What happened
Avive Solutions of San Francisco makes the Avive AED, a portable automated external defibrillator that guides rescuers by voice. Its rhythm detector is a neural network that decides whether a heart rhythm should be shocked. Avive developed it on separate training, validation and test sets drawn from more than 23,000 ECG recordings. FDA reviewers report results on the test set, one record per person, against goals from the American Heart Association. Sensitivity was 98.9% for coarse ventricular fibrillation against a goal above 90%. For rapid ventricular tachycardia it was 99.0% against a goal above 75%. Specificity was 100% for normal sinus rhythm and 99.3% for other rhythms that should not be shocked. FDA approved the AED on 31 October 2022. The evidence for rhythm detection comes from bench tests on recordings. No clinical study of the device in cardiac arrest is reported, and demographic information on the recordings is limited.
Sources
-
Regulatory document, by authors with no tie to the system.
U.S. Food and Drug Administration, Center for Devices and Radiological Health. Summary of Safety and Effectiveness Data (SSED), P210015: Avive Automated External Defibrillator (AED) System. Approved 2022-10-31.
The bench testing demonstrates that the rhythm recognition detector exceeds both the AHA recommendations and the IEC 60601-2-4 standard for the basic safety and essential performance of cardiac defibrillators, for sensitivity and specificity for detecting shockable and non-shockable arrythmias.
Details
AI system
- Workflow
-
The network was developed on non-overlapping training, validation and test sets from more than 23,000 recordings. Test recordings came from defibrillators with multifunction electrodes, one record per person.
Follow-up
- Conditions for replication
-
The patient is unresponsive and not breathing normally. Rescuers use Child Mode for children under 8 years old or under 55 pounds.
Entry details
- Sector
-
Healthcare.
- Purpose
-
Avive Solutions built the Avive AED to let trained and untrained rescuers treat sudden cardiac arrest, with software that decides whether to deliver a shock.
Not yet checked: model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.
Cite this record
Peter H. Vartanian. (2026). Avive AED exceeded AHA goals for spotting shockable rhythms (AIBD-0790). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0790.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)