A secure AI-powered Healthcare 5.0 framework for early detection of MIS-C using intelligent IoH and neuro-fuzzy logic in large-scale IoT environments

Multisystem Inflammatory Syndrome in Children (MIS-C) is a rare post-infectious inflammatory condition associated with SARS-CoV-2. Early recognition is difficult because no single wearable measurement can confirm MIS-C. This study presents a Healthcare 5.0 and Internet of Healthcare framework for early risk screening and continuous monitoring. A Bluetooth Low Energy wearable records temperature, pulse rate, and systolic blood pressure. A layered security design combines lightweight XOR-based device-side message processing with Transport Layer Security, authenticated encryption, secure key management, access control, audit logging, and pseudonymization. An adaptive first-order Takagi–Sugeno neuro-fuzzy model produces a continuous MIS-C risk score that is mapped to low-, moderate-, and high-risk monitoring categories. In a prospective single-center cohort of 120 children from Central Chennai, the model achieved a sensitivity of 0.83, specificity of 0.80, and ROC AUC of 0.88. Six of 36 confirmed MIS-C cases were classified as negative. The framework is intended as a screening and monitoring aid and requires multicenter external validation before clinical deployment.