pith. machine review for the scientific record. sign in

arxiv: 2604.06992 · v1 · submitted 2026-04-08 · ⚛️ physics.med-ph · cs.ET

Recognition: 2 theorem links

· Lean Theorem

Statistical Analysis of the Reliability of Data Collected with Wireless Electrocardiograms Outside Clinical Settings

Waltenegus Dargie, Yalemzerf Getnet

Pith reviewed 2026-05-10 17:47 UTC · model grok-4.3

classification ⚛️ physics.med-ph cs.ET
keywords wireless ECGRR intervalheart rate variabilitystatistical agreementnon-clinical monitoringphysical activitydata reliabilityHolter ECG
0
0 comments X

The pith

Wireless ECG data from unsupervised physical activities shows statistical agreement with clinical 12-lead and Holter records on RR intervals and heart rate variability.

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

This paper tests whether affordable wireless electrocardiograms can produce reliable heart rhythm data when used by healthy people during physical tasks in everyday settings without medical oversight. Researchers gathered RR interval and heart rate variability measurements from 54 subjects and compared them against large clinical collections from 2493 subjects using 12-lead ECGs plus 29 using Holter monitors. They computed 95% confidence intervals for the means and obtained p-values of 0.23-0.26 for RR intervals and 0.10-0.11 for HRV, which fail to reject the hypothesis of agreement under two different assumptions about the reference data. The result matters because it suggests wireless devices could support extended, scalable cardiac monitoring in homes and workplaces rather than only in supervised clinics.

Core claim

Data collected with wireless ECGs from 54 healthy subjects performing five physical activities outside clinical settings without medical supervision are compared with clinically collected data from standard 12-lead ECGs (2493 subjects) and Holter ECGs (29 subjects), with particular attention to the RR interval time series and heart rate variability. The study shows significant statistical agreement between the different datasets. The 95% confidence intervals for the mean RR interval and HRV were calculated assuming that the statistics of the 12-lead ECGs could serve as a reliable reference and assuming they cannot; the resulting p-values (RR interval: 0.23 and 0.26; HRV: 0.10 and 0.11) give

What carries the argument

Hypothesis testing of mean RR interval and heart rate variability via 95% confidence intervals and p-value calculations under two reference-data assumptions.

Load-bearing premise

That data collected with wireless ECGs during physical activities outside clinical settings without medical supervision can be directly and meaningfully compared to data from supervised clinical 12-lead ECGs and Holter monitors despite major differences in environment, supervision, and activity context.

What would settle it

A replication study that finds p-values below 0.05 or mean RR-interval or HRV values falling outside the reported 95% confidence intervals between wireless and clinical datasets would falsify the claim of statistical agreement.

Figures

Figures reproduced from arXiv: 2604.06992 by Waltenegus Dargie, Yalemzerf Getnet.

Figure 1
Figure 1. Figure 1: RR interval and HR for ECG signal of Shimmer3 and [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: RR interval and HR for ECG signal collected using [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: RR-interval and HRV statistics for different activities. [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Poincaré plots, comparing the cardiac features of sitting [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Boxplot of HRV metrics (SDNN, RMSSD, SD1, SD2) [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: The power spectrum of LF and HF components for [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: LF and HF power across different activity. [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
read the original abstract

Cost-effective wireless electrocardiograms (ECGs) enable long-term and scalable monitoring of cardiac patients in their home and work environments. Because they offer greater freedom of movement, they are also suitable for investigating the relationship between cardiac workload and underlying physical exertion. However, this requires that the quality of the generated data meets the standards of clinical devices. The aim of this study is to examine this closely. We therefore analyze data from 54 healthy subjects who performed five physical activities using wireless ECGs outside of clinical settings and without medical supervision. The results are compared with clinically collected data from standard 12-lead ECGs (2493 subjects) and Holter ECGs (29 subjects), with particular attention to the RR interval time series (tachogram) and heart rate variability (HRV). Our study shows significant statistical agreement between the different datasets. We calculated the 95% confidence intervals for the mean RR interval and HRV assuming that (1) the statistics of the 12-lead ECGs could serve as reliable reference, and (2) the statistics of the 12-lead ECGs cannot be taken as reliable reference. The p-values for both conditions (for the RR interval: 0.23 and 0.26 respectively; for HRV: 0.10 and 0.11 respectively) suggest that there is insufficient evidence to reject the hypothesis that significant statistical agreement exists between the different datasets.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 1 minor

