Summary of Proportion Estimation by Masked Learning From Label Proportion, By Takumi Okuo et al.
Proportion Estimation by Masked Learning from Label Proportion
by Takumi Okuo, Kazuya Nishimura, Hiroaki Ito, Kazuhiro Terada, Akihiko Yoshizawa, Ryoma Bise
First submitted to arxiv on: 8 May 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG)
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 proposed method estimates PD-L1 rates by leveraging small amounts of cell-level annotation and proportion annotation. It involves detecting tumor cells using a detection model, then estimating the PD-L1 proportion with a masking technique that incorporates “learning from label proportion.” The method also addresses data imbalance issues with a weighted focal proportion loss. Experimental results on clinical data demonstrate its effectiveness, outperforming comparisons. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research paper proposes a new way to estimate the rate of PD-L1 positive tumor cells. This is important because it helps doctors decide which treatments are best for patients with cancer. The method uses special images and some extra information that’s easy to collect. First, it finds the tumor cells in the image, then it estimates how many of those cells have PD-L1. The method also fixes a problem where there might not be enough data by adjusting its calculations. Tests on real patient data show that this method works well and is better than other methods. |
