Top

Graefe's Archive for Clinical and Experimental Ophthalmology

Published in:

01-03-2020 | Glaucoma | Glaucoma

Detecting glaucoma based on spectral domain optical coherence tomography imaging of peripapillary retinal nerve fiber layer: a comparison study between hand-crafted features and deep learning model

Authors: Ce Zheng, Xiaolin Xie, Longtao Huang, Binyao Chen, Jianling Yang, Jiewei Lu, Tong Qiao, Zhun Fan, Mingzhi Zhang

Published in: Graefe's Archive for Clinical and Experimental Ophthalmology | Issue 3/2020

Abstract

Purpose

To develop a deep learning (DL) model for automated detection of glaucoma and to compare diagnostic capability against hand-craft features (HCFs) based on spectral domain optical coherence tomography (SD-OCT) peripapillary retinal nerve fiber layer (pRNFL) images.

Methods

A DL model with pre-trained convolutional neural network (CNN) based was trained using a retrospective training set of 1501 pRNFL OCT images, which included 690 images from 153 glaucoma patients and 811 images from 394 normal subjects. The DL model was further tested in an independent test set of 50 images from 50 glaucoma patients and 52 images from 52 normal subjects. A customized software was used to extract and measure HCFs including pRNFL thickness in average and four different sectors. Area under the receiver operator characteristics (AROC) curves was calculated to compare the diagnostic capability between DL model and hand-crafted pRNFL parameters.

Results

In this study, the DL model achieved an AROC of 0.99 [CI: 0.97 to 1.00] which was significantly larger than the AROC values of all other HCFs (AROCs 0.661 with 95% CI 0.549 to 0.772 for temporal sector, AROCs 0.696 with 95% CI 0.549 to 0.799 for nasal sector, AROCs 0.913 with 95% CI 0.855 to 0.970 for superior sector, AROCs 0.938 with 95% CI 0.894 to 0.982 for inferior sector, and AROCs 0.895 with 95% CI 0.832 to 0.957 for average).

Conclusion

Our study demonstrated that DL models based on pre-trained CNN are capable of identifying glaucoma with high sensitivity and specificity based on SD-OCT pRNFL images.

Quigley HA (2011) Glaucoma. Lancet 377:1367–1377CrossRef

Weinreb RN, Aung T, Medeiros FA (2014) The pathophysiology and treatment of glaucoma: a review. JAMA 311:1901–1911CrossRef

Sung KR, Kim JS, Wollstein G et al (2011) Imaging of the retinal nerve fibre layer with spectral domain optical coherence tomography for glaucoma diagnosis. Br J Ophthalmol 95:909–914. https://doi.org/10.1136/bjo.2010.186924 CrossRefPubMed

Medeiros FA, Zangwill LM, Bowd C et al (2005) Evaluation of retinal nerve fiber layer, optic nerve head, and macular thickness measurements for glaucoma detection using optical coherence tomography. Am J Ophthalmol 139:44–55CrossRef

Leung CK, Cheung CY, Weinreb RN et al (2009) Retinal nerve fiber layer imaging with spectral-domain optical coherence tomography: a variability and diagnostic performance study. Ophthalmology 116:1257–1263CrossRef

Knight OJ, Chang RT, Feuer WJ et al (2009) Comparison of retinal nerve fiber layer measurements using time domain and spectral domain optical coherent tomography. Ophthalmology 116:1271–1277CrossRef

Asaoka R, Murata H, Hirasawa K et al (2018) Using deep learning and transform learning to accurately diagnose early-onset glaucoma from macular optical coherence tomography images. Am J Ophthalmol. https://doi.org/10.1016/j.ajo.2018.10.007 (1879-1891 (Electronic))CrossRef

Gulshan V, Peng L, Coram M et al (2016) Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA 316(22):2402CrossRef

Shen W, Zhou M, Yang F et al (2015) Multi-scale convolutional neural networks for lung nodule classification. Info Process Med Imaging 24:588–599

10.

Lakhani P, Sundaram B (2017) Deep learning at chest radiography: automated classification of pulmonary tuberculosis by using convolutional neural networks. Radiology 284(2):574–582. https://doi.org/10.1148/radiol.2017162326 CrossRefPubMed

11.

Le MH, Chen J, Wang L et al (2017) Automated diagnosis of prostate cancer in multi-parametric MRI based on multimodal convolutional neural networks. Phys Med Biol 62(16):6491–6514. https://doi.org/10.1088/1361-6560/aa7731 CrossRef

12.

Asaoka R, Murata H, Iwase A et al (2016) Detecting preperimetric glaucoma with standard automated perimetry using a deep learning classifier. Ophthalmology 123(9):1974–1980. https://doi.org/10.1016/j.ophtha.2016.05.029 CrossRefPubMed

13.

Rasband WS. ImageJ, US. National Institutes of Health, Bethesda, Maryland, USA. http://imagej.nih.gov/ij/. Accessed 2 Oct 2019

14.

Shabana N, Aquino MC, See J et al (2012) Quantitative evaluation of anterior chamber parameters using anterior segment optical coherence tomography in primary angle closure mechanisms. Clin Exp Ophthalmol 40(8):792–801. https://doi.org/10.1111/j.1442-9071.2012.02805.x CrossRefPubMed

15.

