Introduction and application of an automatic gait recognition method to diagnose movement disorders that arose of similar causes

Nowadays, researchers try to introduce a more convenient and fast approach (especially a non-invasive one) to diagnose diseases fast and more precisely. Some patient death may be because of wrong diagnosis. Some patients that suffer diseases such as Parkinson disease are known to be dead of wrong diagnosis. Such an approach would help physicians to focus on the correct disease and its treatment and to avoid wasting precious time – that may be critical for the patient – on diagnosis. In this study, we try to develop a new automated approach for classifying (diagnosing) locomotive patients using features that may be extracted from their gait signal. We selected four groups: patients with Huntington’s disease, Parkinson’s disease and Amyotrophic Lateral Sclerosis and a group of healthy subjects. Examining different available classifiers on all proposed features, we have introduced a novel feature with acceptably low error rate using quadratic Bayes classifier.