Linguistic Severity Range Fixation of Vital Signs Using Unsupervised Approach in RHM

Abstract

Automatic abnormality detection in human health status based on the variation in the vital health parameters is a continuous research thrust area. After Covid pandemic the importance of checking the variation in the health status become a part of our regular activities. With the help of artificial intelligence, today many research works have been proposed in abnormality detection. The proposed work is personalized abnormality detection technique based on adaptive unsupervised mechanism and tries to map the health status with the incoming health stream data. The proposed adaptive density-based K-Means fixes the severity range of each vital health parameter of a person and achieved an accuracy rate in fixing the severity range with 91.3% during training and 87.8 % testing respectively.