Awesome

UGPCL

[IJCAI' 22] Uncertainty-Guided Pixel Contrastive Learning for Semi-Supervised Medical Image Segmentation.

Tao Wang, Jianglin Lu, Zhihui Lai, Jiajun Wen and Heng Kong

Installation

Please refer to requirements.txt

Dataset

ACDC dataset can be found in this Link.

Please change the dataset root directory in configs/_datasets/acdc_224.yaml.

Results

ACDC (136 labels | 10%):

Method	Model	Iterations	Batch Size	Label Size	DsC	Ckpt	Config File
UGPCL	UNet-R50	6000	16	8	88.11	Link	configs/comparison_acdc_224_136/ugpcl_unet_r50.yaml
Mean Teacher	UNet-R50	6000	16	8	85.75	Link	configs/comparison_acdc_224_136/mt_unet_r50.yaml

Visualization

Segmentation results:
<img src="pics/preds.jpg" style="zoom: 15%;" />
Pixel features (t-SNE):
<img src="pics/show_feats.jpg" style="zoom: 20%;" />

Reference

Citation

@inproceedings{ijcai2022-201,
  title     = {Uncertainty-Guided Pixel Contrastive Learning for Semi-Supervised Medical Image Segmentation},
  author    = {Wang, Tao and Lu, Jianglin and Lai, Zhihui and Wen, Jiajun and Kong, Heng},
  booktitle = {Proceedings of the Thirty-First International Joint Conference on
               Artificial Intelligence, {IJCAI-22}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Lud De Raedt},
  pages     = {1444--1450},
  year      = {2022},
  month     = {7},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2022/201},
  url       = {https://doi.org/10.24963/ijcai.2022/201},
}