Advertisement

Comparison between support vector machine and deep learning, machine-learning technologies for detecting epiretinal membrane using 3D-OCT

  • Tomoaki Sonobe
  • Hitoshi Tabuchi
  • Hideharu Ohsugi
  • Hiroki Masumoto
  • Naohumi Ishitobi
  • Shoji Morita
  • Hiroki Enno
  • Daisuke Nagasato
Original Paper

Abstract

Purpose

In this study, we compared deep learning (DL) with support vector machine (SVM), both of which use three-dimensional optical coherence tomography (3D-OCT) images for detecting epiretinal membrane (ERM).

Methods

In total, 529 3D-OCT images from the Tsukazaki hospital ophthalmology database (184 non-ERM subjects and 205 ERM patients) were assessed; 80% of the images were divided for training, and 20% for test as follows: 423 training (non-ERM 245, ERM 178) and 106 test (non-ERM 59, ERM 47) images. Using the 423 training images, a model was created with deep convolutional neural network and SVM, and the test data were evaluated.

Results

The DL model’s sensitivity was 97.6% [95% confidence interval (CI), 87.7–99.9%] and specificity was 98.0% (95% CI, 89.7–99.9%), and the area under the curve (AUC) was 0.993 (95% CI, 0.993–0.994). In contrast, the SVM model’s sensitivity was 97.6% (95% CI, 87.7–99.9%), specificity was 94.2% (95% CI, 84.0–98.7%), and AUC was 0.988 (95% CI, 0.987–0.988).

Conclusion

DL model is better than SVM model in detecting ERM by using 3D-OCT images.

Keywords

Deep learning Support vector machine Epiretinal membrane Optical coherence tomography 

Notes

Acknowledgements

I would like to thank M. Miki for creating the figures.

Compliance with ethical standards

Conflict of interest

All authors certify that hey have no affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements) or non-financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee (Tsukazaki hospital) and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from all individual participants included in the study.

