Tissue-border detection in volumetric laser endomicroscopy using bi-directional gated recurrent neural networks

Sanne E. Okel; Fons van der Sommen; Endi Selmanaj; Joost van der Putten; Maarten R. Struyvenberg; Jacques J.G.H.M. Bergman; Peter H.N. De With

doi:10.1117/12.2579751

Tissue-border detection in volumetric laser endomicroscopy using bi-directional gated recurrent neural networks

Sanne E. Okel, Fons van der Sommen, Endi Selmanaj, Joost van der Putten, Maarten R. Struyvenberg, Jacques J.G.H.M. Bergman, Peter H.N. De With

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Abstract

Computer-aided detection (CAD) approaches have shown promising results for early esophageal cancer detection using Volumetric Laser Endoscopy (VLE) imagery. However, the relatively slow and computationally costly tissue segmentation employed in these approaches hamper their clinical applicability. In this paper, we propose to reframe the 2D tissue segmentation problem into a 1D tissue boundary detection problem. Instead of using an encoder-decoder architecture, we propose to follow the tissue boundary using a Recurrent Neural Network (RNN), exploiting the spatio-temporal relations within VLE frames. We demonstrate a near state-of-the-art performance using 18 times less floating point operations, enabling real-time execution in clinical practice.

Original language	English
Title of host publication	Medical Imaging 2021
Subtitle of host publication	Computer-Aided Diagnosis
Editors	Maciej A. Mazurowski, Karen Drukker
Publisher	SPIE
Number of pages	9
ISBN (Electronic)	9781510640238
DOIs	https://doi.org/10.1117/12.2579751
Publication status	Published - 2021
Event	SPIE Medical Imaging 2021 - Online, United States Duration: 15 Feb 2021 → 19 Feb 2021

Publication series

Name	Proceedings of SPIE
Volume	11597
ISSN (Print)	1605-7422

Conference

Conference	SPIE Medical Imaging 2021
Country/Territory	United States
Period	15/02/21 → 19/02/21

Keywords

Barrett's esophagus
Computer aided detection
Deep learning
Recurrent neural network
Volumetric laser endomicroscopy

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1117/12.2579751

Cite this

Okel, S. E., van der Sommen, F., Selmanaj, E., van der Putten, J., Struyvenberg, M. R., Bergman, J. J. G. H. M., & De With, P. H. N. (2021). Tissue-border detection in volumetric laser endomicroscopy using bi-directional gated recurrent neural networks. In M. A. Mazurowski, & K. Drukker (Eds.), Medical Imaging 2021: Computer-Aided Diagnosis Article 115972H (Proceedings of SPIE; Vol. 11597). SPIE. https://doi.org/10.1117/12.2579751

@inproceedings{84bbbda87a7e4e6c869ca8dbf57bd2de,

title = "Tissue-border detection in volumetric laser endomicroscopy using bi-directional gated recurrent neural networks",

abstract = "Computer-aided detection (CAD) approaches have shown promising results for early esophageal cancer detection using Volumetric Laser Endoscopy (VLE) imagery. However, the relatively slow and computationally costly tissue segmentation employed in these approaches hamper their clinical applicability. In this paper, we propose to reframe the 2D tissue segmentation problem into a 1D tissue boundary detection problem. Instead of using an encoder-decoder architecture, we propose to follow the tissue boundary using a Recurrent Neural Network (RNN), exploiting the spatio-temporal relations within VLE frames. We demonstrate a near state-of-the-art performance using 18 times less floating point operations, enabling real-time execution in clinical practice.",

keywords = "Barrett's esophagus, Computer aided detection, Deep learning, Recurrent neural network, Volumetric laser endomicroscopy",

author = "Okel, {Sanne E.} and {van der Sommen}, Fons and Endi Selmanaj and {van der Putten}, Joost and Struyvenberg, {Maarten R.} and Bergman, {Jacques J.G.H.M.} and {De With}, {Peter H.N.}",

