Heart rate (HR) estimation from photoplethysmography (PPG) signals is a key feature of modern wearable devices for health and wellness monitoring. While deep learning models show promise, their performance relies on the availability of large datasets. We present EnhancePPG, a method that enhances state-of-the-art models by integrating self-supervised learning with data augmentation (DA). Our approach combines self-supervised pre-training with DA, allowing the model to learn more generalizable features, without needing more labelled data. Inspired by a U-Net-like autoencoder architecture, we utilize unsupervised PPG signal reconstruction, taking advantage of large amounts of unlabeled data during the pre-training phase combined with data augmentation, to improve state-of-the-art models' performance. Thanks to our approach and minimal modification to the state-of-the-art model, we improve the best HR estimation by 12.2%, lowering from 4.03 Beats-Per-Minute (BPM) to 3.54 BPM the error on PPG-DaLiA. Importantly, our EnhancePPG approach focuses exclusively on the training of the selected deep learning model, without significantly increasing its inference latency.

EnhancePPG: Improving PPG-based Heart Rate Estimation with Self-Supervision and Augmentation / Benfenati, Luca; Belloni, Sofia; Burrello, Alessio; Kasnesis, Panagiotis; Wang, Xiaying; Benini, Luca; Poncino, Massimo; Macii, Enrico; Pagliari, Daniele Jahier. - ELETTRONICO. - (2025), pp. 1-5. (Intervento presentato al convegno 2025 IEEE 7th International Conference on Artificial Intelligence Circuits and Systems (AICAS) tenutosi a Bordeaux (FRA) nel 28-30 April 2025) [10.1109/aicas64808.2025.11173112].

EnhancePPG: Improving PPG-based Heart Rate Estimation with Self-Supervision and Augmentation

Benfenati, Luca;Belloni, Sofia;Burrello, Alessio;Benini, Luca;Poncino, Massimo;Macii, Enrico;Pagliari, Daniele Jahier
2025

Abstract

Heart rate (HR) estimation from photoplethysmography (PPG) signals is a key feature of modern wearable devices for health and wellness monitoring. While deep learning models show promise, their performance relies on the availability of large datasets. We present EnhancePPG, a method that enhances state-of-the-art models by integrating self-supervised learning with data augmentation (DA). Our approach combines self-supervised pre-training with DA, allowing the model to learn more generalizable features, without needing more labelled data. Inspired by a U-Net-like autoencoder architecture, we utilize unsupervised PPG signal reconstruction, taking advantage of large amounts of unlabeled data during the pre-training phase combined with data augmentation, to improve state-of-the-art models' performance. Thanks to our approach and minimal modification to the state-of-the-art model, we improve the best HR estimation by 12.2%, lowering from 4.03 Beats-Per-Minute (BPM) to 3.54 BPM the error on PPG-DaLiA. Importantly, our EnhancePPG approach focuses exclusively on the training of the selected deep learning model, without significantly increasing its inference latency.
2025
979-8-3315-2424-1
File in questo prodotto:
File Dimensione Formato  
enhancePPG_fullpaper_closed.pdf

accesso riservato

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 4.61 MB
Formato Adobe PDF
4.61 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
enhancePPG_fullpaper_open.pdf

accesso aperto

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: Pubblico - Tutti i diritti riservati
Dimensione 1.35 MB
Formato Adobe PDF
1.35 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/3003608