Journals / Afyon Kocatepe Üniversitesi Fen ve Mühendislik Bilimleri Dergisi / 2026 / Cilt: 26 Sayı: 2

Decentralized Disease Prediction: A Federated Learning Perspective

Merkeziyetsiz Hastalık Tahmini: Federe Öğrenme Yaklaşımı

Abdülbaki Demir(MARMARA ÜNİVERSİTESİ)0009-0002-7603-9422
Büşra Büyüktanır(MARMARA ÜNİVERSİTESİ, TEKNOLOJİ FAKÜLTESİ, BİLGİSAYAR MÜHENDİSLİĞİ BÖLÜMÜ)0000-0003-2571-4029
Gozde Karatas Baydogmus(MARMARA ÜNİVERSİTESİ, TEKNOLOJİ FAKÜLTESİ, BİLGİSAYAR MÜHENDİSLİĞİ BÖLÜMÜ)0000-0003-2303-9410
Kazım Yıldız(MARMARA ÜNİVERSİTESİ, TEKNOLOJİ FAKÜLTESİ, BİLGİSAYAR MÜHENDİSLİĞİ BÖLÜMÜ)0000-0001-6999-1410
Pages
337–353
Publication Date
2026-03-27
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

Notably, the duCBA algorithm delivers stable and consistent accuracy (67–71%) across different feature selection methods and varying client numbers, particularly excelling in scenarios with fewer clients.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Machine learning has gained great importance in the healthcare industry due to its ability to extract meaningful information from large datasets.
📊
Key Findings & Data:To address these challenges, this study employs a federated learning-based aggregation algorithm, Data Unaware Classification Based on Association (duCBA), for disease prediction tasks.
💡
Conclusion & Impact:Notably, the duCBA algorithm delivers stable and consistent accuracy (67–71%) across different feature selection methods and varying client numbers, particularly excelling in scenarios with fewer clients.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Federe Öğrenme · İlişkisel Sınıflama · duCBA

Abstract

Machine learning has gained great importance in the healthcare industry due to its ability to extract meaningful information from large datasets. Nonetheless, the sensitive nature of health data presents serious privacy concerns, limiting the feasibility of centralized data collection and the applicability of conventional machine learning algorithms. To address these challenges, this study employs a federated learning-based aggregation algorithm, Data Unaware Classification Based on Association (duCBA), for disease prediction tasks. Model training was conducted using two publicly available health datasets: Diabetes Health Indicators and Heart Attack Prediction. Prior to training, feature selection techniques including Mutual Information, Chi-Square, and Principal Component Analysis (PCA) were applied to generate three distinct datasets. The system architecture comprises a central server, an API server, and multiple clients. Clients locally train models on their own data without sharing it externally. The API server collects these locally trained models and transmits them to the central server. The central server then aggregates them using the duCBA algorithm and disseminates the updated global model back to the clients. This research not only demonstrates the practical viability of duCBA in federated settings but also compares its performance against the widely-used Federated Averaging (FedAvg) algorithm. Experimental results reveal that duCBA achieves comparable performance to FedAvg while offering improved model transparency in line with Explainable Artificial Intelligence (XAI) principles. Notably, the duCBA algorithm delivers stable and consistent accuracy (67–71%) across different feature selection methods and varying client numbers, particularly excelling in scenarios with fewer clients.

