Medical activity monitoring for elderly people using wearable wrist device

This paper discusses the design and implementation of a wearable sensor based framework for medical activity recognition of elderly patients. Tri-axial accelerometers worn on the dominant hand, monitor hand movements to recognize activities. Feature extraction is done based on Hilbert-Huang Transform (HHT) using Empirical Mode Decomposition (EMD). The dataset of activities is collected from 10 subjects and comprises of the following activities: opening a pill box, popping a pill into the mouth, drinking water, taking medicine in a syringe and self-injecting. Activities are classified using decision tree, k-nearest neighbor and support vector machine classifiers. The activity recognition system is integrated with a GSM modem that is used to automatically prompt appropriate alerts to caretakers about the patient's physical activities. This will have huge potential in reducing the cost of providing health care services to the elderly.