Brain tumor detection in MRI images using Adaptive-ANFIS classifier with segmentation of tumor and edema

The brain is a significant organ that controls all activities of the body parts. A Brain Tumor (BT) is a group of tissues, which are structured by the gradual accumulation of irregular cells. The tasks namely (a) identification, (b) segmentation along with (c) identification of the infected region in BT utilizing MRIs is slow and wearisome. Here, BT Detection (BTD) in MRI images has been proposed utilizing Adaptive-ANFIS (Adaptive-Adaptive Neuro-Fuzzy Inference System) classifier with the segmentation of tumor along with edema. Primarily, the input RGB is transmuted into a Grayscale Image (GSI). During pre-processing, the non-brain tissues are eradicated utilizing the Skull Stripping Algorithm (SSA). Next, the resulted image is segmented into ‘2’ parts: (a) tumor and (b) edema utilizing Modified Region Growing (MRG) and Otsu’s thresholding. Then, as of the segmented “tumor part” image, the DWT, WST, Edge, and color histogram features are extracted. Then, the required features are selected as of the extracted features by employing the MGWO algorithm. After that, those features being selected are given to the Adaptive-ANFIS, which categorizes the tumor as (a) Benign and (b) Malignant. The adaptive process is conducted by the K-Means Clustering (KMC) algorithm. Experiential results examined the performances defined by the proposed as well as the prevailing system regarding specificity, accuracy, sensitivity, recall, and precision.