Skip to main content

Advertisement

SpringerLink
Log in

Generative adversarial network-based deep learning approach in classification of retinal conditions with optical coherence tomography images

  • Medical Ophthalmology
  • Published:
Graefe's Archive for Clinical and Experimental Ophthalmology Aims and scope Submit manuscript

Abstract

Purpose

To determine whether a deep learning approach using generative adversarial networks (GANs) is beneficial for the classification of retinal conditions with Optical coherence tomography (OCT) images.

Methods

Our study utilized 84,452 retinal OCT images obtained from a publicly available dataset (Kermany Dataset). Employing GAN, synthetic OCT images are produced to balance classes of retinal disorders. A deep learning classification model is constructed using pretrained deep neural networks (DNNs), and outcomes are evaluated using 2082 images collected from patients who visited the Department of Ophthalmology and the Department of Endocrinology and Metabolism at the Tri-service General Hospital in Taipei from January 2017 to December 2021.

Results

The highest classification accuracies accomplished by deep learning machines trained on the unbalanced dataset for its training set, validation set, fivefold cross validation (CV), Kermany test set, and TSGH test set were 97.73%, 96.51%, 97.14%, 99.59%, and 81.03%, respectively. The highest classification accuracies accomplished by deep learning machines trained on the synthesis-balanced dataset for its training set, validation set, fivefold CV, Kermany test set, and TSGH test set were 98.60%, 98.41%, 98.52%, 99.38%, and 84.92%, respectively. In comparing the highest accuracies, deep learning machines trained on the synthesis-balanced dataset outperformed deep learning machines trained on the unbalanced dataset for the training set, validation set, fivefold CV, and TSGH test set.

Conclusions

Overall, deep learning machines on a synthesis-balanced dataset demonstrated to be advantageous over deep learning machines trained on an unbalanced dataset for the classification of retinal conditions.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6

Similar content being viewed by others

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Abbreviations

GAN :

Generative adversarial network

OCT :

Optical coherence tomography

CV :

Cross validation

ADA :

Adaptive discriminator augmentation

TSGH :

Tri-service General Hospital

CNV :

Choroidal neovascularization

DME :

Diabetic macular edema

DNNs :

Deep neural networks

FFHQ :

Flickr-Faces-High Quality

FFHQ512:

Flickr-Faces-HQ (FFHQ) Dataset with resolution 512 × 512

FID :

Frechet inception distance

b :

Pixel blitting

bg :

Pixel blitting, geometric transformation

bgc :

Pixel blitting, geometric transformation, color transformation

bgcf :

Pixel blitting, geometric transformation, color transformation, image-space filtering

bgcfn :

Pixel blitting, geometric transformation, color transformation, image-space filtering, additive noise

bgcfnc :

Pixel blitting, geometric transformation, color transformation, image-space filtering, additive noise, cutout

ADAM :

Adaptive moment estimation

MCC :

Matthews correlation coefficient

ROC :

Receiver operating characteristic curve

AUC :

Area under the ROC curve

UK :

United Kingdom

References

  1. Bourne RR, Steinmetz JD, Saylan M et al (2021) Causes of blindness and vision impairment in 2020 and trends over 30 years, and prevalence of avoidable blindness in relation to VISION 2020: the right to sight: an analysis for the Global Burden of Disease Study. Lancet Glob Health 9:e144–e160. https://doi.org/10.1016/S2214-109X(20)30489-7

    Article  Google Scholar 

  2. Rauch R, Weingessel B, Maca SM, Vecsei-Marlovits PV (2012) Time to first treatment: the significance of early treatment of exudative age-related macular degeneration. Retina 32:1260–1264. https://doi.org/10.1097/IAE.0b013e3182018df6

    Article  PubMed  Google Scholar 

  3. Ho AC, Albini TA, Brown DM et al (2017) The potential importance of detection of neovascular age-related macular degeneration when visual acuity is relatively good. JAMA Ophthalmology 135:268–273. https://doi.org/10.1001/jamaophthalmol.2016.5314

    Article  PubMed  Google Scholar 

  4. Good WV, Early Treatment for Retinopathy of Prematurity Cooperative Group (2004) Final results of the early treatment for retinopathy of prematurity (ETROP) randomized trial. Trans Am Ophthalmol Soc 102:233–248

    Google Scholar 

  5. Early Treatment For Retinopathy Of Prematurity Cooperative Group (2003) Revised indications for the treatment of retinopathy of prematurity: results of the early treatment for retinopathy of prematurity randomized trial. Arch Ophthalmol 121:1684–1694. https://doi.org/10.1001/archopht.121.12.1684

    Article  Google Scholar 

  6. Neely DC, Bray KJ, Huisingh CE et al (2017) Prevalence of undiagnosed age-related macular degeneration in primary eye care. JAMA Ophthalmol 135:570–575. https://doi.org/10.1001/jamaophthalmol.2017.0830

