Deep learning for vision screening in resource-limited settings: development of multi-branch CNN for refractive error detection based on smartphone image

IntroductionUncorrected refractive errors are a leading cause of preventable vision impairment globally, particularly affecting individuals in low-resource regions where timely diagnosis and screening access remain significant challenges despite the availability of economical treatments.AimThis stud...

Full description

Saved in:
Bibliographic Details
Main Authors: Muhammad Syauqie, Harry Patria, Sutanto Priyo Hastono, Kemal Nazaruddin Siregar, Nila Djuwita Farieda Moeloek
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-07-01
Series:Frontiers in Computer Science
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fcomp.2025.1576958/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849728303807594496
author Muhammad Syauqie
Muhammad Syauqie
Harry Patria
Sutanto Priyo Hastono
Kemal Nazaruddin Siregar
Nila Djuwita Farieda Moeloek
author_facet Muhammad Syauqie
Muhammad Syauqie
Harry Patria
Sutanto Priyo Hastono
Kemal Nazaruddin Siregar
Nila Djuwita Farieda Moeloek
author_sort Muhammad Syauqie
collection DOAJ
description IntroductionUncorrected refractive errors are a leading cause of preventable vision impairment globally, particularly affecting individuals in low-resource regions where timely diagnosis and screening access remain significant challenges despite the availability of economical treatments.AimThis study introduces a novel deep learning-based system for automated refractive error classification using photorefractive images acquired via a standard smartphone camera.MethodsA multi-branch convolutional neural network (CNN) was developed and trained on a dataset of 2,139 corneal images collected from an Indonesian public eye hospital. The model was designed to classify refractive errors into four categories: significant myopia, significant hypermetropia, insignificant refractive error, and not applicable to classified. Grad-CAM visualization was employed to provide insights into the model’s interpretability.ResultsThe 3-branch CNN architecture demonstrated superior performance, achieving an overall test accuracy of 91%, precision of 96%, and recall of 98%, with an area under the curve (AUC) score of 0.9896. Its multi-scale feature extraction pathways were pivotal in effectively addressing overlapping red reflex patterns and subtle variations between classes.ConclusionThis study establishes the feasibility of smartphone-based photorefractive assessment integrated with artificial intelligence for scalable and cost-effective vision screening. By training the CNN model with a real-world dataset representative of Southeast Asian populations, this system offers a reliable solution for early refractive error detection with significant implications for improving accessibility to eye care services in resource-limited settings.
format Article
id doaj-art-2f9b0d1c966149b4b3a5ab73b88dfb22
institution DOAJ
issn 2624-9898
language English
publishDate 2025-07-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Computer Science
spelling doaj-art-2f9b0d1c966149b4b3a5ab73b88dfb222025-08-20T03:09:35ZengFrontiers Media S.A.Frontiers in Computer Science2624-98982025-07-01710.3389/fcomp.2025.15769581576958Deep learning for vision screening in resource-limited settings: development of multi-branch CNN for refractive error detection based on smartphone imageMuhammad Syauqie0Muhammad Syauqie1Harry Patria2Sutanto Priyo Hastono3Kemal Nazaruddin Siregar4Nila Djuwita Farieda Moeloek5Department of Ophthalmology, Faculty of Medicine, Universitas Andalas, Padang, IndonesiaDepartment of Biostatistics and Population Studies, Faculty of Public Health, Universitas Indonesia, Depok, IndonesiaStrathclyde Business School, University of Strathclyde, Glasgow, United KingdomDepartment of Biostatistics and Population Studies, Faculty of Public Health, Universitas Indonesia, Depok, IndonesiaDepartment of Biostatistics and Population Studies, Faculty of Public Health, Universitas Indonesia, Depok, IndonesiaDepartment of Ophthalmology, Faculty of Medicine, Universitas Indonesia, Cipto Mangunkusumo Hospital, Jakarta, IndonesiaIntroductionUncorrected refractive errors are a leading cause of preventable vision impairment globally, particularly affecting individuals