note = "Publisher Copyright: {\textcopyright} COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.; SPIE Medical Imaging 2021 ; Conference date: 15-02-2021 Through 19-02-2021",

year = "2021",

doi = "10.1117/12.2579751",

language = "English",

series = "Proceedings of SPIE",

publisher = "SPIE",

editor = "Mazurowski, {Maciej A.} and Karen Drukker",

booktitle = "Medical Imaging 2021",

address = "United States",

}

Okel, SE, van der Sommen, F, Selmanaj, E, van der Putten, J, Struyvenberg, MR, Bergman, JJGHM & De With, PHN 2021, Tissue-border detection in volumetric laser endomicroscopy using bi-directional gated recurrent neural networks. in MA Mazurowski & K Drukker (eds), Medical Imaging 2021: Computer-Aided Diagnosis., 115972H, Proceedings of SPIE, vol. 11597, SPIE, SPIE Medical Imaging 2021, California, United States, 15/02/21. https://doi.org/10.1117/12.2579751

Tissue-border detection in volumetric laser endomicroscopy using bi-directional gated recurrent neural networks. / Okel, Sanne E.; van der Sommen, Fons; Selmanaj, Endi et al.
Medical Imaging 2021: Computer-Aided Diagnosis. ed. / Maciej A. Mazurowski; Karen Drukker. SPIE, 2021. 115972H (Proceedings of SPIE; Vol. 11597).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Tissue-border detection in volumetric laser endomicroscopy using bi-directional gated recurrent neural networks

AU - Okel, Sanne E.

AU - van der Sommen, Fons

AU - Selmanaj, Endi

AU - van der Putten, Joost

AU - Struyvenberg, Maarten R.

AU - Bergman, Jacques J.G.H.M.

AU - De With, Peter H.N.

PY - 2021

Y1 - 2021

N2 - Computer-aided detection (CAD) approaches have shown promising results for early esophageal cancer detection using Volumetric Laser Endoscopy (VLE) imagery. However, the relatively slow and computationally costly tissue segmentation employed in these approaches hamper their clinical applicability. In this paper, we propose to reframe the 2D tissue segmentation problem into a 1D tissue boundary detection problem. Instead of using an encoder-decoder architecture, we propose to follow the tissue boundary using a Recurrent Neural Network (RNN), exploiting the spatio-temporal relations within VLE frames. We demonstrate a near state-of-the-art performance using 18 times less floating point operations, enabling real-time execution in clinical practice.

AB - Computer-aided detection (CAD) approaches have shown promising results for early esophageal cancer detection using Volumetric Laser Endoscopy (VLE) imagery. However, the relatively slow and computationally costly tissue segmentation employed in these approaches hamper their clinical applicability. In this paper, we propose to reframe the 2D tissue segmentation problem into a 1D tissue boundary detection problem. Instead of using an encoder-decoder architecture, we propose to follow the tissue boundary using a Recurrent Neural Network (RNN), exploiting the spatio-temporal relations within VLE frames. We demonstrate a near state-of-the-art performance using 18 times less floating point operations, enabling real-time execution in clinical practice.

KW - Barrett's esophagus

KW - Computer aided detection

KW - Deep learning

KW - Recurrent neural network

KW - Volumetric laser endomicroscopy

UR - http://www.scopus.com/inward/record.url?scp=85103693321&partnerID=8YFLogxK

U2 - 10.1117/12.2579751

DO - 10.1117/12.2579751

M3 - Conference contribution

AN - SCOPUS:85103693321

T3 - Proceedings of SPIE

BT - Medical Imaging 2021

A2 - Mazurowski, Maciej A.

A2 - Drukker, Karen

PB - SPIE

T2 - SPIE Medical Imaging 2021

Y2 - 15 February 2021 through 19 February 2021

ER -

Tissue-border detection in volumetric laser endomicroscopy using bi-directional gated recurrent neural networks

Abstract

Publication series

Conference

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this