Age Differences Classification Associated with Corpus Callosum Measurement

• Published in: 2022 IEEE International Conference on Computing, ICOCO 2022
• Main Author: Ibrahim S.; Jelaini N.A.S.M.; Md Ghani N.A.; Janor R.M.; Ali M.H.
• Format: Conference paper
• Language: English
• Published: Institute of Electrical and Electronics Engineers Inc. 2022
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85148441274&doi=10.1109%2fICOCO56118.2022.10031802&partnerID=40&md5=822cbad2b86a7ff29c24d7b28aaaac49

A medical visualization is a tool used in medicine to detect aspects of the human body in terms of digital health. The corpus callosum is a large white matter structure that separates the two hemispheres of the brain. It is an extremely essential structural and functional component of the brain. Assessing the corpus callosum measurement could reveal the information on age differences category of each individual, as well as atypical growth such as multiple sclerosis (MS), Alzheimer's, and autism spectrum disorder (ASD). Thus, this study proposed the use of Magnetic Resonance Imaging (MRI) sagittal brain images to classify age differences associated with corpus callosum measurement. Three age differences were studied; children (0-10 years), adolescent (10-18 years), and adult (18-25 years). The present results provided evidence that adult and children differ in terms of developmental trajectories for the brain structure, with significant age-related changes discernable from infancy to early adulthood. A few steps of MRI corpus callosum image collection, Median Filtering image enhancement, Otsu binarization, and K-Means clustering segmentation, corpus callosum measurement, and Support Vector Machine (SVM) classification were involved. The performance of the corpus callosum classification was evaluated using a confusion matrix. The overall mean percentage of accuracy reflected a very high accuracy which are 97.72%, 95.56%, and 97.72% for children, adolescent, and adult respectively. It can be deduced that the proposed techniques of corpus callosum classification are found to be successful. © 2022 IEEE.