Identification and Classification of Brain Tumors with Optimized Neural Network and Canny Edge Detection Algorithm

Abstract

Radiological presentation, clinical signs and frequent histopathology are currently identified and treated for brain tumors. Magnetic resonance imaging (MRI) is an effective non-invasive technique for the anatomical examination of brain tumors. Huge diagnostic issues, such as the grade and form of the tumor, are still difficult to solve using MRI. In recent years, brain tumor disclosure using MRI images has been a powerful area of clinical research. MRI is an efficient method for the safe visualization of an internal structure within a body. This includes the ability to record signals that can differentiate between divergent 'soft' tissues (like grey matter and also white matter). A brain tumor is a very pernicious disease that causes many people to die. In addition, the detection and stratification system should also be available, so it could be diagnosed at earlier stages. In addition, an intuitive and simple approach is to implement closed canny edge detection. In addition, each time, it introduces closed lines around regions. The shape, intensity and texture are then extracted by feature extraction methods from the apportioned image attributes. The value of the extracted characteristics is then entered into the ANN classifier to stratify the normal and abnormal images.