Forecasting left ventricular systolic dysfunction in heart failure with artificial intelligence

Background

Objective assessment of left ventricular function remains a key prognosticator that is used to guide therapeutic decisions for patients with heart failure (HF). However, the Left ventricular ejection fraction (LVEF) is dynamic, with worsening LVEF linked to increased morbidity and mortality. Identifying patients at risk of LVEF decline would improve prognostication and enable timely therapeutic intervention.

Methods

We developed a deep learning model to P redict changes in left ventric UL ar S ystolic function from E CGs of patients who have H eart F ailure (PULSE-HF). The model integrates 12-lead ECG waveforms with a patient’s history of prior LVEF measurements to calculate the likelihood that the LVEF will be less than 40% during the year after the ECG is obtained. The model is retrospectively developed and tested using data from one hospital and validated on retrospective cohorts from two different hospitals.

Findings

PULSE–HF demonstrates strong discriminatory ability with respect to forecasting whether the LVEF would fall below 40% within the next year, achieving AUROCs of 92% across all three HF cohorts. Among HF patients with baseline LVEF above 40%, PULSE–HF effectively identified those at risk of worsening LVEF with AUROCs of 89%, across all three datasets. PULSE–HF’s discriminatory ability remained consistently high across a range of subgroups with different comorbidities and regardless of medical therapy. Assuming an underlying prevalence of LVEF worsening of 10% per year, PULSE-HF negative predictive values over 97% assuming an underlying sensitivity of 80%. Lastly, we demonstrate that a lead I version of PULSE–HF has a performance virtually indistinguishable from the model that uses all 12 ECG leads

Interpretation

PULSE–HF robustly predicts worsening LVEF in patients who have a prior diagnosis of HF. The method provides a platform for identifying patients who are at increased risk of worsening systolic dysfunction.

Funding

This work was supported, in part, by a grant from Quanta Computers.

Research in context

Evidence before this study:

Heart failure (HF) remains a leading cause of morbidity and mortality across the globe. Left ventricular ejection fraction (LVEF) plays an important role in prognostication and treatment of HF. Moreover, LVEF is dynamic, with worsening LVEF being associated with increased mortality and poor outcomes compared to HF patients with stable LVEF. Identifying HF patients at risk of future decline in LVEF would improve prognostication and facilitate the timely delivery of appropriate medical therapy. However, models that use the electronic health record (EHR) to forecast declines in LVEF have demonstrated modest discriminatory ability. For instance, a model incorporating echocardiographic features, demographics, and comorbidities from the EHR had AUROCs between 54%–68% for predicting a 30% decline in LVEF within one year after HF diagnosis. Electrocardiograms (ECG) may contain information about LVEF, as deep learning models of ECGs have been used to estimate the contemporaneous LVEF. Nevertheless, to date, the ECG has not been used to forecast forecast future changes in LVEF have not been described. Overall, no robust method exists for identifying HF patients at high risk of future LVEF decline.

Added value of this study:

We developed a deep learning model, called PULSE–HF, to P redict changes in left ventric UL ar S ystolic function from E CGs of patients who have H eart F ailure. The model leverages both the 12-lead ECG and history of prior LVEF measurements to identify patients at risk of having a reduced LVEF (<40%) within the year after the ECG was obtained. We also developed a single lead version of PULSE-HF, based on lead I, that has similar performance relative to the model that uses all 12 leads.

Implications of all the available evidence:

The ECG contains information that can be leveraged to forecast changes in LVEF. As ECGs are typically obtained for HF patients during yearly follow-up visits, we envision that the method could be used annually to identify patients who are at the highest risk of future LVEF decline. Such screening could be routinely integrated into existing clinical workflow with minimal changes to clinical practice. The prospect that similar performance can be achieved with a single lead model (lead-I) suggests that it may be possible to do such screening using wearable or pocket ECG monitoring devices. Prospective studies are needed to better define the utility of PULSE–HF in guiding HF management decisions, optimizing prediction windows, and integrating the model into clinical practice.