FDA Clears iPredict-DR AI Software for Diabetic Retinopathy Screening
The U.S. Food and Drug Administration (FDA) has granted 510(k) clearance to iPredict-DR, an artificial intelligence-based software platform designed to detect more than mild diabetic retinopathy in adults living with diabetes. Developed by iHealthScreen, the diagnostic tool utilizes automated image analysis of retinal fundus scans to identify signs of the condition, which remains a leading cause of vision loss and blindness among working-age populations worldwide.
- iPredict-DR is cleared to identify diabetic retinopathy levels exceeding “mild” severity in patients with diabetes.
- The software integrates with the iCare DRSplus camera, allowing for image acquisition by staff with minimal specialized ophthalmologic training.
- Clearance was predicated on clinical validation data confirming the system’s diagnostic performance, safety, and operational usability.
Clinical Pathogenesis and the Role of AI Screening
Diagnostic Performance and Clinical Validation
The regulatory clearance was supported by data from a structured clinical validation trial. The system is engineered to function with the iCare DRSplus camera, a non-mydriatic imaging device that simplifies the collection process. Unlike traditional indirect ophthalmoscopy, which requires a highly trained clinician to perform a dilated fundus exam, this AI-driven approach allows for image capture by minimally skilled health care workers.
The validation trial prioritized "diagnostic performance, safety and usability," criteria mandated by the FDA for software-as-a-medical-device (SaMD) classifications.
Addressing Regulatory and Operational Hurdles
Future Directions in Retinal Care
Disclaimer: The information provided in this article is for educational and scientific communication purposes only and does not constitute medical advice. Always consult with a qualified healthcare provider regarding any medical condition, diagnosis, or treatment plan.