Key result
Single-frame echocardiogram AI accurately classifies LVEF ≤40% with an AUC >0.90.
Why the study?
Previous AI models estimating LVEF have typically used video clips as input, which can be computationally intensive, motivating development of a model using static frames.
Does a two-dimensional convolutional neural network model accurately classify LVEF ≤40% versus >40% using single static frames from echocardiogram videos?
Observational (n=36,676)
Yes
Does a two-dimensional convolutional neural network model accurately classify LVEF ≤40% versus >40% using single static frames from echocardiogram videos?
Effect estimate: AUC >0.90
An AI model using single static echocardiogram frames can accurately classify LVEF ≤40% versus >40%, offering a computationally efficient alternative to video-based models for point-of-care practice.
May support efficient LVEF screening from static frames in expert settings; leaves open prospective validation before clinical adoption.
BACKGROUND: Artificial intelligence (AI) is poised to transform point-of-care practice by providing rapid snapshots of cardiac functioning. Although previous AI models have been developed to estimate left ventricular ejection fraction (LVEF), they have typically used video clips as input, which can be computationally intensive. In the current study, we aimed to develop an LVEF estimation model that takes in static frames as input. METHODS: Using retrospective transthoracic echocardiography (TTE) data from Mayo Clinic Rochester and Mayo Clinic Health System sites (training: n=19 627; interval validation: n=862), we developed a two-dimensional convolutional neural network model that provides an LVEF estimate associated with an input frame from an echocardiogram video. We then evaluated model performance for Mayo Clinic TTE data (Rochester, n=1890; Arizona, n=1695; Florida, n=1862), the EchoNet-Dynamic TTE dataset (n=10 015), a prospective cohort of patients from whom TTE and handheld cardiac ultrasound (HCU) were simultaneously collected (n=625), and a prospective cohort of patients from whom HCU clips were collected by expert sonographers and novice users (n=100, distributed across three external sites). FINDINGS: We observed consistently strong model performance when estimates from single frames were averaged across multiple video clips, even when only one frame was taken per video (for classifying LVEF ≤40% vs LVEF>40%, area under the receiver operating characteristic curve [AUC]>0·90 for all datasets except for HCU clips collected by novice users, for which AUC>0·85). We also observed that LVEF estimates differed slightly depending on the phase of the cardiac cycle when images were captured. INTERPRETATION: When aiming to rapidly deploy such models, single frames from multiple videos might be sufficient for LVEF classification. Furthermore, the observed sensitivity to the cardiac cycle offers some insights on model performance from an explainability perspective. FUNDING: Internal institutional funds provided by the Department of Cardiovascular Medicine, Mayo Clinic, Rochester, MN, USA.
No takes yet. Share an insight, caveat, or question.
Malins et al. (2025) conducted an observational in Left ventricular ejection fraction estimation (n=36,676). Two-dimensional convolutional neural network model for LVEF estimation from single frames was evaluated on Classifying LVEF ≤40% vs LVEF>40% (AUC >0.90). An AI model using single frames from echocardiogram videos accurately classified LVEF ≤40% versus >40%, achieving an AUC >0.90 across most datasets and >0.85 for novice user clips.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: