Screening for diabetic retinopathy with artificial intelligence in a primary care setting: a comparative cost...

Area of Science:

• Ophthalmology
• Artificial Intelligence
• Health Economics

Background:

• Diabetic retinopathy (DR) is a leading cause of vision loss in diabetic patients.
• Early detection and treatment of DR are crucial to prevent blindness.
• Current screening methods often rely on physician referrals, which can be resource-intensive.

Purpose of the Study:

• To conduct a comparative cost analysis of autonomous artificial intelligence (AI)-based screening for diabetic retinopathy (DR).
• To compare the costs of AI-based DR screening with the traditional physician-based referral system in a primary care setting.

Main Methods:

• A cost analysis was performed comparing autonomous AI-based DR screening with physician-based screening.
• Data sources included published costs, provincial billing codes, and clinical study data.
• Costs analyzed encompassed direct operational expenses, physician fees, and indirect patient time costs.

Main Results:

• Autonomous AI screening successfully screened 202 participants, with a 93.6% completion rate.
• The AI-based approach cost $13,647.84 per 100 patients, while physician-based screening cost $28,238.09 per 100 patients.
• The cost per diagnosed DR case was 52% lower with AI ($620.36) compared to physician-based screening ($1283.55).

Conclusions:

• Autonomous AI-based DR screening demonstrates significant cost advantages over traditional physician-based methods.
• The lower cost per diagnosed case highlights the potential benefits of implementing AI for DR screening in primary care.
• AI-based screening offers a more economically viable solution for managing diabetic retinopathy.