Brain Tumor Diagnosis with MCNN-Based MRI Image Analysis

Brain tumors are critical life-threatening medical condition that requires timely and accurate diagnosis for effective treatment. Magnetic Resonance Imaging (MRI) is a widely used and non-invasive medical imaging technique for the detection and diagnosis of brain tumors. In recent years, deep learning approaches, particularly Convolutional Neural Networks (CNNs), have shown remarkable success in medical image analysis, including brain tumor detection. This paper presents a novel approach for brain tumor detection using a Modified Convolutional Neural Network (MCNN) on MRI images. The proposed solution will utilize a deep learning architecture that employs Convolutional Neural Networks (CNNs) for feature extraction and classification. The MCNN architecture consists of a deep Convolutional Neural Networks with a unique combination of convolutional layers, pooling layers, and fully connected layers. Furthermore, we introduce several modifications to the traditional CNN architecture, including the additional layers to improve feature extraction and spatial attention. These modifications aim to address the challenges associated with the complex and subtle nature of brain tumor images in MRI scans. The developed system will be evaluated using standard metrics such as accuracy, sensitivity, specificity, and F1 score. The results will be compared to existing methods for brain tumor detection to demonstrate the effectiveness and potential clinical utility of the proposed approach. The proposed model's superior performance highlights its potential to assist healthcare professionals in early and accurate brain tumor diagnosis, ultimately contributing to better patient care and outcomes. In the proposed CNN model, we observed the average accuracy value on the training data is 98%, with an average loss value of 0.14181. However, the findings on the test data show a significant difference: the average accuracy value on the test data is 90%, with an average loss value of 0.44037.