Radiomics for Parkinson's Disease Classification Using Advanced Texture-Based Biomarkers

Abstract

Parkinson's disease (PD) is one of the neurodegenerative diseases whose complete cure is not found to date. Therapies and medications are supportive methods to deal with symptoms. There is always a requirement of medical domain expertise to diagnose PD manually. Since manual diagnosis leads to a time-consuming process, an automatic technique has always been useful in such complex tasks. Magnetic resonance imaging (MRI) based computer-aided diagnosis helps medical experts to diagnose PD more precisely and fast. Texture-based radiomic analysis is carried out on 3D MRI scans of T1 weighted and resting-state modalities. 43 subjects from Neurocon dataset and 40 subjects from Tao Wu dataset were examined, which consisted of 36 scans of healthy controls and 47 scans of Parkinson's patients. Total 360 2D MRI images are selected among around 17000 slices of T1-weighted and resting scans of selected 72 subjects. Local binary pattern (LBP) method was applied with custom variants to acquire advanced textural biomarkers from MRI images. LBP histogram formation helped to learn discriminative local patterns to detect and classify Parkinson’s disease. Using recursive feature elimination (RFE), the data dimensions of around 150-300 LBP histogram features were reduced to 13-21 most significant features based on the score of each feature, and these important features were analysed using supervised machine learning algorithms like support vector machine (SVM) and random forest. The variant-I of LBP has performed well with the highest test accuracy of 83.33%, precision of 84.62%, recall of 91.67%, and f1-score of 88%. Classification accuracies were obtained from 61.11% to 83.33% and area under the curve-receiver operating characteristics (AUC-ROC) values range from 0.43 to 0.86 using four variants of LBP.