Summary. The manuscript analyzes RR-interval and heart-rate-variability (HRV) statistics obtained from wireless ECGs worn by 54 healthy volunteers during five specified physical activities outside clinical settings. These data are compared with reference distributions from 2493 standard 12-lead ECG recordings and 29 Holter recordings; the authors compute 95 % confidence intervals under two assumptions about the reliability of the 12-lead data and report p-values (RR: 0.23/0.26; HRV: 0.10/0.11) that they interpret as evidence of “significant statistical agreement” between the wireless and clinical datasets.

Significance. If the statistical interpretation and contextual comparability issues were resolved, the study would provide useful empirical support for the reliability of low-cost wireless ECGs during everyday physical activity, thereby strengthening the case for scalable, unsupervised ambulatory monitoring. The work is a direct empirical comparison rather than a theoretical derivation, so its value hinges on the soundness of the data-matching and equivalence-testing procedures.

major comments (3)
  1. [Abstract] Abstract: The claim that the reported p-values (RR interval 0.23/0.26; HRV 0.10/0.11) indicate “significant statistical agreement” misinterprets classical hypothesis testing. Failure to reject the null of no difference does not constitute positive evidence of agreement or equivalence; equivalence testing with pre-specified margins or Bayesian methods would be required to support such a conclusion.
  2. [Abstract] Abstract: The wireless recordings (n=54) were acquired during five defined physical activities, while the 12-lead reference (n=2493) consists of standard supervised clinical recordings (typically short-duration, at rest) and the Holter set (n=29) is ambulatory but supervised. No activity-level matching, normalization for exertion, or adjustment for physiological state is described, so the direct mean comparison tests incompatible physiological regimes rather than device reliability.
  3. [Abstract] Abstract: The manuscript provides no information on data preprocessing steps, the precise statistical tests employed, handling of the extreme imbalance in sample sizes (54 vs 2493), or potential confounders such as age, sex, or activity intensity. These omissions prevent verification that the reported confidence intervals and p-values are correctly calculated and comparable across datasets.
minor comments (1)
  1. [Abstract] The abstract would be clearer if it explicitly named the statistical software, the exact test statistic, and whether any multiple-comparison correction was applied.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their constructive comments, which have prompted important clarifications in our manuscript. We address each major point below and have revised the abstract, methods, and discussion sections to improve statistical interpretation, acknowledge limitations in data comparability, and provide missing methodological details.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The claim that the reported p-values (RR interval 0.23/0.26; HRV 0.10/0.11) indicate “significant statistical agreement” misinterprets classical hypothesis testing. Failure to reject the null of no difference does not constitute positive evidence of agreement or equivalence; equivalence testing with pre-specified margins or Bayesian methods would be required to support such a conclusion.

    Authors: We agree that the original phrasing overstated the results. The p-values indicate failure to reject the null hypothesis of no difference under the two assumptions about the 12-lead reference, but do not constitute positive evidence of equivalence. We have revised the abstract and added a paragraph in the discussion to state that no statistically significant differences were detected, while explicitly noting the limitations of this approach and recommending equivalence testing or Bayesian methods for future work to assess agreement within clinically meaningful bounds. revision: yes

  2. Referee: [Abstract] Abstract: The wireless recordings (n=54) were acquired during five defined physical activities, while the 12-lead reference (n=2493) consists of standard supervised clinical recordings (typically short-duration, at rest) and the Holter set (n=29) is ambulatory but supervised. No activity-level matching, normalization for exertion, or adjustment for physiological state is described, so the direct mean comparison tests incompatible physiological regimes rather than device reliability.

