基于深度学习模型进行屈光不正与玻璃膜疣的相关性分析

doi:10.3760/cma.j.cn115909-20191106-00299

摘要
图/表
参考文献(25)
相关文章 (15)

全文: PDF 1732KB HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要目的：采用深度学习模型对彩色眼底图像中的玻璃膜疣进行自动判读，并进一步探讨屈光不正与玻璃膜疣的相关关系。方法：病例对照研究。选取2017年1─12月在宁波市医疗中心李惠利医院行健康体检且年龄大于50岁的参与者，共1 035例（2 055眼）的彩色眼底图像，使用DeepSeeNet深度学习模型对彩色眼底图像进行自动分析判读，按玻璃膜疣大小进行分组。同时随机选择392张（19.1%）彩色眼底图像采用人工阅片方法对玻璃膜疣大小进行判读，采用Cohen's Kappa检验比较2种判读方法的一致性。根据电脑验光数据计算等效球镜度（SE），并分为中重度远视（>+3.0 D）、轻度远视（+0.51～+3.0 D）、正视（-0.5～+0.5 D）、轻度近视（-3.0～-0.51 D）、中重度近视（<-3.0 D）。应用 Logistic回归模型进行屈光不正与玻璃膜疣的相关性分析。结果：深度学习模型对玻璃膜疣大小判读与人工阅片结果具有高度一致性（κ＝0.67，P<0.001）。在矫正性别、年龄等因素后，SE向正方向增加与大玻璃膜疣发生风险相关（OR=1.03，95%CI：1.01～1.04，P<0.001）。中高度近视为玻璃膜疣的保护因素，与正视者相比，中重度近视者发生大玻璃膜疣的风险为0.89（95%CI：0.82～0.97）；相反，中高度远视为危险因素，与正视者相比其发生大玻璃膜疣的风险为1.20（95%CI：1.03～1.39）。结论：屈光不正与玻璃膜疣的发生相关。深度学习作为一种新型技术，除了可以增强医师临床诊断的速度和精准性，同样可以为年龄相关性黄斑变性相关的科研提供线索。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章

关键词 ：屈光不正, 黄斑变性, 年龄相关, 视网膜玻璃膜疣, 深度学习, 人工智能

Abstract：Objective: To classify drusen size in color fundus photographs by using an automated deep learning model and to investigate the association between refractive error and drusen. Methods: This was a casecontrol study. There were 2 055 color fundus photographs and refractive status were obtained from 1 035 participants aged 50 years and older who underwent physical examinations in Ningbo Medical Center, Lihuili Hospital from January 2017 to December 2017. DeepSeeNet, a deep learning mode, was used to detect drusen size from color fundus photographs. Three hundred ninety-two fundus photographs were randomly selected and assessed manually by one retinal specialist. Cohen's kappa was used to evaluate the consistency between the DeepSeeNet and the retinal specialist. The spherical equivalent refraction was calculated as adiopter (D) according to the computer optometry data, and classified as emmetropia (-0.5- +0.5 D), mild myopia (-3.0--0.51 D) or hyperopia (+0.51-+3.0 D), and moderate to severe myopia (<-3.0 D) or hyperopia (>+3.0 D). Statistical analysis was performed using R statistical software, and a logistic regression model was used to analyze the association between refractive error and drusen size. Results: The classification results of the deep learning model were highly consistent with those of the retinal specialist (κ=0.67, P<0.001). After adjustment for confounding factors, the increase in spherical equivalent was associated with an increased risk of large drusen (OR=1.03, 95%CI: 1.01-1.04, P<0.001). Compared with emmetropia, moderate to high myopia was associated with a lower OR of large drusen (OR=0.89, 95%CI: 0.82-0.97); moderate to high hyperopia was associated with a higher OR of large drusen (OR=1.20, 95%CI: 1.03-1.39). Conclusions: Refractive error is associated with the development of drusen. These results indicate that the use of a deep learning model not only enhances the speed and accuracy of clinical diagnosis, but also provides clues for age-related macular disease related scientific research.

