Automated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning

Abdominal aortic aneurysms (AAA) may remain clinically silent until they enlarge and patients present with a potentially lethal rupture. This necessitates early detection and elective treatment. The goal of this study was to develop an easy-to-train algorithm which is capable of automated AAA screen...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Alena-K. Golla, Christian Tönnes, Tom Russ, Dominik F. Bauer, Matthias F. Froelich, Steffen J. Diehl, Stefan O. Schoenberg, Michael Keese, Lothar R. Schad, Frank G. Zöllner, Johann S. Rink
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	deep learning computed X ray tomography abdominal aortic aneurysm image classification interpretable artificial intelligence Medicine (General) R5-920
Acceso en línea:	https://doaj.org/article/03bf2908103447d88829b25b0ecaa4b7
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:03bf2908103447d88829b25b0ecaa4b7
record_format	dspace
spelling	oai:doaj.org-article:03bf2908103447d88829b25b0ecaa4b72021-11-25T17:21:52ZAutomated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning10.3390/diagnostics111121312075-4418https://doaj.org/article/03bf2908103447d88829b25b0ecaa4b72021-11-01T00:00:00Zhttps://www.mdpi.com/2075-4418/11/11/2131https://doaj.org/toc/2075-4418Abdominal aortic aneurysms (AAA) may remain clinically silent until they enlarge and patients present with a potentially lethal rupture. This necessitates early detection and elective treatment. The goal of this study was to develop an easy-to-train algorithm which is capable of automated AAA screening in CT scans and can be applied to an intra-hospital environment. Three deep convolutional neural networks (ResNet, VGG-16 and AlexNet) were adapted for 3D classification and applied to a dataset consisting of 187 heterogenous CT scans. The 3D ResNet outperformed both other networks. Across the five folds of the first training dataset it achieved an accuracy of 0.856 and an area under the curve (AUC) of 0.926. Subsequently, the algorithms performance was verified on a second data set containing 106 scans, where it ran fully automated and resulted in an accuracy of 0.953 and an AUC of 0.971. A layer-wise relevance propagation (LRP) made the decision process interpretable and showed that the network correctly focused on the aortic lumen. In conclusion, the deep learning-based screening proved to be robust and showed high performance even on a heterogeneous multi-center data set. Integration into hospital workflow and its effect on aneurysm management would be an exciting topic of future research.Alena-K. GollaChristian TönnesTom RussDominik F. BauerMatthias F. FroelichSteffen J. DiehlStefan O. SchoenbergMichael KeeseLothar R. SchadFrank G. ZöllnerJohann S. RinkMDPI AGarticledeep learningcomputed X ray tomographyabdominal aortic aneurysmimage classificationinterpretable artificial intelligenceMedicine (General)R5-920ENDiagnostics, Vol 11, Iss 2131, p 2131 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	deep learning computed X ray tomography abdominal aortic aneurysm image classification interpretable artificial intelligence Medicine (General) R5-920
spellingShingle	deep learning computed X ray tomography abdominal aortic aneurysm image classification interpretable artificial intelligence Medicine (General) R5-920 Alena-K. Golla Christian Tönnes Tom Russ Dominik F. Bauer Matthias F. Froelich Steffen J. Diehl Stefan O. Schoenberg Michael Keese Lothar R. Schad Frank G. Zöllner Johann S. Rink Automated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning
description	Abdominal aortic aneurysms (AAA) may remain clinically silent until they enlarge and patients present with a potentially lethal rupture. This necessitates early detection and elective treatment. The goal of this study was to develop an easy-to-train algorithm which is capable of automated AAA screening in CT scans and can be applied to an intra-hospital environment. Three deep convolutional neural networks (ResNet, VGG-16 and AlexNet) were adapted for 3D classification and applied to a dataset consisting of 187 heterogenous CT scans. The 3D ResNet outperformed both other networks. Across the five folds of the first training dataset it achieved an accuracy of 0.856 and an area under the curve (AUC) of 0.926. Subsequently, the algorithms performance was verified on a second data set containing 106 scans, where it ran fully automated and resulted in an accuracy of 0.953 and an AUC of 0.971. A layer-wise relevance propagation (LRP) made the decision process interpretable and showed that the network correctly focused on the aortic lumen. In conclusion, the deep learning-based screening proved to be robust and showed high performance even on a heterogeneous multi-center data set. Integration into hospital workflow and its effect on aneurysm management would be an exciting topic of future research.
format	article
author	Alena-K. Golla Christian Tönnes Tom Russ Dominik F. Bauer Matthias F. Froelich Steffen J. Diehl Stefan O. Schoenberg Michael Keese Lothar R. Schad Frank G. Zöllner Johann S. Rink
author_facet	Alena-K. Golla Christian Tönnes Tom Russ Dominik F. Bauer Matthias F. Froelich Steffen J. Diehl Stefan O. Schoenberg Michael Keese Lothar R. Schad Frank G. Zöllner Johann S. Rink
author_sort	Alena-K. Golla
title	Automated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning
title_short	Automated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning
title_full	Automated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning
title_fullStr	Automated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning
title_full_unstemmed	Automated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning
title_sort	automated screening for abdominal aortic aneurysm in ct scans under clinical conditions using deep learning
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/03bf2908103447d88829b25b0ecaa4b7
work_keys_str_mv	AT alenakgolla automatedscreeningforabdominalaorticaneurysminctscansunderclinicalconditionsusingdeeplearning AT christiantonnes automatedscreeningforabdominalaorticaneurysminctscansunderclinicalconditionsusingdeeplearning AT tomruss automatedscreeningforabdominalaorticaneurysminctscansunderclinicalconditionsusingdeeplearning AT dominikfbauer automatedscreeningforabdominalaorticaneurysminctscansunderclinicalconditionsusingdeeplearning AT matthiasffroelich automatedscreeningforabdominalaorticaneurysminctscansunderclinicalconditionsusingdeeplearning AT steffenjdiehl automatedscreeningforabdominalaorticaneurysminctscansunderclinicalconditionsusingdeeplearning AT stefanoschoenberg automatedscreeningforabdominalaorticaneurysminctscansunderclinicalconditionsusingdeeplearning AT michaelkeese automatedscreeningforabdominalaorticaneurysminctscansunderclinicalconditionsusingdeeplearning AT lotharrschad automatedscreeningforabdominalaorticaneurysminctscansunderclinicalconditionsusingdeeplearning AT frankgzollner automatedscreeningforabdominalaorticaneurysminctscansunderclinicalconditionsusingdeeplearning AT johannsrink automatedscreeningforabdominalaorticaneurysminctscansunderclinicalconditionsusingdeeplearning
_version_	1718412450087305216

Automated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning

Ejemplares similares