Advancements in Brain Tumor Detection: Exploring Diverse CNN Architectures

Machine Learning and Deep learning are some of the powerful tools to analyze and process medical data. The data generated by the medical sector is huge in volume, at a high velocity and needs to be processed properly to mine as much information and insight as possible. This paper talks about various Convolutional Neural Networks (CNN) approaches for analyzing Magnetic Resonance Images (MRI) and CT scan images , made to diagnose neurological diseases, we have used various deep learning architectures like LeNet, AlexNet, ResNet , VGG-19 and InceptionNet-V3 to detect brain tumors from medical images. After selecting that architecture we trained those models and checked each model's performance with the testing images from the dataset. The experiment results draw a comparative study and overall accuracy of the selected architecture and evaluate the bestperforming one.