Statistical Review of Health Monitoring Models for Real-Time Hospital Scenarios

Abstract

Health Monitoring System Models (HMSMs) need speed, efficiency, and security to work. Cascading components ensure data collection, storage, communication, retrieval, and privacy in these models. Researchers propose many methods to design such models, varying in scalability, multidomain efficiency, flexibility, usage and deployment, computational complexity, cost of deployment, security level, feature usability, and other performance metrics. Thus, HMSM designers struggle to find the best models for their application-specific deployments. They must test and validate different models, which increases design time and cost, affecting deployment feasibility. This article discusses secure HMSMs' application-specific advantages, feature-specific limitations, context-specific nuances, and deployment-specific future research scopes to reduce model selection ambiguity. The models based on the Internet of Things (IoT), Machine Learning Models (MLMs), Blockchain Models, Hashing Methods, Encryption Methods, Distributed Computing Configurations, and Bioinspired Models have better Quality of Service (QoS) and security than their counterparts. Researchers can find application-specific models. This article compares the above models in deployment cost, attack mitigation performance, scalability, computational complexity, and monitoring applicability. This comparative analysis helps readers choose HMSMs for context-specific application deployments. This article also devises performance measuring metrics called Health Monitoring Model Metrics (HM3) to compare the performance of various models based on accuracy, precision, delay, scalability, computational complexity, energy consumption, and security.