Summary of Semi-supervised Medical Image Segmentation Method Based on Cross-pseudo Labeling Leveraging Strong and Weak Data Augmentation Strategies, by Yifei Chen et al.
Semi-supervised Medical Image Segmentation Method Based on Cross-pseudo Labeling Leveraging Strong and Weak Data Augmentation Strategies
by Yifei Chen, Chenyan Zhang, Yifan Ke, Yiyu Huang, Xuezhou Dai, Feiwei Qin, Yongquan Zhang, Xiaodong Zhang, Changmiao Wang
First submitted to arxiv on: 17 Feb 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel semi-supervised learning model, DFCPS, is proposed for medical image segmentation. This approach incorporates the Fixmatch concept and leverages data augmentation to enhance performance and generalizability. The model design emphasizes pseudo-label generation, filtration, and refinement processes. A new concept of cross-pseudo-supervision integrates consistency learning with self-training, allowing the model to fully utilize pseudo-labels from multiple perspectives. Experimental results on the Kvasir-SEG dataset demonstrate superior performance across four subdivisions with varying proportions of unlabeled data. The proposed model outperforms baseline and advanced models, showcasing its effectiveness in medical image segmentation. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Medical images are hard to analyze because they’re tricky to collect, expensive to label, have bad signals, and are very complex. Scientists created a new way to train computers using pictures that aren’t labeled. They called it DFCPS. This method uses tricks to make the computer learn more from the unlabeled pictures. It also makes sure the computer is careful when making predictions about those pictures. The scientists tested their method on a big dataset and found out it works better than other methods. Now you can find the code used for this experiment online. |
Keywords
» Artificial intelligence » Data augmentation » Image segmentation » Self training » Semi supervised