Zheng C, de Leon JM, Cheung CY et al (2016) Determinants of pupil diameters and pupil dynamics in an adult Chinese population. Graefes Arch Clin Exp Ophthalmol 254(5):929–936. https://doi.org/10.1007/s00417-016-3272-7 CrossRefPubMed

16.

Chen B, Gao E, Chen H et al (2016) Profile and determinants of retinal optical intensity in normal eyes with spectral domain optical coherence tomography. PLoS One 11(2):1–16. https://doi.org/10.1371/journal.pone.0148183 CrossRef

17.

Russakovsky O, Deng J, Su H et al (2015) ImageNet large scale visual recognition challenge. Int J Comput Vis 115.3(2015):211–252. https://doi.org/10.1007/s11263-015-0816-y CrossRef

18.

Szegedy C, Vanhoucke V, Ioffe S, et al. (2016) Rethinking the inception architecture for computer vision. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2818-2826.doi:https://doi.org/10.1109/CVPR.2016.308

19.

Zhou B, Khosla A, Lapedriza A, et al. (2015) Learning deep features for discriminative localization. CVPR'16 (arXiv:1512.04150, 2015)

20.

DeLong ER, DeLong DM, Clarke-Pearson DL (1988) Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 44(3):837–845CrossRef

21.

Rao HL, Zangwill LM, Weinreb RN et al (2010) Comparison of different spectral domain optical coherence tomography scanning areas for glaucoma diagnosis. Ophthalmology 117(9):1692–1699.e1. https://doi.org/10.1016/j.ophtha.2010.01.031 CrossRefPubMed

22.

Seong M, Sung KR, Choi EH et al (2010) Macular and peripapillary retinal nerve fiber layer measurements by spectral domain optical coherence tomography in normal-tension glaucoma. Investig Ophthalmol Vis Sci 51(3):1446–1452. https://doi.org/10.1167/iovs.09-4258 CrossRef

23.

Wang X, Peng Y, Lu L, et al. (2017) ChestX-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 3462-3471. https://doi.org/10.1109/CVPR.2017.369

24.

Yosinski J, Clune J, Nguyen A, et al. (2015) Understanding neural networks through deep visualization. arXiv preprint. https://arxiv.org/abs/1506.06579. Published June 22, 2015. Accessed Augest 12, 2019

25.

Vermeer KA, van der Schoot J, Lemij HG et al (2012) RPE-normalized RNFL attenuation coefficient maps derived from volumetric OCT imaging for glaucoma assessment. Investig Ophthalmol Vis Sci 53(10):6102–6108. https://doi.org/10.1167/iovs.12-9933 CrossRef

26.

Xu H, Zhai R, Zong Y et al (2018) Comparison of retinal microvascular changes in eyes with high-tension glaucoma or normal-tension glaucoma: a quantitative optic coherence tomography angiographic study. Graefes Arch Clin Exp Ophthcalmol 256(6):1179–1186CrossRef

27.

Rolle T, Briamonte C, Curto D et al (2011) Ganglion cell complex and retinal nerve fiber layer measured by fourier-domain optical coherence tomography for early detection of structural damage in patients with preperimetric glaucoma. Clin Ophthalmol 5:961–969. https://doi.org/10.2147/OPTH.S20249 CrossRefPubMedPubMedCentral

28.

Kim S Y , Park H Y L , Park C K . (2012) The effects of peripapillary atrophy on the diagnostic ability of stratus and cirrus oct in the analysis of optic nerve head parameters and disc size [J]. Invest Ophthalmol Vis Sci 53(8)CrossRef

Title: Detecting glaucoma based on spectral domain optical coherence tomography imaging of peripapillary retinal nerve fiber layer: a comparison study between hand-crafted features and deep learning model
Authors: Ce Zheng
Xiaolin Xie
Longtao Huang
Binyao Chen
Jianling Yang
Jiewei Lu
Tong Qiao
Zhun Fan
Mingzhi Zhang
Publication date: 01-03-2020
Publisher: Springer Berlin Heidelberg
Keyword: Glaucoma
Published in: Graefe's Archive for Clinical and Experimental Ophthalmology / Issue 3/2020
Print ISSN: 0721-832X
Electronic ISSN: 1435-702X
DOI: https://doi.org/10.1007/s00417-019-04543-4

At a glance: The STEP trials

Springer Medicine

Detecting glaucoma based on spectral domain optical coherence tomography imaging of peripapillary retinal nerve fiber layer: a comparison study between hand-crafted features and deep learning model

Abstract

Purpose

Methods

Results

Conclusion

At a glance: The STEP trials

Springer Medicine

Abstract

Purpose

Methods

Results

Conclusion

Please log in to get access to this content

Other articles of this Issue 3/2020

Short-term effects of caffeine intake on anterior chamber angle and intraocular pressure in low caffeine consumers

Effect of light and diurnal variation on macular thickness in X-linked retinoschisis: a case series

Refractive status during pregnancy in the United States: results from NHANES 2005–2008

Analysis of blink activity and anomalous eyelid movements in patients with hemifacial spasm

Microvascular retinal changes in pre-clinical diabetic retinopathy as detected by optical coherence tomographic angiography

The effect of colonoscopy on intraocular pressure: an observational prospective study