Özet

Makine öğrenmesi, büyük ölçekli verilerden anlamlı bilgiler çıkarma yeteneği sayesinde sağlık sektöründe giderek artan bir öneme sahip olmuştur. Ancak sağlık verilerinin hassas doğası, gizlilik endişelerini beraberinde getirmekte ve verilerin merkezi sunucularda toplanmasını, dolayısıyla geleneksel makine öğrenmesi algoritmalarının uygulanabilirliğini kısıtlamaktadır. Bu çalışmada, söz konusu sorunlara çözüm olarak, federated learning (federe öğrenme) temelli bir birleştirme algoritması olan Data Unaware Classification Based on Association (duCBA) hastalık tahmini amacıyla kullanılmıştır. Model eğitimi, Diabetes Health Indicators ve Heart Attack Prediction adlı iki açık erişimli sağlık veriseti kullanılarak gerçekleştirilmiştir. Eğitime başlamadan önce Mutual Information, Chi-Square ve Principal Component Analysis (PCA) gibi öznitelik seçimi yöntemleri uygulanarak üç farklı veri kümesi oluşturulmuştur. Sistem mimarisi; bir merkezî sunucu, bir API sunucusu ve birden fazla istemciden oluşmaktadır. İstemciler, verilerini paylaşmadan kendi cihazlarında yerel modelleri eğitmektedir. API sunucusu, bu yerel modelleri toplayarak merkezî sunucuya iletmektedir. Merkezî sunucu ise bu modelleri duCBA algoritması ile birleştirerek güncel küresel modeli istemcilere tekrar dağıtmaktadır. Bu çalışma, duCBA algoritmasının federe ortamlarda uygulanabilirliğini ortaya koymakla kalmayıp, yaygın olarak kullanılan Federated Averaging (FedAvg) algoritması ile performans karşılaştırmasını da içermektedir. Sonuçlar, duCBA’nın FedAvg ile benzer performans sergilediğini ve aynı zamanda Açıklanabilir Yapay Zekâ (XAI) ilkeleriyle uyumlu olarak modelin karar süreçlerini şeffaf biçimde izlemeye olanak tanıdığını göstermektedir. Farklı öznitelik seçimi yöntemleri ve istemci sayılarıyla yapılan testlerde, özellikle az sayıda istemcinin bulunduğu sistemlerde duCBA algoritmasının istikrarlı ve dengeli bir başarı sağladığı ve %67–71 arasında tutarlı doğruluk oranı elde ettiği gözlemlenmiştir.

🔖

Cite This Article

Demir, A., Kulaksız, A. Y., Büyüktanır, B., Baydogmus, G. K., Yıldız, K. (2026). Decentralized Disease Prediction: A Federated Learning Perspective. Afyon Kocatepe Üniversitesi Fen ve Mühendislik Bilimleri Dergisi, Cilt: 26 Sayı: 2, 337–353. https://doi.org/10.35414/akufemubid.1679417

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (42)

