Deep learning-enabled ECG detection of LVH predicted incident AMI, HF, and AFib with hazard ratios of 2.67, 3.15, and 2.23 respectively, comparable to echocardiography-defined LVH in 47,007 patients in Taiwan.
Observational (n=47,007)
Yes
Does a deep learning-enabled ECG system accurately detect left ventricular hypertrophy and predict incident acute myocardial infarction, heart failure, and atrial fibrillation compared to echocardiography?
An AI-enabled 12-lead ECG system can accurately detect left ventricular hypertrophy and provide long-term prognostic risk stratification for cardiovascular events that is comparable to echocardiography.
Effect estimate: Hazard ratios: AMI HR 2.67, HF HR 3.15, AFib HR 2.23 for AI-ECG-LVH compared to non-LVH; Echocardiography LVH HRs: AMI 2.76, HF 3.78, AFib 2.25
Left ventricular hypertrophy (LVH) is a common condition with a prevalence of 15%-20% in general population. Prior studies have suggested that deep learning model (DLM)-enabled electrocardiogram (ECG) systems can aid LVH detection and cardiovascular risk assessment; however, conventional manual ECG criteria have limited sensitivity and their prognostic utility remains suboptimal. Therefore, this study aimed to develop a DLM-enabled ECG system to detect LVH and evaluate its prognostic associations with incident cardiovascular outcomes. A total of 40,736 patients from hospital A were used for model development (training and tuning) and internal validation (29,595/5,935/5,206 patients, respectively), and 6,271 patients from hospital B were used for external validation. LVH was defined by left ventricular mass index (LVMI) derived from echocardiography. Prognostic outcomes included new-onset acute myocardial infarction (AMI), heart failure (HF), and atrial fibrillation (AFib). In the external validation set, our AI-ECG-LVH model achieved area under the receiver operating characteristic curve (AUC) values of 0.82 in males and 0.77 in females. Furthermore, the hazard ratios for incident AMI, HF, and AFib were 2.67, 3.15, and 2.23 for AI-ECG-LVH, compared with 2.76, 3.78, and 2.25 for echocardiography-defined LVH (ECHO-LVH). Our AI-ECG-LVH model may provide a straightforward, affordable, and noninvasive approach for LVH screening and first-contact risk stratification.
Yang et al. (Wed,) conducted a observational in Patients with paired ECG and echocardiographic measurements to detect left ventricular hypertrophy and predict cardiovascular outcomes (n=47,007). Deep learning-enabled ECG system to detect LVH vs. Echocardiography-defined LVH (ECHO-LVH) was evaluated on Incident cardiovascular outcomes including acute myocardial infarction (AMI), heart failure (HF), and atrial fibrillation (AFib) after LVH detection (Hazard ratios: AMI HR 2.67, HF HR 3.15, AFib HR 2.23 for AI-ECG-LVH compared to non-LVH; Echocardiography LVH HRs: AMI 2.76, HF 3.78, AFib 2.25). Deep learning-enabled ECG detection of LVH predicted incident AMI, HF, and AFib with hazard ratios of 2.67, 3.15, and 2.23 respectively, comparable to echocardiography-defined LVH in 47,007 patients in Taiwan.