Rough set based classifications of Parkinson's patients gaits

Abstract

Motion capture (MoCap) technology becomes recently often used in neurological applications, especially for diagnosis of gait abnormalities. In this paper we present several different approaches to compute important features of gait abnormalities. This is a continuation of our previous experimental results concerning examination of Parkinson's disease (PD) with bilateral subthalamic nucleus stimulation (DBS) patient in the MoCap laboratory. At first, we calculate mean changes of the gait as effects of medication and DBS. We present these changes as phase plots suggesting different dynamics in different patients. In the second part, we apply AI approach related to application of the Rough Set Theory in order to generate decision rules for all our patients and all experiments. We have tested these rules by comparing training and test sets. © 2014 Springer International Publishing Switzerland.