Fully convolutional boundary regression for retina OCT segmentation

Yufan He, Aaron Carass, Yihao Liu, Bruno M. Jedynak, Sharon D. Solomon, Shiv Saidha, Peter A. Calabresi, Jerry L. Prince

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A major goal of analyzing retinal optical coherence tomography (OCT) images is retinal layer segmentation. Accurate automated algorithms for segmenting smooth continuous layer surfaces, with correct hierarchy (topology) are desired for monitoring disease progression. State-of-the-art methods use a trained classifier to label each pixel into background, layer, or surface pixels. The final step of extracting the desired smooth surfaces with correct topology are mostly performed by graph methods (e.g. shortest path, graph cut). However, manually building a graph with varying constraints by retinal region and pathology and solving the minimization with specialized algorithms will degrade the flexibility and time efficiency of the whole framework. In this paper, we directly model the distribution of surface positions using a deep network with a fully differentiable soft argmax to obtain smooth, continuous surfaces in a single feed forward operation. A special topology module is used in the deep network both in the training and testing stages to guarantee the surface topology. An extra deep network output branch is also used for predicting lesion and layers in a pixel-wise labeling scheme. The proposed method was evaluated on two publicly available data sets of healthy controls, subjects with multiple sclerosis, and diabetic macular edema; it achieves state-of-the art sub-pixel results.

Original languageEnglish (US)
Title of host publicationMedical Image Computing and Computer Assisted Intervention – MICCAI 2019 - 22nd International Conference, Proceedings
EditorsDinggang Shen, Pew-Thian Yap, Tianming Liu, Terry M. Peters, Ali Khan, Lawrence H. Staib, Caroline Essert, Sean Zhou
PublisherSpringer
Pages120-128
Number of pages9
ISBN (Print)9783030322380
DOIs
StatePublished - Jan 1 2019
Event22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019 - Shenzhen, China
Duration: Oct 13 2019Oct 17 2019

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11764 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019
CountryChina
CityShenzhen
Period10/13/1910/17/19

Fingerprint

Optical Coherence Tomography
Retina
Optical tomography
Segmentation
Regression
Topology
Pixels
Pixel
Labeling Scheme
Multiple Sclerosis
Graph Cuts
Sub-pixel
Smooth surface
Graph in graph theory
Feedforward
Progression
Shortest path
Pathology
Differentiable
Branch

Keywords

  • Deep learning segmentation
  • Retina OCT
  • Surface segmentation

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

He, Y., Carass, A., Liu, Y., Jedynak, B. M., Solomon, S. D., Saidha, S., ... Prince, J. L. (2019). Fully convolutional boundary regression for retina OCT segmentation. In D. Shen, P-T. Yap, T. Liu, T. M. Peters, A. Khan, L. H. Staib, C. Essert, ... S. Zhou (Eds.), Medical Image Computing and Computer Assisted Intervention – MICCAI 2019 - 22nd International Conference, Proceedings (pp. 120-128). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11764 LNCS). Springer. https://doi.org/10.1007/978-3-030-32239-7_14

Fully convolutional boundary regression for retina OCT segmentation. / He, Yufan; Carass, Aaron; Liu, Yihao; Jedynak, Bruno M.; Solomon, Sharon D.; Saidha, Shiv; Calabresi, Peter A.; Prince, Jerry L.

Medical Image Computing and Computer Assisted Intervention – MICCAI 2019 - 22nd International Conference, Proceedings. ed. / Dinggang Shen; Pew-Thian Yap; Tianming Liu; Terry M. Peters; Ali Khan; Lawrence H. Staib; Caroline Essert; Sean Zhou. Springer, 2019. p. 120-128 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11764 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

