Discrimination of fatigue in walking patterns

The use of accelerometers has proven to be effective in patient monitoring. The purpose of this work is to evaluate the presence of fatigue in a subject, using a wearable sensor incorporating an accelerometer. The acceleration pattern is recorded during daily activity (walking) both in presence and in absence of physical fatigue. The recorded signals are analyzed to extract a set of features, which are then classified using an artificial neural network. Results indicate the network is able to successfully discriminate the presence of fatigue related to physical activity in the subject.