    Article  PubMed  PubMed Central  Google Scholar 

  7. Statham MO, Sharma A, Pane AR (2008) Misdiagnosis of acute eye diseases by primary health care providers: incidence and implications. Med J Aust 189:402–404. https://doi.org/10.5694/j.1326-5377.2008.tb02091.x

    Article  PubMed  Google Scholar 

  8. Yip H, Crock C, Chan E (2020) Diagnostic error in an ophthalmic emergency department. Diagnosis (Berl) 7:129–131. https://doi.org/10.1515/dx-2019-0047

    Article  PubMed  Google Scholar 

  9. Gelston CD, Patnaik JL (2019) Ophthalmology training and competency levels in care of patients with ophthalmic complaints in United States internal medicine, emergency medicine and family medicine residents. J Educ Eval Health Prof 16:25. https://doi.org/10.3352/jeehp.2019.16.25

    Article  PubMed  PubMed Central  Google Scholar 

  10. Alotaibi AK, Alsalim A, Alruwaili F et al (2019) Burnout during ophthalmology residency training: a national survey in Saudi Arabia. Saudi J Ophthalmol 33:130–134

    Article  PubMed  PubMed Central  Google Scholar 

  11. Cheung R, Yu B, Iordanous Y, Malvankar-Mehta MS (2021) The prevalence of occupational burnout among ophthalmologists: a systematic review and meta-analysis. Psychol Rep 124:2139–2154. https://doi.org/10.1177/0033294120954135

    Article  PubMed  Google Scholar 

  12. Rosenblatt TR, Vail D, Saroj N et al (2021) Increasing incidence and prevalence of common retinal diseases in retina practices across the United States. Ophthalmic Surg Lasers Imaging Retina 52:29–36. https://doi.org/10.3928/23258160-20201223-06

    Article  PubMed  Google Scholar 

  13. Li JQ, Welchowski T, Schmid M et al (2020) Prevalence and incidence of age-related macular degeneration in Europe: a systematic review and meta-analysis. Br J Ophthalmol 104:1077–1084. https://doi.org/10.1136/bjophthalmol-2019-314422

    Article  PubMed  Google Scholar 

  14. Ansah JP, Koh V, de Korne DF et al (2018) Projection of eye disease burden in Singapore. Ann Acad Med Singap 47:13–28

    Article  PubMed  Google Scholar 

  15. Lee R, Wong TY, Sabanayagam C (2015) Epidemiology of diabetic retinopathy, diabetic macular edema and related vision loss. Eye Vis (Lond) 2:17. https://doi.org/10.1186/s40662-015-0026-2

    Article  PubMed  Google Scholar 

  16. Jonas JB, Cheung CMG, Panda-Jonas S (2017) Updates on the epidemiology of age-related macular degeneration. Asia Pac J Ophthalmol (Phila) 6:493–497. https://doi.org/10.22608/APO.2017251

  17. Medical Advisory Secretariat (2009) Optical coherence tomography for age-related macular degeneration and diabetic macular edema: an evidence-based analysis. Ont Health Technol Assess Ser 9:1–22

    PubMed Central  Google Scholar 

  18. Fujimoto J, Swanson E (2016) The development, commercialization, and impact of optical coherence tomography. Invest Ophthalmol Vis Sci 57:OCT1-OCT13. https://doi.org/10.1167/iovs.16-19963

  19. Arevalo JF, Lasave AF, Arias JD et al (2013) Clinical applications of optical coherence tomography in the posterior pole: the 2011 José Manuel Espino Lecture - Part I. Clin Ophthalmol 7:2165–2179. https://doi.org/10.2147/OPTH.S51098

    Article  PubMed  PubMed Central  Google Scholar 

  20. Lee CS, Baughman DM, Lee AY (2017) Deep learning is effective for the classification of OCT images of normal versus age-related macular degeneration. Ophthalmol Retina 1:322–327. https://doi.org/10.1016/j.oret.2016.12.009

    Article  PubMed  PubMed Central  Google Scholar 

  21. Li F, Chen H, Liu Z et al (2019) Fully automated detection of retinal disorders by image-based deep learning. Graefes Arch Clin Exp Ophthalmol 257:495–505. https://doi.org/10.1007/s00417-018-04224-8

    Article  PubMed  Google Scholar 

  22. Motozawa N, An G, Takagi S et al (2019) Optical coherence tomography-based deep-learning models for classifying normal and age-related macular degeneration and exudative and non-exudative age-related macular degeneration changes. Ophthalmol Ther 8:527–539. https://doi.org/10.1007/s40123-019-00207-y

