Comparison Analysis of Brain Image Classification Based on Thresholding Segmentation With Convolutional Neural Network

Brain tumor is one of the most fatal diseases that can afflict anyone regardless gender or age necessitating prompt and accurate treatment as well early discovery symptoms. tumors be identified using Magnetic Resonance Imaging (MRI) to detect abnormal tissue cell development in brain surrounding brain. Biopsy another option, but it takes approximately 10 15 days after inspection, so technology required classify image. The goal this study conduct a comparative analysis greatest accuracy value attained while classifying segmentation with thresholding versus without on CNN method. Images are assigned threshold values 150, 100, 50. dataset consists 7023 MRI scans four types tumors: glioma, notumor, meningioma, pituitary. Without utilising segmentation, classification yielded highest degree accuracy, 92%. At by received score 88%. This demonstrates during model preprocessing less effective for image