Brain tumour Detection using a Modified Convolutional Neural Network

Abstract

Brain tumors represent a formidable healthcare challenge worldwide, and their early detection is critical for timely intervention and improved patient outcomes. This survey research explores the realm of deep learning algorithm with direct attention to convolutional neural networks (CNN) for the timely detection of patients with brain tumor. The paper explores the 5-layer cnn model, which has been a widely used approach in previous research, with a 9-layer CNN-model within the framework of brain tumour detection. a comparative analysis is presented, highlighting the strengths and limitations, architectural intricacies, model complexity, performance assessments and other significant features. The findings from the 5-layer cnn serve as a baseline for comparison in our research. This study uses the 2020 brat's dataset and the experimental result are the proof the model's efficacy that surpasses the performance of existing models with an accuracy of 99.74%, previously unmatched. Paper concludes by emphasizing the relevance and importance of utilizing deep learning, specifically a 9-layer CNN model, for early and accurate detection of brain tumours. This research contributes valuable insights to the field of medical imaging and deep learning, providing a basis for future work in the development of more accurate and effective tools for early brain tumor detection, ultimately leading to improved patient care and outcomes.