NeuroVision: Regional Brain Tumor Detection Web App
Fall 2023
Tools:
NeuroVision is a web-based machine-learning application designed to detect brain tumors from MRI scans.
The project uses a U-Net convolutional neural network optimized for medical imaging segmentation, capable
of identifying tumors and predicting their size and shape with high accuracy. The model achieved an accuracy of 99.48% with
a loss rate of 1.64% on validation dataset.The web app allows users to
upload MRI scans and receive predictions in real-time.
My role in the project was developing and training the model, optimizing its architecture
for improved performance, and fine-tuning the hyperparameters. Additionally, I worked on integrating the
Flask backend with the React frontend, to create a connection between the model and the user interface.
Check out the YouTube video for a demo, or visit the GitHub repository to install the program and try it yourself!
