Post

New research · Ophthalmology
Translational vision science & technology · 5d
AI / informaticsTranslational vision science & technology · 2026

Validation of a Machine Learning Approach to the Analysis of Multifocal Electroretinograms for Hydroxychloroquine Retinopathy.

Godfrey Wong, Gareth Mercer, Brian G Ballios, Tom Wright
Read paper
OphthalmologyAI / informatics

Machine learning algorithm accurately rules out hydroxychloroquine retinopathy in new patients.

Validation of a Machine Learning Approach to the Analysis of Multifocal Electroretinograms for Hydroxychloroquine Retinopathy.

Godfrey Wong … Tom Wright
Translational vision science & technology · 2026
Purpose

This study evaluated the clinical utility of the Multifocal Electroretinogram Classification Interface (Multifocal Electroretinogram Classification Interface) algorithm by assessing its ability to predict hydroxychloroquine retinopathy compared with diagnoses derived from American Academy of Ophthalmology (Academy of Ophthalmology) guidelines.

Methods

Referred for hydroxychloroquine toxicity screening underwent perimetry, spectral-domain optical coherence tomography, fundus autofluorescence photography, and multifocal electroretinogram (mfERG) testing.

0.987
Results
0.987
The algorithm was very good at ruling out eye damage in new patients
More results

Using the decision trees, the temporal (n = 145) and novel (n = 300) datasets showed retinopathy prevalences of 12.4% and 9.7%, respectively.

Multifocal Electroretinogram Classification Interface yielded 63/145 (43.4%) and 115/300 (38.3%) false-positive cases, respectively.

“
Conclusion · 1 of 2

Multifocal Electroretinogram Classification Interface retained high sensitivity and negative predictive value when validated against temporally distinct cohorts using a clinical definition of hydroxychloroquine toxicity.

Conclusion · 2 of 2

This suggests Multifocal Electroretinogram Classification Interface's potential as a screening tool for hydroxychloroquine retinopathy.

Read paper
0 comments

No comments yet. Be the first.

Related papers

LatestFoundational
AI / Informatics
0·962 for OCT
MerMED-FM was very accurate at diagnosing diseases using eye scans
AI / Informatics
0.98
Artificial intelligence's eyelid-height measurements matched doctors' manual measurements almost perfectly
AI / Informatics
92.1%
combining eye scans and photos correctly told benign from cancerous lesions apart nearly every time
Cohort Study
28.6%
recurred locally in more than 1 in 4 patients over years of follow-up
Cohort Study
-0.395
excision group's post-op eyelid fullness score was lower - greater improvement
Cohort Study
86.7%
of infants probed after 12 months still had unresolved tear duct blockage
Observational
16%
eyelid tissue in rosacea patients showed less of this key repair-signaling protein inside cell nuclei
Cohort Study
25%
about 1 in 4 treated cases had symptoms return after improving