Deep learning-enabled, computed-tomography-based race- and sex-specific epicardial adipose tissue thresholds for cardiovascular risk stratification
Bibliographic record
Abstract
While the volume of epicardial adipose tissue (EAT) has been linked to various conditions and showed a prognostic value of cardiovascular events, an insufficient effort was put into proposing clear, race- and sex-specific thresholds of high-risk and low-risk EAT amounts. Also, previous works focused on the absolute volume of EAT in milliliters, adjusted to the body size of the patients, and none has proposed a relative measure of what is the amount of EAT for a given patient also in reference to the size of the patient’s heart. In this retrospective study, computed tomography attenuation correction (CTAC) scans from 20,587 patients (58% male; 84% White, 13% Black, 3% Asian) from 6 sites were used. With the previously described deep learning (DL)-based method, EAT values per patient were calculated, and subsequently normalized using either the DL-obtained cardiac volume mask, body size, or both. The association between the percent of EAT in reference to the heart size and incident mortality or nonfatal myocardial infarction (MI) was evaluated with Cox models adjusted for age, sex, body mass index, race, cigarette smoking status, hypertension, diabetes, family history of coronary artery disease, dyslipidemia, and left ventricle size. In total, 4,133 events were observed over a period of 6-year follow-up. Our analysis showed that when the amount of EAT was normalized jointly for body surface area and heart size, the predictive effect of EAT was diminished (p=0.85). Patients with higher EAT density, however, were more likely to experience death or MI during follow-up (p⪅0.001). These results were also confirmed in an analysis with race- and sex-specific thresholds. Additionally, per-race analysis showed that the density was predictive only for White patients (both when using global, as well as race- and sex-specific thresholds). To conclude, we showed that the previously observed effect of the EAT indexed for body size predicting cardiovascular risk was associated with patients’ cardiac sizes. Nevertheless, in our analysis EAT density was still associated with a 6-year risk of death or nonfatal myocardial infarction, and this effect was significant only for a single race.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".