Next-Generation Teleophthalmology: AI-enabled Quality Assessment Aiding Remote Smartphone-based Consultation

Area of Science:

• Ophthalmology
• Artificial Intelligence
• Medical Imaging

Background:

• Blindness and eye diseases pose a global health challenge, particularly in low- and middle-income countries.
• Teleophthalmology and smartphone-based imaging (e.g., Grabi attachment) expanded during the COVID-19 pandemic.
• Patient-captured eye images often lack sufficient clinical detail for accurate diagnosis, causing treatment delays.

Purpose of the Study:

• To develop and test an AI-based system for assessing the quality of patient-captured eye images.
• To provide instant feedback mimicking clinical judgment to improve image usability.
• To address a critical component of AI-driven teleophthalmology for enhanced eye care.

Main Methods:

• A hierarchical approach was used to address the complex problem of image quality assessment.
• The AI system was trained and tested on patient-captured eye images.
• The system was designed to mimic the quality judgments of clinical experts.

Main Results:

• The study demonstrates a proof of concept for an AI-based quality assessment system.
• The system is capable of evaluating a non-trivial aspect of image quality.
• The AI system provides instant feedback, a key feature for real-time clinical support.

Conclusions:

• AI-powered quality assessment can significantly enhance the utility of smartphone-based eye imaging in teleophthalmology.
• Instant feedback systems can help overcome limitations of user-generated medical images.
• This technology holds potential for improving eye care accessibility and efficiency, especially in resource-limited settings.