Cybersecurity-enabled federated learning approach for digital healthcare

Document Type

Book Chapter

Source of Publication

Explainable Artificial Intelligence Xai for Next Generation Cybersecurity Concepts Challenges and Applications

Publication Date

10-15-2025

Abstract

With the rapid integration of artificial intelligence into cybersecurity, smart and digitized systems have significantly enhanced the ability to detect and prevent security threats. However, the increasing reliance on distributed AI systems has also introduced critical challenges related to data leakage, system vulnerabilities, and computational overhead. This chapter addresses the persistent problem of securing sensitive healthcare data in federated learning environments, where centralized data aggregation is avoided. The objective is to develop a cybersecurity-enabled federated learning framework that ensures privacy-preserving, secure, and scalable intrusion detection in digital healthcare systems. The proposed framework incorporates four core modules: secure data encryption and transmission, participant authentication and authorization, privacy-preserving model aggregation using homomorphic encryption or secure multi-party computation, and anomaly detection with intrusion prevention mechanisms. Deep learning models, specifically CNNs, are employed within the federated setting to enhance detection accuracy. Key contributions include maintaining data privacy without sacrificing model performance, enabling distributed training while preserving data ownership, and integrating proactive anomaly detection. Experimental results using the CIC IDS 2017 and IoT Healthcare Security datasets show the proposed model outperforms centralized systems, achieving accuracy and precision rates above 99%. Despite its effectiveness, the model faces limitations related to communication latency and computational complexity in real-time healthcare systems. Future research will focus on optimizing resource efficiency, extending the framework to more diverse IoT healthcare datasets, and incorporating adaptive threat intelligence to respond to evolving cybersecurity risks.

ISBN

[9781837240319, 9781837240326]

Publisher

The Institution of Engineering and Technology

First Page

259

Last Page

278

Disciplines

Computer Sciences

Keywords

Privacy-Preserving Technologies in Data

Scopus ID

105026740013

Indexed in Scopus

yes

Open Access

no

Link to Full Text

Share

COinS