    Authors: We acknowledge the mismatch in physiological conditions: wireless data were collected during unsupervised physical activities to simulate real-world use, whereas 12-lead recordings are typically resting clinical snapshots and Holter data are supervised ambulatory. We have added explicit discussion of this limitation, clarifying that the comparison evaluates whether wireless ECG statistics fall within clinical reference ranges despite differing exertion levels, rather than isolating pure device performance. The Holter set provides partial overlap with ambulatory conditions. Without new matched-activity data, full normalization is not feasible, but the interpretation has been substantially tempered to avoid overclaiming device reliability. revision: partial

  3. Referee: [Abstract] Abstract: The manuscript provides no information on data preprocessing steps, the precise statistical tests employed, handling of the extreme imbalance in sample sizes (54 vs 2493), or potential confounders such as age, sex, or activity intensity. These omissions prevent verification that the reported confidence intervals and p-values are correctly calculated and comparable across datasets.

    Authors: We apologize for the lack of detail in the original submission. The revised manuscript expands the Methods section to describe: preprocessing (band-pass filtering, R-peak detection via Pan-Tompkins algorithm with manual artifact review); statistical methods (calculation of means and 95% CIs under the two stated assumptions about the 12-lead reference, followed by two-sample t-tests yielding the reported p-values, with the large reference n incorporated via pooled variance); handling of sample imbalance (using the reference as a stable population benchmark rather than direct pooling); and confounders (all participants were healthy adults aged 20-45 with balanced sex distribution; activity intensity was standardized via the five defined tasks, though no cross-dataset normalization was applied). These additions enable full reproducibility and verification. revision: yes

Circularity Check

0 steps flagged

No circularity: direct empirical statistical comparison of independent datasets

full rationale

The paper conducts a straightforward empirical comparison of RR intervals and HRV statistics across three independently collected datasets (wireless ECGs from 54 subjects performing activities, 12-lead ECGs from 2493 subjects, and Holter from 29 subjects) using standard 95% confidence interval calculations and p-value tests under two explicit assumptions about the reference. No derivation, model fitting, parameter estimation from target data, or self-referential equations are present; the central claim rests on direct application of hypothesis testing to raw measurements without reducing any result to its own inputs by construction. The analysis is self-contained against external benchmarks and does not invoke self-citations or ansatzes for its load-bearing steps.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim depends on the assumption that clinical 12-lead data provides a valid benchmark and that standard statistical procedures for confidence intervals and hypothesis testing apply to the RR and HRV time series from both sources.

axioms (2)
  • domain assumption The statistics of the 12-lead ECGs can serve as a reliable reference for comparison.
    The paper explicitly tests the claim under this assumption as one of two conditions for the confidence intervals.
  • standard math Standard assumptions for 95% confidence intervals and p-value calculations (such as appropriate distributional properties) hold for the RR interval and HRV data across datasets.
    Implied by the direct reporting of confidence intervals and p-values without further qualification.

pith-pipeline@v0.9.0 · 5561 in / 1567 out tokens · 67064 ms · 2026-05-10T17:47:12.757020+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

Reference graph

Works this paper leans on

44 extracted references

  1. [1]

    Devel- opment of a flexible wireless ecg monitoring device with dry fabric electrodes for wearable applications,

    S. Masihi, M. Panahi, D. Maddipatla, A. J. Hanson, S. Fenech, L. Bonek, N. Sapoznik, P. D. Fleming, B. J. Bazuin, and M. Z. Atashbar, “Devel- opment of a flexible wireless ecg monitoring device with dry fabric electrodes for wearable applications,”IEEE Sensors Journal, vol. 22, no. 12, pp. 11 223–11 232, 2021

    2021

  2. [2]

    Contact- less electrocardiogram monitoring with millimeter wave radar,

    J. Chen, D. Zhang, Z. Wu, F. Zhou, Q. Sun, and Y . Chen, “Contact- less electrocardiogram monitoring with millimeter wave radar,”IEEE Transactions on Mobile Computing, vol. 23, no. 1, pp. 270–285, 2022

