Segmentation with artificial intelligence and automated calculation of the corpus callosum index in multiple sclerosis.

Abstract

Objectives: To determine the corpus callosum index (CCI) differences between chronic phase multiple sclerosis (MS) patients and healthy individuals and to evaluate the corpus callosum segmentation in MS patients using artificial intelligence technologies. The CCI can be reliably measured on magnetic resonance imaging (MRI) and has been proposed as a possible marker of brain atrophy in MS.

Methods: In this study, 150 MS patients (disease duration 12.6±5.9 years) and 150 healthy control subjects were scanned. Corpus callosum index was manually measured from the mid-sagittal slices on MRI. A deep learning architecture-based U-Net model was used for automatic corpus callosum segmentation from 2D brain MRI.

Results: The CCI score was calculated as mean 0.274 in the patient group and 0.382 in the control group (p=0.01). According to the ROC analysis, it was observed that the CCI measurement had a discrimination rate of 98.3% between groups with a cut-off value of 0.334. Sensitivity and specificity were calculated as 94%. The mean CCI calculated automatically after segmentation in the patient group was 0.286.

Conclusion: Corpus callosum index is a method with high sensitivity and specificity in respect of determining corpus callosum atrophy in patients with MS in the chronic phase. Artificial intelligence technologies such as segmentation, machine learning, and deep learning to determine corpus callosum atrophy were seen to be successful in MS patients and the automatically calculated CCI score was successful in showing atrophy.