    Article  PubMed  PubMed Central  Google Scholar 

  23. Schlegl T, Waldstein SM, Bogunovic H et al (2018) Fully automated detection and quantification of macular fluid in OCT using deep learning. Ophthalmology 125:549–558. https://doi.org/10.1016/j.ophtha.2017.10.031

    Article  PubMed  Google Scholar 

  24. Wu Q, Zhang B, Hu Y et al (2021) Detection of morphologic patterns of diabetic macular edema using a deep learning approach based on optical coherence tomography images. Retina 41:1110–1117. https://doi.org/10.1097/IAE.0000000000002992

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Virgili G, Menchini F, Murro V, etal., (2011) Optical coherence tomography (OCT) for detection of macular oedema in patients with diabetic retinopathy. Cochrane Database Syst Rev 7:CD008081. https://doi.org/10.1002/14651858.CD008081.pub2

  26. Lu W, Tong Y, Yu Y et al (2018) Deep learning-based automated classification of multi- categorical abnormalities from optical coherence tomography images. Transl Vis Sci Technol 7:41. https://doi.org/10.1167/tvst.7.6.41

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Yoon J, Han J, Park JI et al (2020) Optical coherence tomography-based deep-learning model for detecting central serous chorioretinopathy. Sci Rep 10:18852. https://doi.org/10.1038/s41598-020-75816-w

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  28. Buda M, Maki A, Mazurowski MA (2018) A systematic study of the class imbalance problem in convolutional neural networks. Neural Netw 106:249–259. https://doi.org/10.1016/j.neunet.2018.07.011

    Article  PubMed  Google Scholar 

  29. Krawczyk B (2016) Learning from imbalanced data: open challenges and future directions. Prog Artif Intell 5:221–232. https://doi.org/10.1007/s13748-016-0094-0

    Article  Google Scholar 

  30. He H, Garcia EA (2009) Learning from imbalanced data. IEEE Trans Knowl Data Eng 21(9):1263–1284. https://doi.org/10.1109/TKDE.2008.239

    Article  Google Scholar 

  31. Kuwayama S, Ayatsuka Y, Yanagisono D et al (2019) Automated detection of macular diseases by optical coherence tomography and artificial intelligence machine learning of optical coherence tomography images. J Ophthalmol 2019:6319581. https://doi.org/10.1155/2019/6319581

    Article  PubMed  PubMed Central  Google Scholar 

  32. Yoo TK, Choi JY, Seo JG et al (2019) The possibility of the combination of OCT and fundus images for improving the diagnostic accuracy of deep learning for age-related macular degeneration: a preliminary experiment. Med Biol Eng Comput 57:677–687. https://doi.org/10.1007/s11517-018-1915-z

    Article  PubMed  Google Scholar 

  33. Thakoor KA, Li X, Tsamis E et al (2021) Strategies to improve convolutional neural network generalizability and reference standards for glaucoma detection from OCT scans. Transl Vis Sci Technol 10:16. https://doi.org/10.1167/tvst.10.4.16

    Article  PubMed  PubMed Central  Google Scholar 

  34. Choi KJ, Choi JE, Roh HC et al (2021) Deep learning models for screening of high myopia using optical coherence tomography. Sci Rep 11:21663. https://doi.org/10.1038/s41598-021-00622-x

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Elgendi M, Nasir MU, Tang Q et al (2021) The effectiveness of image augmentation in deep learning networks for detecting COVID-19: a geometric transformation perspective. Front Med (Lausanne) 8:629134. https://doi.org/10.3389/fmed.2021.629134

    Article  PubMed  Google Scholar 

  36. Zheng R, Liu L, Zhang S et al (2018) Detection of exudates in fundus photographs with imbalanced learning using conditional generative adversarial network. Biomed Opt Express 9:4863–4878. https://doi.org/10.1364/BOE.9.004863

    Article  PubMed  PubMed Central  Google Scholar 

  37. Guan S, Loew M (2019) Breast cancer detection using synthetic mammograms from generative adversarial networks in convolutional neural networks. J Med Imaging (Bellingham) 6:031411. https://doi.org/10.1117/1.JMI.6.3.031411

    Article  PubMed  Google Scholar 

  38. Xiao Y, Wu J, Lin Z (2021) Cancer diagnosis using generative adversarial networks based on deep learning from imbalanced data. Comput Biol Med 135:104540. https://doi.org/10.1016/j.compbiomed.2021.104540

    Article  PubMed  Google Scholar 

  39. Zheng C, Xie X, Zhou K et al (2020) Assessment of generative adversarial networks model for synthetic optical coherence tomography images of retinal disorders. Transl Vis Sci Technol 9:29. https://doi.org/10.1167/tvst.9.2.29

    Article  PubMed  PubMed Central  Google Scholar 

  40. Karras T, Aittala M, Hellsten J, etal., (2020) Training generative adversarial networks with limited data. Adv Neural Inf Process Syst 33:1–15. https://doi.org/10.48550/arXiv.2006.06676

