Magnetic Resonance Imaging (MRI) is widely used for detecting abnormal brain tissues; however, accurate tumor identification and classification remain challenging due to image variability and noise. This study proposes an efficient framework for automated brain tumor detection and classification. Initially, MRI images undergo preprocessing, including denoising and skull stripping, to enhance image quality. Tumor segmentation is performed using an optimized active contour model with a tuned weighting factor for precise boundary extraction. Discrete Wavelet Transform (DWT)-based features are then extracted to capture discriminative spatial–frequency information. These features are classified using an Optimized Deep Belief Network (DBN), where network weights are fine-tuned using the Improved Bat Algorithm (BA). Compared to conventional optimizers, the BA-based optimization provides faster convergence, better global search capability, and reduced risk of local minima, leading to improved classification performance. The proposed model is evaluated on a benchmark dataset using metrics such as accuracy, precision, sensitivity, and specificity. Experimental results demonstrate that the proposed approach achieves superior performance and robustness compared to existing methods for brain tumor detection.
Ramasamy et al. (Tue,) studied this question.