Detection of the Imbalance Step Length using the Decision Tree

Step balance is one of the elements that will be examined for stroke patients' walking training. The step imbalance indicates difference in walking between both the right and left legs and can be used to identify patient training progress. The purpose of this study is classifying the step imbalance which later can be used for monitoring the progress during walking training. An MPU6500-based inertial measurement unit (IMU) sensor is attached to the patient's leg to monitor the leg's position. The IMU sensor data is acquired with the sampling rate of 0.01 second. The IMU sensor's signal is then preprocessed and exported to obtain the signal feature. The signal feature including the positive and negative peak of the acceleration and angular velocity data are used in the study. Based on these features, the decision tree will be used to classify the user's step into balance or imbalance. The system yields a good result with the classification error 15.3%. The proposed method could be implemented to classify the imbalance steps while conducting walking training.