Detection of Brain Tumor
This project is about creating an image classification model to detect whether an MRI image of a brain has a tumor or not, created using Fast ai, which is a high-level deep learning library built on top of Py Torch.
Abstract
Brain tumor detection and segmentation are important tasks in medical image analysis. This project is about creating an image classification model to detect whether an MRI image of a brain has a tumor or not. The model is created using Fast ai, which is a high-level deep learning library built on top of Py Torch. The dataset used in this project contains MRI images of brains with and without tumors. The model is trained using transfer learning with ResNet18 and ResNet34 as the base architectures. After training the model, it is exported and used to make predictions on new images using a simple web interface built with widgets. Keywords: Brain tumor detection, segmentation, medical image analysis, Fast.ai, deep learning, transfer learning, ResNet18, ResNet34, MRI images, web interface