Identification of texture MRI brain abnormalities on first-episode psychosis and clinical high-risk subjects using explainable artificial intelligence

Abstract

Structural MRI studies in first-episode psychosis and the clinical high-risk state have consistently shown volumetric abnormalities. Aim of the present study was to introduce radiomics texture features in identification of psychosis. Radiomics texture features describe the interrelationship between voxel intensities across multiple spatial scales capturing the hidden information of underlying disease dynamics in addition to volumetric changes. Structural MR images were acquired from 77 first-episode psychosis (FEP) patients, 58 clinical high-risk subjects with no later transition to psychosis (CHR_NT), 15 clinical high-risk subjects with later transition (CHR_T), and 44 healthy controls (HC). Radiomics texture features were extracted from non-segmented images, and two-classification schemas were performed for the identification of FEP vs. HC and FEP vs. CHR_NT. The group of CHR_T was used as external validation in both schemas. The classification of a subject's clinical status was predicted by importing separately (a) the difference of entropy feature map and (b) the contrast feature map, resulting in classification balanced accuracy above 72% in both analyses. The proposed framework enhances the classification decision for FEP, CHR_NT, and HC subjects, verifies diagnosis-relevant features and may potentially contribute to identification of structural biomarkers for psychosis, beyond and above volumetric brain changes.

Fig. 1. Histogram equalization to adjust the intensity values.

Representation of a the discretization function and b the initial intensities and the adjusted intensities using histogram equalization. The brain MRI in SPM12 for the c initial MR image and d transformed MR image using the histogram equalization.

Fig. 2. Workflow chart.

Workflow for the calculation of the texture feature maps.

Fig. 3. Visualization of the relevance of the voxels in each class for the classification schema (a).

We demonstrated the smoothed PR with a 7 × 7 × 7 Gaussian kernel of the correct classified subjects of each group against the other in classification schema (a), FEP vs. HC for the registered texture feature map: a difference of entropy and b contrast. The red (cluster 1), blue (cluster 2), and green (cluster 3) color corresponds to the sorted clusters according to the number of subjects belong to each cluster.

Fig. 4. Visualization of the relevance of the voxels in each class for the classification schema (b).

We demonstrated the smoothed PR with a 7 × 7 × 7 Gaussian kernel of the correct classified subjects of each group against the other in classification schema (b), FEP vs. CHR_NT for the registered texture feature map: a difference of entropy and b contrast. The red (cluster 1), blue (cluster 2), and green (cluster 3) color corresponds to the sorted clusters according to the number of subjects belong to each cluster.

Fig. 5. Visualization of the relevance of the voxels for CHR_T.

Classification of CHR_T as FEP using the difference of entropy in classification schema (b). The red (cluster 1) and blue (cluster 2) color corresponds to the sorted clusters according to the number of subjects. The smoothed PR with a 7 × 7 × 7 Gaussian kernel is presented.