Key words： refractive errors macular degeneration, age-related retinal drusen deep learning artificial intelligence

收稿日期: 2019-11-06

通讯作者: 顾虹（ORCID：0000-0001-5691-4571），Email：guhong_1984@126.com

引用本文:

顾虹沈肇萌李志国. 基于深度学习模型进行屈光不正与玻璃膜疣的相关性分析[J]. 中华眼视光学与视觉科学杂志, 2020, 22(5): 360-366. Hong Gu, Zhaomeng Shen, Zhiguo Li. Correlation Analysis between Refractive Error and the Risk of Drusen Based on a Deep Learning Model. Chinese Journal of Optometry Ophthalmology and Visual science, 2020, 22(5): 360-366. DOI: 10.3760/cma.j.cn115909-20191106-00299

链接本文:

https://www.cjoovs.com/CN/10.3760/cma.j.cn115909-20191106-00299 或 https://www.cjoovs.com/CN/Y2020/V22/I5/360

[1]	Shen L, Melles RB, Metlapally R, et al. The association of refractive error with glaucoma in a multiethnic population. Ophthalmology, 2016, 123(1): 92-101. DOI: 10.1016/ j.ophtha.2015.07.002.
[2]	项勇刚, 夏凌云, 张勇, 等. 中国人近视与原发性开角型青光眼相关性的Meta分析. 临床眼科杂志, 2014, 22(3): 259-262. DOI: 10.3969/j.issn.1006-8422. 2014.03.024.
[3]	Man REK, Gan ATL, Gupta G, et al. Is myopia associated with the incidence and progression of diabetic retinopathy? Am J Ophthalmol, 2019, 208: 226-233. DOI: 10.1016/ j.ajo.2019.05.012.
[4]	贾杰瑞, 原慧萍. 屈光不正和眼球生物学参数与糖尿病视网膜病变的相关性. 国际眼科杂志, 2018, 18(3): 466-469. DOI: 10.3980/j.issn.1672-5123.2018.3.14.
[5]	Lavanya R, Kawasaki R, Tay WT, et al. Hyperopic refractive error and shorter axial length are associated with age-related macular degeneration: The Singapore Malay Eye Study. Invest Ophthalmol Vis Sci, 2010, 51(12): 6247-6252. DOI: 10.1167/ iovs.10-5229.
[6]	Mao FF, Yang XH, Yang K, et al. Six year incidence and risk factors for age related macular degeneration in a rural chinese population: The Handan Eye study. Invest Ophthalmol Vis Sci, 2019, 60(15): 4966-4917. DOI: 10.1167/iovs.19-27325.
[7]	Lin SC, Singh K, Chao DL, et al. Refractive error and the risk of age-related macular degeneration in the South Korean population. Asia Pac J Ophthalmol, 2016, 5(2): 115-121. DOI: 10.1097/APO.0000000000000169.
[8]	Wang CP, Lai CH, Huang EJ, et al. Axial length and subfoveal choroidal thickness in individuals with age-related macular degeneration. Taiwan J Ophthalmol, 2015, 5(4): 169-176. DOI: 10.1016/j.tjo.2015.10.002.
[9]	Fisher DE, Klein BE, Wong TY, et al. Incidence of age-related macular degeneration in a Multi-Ethnic United States Population: The Multi-Ethnic Study of Atherosclerosis. Ophthalmology, 2016, 123(6): 1297-1308. DOI: 10.1016/ j.ophtha.2015.12.026.
[10]	Vitale S, Clemons TE, Agron E, et al. Evaluating the Validity of the Age-Related Eye Disease Study Grading Scale for Age-Related Macular Degeneration: AREDS2 Report 10. JAMA Ophthalmol, 2016, 134(9): 1041-1047. DOI: 10.1001/ jamaophthalmol.2016.2383.
[11]	王雯秋, 孙晓东. 人工智能辅助眼底影像——机遇与挑战并存. 中华眼视光学与视觉科学杂志, 2019, 21(3): 166-169. DOI: 10.3760/cma.j.issn.1674-845X.2019.03.002.
[12]	Burlina PM, Joshi N, Pekala M, et al. Automated Grading of Age-Related Macular Degeneration From Color Fundus Images Using Deep Convolutional Neural Networks. JAMA Ophthalmol, 2017, 135(11): 1170-1176. DOI: 10.1001/ jamaophthalmol.2017.3782.
