The IoMT Security Trilemma: A Systematic Review of Privacy, Utility, and Resource Efficiency in Federated Edge Learning
Abstract
Federated Learning (FL) is widely proposed as a privacy-preserving solution for healthcare AI. However, its deployment on Internet of Medical Things (IoMT) devices introduces critical tensions between privacy guarantees, task utility, and resource efficiency. Heavyweight privacy-enhancing technologies (PETs) such as differential privacy, homomorphic encryption, and secure aggregation challenge edge devices with constrained battery, compute, and bandwidth capacity. This systematic review synthesizes reported performance data from FL deployments on IoMT hardware to map the practical cost of security. We analyze trade-offs between privacy, accuracy, and resource usage across a range of PETs and hardware platforms. Our findings show that privacy costs are unavoidable but highly context-dependent: lightweight PETs combined with architectural co-design can be validated on real IoMT devices with modest utility loss, while crypto-heavy methods remain largely gateway-bound or simulation-based.
How strong is the evidence?
Evidence Hub reads this as a systematic review. Its evidence level is 1 of 5 (Highest · pooled evidence). Pooled evidence from several studies sits at the top of the evidence pyramid.
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.