Learning from the Truth: Fully Automatic Ground Truth Generation for Training of Medical Deep Learning Networks

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

Automatic medical image analysis has become an invaluable tool in the different treatment stages of diseases. Especially medical image segmentation plays a vital role, since segmentation is often the initial step in an image analysis pipeline. Convolutional neural networks (CNNs) have rapidly become a state of the art method for many medical image analysis tasks, such as segmentation. However, in the medical domain, the use of CNNs is limited by a major bottleneck: the lack of training data sets for supervised learning. Although millions of medical images have been collected in clinical routine, relevant annotations for those images are hard to acquire. Generally, annotations are created (semi-)manually by experts on a slice-by-slice basis, which is time consuming and tedious. Therefore, available annotated data sets are often too small for deep learning techniques. To overcome these problems, we proposed a novel method to automatically generate ground truth annotations by exploiting positron emission tomography (PET) data acquired simultaneously with computed tomography (CT) scans in combined PET/CT systems.
Original languageEnglish
Title of host publicationProceedings of the Joint ARW & OAGM Workshop 2019
EditorsAndreas Pichler, Peter M. Roth, Robert Slabatnig, Gernot Stübl, Markus Vincze
Place of PublicationGraz
PublisherVerlag der Technischen Universität Graz
Pages173-174
ISBN (Electronic) 978-3-85125-663-5
DOIs
Publication statusPublished - 2019
EventARW & OAGM Workshop 2019: Austrian Robotics Workshop and OAGM Workshop 2019 - Steyr, Austria
Duration: 9 May 201910 May 2019

Conference

ConferenceARW & OAGM Workshop 2019
CountryAustria
CitySteyr
Period9/05/1910/05/19

Fingerprint

Image analysis
Positron emission tomography
Tomography
Neural networks
Supervised learning
Image segmentation
Pipelines
Deep learning

Cite this

Gsaxner, C., Roth, P. M., Wallner, J., & Egger, J. (2019). Learning from the Truth: Fully Automatic Ground Truth Generation for Training of Medical Deep Learning Networks. In A. Pichler, P. M. Roth, R. Slabatnig, G. Stübl, & M. Vincze (Eds.), Proceedings of the Joint ARW & OAGM Workshop 2019 (pp. 173-174). Graz: Verlag der Technischen Universität Graz. https://doi.org/10.3217/978-3-85125-663-5

Learning from the Truth: Fully Automatic Ground Truth Generation for Training of Medical Deep Learning Networks. / Gsaxner, Christina; Roth, Peter M.; Wallner, Jürgen; Egger, Jan.

Proceedings of the Joint ARW & OAGM Workshop 2019. ed. / Andreas Pichler; Peter M. Roth; Robert Slabatnig; Gernot Stübl; Markus Vincze. Graz : Verlag der Technischen Universität Graz, 2019. p. 173-174.

