arxiv: 2604.07521 · v1 · submitted 2026-04-08 · 📡 eess.SP

Recognition: 2 theorem links

· Lean Theorem

ospEDA: Orthogonal Subspace Projection for Electrodermal Activity Decomposition

Yongbin Lee , Youngsun Kong , Ki H. Chon

Authors on Pith no claims yet

Pith reviewed 2026-05-10 17:07 UTC · model grok-4.3

classification 📡 eess.SP

keywords electrodermal activityEDA decompositiontonic phasic separationorthogonal subspace projectionsympathetic nervous activityphysiological signal processingnoise-robust decompositionphasic driver estimation

0 comments

The pith

ospEDA decomposes electrodermal activity into tonic and phasic components using orthogonal subspace projection after valley detection for the tonic part.

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

The paper presents ospEDA as a method to separate EDA signals into a slowly varying tonic component and a fast phasic component that reflects sympathetic nerve bursts. It combines valley detection to estimate the tonic baseline for noise resistance, orthogonal subspace projection to pull out the phasic part while handling differences between people, and non-negative least squares deconvolution to recover the underlying driver. This matters because accurate decomposition lets researchers and clinicians better quantify arousal, stress, and pain from skin conductance recordings that are often noisy or vary across users. The evaluations on simulated data with added noise and five real datasets show lower error rates and better detection of sympathetic events than six prior approaches. If the improvements hold, the method could support more reliable physiological monitoring in clinical and wearable settings.

Core claim

By estimating the tonic component through physiologically motivated valley detection, extracting the phasic component with orthogonal subspace projection to handle inter-subject variability, and recovering the phasic driver via non-negative least squares deconvolution with ridge regularization, ospEDA produces lower root mean square error on tonic and phasic estimates than existing methods in 10 dB and 20 dB SNR simulations, higher F1 scores for sympathetic nerve activity detection, an AUROC of 0.766 for stimulus classification, and effect sizes ω² > 0.14 across all five real-world datasets.

What carries the argument

Orthogonal subspace projection applied to the phasic residual after valley-based tonic estimation, followed by NNLS deconvolution, to isolate sympathetic bursts while accommodating signal morphology differences.

If this is right

Lower RMSE on both tonic and phasic components under moderate to high noise levels compared with prior decomposition techniques.
Higher F1 scores for detecting sympathetic nerve activity at 10 dB, 20 dB, and 30 dB SNR.
Consistent effect sizes above 0.14 and stimulus classification AUROC of 0.766 on multiple real-world EDA collections.
More stable phasic driver estimates that support downstream analysis of arousal and stress responses.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same valley-plus-projection pipeline could be tested on other slowly varying physiological signals that need baseline removal, such as respiration or blood pressure.
If the method works without per-user tuning, it may reduce the need for extensive calibration in consumer wearable devices that record skin conductance.
Performance gains in noisy conditions suggest the approach could extend to ambulatory recordings where motion artifacts are common.
Integrating the estimated phasic driver with heart-rate variability features might improve multimodal models of sympathetic activation.

Load-bearing premise

That valley detection reliably identifies the true tonic level across people with different skin conductance shapes and that the orthogonal projection step removes phasic content without creating new artifacts or needing per-dataset adjustments.

What would settle it

A new set of simulated EDA traces at 15 dB SNR or a sixth real dataset where ospEDA's phasic RMSE exceeds the best competing method or its F1 score for sympathetic detection falls below the next-best algorithm.

Figures

Figures reproduced from arXiv: 2604.07521 by Ki H. Chon, Yongbin Lee, Youngsun Kong.

