Brain Tumor Detection and Classification using Deep Learning

A brain tumor is a growth of abnormal brain cells, some of which develop into tumors. Brain tumors are among the most well-known causes of brain dysfunction. Magnetic resonance imaging is the most accurate and consistent technique for finding brain tumors (MRI). The scans generate a lot of images. The radiologists evaluate these images. Manual assessment takes time and increases the risk of making a false cancer diagnosis. In recent years, decision-making has been practiced in practically every sector of the economy, including financial services, healthcare, marketing, and agreements. Deep learning has become a recognized quality in these areas. For more accurate brain tumor prediction and better patient care, machine learning and deep learning algorithms are used to MRI data. The radiologist can make speedy decisions thanks to these results. In this study, brain cancers are detected and categorized into different groups using a convolution neural network (CNN). We put forth a model that categorizes brain cancers using pre-trained convolutional neural networks. Four classes make up the dataset, three of which are tumor types and one of which is non-tumor. The networks used are ResNet50 and EfficientNet. The size of the Dataset is also increased through Data Augmentation. The accuracy of various networks is used to evaluate models performance