Identification of ocular refraction using a novel intelligent retinoscopy system

Abstract

Abstract Background: The evaluation of refraction is indispensable in ophthalmic clinics, generally requiring a refractor or retinoscopy under cycloplegia. Retinal fundus photographs (RFPs) supply a wealth of information related to the human eye and might provide a new approach that is more convenient and objective. Here, we aimed to develop and validate a fusion model-based intelligent retinoscopy system (FMIRS) to identify ocular refraction via RFPs and compare with the cycloplegic refraction. ​In this population-based comparative study, we retrospectively collected 11,973 RFPs from May 1, 2020 to November 20, 2021. The FMIRS was constructed, and the performance of the regression models of sphere and cylinder was evaluated. The accuracy, sensitivity, specificity, area under the receiver operating characteristic curve, and F1-score were used to evaluate the classification model of the cylinder axis. Results: Overall, 11,973 images were included. For sphere and cylinder, the mean absolute error values between the FMIRS and cycloplegic refraction were 0.50 D and 0.31 D, representing an increase of 29.41% and 26.67%, respectively, when compared with those of the single models. The correlation coefficients (r) were 0.949 and 0.807, respectively. For axis analysis, the accuracy, specificity, sensitivity, and area under the curve value of the classification model were 0.89, 0.941, 0.882, and 0.814, respectively, and the F1-score was 0.88. Conclusions: The FMIRS successfully identified ocular refraction accurately in sphere, cylinder, and axis, and it showed good agreement with the cycloplegic refraction. The RFPs can not only provide comprehensive fundus information but also the refraction state of the eye, emphasising their potential clinical value.