Aim of the study:
The aim of this study was to examine Magnetic Resonance (MR) images of the dorsal striatum, which is known to play a role in the etiopathogenesis of obsessive-compulsive disorder (OCD), using a texture analysis method.

Subject or material and methods:
Twenty-eight OCD patients with brain MRI images stored in the hospital Picture Archiving and Communication System were included. The control group consisted of 28 healthy individuals. T1 weighted MRI images were obtained with 1.5T systems. All regions of the putamen, and the caudate nucleus were selected individually from the axial images that best represented the anatomy, without exceeding their borders. Symmetrical locations were used as separate samples (56 samples each of the caudate and putamen for both groups).

Results:
Most of the parameters for both locations showed statistically significant differences between the patient and controls. ROC curve analysis showed high sensitivity and specificity for the mean values for both nuclei. Logistic regression analyses indicated that the calculated parameters could predict the probability of participants having OCD.

Discussion:
The current methodology provided better results compared to structural approaches. It identified the structures that could not be visually assessed based on microstructural information and provided more valuable information when compared to shape-based measurements.

Conclusions:
Tissue parameters in the dorsal striatum of OCD patients were found to be different from healthy controls. Thus, tissue analyses can be a useful technique for demonstrating tissue changes in the dorsal striatum using MRI images of OCD patients.