Evaluation of HRV estimation algorithms from PPG data using neural networks

Koch R, Pfeiffer N, Lang N, Struck M, Amft O, Eskofier B, Wittenberg T (2020)

Publication Language: English

Publication Type: Journal article, Original article

Publication year: 2020

Journal

Current Directions in Biomedical Engineering De Gruyter

Book Volume: 6

Article Number: 20203130

Journal Issue: 3

URI: https://www.degruyter.com/document/doi/10.1515/cdbme-2020-3130/html

DOI: 10.1515/cdbme-2020-3130

Open Access Link: https://www.degruyter.com/document/doi/10.1515/cdbme-2020-3130/html

Abstract

Heart rate variability (HRV) is a powerful measure to gain information on the activation of the central nervous system and is thus a strong indicator for the overall health and emotional state of a person. Currently, the gold standard for HRV analysis is the examination of R-peaks in electrocardiograms (ECG), which requires a placement of electrodes on the torso. This is often impracticable, especially for the use in daily routines or 24/7 measurements. Photoplethysmograms (PPG) are an alternative to ECG assessment and are easier to acquire, e.g. by using fitness trackers or smart watches. Nevertheless, PPG data is more susceptible to motion artifacts. Hence, goal of this work is to develop and evaluate an artificial neural network (ANN) approach to estimate the R-peak locations in complex PPG signals. Public data collections were used as benchmark to compare our ANN-based approach to state-of-the-art methods. Results show that ANNs can improve HRV estimation during motion. HRV estimations from baseline methods (decision-tree based and automatic multiscalebased peak detection) were compared with the best performing neural network (3L-GRU) using the TROIKA dataset with respect to reference parameters obtained from a manual selection of the peaks in ECG data. In most cases, the neural network based HRV estimation was closer to the reference HRV compared to baseline methods (lower μ and σ) Also, σ is smaller for the best performing ANN approach across most HRV parameters. Inclusion of another PPG or acceleration channel did not affect HRV estimation. Although, the neural network learning approach outperforms conventional methods, the examined PPG-based HRV estimation has still accuracy limitations. Nonetheless, the proposed estimation approach opens up new directions for further improvement.

Authors with CRIS profile

Oliver Amft Lehrstuhl für Digital Health Björn Eskofier Machine Learning and Data Analytics Lab Thomas Wittenberg

Involved external institutions

Fraunhofer-Institut für Integrierte Systeme und Bauelementetechnologie (IISB) DE Germany (DE) Fraunhofer-Institut für Integrierte Schaltungen (IIS) DE Germany (DE)

How to cite

APA:

Koch, R., Pfeiffer, N., Lang, N., Struck, M., Amft, O., Eskofier, B., & Wittenberg, T. (2020). Evaluation of HRV estimation algorithms from PPG data using neural networks. Current Directions in Biomedical Engineering, 6(3). https://dx.doi.org/10.1515/cdbme-2020-3130

MLA:

Koch, R., et al. "Evaluation of HRV estimation algorithms from PPG data using neural networks." Current Directions in Biomedical Engineering 6.3 (2020).

BibTeX: Download