Exploit fully automatic low-level segmented PET data for training high-level deep learning algorithms for the corresponding CT data

Publikation: Beitrag in einer FachzeitschriftArtikelForschungBegutachtung

Abstract

We present an approach for fully automatic urinary bladder segmentation in CT images with artificial neural networks in this study. 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. Since deep neural networks have made a large impact on the field of image processing in the past years, we use two different deep learning architectures to segment the urinary bladder. Both of these architectures are based on pre-trained classification networks that are adapted to perform semantic segmentation. Since deep neural networks require a large amount of training data, specifically images and corresponding ground truth labels, we furthermore propose a method to generate such a suitable training data set from Positron Emission Tomography/Computed Tomography image data. This is done by applying thresholding to the Positron Emission Tomography data for obtaining a ground truth and by utilizing data augmentation to enlarge the dataset. In this study, we discuss the influence of data augmentation on the segmentation results, and compare and evaluate the proposed architectures in terms of qualitative and quantitative segmentation performance. The results presented in this study allow concluding that deep neural networks can be considered a promising approach to segment the urinary bladder in CT images.
Originalspracheenglisch
Seiten (von - bis)e0212550
FachzeitschriftPLoS ONE
PublikationsstatusVeröffentlicht - 2019

Fingerprint

neural networks
Learning algorithms
Urinary Bladder
Positron emission tomography
learning
bladder
Learning
positron-emission tomography
Image analysis
image analysis
Image segmentation
Semantics
Positron-Emission Tomography
Tomography
Labels
Image processing
Pipelines
computed tomography
Neural networks
Deep learning

Dies zitieren

Exploit fully automatic low-level segmented PET data for training high-level deep learning algorithms for the corresponding CT data. / Gsaxner, Christina; Roth, Peter M.; Egger, Jan; Wallner, Jürgen.

in: PLoS ONE, 2019, S. e0212550.

Publikation: Beitrag in einer FachzeitschriftArtikelForschungBegutachtung

@article{569b3ddcd5bd42d8bb5276746aff6b48,
title = "Exploit fully automatic low-level segmented PET data for training high-level deep learning algorithms for the corresponding CT data",
abstract = "We present an approach for fully automatic urinary bladder segmentation in CT images with artificial neural networks in this study. 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. Since deep neural networks have made a large impact on the field of image processing in the past years, we use two different deep learning architectures to segment the urinary bladder. Both of these architectures are based on pre-trained classification networks that are adapted to perform semantic segmentation. Since deep neural networks require a large amount of training data, specifically images and corresponding ground truth labels, we furthermore propose a method to generate such a suitable training data set from Positron Emission Tomography/Computed Tomography image data. This is done by applying thresholding to the Positron Emission Tomography data for obtaining a ground truth and by utilizing data augmentation to enlarge the dataset. In this study, we discuss the influence of data augmentation on the segmentation results, and compare and evaluate the proposed architectures in terms of qualitative and quantitative segmentation performance. The results presented in this study allow concluding that deep neural networks can be considered a promising approach to segment the urinary bladder in CT images.",
author = "Christina Gsaxner and Roth, {Peter M.} and Jan Egger and J{\"u}rgen Wallner",
year = "2019",
language = "English",
pages = "e0212550",
journal = "PLoS ONE",
issn = "1932-6203",
publisher = "Public Library of Science",

}

TY - JOUR

T1 - Exploit fully automatic low-level segmented PET data for training high-level deep learning algorithms for the corresponding CT data

AU - Gsaxner, Christina

AU - Roth, Peter M.

AU - Egger, Jan

AU - Wallner, Jürgen

PY - 2019

Y1 - 2019

N2 - We present an approach for fully automatic urinary bladder segmentation in CT images with artificial neural networks in this study. 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. Since deep neural networks have made a large impact on the field of image processing in the past years, we use two different deep learning architectures to segment the urinary bladder. Both of these architectures are based on pre-trained classification networks that are adapted to perform semantic segmentation. Since deep neural networks require a large amount of training data, specifically images and corresponding ground truth labels, we furthermore propose a method to generate such a suitable training data set from Positron Emission Tomography/Computed Tomography image data. This is done by applying thresholding to the Positron Emission Tomography data for obtaining a ground truth and by utilizing data augmentation to enlarge the dataset. In this study, we discuss the influence of data augmentation on the segmentation results, and compare and evaluate the proposed architectures in terms of qualitative and quantitative segmentation performance. The results presented in this study allow concluding that deep neural networks can be considered a promising approach to segment the urinary bladder in CT images.

AB - We present an approach for fully automatic urinary bladder segmentation in CT images with artificial neural networks in this study. 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. Since deep neural networks have made a large impact on the field of image processing in the past years, we use two different deep learning architectures to segment the urinary bladder. Both of these architectures are based on pre-trained classification networks that are adapted to perform semantic segmentation. Since deep neural networks require a large amount of training data, specifically images and corresponding ground truth labels, we furthermore propose a method to generate such a suitable training data set from Positron Emission Tomography/Computed Tomography image data. This is done by applying thresholding to the Positron Emission Tomography data for obtaining a ground truth and by utilizing data augmentation to enlarge the dataset. In this study, we discuss the influence of data augmentation on the segmentation results, and compare and evaluate the proposed architectures in terms of qualitative and quantitative segmentation performance. The results presented in this study allow concluding that deep neural networks can be considered a promising approach to segment the urinary bladder in CT images.

M3 - Article

SP - e0212550

JO - PLoS ONE

JF - PLoS ONE

SN - 1932-6203

ER -