3D MRI Based Brain Tumor Detection and Classification Using Multi Model Algorithms and Prediction by Modal Value

Abstract

Magnetic Resonance Images(MRI) are essential means and have a vital role in diagnosing Braintumors. Even though Machine Learning algorithms give accuracy to some extent, still accurate prediction cannot be achieved in the medical field for the classification of Brain tumors. In this paper, Multiple algorithms are used to improve the prediction accuracy. 3D MRI images are used since they are able to offer more features than 2D images. Segmentation is carried out using watershed algorithm. The GLCM (Grey Level Co-occurrence Matrix) is used for feature extraction of the segmented region. In this paper modal values of three algorithms are used. In a dataset, a value that occurs number of times is termed as the mode or modal value. Three Machine Learning algorithms (SVM, KNN,CNN) are used and the Modal Value Prediction (MVP) is the Final Prediction achieved based on Modal value.