Limited reliability of radiographic assessment of spinal progression in ankylosing spondylitis
Bibliographic record
Abstract
Objectives: Conventional radiography is key to assessing AS-related spinal involvement and has become increasingly important given that spinal fusion may continue under biologic therapy. We aimed to compare the reliability of radiographic scoring of the spine by using different approaches to understand how different readers agree on overall scores and on individual findings. Method: Six investigators scored 68 plain radiographs of the cervical and lumbar spine of 34 patients with a 2-year interval, for erosions, sclerosis, squaring, syndesmophytes and ankyloses using the Spondyloarthritis Radiography (SPAR) module. The intraclass correlation coefficients were calculated compared with two gold standards. The reproducibility of each finding in 1632 vertebral corners and new syndesmophytes in each corner was calculated by kappa analysis and positive agreement rates. Results: The intraclass correlation coefficients mostly revealed good to excellent agreement with the gold standards (0.69-0.95). The kappa analysis showed worse agreement, being relatively higher for syndesmophytes (0.163-0.559) and ankylosis (0.48-0.95). Positive agreement rates showed that erosions were never detected at the same vertebral corner by two readers (positive agreement rate: 0%). The mean (range) positive agreement rates were 10.1% (0-27.7%) for sclerosis and 19.2% (0-59.7%) for squaring, and were higher for syndesmophytes [38.8% (21.4-62.5%)] and ankylosis [77.3% (64-95.3%)]. Conclusion: Our results show that there is a poor agreement on the presence of grade 1 lesions included in the Modified Stoke Ankylosing Spondylitis Spine Score-mostly for erosions and sclerosis-which may increase the measurement error. The currently used definitions of reliability have a risk of overestimating reproducibility.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.034 | 0.063 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".