TOPLINE:
Only a small proportion of patients with diabetes have undergone artificial intelligence (AI)–based screening for diabetic retinopathy since 2021, despite approval from the US Food and Drug Administration; traditional remote imaging methods for the detection of the eye disorder remain more prevalent.
METHODOLOGY:
- Researchers conducted the retrospective cohort study using the TriNetX federated database across 62 healthcare organizations in the United States to evaluate national trends in the AI-based detection of diabetic retinopathy.
- They examined the records of patients with diabetes from January 2019 to December 2023, with data analysis completed in May 2024.
- Usage rates of a Current Procedural Terminology code for AI-based screening were compared with those of traditional codes for remote eye imaging and imaging modalities from secondary referrals.
- A total of 4,959,890 patients with diabetes (mean age, 64 years; 48% women) were included, with 209,673 unique patients receiving at least one of the targeted codes.
TAKEAWAY:
- AI-based imaging was used in only 0.09% of all patients with diabetes, indicating limited adoption.
- Use of traditional methods for remote imaging increased by 185.4% between 2021 and 2023.
- The overall use of remote imaging modalities increased by 90.16% from 2021 to 2023 (P
- Among patients who received AI imaging, more than 80% were from the South, and almost half were Black.
IN PRACTICE:
“Broader adoption [of AI imaging] may require support to help physicians and organizations integrate these systems into existing workflows,” the authors of the study wrote. The findings “support further evaluation of imaging practices to develop targeted strategies for improving diabetic eye imaging rates and patient outcomes,” they added.
SOURCE:
The study was led by Shreya A. Shah, BS, of the Byers Eye Institute at Stanford University School of Medicine in Palo Alto, California. It was published online on October 31, 2024, in JAMA Ophthalmology.
LIMITATIONS:
No limitations were reported for this study.
DISCLOSURES:
Some authors acknowledged receiving Departmental Core Grants from the National Eye Institute and Research to Prevent Blindness and were supported in part by an unrestricted grant from Research to Prevent Blindness to the New York University Department of Ophthalmology. One author reported receiving personal fees from Zeiss and Google outside the submitted work. Another author reported receiving consulting fees and funding and serving as a consultant for various sources.
This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.
Source link : https://www.medscape.com/viewarticle/artificial-intelligence-based-detection-diabetic-retinopathy-2024a1000k5x?src=rss
Author :
Publish date : 2024-11-05 11:22:31
Copyright for syndicated content belongs to the linked Source.