Key result
Machine learning of single-lead ECG and pulse wave data detects reduced EF with ~0.93 AUC.
Its licence doesn't allow us to show a copy here.
Why the study?
Analysis of a single-channel electrocardiogram may be used as a screening method to detect systolic and diastolic left ventricular dysfunction, but effective machine learning-based methods were needed.
Does a machine learning model based on single-channel ECG and pulse wave recording accurately detect left ventricular systolic and diastolic dysfunction in adult patients?
Observational (n=375)
Does a machine learning model based on single-channel ECG and pulse wave recording accurately detect left ventricular systolic and diastolic dysfunction in adult patients?
Effect estimate: AUC 0.931
Machine learning algorithms applied to single-lead ECG and pulse wave data can accurately screen for left ventricular systolic and diastolic dysfunction, potentially simplifying early heart failure detection.
May support wearable screening for reduced EF and diastolic dysfunction; hypothesis-generating pending prospective validation.
Background Analysis of a single-channel electrocardiogram can potentially be used as a screening method to detect systolic and diastolic dysfunction of the left ventricle . The purpose of our study was to develop a new screening method for detecting a decrease in systolic and/or diastolic function of the left ventricle based on single-channel ECG and pulse wave recording using machine learning methods. Materials and methods The study prospectively included 375 patients aged 18 years and above. Every participant underwent a transthoracic echocardiographic study with the aim of determining the systolic (EF below 52% for male and 54% for female) and diastolic dysfunction of the left ventricle, as well as, registering the 3 min single channel electrocardiogram and pulse wave using a portative electrocardiogram device. Spectral analysis of the electrocardiogram was performed which was based on the Fourier transform . Results For ejection fraction decrease: Lasso regression showed a sensitivity of 92,3%, specificity of 90,2% (AUC = 0.922); Random Forest Classifier sensitivity − 100%, specificity − 84,3% (AUC = 0.931). For diastolic dysfunction : Lasso regression sensitivity − 86,2%, specificity − 84,5% (AUC 0.878); random forest classifier sensitivity − 82,8%, specificity 89,3% (AUC 0.854). Algorithm approbation has shown diagnostic accuracy of 95,6% left ventricular diastolic dysfunction of 2–3 grade. Conclusions Machine learning models, based on the single lead ECG and pulse wave parameters, as well as age and gender may simplify screening diagnostics of ejection fraction decrease and diastolic dysfunction prior to echocardiographic study for in time heart failure diagnostics with high accuracy.
No takes yet. Share an insight, caveat, or question.
Kuznetsova et al. (2023) conducted an observational in Left ventricular dysfunction (n=375). Machine learning models based on single-channel ECG and pulse wave recording vs. Transthoracic echocardiography was evaluated on Detection of ejection fraction decrease (EF <52% for male and <54% for female) (AUC 0.931). Machine learning models based on single-lead ECG and pulse wave parameters detected ejection fraction decrease with an AUC up to 0.931 and diastolic dysfunction with an AUC up to 0.878.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: