EUVIMEDEUVIMED entdecken
Multizentrische StudieVergleichsstudie

Deep learning based CT grading system for sacroiliitis: a multi-center studydemonstrating superior accuracy and efficiency compared to human readers

Mengling Xu, Zhihao Li, Ningning Ding, Zhiteng Li, Zhenzhong Ma, Yibin Xi, Chao Li, Hua Guo, Jian Yang

European Journal of Radiology · 2026

DOI: 10.1016/j.ejrad.2025.112557 ↗

Kurz gefasst

Multizentrische Studie in der Zeitschrift European Journal of Radiology, 2026. Verfasst von Mengling Xu und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, Xi'an Fifth Hospital, Xi'an, China. Electronic address: 18709676155@163.com. Hauptthemen laut MEDLINE: Deep Learning, Sacroiliitis, Tomography, X-Ray Computed, Radiographic Image Interpretation, Computer-Assisted. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 33 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND: To develop and validate a deep convolutional neural network (DCNN) for automated sacroiliitis grading in axial spondyloarthritis (axSpA) using CT images. METHODS: A total of 1,590 patients were enrolled, including 1341 axSpA patients from a Rheumatology Specialist Hospital for 3D-ResNet50 model development. The model was then evaluated on an internal validation set of 130 patients and external sets from two tertiary hospitals (n = 249). Diagnostic ssensitivity and reading time were assessed for six readers (two rheumatologists, two junior radiologists, and two senior radiologists). Participant-assigned sacroiliitis Grades were compared with model predictions, and the area under the curve (AUC) was analyzed using the DeLong test. RESULTS: The 3D-ResNet50 model achieved overall sample-level diagnostic accuracies of 89.6 %, 89.9 %, 88.5 %, and 87.1 % in the training, testing, internal validation, and external validation sets, respectively.The AUCs for Grade I, II, III, and IV in external validation set were 0.93, 0.89, 0.92 and 0.88, all exceeding those of readers (AUC: 0.65-0.93), with the most significant improvement at Grade II (p 0.05 for Grade IV). The model significantly reduced diagnostic time per case (2.74 ± 0.15 s) compared to readers (119.4 ± 42.0 s). CONCLUSION: The 3D-ResNet50 model accurately grades sacroiliitis with superior sensitivity and efficiency compared to rheumatologists and radiologists. Its strong performance supports its integration into clinical practice to assist in axSpA diagnosis.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningSacroiliitisTomography, X-Ray ComputedRadiographic Image Interpretation, Computer-AssistedHumansFemaleMaleAdultMiddle AgedReproducibility of ResultsObserver VariationSensitivity and SpecificitySeverity of Illness Index

Stichwörter der Autor:innen

3D ResNetCTMulti-reader multi-centerSacroiliac arthritisaxSpA

Autor:innen und Institutionen

  • Department of Radiology, Xi'an Fifth Hospital, Xi'an, China. Electronic address: 18709676155@163.com.
  • Institute of Image Processing and Pattern Recognition, School of Information and Communications Engineering, Xi'an Jiaotong University, Xi'an, China; Department of Radiology, The First Affiliated Hospital of Xi'an Jiaot…
  • Department of Radiology, The First Affiliated Hospital of Xi'an Jiaotong University, Xi'an, China; Shaanxi Engineering Research Center of Computational Imaging and Medical Intelligence, Xi'an, China; Xi'an Key Laborator…
  • Department of Rheumatology and Immunology, Xi'an Fifth Hospital, Xi'an, China. Electronic address: 15804066356@163.com.
  • Department of Radiology, Zizhou County People's Hospital, Yulin, China. Electronic address: mzyxlm0505@163.com.
  • Department of Radiology, Xi'an People's Hospital (Xi'an Fourth Hospital), Xi'an, China. Electronic address: xyb1113@qq.com.

Zitieren

APA

Xu, M., Li, Z., Ding, N., Li, Z., Ma, Z., Xi, Y., Li, C., Guo, H., & Yang, J. (2026). Deep learning based CT grading system for sacroiliitis: a multi-center studydemonstrating superior accuracy and efficiency compared to human readers. European Journal of Radiology. https://doi.org/10.1016/j.ejrad.2025.112557

Vancouver

Xu M, Li Z, Ding N, Li Z, Ma Z, Xi Y, et al. Deep learning based CT grading system for sacroiliitis: a multi-center studydemonstrating superior accuracy and efficiency compared to human readers. European Journal of Radiology. 2026. doi:10.1016/j.ejrad.2025.112557

BibTeX
@article{Xu2026,
  author = {Xu, Mengling and Li, Zhihao and Ding, Ningning and Li, Zhiteng and Ma, Zhenzhong and Xi, Yibin and Li, Chao and Guo, Hua and Yang, Jian},
  title = {Deep learning based CT grading system for sacroiliitis: a multi-center studydemonstrating superior accuracy and efficiency compared to human readers},
  journal = {European Journal of Radiology},
  year = {2026},
  doi = {10.1016/j.ejrad.2025.112557}
}
Kostenlos starten →