Summary of Uncertainty-aware Evidential Fusion-based Learning For Semi-supervised Medical Image Segmentation, by Yuanpeng He et al.
Uncertainty-aware Evidential Fusion-based Learning for Semi-supervised Medical Image Segmentation
by Yuanpeng He, Lijian Li
First submitted to arxiv on: 9 Apr 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 The paper proposes an evidential deep learning framework for medical segmentation tasks, aiming to address limitations in existing semi-supervised methods. It integrates predictive results from mixed and original samples to reallocate confidence degrees and uncertainty measures, using a probability assignment fusion rule based on traditional evidence theory. Additionally, the authors design a voxel-level asymptotic learning strategy that combines information entropy with fused uncertainty measure to estimate predictions more precisely. The approach focuses on learning features that are difficult to master by gradually paying attention to prediction results with high uncertainty. Experimental results on LA, Pancreas-CT, ACDC, and TBAD datasets demonstrate the superiority of this proposed method over existing state-of-the-art approaches. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper tries to make medical image segmentation better. Right now, some methods can do it well but only consider one way to be sure about what they’re doing. This isn’t enough because it doesn’t solve all problems related to trust. The new approach uses a special type of learning called evidential deep learning and combines the results from different sources to make a more informed decision. It also tries to learn more by paying attention to things that are hard to figure out. The results show that this method works better than what’s been done before on several types of medical images. |
Keywords
» Artificial intelligence » Attention » Deep learning » Image segmentation » Probability » Semi supervised
