A Novel Brain Tumor Detection Approach: Integrating Deep Learning and Support Vector Machines

Brain tumors are among the most common causes of death in humans. Early and accurate detection of brain cancers is critical for effective treatment. Imaging techniques such as computed tomography, magnetic resonance imaging, X-rays, and ultrasound are used as a preliminary reference by illness experts. Different learning strategies have been employed in the field of health to diagnose diseases early, reduce the intensity of experts, and minimize diagnostic errors. Image processing studies in brain research have begun to provide successful findings in recent years, thanks to the developed of machine learning and deep learning models. In this study, as a novelty to the studies in the literature, a hybrid algorithm is proposed that features were extracted with pre-trained based CNN, classification was made with SVM based different kernels. As a result, the brain tumors were detected with 98% classification performance.