A novel approach to the classification of scapular kinematics

Download

Microsoft Word - Presentation Download (119Kb)

Description/Abstract

The complex 3-dimensional rotation of the scapula and the extensive kinematic data produced during motion analysis makes it difficult to objectively classify participants as having movement faults. Classification based on Demspter-Shafer theory of evidence (DST) has been used previously in gait analysis to overcome similar issues, but has not yet been adopted in scapular kinematics. The aim of the study was to adopt the DST method of classifying participants during a motor control test of the scapula. Scapular kinematics were recorded in two groups of participants: healthy control (n=12) and history of shoulder pain (n=6). Six kinematic variables were used as inputs into the DST classifier. The DST successfully classified 5 out of the 6 participants with a history of pain as having movement faults with an accuracy of 72.2% using a leave-one-out cross validation approach. This study demonstrates the use of DST classifier to successfully classify the majority of participants when measuring scapular kinematics during a motor control test.