Parkinsons disease classification using wavelet transform based feature extraction of gait data

Parkinsons disease (PD) is one of the most common degenerative disease of the nervous system. In this paper, a method is proposed to distinguish PD patients from healthy people using wavelet transform based feature extraction of gait characteristics. This approach could help the physicians for early diagnosis of the disease and to start the treatment at the beginning stage of the disease. Wavelet Transform (WT) of preprocessed gait data was performed and statistical features were extracted from these coefficients. Artificial Neural Network (ANN) based classifier has been used in this paper and performance of the model with various efficient Back Propagation (BP) algorithms were evaluated. A comparative study on classification using different classifiers such as Support Vector Machine (SVM), Naive Bayes classifier has also been done. Experimental results have demonstrated very good performance on classifying PD patients.