Main Article Content
MRI (Magnetic Resonance Imaging) is a crucial medical imaging modality for diagnosing and characterizing brain tumors. Extracting informative features from MRI brain images is essential for accurate tumor analysis and decision-making. Traditional Principal Component Analysis (PCA) has been widely used for feature extraction, but its linear assumption limits its ability to capture the complex nonlinear structures present in MRI images. To overcome this limitation, an improved version of PCA, known as Kernel PCA or KPCA, has been introduced. This research investigates the application of KPCA for MRI brain image feature extraction, aiming to enhance the representation of intricate tumor patterns and structures.