Related Experiment Video
Updated: May 8, 2026

Using Optical Coherence Tomography and Optokinetic Response As Structural and Functional Visual System Readouts in Mice and Rats
Published on: January 10, 2019
An artificial intelligence cloud platform for OCT-based retinal anomalies screening system in real clinical
Xinjian Chen1,2, Jingtao Wang3, Tianwei Qian4,5,6,7,8
1School of Electronics and Information Engineering, Soochow University, Suzhou, China. xjchen@suda.edu.cn.
An AI platform for Optical Coherence Tomography (OCT) retinal anomaly screening significantly improves remote diagnostics in China. This AI-PORAS system matches ophthalmologist performance, aiding healthcare decisions for millions at risk of vision loss.
Area of Science:
- Ophthalmology
- Artificial Intelligence
- Medical Imaging
Background:
- Millions globally experience retinal anomalies, risking irreversible vision loss.
- China faces a critical shortage of ophthalmologists, particularly in rural areas.
- Existing diagnostic methods are strained by the high patient-to-specialist ratio.
Purpose of the Study:
- To develop and validate an Artificial Intelligence Cloud Platform for OCT-based Retinal Anomalies Screening (AI-PORAS).
- To enhance remote diagnostic efficiency for 15 common retinal anomalies.
- To provide data for improved healthcare decision-making and treatment planning.
Main Methods:
- Development of the AI-PORAS platform utilizing Artificial Intelligence.
- Training and validation on a large dataset of Optical Coherence Tomography (OCT) B-scan images.
- Deployment to 207 medical institutions across 29 Chinese provinces for real-world application.
Main Results:
- AI-PORAS demonstrated high accuracy (93.16%) and AUC (93.64%) in detecting 15 retinal anomalies.
- Performance metrics (FPR 6.82%, FNR 7.87%) were comparable to attending ophthalmologists.
- Validation involved 165,384 eyes and 3,551,959 OCT B-scan slices.
Conclusions:
- AI-PORAS effectively screens for retinal anomalies using OCT images, addressing specialist shortages.
- The platform's performance rivals that of experienced ophthalmologists, enhancing remote diagnostic capabilities.
- Statistical analysis of 116,717 remotely diagnosed patients offers valuable insights for public health strategies.
More Related Videos
Related Concept Videos
The Retina
Glaucoma: Overview
Open Angle Glaucoma: Treatment
Drugs such as carbonic anhydrase inhibitors, α2- and...
Angle Closure Glaucoma: Treatment
Photoreceptors and Visual Pathways
Diabetic Retinopathy

