Intended Use

Software intended to aid clinicians in identifying individuals with Left Ventricular Ejection Fraction (LVEF) ≤ 40% from ECG and heart sounds in patients at risk for heart failure, including those with coronary artery disease, diabetes, cardiomyopathy, hypertension, and obesity; for use in adults 18 years or older.

Technology

The device uses machine learning algorithms and signal processing to analyze ECG and phonocardiogram signals collected by FDA-cleared Eko Stethoscopes, processing data via a web-based API with deep convolutional neural networks to classify LVEF ≤ 40%; it outputs likelihood notifications for clinical referral, not diagnostic confirmation.

Performance

Performance validation was conducted on a proprietary database with 3,456 paired ECG and heart sound recordings, validated against gold standard echocardiogram measurements. Sensitivity was 74.7% and specificity 77.5% for detecting LVEF ≤ 40%. Testing included significant demographic variation analysis and software verification per IEC 62304 standards.

Predicate Devices

No predicate devices specified

Device Timeline

  • 1

    Submission

    10/6/2023

    5 months
  • 2

    FDA Approval

    3/28/2024

Ready to Sharpen Your Edge?

Join hundreds of your peers who rely on RadAI Slice. Get the essential weekly briefing that empowers you to navigate the future of radiology.

We respect your privacy. Unsubscribe at any time.