PortalPublikationenDeePSC: A Deep Learning Model for Automated Diagnosis of Pri...

DeePSC: A Deep Learning Model for Automated Diagnosis of Primary Sclerosing Cholangitis at Two-dimensional MR Cholangiopancreatography

Standard

DeePSC: A Deep Learning Model for Automated Diagnosis of Primary Sclerosing Cholangitis at Two-dimensional MR Cholangiopancreatography. / Ragab, Haissam; Westhaeusser, Fabian; Ernst, Anne; Yamamura, Jin; Fuhlert, Patrick; Zimmermann, Marina; Sauerbeck, Julia; Shenas, Farzad; Özden, Cansu; Weidmann, Anna; Adam, Gerhard; Bonn, Stefan; Schramm, Christoph.

in: RADIOL-ARTIF INTELL, Jahrgang 5, Nr. 3, 05.2023, S. e220160.

Publikationen: SCORING: Beitrag in Fachzeitschrift/Zeitung › SCORING: Zeitschriftenaufsatz › Forschung › Begutachtung

Harvard

Ragab, H, Westhaeusser, F, Ernst, A, Yamamura, J, Fuhlert, P, Zimmermann, M, Sauerbeck, J, Shenas, F, Özden, C, Weidmann, A, Adam, G, Bonn, S & Schramm, C 2023, 'DeePSC: A Deep Learning Model for Automated Diagnosis of Primary Sclerosing Cholangitis at Two-dimensional MR Cholangiopancreatography', RADIOL-ARTIF INTELL, Jg. 5, Nr. 3, S. e220160. https://doi.org/10.1148/ryai.220160

APA

Ragab, H., Westhaeusser, F., Ernst, A., Yamamura, J., Fuhlert, P., Zimmermann, M., Sauerbeck, J., Shenas, F., Özden, C., Weidmann, A., Adam, G., Bonn, S., & Schramm, C. (2023). DeePSC: A Deep Learning Model for Automated Diagnosis of Primary Sclerosing Cholangitis at Two-dimensional MR Cholangiopancreatography. RADIOL-ARTIF INTELL, 5(3), e220160. https://doi.org/10.1148/ryai.220160

Vancouver

Ragab H, Westhaeusser F, Ernst A, Yamamura J, Fuhlert P, Zimmermann M et al. DeePSC: A Deep Learning Model for Automated Diagnosis of Primary Sclerosing Cholangitis at Two-dimensional MR Cholangiopancreatography. RADIOL-ARTIF INTELL. 2023 Mai;5(3):e220160. https://doi.org/10.1148/ryai.220160

Bibtex

@article{5b8bb11009414f1f8f74a392475ed653,
title = "DeePSC: A Deep Learning Model for Automated Diagnosis of Primary Sclerosing Cholangitis at Two-dimensional MR Cholangiopancreatography",
abstract = "PURPOSE: To develop, train, and validate a multiview deep convolutional neural network (DeePSC) for the automated diagnosis of primary sclerosing cholangitis (PSC) on two-dimensional MR cholangiopancreatography (MRCP) images.MATERIALS AND METHODS: This retrospective study included two-dimensional MRCP datasets of 342 patients (45 years ± 14 [SD]; 207 male patients) with confirmed diagnosis of PSC and 264 controls (51 years ± 16; 150 male patients). MRCP images were separated into 3-T (n = 361) and 1.5-T (n = 398) datasets, of which 39 samples each were randomly chosen as unseen test sets. Additionally, 37 MRCP images obtained with a 3-T MRI scanner from a different manufacturer were included for external testing. A multiview convolutional neural network was developed, specialized in simultaneously processing the seven images taken at different rotational angles per MRCP examination. The final model, DeePSC, derived its classification per patient from the instance expressing the highest confidence in an ensemble of 20 individually trained multiview convolutional neural networks. Predictive performance on both test sets was compared with that of four licensed radiologists using the Welch t test.RESULTS: DeePSC achieved an accuracy of 80.5% ± 1.3 (sensitivity, 80.0% ± 1.9; specificity, 81.1% ± 2.7) on the 3-T and 82.6% ± 3.0 (sensitivity, 83.6% ± 1.8; specificity, 80.0% ± 8.9) on the 1.5-T test set and scored even higher on the external test set (accuracy, 92.4% ± 1.1; sensitivity, 100.0% ± 0.0; specificity, 83.5% ± 2.4). DeePSC outperformed radiologists in average prediction accuracy by 5.5 (P = .34, 3 T) and 10.1 (P = .13, 1.5 T) percentage points.CONCLUSION: Automated classification of PSC-compatible findings based on two-dimensional MRCP was achievable and demonstrated high accuracy on internal and external test sets.Keywords: Neural Networks, Deep Learning, Liver Disease, MRI, Primary Sclerosing Cholangitis, MR Cholangiopancreatography Supplemental material is available for this article. {\textcopyright} RSNA, 2023.",
author = "Haissam Ragab and Fabian Westhaeusser and Anne Ernst and Jin Yamamura and Patrick Fuhlert and Marina Zimmermann and Julia Sauerbeck and Farzad Shenas and Cansu {\"O}zden and Anna Weidmann and Gerhard Adam and Stefan Bonn and Christoph Schramm",
note = "{\textcopyright} 2023 by the Radiological Society of North America, Inc.",
year = "2023",
month = may,
doi = "10.1148/ryai.220160",
language = "English",
volume = "5",
pages = "e220160",
journal = "RADIOL-ARTIF INTELL",
issn = "2638-6100",
publisher = "Radiological Society of North America Inc.",
number = "3",

