Back to Evidence Hub
Case Study S044
2023 Release

Ebrahimkhani 2023: SCG for Aortic Valve Stenosis

D. Ebrahimkhani et al.
Access Paper

Quick Conclusion: Demonstrates the capability of SCG to capture high-fidelity hemodynamic information previously thought to require advanced imaging.


📊 Key Accuracy Metrics

MetricResult
AVS Detection (AUC)0.99
Peak Velocity Correlation (r)0.85
Mean Pressure Gradient Correlation (r)0.83

🔍 Study Analysis

Objective & Population

Technical Development / Deep Learning Validation. Cohort: Patients with Aortic Valve Stenosis (AVS) (N=22) and controls (N=10) (N=32).

What it Supports

The study supports the use of SCG and deep learning to non-invasively detect Aortic Valve Stenosis and predict complex hemodynamic metrics (like peak velocity and pressure gradients) usually obtained via 4D Flow MRI.

What it Does Not Support

The study does not support the replacement of clinical MRI for all patients but demonstrates SCG as a low-cost screening and monitoring alternative.


🛠 Technical Context

Study Snapshot

Metadata Summary

Target Population

Patients with Aortic Valve Stenosis (AVS) (N=22) and controls (N=10)

N

Sample Size

32 Subjects

Validated Metric

0.99

Critical Appraisal
supporting

Validated the use of SCG and Deep Learning to predict aortic hemodynamics and detect Aortic Valve Stenosis.