Smartwatch AFib Detection Accuracy
Reviewed by Sameer Sontakey, Co-founder and Chief AI & Algorithms Officer at Sensor Bio. ChatPPG is independent and device-neutral; it is not a Sensor Bio product.
The Fitbit Heart Study reported a 98.2% positive predictive value for concurrent AF; the Apple Heart Study reported 0.84. These used different definitions, so they are listed side by side, not ranked.
Which wearable is most accurate for AFib detection?
13 published results from 6 studies in the wearable accuracy database. Each ranking below compares devices only on the same statistic against the same kind of reference. Results from different statistics or references are never ranked together.
Evidence by brand
- Apple Watch4 studies, n = 419622
- OuraNo published validation found
- WHOOPNo published validation found
- GarminNo published validation found
- Fitbit3 studies, n = 455973
- Pixel WatchNo published validation found
- Samsung1 study, n = 201
- PolarNo published validation found
- Withings1 study, n = 201
- AmazfitNo published validation found
- UltrahumanNo published validation found
- COROSNo published validation found
Not enough comparable results to rank devices on AFib detection. The studies report different statistics, or report sensitivity and specificity, which trade off against each other and are listed by study below instead.
Head-to-head results within single studies
Devices tested side by side in one study, on the same people and protocol.
Reported result (not ranked)
PPV of irregular rhythm detection for concurrent AF on ECG patch
Lubitz SA, et al. (2022), Fitbit Heart Study, Circulation. doi:10.1161/CIRCULATIONAHA.122.060291
455,699 participants · Reference: 1-week ambulatory ECG patch · Irregular heart rhythm detection, PPV for concurrent AF
- •Fitbit (PPG algorithm)95% CI 95.5–99.5%98.2%
Reported result (not ranked)
PPV of irregular pulse notification for simultaneous AF on ECG patch
Perez MV, et al. (2019), Apple Heart Study, New England Journal of Medicine. doi:10.1056/NEJMoa1901183
419,297 participants · Reference: ECG patch worn up to 7 days · Irregular pulse notification, PPV for AF on simultaneous ECG patch
- •Apple Watch (PPG irregular rhythm)Brand guide95% CI 0.76–0.920.84
Reported result (not ranked)
Automated AF read vs 12-lead ECG (interpretable recordings)
Bumgarner JM, et al. (2018), Journal of the American College of Cardiology. doi:10.1016/j.jacc.2018.03.003
100 patients presenting for cardioversion · Reference: Physician-interpreted 12-lead ECG · Kardia Band (single-lead ECG on Apple Watch) automated AF read
- •Kardia Band on Apple Watch (single-lead ECG)Sens 93% / Spec 84%57 of 169 recordings non-interpretable
All published results
Every AFib result in the database, including statistics that cannot be ranked. Download the full table as CSV for all rows. From the wearable accuracy database (v1.0.0).
Atrial fibrillation detection6 studies
| Device tested | Result | Reference | Sample | Study |
|---|---|---|---|---|
| Apple Watch Series 6smartwatch ECG app, inconclusive tracings counted | 85% (spec 75%)Sensitivity | 12-lead ECG | n = 201, Tertiary cardiology service, single-lead ECG algorithms | Mannhart et al. (2023)Funding not reported |
| Apple Watch Series 6smartwatch ECG app, inconclusive tracings counted | 75%Specificity | 12-lead ECG | n = 201, Tertiary cardiology service, single-lead ECG algorithms | Mannhart et al. (2023)Funding not reported |
| Samsung Galaxy Watch3smartwatch ECG app, inconclusive tracings counted | 85% (spec 75%)Sensitivity | 12-lead ECG | n = 201, Tertiary cardiology service, single-lead ECG algorithms | Mannhart et al. (2023)Funding not reported |
| Samsung Galaxy Watch3smartwatch ECG app, inconclusive tracings counted | 75%Specificity | 12-lead ECG | n = 201, Tertiary cardiology service, single-lead ECG algorithms | Mannhart et al. (2023)Funding not reported |
| Withings ScanWatchsmartwatch ECG app, inconclusive tracings counted | 58% (spec 75%)Sensitivity | 12-lead ECG | n = 201, Tertiary cardiology service, single-lead ECG algorithms | Mannhart et al. (2023)Funding not reported |
| Withings ScanWatchsmartwatch ECG app, inconclusive tracings counted | 75%Specificity | 12-lead ECG | n = 201, Tertiary cardiology service, single-lead ECG algorithms | Mannhart et al. (2023)Funding not reported |
| Fitbit Sensesmartwatch ECG app, inconclusive tracings counted | 66% (spec 79%)Sensitivity | 12-lead ECG | n = 201, Tertiary cardiology service, single-lead ECG algorithms | Mannhart et al. (2023)Funding not reported |
| Fitbit Sensesmartwatch ECG app, inconclusive tracings counted | 79%Specificity | 12-lead ECG | n = 201, Tertiary cardiology service, single-lead ECG algorithms | Mannhart et al. (2023)Funding not reported |
