Characterization of Corpus Callosum using Magnetic Resonance Imaging for Sudanese Population

Abstract

Morphometric of the corpus callosum (CC) are important to have normative values according to gender, age and race/ethnicity. The present study examined the correlation between age, gender, and CC morphometrical data, across ages <10> 60 years old to characterize normal developmental alternations in order to be a standard local reference for Sudanese. At issue the objectives are also to examine whether CC index continues to increase throughout life, whether there are regional differences in measurements of CC maturation, and whether these outlines are sexually demographic. As well this study concern to characterize the corpus collosum parts to splenium, trunk and genu using first order statisticand extract classification features from MR images. The CC on magnetic resonance midsagittal T1 weighted images was measured in 233 normal Sudanese subjects, (126 were males constituting 54.1% and 107 were females constituting 45.9%) admitted to Royal care international hospital, and scanned with MRI scanner of 1.5 Tesla (Toshiba ) during the period extended from 2014 – 2017. Considering age and gender; Fronto occipital maximum brain length, Thickness of CC compartments at its maximum level for rostrum, genu, body/ trunk and splenium, CC greatest anteroposterior (AP) diameter, fronto-corpus callosum length, occipito-corpus callosum length and corpus callosum index (CCI) were measured as well the first order statistic (FOS) techniques included eight’s features which are mean, variance, coarseness, skewness, kurtosis, energy and entropy to find the gray level variation in MR images it complements the FOS features extracted from MR images with variation of gray level in pixels and estimate the size variated of the sub patterns. Data were analyzed using statistical package of social science (SPSS) program (Version.16), for texture analysis analyzing the image with interactive data language (IDL) software to measure the grey level variation of images. Result showed that all dimensions of CC compartments in males were (rostrum =1.27mm, genu =11.33mm, trunk =5.91 and splenium =10.31mm ) and for female were (rostrum =1.21mm, genu =11.83mm, trunk =6.56 and splenium =11.04 mm) and brain dimensions had significant relation with increasing age except for occipito-corpus callosum length where no significant relation was detected (p=0.126). The gender has an impact on the changes detected in the brain and CC compartments except for the rostrum, genu, callosal AP maximum diameter. Brain dimensions were significantly larger in males than in females at (P ≤ 0.05). Another findings were found in the CC trunk, splenium maximum thickness and CCI, where females were greater than males with significant difference at P= 0.000, 0.011 and 0.031 respectively .The CCI increased with age and then decreased thereafter. There was also a positive linear relationship between the AP length of the CC and the fronto-corpus callosum length. Regression equation for predicting the length of the CC and morphometric index as local reference for normative data of CC during maturation in the Sudanese population in both genders at similar age classes have been established. The texture analysis results showed that the first order statistic and features give classification accuracy of corpus collosum parts for splenium 100.0%, trunk76.5% and the genu classification accuracy 97.0%. The overall classification accuracy of corpus collosum area 96.2%.These relationships are stored in a texture dictionary that can be later used to automatically annotate new MR images with the appropriate corpus collosum area names.