Analysis of sub-cerebellar regions in patients with Chiari Malformations

Chiari Malformations are serious neurological defects involving herniation of hindbrain tissues such as cerebellar tonsils, brainstem and IV ventricle into the spinal canal through the foramen magnum. By the severity of cerebellar descent, these malformations are classified into four different types. Clinically the least obvious and the mildest one is named as type I and defined as the descent of cerebellar tonsils into the cervical canal more than 5 mm. Magnetic Resonance Images (MRI) of brain in the sagittal plane provides the best clues in the diagnosis of the Chiari Malformation type I (CM-I). Previous studies investigated the morphological characteristics of cerebellum and nearby regions such as brain stem and fourth ventricle. Aim of this study is to analyze the cerebellar regions in chiari patients and healthy controls to search for the discriminative properties between the two groups. Sagittal brain MRI of eleven chiari patients and gender matched controls were used in order to examine the area of sub-cerebellar tissues such as gray matter (GM) and white matter (WM) and the area ratio between GM and WM. A graphical user interface (GUI) for implementing image processing techniques was developed using MATLAB environment. By means of GUI, the region embracing the whole cerebellum tissue on the mid-sagittal MR images were manually extracted. In addition, using Statistical Parametric Mapping (SPM) package the MRI slices were segmented into GM and WM tissues. Using the extracted cerebellum region as a mask, the cerebellar GM and WM tissues were achieved and the corresponding areas were computed by counting the number of pixels on each GM and WM slice. According to the statistical results, it has been found that cerebellar GM areas of the patients are significantly higher than the values of controls. As a consequence, this approach may provide a discriminative feature between patients with CM-I and health control subjects. Keywords : Chiari malformation, magnetic resonance imaging, segmentation, gray matter, white matter