  1. Albshaier, L., Almarri, S. and Albuali, A., 2025. Federated Learning for Cloud and Edge Security: A Systematic Review of Challenges and AI Opportunities. Electronics, 14, 5, 1019. https://doi.org/10.3390/electronics14051019
  2. Ali, M.K., Pearson-Stuttard, J., Selvin, E. and Gregg, E.W., 2022. Interpreting global trends in type 2 diabetes complications and mortality. Diabetologia, 65, 1, 3-13. https://doi.org/10.1007/s00125-021-05585-2
  3. Bian, J., Arafat, A.A., Xiong, H., Li, J., Li, L., Chen, H., Wang, J., Dou, D. and Guo, Z., 2022. Machine Learning in Real-Time Internet of Things (IoT) Systems: A Survey. IEEE Internet of Things Journal, 9, 11, 8364-8386. https://doi.org/10.1109/JIOT.2022.3161050
  4. Bhatt, H., Jadav, N.K., Kumari, A., Gupta, R., Tanwar, S., Polkowski, Z., Tolba, A. and Hassanein, A.S., 2024. Artificial neural network-driven federated learning for heart stroke prediction in healthcare 4.0 underlying 5G. Concurrency and Computation: Practice and Experience, 36, 3, e7911. https://doi.org/10.1002/cpe.7911
  5. Buyuktanir, B., Altinkaya, Ş., Karatas Baydogmus, G. and Yildiz, K., 2025. Federated learning in intrusion detection: advancements, applications, and future directions. Cluster Computing, 28, 7, 473. https://doi.org/10.1007/s10586-025-05325-w
  6. Büyüktanir, B., Çıplak, Z., Çil, A. E., Yakar, Ö., Adoum, M.B., & Yıldız, K. 2025. DDoS_FL: Federated learning architecture approach against DDoS attack. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, 31, 6, 1004-1018. https://dergipark.org.tr/en/pub/pajes/article/1813231
  7. Büyüktanır, B., Yıldız, K., Ülkü, E.E. and Büyüktanır, T., 2023. du-CBA: Data-agnostic and incremental classification-based association rules extraction architecture. Journal of the Faculty of Engineering and Architecture of Gazi University, 38, 3, 1919-1929. https://doi.org/10.17341/gazimmfd.1067527
  8. Chen, S., 2023. Comparison of machine learning algorithms and feature visualization analysis for diabetes risk prediction. Journal of Physics: Conference Series, 2646, 1, 012013. https://doi.org/10.1088/1742-6596/2646/1/012013
  9. Choudhury, O., Gkoulalas-Divanis, A., Salonidis, T., Sylla, I., Park, Y., Hsu, G. and Das, A., Differential Privacy-enabled Federated Learning for Sensitive Health Data. https://doi.org/10.48550/arXiv.1910.02578
  10. Çıplak, Z., Yıldız, K. and Altınkaya, Ş., 2025. FEDetect: A Federated Learning-Based Malware Detection and Classification Using Deep Neural Network Algorithms. Arabian Journal for Science and Engineering, 50, 19, 16107-16134. https://doi.org/10.1007/s13369-025-10043-x
  11. Demir, A., Kulaksız, A.Y., Büyüktanır, B., Baydoğmuş, G.K. and Yıldız, K., (2024). Model Training and Real World Analysis Using Health Data with Federated Learning. 2024 International Open Source Conference (UAKK). İstanbul, Türkiye, 103-108.
  12. Dian, F.J., Vahidnia, R. and Rahmati, A., 2020. Wearables and the Internet of Things (IoT), Applications, Opportunities, and Challenges: A Survey. IEEE Access, 8, 69200-69211. https://doi.org/10.1109/ACCESS.2020.2986329
  13. Durga, R. and Poovammal, E., (2021). Federated Learning Model for Healthchain System. 2021 6th IEEE International Conference on Recent Advances and Innovations in Engineering (ICRAIE). Kedah, Malezya, 1-6.
  14. Ezzeldin, Y.H., Yan, S., He, C., Ferrara, E. and Avestimehr, A.S., 2023. FairFed: Enabling Group Fairness in Federated Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 37, 6, 7494-7502. https://doi.org/10.1609/aaai.v37i6.25911
  15. Gupta, S., Kumar, P., Srivastava, N.V., Kumar, A. and Chaurasia, B.K., (2024). Heart Disease Prediction Using Federated Learning. Proceedings of International Conference on Recent Innovations in Computing. Singapore, 35-47.
  16. Hama Saeed, M.A., 2023. Diabetes type 2 classification using machine learning algorithms with up-sampling technique. Journal of Electrical Systems and Information Technology, 10, 1, 8. https://doi.org/10.1186/s43067-023-00074-5
  17. Horvath, A.N., Berchier, M., Nooralahzadeh, F., Allam, A. and Krauthammer, M., 2023. Exploratory Analysis of Federated Learning Methods with Differential Privacy on MIMIC-III. arXiv preprint arXiv:2302.04208. https://doi.org/10.48550/arXiv.2302.04208
  18. Kalasampath, K., Spoorthi, K.N., Sajeev, S., Kuppa, S.S., Ajay, K. and Maruthamuthu, A., 2025. A Literature Review on Applications of Explainable Artificial Intelligence (XAI). IEEE Access, 13, 41111-41140. https://doi.org/10.1109/ACCESS.2025.3546681
  19. Kaur, H., Rani, V., Kumar, M., Sachdeva, M., Mittal, A. and Kumar, K., 2024. Federated learning: a comprehensive review of recent advances and applications. Multimedia Tools and Applications, 83, 18, 54165-54188. https://doi.org/10.1007/s11042-023-17737-0
  20. Kucur, E. N., Buyuktanir, T., Ugurelli, M., & Yildiz, K. (2025). Privacy-Preserving Machine Learning Techniques: Cryptographic Approaches, Challenges, and Future Directions. Applied Sciences, 16(1), 277. https://doi.org/10.3390/app16010277
