Brain Tumor Classification and Detection Using Machine Learning Algorithm

• Published in: 2022 International Conference on Augmented Intelligence and Sustainable Systems (ICAISS) pp. 366 - 373
• Main Authors: Barakala, Monisha, Attada, Venkata Ramana, Rajan, Cristin
• Format: Conference Proceeding
• Language: English
• Published: IEEE 24.11.2022
• Subjects: Adaptive Bilateral Filter; Brain Tumor; Convolution Neural Network; Feature extraction; Magnetic resonance imaging; Magnetic separation; Sensitivity and specificity; Thresholding (Imaging); Training; X-rays
• DOI: 10.1109/ICAISS55157.2022.10011031

The human brain is the main part of the humanoid system that makes it work. Brain tumors are caused by cells that grow and divide in the brain in ways that aren't supposed to. As brain tumors grow, they can turn into brain cancer. Computer vision is very important to human health because it takes away the need for people to make accurate decisions. When it comes to magnetic resonance imaging (MRI), CT scans, X-rays, and MRI scans are the most common and safest ways to get an image. MRI can find things that are very small. In our paper, we want to talk about the different ways that brain MRI can be used to find brain cancer. In this study, we used the bilateral filter (BF) to remove noises from an MR image before it was processed. The tumor region was then found using the binary thresholding and Convolution Neural Network (CNN) segmentation techniques. There are training datasets, test datasets, and validation datasets. We will be able to tell from our machine if the subject has a brain tumor or not. Several measures of performance, such as accuracy, sensitivity, and specificity, will be used to look at the results. These dense layers extract features and all features are passed to a fully connected layer. Dense network extract features more efficiently from brain MRI. This work is experimented on MRI as MRI provides more details of cell structure and functions. It is hoped that the proposed work will do better than similar works.