  41. Kermany DS, Goldbaum M, Cai W et al (2018) Identifying medical diagnoses and treatable diseases by image-based deep learning. Cell 172:1122-1131.e9. https://doi.org/10.1016/j.cell.2018.02.010

    Article  CAS  PubMed  Google Scholar 

  42. The MathWorks, Inc (2022) Pretrained Deep Neural Networks - MATLAB & Simulink. Help Center https://www.mathworks.com/help/deeplearning/ug/pretrained-convolutional-neural-networks.html Accessed 2 May 2022

  43. Tsuji T, Hirose Y, Fujimori K et al (2020) Classification of optical coherence tomography images using a capsule network. BMC Ophthalmol 20:114. https://doi.org/10.1186/s12886-020-01382-4

    Article  PubMed  PubMed Central  Google Scholar 

  44. Chen YM, Huang WT, Ho WH, Tsai JT (2021) Classification of age-related macular degeneration using convolutional-neural-network-based transfer learning. BMC Bioinformatics 22:99. https://doi.org/10.1186/s12859-021-04001-1

    Article  PubMed  PubMed Central  Google Scholar 

  45. Ho WH, Huang TH, Yang PY et al (2021) Artificial intelligence classification model for macular degeneration images: a robust optimization framework for residual neural networks. BMC Bioinformatics 22:148. https://doi.org/10.1186/s12859-021-04085-9

    Article  PubMed  PubMed Central  Google Scholar 

  46. Sunija A, Kar S, Gayathri S et al (2021) Octnet: a lightweight cnn for retinal disease classification from optical coherence tomography images. Comput Methods Programs Biomed 200:105877. https://doi.org/10.1016/j.cmpb.2020.105877

    Article  Google Scholar 

  47. Silvestri G, Williams MA, McAuley C et al (2012) DRUSEN prevalence and pigmentary changes in Caucasians aged 18–54 years. Eye (Lond) 26:1357–1362. https://doi.org/10.1038/eye.2012.165

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

Not applicable

Funding

The design and part-writing costs of the study are funded by the Ministry of Science and Technology, Taiwan (MOST 108–3111-Y-016–012) and costs of collection, analysis and interpretation of data and part-writing are funded by the Ministry of National Defense-Medical Affairs Bureau (MND-MAB-D-111097). Publication costs are funded by the authors.

Author information

Authors and Affiliations

  1. School of Medicine, National Defense Medical Center, Taipei, Taiwan

    Ling-Chun Sun

  2. Department of Ophthalmology, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan

    Shu-I. Pao

  3. Department of Ophthalmology, Song-Shan Branch of Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan

    Ke-Hao Huang

  4. Graduate Institute of Life Sciences, National Defense Medical Center, Taipei, Taiwan

    Chih-Yuan Wei

  5. Medical Informatics Office, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan

    Ke-Feng Lin

  6. School of Public Health, National Defense Medical Center, Taipei, Taiwan

    Ke-Feng Lin

  7. Department of Biomedical Engineering, National Defense Medical Center, No.161, Sec.6, Minchiuan E. Rd., Neihu Dist, Taipei, 11490, Taiwan

    Ping-Nan Chen

Authors
  1. Ling-Chun Sun
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Shu-I. Pao
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Ke-Hao Huang
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Chih-Yuan Wei
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Ke-Feng Lin
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Ping-Nan Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Drs. Chen and K.-F. Lin had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: Sun, K.-F. Lin, Chen. Acquisition, analysis, or interpretation of data: Sun, Pao, Huang, Wei, K.-F. Lin, Chen. Drafting of the manuscript: Sun, Chen. Critical revision of the manuscript for important intellectual content: Sun, Pao, Huang, Wei, K.-F. Lin, Chen. Statistical analysis: Sun, Wei, K.-F. Lin, Chen. Obtained funding: Chen. Administrative, technical, or material support: Supervision: K.-F. Lin, Chen.

Corresponding author

Correspondence to Ping-Nan Chen.

Ethics declarations

Ethics approval and consent to participate

The experimental protocol was approved by the Tri-Service General Hospital (TSGH) human ethics committee under registration number IRB: 1–108-05–082. Images were retrospectively obtained from the Department of Ophthalmology and the Department of Endocrinology and Metabolism at TSGH and were anonymized; thus, informed consent was not required.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Sun, LC., Pao, SI., Huang, KH. et al. Generative adversarial network-based deep learning approach in classification of retinal conditions with optical coherence tomography images. Graefes Arch Clin Exp Ophthalmol 261, 1399–1412 (2023). https://doi.org/10.1007/s00417-022-05919-9

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00417-022-05919-9

Keywords

Search

Navigation