Pith. sign in

REVIEW 4 major objections 6 minor 55 references

Effect of Photoplethysmogram Artifacts on Uncertainty in Atrial Fibrillation Detection

T0 review · 4 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper demonstrates that PPG motion artifacts affect atrial-fibrillation detectors in opposite ways: the feature-based detector is more robust to short artifacts but becomes increasingly uncertain as artifacts lengthen, while the…

desk verdict A genuinely systematic simulation study of how artifact type and duration affect uncertainty in two PPG-based AF detectors, though the headline raw-vs-feature asymmetry may be partly baked into the artifact insertion method and the signal-quality feature. read the letter →

arxiv 2608.09204 v1 pith:L2HMULWN submitted 2026-08-10 eess.SP

Andrius Sološenko , Enikö Vargova , Loïc Coquelin , Vaidotas Marozas , Andrius Petrėnas This is my paper · ORCID
classification eess.SP
keywords atrialfibrillationdetectionphotoplethysmogramPPGartifactsuncertaintyquantificationconformalpredictionMonteCarlodropoutwearablesensorsdeeplearning
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper sets out to establish that motion artifacts in wrist photoplethysmogram (PPG) signals change the uncertainty of atrial fibrillation (AF) detectors in a detector-specific and artifact-specific way. On clean signals the feature-based detector $D_f$ outperforms the raw-signal convolutional network $D_r$ (sensitivity/specificity 0.94/0.94 versus 0.92/0.86), but this ordering depends on what corrupts the signal. Under 12-s device displacement, $D_r$'s sensitivity drops by 0.52 while $D_f$'s drops by 0.31; $D_f$ is steadier for short artifacts yet shows rising uncertainty with longer artifact durations, whereas $D_r$ degrades abruptly as soon as an artifact appears and stays roughly flat in duration. Device displacement is the most damaging artifact type, and conformal prediction with 90% coverage removes up to 12% of false positives for $D_r$ and up to 64% for $D_f$. If these patterns hold in real wearables, artifact-aware detector selection and uncertainty-based alert suppression could make PPG screening for AF less noisy.

What carries the argument

The load-bearing object is a controlled artifact-injection experiment built on a validated PPG simulator. The simulator generates 25-s AF and non-AF waveforms from beat-to-beat interval series, and four artifact types, device displacement, forearm motion, hand motion, and poor contact, are produced by filtering white noise with spectral slopes and RMS amplitudes fitted to real wrist-PPG artifacts and added for 4, 8, or 12 s to 20,000 baseline signals, yielding 260,000 investigation signals. Uncertainty is measured three ways: threshold-based error rate $E$ at thresholds 0.5, 0.7, and 0.9; conformal prediction, which returns a decision set whose size gives the proportion of uncertain decisions $U$ at coverage $\gamma$; and Monte Carlo dropout, where the detector is run 50 times with dropout active and the average predictive entropy $\bar{H}$ is computed. These measurements are compared against artifact-free baselines to produce the sensitivity/specificity deltas and false-positive reductions that carry the conclusion.

What would settle it

Run both detectors on real wrist-PPG recordings with synchronized ECG in free-living conditions, annotate artifact type and duration, and compare the sensitivity drops and conformal-prediction uncertainty to the simulated values; the central claim would fail if device displacement is not the worst artifact, or if the feature-based detector does not grow more uncertain with artifact duration while the raw-signal CNN stays roughly flat.

Watch

Extended reading notes

Core claim

The central claim is that uncertainty in PPG-based AF detection is not a single property of the signal or detector but a joint property of artifact type, artifact duration, and detector design. The feature-based detector $D_f$, which uses seven rhythm-irregularity features plus a signal quality index, predicts across the full range $[0,1]$ and reacts to longer artifacts by flattening its prediction distribution and increasing the conformal-prediction uncertainty set. The raw-signal CNN $D_r$ produces outputs in a narrower range, so it loses sensitivity immediately when any artifact is present, most severely for device displacement, where sensitivity falls by 0.52 at 12 s, but its uncertainty barely moves with artifact duration. The paper further shows that the proportion of uncertain decisions under conformal prediction depends strongly on the coverage level and the detector's output range, and that suppressing non-singleton decisions reduces false positives substantially for both detectors. The conclusion is that artifact type and duration should be treated as first-class inputs when choosing an AF detector and deciding whether to alert.