**Figure 1.** Figure 1: An example of the simulated EDA signal. The figure displays the ground-truth tonic component, discrete sympathetic stimuli, and the composite EDA signal corrupted by varying levels of noise (clean, 30 dB, 20 dB, and 10 dB SNR) [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗

read the original abstract

Electrodermal activity (EDA) is a widely used physiological signal for assessing sympathetic nervous activity, such as arousal, stress, and pain. However, reliable decomposition into tonic and phasic components remains challenging, particularly in noisy environments and across individuals with varying signal morphologies and stimulus responses. We propose ospEDA, a novel Orthogonal Subspace Projection (OSP) based method for EDA decomposition. The method integrates (1) tonic estimation via physiologically motivated valley detection for noise robustness; (2) phasic extraction using OSP to accommodate inter subject variability; and (3) phasic driver estimation through non-negative least squares (NNLS) deconvolution with ridge regularization. We evaluated ospEDA on five real-world datasets and one simulated EDA dataset with ground-truth components, comparing its performance against six existing methods. In simulations with a 20 dB signal to noise ratio (SNR), ospEDA achieved the lowest root mean square error (RMSE) for estimated tonic (0.131) and phasic (0.132) components. Under noisier conditions (10 dB SNR), it maintained superior phasic RMSE (0.293), Pearson correlation (0.782), and R^2 (0.979) values. Furthermore, ospEDA consistently provided the highest F1 scores (0.573, 0.617, 0.638) for sympathetic nerve activity detection across 10, 20, and 30 dB SNR levels, respectively, compared to existing methods. On the real world datasets, ospEDA achieved a stimulus classification AUROC of 0.766 and consistently maintained strong effect sizes ({\omega}^2>0.14) across all five datasets. Overall, ospEDA represents a promising framework for EDA decomposition, showing generally consistent performance and reliable phasic driver estimation under the varying noise conditions, with potential utility for real world physiological monitoring applications.

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.

Desk Editor's Note private letter to a colleague

ospEDA puts valley detection, orthogonal subspace projection, and ridge NNLS together for EDA decomposition and reports better RMSE, F1, and AUROC numbers than six prior methods on noisy simulations plus five real datasets, but the robustness of the valley step to varying signal shapes is untested.

read the letter

ospEDA is a new pipeline for breaking down electrodermal activity into tonic and phasic parts using valley detection for the tonic level, orthogonal subspace projection to pull out the phasic component, and ridge-regularized non-negative least squares for the driver signal. It shows stronger numbers than six other methods on noisy simulations and real data. The combination itself is what stands out as new here. The paper does a decent job laying out the steps and running head-to-head tests. On the simulated data at 20 dB SNR it gets the lowest RMSE for both components, and it keeps an edge at 10 dB too. On the real datasets it hits an AUROC of 0.766 for stimulus classification and solid effect sizes across the board. The soft spot is the reliance on valley detection for tonic estimation. That step assumes the valleys reliably mark the tonic level even when signals look different from person to person or under varying conditions. If that assumption slips, the orthogonal projection step gets the wrong subspace and the rest follows. The abstract and results do not show any targeted checks for how well this holds across morphologies, so the generalizability claim rests on the tested cases alone. The simulation wins could be tied to how the data was generated. This paper is aimed at people who work with EDA in psychology, HCI, or monitoring applications and want a practical decomposition tool that handles noise better. A reader who needs to pick or improve a method for real-world signals would get concrete comparisons to use. It has enough structure and evidence to warrant a serious referee, even if revisions are needed on the validation side. I would send it to peer review and ask the authors to add sensitivity tests on different signal shapes and more detail on the exact parameter settings and statistical procedures.

Referee Report

3 major / 2 minor

Summary. The manuscript proposes ospEDA, a method for decomposing electrodermal activity (EDA) signals into tonic and phasic components via orthogonal subspace projection (OSP). It integrates physiologically motivated valley detection for tonic estimation, OSP to accommodate inter-subject variability in phasic extraction, and non-negative least squares (NNLS) deconvolution with ridge regularization for phasic driver estimation. The approach is evaluated against six existing methods on one simulated dataset (with ground-truth components at 10–30 dB SNR) and five real-world datasets, claiming lowest RMSE (0.131 tonic / 0.132 phasic at 20 dB), superior phasic RMSE/correlation/R² at 10 dB, highest F1 scores (0.573–0.638) for sympathetic nerve activity detection, AUROC of 0.766, and ω² > 0.14 on real data.

