MR Görüntülerinde Federasyonlu Öğrenme Kullanılarak Hepatosit Karsinomu - HCC Karaciğer Kanseri Tanısı

Öz

Son yıllarda, Federasyonlu Öğrenme, veri gizliliğini korurken merkezi olmayan veri kaynakları arasında makine öğrenimi modellerini eğitmek için güçlü bir paradigma olarak ortaya çıkmıştır. Bu çalışma, hepatosellüler karsinom vakalarının kontrastla geliştirilmiş görüntülerini sağlayan ATLAS veri setinden elde edilen Manyetik Rezonans Görüntüleme kullanılarak karaciğer tümörlerinin sınıflandırılması için bir Federasyonlu Öğrenme çerçevesi önermektedir. Federasyonlu ortamda Evrişimli Sinir Ağı, EfficientNet, MobileNetV3, ResNet50 ve VGG16 mimarileri kullanılarak karşılaştırmalı bir değerlendirme yapılmıştır. Bu modeller arasında, EfficientNet tabanlı Federasyonlu Öğrenme uygulaması, %93,75'lik bir doğruluk ve %99,19'luk bir ROC-AUC puanına ulaşarak üstün bir performans elde etmiştir. Sonuçlar, federasyonlu yaklaşımların hasta verilerinin gizliliğini sağlarken merkezi öğrenmeye benzer performans seviyelerine ulaşabileceğini göstermektedir. Bu çalışma, Federasyonlu Öğrenmenin hassas tıbbi görüntüleme görevlerinde uygulanabilirliğini vurgulamakta ve gizliliği koruyan işbirlikçi model geliştirme potansiyelini vurgulamaktadır. Gelecekteki çalışmalar, heterojen klinik ortamlarda gerçek dünya dağıtımını ve ölçeklenebilirliğini araştırabilir.

Anahtar Kelimeler

• Federasyonlu Öğrenme
• FedAvg
• Karaciğer Tümörü Sınıflandırması
• Evrişimli Sinir Ağları
• MRI

Diagnosis of Hepatocellular Carcinoma - HCC Liver Cancer Using Federated Learning on MR Images

Öz

In recent years, Federated Learning has emerged as a powerful paradigm for training machine learning models across decentralized data sources while preserving data privacy. This study proposes a Federated Learning framework for the classification of liver tumors using Magnetic Resonance Imaging obtained from the ATLAS dataset, which provides contrast-enhanced images of hepatocellular carcinoma cases. A comparative evaluation was performed utilizing Convolutional Neural Network, EfficientNet, MobileNetV3, ResNet50, and VGG16 architectures within the federated environment. Among these models, the Federated Learning implementation based on EfficientNet achieved superior performance, reaching an accuracy of 93.75% and a ROC-AUC score of 99.19%. The results demonstrate that federated approaches can attain performance levels comparable to centralized learning while ensuring patient data confidentiality. This study highlights the applicability of Federated Learning in sensitive medical imaging tasks and emphasizes its potential for privacy-preserving collaborative model development. Future work may explore real-world deployment and scalability across heterogeneous clinical settings.

Anahtar Kelimeler

• Federated Learning
• FedAvg
• Liver Tumor Classification
• Convolutional Neural Networks
• MRI

