Classification of Patients with Disorder of Consciousness Based on DTI Sequence Analysis
Abstract—In this paper, a method is proposed for classification of patients with disorder of consciousness (DOC) based on the diffusion tensor imaging (DTI) sequences analysis. The patients are divided into vegetative state (VS) and minimally consciousness state (MCS). Firstly, tract-based spatial statistics (TBSS) was applied to find the regions of interest (ROIs), and the values of fractional anisotropy (FA), mean diffusivity (MD) of ROIs were calculated subsequently. Secondly, statistical analysis, including t-test and Spearman correlation analysis were used to obtain the parameters with significant difference between VS and MCS and to extract the parameters significantly correlated to Coma Recovery Scale-Revised (CRS-R) scores. Finally, a classifier based on support vector machine (SVM) was trained with parameters of ROIs. Results show that a 92.31% accuracy was achieved with age and gender as extra classification features, and the confidence of classification result can be used to evaluate the level of consciousness of patients.