Significance. If the performance advantages prove robust and reproducible, ospEDA could offer a practical advance in EDA decomposition for noisy environments and variable morphologies, supporting more reliable sympathetic activity monitoring in stress, arousal, and pain applications. The use of linear-algebraic subspace projection grounded in physiological priors is a constructive contribution, but the central claims depend on unverified assumptions about valley detection stability.

major comments (3)

[Methods] Methods (tonic estimation subsection): The valley detection procedure for tonic level estimation is described only as 'physiologically motivated' without algorithmic details, thresholds, or pseudocode; this step is load-bearing because misalignment in the orthogonal complement directly biases the subsequent OSP phasic isolation and NNLS deconvolution.
[Results] Results (simulation experiments): The reported RMSE values (tonic 0.131, phasic 0.132 at 20 dB SNR) and F1 scores (up to 0.638) are presented as evidence of superiority, yet the manuscript provides neither the exact simulation parameters, ridge regularization strength, nor any sensitivity analysis to morphological variations, leaving open the possibility that gains are tied to the specific data distribution rather than intrinsic method properties.
[Evaluation] Evaluation (real-world datasets section): The AUROC of 0.766 and consistent ω² > 0.14 are claimed across five datasets, but without explicit data-exclusion criteria, statistical test details, or cross-morphology validation, it is impossible to confirm that the orthogonal projection accommodates inter-subject variability without introducing artifacts or requiring per-dataset tuning.

minor comments (2)

[Abstract] Abstract: The six baseline methods are not named; listing them would allow readers to immediately contextualize the comparisons.
[Results] Notation: The symbol ω² for effect size appears without definition or reference on first use; a brief parenthetical explanation would improve accessibility.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback on our manuscript. We have reviewed each major comment carefully and provide point-by-point responses below. We agree that additional methodological details, simulation parameters, and evaluation transparency are warranted to strengthen the presentation and will revise the manuscript accordingly.

read point-by-point responses

Referee: [Methods] Methods (tonic estimation subsection): The valley detection procedure for tonic level estimation is described only as 'physiologically motivated' without algorithmic details, thresholds, or pseudocode; this step is load-bearing because misalignment in the orthogonal complement directly biases the subsequent OSP phasic isolation and NNLS deconvolution.

Authors: We agree that the current description of the valley detection procedure is insufficiently detailed. In the revised manuscript we will expand the tonic estimation subsection to include the full algorithmic specification, explicit thresholds for valley identification, the precise criteria for selecting physiologically motivated valleys, and pseudocode. This addition will directly address the concern that misalignment could bias the orthogonal complement and subsequent steps, allowing readers to reproduce and evaluate the stability of the tonic estimate. revision: yes
Referee: [Results] Results (simulation experiments): The reported RMSE values (tonic 0.131, phasic 0.132 at 20 dB SNR) and F1 scores (up to 0.638) are presented as evidence of superiority, yet the manuscript provides neither the exact simulation parameters, ridge regularization strength, nor any sensitivity analysis to morphological variations, leaving open the possibility that gains are tied to the specific data distribution rather than intrinsic method properties.

Authors: We acknowledge that the simulation section currently omits the exact generative parameters, the numerical value of the ridge regularization strength, and any sensitivity analysis across morphological variations. In the revision we will add a dedicated subsection listing all simulation parameters (including SNR generation, component amplitudes, and noise models), report the ridge parameter used, and include a new sensitivity analysis that varies morphological features (e.g., phasic burst width and tonic drift rates) to demonstrate that the reported RMSE and F1 improvements are not artifacts of the particular data distribution. revision: yes
Referee: [Evaluation] Evaluation (real-world datasets section): The AUROC of 0.766 and consistent ω² > 0.14 are claimed across five datasets, but without explicit data-exclusion criteria, statistical test details, or cross-morphology validation, it is impossible to confirm that the orthogonal projection accommodates inter-subject variability without introducing artifacts or requiring per-dataset tuning.