Load-bearing premise

The load-bearing premise is that the simulated artifacts, whose spectral slopes and amplitudes were fitted to wrist-PPG recordings in one prior study, faithfully represent real device displacement, forearm motion, hand motion, and poor contact; if real artifacts differ in amplitude, spectral shape, or timing, the measured detector-specific uncertainty differences may not transfer to wearable use.

Editorial extensions

If this is right

  • Device displacement is the artifact that most degrades both detectors, so AF screening pipelines should detect large-amplitude sensor shifts and treat their output as low-confidence rather than making a binary decision.
  • Discarding predictions that conformal prediction marks as uncertain (two-class or empty sets) cuts false positives by up to 12% for the raw-signal CNN and up to 64% for the feature-based detector at 90% coverage.
  • For short artifacts the feature-based detector is the safer choice, while for long artifacts the raw-signal CNN becomes comparatively more reliable, supporting a gating strategy that chooses between the two detectors based on estimated artifact duration.
  • Uncertainty metrics from conformal prediction are not directly comparable across detectors because the raw-signal CNN's narrower output range inflates its uncertain-decision proportion at high coverage; any alert-suppression rule must be calibrated per detector.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: a wearable could estimate artifact duration from accelerometry or signal quality in real time and route each 25-s window to the detector with lower expected uncertainty, rather than running both and arbitrating by confidence alone.
  • Editorial inference: because the feature-based detector is cheaper and its false positives are reduced far more by conformal suppression, an on-device implementation could lean on it during quiet periods and fall back to the convolutional network when motion is sustained.
  • Editorial inference: the simulated artifacts omit physiological fluctuations such as respiration-induced baseline drift and vasomotor changes, so a field test with real annotated artifacts might shift the duration crossover point at which the raw-signal CNN becomes the more reliable detector.
  • Editorial inference: because calibration used artifact-free simulated signals while training used real patient signals, a natural extension is to calibrate conformal prediction on real artifact-contaminated data and measure whether the false-positive reductions change.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper quantifies how four simulated PPG artifact types (device displacement, forearm motion, hand motion, poor contact) and three durations (4, 8, 12 s) affect the uncertainty of two AF detectors: a CNN operating on 25-s raw PPG signals (Dr) and a feed-forward network operating on engineered rhythm-irregularity features plus a signal quality index (Df). The investigation uses 260,000 simulated PPG signals generated with an extended version of the authors' open-access PPG simulator. Uncertainty is assessed with a threshold-based error rate, conformal prediction, and Monte Carlo dropout. The central claim is that Df is more robust to short artifacts but becomes increasingly uncertain as artifact duration grows, whereas Dr exhibits an abrupt performance drop when artifacts occur but is less sensitive to artifact duration; device displacement causes the largest increase in uncertainty. The paper further proposes a dual-detector gating strategy and reports that conformal-prediction-based alert suppression reduces false positives by up to 12% for Dr and up to 64% for Df.

Significance. If the reported detector-specific behavior transfers to real wearable use, the findings are practically valuable because they suggest an artifact-aware detector-selection and uncertainty-based alert-suppression strategy for PPG-based AF screening. The study has clear strengths: it systematically varies artifact type and duration, evaluates two architecturally different detectors, uses three complementary uncertainty measures, and candidly discusses limitations such as the simulated nature of the data, the exchangeability assumption in conformal prediction, and the narrow output range of Dr. It also builds on an openly available simulator and prior validated models, which facilitates reproducibility. However, the significance is tempered by the fact that the investigation dataset is entirely simulated and the artifact model parameters come from the authors' own prior work [13] without independent validation on real artifact-contaminated wrist PPG.

