Clustering analysis based on automated electrocardiographic measurements to identify prognostically distinct phenotypes in patients hospitalized for heart failure: a retrospective cohort study
Bibliographic record
Abstract
Abstract Funding Acknowledgements Type of funding sources: None. Background Heart failure (HF) is a heterogeneous disease with complex structural and electrophysiological derangements of the heart. Attempts to classify HF from the electrophysiological perspective are lacking. Purpose To use electrocardiographic (ECG) data for phenotypic classification of patients with HF. Methods In this retrospective cohort study, all adult patients hospitalized for HF during 2010-2016 at a tertiary center were included. Automated measurements of the first ECG obtained during the index admission were recorded. K-means clustering using premorbid conditions and selected ECG measurements were used to classify the cohort into four mutually exclusive clusters. The primary (all-cause and cardiovascular mortality) and secondary (ventricular arrhythmia (VA)) outcomes were compared between clusters using Cox regression analysis. Results In total, 2849 patients (1363 males, age 75.1 ± 13.4 years) were included. Over a mean follow-up period of 5.37 ± 4.10 years, all-cause and cardiovascular mortality occurred in 2071 (72.7%) and 600 (21.1%) patients respectively, while VA occurred in 110 patients (3.9%). Cluster 1 was characterised by a low heart rate and low ventricular activation time (VAT). Cluster 2 was characterised by old age, low absolute QRS area, and high QTc and QT dispersion. Cluster 3 was characterised by young age, and left ventricular hypertrophy (LVH), and few had history of VA. Cluster 4 was characterised by wide QRS, hypertension, ischaemic heart disease, high VAT, and high absolute T wave area. Cluster 4 had the highest and cluster 1 the lowest risks of all-cause (hazard ratio (HR) 2.96 [1.09, 1.50], p = 0.003; Figure A) and cardiovascular mortality (HR 2.90 [1.15, 2.11], p = 0.004; Figure B). Meanwhile, cluster 2 had the highest risk of VA (HR 2.23 [1.09, 3.85], p = 0.025; Figure C) while clusters 1 and 3 similarly had the lowest risks. Conclusion HF presents with clinically and electrophysiologically distinct phenotypes. Clustering analysis is useful in identifying HF phenotypes which are prognostically significant. Abstract Figures A, B, and C
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 | 0.003 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.003 |
| Science and technology studies | 0.001 | 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".