An Adaptive Federated Machine Learning Framework for Privacy-Preserving Prediction of Type 2 Diabetes Mellitus across Healthcare Institutions.

There are challenges in collaboration between health care institutions in the areas of confidentiality, governance considerations and information system integration that hinder the accurate prediction of Type 2 diabetes mellitus to enable timely preventive interventions. This is further complicated by differences in the populations of patients, number of patients in datasets, the computing resources available and participation. In this study, a novel adaptive federated machine learning approach for privacy-preserving diabetes onset prediction within three years is proposed. The methodological design calls for 5,000 synthetic records of adult patients, across five simulated hospitals. Preprocessing consists of duplicate removal, harmonisation of variables, missing value imputation, categorical encoding, numerical scaling, and class weighting, and is only applied to the training data. Prediction model gets local training, aggregation weight based on model validation performance and update reliability and participation history. Differential privacy and secure aggregation are defined as complementary protection measures. Discrimination, sensitivity, F1 score, calibration, institutional consistency, and computational efficiency are used as evaluation metrics for comparison between local-only, centralised, FedAvg, FedProx, and adaptive approaches. Illustrative performance targets are an AUROC of 0.9000, sensitivity of 0.8200, and F1 score of 0.8500. An efficiency illustration shows that the training time is reduced by 25%, the volume of communication is reduced by 33.3%, and convergence is achieved in 60 rounds, when compared to the efficiency illustrated with FedAvg. These numbers are not experimentally validated and differences in institution performance, confidence intervals and numerical privacy budgets are not measured. The contribution is an integrated design connecting adaptive collaboration, explicit privacy controls, and comprehensive evaluation. To draw conclusions on effectiveness and readiness for deployment in different clinical contexts, the implementation needs to be reproducible, privacy accounting must be done, and external clinical validation must be conducted.

There are challenges in collaboration between health care institutions in the areas of confidentiality, governance considerations and information system integration that hinder the accurate prediction of Type 2 diabetes mellitus to enable timely preventive interventions. This is further complicated by differences in the populations of patients, number of patients in datasets, the computing resources available and participation. In this study, a novel adaptive federated machine learning approach for privacy-preserving diabetes onset prediction within three years is proposed. The methodological design calls for 5,000 synthetic records of adult patients, across five simulated hospitals. Preprocessing consists of duplicate removal, harmonisation of variables, missing value imputation, categorical encoding, numerical scaling, and class weighting, and is only applied to the training data. Prediction model gets local training, aggregation weight based on model validation performance and update reliability and participation history. Differential privacy and secure aggregation are defined as complementary protection measures. Discrimination, sensitivity, F1 score, calibration, institutional consistency, and computational efficiency are used as evaluation metrics for comparison between local-only, centralised, FedAvg, FedProx, and adaptive approaches. Illustrative performance targets are an AUROC of 0.9000, sensitivity of 0.8200, and F1 score of 0.8500. An efficiency illustration shows that the training time is reduced by 25%, the volume of communication is reduced by 33.3%, and convergence is achieved in 60 rounds, when compared to the efficiency illustrated with FedAvg. These numbers are not experimentally validated and differences in institution performance, confidence intervals and numerical privacy budgets are not measured. The contribution is an integrated design connecting adaptive collaboration, explicit privacy controls, and comprehensive evaluation. To draw conclusions on effectiveness and readiness for deployment in different clinical contexts, the implementation needs to be reproducible, privacy accounting must be done, and external clinical validation must be conducted.

1. Aagaard, A., Röttger, R., Johnson, E. K., & Olsen, K. R. (2025). Comparing the predictive performance of diabetes complications using administrative health data and clinical data. Scientific Reports, 15, Article 33035. https://doi.org/10.1038/s41598-025-18079-7 [Google Scholar] [Crossref]

2. Arkan Yurtoglu, R., Dervişoğlu, H., Aktaş, M., Daşçılar, M., Halepmollası, R., Baz, İ., & Haklidir, M. (2024, November). Federated learning for diabetes prediction in healthcare [Conference presentation]. IV International Congress on Artificial Intelligence in Health, Izmir, Türkiye. [Google Scholar] [Crossref]

3. Deng, Y., Kamani, M. M., & Mahdavi, M. (2020). Adaptive personalized federated learning [Preprint]. arXiv. https://doi.org/10.48550/arXiv.2003.13461 [Google Scholar] [Crossref]

4. Dhade, P., & Shirke, P. (2023). Federated learning for healthcare: A comprehensive review. Engineering Proceedings, 59(1), Article 230. https://doi.org/10.3390/engproc2023059230 [Google Scholar] [Crossref]

5. Goyal, R., Singhal, M., & Jialal, I. (2023). Type 2 diabetes. In StatPearls. StatPearls Publishing. https://www.ncbi.nlm.nih.gov/books/NBK513253/ [Google Scholar] [Crossref]

6. Herath, H. M. S. S., Herath, H. M. K. K. M. B., Madhusanka, B. G. D. A., & Guruge, L. G. P. K. (2024). Data protection challenges in the processing of sensitive data. In C. Hewage, L. Yasakethu, & D. N. K. Jayakody (Eds.), Data protection (pp. 155–179). Springer. https://doi.org/10.1007/978-3-031-76473-8_8 [Google Scholar] [Crossref]

