TY - JOUR
T1 - Anatomy-centred deep learning improves generalisability and progression prediction in radiographic sacroiliitis detection
AU - Dorfner, Felix J.
AU - Vahldiek, Janis L.
AU - Donle, Leonhard
AU - Zhukov, Andrei
AU - Xu, Lina
AU - Häntze, Hartmut
AU - Makowski, Marcus R.
AU - Aerts, Hugo J.W.L.
AU - Proft, Fabian
AU - Rios Rodriguez, Valeria
AU - Rademacher, Judith
AU - Protopopov, Mikhail
AU - Haibel, Hildrun
AU - Hermann, Kay Geert
AU - Diekhoff, Torsten
AU - Adams, Lisa C.
AU - Torgutalp, Murat
AU - Poddubnyy, Denis
AU - Bressem, Keno K.
N1 - Publisher Copyright:
© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ Group.
PY - 2024/12/23
Y1 - 2024/12/23
N2 - PURPOSE: To examine whether incorporating anatomy-centred deep learning can improve generalisability and enable prediction of disease progression. METHODS: This retrospective multicentre study included conventional pelvic radiographs of four different patient cohorts focusing on axial spondyloarthritis collected at university and community hospitals. The first cohort, which consisted of 1483 radiographs, was split into training (n=1261) and validation (n=222) sets. The other cohorts comprising 436, 340 and 163 patients, respectively, were used as independent test datasets. For the second cohort, follow-up data of 311 patients was used to examine progression prediction capabilities. Two neural networks were trained, one on images cropped to the bounding box of the sacroiliac joints (anatomy-centred) and the other one on full radiographs. The performance of the models was compared using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity and specificity. RESULTS: On the three test datasets, the standard model achieved AUC scores of 0.853, 0.817, 0.947, with an accuracy of 0.770, 0.724, 0.850. Whereas the anatomy-centred model achieved AUC scores of 0.899, 0.846, 0.957, with an accuracy of 0.821, 0.744, 0.906, respectively. The patients who were identified as high risk by the anatomy-centred model had an OR of 2.16 (95% CI 1.19, 3.86) for having progression of radiographic sacroiliitis within 2 years. CONCLUSION: Anatomy-centred deep learning can improve the generalisability of models in detecting radiographic sacroiliitis. The model is published as fully open source alongside this study.
AB - PURPOSE: To examine whether incorporating anatomy-centred deep learning can improve generalisability and enable prediction of disease progression. METHODS: This retrospective multicentre study included conventional pelvic radiographs of four different patient cohorts focusing on axial spondyloarthritis collected at university and community hospitals. The first cohort, which consisted of 1483 radiographs, was split into training (n=1261) and validation (n=222) sets. The other cohorts comprising 436, 340 and 163 patients, respectively, were used as independent test datasets. For the second cohort, follow-up data of 311 patients was used to examine progression prediction capabilities. Two neural networks were trained, one on images cropped to the bounding box of the sacroiliac joints (anatomy-centred) and the other one on full radiographs. The performance of the models was compared using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity and specificity. RESULTS: On the three test datasets, the standard model achieved AUC scores of 0.853, 0.817, 0.947, with an accuracy of 0.770, 0.724, 0.850. Whereas the anatomy-centred model achieved AUC scores of 0.899, 0.846, 0.957, with an accuracy of 0.821, 0.744, 0.906, respectively. The patients who were identified as high risk by the anatomy-centred model had an OR of 2.16 (95% CI 1.19, 3.86) for having progression of radiographic sacroiliitis within 2 years. CONCLUSION: Anatomy-centred deep learning can improve the generalisability of models in detecting radiographic sacroiliitis. The model is published as fully open source alongside this study.
KW - Axial Spondyloarthritis
KW - Classification
KW - Machine Learning
KW - Spondylitis, Ankylosing
UR - http://www.scopus.com/inward/record.url?scp=85213936837&partnerID=8YFLogxK
U2 - 10.1136/rmdopen-2024-004628
DO - 10.1136/rmdopen-2024-004628
M3 - Article
C2 - 39719299
AN - SCOPUS:85213936837
SN - 2056-5933
VL - 10
JO - RMD Open
JF - RMD Open
IS - 4
ER -