An automated algorithm for the detection of cortical interruptions and its underlying loss of trabecular bone; a reproducibility study
Bibliographic record
Abstract
We developed a semi-automated algorithm that detects cortical interruptions in finger joints using high-resolution peripheral quantitative computed tomography (HR-pQCT), and extended it with trabecular void volume measurement. In this study we tested the reproducibility of the algorithm using scan/re-scan data. Second and third metacarpophalangeal joints of 21 subjects (mean age 49 (SD 11) years, 17 early rheumatoid arthritis and 4 undifferentiated arthritis, all diagnosed < 1 year ago) were imaged twice by HR-pQCT on the same day with repositioning between scans. The images were analyzed twice by one operator (OP1) and once by an additional operator (OP2), who independently corrected the bone contours when necessary. The number, surface and volume of interruptions per joint were obtained. Intra- and inter-operator reliability and intra-operator reproducibility were determined by intra-class correlation coefficients (ICC). Intra-operator reproducibility errors were determined as the least significant change (LSC SD ). Per joint, the mean number of interruptions was 3.1 (SD 3.6), mean interruption surface 4.2 (SD 7.2) mm 2 , and mean interruption volume 3.5 (SD 10.6) mm 3 for OP1. Intra- and inter-operator reliability was excellent for the cortical interruption parameters (ICC ≥0.91), except good for the inter-operator reliability of the interruption surface (ICC = 0.70). The LSC SD per joint was 4.2 for the number of interruptions, 5.8 mm 2 for interruption surface, and 3.2 mm 3 for interruption volume. The algorithm was highly reproducible in the detection of cortical interruptions and their volume. Based on the LSC findings, the potential value of this algorithm for monitoring structural damage in the joints in early arthritis patients needs to be tested in clinical studies.
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 | 0.009 | 0.026 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".