Philips Ultrasound, LLC · Class II · Cleared May 21, 2025
| K-number | K243793 |
| Device name | EPIQ Series Diagnostic Ultrasound System; Affiniti Series Diagnostic Ultrasound System |
| Applicant | Philips Ultrasound, LLC |
| Product code | IYN |
| Device class | Class II |
| Decision date | May 21, 2025 |
| Decision | Substantially Equivalent |
| Regulation | 892.1550 |
The R-Trigger software application is an AI-based algorithm added to Philips EPIQ and Affiniti diagnostic ultrasound systems to automatically detect R-wave peaks from cardiac ultrasound clips without requiring an ECG signal. It serves as a backup method to enable AutoStrain and AutoMeasure cardiac clinical applications when ECG input is unavailable or unusable, supporting diagnostic ultrasound imaging and fluid flow analysis across numerous anatomical regions and clinical applications.
The R-Trigger algorithm provides a non-ECG-based alternative to the existing ECG-based R-trigger detection method performed by the on-cart physio board. There are no hardware changes; the feature is software-only and operates as a workflow enhancement in the background. When ECG signal is not available, the new algorithm detects R-trigger events for input into AutoMeasure and AutoStrain applications, while ECG-based detection remains the preferred method when available.
IEC 62304 (Medical device software – Software life cycle processes, 2006 + A 2015) and ISO 14971 (Medical devices – Application of risk management to medical devices, 2019). The devices comply with the FDA's February 2023 guidance document 'Marketing Clearance of Diagnostic Ultrasound Systems and Transducers.' Track 3 device classification applies.
The R-Trigger feature maintains identical indications for use, intended users, clinical environments, FDA classification, product codes, and regulations as the predicate (K240850). A retrospective validation study of 7,309 cardiac clips from 3,964 subjects demonstrated that the R-Trigger AI algorithm achieved high agreement with ECG-based R-trigger detection across all 35 pre-defined hypotheses, with mean differences and 95% confidence intervals falling within clinically acceptable ranges for all parameters tested. Performance was consistent across diverse demographics (age, race, gender, BMI), clinical status, and geographic regions, establishing that the software-only modification does not raise new safety or effectiveness questions.
View the full FDA submission: accessdata.fda.gov