7. Horst, A., Loustalot, P., Yoganathan, S., Li, T., Xu, J., Tong, W., Schneider, D., Löffler-Perez, N., Di Renzo, E., & Renaudin, M. (2025). Federated learning: A privacy-preserving approach to data-centric regulatory cooperation. Frontiers in Drug Safety and Regulation, 5, Article 1579922. [Google Scholar] [Crossref]

8. https://doi.org/10.3389/fdsfr.2025.1579922 [Google Scholar] [Crossref]

9. Jayalakshmi, R., & Tamilvizhi, T. (2025). Privacy preservation in diabetic disease prediction using federated learning based on efficient cross stage recurrent model. Scientific Reports, 15, Article 37258. https://doi.org/10.1038/s41598-025-21229-6 [Google Scholar] [Crossref]

10. Kawalkar, U., Gaidhane, A., Aggarwal, S., Mankar, A., Ambadekar, N., & Kogade, P. (2025). Challenges and strategies for early diagnosis and management of type 2 diabetes mellitus at the primary healthcare level in rural Western Vidarbha, India: A qualitative study. BMJ Public Health, 3(2), Article e003126. https://doi.org/10.1136/bmjph-2025-003126 [Google Scholar] [Crossref]

11. Koch, A., Brust, L., & Weigl, M. (2024). Barriers and facilitators of patient engagement activities to improve patient safety in healthcare organizations: A Delphi-based expert survey. Risk Management and Healthcare Policy, 17, 3217–3226. https://doi.org/10.2147/RMHP.S489522 [Google Scholar] [Crossref]

12. Ravi, R., & Olickal, J. J. (2026). Machine learning-based risk prediction models for type 2 diabetes in primary care: A scoping review. Frontiers in Public Health, 14, Article 1946070. [Google Scholar] [Crossref]

13. https://doi.org/10.3389/fpubh.2026.1946070 [Google Scholar] [Crossref]

14. Rieke, N., Hancox, J., Li, W., Milletarì, F., Roth, H. R., Albarqouni, S., Bakas, S., Galtier, M. N., Landman, B. A., Maier-Hein, K., Ourselin, S., Sheller, M., Summers, R. M., Trask, A., Xu, D., Baust, M., & Cardoso, M. J. (2020). The future of digital health with federated learning. npj Digital Medicine, 3, Article 119. https://doi.org/10.1038/s41746-020-00323-1 [Google Scholar] [Crossref]

15. Sheller, M. J., Edwards, B., Reina, G. A., Martin, J., Pati, S., Kotrotsou, A., Milchenko, M., Xu, W., Marcus, D., Colen, R. R., & Bakas, S. (2020). Federated learning in medicine: Facilitating multi-institutional collaborations without sharing patient data. Scientific Reports, 10, Article 12598. https://doi.org/10.1038/s41598-020-69250-1 [Google Scholar] [Crossref]

16. SM, M. (2025). Diabetes prediction using federated learning. International Journal of Scientific Research in Engineering and Management, 9(1), 1–9. https://doi.org/10.55041/IJSREM41133 [Google Scholar] [Crossref]

17. Teo, Z. L., Jin, L., Liu, N., Li, S., Miao, D., Zhang, X., Ng, W. Y., Tan, T. F., Lee, D. M., Chua, K. J., Heng, J., Liu, Y., Goh, R. S. M., & Ting, D. S. W. (2024). Federated machine learning in healthcare: A systematic review on clinical applications and technical architecture. Cell Reports Medicine, 5(2), Article 101419. https://doi.org/10.1016/j.xcrm.2024.101419 [Google Scholar] [Crossref]

18. Usman, M., Bernardi, M. L., & Cimitile, M. (2025). Introducing a quality-driven approach for federated learning. Sensors, 25(10), Article 3083. https://doi.org/10.3390/s25103083 [Google Scholar] [Crossref]

19. Wang, Q., Cheng, H., Jiang, S., Zhang, L., Liu, X., Chen, P., Liu, J., Li, Y., Liu, X., Wang, L., Li, Z., Cai, G., Chen, X., & Dong, Z. (2024). The relationship between diabetic retinopathy and diabetic nephropathy in type 2 diabetes. Frontiers in Endocrinology, 15, Article 1292412. [Google Scholar] [Crossref]

20. https://doi.org/10.3389/fendo.2024.1292412 [Google Scholar] [Crossref]

21. Yang, X., Yao, M., Huang, J., Cheng, Z., & Sun, T. (2026). Machine learning predicts diabetes risk in high-risk populations: Analysis of National Health and Nutrition Examination Survey data. Archives of Medical Science, 22(2), 708–720. https://doi.org/10.5114/aoms/209547 [Google Scholar] [Crossref]

22. Corrections: Yang et al. uses 2026, the year of its assigned journal issue, rather than its 2025 online publication date. Issue 2/2026 vol. 22 Dhade and Shirke uses 2023, following the publisher’s recommended citation. The Manjunath entry retains “SM” as the indexed surname; confirm the name order against the original article before submission. [Google Scholar] [Crossref]

• Classifying News Articles on Nigerian Newspapers Using Machine Learning
• Approaching Early Dyslexia Prediction with Ensemble Machine Learning: A Secondary Empirical Analysis of the Rello Et Al. Public Behavioural Dataset
• Machine Learning Modelling of Geostationary SBAS Satellite Orbit Errors Observed from an Equatorial West African Station
• A Comparative Study of CNN and Vision Transformer Models in Alzheimer's Detection
• A Stacking Ensemble Machine Learning Framework for Predicting Academic Performance and Identifying At-Risk Students in Rural Secondary Schools: Evidence from Aurangabad District, Maharashtra