A Prediction Method of Elderly Cognitive Impairment Based on Edge Intelligence and Multimodal Perception

Abstract

The situation of population aging in our country is severe, and the incidence of cognitive impairment in the elderly is gradually increasing, which seriously affects the lives of the elderly. Monitoring the daily behavior of the elderly in the early stage of cognitive impairment, proactively finding out the signs of cognitive impairment, and providing corresponding cognitive training in time can effectively delay the development of cognitive impairment. To find a prediction method which can identify the signs of cognitive impairment under the premise of protecting user privacy, this article uses the Internet of Things system to monitor and record the daily activities of the elderly. Then Back Propagation Neural Network (BPNN) is utilized to assist the diagnosis of cognitive impairment. This paper carried out experiments to verify the accuracy rate of acquisition and that of recognition by BPNN. The experimental results show that the system has better feasibility and higher recognition accuracy.