Summary of Etscl: An Evidence Theory-based Supervised Contrastive Learning Framework For Multi-modal Glaucoma Grading, by Zhiyuan Yang et al.
ETSCL: An Evidence Theory-Based Supervised Contrastive Learning Framework for Multi-modal Glaucoma Grading
by Zhiyuan Yang, Bo Zhang, Yufei Shi, Ningze Zhong, Johnathan Loh, Huihui Fang, Yanwu Xu, Si Yong Yeo
First submitted to arxiv on: 19 Jul 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 A novel computer-aided glaucoma diagnosis framework, ETSCL, is proposed to integrate color fundus photography (CFP) and optical coherence tomography (OCT) modalities for improved diagnostic accuracy. The framework consists of a contrastive feature extraction stage and a decision-level fusion stage, utilizing supervised contrastive loss and the Frangi vesselness algorithm as preprocessing steps. An evidence theory-based multi-modality classifier is employed to combine multi-source information with uncertainty estimation. Experimental results demonstrate state-of-the-art performance. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Glaucoma is a leading cause of vision impairment. Digital imaging techniques like color fundus photography (CFP) and optical coherence tomography (OCT) provide noninvasive methods for diagnosis. Researchers have developed computer-aided glaucoma diagnosis methods that use both CFP and OCT. However, it’s hard to get reliable features from medical images because they look similar, and there isn’t a balanced mix of data from different sources. To solve these problems, scientists created a new framework called ETSCL. It includes two stages: one for extracting features and another for combining information from multiple sources with uncertainty estimates. The results show that this method works really well. |
Keywords
» Artificial intelligence » Contrastive loss » Feature extraction » Supervised