| Apple Watch + KardiaBand (SmartRhythm 2.0)AF episodes ≥ 1 h | 97.5% episode sensitivity (PPV 39.9%)Sensitivity | Insertable cardiac monitor | n = 24, Ambulatory, 31,348.9 h of recordings | Wasserlauf et al. (2019)Funding not reported |
| Apple Watch (irregular rhythm notification)PPG irregular pulse notification | PPV 0.84 (95% CI 0.76–0.92)Positive predictive value | ECG patch | n = 419297, Siteless pragmatic study; ECG patch after notification | Perez et al. (2019)Maker funded |
| Fitbit (PPG irregular rhythm algorithm)PPG irregular rhythm detection | PPV 98.2% (95% CI 95.5–99.5%)Positive predictive value | ECG patch | n = 455699, Remote trial; 1-week ECG patch after detection | Lubitz et al. (2022)Maker funded |
| Kardia Band on Apple Watch (single-lead ECG)automated ECG read; 57 of 169 recordings non-interpretable | Sens 93% / spec 84% (interpretable recordings)Sensitivity | 12-lead ECG | n = 100, Before and after cardioversion | Bumgarner et al. (2018)Funding not reported |
- Apple Watch Series 685% (spec 75%)
Sensitivity, smartwatch ECG app, inconclusive tracings counted · vs 12-lead ECG · n = 201, Tertiary cardiology service, single-lead ECG algorithms
Mannhart et al. (2023) Funding not reported
Specificity, smartwatch ECG app, inconclusive tracings counted · vs 12-lead ECG · n = 201, Tertiary cardiology service, single-lead ECG algorithms
Mannhart et al. (2023) Funding not reported
- Samsung Galaxy Watch385% (spec 75%)
Sensitivity, smartwatch ECG app, inconclusive tracings counted · vs 12-lead ECG · n = 201, Tertiary cardiology service, single-lead ECG algorithms
Mannhart et al. (2023) Funding not reported
Specificity, smartwatch ECG app, inconclusive tracings counted · vs 12-lead ECG · n = 201, Tertiary cardiology service, single-lead ECG algorithms
Mannhart et al. (2023) Funding not reported
- Withings ScanWatch58% (spec 75%)
Sensitivity, smartwatch ECG app, inconclusive tracings counted · vs 12-lead ECG · n = 201, Tertiary cardiology service, single-lead ECG algorithms
Mannhart et al. (2023) Funding not reported
Specificity, smartwatch ECG app, inconclusive tracings counted · vs 12-lead ECG · n = 201, Tertiary cardiology service, single-lead ECG algorithms
Mannhart et al. (2023) Funding not reported
- Fitbit Sense66% (spec 79%)
Sensitivity, smartwatch ECG app, inconclusive tracings counted · vs 12-lead ECG · n = 201, Tertiary cardiology service, single-lead ECG algorithms
Mannhart et al. (2023) Funding not reported
- Fitbit Sense79%
Specificity, smartwatch ECG app, inconclusive tracings counted · vs 12-lead ECG · n = 201, Tertiary cardiology service, single-lead ECG algorithms
Mannhart et al. (2023) Funding not reported
- Apple Watch + KardiaBand (SmartRhythm 2.0)97.5% episode sensitivity (PPV 39.9%)
Sensitivity, AF episodes ≥ 1 h · vs Insertable cardiac monitor · n = 24, Ambulatory, 31,348.9 h of recordings
Wasserlauf et al. (2019) Funding not reported
- Apple Watch (irregular rhythm notification)PPV 0.84 (95% CI 0.76–0.92)
Positive predictive value, PPG irregular pulse notification · vs ECG patch · n = 419297, Siteless pragmatic study; ECG patch after notification
Perez et al. (2019) Maker funded
- Fitbit (PPG irregular rhythm algorithm)PPV 98.2% (95% CI 95.5–99.5%)
Positive predictive value, PPG irregular rhythm detection · vs ECG patch · n = 455699, Remote trial; 1-week ECG patch after detection
Lubitz et al. (2022) Maker funded
- Kardia Band on Apple Watch (single-lead ECG)Sens 93% / spec 84% (interpretable recordings)
Sensitivity, automated ECG read; 57 of 169 recordings non-interpretable · vs 12-lead ECG · n = 100, Before and after cardioversion
Bumgarner et al. (2018) Funding not reported
1 more result in the database.
Read before comparing
- Not ranked: PPV depends on how common AF is in each population and on the definition of a positive alert.
- Kardia Band is an ECG feature, not PPG, and was tested in patients already known to have AF.
- Rankings combine studies only when they report the same statistic against the same kind of reference, and still differ in protocol and population. Treat close values as ties.
Download the wearable accuracy table (CSV)
Every leaderboard row with study, DOI, reference standard and sample size. We'll email you the link.
FAQ
Which smartwatch is best at detecting AFib?
No head-to-head study exists. Separately, Fitbit reported 98.2% PPV (Lubitz 2022) and Apple reported 0.84 PPV (Perez 2019) under different protocols.
Can a smartwatch diagnose AFib?
No. In the Apple Heart Study only 34% of notified participants had AF on a later ECG patch; alerts need ECG confirmation by a clinician.