Skip to content

tmb2k01/medical-image-segmentation

Repository files navigation

Brain Tumor Segmentation - Partners in Prediction

Contributors

Project Overview

This project focuses on Brain Tumor Segmentation using deep learning techniques. We are leveraging the BraTS 2017 dataset, which contains a variety of MRI scans of brain tumor patients. The primary goal of the project is to develop a model that can accurately segment different tumor regions in brain MRI images, helping in the diagnosis and treatment planning for patients.

Our project aims to preprocess this data, train a segmentation model, and evaluate its performance.

Documentation

For more details on specific parts of our project, please refer to the following:

  • Baseline Model: Description of the baseline model we use.
  • Data Acquisition Guide: Steps to acquire the BraTS 2017 dataset.
  • Docker Guide: Instructions for setting up and running the project in a Docker environment.
  • Evaluation Criteria Definition: Detailed description of the metrics and criteria used to evaluate model performance.
  • Model Evaluation: Results and analysis of the performance of the final model.
  • Project Structure: A detailed overview of the project's structure and organization.
  • References: A list of all references we've used.
  • User Guide: Instructions for starting the Gradio interface powered by the machine learning model we developed. This document contains everything you need to start the project (the teacher should read this).

About

In this project, you'll dive into the idea of using multiple models together, known as model ensembles, to make our deep learning solutions more accurate. They are a reliable approach to improve accuracy of a deep learning solution for the added cost of running multiple networks.

Resources

Stars

1 star

Watchers

2 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages