Brain Tumor Detection Using CNN

Abstract

This paper presents a comprehensive approach for brain tumor detection using a dataset comprising MRI images. The process involves dataset preprocessing, splitting into training and testing sets, image extraction, and labelling. A Convolutional Neural Network (CNN) model with 23 layers is proposed for the detection task, alongside an exploration of the VGG16 model for comparison. The CNN architecture is meticulously designed to extract intricate features from the MRI images. The proposed model's architecture and performance are thoroughly analysed and compared with the VGG16 model. Results indicate promising detection accuracy, demonstrating the effectiveness of the proposed approach in aiding medical professionals in diagnosing brain tumors accurately and efficiently. Furthermore, insights gained from the reflection on the 23 layers CNN architecture provide valuable perspectives for future advancements in medical image analysis using deep learning techniques. Finally it uses deep learning based Depth wise Separable Convolution Neural Network to detect the tumor based on the MRI images.