Machine learning for patient selection in corticosteroid decision making in knee osteoarthritis: A feasibility model
Bibliographic record
Abstract
BACKGROUND: Relieving pain is central to the early management of knee osteoarthritis, with a plethora of pharmacological agents licensed for this purpose. Intra-articular corticosteroid injections are a widely used option, albeit with variable efficacy. AIM: To develop a machine learning (ML) model that predicts which patients will benefit from corticosteroid injections. METHODS: Data from two prospective cohort studies [Osteoarthritis (OA) Initiative and Multicentre OA Study] was combined. The primary outcome was patient-reported pain score following corticosteroid injection, assessed using the Western Ontario and McMaster Universities OA pain scale, with significant change defined using minimally clinically important difference and meaningful within person change. A ML algorithm was developed, utilizing linear discriminant analysis, to predict symptomatic improvement, and examine the association between pain scores and patient factors by calculating the sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and F2 score. RESULTS: A total of 330 patients were included, with a mean age of 63.4 (SD: 8.3). The mean Western Ontario and McMaster Universities OA pain score was 5.2 (SD: 4.1), with only 25.5% of patients achieving significant improvement in pain following corticosteroid injection. The ML model generated an accuracy of 67.8% (95% confidence interval: 64.6%-70.9%), F1 score of 30.8%, and an area under the curve score of 0.60. CONCLUSION: The model demonstrated feasibility to assist clinicians with decision-making in patient selection for corticosteroid injections. Further studies are required to improve the model prior to testing in clinical settings.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".