https://github.com/sudohainguyen/medical-image-segmentation

https://github.com/sudohainguyen/medical-image-segmentation

Last synced: 3 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/sudohainguyen/medical-image-segmentation
• Owner: sudohainguyen
• Archived: true
• Created: 2020-07-06T06:13:11.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2020-10-15T16:52:56.000Z (about 4 years ago)
• Last Synced: 2024-04-10T21:49:01.519Z (7 months ago)
• Language: Python
• Homepage:
• Size: 32.2 KB
• Stars: 0
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome_medical - medical-image-segmentation

README

        
# Medical Image segmentation

This project is about my personal experiments with various of Neural Network for medical image segmentation (organs, brain, tumors,...) implemented in PyTorch and PyTorch Lightning.

## Dataset

- [x] Dataset B

- [ ] LITS 2017

## Neural Nets

- [x] UNet

- [x] AttentionUNet

- [x] Nested-UNet

- [ ] STAN ([Small tumor-aware network](https://arxiv.org/pdf/2002.01034.pdf))

- [ ] MSDNet ([Mixed-scale dense convolutional network](https://www.pnas.org/content/pnas/115/2/254.full.pdf))

## Attention mechanism

- [x] Attention Gated ([paper](https://openreview.net/pdf?id=Skft7cijM))

- [ ] Grid Attention

- [ ] Axial Attention ([paper](https://arxiv.org/abs/2003.07853))

## Loss function

- [x] Dice loss

- [ ] Lovasz loss

- [ ] Tversky loss

## Demo