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Gsaxner, C, Roth, PM, Wallner, J & Egger, J 2019, Learning from the Truth: Fully Automatic Ground Truth Generation for Training of Medical Deep Learning Networks. in A Pichler, PM Roth, R Slabatnig, G Stübl & M Vincze (eds), Proceedings of the Joint ARW & OAGM Workshop 2019. Verlag der Technischen Universität Graz, Graz, pp. 173-174, ARW & OAGM Workshop 2019, Steyr, Austria, 9/05/19. https://doi.org/10.3217/978-3-85125-663-5
Gsaxner C, Roth PM, Wallner J, Egger J. Learning from the Truth: Fully Automatic Ground Truth Generation for Training of Medical Deep Learning Networks. In Pichler A, Roth PM, Slabatnig R, Stübl G, Vincze M, editors, Proceedings of the Joint ARW & OAGM Workshop 2019. Graz: Verlag der Technischen Universität Graz. 2019. p. 173-174 https://doi.org/10.3217/978-3-85125-663-5
Gsaxner, Christina ; Roth, Peter M. ; Wallner, Jürgen ; Egger, Jan. / Learning from the Truth: Fully Automatic Ground Truth Generation for Training of Medical Deep Learning Networks. Proceedings of the Joint ARW & OAGM Workshop 2019. editor / Andreas Pichler ; Peter M. Roth ; Robert Slabatnig ; Gernot Stübl ; Markus Vincze. Graz : Verlag der Technischen Universität Graz, 2019. pp. 173-174
@inproceedings{c38c4a49decd424aaa7e468a5d1e2551,
title = "Learning from the Truth: Fully Automatic Ground Truth Generation for Training of Medical Deep Learning Networks",
abstract = "Automatic medical image analysis has become an invaluable tool in the different treatment stages of diseases. Especially medical image segmentation plays a vital role, since segmentation is often the initial step in an image analysis pipeline. Convolutional neural networks (CNNs) have rapidly become a state of the art method for many medical image analysis tasks, such as segmentation. However, in the medical domain, the use of CNNs is limited by a major bottleneck: the lack of training data sets for supervised learning. Although millions of medical images have been collected in clinical routine, relevant annotations for those images are hard to acquire. Generally, annotations are created (semi-)manually by experts on a slice-by-slice basis, which is time consuming and tedious. Therefore, available annotated data sets are often too small for deep learning techniques. To overcome these problems, we proposed a novel method to automatically generate ground truth annotations by exploiting positron emission tomography (PET) data acquired simultaneously with computed tomography (CT) scans in combined PET/CT systems.",
author = "Christina Gsaxner and Roth, {Peter M.} and J{\"u}rgen Wallner and Jan Egger",
year = "2019",
doi = "10.3217/978-3-85125-663-5",
language = "English",
pages = "173--174",
editor = "Andreas Pichler and Roth, {Peter M.} and Robert Slabatnig and Gernot St{\"u}bl and Markus Vincze",
booktitle = "Proceedings of the Joint ARW & OAGM Workshop 2019",
publisher = "Verlag der Technischen Universit{\"a}t Graz",

}

TY - GEN

T1 - Learning from the Truth: Fully Automatic Ground Truth Generation for Training of Medical Deep Learning Networks

AU - Gsaxner, Christina

AU - Roth, Peter M.

AU - Wallner, Jürgen

AU - Egger, Jan

PY - 2019

Y1 - 2019

N2 - Automatic medical image analysis has become an invaluable tool in the different treatment stages of diseases. Especially medical image segmentation plays a vital role, since segmentation is often the initial step in an image analysis pipeline. Convolutional neural networks (CNNs) have rapidly become a state of the art method for many medical image analysis tasks, such as segmentation. However, in the medical domain, the use of CNNs is limited by a major bottleneck: the lack of training data sets for supervised learning. Although millions of medical images have been collected in clinical routine, relevant annotations for those images are hard to acquire. Generally, annotations are created (semi-)manually by experts on a slice-by-slice basis, which is time consuming and tedious. Therefore, available annotated data sets are often too small for deep learning techniques. To overcome these problems, we proposed a novel method to automatically generate ground truth annotations by exploiting positron emission tomography (PET) data acquired simultaneously with computed tomography (CT) scans in combined PET/CT systems.

AB - Automatic medical image analysis has become an invaluable tool in the different treatment stages of diseases. Especially medical image segmentation plays a vital role, since segmentation is often the initial step in an image analysis pipeline. Convolutional neural networks (CNNs) have rapidly become a state of the art method for many medical image analysis tasks, such as segmentation. However, in the medical domain, the use of CNNs is limited by a major bottleneck: the lack of training data sets for supervised learning. Although millions of medical images have been collected in clinical routine, relevant annotations for those images are hard to acquire. Generally, annotations are created (semi-)manually by experts on a slice-by-slice basis, which is time consuming and tedious. Therefore, available annotated data sets are often too small for deep learning techniques. To overcome these problems, we proposed a novel method to automatically generate ground truth annotations by exploiting positron emission tomography (PET) data acquired simultaneously with computed tomography (CT) scans in combined PET/CT systems.

U2 - 10.3217/978-3-85125-663-5

DO - 10.3217/978-3-85125-663-5

M3 - Conference contribution

SP - 173

EP - 174

BT - Proceedings of the Joint ARW & OAGM Workshop 2019

A2 - Pichler, Andreas

A2 - Roth, Peter M.

A2 - Slabatnig, Robert

A2 - Stübl, Gernot

A2 - Vincze, Markus

PB - Verlag der Technischen Universität Graz

CY - Graz

ER -