An AI-based Framework for Diagnostic Quality Assessment of Ankle Radiographs

Dominik Mairhöfer; Manuel Laufer; Paul Martin Simon; Malte Sieren; Arpad Bischof; Thomas Käster; Erhardt Barth; Jörg Barkhausen; Thomas Martinetz

An AI-based Framework for Diagnostic Quality Assessment of Ankle Radiographs

Dominik Mairhöfer, Manuel Laufer, Paul Martin Simon, Malte Sieren, Arpad Bischof, Thomas Käster, Erhardt Barth, Jörg Barkhausen, Thomas Martinetz

9 Citations (Scopus)

Abstract

The quality of radiographs is of major importance for diagnosis and treatment planning. While most research regarding automated radiograph quality assessment uses technical features such as noise or contrast, we propose to use anatomical structures as more appropriate features. We show that based on such anatomical features, a modular deep-learning framework can serve as a quality control mechanism for the diagnostic quality of ankle radiographs. For evaluation, a dataset consisting of 950 ankle radiographs was collected and their quality was labeled by radiologists. We obtain an average accuracy of 94.1%, which is better than the expert radiologists are on average.

Original language	English
Title of host publication	Proceedings of the Fourth Conference on Medical Imaging with Deep Learning
Editors	Mattias Heinrich, Qi Dou, Marleen de Bruijne, Jan Lellmann, Alexander Schläfer, Floris Ernst
Number of pages	13
Volume	143
Publisher	PMLR
Publication date	01.07.2021
Pages	484-496
Publication status	Published - 01.07.2021

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being
SDG 4 Quality Education
SDG 9 Industry, Innovation, and Infrastructure
SDG 11 Sustainable Cities and Communities
SDG 12 Responsible Consumption and Production
SDG 14 Life Below Water
SDG 15 Life on Land

Research Areas and Centers

Centers: Center for Artificial Intelligence Luebeck (ZKIL)
Research Area: Intelligent Systems

DFG Research Classification Scheme

4.43-05 Image and Language Processing, Computer Graphics and Visualisation, Human Computer Interaction, Ubiquitous and Wearable Computing

Access to Document

https://webmail.inb.uni-luebeck.de/inb-publications/pdfs/MaLaSiSiBiKaBaBaMa21.pdf

Cite this

Mairhöfer, D., Laufer, M., Simon, P. M., Sieren, M., Bischof, A., Käster, T., Barth, E., Barkhausen, J., & Martinetz, T. (2021). An AI-based Framework for Diagnostic Quality Assessment of Ankle Radiographs. In M. Heinrich, Q. Dou, M. de Bruijne, J. Lellmann, A. Schläfer, & F. Ernst (Eds.), Proceedings of the Fourth Conference on Medical Imaging with Deep Learning (Vol. 143, pp. 484-496). (Proceedings of Machine Learning Research). PMLR. https://webmail.inb.uni-luebeck.de/inb-publications/pdfs/MaLaSiSiBiKaBaBaMa21.pdf

Mairhöfer, Dominik ; Laufer, Manuel ; Simon, Paul Martin et al. / An AI-based Framework for Diagnostic Quality Assessment of Ankle Radiographs. Proceedings of the Fourth Conference on Medical Imaging with Deep Learning. editor / Mattias Heinrich ; Qi Dou ; Marleen de Bruijne ; Jan Lellmann ; Alexander Schläfer ; Floris Ernst. Vol. 143 PMLR, 2021. pp. 484-496 (Proceedings of Machine Learning Research).

@inproceedings{798da295869d4f5e95df96d9507fe2e6,

title = "An AI-based Framework for Diagnostic Quality Assessment of Ankle Radiographs",

abstract = "The quality of radiographs is of major importance for diagnosis and treatment planning. While most research regarding automated radiograph quality assessment uses technical features such as noise or contrast, we propose to use anatomical structures as more appropriate features. We show that based on such anatomical features, a modular deep-learning framework can serve as a quality control mechanism for the diagnostic quality of ankle radiographs. For evaluation, a dataset consisting of 950 ankle radiographs was collected and their quality was labeled by radiologists. We obtain an average accuracy of 94.1\%, which is better than the expert radiologists are on average.",

author = "Dominik Mairh{\"o}fer and Manuel Laufer and Simon, \{Paul Martin\} and Malte Sieren and Arpad Bischof and Thomas K{\"a}ster and Erhardt Barth and J{\"o}rg Barkhausen and Thomas Martinetz",

year = "2021",

month = jul,

day = "1",

language = "English",

volume = "143",

series = "Proceedings of Machine Learning Research",

publisher = "PMLR",

pages = "484--496",

editor = "Mattias Heinrich and Qi Dou and \{de Bruijne\}, Marleen and Jan Lellmann and Alexander Schl{\"a}fer and Floris Ernst",

booktitle = "Proceedings of the Fourth Conference on Medical Imaging with Deep Learning",

