Key result
Random forest ECG model detects LVSD with 0.92 AUC, matching deep learning waveform performance.
Why the study?
Deep learning models detect left ventricular systolic dysfunction from ECG waveforms accurately, but are difficult to interpret and deploy broadly in clinical settings.
Do simple models based on standard electrocardiogram measurements detect left ventricular systolic dysfunction (LVEF < 35%) with similar accuracy to deep learning models?
Observational (n=40,994)
Yes
Do simple models based on standard electrocardiogram measurements detect left ventricular systolic dysfunction (LVEF < 35%) with similar accuracy to deep learning models?
Absolute Event Rate: 0.92% vs 0.94%
p-value: p=0.08
Simple, interpretable models based on standard ECG measurements can detect left ventricular systolic dysfunction with accuracy approaching that of complex deep learning models and superior to NT-proBNP.
May support interpretable ECG models for LVSD screening; leaves open prospective validation before clinical adoption.
Importance Deep learning methods have recently gained success in detecting left ventricular systolic dysfunction (LVSD) from electrocardiogram waveforms. Despite their impressive accuracy, they are difficult to interpret and deploy broadly in the clinical setting. Objective To determine whether simpler models based on standard electrocardiogram measurements could detect LVSD with similar accuracy to deep learning models. Design Using an observational dataset of 40,994 matched 12-lead electrocardiograms (ECGs) and transthoracic echocardiograms, we trained a range of models with increasing complexity to detect LVSD based on ECG waveforms and derived measurements. We additionally evaluated models in two independent cohorts from different medical centers, vendors, and countries. Setting The training data was acquired from Stanford University Medical Center. External validation data was acquired from Cedars-Sinai Medical Center and the UK Biobank. Exposures The performance of models based on ECG waveforms in their detection of LVSD, as defined by ejection fraction below 35%. Main outcomes The performance of the models as measured by area under the receiver operator characteristic curve (AUC) and other measures of classification accuracy. Results The Stanford dataset consisted of 40,994 matched ECGs and echocardiograms, the test set having an average age of 62.13 (17.61) and 55.20% Male patients, of which 9.72% had LVSD. We found that a random forest model using 555 discrete, automated measurements achieves an area under the receiver operator characteristic curve (AUC) of 0.92 (0.91-0.93), similar to a deep learning waveform model with an AUC of 0.94 (0.93-0.94). Furthermore, a linear model based on 5 measurements achieves high performance (AUC of 0.86 (0.85-0.87)), close to a deep learning model and better than NT-proBNP (0.77 (0.74-0.79)). Finally, we find that simpler models generalize better to other sites, with experiments at two independent, external sites. Conclusion Our study demonstrates the value of simple electrocardiographic models which perform nearly as well as deep learning models while being much easier to implement and interpret.
No takes yet. Share an insight, caveat, or question.
Hughes et al. (2024) conducted an observational in Left ventricular systolic dysfunction (LVSD) (n=40,994). Random forest model using 555 discrete automated ECG measurements vs. Deep learning waveform model was evaluated on Detection of LVSD (ejection fraction < 35%) measured by AUC (95% CI 0.91-0.93, p=0.08). A random forest model using 555 discrete automated ECG measurements detected left ventricular systolic dysfunction with an AUC of 0.92, performing similarly to a deep learning waveform model.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: