What happened
Apple's AFib History Feature analyzes pulse data from the Apple Watch optical sensor. It estimates the share of time that a user with atrial fibrillation spent in AFib over the past week. FDA qualified it as a medical device development tool, so makers of cardiac ablation devices can use its weekly estimates as a secondary endpoint in trials. The evidence came from Apple's prospective clinical study of 280 subjects with an ECG patch as the reference. The average difference in weekly burden was 0.67%, and 92.9% of subjects were within ±5%. FDA wrote that developers will be able to capture this information more easily and consistently in their studies. FDA issued its qualification summary in May 2024. In 36 subjects with a prior ablation, the upper limit of agreement exceeded the 15% criterion. The feature does not identify atrial flutter and works only while the watch is worn.
Sources
-
Regulatory document, by authors with no tie to the system.
The performance data summarized here demonstrate the effectiveness of the AFib History Feature to provide clinically acceptable retrospective weekly AFib burden estimates, thus meeting the objective qualification criteria for the AFib History feature as a biomarker test
Details
AI system
- Workflow
-
The feature analyzed beat-to-beat intervals from the Apple Watch photoplethysmography sensor over seven days of wear. A Cardea SOLO wireless ECG patch gave the reference AFib burden.
Follow-up
- Conditions for replication
-
The tool serves only as a secondary endpoint in studies of cardiac ablation devices. The participant wears the watch for at least seven days before an estimate is given.
Entry details
- Sector
-
Healthcare.
- Purpose
-
Apple built the AFib History Feature to give users aged 22 and over with diagnosed atrial fibrillation a retrospective weekly estimate of their AFib burden from Apple Watch pulse data.
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). FDA qualified the AFib History Feature for ablation device trials (AIBD-0810). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0810.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)