Hybrid Brain Tumor Detection in MRI Images Using Fuzzy C Means and SVM A Fig share Dataset Based Study

They aim to develop a hybrid numerical framework that can analyze MRI data for the purpose of diagnosing brain cancers. The primary objective was to enhance detection accuracy through the integration of fuzzy C-means clustering with support vector machine classification. The study utilized 3,062 T1-weighted MRI scans that were originally categorized as gliomas, meningiomas, or pituitary tumors. Our may find this dataset on Figshare. The most common type of malignancies found by exploratory data analysis (EDA) were gliomas, pituitary tumors, and meningiomas. The present research employed count and distribution graphs to illustrate this, and additionally examined MRI data from each group to determine if structural changes had occurred. The FCM approach changed all the photos to black and white, made them smaller, got rid of Gaussian noise, made their brightness the same, and divided them into groups during the preparation phase. The segmentation was more accurate because the fuzziness coefficient for FCM could be changed based on the contrast of the image. Applied a fine-tuned support vector machine (SVM) equipped with a radial basis function (RBF) kernel to collect and organize statistical or textural data, including contrast, correlation, energy, or entropy. The proposed hybrid FCM-SVM model outperformed all other options with an F1-score of 96.37%, accuracy of 97.36%, precision of 96.42%, or recall of 95.72%. Because of this, it becomes clear that it is resilient and shows promise as an automated method for finding brain tumors.