Exploit fully automatic low-level segmented PET data for training high-level deep learning algorithms for the corresponding CT data.
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,...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Public Library of Science (PLoS)
2019-01-01
|
Series: | PLoS ONE |
Online Access: | https://doi.org/10.1371/journal.pone.0212550 |
id |
doaj-ba1cb6d8218245d499a9d3a64c40f3c9 |
---|---|
record_format |
Article |
spelling |
doaj-ba1cb6d8218245d499a9d3a64c40f3c92021-03-03T20:50:40ZengPublic Library of Science (PLoS)PLoS ONE1932-62032019-01-01143e021255010.1371/journal.pone.0212550Exploit fully automatic low-level segmented PET data for training high-level deep learning algorithms for the corresponding CT data.Christina GsaxnerPeter M RothJürgen WallnerJan EggerWe 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.https://doi.org/10.1371/journal.pone.0212550 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Christina Gsaxner Peter M Roth Jürgen Wallner Jan Egger |
spellingShingle |
Christina Gsaxner Peter M Roth Jürgen Wallner Jan Egger Exploit fully automatic low-level segmented PET data for training high-level deep learning algorithms for the corresponding CT data. PLoS ONE |
author_facet |
Christina Gsaxner Peter M Roth Jürgen Wallner Jan Egger |
author_sort |
Christina Gsaxner |
title |
Exploit fully automatic low-level segmented PET data for training high-level deep learning algorithms for the corresponding CT data. |
title_short |
Exploit fully automatic low-level segmented PET data for training high-level deep learning algorithms for the corresponding CT data. |
title_full |
Exploit fully automatic low-level segmented PET data for training high-level deep learning algorithms for the corresponding CT data. |
title_fullStr |
Exploit fully automatic low-level segmented PET data for training high-level deep learning algorithms for the corresponding CT data. |
title_full_unstemmed |
Exploit fully automatic low-level segmented PET data for training high-level deep learning algorithms for the corresponding CT data. |
title_sort |
exploit fully automatic low-level segmented pet data for training high-level deep learning algorithms for the corresponding ct data. |
publisher |
Public Library of Science (PLoS) |
series |
PLoS ONE |
issn |
1932-6203 |
publishDate |
2019-01-01 |
description |
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. |
url |
https://doi.org/10.1371/journal.pone.0212550 |
work_keys_str_mv |
AT christinagsaxner exploitfullyautomaticlowlevelsegmentedpetdatafortraininghighleveldeeplearningalgorithmsforthecorrespondingctdata AT petermroth exploitfullyautomaticlowlevelsegmentedpetdatafortraininghighleveldeeplearningalgorithmsforthecorrespondingctdata AT jurgenwallner exploitfullyautomaticlowlevelsegmentedpetdatafortraininghighleveldeeplearningalgorithmsforthecorrespondingctdata AT janegger exploitfullyautomaticlowlevelsegmentedpetdatafortraininghighleveldeeplearningalgorithmsforthecorrespondingctdata |
_version_ |
1714820289040220160 |