Caption Interpretation Automated Ejection Fraction Software
Caption Health, IncCaption Interpretation Automated Ejection Fraction Software is an AI-powered tool that processes cardiac ultrasound images to automatically estimate the left ventricular ejection fraction, a critical measure of heart function. It helps clinicians quickly and accurately evaluate cardiac health by analyzing ultrasound video clips, selecting the best quality views, and calculating ejection fraction with confidence metrics. This supports clinical decision-making in adult cardiac evaluations.
Intended Use
Used to process previously acquired transthoracic cardiac ultrasound images to provide automated estimation of left ventricular ejection fraction to assist clinicians in cardiac evaluation in adult patients.
Technology
Software-only device employing deep learning algorithms to classify ultrasound video clips (PLAX, AP4, AP2 views), assess image quality, and estimate left ventricular ejection fraction with associated confidence metrics. Operates as an add-in to DICOM PACS, ultrasound device, or personal computer. Includes a predetermined change control plan (PCCP) for future algorithm improvements.
Performance
Validated on 186 patient studies in a retrospective multicenter study. Achieved strong performance with root mean square deviation (RMSD) of approximately 7.2% between automated and expert ejection fraction measurements, exceeding predefined acceptance criteria. Demonstrated high clip annotation accuracy (>97% positive predictive value and >96% sensitivity). Confidence metrics for expected error were verified. Performance was consistent across patient demographics, ultrasound systems, and cardiac function subgroups.
Device Timeline
- 1
Submission
9/28/2022
4 months - 2
FDA Approval
2/24/2023
Other devices from Caption Health, Inc
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.