References

  1. 1.
    Okamoto F, Sugiura Y, Okamoto Y, Hiraoka T, Oshika T (2015) Inner nuclear layer thickness as a prognostic factor for metamorphopsia after epiretinal membrane surgery. Retina 35(10):1–8CrossRefGoogle Scholar
  2. 2.
    Okamoto F, Okamoto Y, Hiraoka T, Oshika T (2009) Effect of vitrectomy for epiretinal membrane on visual function and vision-related quality of life. Am J Ophthalmol 147(5):869–874CrossRefPubMedGoogle Scholar
  3. 3.
    Gomes LN, Corcostegui I, Fine HF, Chang S (2009) Subfoveal pigment changes in patients with longstanding epiretinal membranes. Am J Ophthalmol 147(5):865–868CrossRefPubMedGoogle Scholar
  4. 4.
    Smiddy WE, Maguire AM, Green WR, Michels RG, de la Cruz Z, Enger C, Jaeger M, Rice TA (1989) Idiopathic epiretinal membranes. Ultra-structural characteristics and clinicopathologic correlation. Ophthalmology 96(6):811–821CrossRefPubMedGoogle Scholar
  5. 5.
    Inoue M, Kadonosono K (2014) Macular diseases: epiretinal membrane. Dev Ophthalmol 54:159–163CrossRefPubMedGoogle Scholar
  6. 6.
    Mitchell P, Smith W, Chey T, Wang JJ, Chang A (1997) Prevalence and association of epiretinal membranes. The Blue Mountains eye study, Australia. Ophthalmology 104(6):1033–1040CrossRefPubMedGoogle Scholar
  7. 7.
    Bouwens MD, Meurs JC (2003) Sine Amsler Charts: a new method for the follow-up of metamorphopsia in patients undergoing macular pucker surgery. Grafes Arch Clin Exp Ophthalmol 241(2):89–93CrossRefGoogle Scholar
  8. 8.
    Kinoshita T, Imaizumi H, Okushiba U, Miyamoto H, Ogino T, Mitamura Y (2012) Time course of changes in metamorphopsia, visual acuity, and OCT parameters after successful epiretinal membrane surgery. Invest Ophthalmol Vis Sci 53(7):3592–3597CrossRefPubMedGoogle Scholar
  9. 9.
    Kim JH, Kang SW, Kong MG, Ha HS (2013) Assessment of retinal layers and visual rehabilitation after epiretinal membrane removal. Graefes Arch Clin Exp Ophthalmol 251(4):1055–1064CrossRefPubMedGoogle Scholar
  10. 10.
    Huang D, Swanson EA, Lin CP, Schuman JS, Stinson WG, Chang W, Hee MR, Flotte T, Gregory K, Puliafito CA, Fujimoto JG (1991) Optical coherence tomography. Science 254(5035):1178–1181CrossRefPubMedPubMedCentralGoogle Scholar
  11. 11.
    Drexler W, Liu M, Kumar A, Kamali T, Unterhuber A, Leitgeb RA (2014) Optical coherence tomography today: speed, contrast, and multimodality. J Biomed Opt 19(7):071412CrossRefPubMedGoogle Scholar
  12. 12.
    Legarreta JE, Gregori G, Knighton RW, Punjabi OS, Lalwani GA, Puliafito CA (2008) Three-dimensional spectral-domain optical coherence tomography images of the retina in the presence of epiretinal membranes. Am J Ophthamol 145(6):1023–1030CrossRefGoogle Scholar
  13. 13.
    Lingyun G, Mingquan Y, Changrong W (2017) Cancer classification based on support vector machine optimized by particle swarm optimization and artificial bee colony. Molecules 22(12):pii: E2086CrossRefGoogle Scholar
  14. 14.
    LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521:436–444CrossRefGoogle Scholar
  15. 15.
    Liu S, Liu S, Cai W, Che H, Pujol S, Kikinis R, Feng D, Fulham MJ (2015) Multimodal neuroimaging feature learning for multiclass diagnosis of Alzheimer’s disease. IEEE Trans Biomed Eng 62:1132–1140CrossRefPubMedGoogle Scholar
  16. 16.
    Litjens G, Sánchez CI, Timofeeva N, Hermsen M, Nagtegaal I, Kovacs I, de Hulsbergen-van KC, Bult P, van Ginneken B, van der Laak J (2016) Deep learning as a tool for increased accuracy and efficiency of histopathological diagnosis. Sci Rep 6:26286CrossRefPubMedPubMedCentralGoogle Scholar
  17. 17.
    Gulshan V, Peng L, Coram M, Stumpe MC, Wu D, Narayanaswamy A, Venugopalan S, Widner K, Madams T, Cuadros J, Kim R, Raman R, Nelson PC, Mega JL, Webster DR (2016) Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA 316:2402–2410CrossRefPubMedGoogle Scholar
  18. 18.
    Pinaya WH, Gadelha A, Doyle OM, Noto C, Zugman A, Cordeiro Q, Jackowski AP, Bressan RA, Sato JR (2016) Using deep belief network modelling to characterize differences in brain morphometry in schizophrenia. Sci Rep 6:38897CrossRefPubMedPubMedCentralGoogle Scholar
  19. 19.
    Krizhevsky A, Sutskever I, Hinton GE (2012) ImageNet classification with deep convolutional neural networks. In: Proceedings of NIPS, pp 1097–1105Google Scholar
  20. 20.
    Gargeya R, Leng T (2017) Automated identification of diabetic retinopathy using deep learning. Ophthalmology 124(7):962–969CrossRefPubMedGoogle Scholar
  21. 21.
    Glorot X, Bordes A, Bengio Y (2011) Deep sparse rectifier neural networks. In: Proceedings of the fourteenth international conference on artificial intelligence and statistics, pp 315–323Google Scholar
  22. 22.
    Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. JMLR 15(1):1929–1958Google Scholar
  23. 23.
    Qian N (1999) On the momentum term in gradient descent learning algorithms. Neural Netw 12(1):145–151CrossRefPubMedGoogle Scholar
  24. 24.
    Nesterov Y (1983) A method for unconstrained convex minimization problem with the rate of convergence O (1/k^ 2). Doklady AN USSR 269:543–547Google Scholar
  25. 25.
    Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, Vanderplas J, Passos A, Cournapeau D, Brucher M, Perrot M, Duchesnay E (2011) Scikit-learn: machine learning in Python. J Mach Learn Res 12:2825–2830Google Scholar
  26. 26.
    Schisterman EF, Faraggi D, Reiser B, Hu J (2008) Youden Index and the optimal threshold for markers with mass at zero. Stat Med 27(2):297–315CrossRefPubMedPubMedCentralGoogle Scholar
  27. 27.
    Deng J, Dong W, Socher R, Li L-J, Li K, Fei-Fei L (2009) ImageNet: a large-scale hierarchical image database. In: IEEE conference on computer vision and pattern recognition, CVPR 2009Google Scholar
  28. 28.
    Alsaih K, Lemaitre G, Rastgoo M (2017) Machine learning techniques for diabetic macular edema (DME) classification on SD-OCT image. BioMed Eng Online 16:68CrossRefPubMedPubMedCentralGoogle Scholar
  29. 29.
  30. 30.
    Ohsugi H, Tabuchi H, Enno H, Ishitobi N (2017) Accuracy of deep learning, a machine-learning technology, using ultra-wide-field fundus ophthalmoscopy for detecting rhegmatogenous retinal detachment. Sci Rep 7(1):9425.  https://doi.org/10.1038/s41598-017-09891-x CrossRefPubMedPubMedCentralGoogle Scholar
  31. 31.
    Asaoka R, Murata H, Iwase A, Araie M (2016) Detecting preperimetric glaucoma with standard automated perimetry using a deep learning classifier. Ophthalmology 123(9):1974–1980CrossRefPubMedGoogle Scholar
  32. 32.
    Kim JS, Chhablani J, Chan K, Cheng L, Kozak I, Hartmann K, Freeman WR (2012) Retinal adherence and fibrillary surface changes correlate with surgical difficulty of epiretinal membrane removal. Am J Ophthalmol 153(4):692–697CrossRefPubMedGoogle Scholar

Copyright information

© Springer Nature B.V. 2018

Authors and Affiliations

  1. 1.Department of OphthalmologyTsukazaki HospitalHimejiJapan
  2. 2.Research Group of Intelligent Cybernetics and Computer Science Graduate School of EngineeringUniversity of HyogoHimejiJapan
  3. 3.Rist IncTokyoJapan

Personalised recommendations