Efficient Brain Tumor Detection with a Lightweight Customized CNN Approach

Brain tumor detection is a critical task in medical imaging, requiring accurate and efficient classification methods. In this work, we propose a lightweight customized Convolutional Neural Network (CNN) model designed to achieve high accuracy with minimal computational complexity. Our model significantly reduces the number of tunable parameters while maintaining robust feature extraction capabilities. Evaluation results show the proposed system reaches 98.03% accuracy in classification tasks, demonstrating significant improvement over existing methods., outperforming several predefined deep learning models. The reduced compuational overhead makes the model suitable for real-time and resource-constrained environments, such as edge devices and mobile healthcare applications. This study highlights the potential of customized CNN architectures in advancing brain tumor diagnosis, ensuring both efficiency and accuracy in medical image analysis.