major comments (4)
  1. [Sec. 2.2 and Fig. 3] The headline asymmetry between Dr and Df may be a consequence of the artifact insertion procedure rather than detector architecture. In Sec. 2.2, artifacts are generated as filtered noise, zero-padded outside the artifact interval, and added directly to the clean PPG at the same time point, creating rectangular onset and offset discontinuities. A CNN with 64-sample kernels could respond to these sharp edges regardless of artifact length, which would explain Dr's 'abrupt but duration-insensitive' behavior. At the same time, Df uses as an input the signal quality index defined as the proportion of usable signal, a quantity that decreases mechanically with artifact duration, as the Discussion partly acknowledges in Sec. 4. To make the central claim convincing, the authors should run control experiments that separate these effects, for example by applying ramped artifact envelopes, removing the SQI feature from Df, or feeding the SQI to Dr through a late-fusion layer.
  2. [Sec. 2.2, Table 1, and Discussion] All investigation data are generated by an artifact model whose parameters (spectral slope, gamma amplitude parameters, transition probabilities) are taken from the authors' prior work [13] and are not revalidated against independent real artifact-contaminated wrist PPG. Since the reported ranking of artifact types and the duration dependence are the central quantitative findings, the conclusions are only as strong as this model. The paper should either validate the simulated artifacts against an independent real-world dataset or provide a sensitivity analysis showing that the qualitative conclusions are stable under plausible perturbations of the artifact parameters.
  3. [Sec. 6.2.2 and Figs. 5-6] The conformal prediction analysis calibrates on artifact-free simulated signals but evaluates on artifact-contaminated signals, so the exchangeability assumption that guarantees coverage is violated. The authors acknowledge this in Sec. 4, but the reported '90% coverage' and the false-positive reduction percentages are still presented as quantitative results. Because the coverage property does not formally hold under this distribution shift, U and the false-positive reductions in Figs. 5-6 are descriptive rather than guaranteed. The authors should either apply a distribution-aware method such as weighted conformal prediction, or explicitly report the empirical coverage on the shifted test set and discuss how the false-positive reduction figures would change.
  4. [Figs. 3-6] The central claims about 'abrupt' versus 'progressive' degradation and about detector-specific differences are based on point estimates without confidence intervals or repeated-seed analyses. Although the dataset is large, the underlying signals are simulated and the detectors have stochastic training procedures; the absence of any error bars makes it difficult to judge whether the reported differences between detectors and between 4 s, 8 s, and 12 s artifacts are larger than run-to-run variability. Reporting confidence intervals, or at least a repeated-seed analysis for a subset of conditions, would materially strengthen the qualitative conclusions.
minor comments (6)
  1. [Sec. 2.2] The manuscript states that each artifact is inserted at the same time point but does not specify what that time point is; please state the exact insertion time or justify that the choice does not affect the results.
  2. [Figs. 2, 3, 5, 7] The figures use shades of gray to denote artifact durations, but the mapping between shade and duration is not always given numerically in the figure or caption; a colorbar or explicit legend would improve readability.
  3. [Sec. 3.4 and Table 2] The explanation of the non-monotonic behavior of U for Dr at high coverage levels (γ = 0.99) is clear from the text, but Table 2 would benefit from a column showing the output range of each detector, since the restricted range of Dr is the cause of the effect.
  4. [Sec. 6.1.2] The definition of the signal quality index as 'the proportion of usable signal' depends on the template correlation threshold of 0.7; the authors should state whether the SQI is computed per window or per pulse and how borderline correlations are handled.
  5. [References] Reference [45] in the bibliography lists the publisher as 'Springer, ???'; the missing place or publisher information should be completed.
  6. [Sec. 4] The sentence 'Dr behaves differently, producing less confident predictions for the non-AF class' is somewhat ambiguous because the preceding discussion concerns the range of outputs rather than confidence; consider rewording to distinguish output range from calibration.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity; the central duration-asymmetry finding is measured, not derived from definitions, though the SQI feature is an acknowledged confound.

full rationale

The paper's results are measured behaviors of two fixed detectors under a controlled, simulated artifact protocol; there is no derivation that predicts an outcome from a fitted parameter. The artifact model and PPG simulator are cited from prior work by the same group ([13], [15]), and the detectors come from [6], [43]; this is self-reliance, not circularity, because the cited works supply inputs such as signal models, artifact distributions, and architectures, rather than conclusions. The claim that Df's duration dependence might be driven by the signal quality index input is explicitly acknowledged in Sec. 4: 'Df uses the signal quality index as an input, which may contribute to the gradual increase in detection error for longer artifacts.' However, that is a mechanistic confound, not a circular reduction: SQI is a computed feature that correlates with artifact duration by design, and the detector's mapping from that feature to output uncertainty is learned and measured, not imposed by the paper's definitions. Conformal prediction and Monte Carlo dropout are standard post-hoc estimators applied to fixed detector outputs. The stated limitation that the investigation relies on simulated signals and artifacts is a generalizability concern, not a circular reduction. No equation in the paper is equivalent to its input by construction, and no load-bearing claim is justified solely by a self-citation.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central claim rests on the realism of simulated artifacts and signals, on conformal prediction's exchangeability assumption, and on transfer from a cardiac rehabilitation training cohort; none of these is independently verified within this paper.

free parameters (3)
  • Artifact model spectral slope means and standard deviations = Table 1: device displacement -32.34/6.03; forearm motion -29.39/5.71; hand motion -25.45/4.13; poor contact -18.12/4.10
    Taken from the authors' prior artifact modeling study [13]; these set the frequency content of each simulated artifact and therefore shape the measured detector responses.
  • Artifact amplitude gamma distribution parameters = Table 1: device displacement 0.88/0.04; forearm motion 1.44/0.31; hand motion 1.40/0.45; poor contact 2.01/1.24
    Control artifact amplitude scaling in the simulator; drawn from [13] and not re-estimated here, and they drive the finding that device displacement has the largest effect.
  • Signal quality index correlation threshold = 0.7
    Hand-chosen threshold from prior work [6] that determines the quality feature used by Df; it affects how artifact contamination is quantified and influences the feature detector's behavior.
assumptions (5)
  • domain assumption Artifact simulation parameters and model from [13] produce artifacts representative of real wrist PPG during daily activities.
    All artifact-effect conclusions are computed on 260,000 simulated signals; the authors state that simulations may not cover the full complexity of real-world data (Section 4 limitation).
  • domain assumption Simulated PPG morphology and AF/non-AF beat intervals are sufficiently realistic for detector evaluation.
    Section 2.2 builds the investigation dataset from the PhysioNet simulator [15,16]; real-world transfer depends on this realism claim.
  • domain assumption Calibration and test exchangeability for conformal prediction.
    Section 6.2.2 calibrates only on artifact-free simulated signals yet applies to artifact-contaminated signals; Section 4 acknowledges this violates the formal exchangeability assumption.
  • domain assumption Training data from older cardiac rehabilitation patients transfers to the simulated test population and broader wearable users.
    Section 2.1 describes the training cohort; Section 4 notes that younger or healthier individuals, different skin tones, and other sensors may produce different signal and artifact characteristics.
  • standard math Monte Carlo dropout with K=50 and dropout rate 0.5 provides a meaningful uncertainty estimate.
    Section 6.2.3 uses the Bayesian approximation of dropout from [55]; this is a standard technique, not introduced by the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Effect of Photoplethysmogram Artifacts on Uncertainty in Atrial Fibrillation Detection." pith.science (2026). https://pith.science/paper/L2HMULWN

@misc{pith2026260809204,
  author       = {Pith},
  title        = {Pith review of: Effect of Photoplethysmogram Artifacts on Uncertainty in Atrial Fibrillation Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/L2HMULWN}},
  note         = {Machine review of arXiv:2608.09204}
}
abstract

Detection of atrial fibrillation (AF) from photoplethysmogram (PPG) is highly sensitive to artifacts, yet their effect on uncertainty of different AF detectors remains poorly understood. This work aims to quantify how different PPG artifact types affect the uncertainty of AF detectors. Two machine learning approaches to AF detection were explored: one using 25-s PPG signals as input ($\mathcal{D}_{r}$) and another using AF-related rhythm irregularity features ($\mathcal{D}_{f}$). The detectors were trained on wrist PPG signals acquired during cardiac rehabilitation and then systematically evaluated on 260,000 PPG signals containing controlled artifact types and durations. Uncertainty was quantified using a threshold-based error rate, conformal prediction, and Monte Carlo dropout. Using artifact-free PPG signals, $\mathcal{D}_{f}$ outperforms $\mathcal{D}_{r}$ with sensitivity/specificity of 0.94/0.94 versus 0.92/0.86. Relative to artifact-free performance, sensitivity/specificity drops by 0.52/0.03, 0.25/0.03, 0.16/0.02, and 0.07/0.02 using $\mathcal{D}_{r}$ for 12-s artifacts of device displacement, forearm motion, hand motion, and poor contact respectively. For $\mathcal{D}_{f}$, the respective drops are 0.31/0.07, 0.19/0.11, 0.14/0.15, and 0.13/0.21 for the same artifacts. $\mathcal{D}_{f}$ is more robust to short artifacts but exhibits increasing uncertainty with longer artifact durations, whereas $\mathcal{D}_{r}$ shows an abrupt performance drop when artifacts occur but is less sensitive to artifact duration. Applying conformal prediction with 90\% coverage reduces the false-positive rate by up to 12\% for $\mathcal{D}_{r}$ and up to 64\% for $\mathcal{D}_{f}$. Artifact type and duration have detector-specific effects on AF detection uncertainty. Device displacement causes the largest increase in uncertainty.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