[13]	Burlina PM, Joshi N, Pacheco KD, et al. Use of deep learning for detailed severity characterization and estimation of 5-year risk among patients with age-related macular degeneration. JAMA Ophthalmol, 2018, 136(12): 1359-1366. DOI: 10.1001/ jamaophthalmol.2018.4118.
[14]	Grassmann F, Mengelkamp J, Brandl C, et al. A deep learning algorithm for prediction of age-related eye disease study severity scale for age-related macular degeneration from color fundus photography. Ophthalmology, 2018, 125(9): 1410-1420. DOI: 10.1016/j.ophtha.2018.02.037.
[15]	Pead E, Megaw R, Cameron J, et al. Automated detection of age-related macular degeneration in color fundus photography: a systematic review. Surv Ophthalmol, 2019, 64(4): 498-511.DOI:10.1016/j.survophthal.2019.02.003.
	DOI: 10.1016/j.survophthal.2019.02.003.
[16]	Peng Y, Dharssi S, Chen Q, et al. DeepSeeNet: A Deep Learning Model for Automated Classification of Patient-based Agerelated Macular Degeneration Severity from Color Fundus Photographs. Ophthalmology, 2019, 126(4): 565-575. DOI: 10.1016/j.ophtha.2018.11.015.
[17]	Ferris FL, Davis MD, Clemons TE, et al. A simplified severity scale for age-related macular degeneration: AREDS Report No. 18. Arch Ophthalmol, 2005, 123(11): 1570-1574. DOI: 10.1001/ archopht.123.11.1570.
[18]	Keenan TD, Dharssi S, Peng Y, et al. A deep learning approach for automated detection of geographic atrophy from color fundus photographs. Ophthalmology, 2019, 126(11): 1533-1540. DOI: 10.1016/j.ophtha.2019.06.005.
[19]	Klein R, Klein BE, Tomany S C, et al. Ten-year incidence and progression of age-related maculopathy: The Beaver Dam eye study. Ophthalmology, 2002, 109(10): 1767-1779. DOI: 10.1016/s0161-6420(02)01146-6.
[20]	Wang JJ, Foran S, Smith W, et al. Risk of age-related macular degeneration in eyes with macular drusen or hyperpigmentation: the Blue Mountains Eye Study cohort. Arch Ophthalmol, 2003, 121(5): 658-663. DOI: 10.1001/archopht.121.5.658.
[21]	Age-Related Eye Disease Study Research Group. Risk factors associated with age-related macular degeneration. A casecontrol study in the age-related eye disease study: Age-Related Eye Disease Study Report Number 3. Ophthalmology, 2000, 107(12): 2224-2232. DOI: 10.1016/s0161-6420(00)00409-7.
[22]	Fraser-Bell S, Choudhury F, Klein R, et al. Ocular risk factors for age-related macular degeneration: the Los Angeles Latino Eye Study. Am J Ophthalmol, 2010, 149(5): 735-740. DOI: 10.1016/j.ajo.2009.11.013.
[23]	Xu L, Li Y, Wang S, et al. Characteristics of highly myopic eyes: the Beijing Eye Study. Ophthalmology, 2007, 114(1): 121-126. DOI: 10.1016/j.ophtha.2006.05.071.
[24]	Friedman E. A hemodynamic model of the pathogenesis of age-related macular degeneration. Am J Ophthalmol, 1997, 124(5): 677-682. DOI: 10.1016/s0002-9394(14)70906-7.
[25]	Pallikaris IG, Kymionis GD, Ginis HS, et al. Ocular rigidity in patients with age-related macular degeneration. Am J Ophthalmol, 2006, 141(4): 611-615. DOI:10.1016/j.ajo.2005.11.010.