Self-supervised autoencoder network for robust heart rate extraction from noisy photoplethysmogram: Applying blind source separation to biosignal analysis

Biosignals can be viewed as mixtures measuring particular physiological events, and blind source separation (BSS) aims to extract underlying source signals from mixtures. This paper proposes a self-supervised multi-encoder autoencoder (MEAE) to separate heartbeat-related source signals from photople...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Computers in biology and medicine Ročník 199; s. 111319
Hlavní autoři: Webster, Matthew B., Lee, Dongheon, Lee, Joonnyong
Médium: Journal Article
Jazyk:angličtina
Vydáno: United States Elsevier Ltd 01.12.2025
Témata:
Biosignal processing
Blind source separation
Denoising
Heart rate detection
Photoplethysmogram
Self-supervised learning
Biosignal processing
Heart rate detection
Blind source separation
Self-supervised learning
Photoplethysmogram
Denoising
ISSN:0010-4825, 1879-0534, 1879-0534
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Biosignals can be viewed as mixtures measuring particular physiological events, and blind source separation (BSS) aims to extract underlying source signals from mixtures. This paper proposes a self-supervised multi-encoder autoencoder (MEAE) to separate heartbeat-related source signals from photoplethysmogram (PPG), enhancing heart rate (HR) detection in noisy PPG data. The MEAE is trained on PPG signals from a large open polysomnography database without any pre-processing or data selection. The trained network is then applied to a noisy PPG dataset collected during the daily activities of nine subjects and a surgical dataset comprising 4,681 patients. The extracted heartbeat-related source signal significantly improves HR detection as compared to the original PPG. The absence of pre-processing and the self-supervised nature of the proposed method, combined with its strong performance, highlight the potential of MEAE for BSS in biosignal analysis. •Biosignals are often difficult to analyze due to presence of unknown noises including motion artifacts.•Blind source separation via self-supervised autoencoder allows to separate out information from biosignals.•Blind source separation applied to noisy photoplethysmogram successfully isolated heartbeat related source.•Blind source separation without any directives (supervision in deep learning) can result in extraction of information present in the training data.•Analyses of large biosignal databases can benefit from self-supervised blind source separation due to the lack of pre-processing or data selection.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0010-4825
1879-0534
1879-0534
DOI:10.1016/j.compbiomed.2025.111319