2026· Computer Modeling in Engineering & Sciences· Vol 147, pp. 1-10· 0 citations· 38 references
TL;DR
RP-IoMT is proposed, a robust and privacy-preserving FL framework that integrates secure multi-party computation (MPC), zero-knowledge proof-based gradient verification, and robust aggregation mechanisms to ensure both the correctness and integrity of model updates while maintaining strong privacy guarantees in adversarial IoMT environments.
Abstract
: Federated learning (FL) has emerged as a promising approach for enabling collaborative model training across distributed Internet of Medical Things (IoMT) devices without sharing sensitive data. Existing FL frameworks face significant challenges in healthcare settings, including vulnerability to adversarial attacks, lack of verifiable update integrity, and limited robustness under heterogeneous data distributions. These limitations hinder reliable deployment in critical medical applications. To address these challenges, this paper proposes RP-IoMT, a robust and privacy-preserving FL framework that integrates secure multi-party computation (MPC), zero-knowledge proof-based gradient verification, and robust aggregation mechanisms. The objective of this work is to ensure both the correctness and integrity of model updates while maintaining strong privacy guarantees in adversarial IoMT environments. RP-IoMT enforces bounded client updates using a zero-knowledge clipping protocol (ZKClip), performs secure aggregation using threshold-based MPC, and incorporates robust filtering techniques to mitigate poisoning and backdoor attacks. Experimental results on healthcare datasets demonstrate that RP-IoMT achieves improved predictive performance, reduced attack success rates, and stable convergence under both independent and identically distributed (IID) and non-IID conditions. These results indicate that the proposed framework provides a practical and reliable solution for secure and robust FL in real-world medical Internet of Things (IoT)
Federated learning (FL) enables collaborative model training without sharing raw data, making it well suited to privacy-sensitive healthcare applications. However, malicious model updates and anonymous participation abuse can disrupt optimization and undermine the reliability of downstream medical AI systems. Existing defenses often address robustness or privacy separately, with limited support for accountable authentication and client revocation. To address these limitations, we propose TGSFL, a Trustworthy Group-Signature-Based Federated Learning framework for healthcare large language models (LLMs). The proposed framework integrates dynamic group-signature authentication, per-client local differential privacy, and dual-layer malicious-update detection to identify and suppress adversarial updates before aggregation. It further supports anonymous yet traceable participation and dynamic revocation of persistently malicious clients. Theoretical analysis establishes conditional traceability, dynamic revocation, and per-client local differential privacy. Experiments under multiple Byzantine attack scenarios show that TGSFL improves robustness and training stability over existing baselines while introducing only modest computational overhead, demonstrating its practicality for secure and privacy-preserving medical AI training.
Xinying Liu, Chunhua Jin, Yu-Peng Wu et al.· Journal of King Saud Univers...· 0 citations
Background The rapid expansion of Internet of Medical Things (IoMT) and telehealth platforms has generated vast amounts of patient data suitable for training diagnostic Artificial Intelligence (AI) models. However, strict privacy regulations (HIPAA, GDPR) and the risk of data breaches prevent the centralization of this sensitive information. While Federated Learning (FL) allows for collaborative training without sharing raw patient data, it introduces a critical “trust deficit”: central aggregators cannot verify the integrity of local model updates without inspecting the private data, leaving the system vulnerable to model poisoning and malicious actors. Methods We introduce TeleZK-FL, a privacy-preserving framework designed specifically for resource-constrained telehealth environments. Unlike existing Zero-Knowledge Proof (ZKP) systems that require high-performance computing or offload proof generation to trusted edge servers, TeleZK-FL integrates (1) Post-Training Quantization (PTQ) to compress model gradients from 32-bit floating-point to 8-bit integers, and (2) optimized Look-Up Table (LUT) arguments to generate cryptographic proofs of client-side training integrity directly on edge devices. We evaluated the framework on two clinical modalities—CheXpert (chest X-rays) and PTB-XL (12-lead ECGs)—using simulated medical edge gateways (Raspberry Pi 4) under both homogeneous and heterogeneous data distributions. Results TeleZK-FL accelerates cryptographic proof generation by approximately 25× compared to standard ZK-SNARK implementations, generating full-model proofs in approximately 84 milliseconds per client on edge hardware. It cuts the communication payload by 75% (a 4.0× reduction) while maintaining a diagnostic Area Under the Curve (AUC) of 0.877 on CheXpert and 0.891 on PTB-XL, representing a degradation of only 0.1% and 0.3% respectively compared to the unquantized baselines of 0.878 and 0.894. Conclusion TeleZK-FL establishes the feasibility of verifiable, trustless federated learning on commodity telehealth hardware. By eliminating the computational bottlenecks of server-side proof generation while incurring only 0.1%–0.3% AUC degradation, it provides an efficient, regulation-aligned method for building scalable and secure decentralized medical AI networks. We further discuss an inherent trade-off: the edge-efficient KZG-based construction is classically—not post-quantum—secure, which we position explicitly against recent lattice-based alternatives.
