PSO-BP neural network based on physical health perception of the elderly in retirement home

The recreational community retirement home staff can keep abreast of the elderly's health conditions and provide personalised services. However, how health status is collected and judged is more traditional, and there is a lack of prediction of the healthy development of the elderly. The information and intelligent upgrading of the elderly recreation system is imperative, and neural network modelling based on the Internet of Things and its intelligent optimisation methods is one of the means to do this. Based on 5G IoT technology, this paper adopts BP neural network and particle swarm optimization modelling methods to use the physical health and long-term living environment health indicators collected by IoT terminals as model input information. Through selecting evaluation indicators, the construction of evaluation models, the establishment of a primary database and the determination of evaluation criteria, intelligent analysis and optimisation processes are carried out to form a health posture awareness system for the elderly in recreational community retirement home. A comparison of the simulation results shows that the proposed geriatric health situational awareness system has improved the efficiency and accuracy of its geriatric health awareness capabilities.