End-to-end Automated Deep Neural Network Optimization for PPG-based Blood Pressure Estimation on Wearables

TL;DR

This paper presents an automated neural network design pipeline that creates compact, accurate BP estimation models suitable for resource-limited wearable devices, enabling privacy-preserving on-device processing.

Contribution

It introduces a hardware-aware NAS, pruning, and mixed-precision search pipeline that optimizes DNNs for ultra-low-power wearable SoCs, achieving significant size reduction and accuracy improvements.

Findings

Models achieve up to 7.99% lower error with 7.5x fewer parameters.

All models fit within 512 kB memory on target SoC.

Inference latency is 142 ms with energy consumption of 7.25 mJ.

Abstract

Photoplethysmography (PPG)-based blood pressure (BP) estimation is a challenging task, particularly on resource-constrained wearable devices. However, fully on-board processing is desirable to ensure user data confidentiality. Recent deep neural networks (DNNs) have achieved high BP estimation accuracy by reconstructing BP waveforms or directly regressing BP values, but their large memory, computation, and energy requirements hinder deployment on wearables. This work introduces a fully automated DNN design pipeline that combines hardware-aware neural architecture search (NAS), pruning, and mixed-precision search (MPS) to generate accurate yet compact BP prediction models optimized for ultra-low-power multicore systems-on-chip (SoCs). Starting from state-of-the-art baseline models on four public datasets, our optimized networks achieve up to 7.99% lower error with a 7.5x parameter reduction, or up to 83x fewer parameters with negligible accuracy loss. All models fit within 512 kB of memory on our target SoC (GreenWaves' GAP8), requiring less than 55 kB and achieving an average inference latency of 142 ms and energy consumption of 7.25 mJ. Patient-specific fine-tuning further improves accuracy by up to 64%, enabling fully autonomous, low-cost BP monitoring on wearables.