Abstract:Background Myopia is the leading cause of visual impairment and affects millions of children worldwide. Timely and annual manual optometric screenings of the entire at-risk population improve outcomes, but screening is challenging due to the lack of availability and training of assessors and the economic burden imposed by the screenings. Recently, deep learning and computer vision have shown powerful potential for disease screening. However, these techniques have not been applied to large-scale myopia screening using ocular appearance images. Methods We trained a deep learning system (DLS) for myopia detection using 2,350 ocular appearance images (processed by 7,050 pictures) from children aged 6 to 18. Myopia is defined as a spherical equivalent refraction (SER) [the algebraic sum in diopters (D), sphere + 1/2 cylinder] ≤−0.5 diopters. Saliency maps and gradient class activation maps (grad-CAM) were used to highlight the regions recognized by VGG-Face. In a prospective clinical trial, 100 ocular appearance images were used to assess the performance of the DLS. Results The area under the curve (AUC), sensitivity, and specificity of the DLS were 0.9270 (95% CI, 0.8580–0.9610), 81.13% (95% CI, 76.86–5.39%), and 86.42% (95% CI, 82.30–90.54%), respectively. Based on the saliency maps and grad-CAMs, the DLS mainly focused on eyes, especially the temporal sclera, rather than the background or other parts of the face. In the prospective clinical trial, the DLS achieved better diagnostic performance than the ophthalmologists in terms of sensitivity [DLS: 84.00% (95% CI, 73.50–94.50%) versus ophthalmologists: 64.00% (95% CI, 48.00–72.00%)] and specificity [DLS: 74.00% (95% CI, 61.40–86.60%) versus ophthalmologists: 53.33% (95% CI, 30.00–66.00%)]. We also computed AUC subgroups stratified by sex and age. DLS achieved comparable AUCs for children of different sexes and ages. Conclusions This study for the first time applied deep learning to myopia screening using ocular images and achieved high screening accuracy, enabling the remote monitoring of the refractive status in children with myopia. The application of our DLS will directly benefit public health and relieve the substantial burden imposed by myopia-associated visual impairment or blindness.

High Myopia Detection Method on Fundus Images Based on Curriculum Learning

Automatic Diagnosis of Pediatric High Myopia Via Attention-based Patch Residual Shrinkage Network

A Fundus Image Myopia Diagnosis Model Based on Homogeneous Multimodal Feature Fusion

An Artificial Intelligent Risk Classification Method of High Myopia Based on Fundus Images

Attention to Key Fundus Features: A Prior Knowledge-Guided Deep Learning Network for Pediatric High Myopia Detection

Automatic Detection Of Pathological Myopia And High Myopia On Fundus Images

A diagnostic method of high myopia based on transfer learning

Predicting Optical Coherence Tomography-Derived High Myopia Grades From Fundus Photographs Using Deep Learning

Deep Learning-Based Longitudinal Prediction of Childhood Myopia Progression Using Fundus Image Sequences and Baseline Refraction Data

Learn To Recognize Pathological Myopia In Fundus Images Using Bag-Of-Feature And Sparse Learning Approach

A Curriculum Learning-Based Fully Automated System for Quantification of the Choroidal Structure in Highly Myopic Patients

Development of deep learning-based detecting systems for pathologic myopia using retinal fundus images

A deep learning system for myopia onset prediction and intervention effectiveness evaluation in children

Automatic Identification of Myopia Based on Ocular Appearance Images Using Deep Learning

Automatic Diagnosis of Glaucoma on Color Fundus Images Using Adaptive Mask Deep Network

AI-Based Fully Automatic Analysis of Retinal Vascular Morphology in Pediatric High Myopia

Multi-class classification of pathological myopia based on fundus photography

Deep Learning Detection of Early Retinal Peripheral Degeneration From Ultra-Widefield Fundus Photographs of Asymptomatic Young Adult (17-19 Years) Candidates to Airforce Cadets

Self-supervised learning-enhanced deep learning method for identifying myopic maculopathy in high myopia patients

Deep learning system to predict the 5-year risk of high myopia using fundus imaging in children

Deep Learning for the Detection of Multiple Fundus Diseases Using Ultra-widefield Images