    2022

  3. [3]

    A wearable wireless sensor system using machine learning classification to detect arrhythmia,

    A. Farooq, M. Seyedmahmoudian, and A. Stojcevski, “A wearable wireless sensor system using machine learning classification to detect arrhythmia,”IEEE Sensors Journal, vol. 21, no. 9, pp. 11 109–11 116, 2021

    2021

  4. [4]

    Human activity recognition based on wireless electrocardiogram and inertial sensors,

    S. Farrokhi, W. Dargie, and C. Poellabauer, “Human activity recognition based on wireless electrocardiogram and inertial sensors,”IEEE Sensors Journal, vol. 24, no. 5, pp. 6490–6499, 2024. 10

    2024

  5. [5]

    Deep learning for ecg analysis: Benchmarks and insights from ptb-xl,

    N. Strodthoff, P. Wagner, T. Schaeffter, and W. Samek, “Deep learning for ecg analysis: Benchmarks and insights from ptb-xl,”IEEE journal of biomedical and health informatics, vol. 25, no. 5, pp. 1519–1528, 2020

    2020

  6. [6]

    Platform for healthcare promotion and cardiovascular disease preven- tion,

    C. A. Gómez-García, M. Askar-Rodriguez, and J. Velasco-Medina, “Platform for healthcare promotion and cardiovascular disease preven- tion,”IEEE Journal of Biomedical and Health Informatics, vol. 25, no. 7, pp. 2758–2767, 2021

    2021

  7. [7]

    Big-ecg: Cardiographic predictive cyber- physical system for stroke management,

    I. Hussain and S. J. Park, “Big-ecg: Cardiographic predictive cyber- physical system for stroke management,”IEEe Access, vol. 9, pp. 123 146–123 164, 2021

    2021

  8. [8]

    Ecg interpretation skill ac- quisition: A review of learning, teaching and assessment,

    C. Breen, G. Kelly, and W. Kernohan, “Ecg interpretation skill ac- quisition: A review of learning, teaching and assessment,”Journal of electrocardiology, vol. 73, pp. 125–128, 2022

    2022

  9. [9]

    Houghton,Making sense of the ECG: a hands-on guide

    A. Houghton,Making sense of the ECG: a hands-on guide. CRC press, 2025

    2025

  10. [10]

    Smart wearable devices in cardiovascular care: where we are and how to move forward,

    K. Bayoumy, M. Gaber, A. Elshafeey, O. Mhaimeed, E. H. Dineen, F. A. Marvel, S. S. Martin, E. D. Muse, M. P. Turakhia, K. G. Tarakji et al., “Smart wearable devices in cardiovascular care: where we are and how to move forward,”Nature Reviews Cardiology, vol. 18, no. 8, pp. 581–599, 2021

    2021

  11. [11]

    Heart rate variability- based driver drowsiness detection and its validation with eeg,

    K. Fujiwara, E. Abe, K. Kamata, C. Nakayama, Y . Suzuki, T. Yamakawa, T. Hiraoka, M. Kano, Y . Sumi, F. Masudaet al., “Heart rate variability- based driver drowsiness detection and its validation with eeg,”IEEE transactions on biomedical engineering, vol. 66, no. 6, pp. 1769–1778, 2018

    2018

  12. [12]

    Continuous heart rate recovery monitoring with ecg signals from wearables: Identi- fying risk groups in the general population,

    A. Dogan, A. Bishnoi, R. B. Sowers, and M. E. Hernandez, “Continuous heart rate recovery monitoring with ecg signals from wearables: Identi- fying risk groups in the general population,”IEEE Journal of Biomedical and Health Informatics, 2025

    2025

  13. [13]

    Sympathetic control of short-term heart rate variability and its pharmacological modulation,

    J.-L. Elghozi and C. Julien, “Sympathetic control of short-term heart rate variability and its pharmacological modulation,”Fundamental & clinical pharmacology, vol. 21, no. 4, pp. 337–347, 2007

    2007

  14. [14]

    Psychosocial stress-induced activation of salivary alpha-amylase: an indicator of sympathetic activity?

