Cascade Dual-branch Deep Neural Networks for Retinal Layer and fluid Segmentation of Optical Coherence Tomography Incorporating Relative Positional Map

Da Ma, Donghuan Lu, Morgan Heisler, Setareh Dabiri, Sieun Lee, Gavin Weiguang Ding, Marinko V. Sarunic, Mirza Faisal Beg
; Proceedings of the Third Conference on Medical Imaging with Deep Learning, PMLR 121:493-502, 2020.

Abstract

Optical coherence tomography (OCT) is a non-invasive imaging technology that can provide micrometer-resolution cross-sectional images of the inner structures of the eye. It is widely used for the diagnosis of ophthalmic diseases with retinal alteration such as layer deformation and fluid accumulation. In this paper, a novel framework was proposed to segment retinal layers with fluid presence. The main contribution of this study is two folds: 1) we developed a cascaded network framework to incorporate the prior structural knowledge; 2) we proposed a novel two-path deep neural network which includes both the U-Net architecture as well as the original implementation of the fully convolutional network, concatenated into a final multi-level dilated layer to achieve accurate simultaneous layer and fluid segmentation. Cross validation experiments proved that the proposed network has superior performance comparing with the state-of-the-art methods by up to $3%$, and incorporating the relative positional map structural prior information could further improve the performance (up to $1%$) regardless of the network.

Cite this Paper


BibTeX
@InProceedings{pmlr-v121-ma20a, title = {Cascade Dual-branch Deep Neural Networks for Retinal Layer and fluid Segmentation of Optical Coherence Tomography Incorporating Relative Positional Map}, author = {Ma, Da and Lu, Donghuan and Heisler, Morgan and Dabiri, Setareh and Lee, Sieun and Ding, Gavin Weiguang and Sarunic, Marinko V. and Beg, Mirza Faisal}, pages = {493--502}, year = {2020}, editor = {Tal Arbel and Ismail Ben Ayed and Marleen de Bruijne and Maxime Descoteaux and Herve Lombaert and Christopher Pal}, volume = {121}, series = {Proceedings of Machine Learning Research}, address = {Montreal, QC, Canada}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v121/ma20a/ma20a.pdf}, url = {http://proceedings.mlr.press/v121/ma20a.html}, abstract = {Optical coherence tomography (OCT) is a non-invasive imaging technology that can provide micrometer-resolution cross-sectional images of the inner structures of the eye. It is widely used for the diagnosis of ophthalmic diseases with retinal alteration such as layer deformation and fluid accumulation. In this paper, a novel framework was proposed to segment retinal layers with fluid presence. The main contribution of this study is two folds: 1) we developed a cascaded network framework to incorporate the prior structural knowledge; 2) we proposed a novel two-path deep neural network which includes both the U-Net architecture as well as the original implementation of the fully convolutional network, concatenated into a final multi-level dilated layer to achieve accurate simultaneous layer and fluid segmentation. Cross validation experiments proved that the proposed network has superior performance comparing with the state-of-the-art methods by up to $3%$, and incorporating the relative positional map structural prior information could further improve the performance (up to $1%$) regardless of the network.} }
Endnote
%0 Conference Paper %T Cascade Dual-branch Deep Neural Networks for Retinal Layer and fluid Segmentation of Optical Coherence Tomography Incorporating Relative Positional Map %A Da Ma %A Donghuan Lu %A Morgan Heisler %A Setareh Dabiri %A Sieun Lee %A Gavin Weiguang Ding %A Marinko V. Sarunic %A Mirza Faisal Beg %B Proceedings of the Third Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2020 %E Tal Arbel %E Ismail Ben Ayed %E Marleen de Bruijne %E Maxime Descoteaux %E Herve Lombaert %E Christopher Pal %F pmlr-v121-ma20a %I PMLR %J Proceedings of Machine Learning Research %P 493--502 %U http://proceedings.mlr.press %V 121 %W PMLR %X Optical coherence tomography (OCT) is a non-invasive imaging technology that can provide micrometer-resolution cross-sectional images of the inner structures of the eye. It is widely used for the diagnosis of ophthalmic diseases with retinal alteration such as layer deformation and fluid accumulation. In this paper, a novel framework was proposed to segment retinal layers with fluid presence. The main contribution of this study is two folds: 1) we developed a cascaded network framework to incorporate the prior structural knowledge; 2) we proposed a novel two-path deep neural network which includes both the U-Net architecture as well as the original implementation of the fully convolutional network, concatenated into a final multi-level dilated layer to achieve accurate simultaneous layer and fluid segmentation. Cross validation experiments proved that the proposed network has superior performance comparing with the state-of-the-art methods by up to $3%$, and incorporating the relative positional map structural prior information could further improve the performance (up to $1%$) regardless of the network.
APA
Ma, D., Lu, D., Heisler, M., Dabiri, S., Lee, S., Ding, G.W., Sarunic, M.V. & Beg, M.F.. (2020). Cascade Dual-branch Deep Neural Networks for Retinal Layer and fluid Segmentation of Optical Coherence Tomography Incorporating Relative Positional Map. Proceedings of the Third Conference on Medical Imaging with Deep Learning, in PMLR 121:493-502

Related Material