An AI diagnostic system that detects signs of diabetic retinopathy in retinal images
- Proven to be effective in real-world clinical workflows
- ​Results in less than a minute – no human grader
- Clinically validated with high sensitivity and specificity
How it Works
Easy to use
- Minimal operator training needed
- Uses standard 2-image per eye protocol and accepts most image formats, including DICOM
- Designed with common output formats to directly interface with other medical software and health record systems
Clear Output
Within one minute, the IDx-DR user will know if:
1. Exam quality was insufficient – Low quality images can be retaken immediately, while the patient is still at the camera
2. The patient is negative for referable DR – The patient can be re-screened at the normal interval
3. The patient has signs of referable DR – Outputs of Moderate DR detected or Vision-threatening DR detected indicate further action:
a. Over-read by a human grader
b. Tele-consultation
c. Referral to an ophthalmologist
- Minimal operator training needed
- Uses standard 2-image per eye protocol and accepts most image formats, including DICOM
- Designed with common output formats to directly interface with other medical software and health record systems
Image Quality Feedback
Immediate image quality feedback enables even novice users to capture the high quality images needed for accurate disease detection, reducing the number of patient call-backs.
Insufficient Quality
Sufficient Quality