoadjuvant chemotherapy in breast cancer[J]. World J Surg Oncol, 2021,19(1): 284. DOI: 10. 1186/s12957-021-02398-w.

[32] Ha R, Mutasa S, Karcich J, et al. Predicting Breast Cancer Molecular Subtype with MRI Dataset Utilizing Convolutional Neural Network Algorithm[J]. Journal of Digital Imaging, 2019,32(2): 276-282.DOI:10.1007/s10278-019-00179-2.

[33] Zhang Y, Chen J, Lin Y, et al. Prediction of breast cancer molecular subtypes on DCE-MRI using convolutional neural network with transfer learning between two centers[J]. European Radiology, 2021,31(4):2559-2567.DOI: 10.1007/ s00330-020-07274-x.

[34] Ba Z, Zhang H, Liu A, et al. Combination of DCE-MRI and NME-DWI via Deep Neural Network for Predicting Breast Cancer Molecular Subtypes[J]. Clinical Breast Cancer, 2024,24(5): e417-e427.DOI:10.1016/j.clbc.2024.03.006.

[35] Yu T, Yu R, Liu M, et al. Integrating intratumoral and peritumoral radiomics with deep transfer learning for DCE-MRI breast lesion differentiation: A multicenter study comparing performance with radiologists[J]. European Journal of Radiology, 2024,177: 111556. DOI:10. 1016/j. ejrad. 2024. 111556.

[36] Modi S, Jacot W, Yamashita T, et al. Trastuzumab Deruxtecan in Previously Treated HER2-Low Advanced Breast Cancer[J]. N Engl J Med, 2022,387(1):9-20.DOI:10. 1056/ NEJMoa 220 3690.

[37] Li Q Y, Liang Y, Zhang L, et al. MRI-based habitat analysis for Intratumoral heterogeneity quantification combined with deep learning for HER2 status prediction in breast cancer[J]. Magn Reson Imaging, 2025,122:110429.DOI: 10.1016/j. mri. 2025. 110429.

[38] Park S Y, Kang B J. Combination of shear-wave elastography with ultrasonography for detection of breast cancer and reduction of unnecessary biopsies: a systematic review and meta-analysis[J]. Ultrasonography, 2021,40(3):318-332.DOI: 10. 14366/ usg.20058.

[39] Park C, Trumpour T, Aziz A, et al. Cost-effective, portable, patient-dedicated three-dimensional automated breast ultrasound for point-of-care breast cancer screening[J]. Sci Rep, 2023,13(1): 14390. DOI:10.1038/s41598-023-41424-7.

[40] Martello S E, Xia J, Kusunose J, et al. Ultrafast power doppler ultrasound enables longitudinal tracking of vascular changes that correlate with immune response after radiotherapy[J]. Theranostics, 2024,14(18):6883-6896.DOI:10.7150/ thno. 977 59.

[41] Jiang M, Zhang D, Tang S, et al. Deep learning with convolutional neural network in the assessment of breast cancer molecular subtypes based on US images: a multicenter retrospective study[J]. European Radiology, 2021,31(6):3673-3682. DOI:10.1007/s00330-020-07544-8.

[42] Boulenger A, Luo Y, Zhang C, et al. Deep learning-based system for automatic prediction of triple-negative breast cancer from ultrasound images[J]. Medical & Biological Engineering & Computing, 2023,61(2):567-578. DOI:10.1007/s11517-022-02728-4.

[43] Zhou B, Wang L, Yin H, 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[J]. eBioMedicine, 2021,74: 103684. DOI:10.1016/j.ebiom.2021.103684.

[44] Qian X, Pei J, Han C, et al. A multimodal machine learning model for the stratification of breast cancer risk[J]. Nat Biomed Eng, 2025,9(3):356-370. DOI:10.1038/s41551-024-01302-7.

[45] Huang J, Shi J, Ding S, et al. Deep Learning Model Based on Dual-Modal Ultrasound and Molecular Data for Predicting Response to Neoadjuvant Chemotherapy in Breast Cancer[J]. Academic Radiology, 2023,30:S50-S61.DOI:10. 1016/j. acra. 2023. 03.036.

[46] Wang J, Gao W, Lu M, et al. Development of an interpretable machine learning model for Ki-67 prediction in breast cancer using intratumoral and peritumoral ultrasound radiomics features[J]. Front Oncol, 2023,13: 1290313. DOI: 10. 3389/ fonc. 2023.1290313.

[47] Raptis S, Ilioudis C, Theodorou K. From pixels to prognosis: unveiling radiomics models with SHAP and LIME for enhanced interpretability[J]. Biomed Phys Eng Express, 2024, 10(3). DOI:10.1088/2057-1976/ad34db.

[48] Bauerle T, Dietzel M, Pinker K, et al. Identification of impactful imaging biomarker: Clinical applications for breast and prostate carcinoma[J]. Rofo, 2024,196(4):354-362. DOI: 10. 1055/a-2175-4446.

[49] Crombé A, Kataoka M. Breast cancer molecular subtype prediction: Improving interpretability of complex machine-learning models based on multiparametric-MRI features using SHapley Additive exPlanations (SHAP) methodology[J]. Diagnostic and Interventional Imaging, 2024,105(5):161-162. DOI:10.1016/j.diii.2024.01.008.

[50] Zhou Y, Lin G, Chen W, et al. Multiparametric MRI-based Interpretable Machine Learning Radiomics Model for Distinguishing Between Luminal and Non-luminal Tumors in Breast Cancer: A Multicenter Study[J]. Academic Radiology, 2025,32(7): 3801-3812.DOI:10.1016/j.acra.2025.03.010.

[51] Liu W, Yang Y, Wang X, et al. A Comprehensive Model Outperformed the Single Radiomics Model in Noninvasively Predicting the HER2 Status in Patients with Breast Cancer[J]. Acad Radiol, 2025,32(1):24-36.DOI:10.1016/j. acra. 2024. 07. 048.

[52] Xie X, Zhou H, Ma M, et al. A Deep Learning Model for Predicting Molecular Subtype of Breast Cancer by Fusing Multiple Sequences of DCE-MRI From Two Institutes[J]. Acad Radiol, 2024,31(9):3479-3488.DOI: 10.1016/j. acra. 2024. 03.002.

[53] Zheng S, Yang Z, Du G, et al. Discrimination between HER2-overexpressing, -low-expressing, and -zero-expressing statuses in breast cancer using multiparametric MRI-based radiomics[J]. Eur Radiol, 2024,34(9):6132-6144. DOI: 10.1007/ s00330-024-10641-7.

[54] Chen Y, Chen S, Tang W, et al. Multiparametric MRI Radiomics With Machine Learning for Differentiating HER2-Zero, -Low, and -Positive Breast Cancer: Model Development, Testing, and Interpretability Analysis[J]. AJR Am J Roentgenol, 2025,224(1): e2431717. DOI:10. 2214/ AJR. 24.31717.

[55] Chen H, Liu Y, Zhao J, et al. Quantification of intratumoral heterogeneity using habitat-based MRI radiomics to identify HER2-positive, -low and -zero breast cancers: a multicenter study[J]. Breast Cancer Res, 2024,26(1):160. DOI: 10. 1186/ s13058-024-01921-7.

Downloads

Published

29-10-2025

Issue

Section

Articles

How to Cite

Cao, Y., Ouyang, Y., Wu, Y., & Yang, J. (2025). Deep Learning in Breast Cancer Molecular Subtyping: Recent Advances and Clinical Translation. International Journal of Biology and Life Sciences, 12(2), 47-53. https://doi.org/10.54097/rn3wbj06