Abstract 14708: Cardiovascular Calcium Scoring by Deep Learning is Predictive of Mortality in Emergency Department Patients
Bibliographic record
Abstract
Introduction: Computed tomography (CT) imaging is widely used in the emergency department (ED) setting. Calcifications of the coronary arteries, heart valves, and aorta are common incidental findings that may herald clinical or subclinical cardiovascular disease. Hypothesis: We sought to determine whether the quantitative burden of cardiovascular calcifications, as measured by a CT-based deep learning pipeline, would be predictive of short-term mortality in a diverse population of ED patients. Methods: We conducted a prospective single-center cohort study nested in the Quebec COVID-19 Biobank from March 2020 to September 2021. For the purposes of this study, we enlisted adult patients presenting to the ED with cardiopulmonary symptoms who were tested for COVID-19 and underwent CT imaging of the chest. We used a deep learning model previously developed by our team to automate the quantitative scoring of coronary artery calcification (CAC), aortic valve calcification (AVC), mitral annular calcification (MAC), and thoracic aorta calcification (TAC) from the CT images. These calcium scores were categorized as sex-stratified tertiles plus a zero-score referent category. The primary outcome was all-cause mortality at 30 and 90 days adjusted for age, sex, and COVID-19 status using multivariable logistic regression. Results: The study sample consisted of 731 ED visits among 271 unique patients with a mean age of 66 years and 47% females. COVID-19 illness was the main diagnosis in 29% of ED visits. The prevalence of any quantifiable calcification was 51% for CAC, 33% for AVC, 23% for MAC, and 80% for TAC. The statistically significant adjusted odds ratios for mortality were 2.50 (1.08, 5.81) in the highest AVC tertile at 30 days, 2.73 (1.37 5.47) in the highest CAC tertile at 90 days, and 4.42 (1.01, 19.4) in the highest TAC tertile at 90 days. These odds ratio remained similar after further adjustment for past history of myocardial infarction or heart failure. Conclusions: High calcium scores in the coronary arteries, aortic valve, and thoracic aorta are associated with heightened 30-day mortality in ED patients. Deep learning quantification of calcium scores from clinical CT scans is an opportunistic approach for risk stratification.
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 | 0.001 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 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".