Enhanced Breast Cancer Detection Using Federated Learning With Explainable AI And Multimodal Data Integration
DOI:
https://doi.org/10.29284/9fhefm39Keywords:
Breast Cancer Detection, Federated Learning, Explainable AI, Multimodal Integration, Tumor Segmentation, Convolutional Neural Networks (CNN), Grad-CAM, Privacy-Preserving Machine Learning, AUC-ROC, Clinical Diagnostics.Abstract
Breast cancer remains a leading cause of mortality among women worldwide, necessitating advancements in early detection methods. This study proposes a novel framework for breast cancer detection leveraging federated learning (FL), explainable AI (XAI), and multimodal data integration. The methodology addresses critical challenges, including privacy concerns, model interpretability, and the effective use of diverse data modalities. Utilizing public datasets such as CBIS-DDSM and BCDR, along with private institutional data, the study integrates imaging (mammograms, ultrasounds) and clinical/genetic data to enhance diagnostic precision. Experimental results highlight the superiority of the proposed approach. Federated CNNs achieve an accuracy of 93.5%, surpassing state-of-the-art methods like Shen (89.2%) and Wu (91.3%). Additionally, the model demonstrates an AUC-ROC of 0.96, precision of 0.91, and recall of 0.95, reflecting significant improvements in prediction reliability. Explainable AI techniques such as Grad-CAM provide interpretability, enabling visualization of critical tumor regions. Moreover, segmentation using U-Net achieves a dice coefficient of 0.89 , ensuring precise tumor delineation. The proposed framework's federated architecture preserves data privacy by aggregating locally trained models across institutions, addressing data-sharing barriers while maintaining performance. Multimodal integration further enhances predictions by incorporating genetic and clinical data. The framework also demonstrates scalability and robustness across diverse datasets, with consistent results across institutions (accuracy range: 92-94%). This study demonstrates a significant leap in breast cancer diagnostic technologies by combining federated learning with interpretable and holistic approaches, paving the way for clinically viable, privacy-preserving diagnostic systems.
References
[1] Dhungel, N., Carneiro, G., & Bradley, A. P. "Deep learning and structured prediction for the segmentation of mass in mammograms." IEEE International Conference on Medical Image Computing and Computer-Assisted Intervention, 2015, pp. 605-612.
[2] Shen, L., Margolies, L. R., Rothstein, J. H., Fluder, E., McBride, R., & Sieh, W. "Deep learning to improve breast cancer detection on screening mammography." Scientific Reports, vol. 9, no. 1, 2019, pp. 12495.
[3] Wu, Y., Yang, J., Xu, X., Liu, J., & Yin, Y. "A hybrid deep transfer learning model with machine learning methods for ultrasonic breast cancer diagnosis." BMC Medical Imaging, vol. 20, no. 1, 2020, pp. 21.
[4] Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. "Dermatologist-level classification of skin cancer with deep neural networks." Nature, vol. 542, no. 7639, 2017, pp. 115118.
[5] Tahmasbi, A., Saki, F., & Shokouhi, S. B. "Mass diagnosis in mammography images using novel features and a hierarchical classification scheme." Biomedical Signal Processing and Control, vol. 6, no. 3, 2011, pp. 321-329.
[6] Zhu, W., Lou, Q., Vang, Y. S., & Xie, X. "Deep multi-instance networks with sparse label assignment for whole mammogram classification." IEEE Transactions on Medical Imaging, vol. 38, no. 2, 2019, pp. 669-679.
[7] Ghafoorian, M., Karssemeijer, N., Heskes, T., Majoie, C. B., Sánchez, C. I., & Platel, B. "Deep multiscale location-aware 3D convolutional neural networks for automated detection of lacunes of presumed vascular origin." NeuroImage: Clinical, vol. 14, 2018, pp. 391-399.
[8] Li, H., Zhu, Y., Burnside, E. S., Drukker, K., Hoadley, K. A., Fan, C., ... & Giger, M. L. "MR imaging radiomics signatures for predicting the risk of breast cancer recurrence as given by research versions of clinical pathologic guidelines." Radiology, vol. 294, no. 3, 2020, pp. 491-501.