Kaynakça

1. 1. Singh, A. & Pandey, B. (2016). Diagnosis of liver disease by using least squares support vector machine approach. International Journal of Healthcare Information Systems and Informatics, 11(2), 62-75.
2. 2. Çaviş, T. & Arda, K.N. (2024). Advanced magnetic resonance imaging techniques in the diagnosis of liver diseases. The Turkish Journal of Current Gastroenterology, 26(3), 130-141.
3. 3. Chan, H.P., Samala, R.K., Hadjiiski, L.M. & Zhou, C. (2020). Deep learning in medical image analysis. In: Lee, G., Fujita, H. (eds) Deep Learning in Medical Image Analysis. Advances in Experimental Medicine and Biology, Springer, 181.
4. 4. Kwak, L. & Bai, H. (2023). The role of federated learning models in medical imaging. Radiology: Artificial Intelligence, 5(3), 1-2.
5. 5. Llovet, J.M., Kelley, R.K., Villanueva, A., Singal, A.G., Pikarsky, E., Roayaie, S., Lencioni, R., Koike, K., Rossi, J.Z. & Finn, R.S. (2021). Hepatocellular carcinoma. Nature Rev. Dis. Primers, 7(1), 6-34.
6. 6. Heimbach, J.K., Kulik, L.M., Finn, R.S., Sirlin, C.B., Abecassis, M.M., Roberts, L.R., Zhu, A.X., Murad, M.H. & Marrero J.A. (2018). AASLD guidelines for the treatment of hepatocellular carcinoma. Hepatology, 67(1), 358-380.
7. 7. Shao, Y.Y., Wang, S.Y. & Lin, S.M. (2021). Management consensus guideline for hepatocellular carcinoma: 2020 update on surveillance, diagnosis, and systemic treatment by the Taiwan liver cancer association and the gastroenterological society of Taiwan. J Formos Med Assoc., 120(4), 1051-1060.
8. 8. Chen, H., Gomez, C., Huang, C.M. & Unberath, M. (2022). Explainable medical imaging AI needs human-centered design: guidelines and evidence from a systematic review. npj Digital Medicine, 5(156), 1-15.

