A Hybrid Model of Faster R-CNN and SVM for Tumor Detection and Classification of MRI Brain Images

In the last few decades the death rate based on cancer has increased tremendously, in this case, lung cancer, breast cancer, and brain cancer are at the top rate of these diseases. Among all 7% includes brain cancer. Brain cancer due to the complex architecture and different in size and shape is a challengeable profession in medical image analysis. MRI is a technique of medical imaging for analyzing brain cancer. The traditional methods of machine learning (ML) that needs handcraft and also radiologist specialist to examine. That can lead the procedure on failure and also reduces system effectiveness. Meanwhile in the last few years deep learning (DL) due to good performance more usable in image classification. In the present research study, the MRI images analyzed to identify the tumor area and categorize these areas into being and malignant. Deep-learning is a powerful technique that recently has been used extensively on image classification. Accordingly, Fast R-CNN is an improved deep leaning technique and with a combination of the SVM model has been implemented via the Open CV library in this study. It shows that hybrid Faster R-CNN and SVM yield an accuracy of 98.81%. however, the finding of previous studies that used different techniques got accuracy (R-CNN91.66%, CNN & RBF based SVM-92%, RELM-94.233%, KNN-96.6%, CNN-97.5%), Meanwhile