Efficient Federated LLM Framework for Intelligent IoMT Management

Federated Anomaly Detection for IoMT Networks: Privacy‐Preserving Design, Lightweight Implementation, and Runtime Evaluation

The rapid integration of Internet of Things (IoT) in the healthcare domain has led to the emergence of the Internet of Medical Things (IoMT), which introduces significant benefits in patient monitoring and real‐time medical services. However, IoMT networks are inherently vulnerable due to resource constraints, heterogeneous devices, and sensitivity of medical data. In this paper, we propose a novel federated learning‐based anomaly detection system (Fed‐ADS) designed specifically for IoMT networks. Our system leverages local training of lightweight ML models on resource‐constrained IoMT devices and employs secure model aggregation at the gateway to preserve privacy and avoid centralized data collection. To address real‐world challenges, we implement and evaluate our system on a real IoMT testbed using Raspberry Pi devices under various attack scenarios. Furthermore, we examine the impact of privacy‐preserving techniques such as differential privacy on detection accuracy and system overhead. The runtime evaluation shows that our approach achieves high detection accuracy (over 94%) with minimal CPU and memory usage (under 3%), making it suitable for practical deployment in medical environments.

FedSE-1DSqueezeNet: a lightweight federated intrusion detection system for Internet of Things

FedSE-1DSqueezeNet is proposed, a lightweight federated IDS tailored for resource-constrained IoT environments, designed to optimize feature extraction efficiency under strict resource constraints and achieves detection accuracy exceeding that of state-of-the-art models.

Scalable Data Engineering Architectures for Federated Learning in Decentralized Cloud Environments

The growing adoption of Federated Learning (FL) is reshaping the way machine learning models are trained across distributed, privacy-sensitive datasets. However, the scalable and efficient orchestration of data engineering pipelines in decentralized cloud environments remains a significant challenge. This paper presents a comprehensive architectural framework for scalable data engineering tailored for FL in heterogeneous and resource-constrained environments. By integrating modern distributed computing paradigms, such as Kubernetes-based orchestration, edge-aware data preprocessing, and secure federated communication, we propose a modular architecture that addresses data heterogeneity, scalability, and compliance. A case study in a healthcare IoT scenario validates the performance and flexibility of the proposed system. Our work serves as a blueprint for deploying robust FL systems in real-world decentralized cloud ecosystems.

A Federated Learning Framework for Privacy-Preserving Oral Cancer Screening on Smartphones

Data are the cornerstone of robust AI models. However, in the medical domain, access to reliable data is constrained by regulatory requirements and patient privacy, and clinical oral images are particularly difficult to obtain. Federated learning (FL) mitigates these constraints by enabling collaborative model development across decentralized datasets without centralizing or sharing patient data. This work presents a practical FL framework that supports geographically distributed collaboration among AI healthcare researchers and facilitates the development of robust models for oral cancer screening. Client devices were interconnected via Tailscale to provide secure networking and real-time communication. We implemented the FL workflow using the Flower framework for server-side aggregation, while client deployment and orchestration were configured manually; no enterprise FL platforms were used. To support a smartphone-based screening application, we evaluated lightweight, mobile-friendly architectures including MobileNetV2, MobileNetV3Large, and MobileNetV4-Conv-Small (MNv4-Conv-S). Across the global lightweight models aggregated using FedAvg, the MNv4-Conv-S based global model (GM-V4) achieved the best performance, reaching an AUC of 0.929 and an accuracy of 87%

Federated TinyML and digital twin framework for secure and resilient IoMT-based ICU monitoring

Results indicate that a federated TinyML architecture with lightweight patient-state tracking, validation-based ensemble filtering, differential privacy, and post-quantum-secure communication can support privacy-aware and attack-resilient ICU monitoring experiments in resource-constrained IoMT settings.

Task Offloading in SDN-Fog Architecture Utilizing Hybrid Deep Learning Models for Internet of Medical Things