An Integrated Consumer–Enterprise Federated Intelligence Framework With Post-Quantum Security for Healthcare IoMT

Abstract

This paper addresses the growing challenge of implementing secure, reliable, and scalable federated intelligence on consumer Internet of medical things (IoMT) devices and healthcare enterprise systems. Current solutions generally trade off individual aspects, for example, learning accuracy, cryptographic strength, and blockchain auditability, and do not present an end-to-end framework that ensures privacy preservation, adversarial robustness, post-quantum security, and enterprise governance. We address these shortcomings by introducing CeN-CHAIN, a consumer–enterprise integrated framework that incorporates post-quantum key distribution, homomorphic encryption with minimal computational overhead, differential privacy, blockchain-ensured validation, and federated model lifecycle management into a single architectural model. CeN-CHAIN enables on-device learning at a secure level, incorporating auditing, aggregability, and model custodianship in a heterogeneous IoMT and enterprise setting. Extensive experimental assessment based on evaluated healthcare IoMT dataset shows that CeN-CHAIN attains 96.1% global accuracy, 0.958 F1-score, and convergence in 22 rounds, which simultaneously minimizes the attack success rate (ASR) below 5% and a privacy leakage rate (PLR) below 3%. Although the framework includes sophisticated security levels, it has an approachable overhead, and the blockchain anchoring latency is 22–43 ms, energy usage of 2.15–6.15 J/FL round, and CPU usage of less than 77% in IoMT devices. These findings substantiate the claim that CeN-CHAIN provides a balanced trade-off between learning performance and integrated security under the evaluated healthcare IoMT experimental configuration.