Classification of involuntary hand movements

Involuntary movements of arms and legs reflect neural and metabolic processes in the human body. In this paper the focus is on the properties of physiological tremor, shivering, and tremors caused by physical fatigue measured in fingers of a subject. Three different signal modeling paradigms are compared in the paper using accelerometer data. It is first demonstrated that the data can be modeled as a nearly stationary low-order AR process. Next, it is shown that the different data types can be classified using long-term feature distributions in a naive Bayes classifier. Finally, a comparable performance is obtained when the signal is modeled as a Markov process emitting small prototypical movements or jerks.