3.231.219.178
dgid:
enl:
npi:0
-Advertisement-
-Advertisement-
Retina

Artificial Intelligence shows promise in estimating visual acuity for diabetic macular edema

Posted on

Artificial intelligence (AI) can estimate best-corrected visual acuity (BCVA) directly from fundus photographs in patients with diabetic macular edema (DME), eliminating the need for refraction or subjective visual acuity measurements, according to a new study. The estimated BCVA often fell within 1 to 2 lines on an Early Treatment Diabetic Retinopathy Study (ETDRS) chart.

The use of AI could have significant implications for managing DME by reducing the need for personnel and time required for assessing BCVA, as well as enabling remote imaging.

The study utilized deidentified color fundus images from patients enrolled in the VISTA randomized clinical trial, totaling 7185 images from 459 patients. The primary outcome, evaluated by mean absolute error (MAE), showed promising results. Using the ResNet50 architecture, the MAE for the testing set (n = 641 images) was 9.66 (95% CI, 9.05-10.28).

The study found that 33% of BCVA predictions were within 0 to 5 letters, and 28% were within 6 to 10 letters, indicating a high level of accuracy. When focusing on subsets categorized by baseline BCVA, the MAE was 8.84 letters (95% CI, 7.88-9.81) for BCVA between 100 and 80 letters (20/10 to 20/25), and 7.91 letters (95% CI, 7.28-8.53) for BCVA between 80 and 55 letters (20/32 to 20/80).

Reference
Paul W, Burlina P, Mocharla R, et al. Accuracy of Artificial Intelligence in Estimating Best-Corrected Visual Acuity From Fundus Photographs in Eyes With Diabetic Macular Edema. JAMA Ophthalmol. 2023 Jun 8:e232271. doi: 10.1001/jamaophthalmol.2023.2271. Epub ahead of print. PMID: 37289463; PMCID: PMC10251243.

-Advertisement-
-Advertisement-
-Advertisement-
-Advertisement-
-Advertisement-