Balance Evaluation by Inertial Measurement Unit

Abstract

Falls among the elderly population have become a major health problem in developed countries. Recently, much effort is being devoted to studying their causes and researching methods to predict them. Medical doctors commonly use test protocols and big and expensive medical devices like a posturograph to calculate an equilibrium score, diagnose balance problems, and asses the risk of fall of their patients. In this paper we propose an alternative method to assess the falling risk based on a low-cost wearable device that can be used at home without requiring a visit to the health center. The proposed device uses an Inertial Measurement Unit (IMU) to compute balance values. Our model has been trained on data acquired from 19 elderly patients performing the Romberg test on a Posturograph. The posturograph scores have been used as a ground truth for the training process. The results show that it is possible to detect the posture variations of patients wearing an IMU while performing the Romberg tests.