Brain Tumor Detection Using MRI

ABSTRACT - Detecting and classifying tumors in brain MR images is crucial for early diagnosis and effective treatment. Early detection improves long-term survival rates. This paper presents a new approach for accurate brain tumor segmentation and classification. First, the MR images are enhanced using an anisotropic diffusion filter. Then, an active contour model is applied to detect the tumor, providing smooth and precise contours. For classification, features from the tumor images are extracted using 2-D Daubechies wavelet transform (DWT), and the feature dimensions are reduced with Independent Component Analysis (ICA). Finally, a trained Support Vector Machine (SVM) with different kernels (KSVM) is used to classify tumors as benign or malignant. Key Words: Brain Tumor Detection, Magnetic Resonance Imaging (MRI), Deep Learning, Convolutional Neural Networks (CNN), Image Segmentation, Tumor Classification , Medical Imaging, Computer-Aided Diagnosis