P. Jayaraman, R. Delhibabu· Frontiers in Digital Health· 0 citations
Deep learning is becoming popular in cloud applications and serves to provide intelligent services; data aggregation in a central location makes sensitive information vulnerable to privacy breaches, regulatory infractions, and adversarial manipulation. All modern privacy mechanisms offer partial protection and frequently lack accuracy, scalability, or practicality in their operations. To overcome these limitations, a federated deep learning model is formulated so that secure joint learning can occur without transferring raw data across the domains of ownership. The framework incorporates training that is decentralized, training that uses differential privacy, training that uses secure aggregation, training that uses encrypted communication, and training that uses trust-based anomaly defense to defend against leakage, poisoning, and inference attacks. It also supports heterogeneous and highly non-IID datasets using adaptive coordination and stability-relevant participation regulation and meets emerging data protection requirements. The methods of resource-conscious orchestration and the optimization of communication eliminate overhead without obstructing the effectiveness of learning. The paradigm has therefore formed a privacy-by-design intelligent cloud ecosystem which ensures confidentiality, maintains performance, enhances robustness, and ensures responsible AI implementation in privacy-related sectors of healthcare, finance, governance, and smart infrastructure.
Sribidhya Mohanty, Pallavi Gupta, Anil Pratap Singh et al.· 2026 International Conferenc...· 0 citations
Smart healthcare IoT systems are vulnerable to cyber threats as they deal with sensitive patient information. Problems such as privacy, scalability, and delayed response to threats in distributed healthcare environments challenge centralized security approaches. To mitigate the security challenges of cloud-edge healthcare IoT systems, this paper presents FL-EZTF, a privacy-preserving, Federated Deep Learning and Enhanced Zero Trust Framework. The framework combines federated learning, Enhanced Zero Trust Architecture (E-ZTA), and Secure Access Service Edge (SASE). In this framework, lightweight deep learning models are developed locally at hospitals and various edge nodes without the need to transfer sensitive medical data. In place of raw data, model updates are sent conveniently through a trustaware federated learning process. Simultaneously, E-ZTA performs continuous authentication, micro-segmentation, and access control to rapidly contain threats. The framework is assessed using CIC-IoT-2023, IoT-23, and WESAD datasets. The experimental results show improved accuracy in detection, lower rates of false positives, a significant reduction in the latency of decisions, and enhanced containment as compared to centralized and traditional federated learning.
