Academic Bibliography

 Search 200 years of publications by Ghent University researchers.
Advanced search
1 file | 2.59 MB
Add to list
  1. Search results
  2. Machine learning pipeline for predict...

Machine learning pipeline for predicting bone marrow edema along the sacroiliac joints on magnetic resonance imaging

Joris Roels (UGent) , Ann-Sophie De Craemer (UGent) , Thomas Renson (UGent) , Manouk de Hooge (UGent) , Arne Gevaert (UGent) , Thomas Van Den Berghe (UGent) , Lennart Jans (UGent) , Nele Herregods (UGent) , Philippe Carron (UGent) , Filip Van den Bosch (UGent) , et al.
(2023) ARTHRITIS & RHEUMATOLOGY. 75(12). p.2169-2177
Author
Organization
Project
Abstract
Objective To develop and validate a fully automated machine learning (ML) algorithm that predicts bone marrow oedema (BMO) on a quadrant-level in sacroiliac (SI) joint MRI. Methods A computer vision workflow automatically locates the SI joints, segments regions of interest (ilium and sacrum), performs objective quadrant extraction and predicts presence of BMO, suggestive of inflammatory lesions, on a quadrant-level in semi-coronal slices of T1/T2-weighted MRI scans. Ground truth was determined by consensus among human readers. The inflammation classifier was trained using a ResNet18 backbone and 5-fold cross-validated on scans of spondyloarthritis (SpA) patients (n=279), postpartum (n=71), and healthy subjects (n=114); while independent SpA patient MRIs (n=243) served as test dataset. Patient-level predictions were derived from aggregating quadrant-level predictions, i.e. at least one positive quadrant. Results The algorithm automatically detects the SI joints with a precision of 98.4% and segments ilium/sacrum with an intersection-over-union of 85.6% and 67.9%, respectively. The inflammation classifier performed well in cross-validation: area under the curve (AUC) 94.5%, balanced accuracy (B-ACC) 80.5%, and F1 score 64.1%. In the test dataset, AUC was 88.2%, B-ACC 72.1%, and F1 score 50.8%. On a patient-level, the model achieved a B-ACC of 81.6% and 81.4% in the cross-validation and test dataset, respectively. Conclusion We propose a fully automated ML pipeline that enables objective and standardized evaluation of BMO along the SI joints on MRI. This method has the potential to screen large numbers of (suspected) SpA patients and is a step closer towards artificial intelligence assisted diagnosis and follow-up.
Keywords
Immunology, Rheumatology, Immunology and Allergy, SPONDYLOARTHRITIS, DIAGNOSIS, LESIONS, ALGORITHM

Downloads

  • Download
    Arthritis Rheumatology - 2023 - Roels.pdf
    • full text (Accepted manuscript)
    • |
    • open access
    • |
    • PDF
    • |
    • 2.59 MB

Citation

Please use this url to cite or link to this publication:

MLA
Roels, Joris, et al. “Machine Learning Pipeline for Predicting Bone Marrow Edema along the Sacroiliac Joints on Magnetic Resonance Imaging.” ARTHRITIS & RHEUMATOLOGY, vol. 75, no. 12, 2023, pp. 2169–77, doi:10.1002/art.42650.
APA
Roels, J., De Craemer, A.-S., Renson, T., de Hooge, M., Gevaert, A., Van Den Berghe, T., … Elewaut, D. (2023). Machine learning pipeline for predicting bone marrow edema along the sacroiliac joints on magnetic resonance imaging. ARTHRITIS & RHEUMATOLOGY, 75(12), 2169–2177. https://doi.org/10.1002/art.42650
Chicago author-date
Roels, Joris, Ann-Sophie De Craemer, Thomas Renson, Manouk de Hooge, Arne Gevaert, Thomas Van Den Berghe, Lennart Jans, et al. 2023. “Machine Learning Pipeline for Predicting Bone Marrow Edema along the Sacroiliac Joints on Magnetic Resonance Imaging.” ARTHRITIS & RHEUMATOLOGY 75 (12): 2169–77. https://doi.org/10.1002/art.42650.
Chicago author-date (all authors)
Roels, Joris, Ann-Sophie De Craemer, Thomas Renson, Manouk de Hooge, Arne Gevaert, Thomas Van Den Berghe, Lennart Jans, Nele Herregods, Philippe Carron, Filip Van den Bosch, Yvan Saeys, and Dirk Elewaut. 2023. “Machine Learning Pipeline for Predicting Bone Marrow Edema along the Sacroiliac Joints on Magnetic Resonance Imaging.” ARTHRITIS & RHEUMATOLOGY 75 (12): 2169–2177. doi:10.1002/art.42650.
Vancouver
1.
Roels J, De Craemer A-S, Renson T, de Hooge M, Gevaert A, Van Den Berghe T, et al. Machine learning pipeline for predicting bone marrow edema along the sacroiliac joints on magnetic resonance imaging. ARTHRITIS & RHEUMATOLOGY. 2023;75(12):2169–77.
IEEE
[1]
J. Roels et al., “Machine learning pipeline for predicting bone marrow edema along the sacroiliac joints on magnetic resonance imaging,” ARTHRITIS & RHEUMATOLOGY, vol. 75, no. 12, pp. 2169–2177, 2023.
@article{01H66FHJ47D3J33TF198FM1SW1,
  abstract     = {{Objective
To develop and validate a fully automated machine learning (ML) algorithm that predicts bone marrow oedema (BMO) on a quadrant-level in sacroiliac (SI) joint MRI.

Methods
A computer vision workflow automatically locates the SI joints, segments regions of interest (ilium and sacrum), performs objective quadrant extraction and predicts presence of BMO, suggestive of inflammatory lesions, on a quadrant-level in semi-coronal slices of T1/T2-weighted MRI scans. Ground truth was determined by consensus among human readers. The inflammation classifier was trained using a ResNet18 backbone and 5-fold cross-validated on scans of spondyloarthritis (SpA) patients (n=279), postpartum (n=71), and healthy subjects (n=114); while independent SpA patient MRIs (n=243) served as test dataset. Patient-level predictions were derived from aggregating quadrant-level predictions, i.e. at least one positive quadrant.

Results
The algorithm automatically detects the SI joints with a precision of 98.4% and segments ilium/sacrum with an intersection-over-union of 85.6% and 67.9%, respectively. The inflammation classifier performed well in cross-validation: area under the curve (AUC) 94.5%, balanced accuracy (B-ACC) 80.5%, and F1 score 64.1%. In the test dataset, AUC was 88.2%, B-ACC 72.1%, and F1 score 50.8%. On a patient-level, the model achieved a B-ACC of 81.6% and 81.4% in the cross-validation and test dataset, respectively.

Conclusion
We propose a fully automated ML pipeline that enables objective and standardized evaluation of BMO along the SI joints on MRI. This method has the potential to screen large numbers of (suspected) SpA patients and is a step closer towards artificial intelligence assisted diagnosis and follow-up.}},
  author       = {{Roels, Joris and De Craemer, Ann-Sophie and Renson, Thomas and de Hooge, Manouk and Gevaert, Arne and Van Den Berghe, Thomas and Jans, Lennart and Herregods, Nele and Carron, Philippe and Van den Bosch, Filip and Saeys, Yvan and Elewaut, Dirk}},
  issn         = {{2326-5191}},
  journal      = {{ARTHRITIS & RHEUMATOLOGY}},
  keywords     = {{Immunology,Rheumatology,Immunology and Allergy,SPONDYLOARTHRITIS,DIAGNOSIS,LESIONS,ALGORITHM}},
  language     = {{eng}},
  number       = {{12}},
  pages        = {{2169--2177}},
  title        = {{Machine learning pipeline for predicting bone marrow edema along the sacroiliac joints on magnetic resonance imaging}},
  url          = {{http://doi.org/10.1002/art.42650}},
  volume       = {{75}},
  year         = {{2023}},
}
Altmetric
View in Altmetric
Web of Science
Times cited: