A Novel Methodology for Brain Tumor Detection in MRI Images Using Support Vector Machine

Finding a brain tumor in medical image is never an easy task. The size, form, and location of the tumor varies depending on patient. Therefore, it's crucial to understand the precise size, form, and location of a brain tumor, making its diagnosis a difficult process. Some patients have tumors with a high glioma (HG) type, while others have tumors with a low glioma (LG) type. Thus, it is important to understand the specific characteristics of a tumor to spot it in a medical imaging.The oriented gradient histogram is utilised in this study to extract feature information from medical image of patient tumours kept in a database. SVM is then utilised to distinguish between tumour and no tumor images.SVM performance based on Receiver Operating Characteristics (ROC) testing is used identify brain tumors in patient images. True positive, true negative, false positive, and false negative rates are all included on the ROC. For all of the medical images in the database, accuracy, sensitivity, and specificity values using ROC are calculated.