Unknown authors· International Journal of Eng...· 0 citations
The proliferation of Internet of Medical Things (IoMT) devices has created critical cybersecurity challenges demanding intrusion detection systems that achieve high accuracy across diverse attack taxonomies while preserving patient privacy across institutional boundaries. Existing federated learning (FL) approaches face an inherent tension: Transformer-based architectures achieve strong detection performance but incur quadratic computational complexity and substantial communication overhead, while lightweight classifiers sacrifice representational capacity. Moreover, most FL-based intrusion detection systems lack formal privacy guarantees and robustness against adversarial participants. This paper introduces FedMamba-IoMT, the first federated State Space Model framework for privacy-preserving intrusion detection in IoMT networks, incorporating differential privacy (DP-SGD), Byzantine-resilient aggregation, and multi-level explainability. The proposed architecture reformulates tabular network traffic features as pseudo-sequential tokens processed through stacked selective State Space Model (Mamba) blocks with gated residual connections, achieving linear computational complexity 𝒪(n) with 78% fewer parameters than Transformer alternatives. We design a novel FedMamba aggregation strategy that weights client contributions by a convex combination of dataset proportion and inverse validation loss, augmented with a cosine similarity-based Byzantine filter that detects and excludes malicious model updates. Integration of DP-SGD with Rényi differential privacy accounting provides formal privacy guarantees (ε∈{1.0,2.0,3.0,5.0,8.0}, δ=10−5) while maintaining competitive accuracy. Comprehensive evaluation across three benchmark datasets—Edge-IIoTset (2,219,201 samples, 15 classes), CICIoMT2024 (3,204,537 samples, 19 classes), and Gotham Dataset 2025 (496,191 samples, 8 high-level traffic categories)—demonstrates that FedMamba-IoMT achieves 99.47±0.04%, 99.52±0.04%, and 98.90±0.04% multiclass accuracy without DP, and 98.52%, 98.18%, and 97.16% at ε=3.0, surpassing all prior federated IDS approaches. Byzantine resilience experiments demonstrate that the proposed defense maintains >95% accuracy under 30% malicious clients across label-flipping, model poisoning, and free-rider attacks. Gradient inversion analysis confirms that FedMamba’s compact parameterization (135K parameters, 0.52 MB) provides 2× higher reconstruction error compared to Transformer-based FL, and the integrated SHAP and LIME explainability framework supports regulatory compliance with the FDA’s 2023 cybersecurity guidance for medical devices.
Y. Al-Sharo, Mohammed Tawfik, A. M. Al-madani et al.· PLoS ONE· 0 citations
Federated Learning enables collaborative model training without sharing raw client data, making it attractive for privacy-sensitive domains. However, its performance degrades when local data are non-independent and identically distributed (non-IID) and when malicious clients inject adversarial updates; robust aggregation alone can be insufficient, especially for imbalanced medical datasets where minority-class degradation is masked by overall accuracy. This paper proposes a Median Absolute Deviation (MAD)-based malicious-update filtering framework for non-IID federated learning. The server flags and excludes abnormal client updates before aggregation, using a coordinate-wise median reference and a modified Z-score over cosine distances. Unlike trust-based defenses, it requires no clean server-side dataset and provides an interpretable, per-round diagnostic. We evaluate it on CIFAR-10 and HAM10000 under noise injection and sign flipping against eight robust aggregation baselines, under a unified protocol that reports Accuracy and Macro-F1 at the same validation-selected checkpoint, averaged over three seeds. On CIFAR-10, the proposed method attains a mean Macro-F1 comparable to or better than the strongest baselines under both attacks (75.9% and 67.6%). On HAM10000 under noise injection it achieves the highest mean accuracy (74.0%) with a competitive Macro-F1, whereas under sign flipping it is only mid-ranked. An ablation indicates that removing MAD filtering substantially reduces class-balanced performance on HAM10000 under noise injection, and a diagnostic analysis shows that the cosine-distance signal separates benign and malicious updates strongly under noise injection but weakly under sign flipping. These results characterize both the robustness potential and the boundary conditions of update-level median filtering in non-IID federated learning.
Tae-Wook Kang, Ji-Woo Park, Chulyoung Park et al.· IEEE Access· 0 citations