Generalizable deep learning for photoplethysmography-based blood pressure estimation—A benchmarking study
Machine Learning: Health · 01.12.2025 · 1 (1) · 010501
DOI: 10.1088/3049-477x/ae01a8 ↗Kurz gefasst
Fachartikel in der Zeitschrift Machine Learning: Health, online erschienen am 15.09.2025. Verfasst von Mohammad Moulaeifard und 2 weiteren Autor:innen. Beteiligte Einrichtung: Carl von Ossietzky Universität Oldenburg, Oldenburg, Germany. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: British Heart Foundation, European Partnership on Metrology, co-financed from the European Union’s Horizon Europe Research and Innovation Programme and by the Participating States.
Abstract (Originalsprache)
Photoplethysmography (PPG)-based blood pressure (BP) estimation represents a promising alternative to cuff-based BP measurements. Recently, an increasing number of deep learning (DL) models have been proposed to infer BP from the raw PPG waveform. However, these models have been predominantly evaluated on in-distribution (ID) test sets, which immediately raises the question of the generalizability of these models to external datasets. To investigate this question, we trained five DL models on the recently released PulseDB dataset, provided ID benchmarking results on this dataset, and then assessed their out-of-distribution (OOD) performance on several external datasets. The best model (XResNet1d101) achieved ID mean absolute errors (MAEs) of 9.0 and 5.8 mmHg for systolic and diastolic BP, respectively, on PulseDB with subject-specific calibration, and 13.9 and 8.5 mmHg, respectively, without calibration. The equivalent MAEs on external test datasets without calibration ranged from 10.0 to 18.6 mmHg (SBP) and 5.9 to 10.3 mmHg (DBP). Our results indicate that performance is strongly influenced by the differences in BP distributions between datasets. We investigated a simple way of improving performance through sample-based domain adaptation and put forward recommendations for training models with good generalization properties. With this work, we hope to educate more researchers about the importance and challenges of OOD generalization.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Carl von Ossietzky Universität Oldenburg, Oldenburg, Germany.
- University of Cambridge, Cambridge, United Kingdom.
ORCID-Profile: Mohammad Moulaeifard, Peter H Charlton, Nils Strodthoff
Förderung
- British Heart Foundation
- European Partnership on Metrology, co-financed from the European Union’s Horizon Europe Research and Innovation Programme and by the Participating States
Zitieren
APA
Moulaeifard, M., Charlton, P. H., & Strodthoff, N. (2025). Generalizable deep learning for photoplethysmography-based blood pressure estimation—A benchmarking study. Machine Learning: Health, 1(1), 010501. https://doi.org/10.1088/3049-477x/ae01a8
Vancouver
Moulaeifard M, Charlton PH, Strodthoff N. Generalizable deep learning for photoplethysmography-based blood pressure estimation—A benchmarking study. Machine Learning: Health. 2025;1(1):010501. doi:10.1088/3049-477x/ae01a8
BibTeX
@article{Moulaeifard2025,
author = {Moulaeifard, Mohammad and Charlton, Peter H and Strodthoff, Nils},
title = {Generalizable deep learning for photoplethysmography-based blood pressure estimation—A benchmarking study},
journal = {Machine Learning: Health},
year = {2025},
volume = {1},
number = {1},
pages = {010501},
doi = {10.1088/3049-477x/ae01a8}
}
EUVIMED