Eyenuk artificial intelligence detects diabetic retinopathy more sensitively than dilated exams
At the 2022 Annual Meeting of the American Academy of Ophthalmology, Eyenuk Inc. announced the publication of strong eye art validation results in Sciences of ophthalmologya peer-reviewed journal of the American Academy of Ophthalmology.
According to a press release, the study, titled “Artificial Intelligence Detection of Diabetic Retinopathy: Comparison of EyeArt System Subgroups with Dilated Ophthalmologist Examinations,” evaluated general ophthalmologists, retina specialists and Eyenuk. eye art AI system to detect diabetic retinopathy (DR), one of the leading causes of blindness in working-age adults.
Jennifer I. Lim MD, vice president of ophthalmology, UIC professor emeritus of ophthalmology and director of the retina service at the University of Illinois at Chicago and first author of the publication, offered her assessment of the results.
“Compared to the Reading Center grading, which was the gold standard, the sensitivity for detecting more than mild DR was significantly higher with the eye art AI system than with a general ophthalmologist or clinical retina specialist,” she said in the press release. “Unlike a few cases where general ophthalmologists have missed some cases of vision-threatening diabetic retinopathy, the eye art The AI system did not miss any sight-threatening DR cases. The AI system is an important tool to help us deal with the burden of screening and detecting DR in a timely manner. »
According to a company press release, the study assessed the sensitivity and specificity of the eye art ocular system and dilated examinations performed by general ophthalmologists and retina specialists against the rigorous clinical gold standard of the Early Treatment Diabetic Retinopathy Study (ETDRS) on the same cohort of 521 study participants. The ETDRS gold standard was established by experts at the University of Wisconsin Reading Center using 10 fundus images per eye captured after dilation by certified photographers, while the eye art The system only analyzed 2 images per eye, usually without dilation.
The company also noted that sensitivity, a safety measure (percentage of patients with a disease identified correctly), was 96.4% for the eye art system to identify more than mild DR (mtmDR), while that of dilated ophthalmologist exams was 27.7% on the same cohort of study participants. Specificity, a measure of effectiveness (percentage of patients with no disease identified correctly), was 99.6% by dilated ophthalmologist exams versus 88.4% with the eye art system.
Additionally, the company noted that this result demonstrates that dilated examinations by ophthalmologists are more effective in ruling out the disease, as evidenced by their high specificity. However, the EyeArt system, with its excellent sensitivity, is much better at identifying patients with the disease (at the first lines of care), a critical factor for a screening scenario in which patients are identified for referral and evaluation. further.
Additionally, the company noted that the study also indicated that the eye art The system generated actionable results for over 97% of the eyes, with most (85.3%) not requiring dilation. In contrast, dilated exams provided usable results for 99.9% of eyes, but required all patients to be dilated.
“Given the current low rate of compliance with the recommendation for an annual diabetic retinal examination, this system may be a useful adjunct in the detection of mtmDR and appears to be more accurate than clinical ophthalmoscopy for routine retinal screening. “, Lim and the co-conclude the authors.
Cleared by the FDA in 2020, the eye art The system is now used at more than 200 sites in 18 countries, including 14 US states, to screen more than 60,000 patients and counting. It is the first and only technology to be FDA cleared for the stand-alone detection of referable and vision-threatening diabetic retinopathy.
AAO 2022 Symposium to examine the actual results of using EyeArt
Eyenuk will showcase its AI technologies for diabetic eye screening at AAO 2022. of FDA-Approved AI Technology for Diabetic Retinopathy Screening in Primary Care.”
Ophthalmology leaders from several academic medical centers will discuss their vision and progress in adopting Eyenuk’s AI technologies to improve diabetes eye care and population health.
The speakers of the symposium are:
- RV Paul Chan MD, MSc, MBA, FACS, Chairman, Department of Ophthalmology and Visual Sciences, University of Illinois at Chicago
- Eric D. Hansen, MD Assistant Professor of Ophthalmology and Visual Sciences, University of Utah
- Jeffrey D. Henderer MD, Chair, Department of Ophthalmology, Lewis Katz School of Medicine at Temple University
- Steven Yeh MD, Chairman and Director, Retina and Uveitis, University of Nebraska Medical Center
The AAO Industry Showcase Symposium will take place on Sunday, October 2 at 2:10 p.m. CT, at McCormick Place (booth #1026). From October 1-3, AAO 2022 attendees can watch a demonstration of the eye art AI System and other Eyenuk technologies by visiting booth #2033.