Detailed analysis of PAC characteristics in ambulatory ECG data can improve atrial fibrillation risk prediction
Bibliographic record
Abstract
Abstract Introduction Premature atrial complexes (PACs) are associated with atrial fibrillation (AF) risk, but also common in the general population. The origin and mechanism through which PACs arise lead to different electrocardiographic presentations, and may affect AF risk. We hypothesised that PAC characteristics on ambulatory ECG may predict AF, beyond PAC frequency. Methods We included 12,660 patients with 14-30 days of lead II and III ambulatory ECG recording with a Holter device, without AF during the first 48 hours of recording. We detected QRS complexes using AI and trained a convolutional neural network model to provide automated ECG waveform measurements, using >63,000 annotated heart beats. With this we extracted a total of 89 variables related PAC P-wave morphology, pattern of occurrence in relation to other beats, and PAC-QRS coupling intervals from raw ECG signals. After splitting the dataset randomly into a training (60%) and testing (40%) dataset, we used these PAC variables from the first 48h of recording in a Gaussian mixed model to generate two clusters of patients classified as having either high or low risk of AF of ≥30 seconds in the subsequent ≥12 recording days. The association between the high AF risk cluster and subsequent AF was then tested in Cox regression models, adjusted for age and sex. Results The testing dataset included 5,135 patients (median age 63 years (IQR 49-73), 59% female). AF occurred in 360 (7.0%) patients after a median duration of 7 (IQR 3-14) days. Figure 1 shows the cumulative hazard of AF by high and low risk cluster and PAC frequency. A PAC frequency >100/day was associated with higher AF risk. The high AF-risk cluster included 3,142 patients (61.1%) and had increased AF occurrence in patients with ≤100 daily PACs (62.6% of all patients), HR 1.62, 95% CI 1.03-2.54, p=0.04, but not in patients with >100 PACs (HR 0.89, 95% CI 0.63-1.26, p=0.53), p for interaction = 0.003. Conclusion Clustering analyses based on PAC morphology, pattern of occurrence and coupling interval characteristics can be used to predict AF risk in patients with ≤100 PACs/daycumulative hazards for incident AF
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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".