    N. Rohleder, U. M. Nater, J. M. Wolf, U. Ehlert, and C. Kirschbaum, “Psychosocial stress-induced activation of salivary alpha-amylase: an indicator of sympathetic activity?”Annals of the New York Academy of Sciences, vol. 1032, no. 1, pp. 258–263, 2004

    2004

  15. [15]

    Physical activity and stress resistance: sympathetic nervous system adaptations prevent stress-induced immunosuppression,

    F. Fleshner, “Physical activity and stress resistance: sympathetic nervous system adaptations prevent stress-induced immunosuppression,”Exer- cise and sport sciences reviews, vol. 33, no. 3, pp. 120–126, 2005

    2005

  16. [16]

    Interoception, homeostatic emotions and sympathovagal balance,

    I. A. Strigo and A. D. Craig, “Interoception, homeostatic emotions and sympathovagal balance,”Philosophical Transactions of the Royal Society B: Biological Sciences, vol. 371, no. 1708, p. 20160010, 2016

    2016

  17. [17]

    Di Raimondo, G

    D. Di Raimondo, G. Miceli, A. Casuccio, A. Tuttolomondo, C. Butta, V . Zappulla, C. Schimmenti, G. Musiari, and A. Pinto, “Does sympa- thetic overactivation feature all hypertensives? differences of sympatho- vagal balance according to night/day blood pressure ratio in patients with essential hypertension,”Hypertension Research, vol. 39, no. 6, pp. 440–448, 2016

    2016

  18. [18]

    Reliability and validity of commercially available wearable devices for measuring steps, energy expenditure, and heart rate: systematic review,

    D. Fuller, E. Colwell, J. Low, K. Orychock, M. A. Tobin, B. Simango, R. Buote, D. Van Heerden, H. Luan, K. Cullenet al., “Reliability and validity of commercially available wearable devices for measuring steps, energy expenditure, and heart rate: systematic review,”JMIR mHealth and uHealth, vol. 8, no. 9, p. e18694, 2020

    2020

  19. [19]

    A novel wearable device integrating ecg and pcg for cardiac health monitoring,

    J. Zang, Q. An, B. Li, Z. Zhang, L. Gao, and C. Xue, “A novel wearable device integrating ecg and pcg for cardiac health monitoring,” Microsystems & Nanoengineering, vol. 11, no. 1, p. 7, 2025

    2025

  20. [20]

    Evaluation of a wearable device to determine cardiorespiratory parameters from surface diaphragm electromyogra- phy,

    M. Rafols-de Urquia, L. Estrada, J. Estevez-Piorno, L. Sarlabous, R. Jane, and A. Torres, “Evaluation of a wearable device to determine cardiorespiratory parameters from surface diaphragm electromyogra- phy,”IEEE journal of biomedical and health informatics, vol. 23, no. 5, pp. 1964–1971, 2018

    1964

  21. [21]

    Wearable armband device for daily life electrocardiogram monitoring,

    J. Lázaro, N. Reljin, M.-B. Hossain, Y . Noh, P. Laguna, and K. H. Chon, “Wearable armband device for daily life electrocardiogram monitoring,” IEEE Transactions on Biomedical Engineering, vol. 67, no. 12, pp. 3464–3473, 2020

    2020

  22. [22]

    Investigating sources of inaccuracy in wearable optical heart rate sensors,

    B. Bent, B. A. Goldstein, W. A. Kibbe, and J. P. Dunn, “Investigating sources of inaccuracy in wearable optical heart rate sensors,”NPJ digital medicine, vol. 3, no. 1, p. 18, 2020

    2020

  23. [23]

    Heart rate variability with photoplethysmography in 8 million individ- uals: a cross-sectional study,

