Keno K Bressem1,2, Janis L Vahldiek3, Lisa Adams1,2, Stefan Markus Niehues1, Hildrun Haibel4, Valeria Rios Rodriguez4, Murat Torgutalp4, Mikhail Protopopov4, Fabian Proft4, Judith Rademacher2,4, Joachim Sieper4, Martin Rudwaleit5, Bernd Hamm1, Marcus R Makowski1,6, Kay-Geert Hermann1, Denis Poddubnyy4,7. 1. Department of Radiology, Charité - Universitätsmedizin Berlin, Hindenburgdamm 30, 12203, Berlin, Germany. 2. Berlin Institute of Health, BIH, Berlin, Germany. 3. Department of Radiology, Charité - Universitätsmedizin Berlin, Hindenburgdamm 30, 12203, Berlin, Germany. janis.vahldiek@charite.de. 4. Department of Gastroenterology, Infectious Diseases and Rheumatology, Charité - Universitätsmedizin Berlin, Berlin, Germany. 5. Department of Internal Medicine and Rheumatology, Klinikum Bielefeld Rosenhöhe, Bielefeld, Germany. 6. Department of Diagnostic and Interventional Radiology, School of Medicine, Technical University of Munich, Munich, Germany. 7. Department of Epidemiology, German Rheumatism Research Centre, Berlin, Germany.
Abstract
BACKGROUND: Radiographs of the sacroiliac joints are commonly used for the diagnosis and classification of axial spondyloarthritis. The aim of this study was to develop and validate an artificial neural network for the detection of definite radiographic sacroiliitis as a manifestation of axial spondyloarthritis (axSpA). METHODS: Conventional radiographs of the sacroiliac joints obtained in two independent studies of patients with axSpA were used. The first cohort comprised 1553 radiographs and was split into training (n = 1324) and validation (n = 229) sets. The second cohort comprised 458 radiographs and was used as an independent test dataset. All radiographs were assessed in a central reading session, and the final decision on the presence or absence of definite radiographic sacroiliitis was used as a reference. The performance of the neural network was evaluated by calculating areas under the receiver operating characteristic curves (AUCs) as well as sensitivity and specificity. Cohen's kappa and the absolute agreement were used to assess the agreement between the neural network and the human readers. RESULTS: The neural network achieved an excellent performance in the detection of definite radiographic sacroiliitis with an AUC of 0.97 and 0.94 for the validation and test datasets, respectively. Sensitivity and specificity for the cut-off weighting both measurements equally were 88% and 95% for the validation and 92% and 81% for the test set. The Cohen's kappa between the neural network and the reference judgements were 0.79 and 0.72 for the validation and test sets with an absolute agreement of 90% and 88%, respectively. CONCLUSION: Deep artificial neural networks enable the accurate detection of definite radiographic sacroiliitis relevant for the diagnosis and classification of axSpA.
BACKGROUND: Radiographs of the sacroiliac joints are commonly used for the diagnosis and classification of axial spondyloarthritis. The aim of this study was to develop and validate an artificial neural network for the detection of definite radiographic sacroiliitis as a manifestation of axial spondyloarthritis (axSpA). METHODS: Conventional radiographs of the sacroiliac joints obtained in two independent studies of patients with axSpA were used. The first cohort comprised 1553 radiographs and was split into training (n = 1324) and validation (n = 229) sets. The second cohort comprised 458 radiographs and was used as an independent test dataset. All radiographs were assessed in a central reading session, and the final decision on the presence or absence of definite radiographic sacroiliitis was used as a reference. The performance of the neural network was evaluated by calculating areas under the receiver operating characteristic curves (AUCs) as well as sensitivity and specificity. Cohen's kappa and the absolute agreement were used to assess the agreement between the neural network and the human readers. RESULTS: The neural network achieved an excellent performance in the detection of definite radiographic sacroiliitis with an AUC of 0.97 and 0.94 for the validation and test datasets, respectively. Sensitivity and specificity for the cut-off weighting both measurements equally were 88% and 95% for the validation and 92% and 81% for the test set. The Cohen's kappa between the neural network and the reference judgements were 0.79 and 0.72 for the validation and test sets with an absolute agreement of 90% and 88%, respectively. CONCLUSION: Deep artificial neural networks enable the accurate detection of definite radiographic sacroiliitis relevant for the diagnosis and classification of axSpA.
