Implementation of A New, Mobile Diabetic Retinopathy Screening Model Incorporating Artificial Intelligence in Remote...

Area of Science:

• Ophthalmology
• Public Health
• Medical Technology

Background:

• Diabetic retinopathy (DR) screening is crucial for preventing blindness but faces low rates in remote Western Australia due to reliance on outdated equipment.
• Access to effective DR screening is limited in remote areas, necessitating innovative solutions.

Purpose of the Study:

• To describe and evaluate a novel, mobile diabetic retinopathy screening model integrating artificial intelligence (AI) into routine care.
• To assess the feasibility and effectiveness of AI-assisted DR screening in a remote Australian setting.

Main Methods:

• A prospective, population-based study was conducted in the Pilbara region of Western Australia.
• A mobile screening unit, co-designed with local Aboriginal communities, utilized an automated retinal camera with integrated AI diagnostics.
• Screening was performed by a research officer, with on-the-spot diagnoses and remote clinician support via telehealth.

Main Results:

• DR screening was provided to 78 patients across 9 communities, with 56.4% identifying as Aboriginal or Torres Strait Islander.
• 10.3% of retinal photos revealed referable DR, and 8.4% were ungradable.
• Patient acceptability was high, with 96% reporting satisfaction with the AI system.

Conclusions:

• The AI-assisted mobile DR screening model was culturally safe, acceptable, and effective in a remote Australian context.
• This model demonstrated an 11-fold increase in screening rates compared to previous data.
• AI-assisted screening shows potential to overcome barriers and reduce preventable blindness in underserved remote populations.