[9] Shen, L., Wu, H., & Han, S. "Improving diagnostic accuracy in breast cancer detection using transfer learning techniques." Scientific Reports, vol. 9, no. 1, 2020, pp. 10452.
[10] Wu, P., Yuan, Z., & Zhang, H. "Application of U-Net-based deep learning in breast cancer segmentation tasks." BMC Cancer, vol. 21, no. 1, 2021, pp. 62.
[11] Swamy, R. S., S. C. Kumar, and G. A. Latha. "An efficient skin cancer prognosis strategy using deep learning techniques." Indian Journal of Computer Science and Engineering (IJCSE) 12.1 (2021).
[12] Esteva, A., Thrun, S., & Blau, H. M. "AI-based diagnostic tools for early cancer detection." Nature Reviews Cancer, vol. 20, no. 1, 2020, pp. 23-37.
[13] Sirisati RS, Kumar CS, Latha AG, Kumar BN, a Rao KS. An enhanced multi layer neural network to detect early cardiac arrests. In2021 5th International Conference on Electronics, Communication and Aerospace Technology (ICECA) 2021 Dec 2 (pp. 1514-1518). IEEE.
[14] Zhu, X., Zhang, L., & Yu, Z. "Sparse deep neural networks for breast cancer prediction." IEEE Transactions on Biomedical Engineering, vol. 66, no. 2, 2019, pp. 322-331.
[15] Sirisati RS, Kumar CS, Venuthurumilli P, Ranjith J, Rao KS. Cancer sight: Illuminating the hidden-advancing breast cancer detection with machine learning-based image processing techniques. In2023 International Conference on Sustainable Communication Networks and Application (ICSCNA) 2023 Nov 15 (pp. 1618-1625). IEEE.
[16] Dhungel, N., Carneiro, G., & Bradley, A. P. "Structured prediction models for breast cancer lesion segmentation." IEEE Transactions on Medical Imaging, vol. 34, no. 4, 2015, pp. 1010-1025.
[17] Bulla, Suneetha, et al. "A secure new HRF mechanism for mitigate EDoS attacks." International Journal of Ad Hoc and Ubiquitous Computing 40.1-3 (2022): 20-29.
[18] Ghafoorian, M., Platel, B., & Sánchez, C. I. "Deep learning models for automated detection of vascular lesions." NeuroImage: Clinical, vol. 17, 2018, pp. 391-401.
[19] SwamySirisati, R., Rao, M. S., & Thonukunuri, S. (2020, November). Analysis of Hybrid Fusion-Neural Filter Approach to detect Brain Tumor. In 2020 Sixth International Conference on Parallel, Distributed and Grid Computing (PDGC) (pp. 460-464). IEEE.
[20] Srinivasa Rao, P. S. V., Mekala Srinivasa Rao, and Ranga Swamy Sirisati. "A Hybrid Clinical Data Predication Approach Using Modified PSO." Smart Computing Techniques and Applications: Proceedings of the Fourth International Conference on Smart Computing and Informatics, Volume 2. Singapore: Springer Singapore, 2021.
[21] Swamy, Sirisati Ranga, et al. "Multi-Features Disease Analysis Based Smart Diagnosis for COVID-19." Comput. Syst. Sci. Eng. 45.1 (2023): 869-886.
[22] Wu, C., Deng, W., & Sun, Y. "Improving interpretability in breast cancer diagnostics using Grad-CAM techniques." Medical Image Analysis, vol. 50, 2021, pp. 15-27.
[23] Swamy, S. Ranga, et al. "Dimensionality reduction using machine learning and big data technologies." Int. J. Innov. Technol. Explor. Eng.(IJITEE) 9.2 (2019): 1740-1745.
[24] Li, X., Zhao, H., & Wang, T. "Federated learning frameworks for medical image analysis." IEEE Access, vol. 8, 2020, pp. 10154-10165.
[25] Kumar, C. Srinivasa, et al. "A smart recommendation system for medicine using intelligent NLP techniques." 2022 International Conference on Automation, Computing and Renewable Systems (ICACRS). IEEE, 2022.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
