Comparison of Apple Watch vs KardiaMobile: A Tale of Two Devices
Bibliographic record
Abstract
Background: The Apple Watch Series 4 (AW4) and the KardiaMobile single bipolar lead model (KM) are 2 of the most popular US Food & Drug Administration (FDA)-approved commercial heart trackers. However, a lack of knowledge remains regarding their rhythm-detection accuracy in real-life clinical situations. This paper aims to determine the practicality of using an AW4 or a KM in modern medical practice, by assessing the accuracy of each in identifying heart rhythms and heart rate. Methods: Participants from the Toronto Heart Centre clinic were enrolled from January 2019 to December 2019. They had a 12-lead electrocardiogram (ECG), followed by wearing the AW4 watch (OS 5.3), and pressing on the KM electrode plates, within the span of 5 minutes of one another. Each session involved a 12-lead ECG, an ECG from each device, and AW4's photoplethysmography function (APPG). Results: Of 200 participants, 162 (81%) were in sinus rhythm, and 38 (19%) had atrial fibrillation. The rhythm-detection accuracy for sinus rhythm was 100% for the AW4, and 99.03% for the KM. For atrial fibrillation, accuracy was 90.48% for the AW4, and 100% for the KM. The heart rate accuracy for sinus rhythm was 94.39% for the KM, 90.65% for the APPG, and 96.26% for the Apple ECG function. The heart rate accuracy for atrial fibrillation was 91.30% for the KM, 82.61% for the APPG, and 86.96% for the Apple ECG function. Conclusions: Both the AW4 and the KM could reliably detect rhythm and heart rate in real-life clinical situations. However, a nonsignificant trend occurred toward better rhythm detection and accuracy with KM, compared with AW4. The difference is mainly due to artifacts (eg, tremors) and the fit of the strap for AW4. The findings have important implications for how these consumer devices can be used in real-life clinical settings.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.012 | 0.035 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".