Authors: We agree that greater transparency is required for the real-world evaluation. The revised manuscript will specify the data-exclusion criteria applied to each of the five datasets, provide full details of the statistical procedures used to compute AUROC and ω² (including any multiple-comparison corrections), and add a cross-morphology validation analysis that partitions subjects by signal morphology to verify that the orthogonal subspace projection generalizes without per-dataset retuning or introduction of artifacts. These additions will directly support the claim that the method accommodates inter-subject variability. revision: yes

Circularity Check

0 steps flagged

No circularity: standard techniques applied to EDA decomposition

full rationale

The derivation chain consists of (1) valley detection for tonic level, (2) orthogonal subspace projection to isolate phasic component, and (3) NNLS deconvolution with ridge regularization for the driver. These are conventional linear-algebra and optimization steps whose outputs are not defined in terms of the inputs or fitted parameters from the same data. Performance is assessed via independent simulation ground-truth RMSE/correlation and real-dataset AUROC/effect-size metrics. No self-citation is load-bearing for any central equation, no uniqueness theorem is imported, and no ansatz is smuggled. The method is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review; no explicit free parameters, background axioms, or newly postulated entities are stated. The ridge regularization coefficient in the NNLS step is implicitly a tunable hyper-parameter whose value is not reported.

pith-pipeline@v0.9.0 · 5649 in / 1293 out tokens · 72251 ms · 2026-05-10T17:07:16.869210+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

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

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

tonic estimation via physiologically motivated valley detection... phasic extraction using OSP... NNLS deconvolution with ridge regularization
IndisputableMonolith/Foundation/AlexanderDuality.lean alexander_duality_circle_linking unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

orthogonal subspace projection (OSP) based method... projection matrix P_m onto subspace spanned by V_m

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

Memory-Efficient EDA Denoising via Knowledge Distillation for Wearable IoT Under Severe Motion Artifacts and Underwater Conditions
eess.SP 2026-05 conditional novelty 5.0

Knowledge distillation from a hybrid CNN-Transformer teacher to a depth-wise separable CNN student, combined with realistic motion and environmental augmentation, produces a 15x smaller EDA denoiser that cuts underwat...

Reference graph

Works this paper leans on

30 extracted references · 29 canonical work pages · cited by 1 Pith paper

[1]

Copyright may be transferred without notice, after which this version may no longer be accessible

This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible. ospEDA: Orthogonal Subspace Projection for Electrodermal Activity Decomposition Yongbin Lee, Youngsun Kong, and Ki H. Chon, Fellow, IEEE Abstract— Electrodermal activity (EDA) is a widely use...

2010
[2]

Feature Extraction and Selection for Emotion Recognition from Electrodermal Activity,

J. Shukla, M. Barreda-Ángeles, J. Oliver, G. C. Nandi, and D. Puig, “Feature Extraction and Selection for Emotion Recognition from Electrodermal Activity,” IEEE Trans. Affect. Comput., vol. 12, no. 4, pp. 857–869, Oct. 2021, doi: 10.1109/TAFFC.2019.2901673

work page doi:10.1109/taffc.2019.2901673 2021
[3]

Electrodermal activity in pain assessment and its clinical applications,

Y. Kong and K. H. Chon, “Electrodermal activity in pain assessment and its clinical applications,” Appl. Phys. Rev., vol. 11, no. 3, Sep. 2024, doi: 10.1063/5.0200395

work page doi:10.1063/5.0200395 2024
[4]

Pain Assessment Tool With Electrodermal Activity for Postoperative Patients: Method Validation Study,

S. A. H. Aqajari et al., “Pain Assessment Tool With Electrodermal Activity for Postoperative Patients: Method Validation Study,” JMIR MHealth UHealth, vol. 9, no. 5, p. e25258, May 2021, doi: 10.2196/25258

work page doi:10.2196/25258 2021
[5]

Automated Pain Assessment in Children Using Electrodermal Activity and Video Data Fusion via Machine Learning,

B. T. Susam et al., “Automated Pain Assessment in Children Using Electrodermal Activity and Video Data Fusion via Machine Learning,” IEEE Trans. Biomed. Eng., vol. 69, no. 1, pp. 422–431, Jan. 2022, doi: 10.1109/TBME.2021.3096137

work page doi:10.1109/tbme.2021.3096137 2022
[6]

The Impact of Using Measurements of Electrodermal Activity in the Assessment of Problematic Behaviour in Dementia,

C. A. Melander, B. Kikhia, M. Olsson, B.-M. Wälivaara, and S. Sävenstedt, “The Impact of Using Measurements of Electrodermal Activity in the Assessment of Problematic Behaviour in Dementia,” Dement. Geriatr. Cogn. Disord. Extra, vol. 8, no. 3, pp. 333–347, Oct. 2018, doi: 10.1159/000493339. 13

work page doi:10.1159/000493339 2018
[7]

Acute Stress State Classification Based on Electrodermal Activity Modeling,

A. Greco et al., “Acute Stress State Classification Based on Electrodermal Activity Modeling,” IEEE Trans. Affect. Comput., vol. 14, no. 1, pp. 788–799, Jan. 2023, doi: 10.1109/TAFFC.2021.3055294

work page doi:10.1109/taffc.2021.3055294 2023
[8]

Electrodermal Activity and Heart Rate Variability During Exposure Fear Scripts Predict Trait-Level and Momentary Social Anxiety and Eating-Disorder Symptoms in an Analogue Sample,

C. Christian, E. Cash, D. A. Cohen, C. M. Trombley, and C. A. Levinson, “Electrodermal Activity and Heart Rate Variability During Exposure Fear Scripts Predict Trait-Level and Momentary Social Anxiety and Eating-Disorder Symptoms in an Analogue Sample,” Clin. Psychol. Sci., vol. 11, no. 1, pp. 134–148, Jan. 2023, doi: 10.1177/21677026221083284

work page doi:10.1177/21677026221083284 2023
[9]

Towards real-world wearable sleepiness detection: Electrodermal activity data during speech can identify sleep deprivation,

J. Moon, A. Peitzsch, Y. Kong, P. Seshadri, and K. H. Chon, “Towards real-world wearable sleepiness detection: Electrodermal activity data during speech can identify sleep deprivation,” Comput. Biol. Med., vol. 184, p. 109320, Jan. 2025, doi: 10.1016/j.compbiomed.2024.109320

work page doi:10.1016/j.compbiomed.2024.109320 2025
[10]

Electrodermal activity as a measure of cognitive load: a methodological approach,

M. Buchwald, S. Kupiński, A. Bykowski, J. Marcinkowska, D. Ratajczyk, and M. Jukiewicz, “Electrodermal activity as a measure of cognitive load: a methodological approach,” in 2019 Signal Processing: Algorithms, Architectures, Arrangements, and Applications (SPA), Sep. 2019, pp. 175–179. doi: 10.23919/SPA.2019.8936745

work page doi:10.23919/spa.2019.8936745 2019
[11]

Elevation of spectral components of electrodermal activity precedes central nervous system oxygen toxicity symptoms in divers,

H. F. Posada–Quintero et al., “Elevation of spectral components of electrodermal activity precedes central nervous system oxygen toxicity symptoms in divers,” Commun. Med., vol. 4, no. 1, p. 270, Dec. 2024, doi: 10.1038/s43856-024-00688-4

work page doi:10.1038/s43856-024-00688-4 2024
[12]

Wearables measuring electrodermal activity to assess perceived stress in care: a scoping review,

A. Klimek, I. Mannheim, G. Schouten, E. J. M. Wouters, and M. W. H. Peeters, “Wearables measuring electrodermal activity to assess perceived stress in care: a scoping review,” Acta Neuropsychiatr., vol. 37, p. e19, Jan. 2025, doi: 10.1017/neu.2023.19

work page doi:10.1017/neu.2023.19 2025
[13]

A Wearable Sensor for Unobtrusive, Long-Term Assessment of Electrodermal Activity,

M.-Z. Poh, N. C. Swenson, and R. W. Picard, “A Wearable Sensor for Unobtrusive, Long-Term Assessment of Electrodermal Activity,” IEEE Trans. Biomed. Eng., vol. 57, no. 5, pp. 1243–1252, May 2010, doi: 10.1109/TBME.2009.2038487

work page doi:10.1109/tbme.2009.2038487 2010
[14]