55 extracted references · 51 canonical work pages

  1. [13]

    Paliakaitė, B., Petrėnas, A., Sološenko, A., Marozas, V.: Modeling of artifacts in the wrist photoplethysmogram: Application to the detection of life-threatening arrhythmias. Biomed. Signal Process. Control. 66, 102421 (2021)

    work page 2021

  2. [1]

    NPJ Digit

    Pereira, T., Tran, N., Gadhoumi, K., Pelter, M.M., Do, D.H., Lee, R.J., Colorado, R., Meisel, K., Hu, X.: Photoplethysmography based atrial fibrillation detection: A review. NPJ Digit. Med. 3(1), 3 (2020)

    work page 2020

  3. [2]

    Papalamprakopoulou, Z., Stavropoulos, D., Moustakidis, S., A vgerinos, D., Efremidis, M., Kampaktsis, P.N.: Artificial intelligence-enabled atrial fibrillation detection using smartwatches: Current status and future perspectives. Front. Cardiovasc. Med. 11, 1432876 (2024)

    work page 2024

  4. [3]

    : Large- scale assessment of a smartwatch to identify atrial fibrillation

    Perez, M.V., Mahaffey, K.W., Hedlin, H., Rumsfeld, J.S., Garcia, A., Ferris, T., Balasubramanian, V., Russo, A.M., Rajmane, A., Cheung, L., et al. : Large- scale assessment of a smartwatch to identify atrial fibrillation. N. Engl. J. Med. 381(20), 1909–1917 (2019)

    work page 2019

  5. [4]

    : Mobile photoplethysmographic technology to detect atrial fibrillation

    Guo, Y., Wang, H., Zhang, H., Liu, T., Liang, Z., Xia, Y., Yan, L., Xing, Y., Shi, H., Li, S., et al. : Mobile photoplethysmographic technology to detect atrial fibrillation. J. Am. Coll. Cardiol. 74(19), 2365–2375 (2019)

    work page 2019

  6. [5]

    Lubitz, S.A., Faranesh, A.Z., Selvaggi, C., Atlas, S.J., McManus, D.D., Singer, D.E., Pagoto, S., McConnell, M.V., Pantelopoulos, A., Foulkes, A.S.: Detection of atrial fibrillation in a large population using wearable devices: The Fitbit heart study. Circ. 146(19), 1415–1424 (2022) 18

    work page 2022

  7. [6]

    Sološenko, A., Petrėnas, A., Paliakaitė, B., Sörnmo, L., Marozas, V.: Detection of atrial fibrillation using a wrist-worn device. Physiol. Meas. 40(2), 025003 (2019)

    work page 2019

  8. [7]

    Millán, C.A., Girón, N.A., Lopez, D.M.: Analysis of relevant features from photo- plethysmographic signals for atrial fibrillation classification. Int. J. Environ. Res. Public Health. 17(2), 498 (2020)

    work page 2020

Show all 55 references
  1. [8]

    : Validation of an algorithm for continuous monitoring of atrial fibrillation using a consumer smartwatch

    A vram, R., Ramsis, M., Cristal, A.D., Nathan, V., Zhu, L., Kim, J., Kuang, J., Gao, A., Vittinghoff, E., Rohdin-Bibby, L., et al. : Validation of an algorithm for continuous monitoring of atrial fibrillation using a consumer smartwatch. Heart Rhythm 18(9), 1482–1490 (2021)

    2021

  2. [9]

    Heart Rhythm O2 1(1), 3–9 (2020) https://doi.org/10.1016/j.hroo.2020.02.002

    Aschbacher, K., Yilmaz, D., Kerem, Y., Crawford, S., Benaron, D., Liu, J., Eaton, M., Tison, G.H., Olgin, J.E., Li, Y., Marcus, G.M.: Atrial fibrillation detection from raw photoplethysmography waveforms: A deep learning application. Heart Rhythm O2 1(1), 3–9 (2020) https://do...

