Using Machine Learning to Identify Ophthalmology Subspecialty Care and Advance Workforce Research with the IRIS® Registry (Intelligent Research in Sight).
Jeon Ju Hyun, Lee Ju-Yeun, Elze Tobias, Miller Joan W, Lorch Alice C, Ong Mei-Sing, Wu Ann Chen, Hunter David G, Oke Isdin
AI Summary
Machine learning models accurately identified ophthalmology subspecialists from IRIS Registry data, especially using procedure codes. This can help understand ophthalmic workforce trends and inform eye care policy.
Abstract
Purpose
To develop machine-learning models to identify ophthalmology subspecialists using deidentified patient data from a large database.
Design
Cross-sectional.
Participants
All ophthalmologists participating in the American Academy of Ophthalmology's IRIS® Registry (Intelligent Research in Sight) from 2013 to 2023 were classified under one of the following general or subspecialty categories: comprehensive, cataract, cornea, glaucoma, retina, oculofacial, pediatric, or neuro-ophthalmology.
Methods
We collected the diagnosis, procedure, and prescription codes linked to each ophthalmologist. We performed binary subspecialty classification using random forest models with fivefold cross validation and multispecialty classification using 4 approaches (diagnosis only, procedure only, prescription only, and combined).
Main outcome measures
Model performance was assessed using area under the receiver operating characteristic curve (AUROC), F1 scores, and Matthews correlation coefficient.
Results
The study included 9032 ophthalmologists. Classification accuracy differed by subspecialty (AUROC, retina: 0.981; oculofacial: 0.975; pediatric: 0.972; glaucoma: 0.937; cornea: 0.932; neuro: 0.912; cataract: 0.861; and comprehensive: 0.760). The procedure-only random forest model had better performance (AUROC, 0.903) than the diagnosis-only (0.880) and prescription-only (0.835) model.
Conclusions
Machine learning models leveraging the IRIS Registry can provide a near real-time assessment of the landscape of ophthalmic subspecialty care. Identifying subspecialty physicians through practice patterns may provide valuable insights into the future trends of eye care delivery with implications for workforce research and policy interventions.
Financial disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.
Shields Classification
Key Concepts5
Classification accuracy for identifying ophthalmology subspecialists, assessed using area under the receiver operating characteristic curve (AUROC), differed by subspecialty (retina: 0.981; oculofacial: 0.975; pediatric: 0.972; glaucoma: 0.937; cornea: 0.932; neuro: 0.912; cataract: 0.861; and comprehensive: 0.760) among 9032 ophthalmologists.
The procedure-only random forest model had better performance (AUROC, 0.903) than the diagnosis-only (0.880) and prescription-only (0.835) models for identifying ophthalmology subspecialists among 9032 ophthalmologists.
Machine learning models leveraging the IRIS Registry can provide a near real-time assessment of the landscape of ophthalmic subspecialty care, offering insights into future trends of eye care delivery with implications for workforce research and policy interventions.
Machine learning models were developed to identify ophthalmology subspecialists using deidentified patient data from the IRIS® Registry (Intelligent Research in Sight) from 2013 to 2023.
Binary subspecialty classification was performed using random forest models with fivefold cross validation, and multispecialty classification used 4 approaches (diagnosis only, procedure only, prescription only, and combined) on data from 9032 ophthalmologists.
Related Articles5
Comparison of Deep Learning and Clinician Performance for Detecting Referable Glaucoma from Fundus Photographs in a Safety Net Population.
Cohort StudyArtificial Intelligence and Ophthalmic Clinical Registries.
Systematic ReviewEvaluating a Foundation Artificial Intelligence Model for Glaucoma Detection Using Color Fundus Photographs.
Observational StudyContinuous 24-hour intraocular pressure monitoring in normal Chinese adults using a novel contact lens sensor system.
Observational StudyFeasibility and acceptance of artificial intelligence-based diabetic retinopathy screening in Rwanda.
Observational StudyIs this article assigned to the wrong chapter(s)? Let us know.