9. 9. Song, L., Geoffrey, K. & Kaijian, H. (2020). Bottleneck feature supervised U-net for pixel-wise liver and tumor segmentation. Expert Syst. Appl., 145(5), 1-11.
10. 10. Song, D., Wang, Y., Wang, W. & Wang, Y. (2021). Using deep learning to predict microvascular invasion in hepatocellular carcinoma based on dynamic contrast-enhanced MRI combined with clinical parameters. J. Cancer Res. Clin. Oncol., 147(12), 3757-3767.
11. 11. Srinivasu, P.N., Lakshmi, G.J., Narahari, S.C., Shafi, J., Choi, J. & Ijaz, M.F. (2024). Enhancing medical image classification via federated learning and pre-trained model. Egyptian Informatics Journal, 27(1), 1-16.
12. 12. Roth, H.R., Chang, K., Singh, P., Neumark, N., Li, W., Gupta, V., Gupta, S., Qu, L., Ihsani, A., Bizzo, B.C., Wen, Y., Buch, V., Shah, M., Kitamura, F., Mendonça, M., Lavor, V., Harouni, A., Compas, C., Tetreault, J., Dogra, P., Cheng, Y., Erdal, S., White, R., Hashemian, B., Schultz, T., Zhang, M., McCarthy, A., Yun, B.M., Sharaf, E., Hoebel, K.V., Patel, J.B., Chen, B., Ko, S., Leibovitz, E., Pisano, E.D., Coombs, L., Xu, D., Dreyer, K.J., Dayan, I., Naidu, R.C., Flores, M., Rubin, D. & Cramer, J.K. (2020). Federated learning for breast density classification: a real-world implementation. Domain Adaptation and Representation Transfer, and Distributed and Collaborative Learning, DART 2020, Peru.
13. 13. Bernecker, T., Peters, A., Schlett, C.L., Bamberg, F., Theis, F., Rueckert, D., Weib, J. & Albarqouni, S. (2022). FedNorm: modality-based normalization in federated learning for multi-modal liver segmentation. ArXiv preprint, abs/2205.11096, 1-21.
14. 14. Mahlool, D.H. & Abed, M.H. (2022). Distributed brain tumor diagnosis using a federated learning environment. Bulletin of Electrical Engineering and Informatics, 11(6), 3313-3321.
15. 15. Triverdi, N.K., Shukla, S., Tiwari, R.G., Agarwal, A.K. & Gautam, V. (2023). Liver cancer diagnosis with lightweight federated learning using identically distributed images. 12th International Conference on System Modeling & Advancement in Research Trends (SMART-2023), Moradabad, India.
16. 16. Chai, H., Huang, Y., Xu, L., Song, X., He, M. & Wang, Q. (2024). A decentralized federated learning-based cancer survival prediction method with privacy protection. Heliyon, 10(11), 1-11.
17. 17. Lusnig, L., Sagingalieva, A., Surmach, M., Protasevich, T., Michiu, O., McLoughlin, J., Mansell, C., Petris, G.D., Bonazza, D., Zanconati, F., Melnikov, A. & Cavalli, F. (2024). Hybrid quantum image classification and federated learning for hepatic steatosis diagnosis. Diagnostics, 14(5), 1-16.
18. 18. Shankar, P.U., Rahul, E.S., Rao, K.D., Satish, K., Kumar, U.D., Ravindra, D. & Subbarao, G. (2025). Liver disease prediction using federated learning. International Journal of Innovative Science and Research Technology, 10(4), 880-887.
19. 19. Balla Fofanah, A., Özbilge, E. & Kırsal, Y. (2023). Skin cancer recognition using compact deep convolutional neural network. Çukurova Üniversitesi Mühendislik Fakültesi Dergisi, 38(3), 787-797.
20. 20. Fırat, H. & Üzen, H. (2024). MR görüntülerinden alzheimer hastalığının sınıflandırılması için inception ve sıkma-uyarma ağı tabanlı derin öğrenme modeli. Çukurova Üniversitesi Mühendislik Fakültesi Dergisi, 39(2), 555-567.
21. 21. Quinton, F., Popoff, R., Presles, B., Leclerc, S., Meriaudeau, F., Nodari, G., Lopez, O., Pellegrinelli, J., Chevallier, O., Gignac, D., Vrigneaud, J.-M. & Alberini, J.-L. (2023). A tumour and liver automatic segmentation (ATLAS) dataset on contrast-enhanced magnetic resonance imaging for hepatocellular carcinoma. Data, 8(5), 1-9.
22. 22. Lotfinia, M., Tayebiarasteh, A., Samiei, S., Joodaki, M. & Arasteh, S.T. (2025). Boosting multi-demographic federated learning for chest x-ray analysis using general-purpose self-supervised representations. European Journal of Radiology Artificial Intelligence, 3(1), 1-13.
23. 23. Overman, T. & Klabjan, D. (2025). Continuous-time analysis of federated averaging. ArXiv preprint, abs/2501.18870, 1-25.
24. 24. Yasaka, K., Akai, H., Abe, O. & Kiryu, S. (2018). Deep learning with convolutional neural network for differentiation of liver masses at dynamic contrast-enhanced CT: A preliminary study. Radiology, 286(3), 887-896.
25. 25. Ahn, J.C., Qureshi, T.A., Singal, A.G., Li, D. & Yang, J.D. (2021). Deep learning in hepatocellular carcinoma: current status and future perspectives. World J Hepatol., 13(12), 2039-2051.
26. 26. American College of Radiology (ACR), (2021). Liver imaging reporting and data system (LI-RADS) v2018 manual. Retrieved from https://www.acr.org/Clinical-Resources/Clinical-Tools-and-Reference/Reporting-and-Data-Systems/LI-RADS, Access date: 16/07/2025.
27. 27. Choi, J.Y., Lee, J.M. & Sirlin, C.B. (2014). CT and MR imaging diagnosis and staging of hepatocellular carcinoma: Part II. Extracellular agents, hepatobiliary agents, and ancillary imaging features. Radiology, 273(1), 30-50.
28. 28. Zhu, L., Liu, Z. & Han, S. (2019). Deep leakage from gradients. 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, Canada.
29. 29. Shokri, R., Stronati, M., Song, C. & Shmatikov, V. (2017). Membership inference attacks against machine learning models. 38th IEEE Symposium on Security and Privacy (SP), San Jose, CA.