Artificial intelligence in the interpretation of upper extremity trauma radiographs: a systematic review and meta-analysis
Bibliographic record
Abstract
Background: Upper extremity fractures represent a significant reason for emergency room visits; however, nonexpert readings commonly lead to diagnostic errors, particularly missed fractures. Artificial intelligence (AI) has emerged as a promising tool to aid in fracture detection, but it has been shown to be comparable to physicians at best, so it remains unclear whether there is value in its increasing implementation. This review aims to analyze the existing literature on AI in the identification and interpretation of upper extremity fractures on x-ray and to assess the diagnostic performance of such AI models. Methods: Three databases were searched (MEDLINE, Embase, and CENTRAL) for studies involving AI and imaging in upper extremity orthopedics. The review was conducted in adherence to the Preferred Reporting Items for Systematic reviews and Meta-Analyses guidelines. Inclusion criteria were papers that (1) investigated fractures of the upper extremity, (2) included the use of AI models to identify or augment the identification of fractures on imaging, identify characteristic of images, or classify images, and (3) assessed X-ray, computed tomography, or magnetic resonance imaging identification of fractures. Exclusion criteria were papers that (1) were not published in English, (2) were case reports, conference abstracts, editorials, or review articles, (3) related to hand and wrist orthopedics, and (4) reported upper extremity data integrated with nonupper extremity data. Data on fracture detection accuracy, area under the curve, sensitivity, and specificity were recorded. The Quality Assessment of Diagnostic Accuracy Studies score was used to conduct a quality assessment of all included studies. A meta-analysis was conducted on the sensitivity, specificity, and AI-reader differences in sensitivity and specificity on relevant studies. Results: = 90.50%), respectively. Discussion: The AI models show promising diagnostic accuracy in the detection of upper extremity fractures, but there is significant variability in the results. Future studies should investigate whether factors such as model type or anatomic location influence accuracy in order to guide physicians on where such models will meet a minimum standard of accuracy.
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 | 0.005 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.008 | 0.002 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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".