Entities:
Keywords:
Artificial intelligence; Axial spondyloarthritis; Deep learning; Machine learning; Sacroiliitis
Authors: Anne Boel; Anna Molto; Désirée van der Heijde; Adrian Ciurea; Maxime Dougados; Lianne S Gensler; Maria-José Santos; Eugenio De Miguel; Denis Poddubnyy; Martin Rudwaleit; Astrid van Tubergen; Floris A van Gaalen; Sofia Ramiro Journal: Ann Rheum Dis Date: 2019-07-30 Impact factor: 19.103
Authors: Denis Poddubnyy; Martin Rudwaleit; Hildrun Haibel; Joachim Listing; Elisabeth Märker-Hermann; Henning Zeidler; Jürgen Braun; Joachim Sieper Journal: Ann Rheum Dis Date: 2011-05-27 Impact factor: 19.103
Authors: M Rudwaleit; D van der Heijde; R Landewé; J Listing; N Akkoc; J Brandt; J Braun; C T Chou; E Collantes-Estevez; M Dougados; F Huang; J Gu; M A Khan; Y Kirazli; W P Maksymowych; H Mielants; I J Sørensen; S Ozgocmen; E Roussou; R Valle-Oñate; U Weber; J Wei; J Sieper Journal: Ann Rheum Dis Date: 2009-03-17 Impact factor: 19.103
Authors: Anneke Spoorenberg; Kurt de Vlam; Sjef van der Linden; Maxime Dougados; Herman Mielants; Hille van de Tempel; Désirée van der Heijde Journal: J Rheumatol Date: 2004-01 Impact factor: 4.666
Authors: P Mandl; V Navarro-Compán; L Terslev; P Aegerter; D van der Heijde; M A D'Agostino; X Baraliakos; S J Pedersen; A G Jurik; E Naredo; C Schueller-Weidekamm; U Weber; M C Wick; P A C Bakker; E Filippucci; P G Conaghan; M Rudwaleit; G Schett; J Sieper; S Tarp; H Marzo-Ortega; M Østergaard Journal: Ann Rheum Dis Date: 2015-04-02 Impact factor: 19.103
Authors: Scott Mayer McKinney; Marcin Sieniek; Varun Godbole; Jonathan Godwin; Natasha Antropova; Hutan Ashrafian; Trevor Back; Mary Chesus; Greg S Corrado; Ara Darzi; Mozziyar Etemadi; Florencia Garcia-Vicente; Fiona J Gilbert; Mark Halling-Brown; Demis Hassabis; Sunny Jansen; Alan Karthikesalingam; Christopher J Kelly; Dominic King; Joseph R Ledsam; David Melnick; Hormuz Mostofi; Lily Peng; Joshua Jay Reicher; Bernardino Romera-Paredes; Richard Sidebottom; Mustafa Suleyman; Daniel Tse; Kenneth C Young; Jeffrey De Fauw; Shravya Shetty Journal: Nature Date: 2020-01-01 Impact factor: 49.962
Authors: Lukas Folle; David Simon; Koray Tascilar; Gerhard Krönke; Anna-Maria Liphardt; Andreas Maier; Georg Schett; Arnd Kleyer Journal: Front Med (Lausanne) Date: 2022-03-10
Authors: Johannes Knitza; Lena Janousek; Felix Kluge; Cay Benedikt von der Decken; Stefan Kleinert; Wolfgang Vorbrüggen; Arnd Kleyer; David Simon; Axel J Hueber; Felix Muehlensiepen; Nicolas Vuillerme; Georg Schett; Bjoern M Eskofier; Martin Welcker; Peter Bartz-Bazzanella Journal: Front Med (Lausanne) Date: 2022-07-22