Caption Interpretation Automated Ejection Fraction Software
Automated Radiological Image Processing Software. 510(k) K200621, from Caption Health (Brisbane, CA). Product code QIH, class 2.
As of Nov 19, 2026, K200621 is an FDA 510(k) held by Caption Health, substantially equivalent on Jul 22, 2020 after 135 days, with 1 recall tied to it in FDA's recall database.
- Review time: 135 days
- Recalls: 1
- Cited as a predicate by: 1 device
The Caption Interpretation Automated Ejection Fraction software is used to process previously acquired transthoracic cardiac ultrasound images, to store images, and to manipulate and make measurements on images using an ultrasound device, personal computer, or a compatible DICOM-compliant PACS system in order to provide automated estimation of left ventricular ejection fraction.
- EchoMD Automated Ejection Fraction SoftwareK173780 · Bay Labs, Inc. · Jun 14, 2018
- Caption Interpretation Automated Ejection Fraction SoftwareK210747 · Caption Health · Jan 19, 2022
| Started | Recall | Status |
|---|---|---|
| May 19, 2021 | Z-2111-2021The firm is aware of a issue with ultrasound software that results in unintended video frames being included. This could lead to an incorrect automated ejection fraction (AutoEF) and image quality score (IQS) estimates. | Terminated |
- Clever OneK260372 · Ewoosoft Co., Ltd. · Nov 4, 2026
- PeekMed® webK262667 · Peek Health, S.A. · Nov 4, 2026
- CINA-Aneurysm+K262147 · Avicenna.Ai · Oct 19, 2026
- Ligence HeartK261733 · UAB Ligence · Oct 7, 2026
- NeuroDRK262134 · Deep Reasoning Ai, Inc. · Oct 7, 2026
- QGuideCT (1.0)K260523 · Medis Medical Imaging Systems BV · Oct 5, 2026
- syngo.CT Brain QuantificationK260055 · Siemens Medical Solutions USA, Inc. · Sep 25, 2026
- BoneXpertK262390 · Visiana · Sep 22, 2026
- On FDA's list of AI-enabled devicesFDA AI-enabled medical devices
- Document
- 510(k) summary (PDF) ↗
- Type
- Traditional
- Received
- Mar 9, 2020
- Review panel
- Radiology
- Regulation
- 21 CFR 892.2050
One page per 510(k), De Novo and original PMA. Recalls are those FDA's recall database ties to this number; recalls of related devices that cite another number are not shown. Refreshed weekly. Predicates are read by Arca from the text of FDA's 510(k) summary PDFs, most of them scans: a device number counts when the summary names it as a predicate or as the device it is substantially equivalent to. About 97% of those read were right on a sample we checked by hand; summaries from before 2000 often name predicates without a number, so older devices show fewer. Pivotal studies are read by a model from FDA's De Novo decision summaries and PMA SSEDs; a count or result is shown only when its numbers appear in FDA's text. openFDA device 510(k), PMA and recall files