An improved classification method for fall detection based on Bayesian framework

Aging population is a big challenge in modern healthcare. Injuries caused by fall incidents are great threats to the elderly. As a consequence, fall detection and movement classification have very high research value and application significance. This paper aims to study the optimum feature subset of falls and verify the effectiveness of the proposed features through experiments. A set of twelve features of eleven kinds of activities are extracted from four sensor data firstly. Then, an improved classifier is proposed based on Bayesian framework for feature selection and fall detection. The optimal features will be selected to reduce the number of features required for the classification problem. Finally, the activity types of unknown samples are predicted using the optimal features and the classifier gained above, and the accuracy of classification will be analyzed. It has been verified that the improved classification method can get a higher detection rate and a lower false alarm rate.