Discrimination of atrial fibrillation burden using clinical and cardiac imaging data
Bibliographic record
Abstract
Abstract Background Current literature suggests that atrial fibrillation (AF) burden has prognostic implications in AF patients. However, it is not feasible to assess AF burden in all AF patients. Therefore, we aim to investigate whether clinical and cardiac imaging variables can stratify between a high and low AF burden. Method Data from 170 patients with paroxysmal (61%) or persistent (39%) AF at baseline, enrolled into the prospective, multicenter Swiss-AF Burden study, were analyzed. All patients underwent a 7-day Holter ECG and standardized cardiac magnetic resonance imaging (cMRI) without gadolinium. AF burden, defined as the percentage of time in AF, was dichotomized due to its nonlinear distribution and categorized as low (<10%) or high (≥10%). Both a clinical and cardiac imaging logistic regression model were built using a stepwise selection based on the Akaike Information Criterion (AIC) for each set of predefined variables (Figure). A combined model was then created by merging the clinical and imaging models. All models were adjusted to age and sex. Discriminative performance was evaluated by comparing AUCs. Results The median age was 72 years, and 18% were female. Overall, 26% (n=44) had a high (≥10%) AF burden (Median 0%, IQR 15.3%). The selected variables in the clinical model were BMI, and in the cardiac imaging model LA max volume index, LVED volume index, RA fractional area change and LVEF. The AUCs (95% CI) for the clinical and cardiac imaging models were 0.67 (0.58–0.77) and 0.91 (0.84–0.98), respectively (Figure). Combining both models resulted in an AUC of 0.92 (0.86–0.99), with no substantial improvement over the cardiac imaging model alone. The estimated associations were [OR (95% CI)]: 1.16 (1.05–1.31) for BMI, 1.04 (1.01–1.07) for LA max volume index, 0.94 (0.91–0.97) for LVED volume index, 0.92 (0.87–0.97) for RA fractional area change and 0.88 (0.81–0.94) for LVEF. Conclusion Cardiac imaging variables possess superior discriminatory ability than clinical variables, suggesting their potential as a tool for estimating AF burden.
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 | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 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".