Structured reporting platform improves CAD-RADS assessment
Bibliographic record
Abstract
BACKGROUND: Structured reporting in cardiac imaging is strongly encouraged to improve quality through consistency. The Coronary Artery Disease - Reporting and Data System (CAD-RADS) was recently introduced to facilitate interdisciplinary communication of coronary CT angiography (CTA) results. We aimed to assess the agreement between manual and automated CAD-RADS classification using a structured reporting platform. METHODS: Five readers prospectively interpreted 500 coronary CT angiographies using a structured reporting platform that automatically calculates the CAD-RADS score based on stenosis and plaque parameters manually entered by the reader. In addition, all readers manually assessed CAD-RADS blinded to the automatically derived results, which was used as the reference standard. We evaluated factors influencing reader performance including CAD-RADS training, clinical load, time of the day and level of expertise. RESULTS: Total agreement between manual and automated classification was 80.2%. Agreement in stenosis categories was 86.7%, whereas the agreement in modifiers was 95.8% for "N", 96.8% for "S", 95.6% for "V" and 99.4% for "G". Agreement for V improved after CAD-RADS training (p = 0.047). Time of the day and clinical load did not influence reader performance (p > 0.05 both). Less experienced readers had a higher total agreement as compared to more experienced readers (87.0% vs 78.0%, respectively; p = 0.011). CONCLUSIONS: Even though automated CAD-RADS classification uses data filled in by the readers, it outperforms manual classification by preventing human errors. Structured reporting platforms with automated calculation of the CAD-RADS score might improve data quality and support standardization of clinical decision making.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.004 |
| Bibliometrics | 0.000 | 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".