Enhancing the Robustness of Federated Learning Models for Dementia Diagnosis Against Data Poisoning Attacks

Authors

https://doi.org/10.22105/ahse.v3i2.65

Abstract

Federated Learning (FL) enables privacy preserving dementia diagnosis but is vulnerable to poisoning attacks. We propose a hybrid defense integrating robust aggregation, client behavior analysis, feature consistency validation, and anomaly aware training. Results on ADNI, OASIS-3, and DementiaBank improve robustness while maintaining accuracy.  

Keywords:

Federated learning, Dementia diagnosis, Data poisoning attacks, Backdoor attacks, Robust aggregation

References

  1. [1] world health organization. (2021). Global status report on the public health response to dementia. https://www.who.int/publications/i/item/9789240033245
  2. [2] LaMontagne, P. J., Benzinger, T. L. S., Morris, J. C., Keefe, S., Hornbeck, R., Xiong, C., … ., & Marcus, D. (2019). OASIS-3: Longitudinal neuroimaging, clinical, and cognitive dataset for normal aging and Alzheimer disease. Cold Spring Harbor Laboratory Press. https://doi.org/10.1101/2019.12.13.19014902
  3. [3] Marcus, D. S., Fotenos, A. F., Csernansky, J. G., Morris, J. C., & Buckner, R. L. (2010). Open access series of imaging studies: Longitudinal MRI data in nondemented and demented older adults. Journal of Cognitive Neuroscience, 22(12), 2677–2684. https://doi.org/10.1162/jocn.2009.21407
  4. [4] Sheller, M. J., Edwards, B., Reina, G. A., Martin, J., Pati, S., Kotrotsou, A., … ., & Bakas, S. (2020). Federated learning in medicine: Facilitating multi-institutional collaborations without sharing patient data. Scientific Reports, 10(1), 12598. https://doi.org/10.1038/s41598-020-69250-1
  5. [5] Chen, A. A., Luo, C., Chen, Y., Shinohara, R. T., & Shou, H. (2022). Privacy-preserving harmonization via distributed ComBat. NeuroImage, 248, 118822. https://doi.org/10.1016/j.neuroimage.2021.118822
  6. [6] Bagdasaryan, E., Veit, A., Hua, Y., Estrin, D., & Shmatikov, V. (2020). How to backdoor federated learning. Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics (pp. 2938–2948). PMLR. https://proceedings.mlr.press/v108/bagdasaryan20a.html
  7. [7] Fang, M., Cao, X., Jia, J., & Gong, N. (2020). Local model poisoning attacks to {byzantine-robust} federated learning. 29th Usenix Security Symposium (Usenix Security 20) (pp. 1605–1622). USENIX Association. https://www.usenix.org/conference/usenixsecurity20/presentation/fang
  8. [8] Jack Jr., C. R., Bernstein, M. A., Fox, N. C., Thompson, P., Alexander, G., Harvey, D., … ., & Weiner, M. W. (2008). The Alzheimer’s disease neuroimaging initiative (ADNI): MRI methods. Journal of Magnetic Resonance Imaging, 27(4), 685–691. https://doi.org/10.1002/jmri.21049
  9. [9] Eyigoz, E., Mathur, S., Santamaria, M., Cecchi, G., & Naylor, M. (2020). Linguistic markers predict onset of Alzheimer’s disease. EClinicalMedicine, 28. https://doi.org/10.1016/j.eclinm.2020.100583
  10. [10] Dayan, I., Roth, H. R., Zhong, A., Harouni, A., Gentili, A., Abidin, A. Z., … ., & Li, Q. (2021). Federated learning for predicting clinical outcomes in patients with COVID-19. Nature Medicine, 27(10), 1735–1743. https://doi.org/10.1038/s41591-021-01506-3
  11. [11] Xie, C., Huang, K., Chen, P. Y., & Li, B. (2019). Dba: Distributed backdoor attacks against federated learning. International Conference on Learning Representations. OpenReview.net.
  12. [12] Blanchard, P., El Mhamdi, E. M., Guerraoui, R., & Stainer, J. (2017). Machine learning with adversaries: byzantine tolerant gradient descent. Advances in Neural Information Processing Systems. Curran Associates, Inc. https://proceedings.neurips.cc/paper_files/paper/2017/file/f4b9ec30ad9f68f89b29639786cb62ef-Paper.pdf
  13. [13] Yin, D., Chen, Y., Kannan, R., & Bartlett, P. (2018). Byzantine-robust distributed learning: Towards optimal statistical rates. Proceedings of the 35th International Conference on Machine Learning (pp. 5650–5659). PMLR. https://proceedings.mlr.press/v80/yin18a.html
  14. [14] McMahan, B., Moore, E., Ramage, D., Hampson, S., & 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 (pp. 1273–1282). PMLR. https://proceedings.mlr.press/v54/mcmahan17a.html

Downloads

Published

2026-06-27

Issue

Vol. 3 No. 2 (2026): Annals of Healthcare Systems Engineering

Section

Articles

How to Cite

Ebrahimzadeh, F., Kaveh, S., & Falah Rad, M. (2026). Enhancing the Robustness of Federated Learning Models for Dementia Diagnosis Against Data Poisoning Attacks. Annals of Healthcare Systems Engineering, 3(2), 134-147. https://doi.org/10.22105/ahse.v3i2.65