Prospective validation in a real-world cohort of a deep learning model for left ventricular filling pressure estimation using standard 12-lead ECG
Bibliographic record
Abstract
Abstract Background Left ventricular filling pressure (LVFP) assessment by echocardiography has proven useful in both the diagnosis and prognosis of cardiovascular disease. However, diagnosis may be compromised because of limited resources or expertise and results may show indeterminate status due to inherent limitation associated with the echocardiographic assessment. Deep learning analysis of electrocardiography (ECG) has demonstrated its ability to identify subtle structural and functional abnormalities within the cardiovascular system. A novel artificial intelligence electrocardiogram (AI-ECG) algorithm to assess LVFP from a 12 leads ECG was developed using open-access databases (training dataset: n=14943), including MIMIC-IV and its derived resources, MIMIC-IV-ECG and MIMIC-IV-ECHO, where each ECG was paired to echocardiography. In the internal validation dataset (n=3889), the model demonstrated an area under the receiver-operating characteristic (AUROC) of 0.88 (95% CI 0.87-0.89) for the assessment of elevated LVFP. Methods This study aims to validate within an external prospective cohort a novel AI-ECG algorithm to assess LVFP. We prospectively enrolled 132 patients hospitalized in a cardiology unit from a tertiary care center with a standard 12-lead ECG and a complete echocardiographic assessment made within 2 hours of each other’s. Results Mean age of the cohort was 68.1 ± 11.4 years old and 23.3% were female. 56.1% of the patients suffer from hypertension and 46.3% from coronary artery disease. Prevalence of elevated LVFP in our cohort was 13.8%. The model accurately identified elevated LVFP with AUROC of 0.81 (95% CI 0.70-0.93). Conclusion A deep learning-enabled ECG demonstrates robust performance in identifying patients with elevated LVFP in a real-world clinical setting. This prospective cohort study demonstrated the feasibility of integrating an AI model as a clinical tool to enhance clinicians' assessment of patients' LVFP status from a simple 12 leads ECG. Our model may be used for various clinical applications such as early detection of heart failure, as well as for monitoring patients with heart failure to guide treatment titration and prevent hospitalization. True positive ECG for elevated LVFP
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 | 0.005 | 0.010 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".