Abstract
Choroidal neovascularization (CNV) is a typical clinical manifestation of age-related macular degeneration (AMD) and an important factor leading to blindness in AMD patients. Automated CNV lesion segmentation based on SD-OCT images has important research significance for clinical diagnosis. We propose a data-dependence dual path network (D3PNet) for CNV segmentation by designing an expansive path, a guidance path and a novel feature fusion strategy. In the expansive path, the data-dependent upsampling method and the proposed upsampling strategy would preserve more detail information and make the obtained features more diversified. In the guidance path, a deformable module is proposed to generate the saliency maps and lead the model focusing on the contours. Finally, we design a novel feature fusion method by regarding the saliency maps as the attention mechanism of hierarchical features to amplify the beneficial features and suppress the useless ones. Experimental results demonstrate the superior performances and reliabilities of the proposed network comparing with state-of-the-art methods.
This work was supported by National Natural Science Foundation of China under Grant No. 62072241, and in part by Natural Science Foundation of Jiangsu Province under Grants No. BK20180069, and in part by Six talent peaks project in Jiangsu Province under Grant No. SWYY-056.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Lim, L.S., Mitchell, P., Seddon, J.M., Holz, F.G., Wong, T.Y.: Age-related macular degeneration. Lancet 379(9827), 1728–1738 (2012)
Grossniklaus, H.E., Green, W.R.: Choroidal neovascularization. Am. J. Ophthalmol. 137(3), 496–503 (2004)
Framme, C., Panagakis, G., Birngruber, R.: Effects on choroidal neovascularization after anti-VEGF upload using intravitreal ranibizumab, as determined by spectral domain-optical coherence tomography. Invest. Ophthalmol. Vis. Sci. 51(3), 1671–1676 (2010)
Cavallerano, A.A.: Ophthalmic fluorescein angiography. Optom. Clin. Official Publ. Prentice Soc. 5(1), 1–23 (1996)
Bruyère, E., et al.: Spectral-domain optical coherence tomography of subretinal hyperreflective exudation in myopic choroidal neovascularization. Am. J. Ophthalmol. 160(4), 749–758 (2015)
Li, Y., Niu, S., Ji, Z., Fan, W., Yuan, S., Chen, Q.: Automated choroidal neovascularization detection for time series SD-OCT images. In: Frangi, A.F., Schnabel, J.A., Davatzikos, C., Alberola-López, C., Fichtinger, G. (eds.) MICCAI 2018. LNCS, vol. 11071, pp. 381–388. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-00934-2_43
Lee, N., Laine, A.F., Theodore Smith, R.: Bayesian transductive Markov random fields for interactive segmentation in retinal disorders. In: Dössel, O., Schlegel, W.C. (eds.) World Congress on Medical Physics and Biomedical Engineering, 7–12 September 2009, Munich, Germany, pp. 227–230. Springer, Heidelberg (2009). https://doi.org/10.1007/978-3-642-03891-4_61
Fahmy, A.S., Abdelmoula, W.M., Mahfouz, A.E., Shah, S.M.: Segmentation of choroidal neovascularization lesions in fluorescein angiograms using parametric modeling of the intensity variation. In: 2011 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, pp. 665–668. IEEE (2011)
Tsai, C.-L., Yang, Y.-L., Chen, S.-J., Chan, C.-H., Lin, W.-Y.: Automatic characterization and segmentation of classic choroidal neovascularization using adaboost for supervised learning. In: IEEE Nuclear Science Symposuim & Medical Imaging Conference, pp. 3610–3612. IEEE (2010)
Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3431–3440 (2015)
Ronneberger, O., Fischer, P., Brox, T.: U-net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-24574-4_28
Chen, L.-C., Papandreou, G., Schroff, F., Adam, H.: Rethinking atrous convolution for semantic image segmentation. arXiv preprint arXiv:1706.05587 (2017)
Lin, G., Milan, A., Shen, C., Reid, I.: RefineNet: multi-path refinement networks for high-resolution semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1925–1934 (2017)
Zhang, Y., et al.: MPB-CNN: a multi-scale parallel branch CNN for choroidal neovascularization segmentation in SD-OCT images. OSA Continuum 2(3), 1011–1027 (2019)
Wang, J., et al.: Automated diagnosis and segmentation of choroidal neovascularization in OCT angiography using deep learning. Biomed. Opt. Express 11(2), 927–944 (2020)
Xi, X., et al.: Automated segmentation of choroidal neovascularization in optical coherence tomography images using multi-scale convolutional neural networks with structure prior. Multimedia Syst. 25(2), 95–102 (2019)
Tian, Z., He, T., Shen, C., Yan, Y.: Decoders matter for semantic segmentation: Data-dependent decoding enables flexible feature aggregation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 3126–3135 (2019)
Dai, J., et al.: Deformable convolutional networks. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 764–773 (2017)
Oktay, O., et al.: Attention U-Net: learning where to look for the pancreas. arXiv preprint arXiv:1804.03999 (2018)
Fu, J.: Dual attention network for scene segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 3146–3154 (2019)
Zaiwang, G., et al.: CE-Net: context encoder network for 2D medical image segmentation. IEEE Trans. Med. Imaging 38(10), 2281–2292 (2019)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Ke, J., Ji, Z., Chen, Q., Fan, W., Yuan, S. (2021). Data-Dependence Dual Path Network for Choroidal Neovascularization Segmentation in SD-OCT Images. In: Peng, Y., Hu, SM., Gabbouj, M., Zhou, K., Elad, M., Xu, K. (eds) Image and Graphics. ICIG 2021. Lecture Notes in Computer Science(), vol 12889. Springer, Cham. https://doi.org/10.1007/978-3-030-87358-5_43
Download citation
DOI: https://doi.org/10.1007/978-3-030-87358-5_43
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-87357-8
Online ISBN: 978-3-030-87358-5
eBook Packages: Computer ScienceComputer Science (R0)