Brain Tumor Segmentation on MR and CT Images Using Fuzzy C-Means and Active Contour Methods

Abstract

A brain tumor is a dangerous brain disease that can attack anyone. It can be described as the abnormal growth of cells in or around the brain, leading to impaired brain function. The first step in diagnosing a brain tumor is to perform an MRI (Magnetic Resonance Imaging) scan. The research aims to analyze the segmentation results of brain tumor MRI and CT (Computed Tomography) images using the Fuzzy C-Means and Active Contour methods. The evaluation is based on ROC parameters, including accuracy, dice score, precision, and sensitivity. The methodology involves analyzing data from secondary image sources, using MATLAB for the segmentation process, and evaluating the results of image segmentation by radiologists. Four ROC measurements were used for each method. The segmentation evaluation results for MRI images show that the Fuzzy C-Means method achieved a precision of 0.92; sensitivity of 0.64; dice score of 0.76; and accuracy of 0.61. The Active Contour method, on the other hand, obtained a precision of 0.97; a sensitivity of 0.99; a dice score of 0.98; and an accuracy of 0.96. For CT images, the Fuzzy C-Means method yielded a precision of 0.72; sensitivity of 0.98; dice score of 0.83; and accuracy of 0.71. The Active Contour method obtained a precision of 0.96; a sensitivity of 0.95; a dice score of 0.96; and an accuracy of 0.92. These results indicate that the Active Contour method, especially with MRI images, provides better segmentation performance. In conclusion, the segmentation results from the Active Contour method can be used as additional information for doctors in diagnosing the presence of tumors.