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

2026-04-11

AI Summary

This work addresses the challenge of achieving high-accuracy, low-power, and privacy-preserving photoplethysmography (PPG)-based blood pressure estimation on resource-constrained wearable devices. The authors propose an end-to-end automated deep neural network optimization framework that integrates hardware-aware neural architecture search (NAS), model pruning, and mixed-precision search (MPS). This approach enables, for the first time, fully automatic deployment of compact blood pressure prediction models tailored to ultra-low-power multicore SoCs such as GAP8, while supporting user-specific fine-tuning. Experimental results demonstrate up to an 83× reduction in model parameters, a 7.99% improvement in estimation error, memory footprint under 55 kB, inference latency of only 142 ms, and energy consumption of 7.25 mJ. Personalized fine-tuning further enhances accuracy by up to 64%.

Technology Category

• Machine Learning: Hardware-aware ML
• Search and Optimization: Learning to Search
• Computer Vision: Learning & Optimization for CV

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Security and Privacy: Privacy-enhancing technologies

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.

Problem

• PPG-based blood pressure estimation
• wearable devices
• resource-constrained
• on-board processing
• deep neural networks

Innovation

• hardware-aware neural architecture search
• mixed-precision search
• model pruning
• on-device blood pressure estimation
• ultra-low-power SoC