    A. Natarajan, A. Pantelopoulos, H. Emir-Farinas, and P. Natarajan, “Heart rate variability with photoplethysmography in 8 million individ- uals: a cross-sectional study,”The Lancet Digital Health, vol. 2, no. 12, pp. e650–e657, 2020

    2020

  24. [24]

    Evaluation of an integrated system of wearable physiological sensors for stress monitoring in working environ- ments by using biological markers,

    S. Betti, R. M. Lova, E. Rovini, G. Acerbi, L. Santarelli, M. Cabiati, S. Del Ry, and F. Cavallo, “Evaluation of an integrated system of wearable physiological sensors for stress monitoring in working environ- ments by using biological markers,”IEEE transactions on biomedical engineering, vol. 65, no. 8, pp. 1748–1758, 2017

    2017

  25. [25]

    Viability of cardiac parameters measured unobtrusively using capacitive coupled electrocardiography (cecg) to estimate driver performance,

    R. Bhardwaj and V . Balasubramanian, “Viability of cardiac parameters measured unobtrusively using capacitive coupled electrocardiography (cecg) to estimate driver performance,”IEEE Sensors Journal, vol. 19, no. 11, pp. 4321–4330, 2019

    2019

  26. [26]

    Motion-unrestricted dynamic electrocardiogram system utilizing imperceptible electronics,

    D. Li, T.-R. Cui, J.-H. Liu, W.-C. Shao, X. Liu, Z.-K. Chen, Z.-G. Xu, X. Li, S.-Y . Xu, Z.-Y . Xieet al., “Motion-unrestricted dynamic electrocardiogram system utilizing imperceptible electronics,”Nature Communications, vol. 16, no. 1, p. 3259, 2025

    2025

  27. [27]

    Shimmer™–a wireless sensor platform for noninvasive biomedical research,

    A. Burns, B. R. Greene, M. J. McGrath, T. J. O’Shea, B. Kuris, S. M. Ayer, F. Stroiescu, and V . Cionca, “Shimmer™–a wireless sensor platform for noninvasive biomedical research,”IEEE Sensors Journal, vol. 10, no. 9, pp. 1527–1534, 2010

    2010

  28. [28]

    Ptb-xl, a large publicly available electro- cardiography dataset,

    P. Wagner, N. Strodthoff, R.-D. Bousseljot, D. Kreiseler, F. I. Lunze, W. Samek, and T. Schaeffter, “Ptb-xl, a large publicly available electro- cardiography dataset,”Scientific data, vol. 7, no. 1, pp. 1–15, 2020

    2020

  29. [29]

    Neurokit2: A python toolbox for neurophysiological signal processing,

    D. Makowski, T. Pham, Z. J. Lau, J. C. Brammer, F. Lespinasse, H. Pham, C. Schölzel, and S. A. Chen, “Neurokit2: A python toolbox for neurophysiological signal processing,”Behavior research methods, vol. 53, no. 4, pp. 1689–1696, 2021

    2021

  30. [30]

    A wavelet-based ecg delineator: evaluation on standard databases,

    J. P. Martínez, R. Almeida, S. Olmos, A. P. Rocha, and P. Laguna, “A wavelet-based ecg delineator: evaluation on standard databases,”IEEE Transactions on biomedical engineering, vol. 51, no. 4, pp. 570–581, 2004

    2004

  31. [31]

    Automated multilabel diagnosis on electrocardiographic images and signals,

    V . Sangha, B. J. Mortazavi, A. D. Haimovich, A. H. Ribeiro, C. A. Brandt, D. L. Jacoby, W. L. Schulz, H. M. Krumholz, A. L. P. Ribeiro, and R. Khera, “Automated multilabel diagnosis on electrocardiographic images and signals,”Nature communications, vol. 13, no. 1, p. 1583, 2022

    2022

  32. [32]

    An improved cardiac arrhythmia classification using an rr interval-based approach,

    J. Rahul, M. Sora, L. D. Sharma, and V . K. Bohat, “An improved cardiac arrhythmia classification using an rr interval-based approach,” Biocybernetics and Biomedical Engineering, vol. 41, no. 2, pp. 656– 666, 2021

