A proposal of abnormal condition detection system for elderly people using wireless wearable biosensor

In this paper, we described an abnormal condition detection system that uses a wireless wearable biosensor that can observe triaxial acceleration, ECG, and the temperature of the surface of the body. We implemented a feature that detects falls to determine a hard impact from acceleration data. Our system achieved a fall detection rate of 90.9% with a 2.8% false positive alarm rate and a 9.1% false negative alarm rate. We also implemented a feature that detects an abnormal heart rate and we determined that the feature works correctly.