PPG-based Blood Pressure (BP) estimation is a challenging biosignal processing task for low-power devices such as wearables. State-of-the-art Deep Neural Networks (DNNs) trained for this task implement either a PPG-to-BP signal-to-signal reconstruction or a scalar BP value regression and have been shown to outperform classic methods on the largest and most complex public datasets. However, these models often require excessive parameter storage or computational effort for wearable deployment, exceeding the available memory or incurring too high latency and energy consumption. In this work, we describe a fully-automated DNN design pipeline, encompassing HW-aware Neural Architecture Search (NAS) and Quantization, thanks to which we derive accurate yet lightweight models, that can be deployed on an ultra-low-power multicore System-onChip (SoC), GAP8. Starting from both regression and signal-to-signal state-of-the-art models on four public datasets, we obtain optimized versions that achieve up to 4.99% lower error or 73.3 6% lower size at iso-error. Noteworthy, while the most accurate SoA network on the largest dataset can not fit the GAP8 memory, all our optimized models can; our most accurate DNN consumes as little as 0.37 mJ while reaching the lowest MAE of 8.08 on Diastolic BP estimation.

Burrello, A., Carlucci, F., Pollo, G., Wang, X., Poncino, M., Macii, E., et al. (2024). Optimization and Deployment of Deep Neural Networks for PPG-based Blood Pressure Estimation Targeting Low-power Wearables. Institute of Electrical and Electronics Engineers Inc. [10.1109/biocas61083.2024.10798404].

Optimization and Deployment of Deep Neural Networks for PPG-based Blood Pressure Estimation Targeting Low-power Wearables

Burrello, Alessio;Carlucci, Francesco;Benini, Luca;Pagliari, Daniele Jahier
2024

Abstract

PPG-based Blood Pressure (BP) estimation is a challenging biosignal processing task for low-power devices such as wearables. State-of-the-art Deep Neural Networks (DNNs) trained for this task implement either a PPG-to-BP signal-to-signal reconstruction or a scalar BP value regression and have been shown to outperform classic methods on the largest and most complex public datasets. However, these models often require excessive parameter storage or computational effort for wearable deployment, exceeding the available memory or incurring too high latency and energy consumption. In this work, we describe a fully-automated DNN design pipeline, encompassing HW-aware Neural Architecture Search (NAS) and Quantization, thanks to which we derive accurate yet lightweight models, that can be deployed on an ultra-low-power multicore System-onChip (SoC), GAP8. Starting from both regression and signal-to-signal state-of-the-art models on four public datasets, we obtain optimized versions that achieve up to 4.99% lower error or 73.3 6% lower size at iso-error. Noteworthy, while the most accurate SoA network on the largest dataset can not fit the GAP8 memory, all our optimized models can; our most accurate DNN consumes as little as 0.37 mJ while reaching the lowest MAE of 8.08 on Diastolic BP estimation.
2024
2024 IEEE Biomedical Circuits and Systems Conference, BioCAS 2024
1
5
Burrello, A., Carlucci, F., Pollo, G., Wang, X., Poncino, M., Macii, E., et al. (2024). Optimization and Deployment of Deep Neural Networks for PPG-based Blood Pressure Estimation Targeting Low-power Wearables. Institute of Electrical and Electronics Engineers Inc. [10.1109/biocas61083.2024.10798404].
Burrello, Alessio; Carlucci, Francesco; Pollo, Giovanni; Wang, Xiaying; Poncino, Massimo; Macii, Enrico; Benini, Luca; Pagliari, Daniele Jahier...espandi
File in questo prodotto:
Eventuali allegati, non sono esposti

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/11585/1005032
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact