An Intelligent Wellness Assessment System for the Elderly Healthcare

Abstract

Due to the rapidly increasing aged population in China, the demand for institutional healthcare service is growing. In this paper, we proposed an intelligent system which could use the daily activity data of the elderly living in a stable environment, like nursing home, to predict their wellness conditions. This system has many advantages, such as high flexibility, high reliability, and low cost. Daily activities of the elderly could be recognized by analyzing their location information, and the location data can be collected by using Radio Frequency Identification (RFID) devices. The wellness conditions could be determined by using a novel framework integrating temporal and spatial contextual information. We initiated this research with the development of the intelligent system and conducted the testing experiments at the geriatrics department in the third people's hospital of Lanzhou. Daily location data of 5 male elderly patients were collected for data analysis. A Support Vector Machine (SVM) model and a Back Propagation Neural Network (BPNN) model were constructed to determine the wellness conditions of the subjects. The experimental results were encouraging and it indicated that this proposed intelligent system could assist in providing prompt healthcare services to elderly people.