}

Mairhöfer, D, Laufer, M, Simon, PM, Sieren, M, Bischof, A, Käster, T , Barth, E , Barkhausen, J & Martinetz, T 2021, An AI-based Framework for Diagnostic Quality Assessment of Ankle Radiographs. in M Heinrich, Q Dou, M de Bruijne, J Lellmann, A Schläfer & F Ernst (eds), Proceedings of the Fourth Conference on Medical Imaging with Deep Learning. vol. 143, Proceedings of Machine Learning Research, PMLR, pp. 484-496. <https://webmail.inb.uni-luebeck.de/inb-publications/pdfs/MaLaSiSiBiKaBaBaMa21.pdf>

An AI-based Framework for Diagnostic Quality Assessment of Ankle Radiographs. / Mairhöfer, Dominik; Laufer, Manuel; Simon, Paul Martin et al.
Proceedings of the Fourth Conference on Medical Imaging with Deep Learning. ed. / Mattias Heinrich; Qi Dou; Marleen de Bruijne; Jan Lellmann; Alexander Schläfer; Floris Ernst. Vol. 143 PMLR, 2021. p. 484-496 (Proceedings of Machine Learning Research).

Research output: Chapters in Books/Reports/Conference Proceedings › Conference contribution › peer-review

TY - GEN

T1 - An AI-based Framework for Diagnostic Quality Assessment of Ankle Radiographs

AU - Mairhöfer, Dominik

AU - Laufer, Manuel

AU - Simon, Paul Martin

AU - Sieren, Malte

AU - Bischof, Arpad

AU - Käster, Thomas

AU - Barth, Erhardt

AU - Barkhausen, Jörg

AU - Martinetz, Thomas

PY - 2021/7/1

Y1 - 2021/7/1

N2 - The quality of radiographs is of major importance for diagnosis and treatment planning. While most research regarding automated radiograph quality assessment uses technical features such as noise or contrast, we propose to use anatomical structures as more appropriate features. We show that based on such anatomical features, a modular deep-learning framework can serve as a quality control mechanism for the diagnostic quality of ankle radiographs. For evaluation, a dataset consisting of 950 ankle radiographs was collected and their quality was labeled by radiologists. We obtain an average accuracy of 94.1%, which is better than the expert radiologists are on average.

AB - The quality of radiographs is of major importance for diagnosis and treatment planning. While most research regarding automated radiograph quality assessment uses technical features such as noise or contrast, we propose to use anatomical structures as more appropriate features. We show that based on such anatomical features, a modular deep-learning framework can serve as a quality control mechanism for the diagnostic quality of ankle radiographs. For evaluation, a dataset consisting of 950 ankle radiographs was collected and their quality was labeled by radiologists. We obtain an average accuracy of 94.1%, which is better than the expert radiologists are on average.

M3 - Conference contribution

VL - 143

T3 - Proceedings of Machine Learning Research

SP - 484

EP - 496

BT - Proceedings of the Fourth Conference on Medical Imaging with Deep Learning

A2 - Heinrich, Mattias

A2 - Dou, Qi

A2 - de Bruijne, Marleen

A2 - Lellmann, Jan

A2 - Schläfer, Alexander

A2 - Ernst, Floris

PB - PMLR

ER -

Mairhöfer D, Laufer M, Simon PM, Sieren M, Bischof A, Käster T et al. An AI-based Framework for Diagnostic Quality Assessment of Ankle Radiographs. In Heinrich M, Dou Q, de Bruijne M, Lellmann J, Schläfer A, Ernst F, editors, Proceedings of the Fourth Conference on Medical Imaging with Deep Learning. Vol. 143. PMLR. 2021. p. 484-496. (Proceedings of Machine Learning Research).

An AI-based Framework for Diagnostic Quality Assessment of Ankle Radiographs

Abstract

UN SDGs

Research Areas and Centers

DFG Research Classification Scheme

Access to Document

Fingerprint

Cite this