EUVIMEDEUVIMED entdecken
Open Access

Deep Learning Prediction of Childhood Myopia Progression Using Fundus Image and Refraction Data

Meng-Tian Kang, Yansong Hu, Ningli Wang, Jing Fu, Ankang Zhou, Yuanyuan Liu, Hongbei Meng, Xuemeng Li, Shengbo Wang, Xuhang Chen, Hubin Zhao, Guohua Hu und 6 weitere

JAMA Network Open · 26.01.2026 · 9 (1) · e2553543

DOI: 10.1001/jamanetworkopen.2025.53543 ↗

Kurz gefasst

Fachartikel in der Zeitschrift JAMA Network Open, online erschienen am 26.01.2026. Verfasst von Meng-Tian Kang und 17 weiteren Autor:innen. Beteiligte Einrichtung: Beijing Tongren Eye Center, Beijing Tongren Hospital, Beijing Institute of Ophthalmology, Beijing Key Laboratory of Intelligent Diagnosis Technology …. Hauptthemen laut MEDLINE: Deep Learning, Myopia, Refraction, Ocular, Fundus Oculi. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Importance Childhood myopia is a global health concern with escalating prevalence and can lead to severe irreversible visual impairment. Early prediction of myopia progression is crucial for timely intervention to prevent high myopia and associated complications.

Objective

To develop and validate a quantitative method, based on a deep learning method and using only fundus images and baseline refraction data, to predict both myopia progression trajectory and high myopia risk in schoolchildren. Design, Setting, and Participants This longitudinal school-based cohort study (Anyang Childhood Eye Study) was conducted from February 2012 to May 2018, with annual follow-up examinations from February 2013 to May 2018. Grade 1 students, aged 6 to 9 years, were recruited from 11 randomly selected urban primary schools in Anyang, Henan Province, China. Children who received myopia control treatments, had amblyopia, or underwent strabismus surgery were excluded. Two independent external validation cohorts were used: the Lhasa cohort (predominantly consisting of Tibetan children) and the Beijing cohort (predominantly consisting of Han Chinese children). Data analysis was performed from July 2024 to July 2025. Main Outcomes and Measures Performance of a novel deep learning model, created from combining convolutional neural network (34-layer residual network) and recurrent neural network (long short-term memory network). Area under the curve (AUC) was used to assess myopia and high myopia risk prediction, and mean absolute error (MAE) was used to assess spherical equivalent refraction (SER) prediction. Myopia was defined as SER of –0.5 D or less, and high myopia was defined as –6.0 D or less, using cycloplegic autorefraction.

Results

Among the 3048 children (mean [SD] age, 7.1 [0.4] years; 1716 females [56.3%]) included, the baseline prevalence rate of myopia was 5.71% (174) and high myopia was 0.5% (15). The deep learning model achieved AUC scores of 0.941 (95% CI, 0.936-0.946) for myopia risk prediction and 0.985 (95% CI, 0.982-0.988) for high myopia risk prediction, with an overall MAE of 0.322 D per year for SER prediction. External validation in the Beijing cohort (n = 130; mean [SD] age, 9.9 [3.7] years; 82 males [63.1%]; 128 Han Chinese [98.5%], 2 Man Chu [1.5%]) and the Lhasa cohort (n = 1039; mean [SD] age, 6.8 [0.5] years; 536 males [51.6%]; 29 Han Chinese [2.8%], 1007 Tibetan [96.9%], 3 [0.2%] from other ethnic minority groups) demonstrated maintained cross-ethnicity performance by the model, with an MAE of 0.355 D and 0.261 D per year, respectively. Conclusions and Relevance In this cohort study, a deep learning model, using minimal baseline data, provided highly accurate prediction of myopia and high myopia risk. This deep learning approach suggests the utility of large-scale screening and early-intervention efforts in resource-limited settings.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningMyopiaRefraction, OcularFundus OculiHumansChildFemaleMaleDisease ProgressionChinaPredictive Learning ModelsLongitudinal StudiesPrediction AlgorithmsConvolutional Neural NetworksCohort Studies

Autor:innen und Institutionen

  • Beijing Tongren Eye Center, Beijing Tongren Hospital, Beijing Institute of Ophthalmology, Beijing Key Laboratory of Intelligent Diagnosis Technology and Equipment for Optic Nerve-Related Eye Diseases, National Engineeri…
  • School of Instrumentation and Optoelectronic Engineering, Beihang University, Beijing, China.
  • School of Mechanical Engineering, Beihang University, Beijing, China.
  • Division of Neurosurgery, Department of Clinical Neurosciences, University of Cambridge, Cambridge, United Kingdom.
  • Division of Surgery and Interventional Science, University College London, Stanmore, United Kingdom.
  • Department of Electronic Engineering, The Chinese University of Hong Kong, Shatin, N. T., Hong Kong SAR, China.

Zitieren

APA

Kang, M. T., Hu, Y., Wang, N., Fu, J., Zhou, A., Liu, Y., Meng, H., Li, X., Wang, S., Chen, X., Zhao, H., Hu, G., Wang, W., Dai, Y., Nathan, A., Smielewski, P., Gao, S., & Li, S. M. (2026). Deep Learning Prediction of Childhood Myopia Progression Using Fundus Image and Refraction Data. JAMA Network Open, 9(1), e2553543. https://doi.org/10.1001/jamanetworkopen.2025.53543

Vancouver

Kang MT, Hu Y, Wang N, Fu J, Zhou A, Liu Y, et al. Deep Learning Prediction of Childhood Myopia Progression Using Fundus Image and Refraction Data. JAMA Network Open. 2026;9(1):e2553543. doi:10.1001/jamanetworkopen.2025.53543

BibTeX
@article{Kang2026,
  author = {Kang, Meng-Tian and Hu, Yansong and Wang, Ningli and Fu, Jing and Zhou, Ankang and Liu, Yuanyuan and Meng, Hongbei and Li, Xuemeng and Wang, Shengbo and Chen, Xuhang and Zhao, Hubin and Hu, Guohua and Wang, Wei and Dai, Yanning and Nathan, Arokia and Smielewski, Peter and Gao, Shuo and Li, Shi-Ming},
  title = {Deep Learning Prediction of Childhood Myopia Progression Using Fundus Image and Refraction Data},
  journal = {JAMA Network Open},
  year = {2026},
  volume = {9},
  number = {1},
  pages = {e2553543},
  doi = {10.1001/jamanetworkopen.2025.53543}
}
Kostenlos starten →