Segmentation and Validating the Accuracy of Classification Algorithms in Brain Tumor Magnetic Resonance Images using Machine Learning Approach

Magnetic Resonance Image (MRI) segmentation used for brain tissues extraction white matter and gray matter. These tissues help with many medical image segmentation applications such as radiotherapy planning, clinical diagnosis, treatment planning and Alzheimer disease. However, while finding of small structural brain differences may fundamentally depend on the method used, both accuracy and reliability of different automatic segmentation algorithms are compared. In this article, the performance of the segmentation algorithms provided by SPM8, VBM8, FSL and Free Surfer was counted on replicated and real magnetic resonance imaging data. First, accuracy was measured by matching segmentations of ten simulated and twenty real T1 weighted brain MRI images with equivalent ground truth images. Second, reliability was determined in five T1 images from the same subject and in five T1 images of different subjects. Then the effect of pre-processing stages on segmentation accuracy was examined. VBM8 showed a very high accuracy and a very high reliability. FSL achieved the highest accuracy but demonstrated poor reliability and Free Surfer showed the lowest accuracy, but high reliability. Though, our examination proposes that researchers can enhance their distinct dispensation measures with respect to segmentation excellence and demonstrates passable recital principles.