    2020 doi

  3. [10]

    Kwon, S., Hong, J., Choi, E.-K., Lee, B., Baik, C., Lee, E., Jeong, E.-R., Koo, B.-K., Oh, S., Yi, Y.: Detection of atrial fibrillation using a ring-type wear- able device (cardiotracker) and deep learning analysis of photoplethysmography signals: Prospective observational pro...

    2020 doi

  4. [11]

    JACC Clin

    Antiperovitch, P., Mortara, D., Barrios, J., A vram, R., Yee, K., Khaless, A.N., Cristal, A., Tison, G., Olgin, J.: Continuous atrial fibrillation monitoring from photoplethysmography: Comparison between supervised deep learning and heuristic signal processing. JACC Clin. Elec...

    2024

  5. [12]

    Bench, C., Strodthoff, N., Moulaeifard, M., Aston, P.J., Thompson, A.: Towards trustworthy atrial fibrillation classification from wearables data: Quantifying model uncertainty. Comput. Cardiol. 51 (2024)

    2024

  6. [14]

    In: International Conference on Bio-inspired Systems and Signal Processing, vol

    Paliakaitė, B., Petrėnas, A., Skibarkienė, J., Mickus, T., Daukantas, S., Kubilius, R., Marozas, V.: Towards long-term monitoring of atrial fibrillation using pho- toplethysmography. In: International Conference on Bio-inspired Systems and Signal Processing, vol. 5, pp. 141–14...

    2017

  7. [15]

    Sološenko, A., Petrėnas, A., Marozas, V., Sörnmo, L.: Modeling of the pho- toplethysmogram during atrial fibrillation. Comput. Biol. Med. 81, 130–138 (2017)

    2017

  8. [16]

    PhysioNet 101, 23 19 (2021)

    Sološenko, A., Petrėnas, A., Paliakaitė, B., Marozas, V., Sörnmo, L.: Model for simulating ECG and PPG signals with arrhythmia episodes. PhysioNet 101, 23 19 (2021)

    2021

  9. [17]

    Petrenas, A., Marozas, V., Sološenko, A., Kubilius, R., Skibarkiene, J., Oster, J., Sörnmo, L.: Electrocardiogram modeling during paroxysmal atrial fibrillation: Application to the detection of brief episodes. Physiol. Meas. 38(11), 2058 (2017)

    2017

  10. [18]

    Plos One 20(2), 0314154 (2025)

    Bulut, M.G., Unal, S., Hammad, M., Pławiak, P.: Deep CNN-based detection of cardiac rhythm disorders using PPG signals from wearable devices. Plos One 20(2), 0314154 (2025)

    2025

  11. [19]

    Xiao, Y., Lu, Y., Liu, M., Zeng, R., Bai, J.: A deep feature fusion network for fetal state assessment. Front. Physiol. 13, 969052 (2022)

    2022

  12. [20]

    : Raw photoplethysmogram waveforms versus peak-to-peak intervals for machine learn- ing detection of atrial fibrillation: Does waveform matter? Comput

    Isaksen, J.L., Arildsen, B., Lind, C., Nørregaard, M., Vernooy, K., Schot- ten, U., Jespersen, T., Betz, K., Hermans, A.N., Kanters, J.K., et al. : Raw photoplethysmogram waveforms versus peak-to-peak intervals for machine learn- ing detection of atrial fibrillation: Does wave...

    2025

  13. [21]

    Hüllermeier, E., Waegeman, W.: Aleatoric and epistemic uncertainty in machine learning: An introduction to concepts and methods. Mach. Learn. 110(3), 457– 506 (2021)

    2021

  14. [22]

    Der Kiureghian, A., Ditlevsen, O.: Aleatory or epistemic? Does it matter? Struct. Saf. 31(2), 105–112 (2009)

    2009

  15. [23]

    PLOS Digit

    Charlton, P.H., Marozas, V., Mejía-Mejía, E., Kyriacou, P.A., Mant, J.: Determi- nants of photoplethysmography signal quality at the wrist. PLOS Digit. Health. 4(6), 0000585 (2025)

    2025

  16. [24]

    Guo, C., Pleiss, G., Sun, Y., Weinberger, K.Q.: On calibration of modern neural networks. In: Int. Conf. Mach. Learn., pp. 1321–1330 (2017). PMLR

    2017

  17. [25]

    Vovk, V., Gammerman, A., Saunders, C.: Machine-learning applications of algorithmic randomness. Int. Conf. Mach. Learn. (1999)