in low-resource regions where timely diagnosis and screening access remain significant challenges despite the availability of economical treatments.AimThis study introduces a novel deep learning-based system for automated refractive error classification using photorefractive images acquired via a standard smartphone camera.MethodsA multi-branch convolutional neural network (CNN) was developed and trained on a dataset of 2,139 corneal images collected from an Indonesian public eye hospital. The model was designed to classify refractive errors into four categories: significant myopia, significant hypermetropia, insignificant refractive error, and not applicable to classified. Grad-CAM visualization was employed to provide insights into the model’s interpretability.ResultsThe 3-branch CNN architecture demonstrated superior performance, achieving an overall test accuracy of 91%, precision of 96%, and recall of 98%, with an area under the curve (AUC) score of 0.9896. Its multi-scale feature extraction pathways were pivotal in effectively addressing overlapping red reflex patterns and subtle variations between classes.ConclusionThis study establishes the feasibility of smartphone-based photorefractive assessment integrated with artificial intelligence for scalable and cost-effective vision screening. By training the CNN model with a real-world dataset representative of Southeast Asian populations, this system offers a reliable solution for early refractive error detection with significant implications for improving accessibility to eye care services in resource-limited settings.https://www.frontiersin.org/articles/10.3389/fcomp.2025.1576958/fullrefractive error detectionvision screeningartificial intelligenceconvolutional neural networksmartphonered reflex
spellingShingle Muhammad Syauqie
Muhammad Syauqie
Harry Patria
Sutanto Priyo Hastono
Kemal Nazaruddin Siregar
Nila Djuwita Farieda Moeloek
Deep learning for vision screening in resource-limited settings: development of multi-branch CNN for refractive error detection based on smartphone image
Frontiers in Computer Science
refractive error detection
vision screening
artificial intelligence
convolutional neural network
smartphone
red reflex
title Deep learning for vision screening in resource-limited settings: development of multi-branch CNN for refractive error detection based on smartphone image
title_full Deep learning for vision screening in resource-limited settings: development of multi-branch CNN for refractive error detection based on smartphone image
title_fullStr Deep learning for vision screening in resource-limited settings: development of multi-branch CNN for refractive error detection based on smartphone image
title_full_unstemmed Deep learning for vision screening in resource-limited settings: development of multi-branch CNN for refractive error detection based on smartphone image
title_short Deep learning for vision screening in resource-limited settings: development of multi-branch CNN for refractive error detection based on smartphone image
title_sort deep learning for vision screening in resource limited settings development of multi branch cnn for refractive error detection based on smartphone image
topic refractive error detection
vision screening
artificial intelligence
convolutional neural network
smartphone
red reflex
url https://www.frontiersin.org/articles/10.3389/fcomp.2025.1576958/full
work_keys_str_mv AT muhammadsyauqie deeplearningforvisionscreeninginresourcelimitedsettingsdevelopmentofmultibranchcnnforrefractiveerrordetectionbasedonsmartphoneimage
AT muhammadsyauqie deeplearningforvisionscreeninginresourcelimitedsettingsdevelopmentofmultibranchcnnforrefractiveerrordetectionbasedonsmartphoneimage
AT harrypatria deeplearningforvisionscreeninginresourcelimitedsettingsdevelopmentofmultibranchcnnforrefractiveerrordetectionbasedonsmartphoneimage
AT sutantopriyohastono deeplearningforvisionscreeninginresourcelimitedsettingsdevelopmentofmultibranchcnnforrefractiveerrordetectionbasedonsmartphoneimage
AT kemalnazaruddinsiregar deeplearningforvisionscreeninginresourcelimitedsettingsdevelopmentofmultibranchcnnforrefractiveerrordetectionbasedonsmartphoneimage
AT niladjuwitafariedamoeloek deeplearningforvisionscreeninginresourcelimitedsettingsdevelopmentofmultibranchcnnforrefractiveerrordetectionbasedonsmartphoneimage