Related Experiment Video
Updated: Jul 21, 2025
07:40
Experimental Autoimmune Uveitis: An Intraocular Inflammatory Mouse Model
Published on: January 12, 2022
4.9K
A machine learning framework for the quantification of experimental uveitis in murine OCT.
Youness Mellak1, Alin Achim2, Amy Ward2
1Université Côte d'Azur, INRIA, CNRS, I3S, Sophia Antipolis, France.
Biomedical Optics Express
|July 27, 2023
Summary
This study introduces advanced methods for detecting non-infectious uveitis using optical coherence tomography (OCT) images. The research develops AI models to classify disease stages and identify retinal particles, aiding in vision loss assessment.
Area of Science:
- Ophthalmology
- Medical Imaging
- Artificial Intelligence
Background:
- Non-infectious uveitis is a primary cause of vision loss in adults.
- Accurate detection and staging are crucial for effective treatment.
Purpose of the Study:
- To develop and validate novel methods for uveitis detection and assessment.
- To utilize optical coherence tomography (OCT) and AI for improved diagnostic capabilities.
Main Methods:
- A classification model using OCT images to predict uveitis presence and stage.
- Grad-CAM for visualizing classifier decision-making.
- Comparison of three methods (supervised, MPP, weakly supervised) for retinal particle detection.
- Automated pipeline for 2-D/3-D segmentation, volume reconstruction, and particle distribution analysis.
Main Results:
- The classification model accurately predicts uveitis and its stages from OCT images.
- Weakly supervised segmentation enables automated detection and 3-D reconstruction of retinal particles.
- Analysis of particle distribution reveals clustering patterns indicative of disease severity.
Conclusions:
- The proposed methods offer enhanced capabilities for non-infectious uveitis detection and assessment.
- AI-driven analysis of OCT images and retinal particles can improve disease grading and understanding.