    1999

  18. [26]

    arXiv preprint arXiv:2107.07511 (2021)

    Angelopoulos, A.N., Bates, S.: A gentle introduction to conformal prediction and distribution-free uncertainty quantification. arXiv preprint arXiv:2107.07511 (2021)

    2021 arXiv

  19. [27]

    Tibshirani, R.J., Foygel Barber, R., Candes, E., Ramdas, A.: Conformal predic- tion under covariate shift. Adv. Neural Inf. Process. Syst. 32 (2019)

    2019

  20. [28]

    Yang, Y., Kuchibhotla, A.K., Tchetgen Tchetgen, E.: Doubly robust calibration of prediction sets under covariate shift. J. R. Stat. Soc. Ser. B Stat. Methodol. 86(4), 943–965 (2024) 20

    2024

  21. [29]

    : The 2023 wearable photoplethysmography roadmap

    Charlton, P.H., Allen, J., Bailón, R., Baker, S., Behar, J.A., Chen, F., Clif- ford, G.D., Clifton, D.A., Davies, H.J., Ding, C., et al. : The 2023 wearable photoplethysmography roadmap. Physiol. Meas. 44(11), 111001 (2023)

    2023

  22. [30]

    Pradhan, N., Rajan, S., Adler, A.: Evaluation of the signal quality of wrist-based photoplethysmography. Physiol. Meas. 40(6), 065008 (2019)

    2019

  23. [31]

    Signals 3(2), 249–265 (2022)

    Gasparini, F., Grossi, A., Giltri, M., Bandini, S.: Personalized PPG normaliza- tion based on subject heartbeat in resting state condition. Signals 3(2), 249–265 (2022)

    2022

  24. [32]

    Ding, C., Xiao, R., Do, D.H., Lee, D.S., Lee, R.J., Kalantarian, S., Hu, X.: Log- spectral matching GAN: PPG-based atrial fibrillation detection can be enhanced by GAN-based data augmentation with integration of spectral loss. IEEE J. Biomed. Health Inform. 27(3), 1331–1341 (2023)

    2023

  25. [33]

    Kapuria, S., Minot, P., Kapusta, A., Ikoma, N., Alambeigi, F.: A novel dual layer cascade reliability framework for an informed and intuitive clinician-AI interaction in diagnosis of colorectal cancer polyps. IEEE J. Biomed. Health Inform. 28(4), 2326–2337 (2024)

    2024

  26. [34]

    Papangelou, C., Kyriakidis, K., Natsiavas, P., Chouvarda, I., Malousi, A.: Reliable machine learning models in genomic medicine using conformal prediction. Front. Bioinform. 5, 1507448 (2025)

    2025

  27. [35]

    Kumar, D., Peimankar, A., Sharma, K., Domínguez, H., Puthusserypady, S., Bardram, J.E.: Deepaware: A hybrid deep learning and context-aware heuristics- based model for atrial fibrillation detection. Comput. Methods Programs Biomed. 221, 106899 (2022)

    2022

  28. [36]

    Bashar, S.K., Han, D., Hajeb-Mohammadalipour, S., Ding, E., Whitcomb, C., McManus, D.D., Chon, K.H.: Atrial fibrillation detection from wrist photo- plethysmography signals using smartwatches. Sci. Rep. 9, 15054 (2019) https: //doi.org/10.1038/s41598-019-49092-2

    2019 doi

  29. [37]

    JACC Clin

    Dörr, M., Nohturfft, V., Brasier, N., Bosshard, E., Djurdjevic, A., Gross, S., Raichle, C.J., Rhinisperger, M., Stöckli, R., Eckstein, J.: The watch af trial: Smartwatches for detection of atrial fibrillation. JACC Clin. Electrophysiol. 5(2), 199–208 (2019)

    2019

  30. [38]

    : Accuracy of wristwatch- type photoplethysmography in detecting atrial fibrillation in daily life

    Nonoguchi, N.M., Soejima, K., Goda, A., Nishimura, K., Onozuka, D., Fujita, S., Koyama, F., Takano, Y., Iguchi, S., Sato, H., et al. : Accuracy of wristwatch- type photoplethysmography in detecting atrial fibrillation in daily life. Eur. Heart J.-Digit. Health 3(3), 455–464 (2022)

    2022

  31. [39]