  21. Li, T., Sahu, A.K., Talwalkar, A. and Smith, V., 2020. Federated Learning: Challenges, Methods, and Future Directions. IEEE Signal Processing Magazine, 37, 3, 50-60. https://doi.org/10.1109/MSP.2020.2975749
  22. Li, M., Xu, P., Hu, J., Tang, Z. and Yang, G., 2025. From challenges and pitfalls to recommendations and opportunities: Implementing federated learning in healthcare. Medical Image Analysis, 101, 103497. https://doi.org/10.1016/j.media.2025.103497
  23. Lycklama, H., Burkhalter, L., Viand, A., Küchler, N. and Hithnawi, A., (2023). RoFL: Robustness of Secure Federated Learning. 2023 IEEE Symposium on Security and Privacy (SP). San Francisco, USA, 453-476.
  24. Ma, Y., Woods, J., Angel, S., Polychroniadou, A. and Rabin, T., (2023). Flamingo: Multi-Round Single-Server Secure Aggregation with Applications to Private Federated Learning. 2023 IEEE Symposium on Security and Privacy (SP). San Francisco, USA, 477-496
  25. McMahan, B., Moore, E., Ramage, D., Hampson, S. and Arcas, B.A.y., (2017). Communication-Efficient Learning of Deep Networks from Decentralized Data. Proceedings of the 20th International Conference on Artificial Intelligence and Statistics. Fort Lauderdale, USA, 1273-1282.
  26. Önlü, A.Ö., Akca, B., Buyuktanir, B., Yildiz, K. and Baydogmus, G.K., (2025). An investigation and performance evaluation of aggregation algorithms in federated learning architecture. Innovations and Applications in Smart Systems Conference (ASYU). Istanbul, Turkey.
  27. Ozturk, O., Buyuktanir, B., Baydogmus, G.K. and Yildiz, K., 2025. Differential Privacy in Federated Learning: Mitigating Inference Attacks with Randomized Response. arXiv preprint arXiv:2509.13987. https://doi.org/10.48550/arXiv.2509.13987
  28. Paleyes, A., Urma, R.G. and Lawrence, N.D., 2022. Challenges in Deploying Machine Learning: A Survey of Case Studies. ACM Computing Surveys, 55, 6, 114. https://doi.org/10.1145/3533378
  29. Raj, A., Sharma, V. and Shanu, A.K., (2022). Comparative Analysis of Security and Privacy Technique For Federated Learning In IOT Based Devices. 2022 3rd International Conference on Computation, Automation and Knowledge Management (ICCAKM). Dubai, United Arab Emirates, 1-5.
  30. Sasaki, Y., 2022. A Survey on IoT Big Data Analytic Systems: Current and Future. IEEE Internet of Things Journal, 9, 2, 1024-1036. https://doi.org/10.1109/JIOT.2021.3131724
  31. Shah, D., Patel, S. and Bharti, S.K., 2020. Heart Disease Prediction using Machine Learning Techniques. SN Computer Science, 1, 6, 345. https://doi.org/10.1007/s42979-020-00365-y
  32. Sun, Y., Li, H., Shen, Y., Xie, J., Zhao, Y., Gao, X. and Si, N., (2023). An Asynchronous Federated Learning Algorithm Based on a Backup Update of Model Version Parameters. 2023 3rd International Conference on Electrical Engineering and Control Science (IC2ECS). Nanjing, China, 1538-1544.
  33. Surur, F.M., Mamo, A.A., Gebresilassie, B.G., Mekonen, K.A., Golda, A., Behera, R.K. and Kumar, K., 2025. Unlocking the power of machine learning in big data: a scoping survey. Data Science and Management, 8, 4, 519-535. https://doi.org/10.1016/j.dsm.2025.02.004
  34. Tosun, B.E., Yorulmaz, E., Ergul, F.E., Buyuktanir, B., Baydogmus, G.K. and Yildiz, K., (2025). Comparative Analysis of Relational Classification Algorithms on Benchmark Datasets. 2025 10th International Conference on Computer Science and Engineering (UBMK). 790-795.
  35. Wen, J., Zhang, Z., Lan, Y., Cui, Z., Cai, J. and Zhang, W., 2023. A survey on federated learning: challenges and applications. International Journal of Machine Learning and Cybernetics, 14, 2, 513-535. https://doi.org/10.1007/s13042-022-01647-y
  36. Zhang, C., Xie, Y., Bai, H., Yu, B., Li, W. and Gao, Y., 2021. A survey on federated learning. Knowledge-Based Systems, 216, 106775. https://doi.org/10.1016/j.knosys.2021.106775
  37. Zhang, T., Zhu, T., Gao, K., Zhou, W. and Yu, P.S., 2023. Balancing Learning Model Privacy, Fairness, and Accuracy With Early Stopping Criteria. IEEE Transactions on Neural Networks and Learning Systems, 34, 9, 5557-5569. https://doi.org/10.1109/TNNLS.2021.3129592
  38. Zhou, S. and Li, G.Y., 2023. FedGiA: An Efficient Hybrid Algorithm for Federated Learning. IEEE Transactions on Signal Processing, 71, 1493-1508. https://doi.org/10.1109/TSP.2023.3268845
  39. Internet References
  40. American Heart Association, AHA names biggest advances in cardiovascular research for 2024, https://www.heart.org/en/around-the-aha/aha-names-biggest-advances-in-cardiovascular-research-for-2024, (04.01.2026)
  41. Centers for Disease Control and Prevention (CDC), Diabetes Health Indicators Dataset, http://www.kaggle.com/datasets/alexteboul/diabetes-health-indicators-dataset/data, (04.01.2026)
  42. Fayez, O., Heart Attack Data, https://www.kaggle.com/datasets/omarfayez/heart-attack-data, (04.01.2026)