Summary of Passion: Towards Effective Incomplete Multi-modal Medical Image Segmentation with Imbalanced Missing Rates, by Junjie Shi et al.
PASSION: Towards Effective Incomplete Multi-Modal Medical Image Segmentation with Imbalanced Missing Rates
by Junjie Shi, Caozhi Shang, Zhaobin Sun, Li Yu, Xin Yang, Zengqiang Yan
First submitted to arxiv on: 20 Jul 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 a novel approach called Preference-Aware Self-distillation (PASSION) for incomplete multi-modal medical image segmentation under imbalanced missing rates. The authors formulate this challenging setting, which is far from realistic in clinical scenarios where modalities can have imbalanced missing rates. PASSION consists of pixel-wise and semantic-wise self-distillation to balance the optimization objective of each modality. Additionally, it defines relative preference to evaluate the dominance of each modality during training, which guides task-wise and gradient-wise regularization to balance the convergence rates of different modalities. The proposed approach is validated through experimental results on two publicly available multi-modal datasets, demonstrating its superiority over existing approaches for modality balancing. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper is about a new way to help computers segment medical images when some information is missing. Right now, computers are only good at segmenting images when they have all the information. But in real life, sometimes some information is missing, and that makes it hard for computers to do their job. The authors propose a new approach called PASSION that can handle this problem by balancing the different types of information. They show that PASSION works better than other approaches on two different datasets. |
Keywords
» Artificial intelligence » Distillation » Image segmentation » Multi modal » Optimization » Regularization