    Selder, J.L., Te Kolste, H.J., Twisk, J., Schijven, M., Gielen, W., Allaart, C.P.: 21 Accuracy of a standalone atrial fibrillation detection algorithm added to a pop- ular wristband and smartwatch: Prospective diagnostic accuracy study. J. Med. Internet Res. 25, 44642 (2023)

    2023

  32. [40]

    : Evaluation of an algorithm-guided photoplethysmography for atrial fibrillation burden using a smartwatch

    Zhao, Z., Li, Q., Li, S., Guo, Q., Bo, X., Kong, X., Xia, S., Li, X., Dai, W., Guo, L., et al. : Evaluation of an algorithm-guided photoplethysmography for atrial fibrillation burden using a smartwatch. PACE 47(4), 511–517 (2024)

    2024

  33. [41]

    : Continuous 24-h photoplethysmogram monitoring enables detection of atrial fibrillation

    Väliaho, E.-S., Lipponen, J.A., Kuoppa, P., Martikainen, T.J., Jäntti, H., Ris- sanen, T.T., Castrén, M., Halonen, J., Tarvainen, M.P., Laitinen, T.M., et al. : Continuous 24-h photoplethysmogram monitoring enables detection of atrial fibrillation. Front. physiol. 12, 778775 (2022)

    2022

  34. [42]

    Allen, J.: Photoplethysmography and its application in clinical physiological measurement. Physiol. Meas. 28(3), 1 (2007)

    2007

  35. [43]

    IEEE Trans

    Butkuvienė, M., Petrėnas, A., Sološenko, A., Martin-Yebra, A., Marozas, V., Sörnmo, L.: Considerations on performance evaluation of atrial fibrillation detectors. IEEE Trans. Biomed. Eng. 68(11), 3250–3260 (2021)

    2021

  36. [44]

    arXiv preprint arXiv:1412.6980 (2014)

    Kingma, D.P.: Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)

    2014 arXiv

  37. [45]

    Springer, ??? (2018)

    Sörnmo, L.: Atrial Fibrillation from an Engineering Perspective. Springer, ??? (2018)

    2018

  38. [46]

    Dash, S., Chon, K., Lu, S., Raeder, E.: Automatic real time detection of atrial fibrillation. Ann. Biomed. Eng. 37, 1701–1709 (2009)

    2009

  39. [47]

    Bell Syst

    Shannon, C.E.: A mathematical theory of communication. Bell Syst. Tech. J. 27(3), 379–423 (1948)

    1948

  40. [48]

    Zhou, X., Ding, H., Ung, B., Pickwell-MacPherson, E., Zhang, Y.: Automatic online detection of atrial fibrillation based on symbolic dynamics and Shannon entropy. Biomed. Eng. Online 13, 1–18 (2014)

    2014

  41. [49]

    Tateno, K., Glass, L.: Automatic detection of atrial fibrillation using the coeffi- cient of variation and density histograms of RR and δRR intervals. Med. Biol. Eng. Comput. 39, 664–671 (2001)

    2001

  42. [50]

    Langley, P., Dewhurst, M., Di Marco, L., Adams, P., Dewhurst, F., Mwita, J., Walker, R., Murray, A.: Accuracy of algorithms for detection of atrial fibrillation from short duration beat interval recordings. Med. Eng. Phys. 34(10), 1441–1447 (2012)

    2012

  43. [51]

    Park, J., Lee, S., Jeon, M.: Atrial fibrillation detection by heart rate variability in Poincare plot. Biomed. Eng. Online 8, 1–12 (2009) 22

    2009

  44. [52]

    Lian, J., Wang, L., Muessig, D.: A simple method to detect atrial fibrillation using RR intervals. Am. J. Cardiol. 107(10), 1494–1497 (2011)

    2011

  45. [53]

    Richman, J.S., Moorman, J.R.: Physiological time-series analysis using approxi- mate entropy and sample entropy. Am. J. Physiol. Heart Circ. Physiol. 278(6), 2039–2049 (2000)

    2000

  46. [54]

    Petrėnas, A., Marozas, V., Sörnmo, L.: Low-complexity detection of atrial fib- rillation in continuous long-term monitoring. Comput. Biol. Med. 65, 184–191 (2015)

    2015

  47. [55]

    Gal, Y., Ghahramani, Z.: Dropout as a bayesian approximation: Representing model uncertainty in deep learning. In: Int. Conf. Mach. Learn., pp. 1050–1059 (2016). PMLR 23

    2016

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.