AI-based ECG predicts biventricular dysfunction, dilation in CHD

An ECG enhanced by artificial intelligence (AI) demonstrates its capability in detecting biventricular dysfunction and dilation in patients with congenital heart disease (CHD).

A team of investigators trained and tested a convolutional neural network on paired ECG-cardiovascular magnetic resonance (CMR) ≤30 days apart from patients with and without CHD to diagnose left ventricular (LV) dysfunction (ejection fraction ≤40 percent), right ventricular (RV) dysfunction (ejection fraction ≤35 percent), and LV and RV dilation (end-diastolic volume z-score ≥4).

The investigators performed internal testing and external validation on an outside healthcare system to assess performance using area under receiver-operating curve (AUROC) and are under precision recall curve. The internal and external cohorts consisted of 8,584 ECG-CMR pairs (n=4,941; median CMR age 20.7 years) and 909 ECG-CMR pairs (n=746; median CMR age 25.4 years), respectively.

Model performance was comparable for internal (AUROC: LV dysfunction, 0.87; LV dilation, 0.86; RV dysfunction, 0.88; RV dilation, 0.81) and external validation (AUROC: LV dysfunction, 0.89; LV dilation, 0.83; RV dysfunction, 0.82; RV dilation, 0.80), and lowest in functionally single ventricular patients.

Patients with tetralogy of Fallot who were predicted to have a high risk of ventricular dysfunction had poorer survival (p<0.001).

In addition, “[m]odel explainability via saliency mapping revealed that lateral precordial leads influence all outcome predictions, with high-risk features including QRS widening and T-wave inversions for RV dysfunction/dilation,” according to the investigators.