Key result
An automated machine learning model for tracking the tricuspid annulus on echocardiography provided good diagnostic performance for right ventricular dysfunction (AUC 0.69-0.73) compared to CMR.
Why the study?
Current echocardiographic techniques for right ventricular assessment require additional imaging and manual analysis, highlighting the need for efficient, fully automated quantification using machine learning.
Does an automated machine learning model for tricuspid annulus tracking on echocardiography accurately quantify right ventricular function compared to CMR?
Cross-Sectional (n=101)
Does an automated machine learning model for tricuspid annulus tracking on echocardiography accurately quantify right ventricular function compared to CMR?
Effect estimate: AUC 0.69-0.73
A novel machine learning model for automated tricuspid annulus tracking on echocardiography provides rapid, reproducible, and accurate assessment of right ventricular function compared to CMR.
May support reproducible RV assessment on echocardiography; leaves open prospective outcome validation before clinical use.
PURPOSE: Echocardiography (echo) is widely used for right ventricular (RV) assessment. Current techniques for RV evaluation require additional imaging and manual analysis; machine learning (ML) approaches have the potential to provide efficient, fully automated quantification of RV function. METHODS: An automated ML model was developed to track the tricuspid annulus on echo using a convolutional neural network approach. The model was trained using 7791 image frames, and automated linear and circumferential indices quantifying annular displacement were generated. Automated indices were compared to an independent reference of cardiac magnetic resonance (CMR) defined RV dysfunction (RVEF < 50%). RESULTS: A total of 101 patients prospectively underwent echo and CMR: Fully automated annular tracking was uniformly successful; analyses entailed minimal processing time (<1 second for all) and no user editing. Findings demonstrate all automated annular shortening indices to be lower among patients with CMR-quantified RV dysfunction (all P < .001). Magnitude of ML annular displacement decreased stepwise in relation to population-based tertiles of TAPSE, with similar results when ML analyses were localized to the septal or lateral annulus (all P ≤ .001). Automated segmentation techniques provided good diagnostic performance (AUC 0.69-0.73) in relation to CMR reference and compared to conventional RV indices (TAPSE and S') with high negative predictive value (NPV 84%-87% vs 83%-88%). Reproducibility was higher for ML algorithm as compared to manual segmentation with zero inter- and intra-observer variability and ICC 1.0 (manual ICC: 0.87-0.91). CONCLUSIONS: This study provides an initial validation of a deep learning system for RV assessment using automated tracking of the tricuspid annulus.
No takes yet. Share an insight, caveat, or question.
Beecy et al. (2020) conducted a cross-sectional in Right ventricular dysfunction (n=101). Automated machine learning model for tracking the tricuspid annulus vs. Cardiac magnetic resonance (CMR) and manual segmentation was evaluated on Diagnostic performance for CMR-defined RV dysfunction (RVEF < 50%) (AUC 0.69-0.73). An automated machine learning model for tracking the tricuspid annulus on echocardiography provided good diagnostic performance for right ventricular dysfunction (AUC 0.69-0.73) compared to CMR.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: