Convolutional Neural Networks (CNN) based Brain Tumor Detection in MRI Images

• Published in: International Conference on Smart Systems and Inventive Technology (Online) pp. 976 - 979
• Main Authors: Rajeshkumar, C., Soundar, K.Ruba, Sneha, M., Maheswari, S.Santhana, Lakshmi, M.Subbu, Priyanka, R.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 23.01.2023
• Subjects: Brain modeling; Convolutional Neural Network; Convolutional neural networks; Deep learning; Magnetic resonance imaging; Malignant; Medical Practice; Predictive models; Task analysis; Training
• ISSN: 2832-3017
• DOI: 10.1109/ICSSIT55814.2023.10060968

For the purpose of diagnosing brain tumors, Magnetic Resonance Imaging (MRI) is one of the popular diagnostic methods of choice. To determine if a brain tumor has the potential to become malignant, prompt detection plays a crucial role in medical practice. Image categorization is a common task that may be performed quickly and accurately using deep learning. Because deep learning can be used without relying on an expert in the linked subject, it has been extensively used in a variety of industries, including medical imaging. However, in order to achieve effective classification results, a large quantity of different data is necessary. Among the deep learning methods, Convolutional Neural Network (CNN) is indeed the most often used for image categorization. This study has examined two different CNN models to see which one is best suited for classifying brain tumors in MRI images. Ultimately, a CNN model is trained and prediction accuracy has also been increased to as high as 93%.