Summary of L-mae: Longitudinal Masked Auto-encoder with Time and Severity-aware Encoding For Diabetic Retinopathy Progression Prediction, by Rachid Zeghlache et al.
L-MAE: Longitudinal masked auto-encoder with time and severity-aware encoding for diabetic retinopathy progression prediction
by Rachid Zeghlache, Pierre-Henri Conze, Mostafa El Habib Daho, Yihao Li, Alireza Rezaei, Hugo Le Boité, Ramin Tadayoni, Pascal Massin, Béatrice Cochener, Ikram Brahim, Gwenolé Quellec, Mathieu Lamard
First submitted to arxiv on: 24 Mar 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 proposed longitudinal masked auto-encoder (MAE) uses self-supervised learning (SSL) pre-training strategies for medical imaging tasks, addressing limitations in applying typical SSL to medical images. The MAE incorporates time-aware position embedding and disease progression-aware masking to capture temporal changes and trends in follow-up examinations. This approach is evaluated on the OPHDIAT dataset, a large diabetic retinopathy screening dataset, to predict severity labels based on past time series examinations. Results show that incorporating time-aware position embedding and masking strategies improves predictive ability for deep classification models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper uses special computer learning methods to help doctors better diagnose and track diseases like diabetes-related eye problems. They developed a new way to train these computers, using information from many medical images taken over time. This helps the computers understand how diseases change and progress over time. The researchers tested this method on a large group of medical images and found that it worked much better than usual methods. This could help doctors make more accurate predictions about patients’ conditions. |
Keywords
» Artificial intelligence » Classification » Embedding » Encoder » Mae » Self supervised » Time series
