Implementation of A New, Mobile Diabetic Retinopathy Screening Model Incorporating Artificial Intelligence in Remote Western Australia.
Qiang Li, Jocelyn J Drinkwater, Kerry Woods and 3 others
PMID 40110918WHAT IT FOUND
A van-based AI retinal screening reached remote communities and was accepted by patients: 96% of survey responders were happy with AI, and all rated the experience good or very good.
Key findings
01The mobile service screened 9 communities, obtained consent from 78 participants, and 50 completed the patient survey; 96% said they were happy with AI use, and all rated the overall experience good or very good.
02Of 155 retinal photographs analysed, 126 (81.3%) were non-referable, 16 (10.3%) were referable, and 13 (8.4%) were ungradable.
03A research officer without formal healthcare qualifications performed the screening, while a remote telehealth clinician was available at all times to support and supervise.
STILL TO COME
How it was doneWhat they foundWhat it means for RNs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The paper did not compare this mobile service with usual care. It did not validate the diagnostic accuracy of the AI systems. Only 78 participants consented and only 50 completed the survey, so patient acceptability is based on a small group. The mobile service experienced equipment malfunction, internet problems, staff safety concerns, and cost. Some patients were unsure of their diabetes status, and internet problems may have affected survey completion.
The easy way to misread this
Do not read this as proof that AI diagnoses diabetic retinopathy correctly. The paper says accuracy was not tested here and will be studied separately. It also did not compare screening with usual care or show vision outcomes.