Hybrid Approach for Brain Tumor Detection and Classification Using MRI

Detecting a brain tumor might be the difference between life and death. Because of their enhanced performance, machine learning-based techniques for detecting brain cancers have become popular in recent years. However, for these machine learning-based systems to function well, they require many tagged photos. Such data is generally arduous, time-consuming, and prone to human mistakes, making deploying machine-learning approaches challenging. This study combines conventional machine learning approaches with various tests to diagnose brain tumors. The NLM (Non-Local Mean) filter is used for preprocessing and noise removal. Following denoising, Otsu thresholding is used to partition the tumor to maximize variation between classes. The GLCM (Grey Level Co-occurrence Matrix) approach extracts features. Machine learning (ML) techniques such as K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Decision Tree (DT), and Ensemble are given access to the created feature set. The proposed method achieves good accuracy, sensitivity, and specificity in identifying and categorizing brain tumors, according to experimental data. The suggested method may be a valuable instrument for doctors to precisely identify brain tumors, enabling better patient care and treatment.