He, Y, Carass, A, Liu, Y, Jedynak, BM, Solomon, SD, Saidha, S, Calabresi, PA & Prince, JL 2019, Fully convolutional boundary regression for retina OCT segmentation. in D Shen, P-T Yap, T Liu, TM Peters, A Khan, LH Staib, C Essert & S Zhou (eds), Medical Image Computing and Computer Assisted Intervention – MICCAI 2019 - 22nd International Conference, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11764 LNCS, Springer, pp. 120-128, 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019, Shenzhen, China, 10/13/19. https://doi.org/10.1007/978-3-030-32239-7_14
He Y, Carass A, Liu Y, Jedynak BM, Solomon SD, Saidha S et al. Fully convolutional boundary regression for retina OCT segmentation. In Shen D, Yap P-T, Liu T, Peters TM, Khan A, Staib LH, Essert C, Zhou S, editors, Medical Image Computing and Computer Assisted Intervention – MICCAI 2019 - 22nd International Conference, Proceedings. Springer. 2019. p. 120-128. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-32239-7_14
He, Yufan ; Carass, Aaron ; Liu, Yihao ; Jedynak, Bruno M. ; Solomon, Sharon D. ; Saidha, Shiv ; Calabresi, Peter A. ; Prince, Jerry L. / Fully convolutional boundary regression for retina OCT segmentation. Medical Image Computing and Computer Assisted Intervention – MICCAI 2019 - 22nd International Conference, Proceedings. editor / Dinggang Shen ; Pew-Thian Yap ; Tianming Liu ; Terry M. Peters ; Ali Khan ; Lawrence H. Staib ; Caroline Essert ; Sean Zhou. Springer, 2019. pp. 120-128 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{75f80c5206a64a3384cef6e815fb15ad,
title = "Fully convolutional boundary regression for retina OCT segmentation",
abstract = "A major goal of analyzing retinal optical coherence tomography (OCT) images is retinal layer segmentation. Accurate automated algorithms for segmenting smooth continuous layer surfaces, with correct hierarchy (topology) are desired for monitoring disease progression. State-of-the-art methods use a trained classifier to label each pixel into background, layer, or surface pixels. The final step of extracting the desired smooth surfaces with correct topology are mostly performed by graph methods (e.g. shortest path, graph cut). However, manually building a graph with varying constraints by retinal region and pathology and solving the minimization with specialized algorithms will degrade the flexibility and time efficiency of the whole framework. In this paper, we directly model the distribution of surface positions using a deep network with a fully differentiable soft argmax to obtain smooth, continuous surfaces in a single feed forward operation. A special topology module is used in the deep network both in the training and testing stages to guarantee the surface topology. An extra deep network output branch is also used for predicting lesion and layers in a pixel-wise labeling scheme. The proposed method was evaluated on two publicly available data sets of healthy controls, subjects with multiple sclerosis, and diabetic macular edema; it achieves state-of-the art sub-pixel results.",
keywords = "Deep learning segmentation, Retina OCT, Surface segmentation",
author = "Yufan He and Aaron Carass and Yihao Liu and Jedynak, {Bruno M.} and Solomon, {Sharon D.} and Shiv Saidha and Calabresi, {Peter A.} and Prince, {Jerry L.}",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/978-3-030-32239-7_14",
language = "English (US)",
isbn = "9783030322380",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer",
pages = "120--128",
editor = "Dinggang Shen and Pew-Thian Yap and Tianming Liu and Peters, {Terry M.} and Ali Khan and Staib, {Lawrence H.} and Caroline Essert and Sean Zhou",
booktitle = "Medical Image Computing and Computer Assisted Intervention – MICCAI 2019 - 22nd International Conference, Proceedings",

}

TY - GEN

T1 - Fully convolutional boundary regression for retina OCT segmentation

AU - He, Yufan

AU - Carass, Aaron

AU - Liu, Yihao

AU - Jedynak, Bruno M.

AU - Solomon, Sharon D.

AU - Saidha, Shiv

AU - Calabresi, Peter A.

AU - Prince, Jerry L.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - A major goal of analyzing retinal optical coherence tomography (OCT) images is retinal layer segmentation. Accurate automated algorithms for segmenting smooth continuous layer surfaces, with correct hierarchy (topology) are desired for monitoring disease progression. State-of-the-art methods use a trained classifier to label each pixel into background, layer, or surface pixels. The final step of extracting the desired smooth surfaces with correct topology are mostly performed by graph methods (e.g. shortest path, graph cut). However, manually building a graph with varying constraints by retinal region and pathology and solving the minimization with specialized algorithms will degrade the flexibility and time efficiency of the whole framework. In this paper, we directly model the distribution of surface positions using a deep network with a fully differentiable soft argmax to obtain smooth, continuous surfaces in a single feed forward operation. A special topology module is used in the deep network both in the training and testing stages to guarantee the surface topology. An extra deep network output branch is also used for predicting lesion and layers in a pixel-wise labeling scheme. The proposed method was evaluated on two publicly available data sets of healthy controls, subjects with multiple sclerosis, and diabetic macular edema; it achieves state-of-the art sub-pixel results.

AB - A major goal of analyzing retinal optical coherence tomography (OCT) images is retinal layer segmentation. Accurate automated algorithms for segmenting smooth continuous layer surfaces, with correct hierarchy (topology) are desired for monitoring disease progression. State-of-the-art methods use a trained classifier to label each pixel into background, layer, or surface pixels. The final step of extracting the desired smooth surfaces with correct topology are mostly performed by graph methods (e.g. shortest path, graph cut). However, manually building a graph with varying constraints by retinal region and pathology and solving the minimization with specialized algorithms will degrade the flexibility and time efficiency of the whole framework. In this paper, we directly model the distribution of surface positions using a deep network with a fully differentiable soft argmax to obtain smooth, continuous surfaces in a single feed forward operation. A special topology module is used in the deep network both in the training and testing stages to guarantee the surface topology. An extra deep network output branch is also used for predicting lesion and layers in a pixel-wise labeling scheme. The proposed method was evaluated on two publicly available data sets of healthy controls, subjects with multiple sclerosis, and diabetic macular edema; it achieves state-of-the art sub-pixel results.

KW - Deep learning segmentation

KW - Retina OCT

KW - Surface segmentation

UR - http://www.scopus.com/inward/record.url?scp=85075639317&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075639317&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-32239-7_14

DO - 10.1007/978-3-030-32239-7_14

M3 - Conference contribution

C2 - 31853524

AN - SCOPUS:85075639317

SN - 9783030322380

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 120

EP - 128

BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2019 - 22nd International Conference, Proceedings

A2 - Shen, Dinggang

A2 - Yap, Pew-Thian

A2 - Liu, Tianming

A2 - Peters, Terry M.

A2 - Khan, Ali

A2 - Staib, Lawrence H.

A2 - Essert, Caroline

A2 - Zhou, Sean

PB - Springer

ER -