Spondyloarthritis & psoriatic arthritis
Identification of a machine learning-based diagnostic model for axial spondyloarthritis in rheumatological routine care using a random forest approach
Authors' conclusion
We provide a machine learning-based model that reveals high performance in diagnosing patients with chronic back pain with axSpA versus without axSpA based on information from a tertiary rheumatology practice.
Study context
- Population
- Patients with chronic back pain in rheumatological routine care (tertiary rheumatology practice): Axial spondyloarthritis (axSpA) vs non-axSpA659 axSpA, 280 non-axSpA
- Participants
- 939 patients
- Outcome
- Overall accuracy, sensitivity, specificity and ROC-AUC in the test dataset
Methodological detailsmachine learning, random forest algorithm, training and test datasets at a 7:3 ratio
- Machine learning
- Random forest algorithm
- Training and test datasets at a 7:3 ratio
Performance
As reported in the abstract
Diagnostic performance for axSpA vs non-axSpA in the test dataset
| Test or model | Accuracy | Sensitivity | Specificity | ROC-AUC |
|---|---|---|---|---|
| Model | 0.9234 | 0.9586 | 0.8438 | 0.9717 |
Model
Accuracy0.9234
Sensitivity0.9586
Specificity0.8438
ROC-AUC0.9717
Further findings
Variable contribution to model accuracy
- Human leucocyte antigen B27 (HLA-B27) contributed most to model accuracy, followed by insidious onset of back pain and erosions in the sacroiliac joint
Original publication
Identification of a machine learning-based diagnostic model for axial spondyloarthritis in rheumatological routine care using a random forest approach. RMD Open. Published 27 November 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →