Evaluation of a Novel Deep Learning Based Screening System for Pathologic Myopia

Overview

Journal Int J Ophthalmol

Specialty Ophthalmology

Date 2023 Sep 19

PMID 37724265

Authors

Pei-Fang Ren

Xu-Yuan Tang

Chen-Ying Yu

Li-Li Zhu

Wei-Hua Yang

Ye Shen

Affiliations

Soon will be listed here.

Abstract

Aim: To evaluate the clinical application value of the artificial intelligence assisted pathologic myopia (PM-AI) diagnosis model based on deep learning.

Methods: A total of 1156 readable color fundus photographs were collected and annotated based on the diagnostic criteria of Meta-pathologic myopia (PM) (2015). The PM-AI system and four eye doctors (retinal specialists 1 and 2, and ophthalmologists 1 and 2) independently evaluated the color fundus photographs to determine whether they were indicative of PM or not and the presence of myopic choroidal neovascularization (mCNV). The performance of identification for PM and mCNV by the PM-AI system and the eye doctors was compared and evaluated the relevant statistical analysis.

Results: For PM identification, the sensitivity of the PM-AI system was 98.17%, which was comparable to specialist 1 (=0.307), but was higher than specialist 2 and ophthalmologists 1 and 2 (<0.001). The specificity of the PM-AI system was 93.06%, which was lower than specialists 1 and 2, but was higher than ophthalmologists 1 and 2. The PM-AI system showed the value of 0.904, while the values of specialists 1, 2 and ophthalmologists 1, 2 were 0.968, 0.916, 0.772 and 0.730, respectively. For mCNV identification, the AI system showed the sensitivity of 84.06%, which was comparable to specialists 1, 2 and ophthalmologist 2 (>0.05), and was higher than ophthalmologist 1. The specificity of the PM-AI system was 95.31%, which was lower than specialists 1 and 2, but higher than ophthalmologists 1 and 2. The PM-AI system gave the value of 0.624, while the values of specialists 1, 2 and ophthalmologists 1 and 2 were 0.864, 0.732, 0.304 and 0.238, respectively.

Conclusion: In comparison to the senior ophthalmologists, the PM-AI system based on deep learning exhibits excellent performance in PM and mCNV identification. The effectiveness of PM-AI system is an auxiliary diagnosis tool for clinical screening of PM and mCNV.

Citing Articles

A study on the variability and correlation of ocular biological measurement parameters in adult myopic patients.

Zhou F, Chen N, Qian H, Gong D, Li K Front Med (Lausanne). 2025; 11():1526703.

PMID: 39839655 PMC: 11746085. DOI: 10.3389/fmed.2024.1526703.

Quantitative analysis of retinal vascular parameters changes in school-age children with refractive error using artificial intelligence.

Liu L, Zhong L, Zeng L, Liu F, Yu X, Xie L Front Med (Lausanne). 2025; 11():1528772.

PMID: 39811156 PMC: 11729343. DOI: 10.3389/fmed.2024.1528772.

Quantitative analysis of optic disc changes in school-age children with ametropia based on artificial intelligence.

Liu F, Yu X, Wang Y, Cao M, Xie L, Liu J Int J Ophthalmol. 2023; 16(11):1727-1733.

PMID: 38028515 PMC: 10626368. DOI: 10.18240/ijo.2023.11.01.

References

Hayashi K, Ohno-Matsui K, Shimada N, Moriyama M, Kojima A, Hayashi W . Long-term pattern of progression of myopic maculopathy: a natural history study. Ophthalmology. 2010; 117(8):1595-611, 1611.e1-4. DOI: 10.1016/j.ophtha.2009.11.003. View

Hubbard D, Cox P, Redd T . Assistive applications of artificial intelligence in ophthalmology. Curr Opin Ophthalmol. 2023; 34(3):261-266. PMC: 10065924. DOI: 10.1097/ICU.0000000000000939. View

Wu X, Liu L, Zhao L, Guo C, Li R, Wang T . Application of artificial intelligence in anterior segment ophthalmic diseases: diversity and standardization. Ann Transl Med. 2020; 8(11):714. PMC: 7327317. DOI: 10.21037/atm-20-976. View

Bullimore M, Ritchey E, Shah S, Leveziel N, Bourne R, Flitcroft D . The Risks and Benefits of Myopia Control. Ophthalmology. 2021; 128(11):1561-1579. DOI: 10.1016/j.ophtha.2021.04.032. View

Esteva A, Kuprel B, Novoa R, Ko J, Swetter S, Blau H . Dermatologist-level classification of skin cancer with deep neural networks. Nature. 2017; 542(7639):115-118. PMC: 8382232. DOI: 10.1038/nature21056. View

Tan T, Anees A, Chen C, Li S, Xu X, Li Z . Retinal photograph-based deep learning algorithms for myopia and a blockchain platform to facilitate artificial intelligence medical research: a retrospective multicohort study. Lancet Digit Health. 2021; 3(5):e317-e329. DOI: 10.1016/S2589-7500(21)00055-8. View

Flitcroft D, He M, Jonas J, Jong M, Naidoo K, Ohno-Matsui K . IMI - Defining and Classifying Myopia: A Proposed Set of Standards for Clinical and Epidemiologic Studies. Invest Ophthalmol Vis Sci. 2019; 60(3):M20-M30. PMC: 6735818. DOI: 10.1167/iovs.18-25957. View

Zhao X, Wu Y, Song G, Li Z, Zhang Y, Fan Y . A deep learning model integrating FCNNs and CRFs for brain tumor segmentation. Med Image Anal. 2017; 43:98-111. PMC: 6029627. DOI: 10.1016/j.media.2017.10.002. View

Liu C, Liu L, Lai C, Chou J, Yeh L, Chen K . Multimodal imaging including spectral-domain optical coherence tomography and confocal near-infrared reflectance for characterization of lacquer cracks in highly myopic eyes. Eye (Lond). 2014; 28(12):1437-45. PMC: 4268461. DOI: 10.1038/eye.2014.221. View

10.

Cheung C, Arnold J, Holz F, Park K, Lai T, Larsen M . Myopic Choroidal Neovascularization: Review, Guidance, and Consensus Statement on Management. Ophthalmology. 2017; 124(11):1690-1711. DOI: 10.1016/j.ophtha.2017.04.028. View

11.

Li F, Song D, Chen H, Xiong J, Li X, Zhong H . Development and clinical deployment of a smartphone-based visual field deep learning system for glaucoma detection. NPJ Digit Med. 2020; 3:123. PMC: 7508974. DOI: 10.1038/s41746-020-00329-9. View

12.

Vongphanit J, Mitchell P, Wang J . Prevalence and progression of myopic retinopathy in an older population. Ophthalmology. 2002; 109(4):704-11. DOI: 10.1016/s0161-6420(01)01024-7. View

13.

Tan N, Liu J, Wong D, Lim J, Zhang Z, Lu S . Automatic detection of pathological myopia using variational level set. Annu Int Conf IEEE Eng Med Biol Soc. 2009; 2009:3609-12. DOI: 10.1109/IEMBS.2009.5333517. View

14.

Verkicharla P, Ohno-Matsui K, Saw S . Current and predicted demographics of high myopia and an update of its associated pathological changes. Ophthalmic Physiol Opt. 2015; 35(5):465-75. DOI: 10.1111/opo.12238. View

15.

Lu L, Zhou E, Yu W, Chen B, Ren P, Lu Q . Development of deep learning-based detecting systems for pathologic myopia using retinal fundus images. Commun Biol. 2021; 4(1):1225. PMC: 8548495. DOI: 10.1038/s42003-021-02758-y. View

16.

Wang R, Zuo G, Li K, Li W, Xuan Z, Han Y . Systematic bibliometric and visualized analysis of research hotspots and trends on the application of artificial intelligence in diabetic retinopathy. Front Endocrinol (Lausanne). 2022; 13:1036426. PMC: 9659570. DOI: 10.3389/fendo.2022.1036426. View

17.

Zhang X, Wang C, Jiang H, Jing S, Zhao J, Yu Z . Trends in research related to high myopia from 2010 to 2019: a bibliometric and knowledge mapping analysis. Int J Ophthalmol. 2021; 14(4):589-599. PMC: 8025176. DOI: 10.18240/ijo.2021.04.17. View

18.

Lu L, Ren P, Tang X, Yang M, Yuan M, Yu W . AI-Model for Identifying Pathologic Myopia Based on Deep Learning Algorithms of Myopic Maculopathy Classification and "Plus" Lesion Detection in Fundus Images. Front Cell Dev Biol. 2021; 9:719262. PMC: 8554089. DOI: 10.3389/fcell.2021.719262. View

19.

Ohno-Matsui K, Yoshida T, Futagami S, Yasuzumi K, Shimada N, Kojima A . Patchy atrophy and lacquer cracks predispose to the development of choroidal neovascularisation in pathological myopia. Br J Ophthalmol. 2003; 87(5):570-3. PMC: 1771643. DOI: 10.1136/bjo.87.5.570. View

20.

LeCun Y, Bengio Y, Hinton G . Deep learning. Nature. 2015; 521(7553):436-44. DOI: 10.1038/nature14539. View