No abstract

No abstract

No abstract

Early detection of brain tumors leads to early treatment, which can reduce the mortality rate from brain tumors. Due to the high complexity of brain tissues, manual diagnosis of brain and tumor tissues is very time-consuming and depends on the operator’s condition. There is also a need for experts to examine the images to identify these conditions, which makes the usual and old methods ineffective in the absence of these people. Therefore, the use of automatic techniques for the accurate diagnosis of tumors is very effective. Recently, the examination of brain tumors using magnetic resonance imaging (MRI) techniques has gained much consideration. MRI methods with high abilities in demonstrating the internal organizations of the human body have become one of the most generally applied techniques in the field of brain tumor diagnosis. In the present paper, a new automatic methodology has been introduced for the efficient diagnosis of brain tumors from MRI. In the proposed method, after image preprocessing, a fast method based on Otsu and mathematical morphology is utilized for the segmentation of the section of interest. Then, features of the segmented image have been extracted and the useless ones have been removed based on an optimized feature selection to reduce the method complexity. The optimized method is based on a newly modified design of a metaheuristic, called the Amended Water Strider Algorithm. Then, the features have been injected into an optimized SVM classifier based on the proposed AWSA to get an optimal classification.