A hybrid approach for enhancing breast cancer classification using attention-enhanced ResNet-34 with SVM

Breast cancer remains one of the leading causes of death among women worldwide, and microcalcifications on mammograms play a key role in the early detection of malignant neoplasms. Despite significant progress in the field of computer-aided analysis of medical images, accurate automatic classification of microcalcifications remains a challenge due to the high variability of their morphology and visual features. Microcalcifications, small calcium deposits that appear as bright point structures on mammograms, play an important role in the early detection of the disease. In this paper, we propose a novel hybrid model combining the ResNet-34 architecture supplemented with a convolutional block attention module (CBAM) and a support vector machine (SVM) classifier with a radial basis kernel. The attention module allows us to highlight the most informative spatial regions and feature channels, while the SVM provides high generalization ability even with a limited amount of data. Experiments on the CBIS-DDSM dataset showed that the proposed approach outperforms both the standard ResNet-34 and its hybrid with SVM in accuracy, sensitivity, specificity, and noise robustness. The proposed model achieves 97.47 % accuracy, 96.56 % sensitivity, and 95.17 % specificity, while ResNet-34 achieves 91.63 %, 92.80 %, and 92.87 %, and ResNet-34 with SVM achieves 96.75 %, 94.10 %, and 95.20 %, respectively.

1. Kim J., Harper A., McCormack V., et al. Global Patterns and Trends in Breast Cancer Incidence and Mortality Across 185 Countries. Nature Medicine. 2025;31:1154–1162. https://doi.org/10.1038/s41591-025-03502-3

2. Araújo T., Aresta G., Castro E., et al. Classification of Breast Cancer Histology Images Using Convolutional Neural Networks. PloS ONE. 2017;12(6). https://doi.org/10.1371/journal.pone.0177544

3. Atrey K., Singh B.K., Bodhey N.K., Pachori R.B. Mammography and Ultrasound Based Dual Modality Classification of Breast Cancer Using a Hybrid Deep Learning Approach. Biomedical Signal Processing and Control. 2023;86. https://doi.org/10.1016/j.bspc.2023.104919

4. Varshney T., Verma K., Kaur A., Puri S.K. Hybrid and Optimized Feature Fusion for Enhanced Breast Cancer Classification. Network Modeling Analysis in Health Informatics and Bioinformatics. 2025;14. https://doi.org/10.1007/s13721-025-00573-7

5. Khairandish M.O., Sharma M., Jain V., Chatterjee J.M., Jhanjhi N.Z. A Hybrid CNN-SVM Threshold Segmentation Approach for Tumor Detection and Classification of MRI Brain Images. IRBM. 2022;43(4):290–299. https://doi.org/10.1016/j.irbm.2021.06.003

6. Murty P.S.R.Ch., Anuradha Ch., Naidu P.A., et al. Integrative Hybrid Deep Learning for Enhanced Breast Cancer Diagnosis: Leveraging the Wisconsin Breast Cancer Database and the CBIS-DDSM Dataset. Scientific Reports. 2024;14. https://doi.org/10.1038/s41598-024-74305-8

7. Chugh G., Kumar Sh., Singh N. Survey on Machine Learning and Deep Learning Applications in Breast Cancer Diagnosis. Cognitive Computation. 2021;13(6):1451–1470. https://doi.org/10.1007/s12559-020-09813-6

8. Sarker I.H. Machine Learning: Algorithms, Real-World Applications and Research Directions. SN Computer Science. 2021;2. https://doi.org/10.1007/s42979-021-00592-x

9. Yurdusev A.A., Adem K., Hekim M. Detection and Classification of Microcalcifications in Mammograms Images Using Difference Filter and YOLOv4 Deep Learning Model. Biomedical Signal Processing and Control. 2023;80. https://doi.org/10.1016/j.bspc.2022.104360

10. Resmini R., Silva L., Araujo A.S., Medeiros P., Muchaluat-Saade D., Conci A. Combining Genetic Algorithms and SVM for Breast Cancer Diagnosis Using Infrared Thermography. Sensors. 2021;21(14). https://doi.org/10.3390/s21144802

11. Jalloul R., Chethan H.K., Alkhatib R. A Review of Machine Learning Techniques for the Classification and Detection of Breast Cancer from Medical Images. Diagnostics. 2023;13(14). https://doi.org/10.3390/diagnostics13142460

12. Liao L., Aagaard E.M. An Open Codebase for Enhancing Transparency in Deep Learning-Based Breast Cancer Diagnosis Utilizing CBIS-DDSM Data. Scientific Reports. 2024;14. https://doi.org/10.1038/s41598-024-78648-0

13. Zarif S., Abdulkader H., Elaraby I., Alharbi A., Elkilani W.S., Pławiak P. Using Hybrid Pre-Trained Models for Breast Cancer Detection. PLoS ONE. 2024;19(1). https://doi.org/10.1371/journal.pone.0296912

14. Hernández-Vázquez M.A., Hernández-Rodríguez Ya.M., Cortes-Rojas F.D., Bayareh-Mancilla R., Cigarroa-Mayorga O.E. Hybrid Feature Mammogram Analysis: Detecting and Localizing Microcalcifications Combining Gabor, Prewitt, GLCM Features, and Top Hat Filtering Enhanced with CNN Architecture. Diagnostics. 2024;14(15). https://doi.org/10.3390/diagnostics14151691

15. Afrifa S., Varadarajan V., Appiahene P., Zhang T. A Novel Artificial Intelligence Techniques for Women Breast Cancer Classification Using Ultrasound Images. Clinical and Experimental Obsterics and Gynecology. 2023;50(12). https://doi.org/10.31083/j.ceog5012271

16. Guizani S., Guizani N., Gharsallaoui S. A Hybrid CNN-SVM Prediction Approach for Breast Cancer Ultrasound Imaging. In: 2023 International Wireless Communications and Mobile Computing (IWCMC), 19–23 June 2023, Marrakesh, Morocco. IEEE; 2023. P. 1574–1578. https://doi.org/10.1109/IWCMC58020.2023.10182874