Retinal-Layer Segmentation Using Dilated Convolutions
Visualization and analysis of Spectral Domain Optical Coherence Tomography (SD-OCT) cross-sectional scans has gained a lot of importance in the diagnosis of several retinal abnormalities. Quantitative analytic techniques like retinal thickness and volumetric analysis are performed on cross-sectional images of the retina for early diagnosis and prognosis of retinal diseases. However, segmentation of retinal layers from OCT images is a complicated task on account of certain factors like speckle noise, low image contrast and low signal-to-noise ratio amongst many others. Owing to the importance of retinal layer segmentation in diagnosing ophthalmic diseases, manual segmentation techniques have been proposed and adopted in clinical practice. Nonetheless, manual segmentations suffer from erroneous boundary detection issues. This paper thus proposes a fully automated semantic segmentation technique that uses an encoder–decoder architecture to accurately segment the prominent retinal layers.
KeywordsRetinal layer segmentation Optical coherence tomography Dilated convolutions Deep learning Retina
This work was supported by the Science and Engineering Research Board (Department of Science and Technology, India) through project funding EMR/2016/002677.
The authors would like to thank Vision and Image Processing (VIP) Lab, Department of Biomedical Engineering, Duke University, Durham, NC, USA for providing DME dataset.
- 1.Anger, E.M., Unterhuber, A., Hermann, B., Sattmann, H., Schubert, C., Morgan, J.E., Cowey, A., Ahnelt, P.K., Drexler, W.: Ultrahigh resolution optical coherence tomography of the monkey fovea. identification of retinal sublayers by correlation with semithin histology sections. Exp. Eye Res. 78(6), 1117–1125 (2004)Google Scholar
- 6.Coscas, G.: Optical Coherence Tomography in Age-Related Macular Degeneration. Springer Science & Business Media (2009)Google Scholar
- 8.Fu, T., Liu, X., Liu, D., Yang, Z.: A deep convolutional feature based learning layer-specific edges method for segmenting oct image. In: Ninth International Conference on Digital Image Processing (ICDIP 2017). vol. 10420, p. 1042029. International Society for Optics and Photonics (2017)Google Scholar
- 9.Girish, G.N., Thakur, B., Chowdhury, S.R., Kothari, A.R., Rajan, J.: Segmentation of intra-retinal cysts from optical coherence tomography images using a fully convolutional neural network model. IEEE J. Biomed. Health Inform. (2018)Google Scholar
- 10.Huang, D., Swanson, E.A., Lin, C.P., Schuman, et al.: Optical coherence tomography. Science 254(5035), 1178–1181 (1991)Google Scholar
- 11.Ioffe, S., Szegedy, C.: Batch normalization: accelerating deep network training by reducing internal covariate shift (2015). arXiv preprint arXiv:1502.03167
- 12.Ramachandran, P., Zoph, B., Le, Q.V.: Searching for activation functions. CoRR (2017). arXiv preprint arXiv:1710.05941
- 13.Roy, A.G., Conjeti, S., Carlier, S.G., Dutta, P.K., Kastrati, A., Laine, A.F., Navab, N., Katouzian, A., Sheet, D.: Lumen segmentation in intravascular optical coherence tomography using backscattering tracked and initialized random walks. IEEE J. Biomed. Health Inform. 20(2), 606–614 (2016)CrossRefGoogle Scholar
- 14.Roy, A.G., Conjeti, S., Carlier, S.G., Houissa, K., König, A., Dutta, P.K., Laine, A.F., Navab, N., Katouzian, A., Sheet, D.: Multiscale distribution preserving autoencoders for plaque detection in intravascular optical coherence tomography. In: 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI). IEEE, pp. 1359–1362 (2016)Google Scholar
- 17.Yu, F., Koltun, V.: Multi-scale context aggregation by dilated convolutions (2015). arXiv preprint arXiv:1511.07122