Bodily Electrodermal Representations for Affective Computing,

X. Shui et al., “Bodily Electrodermal Representations for Affective Computing,” IEEE Trans. Affect. Comput., vol. 15, no. 3, pp. 1018–1025, Jul. 2024, doi: 10.1109/TAFFC.2023.3315973

work page doi:10.1109/taffc.2023.3315973 2024
[15]

Detecting Sympathetic Discharges: Comparison of Electrodermal Activity and Skin Sympathetic Nerve Activity in Stimulation-to-Response Time and Recovery Time to Baseline,

F. Baghestani, Y. Kong, I.-P. Chen, W. D’Angelo, and K. H. Chon, “Detecting Sympathetic Discharges: Comparison of Electrodermal Activity and Skin Sympathetic Nerve Activity in Stimulation-to-Response Time and Recovery Time to Baseline,” IEEE Trans. Affect. Comput., pp. 1–9, 2025, doi: 10.1109/TAFFC.2025.3565174

work page doi:10.1109/taffc.2025.3565174 2025
[16]

A continuous measure of phasic electrodermal activity,

M. Benedek and C. Kaernbach, “A continuous measure of phasic electrodermal activity,” J. Neurosci. Methods, vol. 190, no. 1, pp. 80–91, Jun. 2010, doi: 10.1016/j.jneumeth.2010.04.028

work page doi:10.1016/j.jneumeth.2010.04.028 2010
[17]

Decomposition of skin conductance data by means of nonnegative deconvolution,

M. Benedek and C. Kaernbach, “Decomposition of skin conductance data by means of nonnegative deconvolution,” Psychophysiology, vol. 47, no. 4, pp. 647–658, 2010, doi: 10.1111/j.1469-8986.2009.00972.x

work page doi:10.1111/j.1469-8986.2009.00972.x 2010
[18]

cvxEDA: A Convex Optimization Approach to Electrodermal Activity Processing,

A. Greco, G. Valenza, A. Lanata, E. P. Scilingo, and L. Citi, “cvxEDA: A Convex Optimization Approach to Electrodermal Activity Processing,” IEEE Trans. Biomed. Eng., vol. 63, no. 4, pp. 797–804, Apr. 2016, doi: 10.1109/TBME.2015.2474131

work page doi:10.1109/tbme.2015.2474131 2016
[19]

ϵ ⋆: An online coverage path planning algorithm,

F. Hernando-Gallego, D. Luengo, and A. Artés-Rodríguez, “Feature Extraction of Galvanic Skin Responses by Nonnegative Sparse Deconvolution,” IEEE J. Biomed. Health Inform., vol. 22, no. 5, pp. 1385–1394, Sep. 2018, doi: 10.1109/JBHI.2017.2780252

work page doi:10.1109/jbhi.2017.2780252 2018
[20]

Physiological characterization of electrodermal activity enables scalable near real-time autonomic nervous system activation inference,

R. Amin and R. T. Faghih, “Physiological characterization of electrodermal activity enables scalable near real-time autonomic nervous system activation inference,” PLOS Comput. Biol., vol. 18, no. 7, p. e1010275, Jul. 2022, doi: 10.1371/journal.pcbi.1010275

work page doi:10.1371/journal.pcbi.1010275 2022
[21]

A Unified Dynamic Model for the Decomposition of Skin Conductance and the Inference of Sudomotor Nerve Activities,

H. S. Wang, S. Marsella, and M. Pavel, “A Unified Dynamic Model for the Decomposition of Skin Conductance and the Inference of Sudomotor Nerve Activities,” IEEE Trans. Biomed. Eng., vol. 72, no. 3, pp. 1178–1187, Mar. 2025, doi: 10.1109/TBME.2024.3492112

work page doi:10.1109/tbme.2024.3492112 2025
[22]

An improved algorithm for model-based analysis of evoked skin conductance responses,

