Automated Classification of Radiographic Positioning of Hand X-Rays Using a Deep Neural Network
Bibliographic record
Abstract
BACKGROUND: Hand X-rays are ordered in outpatient, inpatient, and emergency settings, the results of which are often initially interpreted by non-radiology trained health care providers. There may be utility in automating upper extremity X-ray analysis to aid with rapid initial analysis. Deep neural networks have been effective in several medical imaging analysis applications. The purpose of this work was to apply a deep learning framework to automatically classify the radiographic positioning of hand X-rays. METHODS: A 152-layer deep neural network was trained using the musculoskeletal radiographs data set. This data set contains 6003 hand X-rays. The data set was filtered to remove pediatric X-rays and atypical views. The X-rays were all labeled as either posteroanterior (PA), lateral, or oblique views. A subset of images was set aside for model validation and testing. Data set augmentation was performed, including horizontal and vertical flips, rotations, as well as modifications in image brightness and contrast. The model was evaluated, and performance was reported as a confusion matrix from which accuracy, precision, sensitivity, and specificity were calculated. RESULTS: The augmented training data set consisted of 80 672 images. Their distribution was 38% PA, 35% lateral, and 27% oblique projections. When evaluated on the test data set, the model performed with overall 96.0% accuracy, 93.6% precision, 93.6% sensitivity, and 97.1% specificity. CONCLUSIONS: Radiographic positioning of hand X-rays can be effectively classified by a deep neural network. Further work will be performed on localization of abnormalities, automated assessment of standard radiographic measures and eventually on computer-aided diagnosis and management guidance of skeletal pathology.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 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".