A0204
Title: Multiclass machine learning classification of functional brain images for Parkinson's disease stage prediction
Authors: Guan-Hua Huang - National Yang Ming Chiao Tung University (Taiwan) [presenting]
Abstract: A data set is analyzed that contains functional brain images from 6 healthy controls and 196 individuals with Parkinson's disease (PD), who were divided into five stages according to illness severity. The goal was to predict patients' PD illness stages by using their functional brain images. We employed the following prediction approaches: multivariate statistical methods (linear discriminant analysis, support vector machine, decision tree, and multilayer perceptron [MLP]), ensemble learning models (random forest [RF] and adaptive boosting), and deep convolutional neural network (CNN). For statistical and ensemble models, principal component analysis was performed to extract features, and synthetic minority over-sampling technique (SMOTE) was used to deal with imbalanced data problems. For CNN modeling, we applied an image augmentation technique to increase and balance data sizes over different disease stages, and we adopted a transfer learning idea to bring pre-trained VGG16 weights and architecture into the model fitting. It was found that MLP and RF were the analytic approaches with the highest prediction accuracy rate for statistical and ensemble models, respectively. Overall, the deep CNN model with pre trained VGG16 weights and architecture outperformed other approaches.