D. R. Bach, K. J. Friston, and R. J. Dolan, “An improved algorithm for model-based analysis of evoked skin conductance responses,” Biol. Psychol., vol. 94, no. 3, pp. 490–497, Dec. 2013, doi: 10.1016/j.biopsycho.2013.09.010

work page doi:10.1016/j.biopsycho.2013.09.010 2013
[23]

Power Spectral Density Analysis of Electrodermal Activity for Sympathetic Function Assessment,

H. F. Posada-Quintero, J. P. Florian, A. D. Orjuela-Cañón, T. Aljama-Corrales, S. Charleston-Villalobos, and K. H. Chon, “Power Spectral Density Analysis of Electrodermal Activity for Sympathetic Function Assessment,” Ann. Biomed. Eng., vol. 44, no. 10, pp. 3124–3135, Oct. 2016, doi: 10.1007/s10439-016-1606-6

work page doi:10.1007/s10439-016-1606-6 2016
[24]

The biovid heat pain database data for the advancement and systematic validation of an automated pain recognition system,

S. Walter et al., “The biovid heat pain database data for the advancement and systematic validation of an automated pain recognition system,” in 2013 IEEE International Conference on Cybernetics (CYBCO), Jun. 2013, pp. 128–131. doi: 10.1109/CYBConf.2013.6617456

work page doi:10.1109/cybconf.2013.6617456 2013
[25]

Automatic Pain Recognition from Video and Biomedical Signals,

P. Werner, A. Al-Hamadi, R. Niese, S. Walter, S. Gruss, and H. C. Traue, “Automatic Pain Recognition from Video and Biomedical Signals,” in 2014 22nd International Conference on Pattern Recognition, Aug. 2014, pp. 4582–4587. doi: 10.1109/ICPR.2014.784

work page doi:10.1109/icpr.2014.784 2014
[26]

Design and Evaluation of Deep Learning Models for Continuous Acute Pain Detection Based on Phasic Electrodermal Activity,

J. O. Pinzon-Arenas, Y. Kong, K. H. Chon, and H. F. Posada-Quintero, “Design and Evaluation of Deep Learning Models for Continuous Acute Pain Detection Based on Phasic Electrodermal Activity,” IEEE J. Biomed. Health Inform., vol. 27, no. 9, pp. 4250–4260, Sep. 2023, doi: 10.1109/JBHI.2023.3291955

work page doi:10.1109/jbhi.2023.3291955 2023
[27]

Using electrodermal activity to validate multilevel pain stimulation in healthy volunteers evoked by thermal grills,

H. F. Posada-Quintero et al., “Using electrodermal activity to validate multilevel pain stimulation in healthy volunteers evoked by thermal grills,” Am. J. Physiol.-Regul. Integr. Comp. Physiol., vol. 319, no. 3, pp. R366–R375, Sep. 2020, doi: 10.1152/ajpregu.00102.2020

work page doi:10.1152/ajpregu.00102.2020 2020
[28]

In: 2021 43rd Annual International Conference of the IEEE En- gineering in Medicine & Biology Society (EMBC)

Y. Kong, H. F. Posada-Quintero, and K. H. Chon, “Female–male Differences Should be Considered in Physical Pain Quantification based on Electrodermal Activity: Preliminary Study,” in 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Nov. 2021, pp. 6941–6944. doi: 10.1109/EMBC46164.2021.9630637

work page doi:10.1109/embc46164.2021.9630637 2021
[29]

Sensitive Physiological Indices of Pain Based on Differential Characteristics of Electrodermal Activity,

Y. Kong, H. F. Posada-Quintero, and K. H. Chon, “Sensitive Physiological Indices of Pain Based on Differential Characteristics of Electrodermal Activity,” IEEE Trans. Biomed. Eng., vol. 68, no. 10, pp. 3122–3130, Oct. 2021, doi: 10.1109/TBME.2021.3065218

work page doi:10.1109/tbme.2021.3065218 2021
[30]

Laird and James H

N. M. Laird and J. H. Ware, “Random-Effects Models for Longitudinal Data,” Biometrics, vol. 38, no. 4, pp. 963–974, 1982, doi: 10.2307/2529876

work page doi:10.2307/2529876 1982