loading page

PEPbench -- Open, Reproducible, and Systematic Benchmarking of Automated Pre-Ejection Period Extraction Algorithms
  • +11
  • Robert Richer,
  • Julia Jorkowitz,
  • Sebastian Stühler,
  • Luca Abel,
  • Miriam Kurz,
  • Marie Oesten,
  • Stefan G. Griesshammer,
  • Nils C. Albrecht,
  • Arne Küderle,
  • Christoph Ostgathe,
  • Alexander Kölpin,
  • Tobias Steigleder,
  • Nicolas Rohleder,
  • Bjoern M. Eskofier
Robert Richer
Friedrich-Alexander-Universitat Erlangen-Nurnberg

Corresponding Author:[email protected]

Author Profile
Julia Jorkowitz
Friedrich-Alexander-Universitat Erlangen-Nurnberg
Author Profile
Sebastian Stühler
Friedrich-Alexander-Universitat Erlangen-Nurnberg
Author Profile
Luca Abel
Friedrich-Alexander-Universitat Erlangen-Nurnberg
Author Profile
Miriam Kurz
Friedrich-Alexander-Universitat Erlangen-Nurnberg
Author Profile
Marie Oesten
Friedrich-Alexander-Universitat Erlangen-Nurnberg
Author Profile
Stefan G. Griesshammer
Universitatsklinikum Erlangen Palliativmedizinische Abteilung
Author Profile
Nils C. Albrecht
Technische Universitat Hamburg
Author Profile
Arne Küderle
Friedrich-Alexander-Universitat Erlangen-Nurnberg
Author Profile
Christoph Ostgathe
Universitatsklinikum Erlangen Palliativmedizinische Abteilung
Author Profile
Alexander Kölpin
Technische Universitat Hamburg
Author Profile
Tobias Steigleder
Universitatsklinikum Erlangen Palliativmedizinische Abteilung
Author Profile
Nicolas Rohleder
Friedrich-Alexander-Universitat Erlangen-Nurnberg
Author Profile
Bjoern M. Eskofier
Friedrich-Alexander-Universitat Erlangen-Nurnberg
Author Profile

Abstract

The pre-ejection period (PEP) is a widely used cardiac parameter in psychophysiological research that reflects the duration between the onset of ventricular depolarization and the opening of the aortic valve and is an established marker of sympathetic nervous system activity. While many algorithms for automated PEP extraction from electrocardiography (ECG) and impedance cardiography (ICG) signals have been proposed in the literature, they have not been systematically compared against each other. This lack of standardized algorithm comparisons is due to the absence of open-source algorithms and annotated datasets for evaluating PEP extraction algorithms. To address this issue, we introduce PEPbench, an open-source Python package with different Q-peak and B-point detection algorithms from the literature that can be combined to form comprehensive PEP extraction pipelines, and a standardized framework for evaluating PEP extraction algorithms. We use PEPbench to systematically compare 90 different algorithm combinations. All combinations are evaluated on datasets from two different studies with manually annotated Q-peaks and B-points, which we make publicly available as the first datasets with reference PEP annotations. Our results show that the algorithms can differ vastly in their performance and that the B-point detection algorithms introduce a considerable amount of error. Thus, we suggest that automated PEP extraction algorithms should be used with caution on a beat-to-beat level as their error rates are still relatively high. This highlights the need for open and reproducible benchmarking frameworks for PEP extraction algorithms to improve the quality of research findings in the field of psychophysiology. With PEPbench, we aim to take a first step towards this goal and encourage other researchers to engage in the evaluation of PEP extraction algorithms by contributing algorithms, data, and annotations. Ultimately, we hope to establish a community-driven platform, fostering innovation and collaboration in the field of psychophysiology and beyond.
16 Jan 2025Submitted to Psychophysiology
22 Jan 2025Submission Checks Completed
22 Jan 2025Assigned to Editor
22 Jan 2025Review(s) Completed, Editorial Evaluation Pending
04 Feb 2025Reviewer(s) Assigned