Read more

April 24, 2023
1 min read
Save

Autonomous AI exams linked with higher adherence to annual diabetic eye disease testing

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

Key takeaways:

  • Adherence to annual diabetic eye disease testing was higher in primary care clinics with autonomous AI exams.
  • Women and patients who were older were more likely to undergo exams.

NEW ORLEANS —The implementation of autonomous artificial intelligence eye exams was associated with a higher adherence to annual diabetic eye disease testing, according to a poster here.

“We wanted to look at the implementation of artificial intelligence screening for diabetic retinopathy,” Jane Huang told Healio/OSN at the Association for Research in Vision and Ophthalmology meeting. “We found that at primary care clinics that have implemented the AI technology, there is a higher adherence rate among the patient base of receiving their annual eye exam screenings.”

Eyes with screen
The implementation of autonomous artificial intelligence eye exams was associated with a higher adherence to annual diabetic eye disease testing, according to a poster here.
Image: Adobe Stock

The retrospective cross-sectional analysis included 22,264 patients with diabetes at Johns Hopkins Medicine, with the primary outcome being patients’ overall adherence to annual diabetic eye disease testing. There were 6,660 patients in the AI group and 15,604 patients in the non-AI group, based on whether the clinic sites had implemented autonomous AI exams.

Overall, adherence to annual screenings was higher among patients in the AI group at 64% compared with 46.3% in the non-AI group (P < .001).

Jane Huang
Jane Huang

Women were more likely to adhere to their annual eye exam screening (P = .013), as well as patients who were older (P < .001), had a higher Adjusted Clinical Groups value (P < .001) and had a lower HbA1c value (P < .001). Patients who had military health insurance also had a higher adherence rate (P < .001).

According to Huang, future research should investigate the effects of autonomous AI screening exams on long-term diabetic outcomes.

“This is really exciting for us just to see that the AI technology has improved access to the annual eye exams that are really important for patients who have been diagnosed with diabetes,” she said.