Detection of complex movement patterns in multivariate kinematic time series for diagnostics in pediatric neurology

Abstract

Evaluation of spontaneous infant movements is an important tool for the detection of neurological impairments. It is necessary to automatically detect movement phases which exhibit certain complex characteristics in order to quantitatively assess these movements. This article presents a method to extract segments of complex movements from multivariate kinematic tracking data. Expert knowledge is represented in a principal component model. Evaluation shows a good concordance between the segments marked by the experts and the results of the automated approach. It is further shown that normal infant movements can be discriminated from pathologic movement patterns.