What question did this study set out to answer?

This research aims to improve prediction accuracy of left ventricular ejection fraction (LVEF) by integrating electrocardiography (ECG) with clinical metadata.

January 10, 2026The Korean Journal of Internal Medicine0 citationsOpen Access

Enhanced prediction of left ventricular ejection fraction using electrocardiography with the addition of clinical metadata

Q: What does this research mean for the field?

Integrating ECG data with clinical metadata improves left ventricular ejection fraction (LVEF) prediction accuracy, achieving an AUC of 0.95 compared to 0.90 for ECG alone. Novelty: ClaimNovelty.NOVEL_FINDING. Consensus alignment: ConsensusAlignment.NEUTRAL.

HPHyun Woong ParkChungnam National University TKTaeseen KangChungnam National University YSYoung-Hoon SeoKonyang University Hospital

Key Result

Integrating ECG data with clinical metadata improved LVEF prediction accuracy, achieving an AUC of 0.95 compared to 0.90 for ECG alone.

Key Points

This research aims to improve prediction accuracy of left ventricular ejection fraction (LVEF) by integrating electrocardiography (ECG) with clinical metadata.
Retrospective analysis of ECG and clinical metadata from two teaching hospitals
Training a deep neural network (EfficientNet B3) to predict LVEF
Assessment of model performance using area under the curve (AUC) and coefficient of determination (R2)
AI model achieved AUC of 0.95 with integrated data; ECG alone had AUC of 0.90
Integration of clinical metadata improved prediction accuracy, especially for heart failure with reduced ejection fraction (HFrEF)
Model specificity was high at 96.9%, but sensitivity was lower at 54.8%, indicating targeted screening potential

Structured PICO

Does integrating clinical metadata (age, NT-proBNP, sodium) with ECG using an AI model improve the prediction of HFrEF compared to ECG alone in patients undergoing echocardiography?

Population

33,800 patients (27,587 in training/validation cohort, 6,213 in test cohort) from two tertiary teaching hospitals in Korea, who had left ventricular ejection fraction (LVEF) measured via transthoracic echocardiography within 7 days of an ECG test.

Intervention

Deep neural network (EfficientNet B3) artificial intelligence model integrating 12-lead ECG data with clinical metadata (age, NT-proBNP, and serum sodium).

Comparator

Deep neural network artificial intelligence model relying on ECG data alone.

Outcome

Prediction accuracy for detecting Heart Failure with reduced Ejection Fraction (HFrEF, LVEF ≤ 40%), assessed by the Area Under the Curve (AUC).surrogate

Integrating minimal clinical metadata (age, NT-proBNP, sodium) with ECG data using a deep neural network significantly improves the accuracy of detecting HFrEF compared to using ECG alone.

Main Result

Absolute Event Rate: 0% vs 0%

Abstract

Background/Aims: Left ventricular ejection fraction (LVEF) is a key echocardiographic parameter for assessing LV systolic function, guiding the management of many cardiovascular diseases, including heart failure (HF). While traditional electrocardiography (ECG) has been widely used in clinical practice, it has limitations in predicting LVEF. This study investigated the impact of integrating ECG data with metadata, such as age, N-terminal pro B-type natriuretic peptide (NT-proBNP), and sodium levels, to enhance the accuracy of LVEF prediction, especially in HF with reduced ejection fraction (HFrEF, LVEF ≤ 40%).Methods: This retrospective study analyzed ECG and metadata from two tertiary teaching hospitals in Korea. A deep neural network (EfficientNet B3) was trained to predict LVEF, incorporating clinical metadata alongside ECG inputs. Model performance was assessed using the area under the curve (AUC) and the coefficient of determination (R2).Results: The artificial intelligence (AI) model achieved an AUC of 0.95 when ECG data were combined with age, NT-proBNP, and sodium levels, outperforming models relying on ECG alone (AUC = 0.90). The integration of metadata significantly improved the prediction accuracy, particularly for HFrEF cases. The specificity of the model remained high (96.9%), but sensitivity was relatively low (54.8%), indicating its potential as a screening tool for HFrEF.Conclusions: The combination of ECG and metadata results using AI enhances the predictive accuracy of HFrEF detection. This approach offers a scalable and noninvasive method for HF screening and risk stratification, particularly in resource-limited settings. Further validation in diverse populations is needed to confirm its clinical utility.

Ask AI

Helpful

Bookmark

View Full Paper