}

RIS

TY - JOUR

T1 - DeePSC: A Deep Learning Model for Automated Diagnosis of Primary Sclerosing Cholangitis at Two-dimensional MR Cholangiopancreatography

AU - Ragab, Haissam

AU - Westhaeusser, Fabian

AU - Ernst, Anne

AU - Yamamura, Jin

AU - Fuhlert, Patrick

AU - Zimmermann, Marina

AU - Sauerbeck, Julia

AU - Shenas, Farzad

AU - Özden, Cansu

AU - Weidmann, Anna

AU - Adam, Gerhard

AU - Bonn, Stefan

AU - Schramm, Christoph

N1 - © 2023 by the Radiological Society of North America, Inc.

PY - 2023/5

Y1 - 2023/5

N2 - PURPOSE: To develop, train, and validate a multiview deep convolutional neural network (DeePSC) for the automated diagnosis of primary sclerosing cholangitis (PSC) on two-dimensional MR cholangiopancreatography (MRCP) images.MATERIALS AND METHODS: This retrospective study included two-dimensional MRCP datasets of 342 patients (45 years ± 14 [SD]; 207 male patients) with confirmed diagnosis of PSC and 264 controls (51 years ± 16; 150 male patients). MRCP images were separated into 3-T (n = 361) and 1.5-T (n = 398) datasets, of which 39 samples each were randomly chosen as unseen test sets. Additionally, 37 MRCP images obtained with a 3-T MRI scanner from a different manufacturer were included for external testing. A multiview convolutional neural network was developed, specialized in simultaneously processing the seven images taken at different rotational angles per MRCP examination. The final model, DeePSC, derived its classification per patient from the instance expressing the highest confidence in an ensemble of 20 individually trained multiview convolutional neural networks. Predictive performance on both test sets was compared with that of four licensed radiologists using the Welch t test.RESULTS: DeePSC achieved an accuracy of 80.5% ± 1.3 (sensitivity, 80.0% ± 1.9; specificity, 81.1% ± 2.7) on the 3-T and 82.6% ± 3.0 (sensitivity, 83.6% ± 1.8; specificity, 80.0% ± 8.9) on the 1.5-T test set and scored even higher on the external test set (accuracy, 92.4% ± 1.1; sensitivity, 100.0% ± 0.0; specificity, 83.5% ± 2.4). DeePSC outperformed radiologists in average prediction accuracy by 5.5 (P = .34, 3 T) and 10.1 (P = .13, 1.5 T) percentage points.CONCLUSION: Automated classification of PSC-compatible findings based on two-dimensional MRCP was achievable and demonstrated high accuracy on internal and external test sets.Keywords: Neural Networks, Deep Learning, Liver Disease, MRI, Primary Sclerosing Cholangitis, MR Cholangiopancreatography Supplemental material is available for this article. © RSNA, 2023.

AB - PURPOSE: To develop, train, and validate a multiview deep convolutional neural network (DeePSC) for the automated diagnosis of primary sclerosing cholangitis (PSC) on two-dimensional MR cholangiopancreatography (MRCP) images.MATERIALS AND METHODS: This retrospective study included two-dimensional MRCP datasets of 342 patients (45 years ± 14 [SD]; 207 male patients) with confirmed diagnosis of PSC and 264 controls (51 years ± 16; 150 male patients). MRCP images were separated into 3-T (n = 361) and 1.5-T (n = 398) datasets, of which 39 samples each were randomly chosen as unseen test sets. Additionally, 37 MRCP images obtained with a 3-T MRI scanner from a different manufacturer were included for external testing. A multiview convolutional neural network was developed, specialized in simultaneously processing the seven images taken at different rotational angles per MRCP examination. The final model, DeePSC, derived its classification per patient from the instance expressing the highest confidence in an ensemble of 20 individually trained multiview convolutional neural networks. Predictive performance on both test sets was compared with that of four licensed radiologists using the Welch t test.RESULTS: DeePSC achieved an accuracy of 80.5% ± 1.3 (sensitivity, 80.0% ± 1.9; specificity, 81.1% ± 2.7) on the 3-T and 82.6% ± 3.0 (sensitivity, 83.6% ± 1.8; specificity, 80.0% ± 8.9) on the 1.5-T test set and scored even higher on the external test set (accuracy, 92.4% ± 1.1; sensitivity, 100.0% ± 0.0; specificity, 83.5% ± 2.4). DeePSC outperformed radiologists in average prediction accuracy by 5.5 (P = .34, 3 T) and 10.1 (P = .13, 1.5 T) percentage points.CONCLUSION: Automated classification of PSC-compatible findings based on two-dimensional MRCP was achievable and demonstrated high accuracy on internal and external test sets.Keywords: Neural Networks, Deep Learning, Liver Disease, MRI, Primary Sclerosing Cholangitis, MR Cholangiopancreatography Supplemental material is available for this article. © RSNA, 2023.

U2 - 10.1148/ryai.220160

DO - 10.1148/ryai.220160

M3 - SCORING: Journal article

C2 - 37293347

VL - 5

SP - e220160

JO - RADIOL-ARTIF INTELL

JF - RADIOL-ARTIF INTELL

SN - 2638-6100

IS - 3

ER -