Corpus callosum texture analysis: a different perspective approach for diagnosing multiple sclerosis

Area of Science:

• Neuroimaging
• Radiology
• Biomedical Engineering

Background:

• Multiple sclerosis (MS) is a demyelinating disease affecting the central nervous system.
• Conventional MRI may not detect subtle microstructural changes in the corpus callosum (CC) in early MS.
• Early diagnosis of MS is crucial for effective management and treatment.

Purpose of the Study:

• To investigate microstructural changes in the corpus callosum (CC) of MS patients using texture analysis (TA).
• To assess the diagnostic value of TA in differentiating MS patients from healthy controls, particularly in the absence of visible lesions.
• To explore TA as a potential tool for early MS detection.

Main Methods:

• Retrospective analysis of midsagittal T2-weighted MRI scans from 54 MS patients and 50 healthy controls.
• Application of histogram-based texture analysis using MATLAB software.
• Statistical comparison of texture parameters and development of a logistic regression model for MS diagnosis prediction.

Main Results:

• Significant differences in histogram-derived texture features were observed between MS patients and controls (p < 0.001).
• The logistic regression model achieved a diagnostic accuracy of 94.23% in identifying MS patients.
• TA successfully identified MS patients despite the absence of radiologically visible lesions on conventional MRI.

Conclusions:

• Texture analysis of the corpus callosum can reveal subtle tissue alterations in MS patients.
• TA presents a promising, non-invasive method for the early diagnosis of MS.
• TA holds potential as a complementary imaging tool in MS diagnostics, warranting further investigation in larger cohorts.