Automatic evaluation of atlantoaxial subluxation in rheumatoid arthritis by a deep learning model
Authors' conclusion
We present the development of a deep learning model for the evaluation of cervical lesions of patients with RA. The model was demonstrably shown to be useful for quantitative evaluation.
Study context
- Population
- Patients with rheumatoid arthritis (RA); X-ray images of the cervical spine
- Participants
- 906 patients4,691 X-ray images
- Comparison
- Deep learning model vs clinicians for atlantodental interval (ADI) and spinal cord (SAC) measurement and assessing atlantoaxial subluxation (AAS) diagnosis
- Outcome
- AAS diagnosis, test set
Methodological detailsdeep learning model development, training, validation and test sets
- Deep learning model development
- Training, validation and test sets
Performance
As reported in the abstract
| Test or model | Sensitivity | Specificity |
|---|---|---|
| ADI by model | 0.86 | 0.57 |
| SAC by model | 0.97 | 0.5 |
Sensitivity
ADI by model0.86
SAC by model0.97
Specificity
ADI by model0.57
SAC by model0.5
- Four coordinates correctly identified
- 99.5%
Original publication
Automatic evaluation of atlantoaxial subluxation in rheumatoid arthritis by a deep learning model. Arthritis Res Ther. Published 25 September 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →