houjunlin/TCSM

Python

Transformation Consistent Self-ensembling Model for Semi-supervised Medical Image Segmentation

Pytorch implementation of TCSM

Paper

Transformation Consistent Self-ensembling Model for Semi-supervised Medical Image Segmentation

Installation

Install Pytorch 1.1.0 and CUDA 9.0
Clone this repo

git clone https://github.com/xmengli999/TCSM
cd TCSM

Data Preparation

Download Skin dataset, REFUGE dataset, Liver CT dataset
Put the data under ./data/

Train

cd scripts_skin
Run sh train_50tcsm.sh to start the training process

Evaluate

Specify the model path in eval.sh
Run sh eval.sh to start the evaluation.

Acknowledgement

Some code is reused from the Pytorch implementation of mean teacher.

Note

Contact: Xiaomeng Li (xmengli999@gmail.com)