Gait Classification Using A Support Vector Machine Algorithm

The gait pattern, as well as the walking process itself, can be an indicator of the overall health of patients. For this reason, it is very important to accurately, clearly and quickly determine the affiliation of the gait pattern (healthy or pathological) and take appropriate measures if necessary. As anterior cruciate ligament injuries are common and may be undetectable, this study presents a classification of gait using a support vector machine (SVM) algorithm. The test data were taken from a Gait LAB laboratory; anterior posterior translation and internal external rotation were used as significant parameters. The classifier performance was evaluated using a confusion matrix. These results showed that the SVM algorithm can be successfully used in tasks of this type of classification.