EUVIMEDEUVIMED entdecken
EvaluationsstudieOpen Access · CC BY 4.0

Evaluation of HRV–PRV agreement with autoencoder-based signal quality assessment in resting and dynamic conditions

Bartosz Olszewski iD, Agnieszka Uryga iD, Cyprian Mataczyński iD, Magdalena Kasprowicz iD

Physiological Measurement · 31.08.2026 · 47 (8) · 085023

DOI: 10.1088/1361-6579/ae95ae ↗

Kurz gefasst

Evaluationsstudie in der Zeitschrift Physiological Measurement, online erschienen am 19.08.2026. Verfasst von Bartosz Olszewski und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedical Engineering, Faculty of Fundamental Problems of Technology, Wroclaw University of Science and Technology, Wroclaw, Poland. Hauptthemen laut MEDLINE: Autoencoder, Heart Rate, Signal Processing, Computer-Assisted, Rest. Die Arbeit nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministerstwo Edukacji i Nauki.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective. Pulse-to-pulse intervals obtained from continuous non-invasive blood pressure (CNBP) signals can be used to derive pulse rate variability (PRV). CNBP signals are susceptible to noise and motion artifacts, which can reduce the reliability of PRV and its agreement with heart rate variability (HRV) derived from electrocardiography (ECG). In this study, we applied a pulse-wise, unsupervised quality assessment method for CNBP signals based on a convolutional autoencoder integrating waveform morphology with pulse-level metadata. The aim was to improve the agreement between PRV and HRV metrics. Approach. Two datasets of CNBP and ECG recordings were analyzed: long baseline measurements (15 volunteers, average 40 min) and short recordings with postural changes (53 volunteers; 5 min squat-stand). A convolutional autoencoder was used for pulse-wise quality assessment. HRV metrics were derived from ECG, and PRV metrics were derived from CNBP before and after quality-based pulse selection (PRV clean ). The analysis included standard time-domain indices (SDNN, RMSSD), frequency-domain measures, and nonlinear Poincaré plot descriptors (SD1, SD2). The agreement between HRV and PRV or PRV clean was evaluated by Bland–Altman analysis. Main results. PRV systematically overestimated variability compared to HRV. During resting conditions, RMSSD and SDNN were lower for HRV (47.7 ± 20.6 ms and 71.4 ± 27.4 ms) than for PRV (61.7 ± 26.7 ms, p = 0.005 and 77.8 ± 30.0 ms, p = 0.003, respectively). After applying quality-based pulse selection, bias decreased from − 14.0 to − 7.3 ms for RMSSD and from − 6.4 to − 3.2 ms for SDNN, improving PRV–HRV agreement. During postural changes, PRV also overestimated HRV, but the effect of quality-based pulse selection was less pronounced. Significance. The proposed approach enhances CNBP signal quality without requiring manual labeling, making it suitable for real-world applications. Improved agreement between PRV and HRV may have important clinical implications, particularly when ECG acquisition is impractical or unavailable.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

AutoencoderHeart RateSignal Processing, Computer-AssistedRestHumansElectrocardiographyFemaleMaleAdultBlood PressureYoung Adult

Stichwörter der Autor:innen

autoencodercontinuous non-invasive blood pressureelectrocardiogramheart rate variabilitypulse rate variability

Autor:innen und Institutionen

  • Department of Biomedical Engineering, Faculty of Fundamental Problems of Technology, Wroclaw University of Science and Technology, Wroclaw, Poland.
  • Hemolens Diagnostics Sp. z o.o., Wroclaw, Poland.

ORCID-Profile: Bartosz Olszewski, Agnieszka Uryga, Cyprian Mataczyński, Magdalena Kasprowicz

Förderung

  • Ministerstwo Edukacji i Nauki

Zitieren

APA

Olszewski, B., Uryga, A., Mataczyński, C., & Kasprowicz, M. (2026). Evaluation of HRV–PRV agreement with autoencoder-based signal quality assessment in resting and dynamic conditions. Physiological Measurement, 47(8), 085023. https://doi.org/10.1088/1361-6579/ae95ae

Vancouver

Olszewski B, Uryga A, Mataczyński C, Kasprowicz M. Evaluation of HRV–PRV agreement with autoencoder-based signal quality assessment in resting and dynamic conditions. Physiological Measurement. 2026;47(8):085023. doi:10.1088/1361-6579/ae95ae

BibTeX
@article{Olszewski2026,
  author = {Olszewski, Bartosz and Uryga, Agnieszka and Mataczyński, Cyprian and Kasprowicz, Magdalena},
  title = {Evaluation of HRV–PRV agreement with autoencoder-based signal quality assessment in resting and dynamic conditions},
  journal = {Physiological Measurement},
  year = {2026},
  volume = {47},
  number = {8},
  pages = {085023},
  doi = {10.1088/1361-6579/ae95ae}
}
Kostenlos starten →