Skip to content
Conference Open access

Explainable and Privacy-Preserving AI in Healthcare using Federated Learning and Homomorphic Encryption

Oct 2026 · International Conference on Innovative Mechanisms for Industry Applications · pp. 799-804 · 0 citations · 19 references

Abstract

Artificial intelligence (AI) in healthcare has had a significant impact on the treatment of diseases, personalized treatment, and clinical decision-making due to its rapid adoption. Nonetheless, the sensitivity of healthcare data poses crucial threats to the privacy of data, data security, and distrust in black-box AI systems. Conventional centralized machine learning methods tend to be inappropriate because of regulatory limitations and risks of information leakage. In response to these challenges, privacy-preserving methods like Federated Learning (FL) and Homomorphic Encryption (HE) have become one of the promising solutions as they provide the possibility to train a model remotely and conduct the computations on encrypted data. However, such methods do not necessarily resolve the problem of interpretability of the models, which is crucial to clinical uptake. In the current review, we provide a detailed discussion of privacy-conserving and explainable AI methods in healthcare through the combination of FL, HE, and Explainable AI (XAI). We suggest a systematic taxonomy of the available methods, classify them according to privacy mechanisms and explainability possibilities, and give a comparative analysis of recent state-of-the-art methods. Moreover, we determine major research gaps, such as the absence of coherent frameworks that provide privacy, efficiency, and interpretability at the same time, and the issues related to scaling, computational costs, and resilience to adversarial attacks. This review will be used to direct future investigations in ensuring that AI-based healthcare systems can be safe, reliable, and effective.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.