Brain Tumor Segmentation using TransUNet with Flask-Based Visualization

Project Domain / Category

Deep Learning based Web Application

Abstract / Introduction

Brain tumor segmentation is a critical task in medical image analysis, enabling precise detection of tumor regions for diagnosis and treatment planning. In this project, you will develop a deep learning-based segmentation model using TransUNet, a hybrid CNN + Transformer-based U-Net model, to automatically segment brain tumors from MRI scans in the BraTS 2023 dataset.

In addition to training and evaluating the segmentation model, you will develop a Flask-based web application that allows users to upload MRI images and visualize the predicted segmentation results alongside ground truth masks.

Functional Requirements:

The following are basic requirements:

·         Implement TransUNet for brain tumor segmentation using PyTorch.

·         Preprocess the BraTS 2023 dataset (multi-modal MRI images).

·         Train the model using Dice Loss + Cross-Entropy Loss and evaluate its performance.

·         Display segmentation results as images with ground truth comparison.

·         Develop a Flask web application where users can upload an MRI scan and receive the predicted segmentation mask.

Note:

For your better understanding the guidelines that how to start and study/code the details of all above points is as follows:

Task 1: Understanding the Dataset

·         Study the U-Net model and TransUNet model.

·         Study the BraTS 2023 dataset (FLAIR, T1, T1ce, T2 modalities).

·         Understand tumor segmentation labels (Whole Tumor, Tumor Core, Enhancing Tumor).

·         Code to visualize sample MRI images and corresponding masks.

Task 2: Preprocessing & Data Augmentation

·         Load and preprocess MRI scans using Nibabel.

·         Normalize images and apply augmentations (flipping, rotation, resizing).

·         Convert segmentation masks into one-hot encoded format.

Task 3: Implementing TransUNet for Brain Tumor Segmentation

·         Implement the TransUNet model using PyTorch.

·         Define loss function (Dice Loss + Cross-Entropy) and optimizer (AdamW).

·         Train the model and monitor performance metrics.