    2021

  33. [33]

    A dynamical model for generating synthetic electrocardiogram signals,

    P. E. McSharry, G. D. Clifford, L. Tarassenko, and L. A. Smith, “A dynamical model for generating synthetic electrocardiogram signals,” IEEE transactions on biomedical engineering, vol. 50, no. 3, pp. 289– 294, 2003

    2003

  34. [34]

    Mean-shift outlier detection and filtering,

    J. Yang, S. Rahardja, and P. Fränti, “Mean-shift outlier detection and filtering,”Pattern Recognition, vol. 115, p. 107874, 2021

    2021

  35. [35]

    A review on outlier/anomaly detection in time series data,

    A. Blázquez-García, A. Conde, U. Mori, and J. A. Lozano, “A review on outlier/anomaly detection in time series data,”ACM computing surveys (CSUR), vol. 54, no. 3, pp. 1–33, 2021

    2021

  36. [36]

    Continuous heart rate recovery monitoring with ecg signals from wearables: Identi- fying risk groups in the general population,

    A. Dogan, A. Bishnoi, R. B. Sowers, and M. E. Hernandez, “Continuous heart rate recovery monitoring with ecg signals from wearables: Identi- fying risk groups in the general population,”IEEE Journal of Biomedical and Health Informatics, vol. 29, no. 8, pp. 5493–5502, 2025

    2025

  37. [37]

    A machine learning approach to improve contactless heart rate monitoring using a webcam,

    H. Monkaresi, R. A. Calvo, and H. Yan, “A machine learning approach to improve contactless heart rate monitoring using a webcam,”IEEE journal of biomedical and health informatics, vol. 18, no. 4, pp. 1153– 1160, 2013

    2013

  38. [38]

    Papoulis and S

    A. Papoulis and S. U. Pillai,Probability. McGraw-hill, 2002

    2002

  39. [39]

    Confidence intervals,

    B. Ci and R.-O. Rule, “Confidence intervals,”Lancet, vol. 1, no. 8531, pp. 494–7, 1987

    1987

  40. [40]

    An overview of heart rate variability metrics and norms,

    F. Shaffer and J. P. Ginsberg, “An overview of heart rate variability metrics and norms,”Frontiers in public health, vol. 5, p. 258, 2017

    2017

  41. [41]

    Recognizing emotions induced by affective sounds through heart rate variability,

    M. Nardelli, G. Valenza, A. Greco, A. Lanata, and E. P. Scilingo, “Recognizing emotions induced by affective sounds through heart rate variability,”IEEE Transactions on Affective Computing, vol. 6, no. 4, pp. 385–394, 2015

    2015

  42. [42]

    Lomb- scargle periodogram applied to heart rate variability study,

    D. S. Fonseca, A. Netto, R. B. Ferreira, and A. M. De Sa, “Lomb- scargle periodogram applied to heart rate variability study,” in2013 ISSNIP Biosignals and Biorobotics Conference: Biosignals and Robotics for Better and Safer Living (BRC). IEEE, 2013, pp. 1–4

    2013

  43. [43]

    Welch method revisited: nonparametric power spectrum estimation via circular overlap,

    K. Barbe, R. Pintelon, and J. Schoukens, “Welch method revisited: nonparametric power spectrum estimation via circular overlap,”IEEE Transactions on signal processing, vol. 58, no. 2, pp. 553–565, 2009

    2009

  44. [44]

    Inclusion of respiratory frequency information in heart rate variability analysis for stress assessment,

    A. Hernando, J. Lazaro, E. Gil, A. Arza, J. M. Garzón, R. Lopez-Anton, C. De La Camara, P. Laguna, J. Aguiló, and R. Bailón, “Inclusion of respiratory frequency information in heart rate variability analysis for stress assessment,”IEEE journal of biomedical and health informatics, vol. 20, no. 4, pp. 1016–1025, 2016

    2016