Tiny-HR: Towards an interpretable machine learning pipeline for heart rate estimation on edge devices

TL;DR

This paper presents a lightweight, interpretable machine learning pipeline for heart rate estimation on low-power edge devices, achieving significant energy and time efficiency improvements over traditional methods.

Contribution

The authors develop a compact ML pipeline under 40 kB for heart rate estimation on edge devices, combining neural networks and signal processing for improved efficiency.

Findings

Reduces energy consumption by 82% compared to traditional algorithms.

Decreases inference time by 28% on ESP32 devices.

Achieves a mean absolute error of 3.28 beats per minute.

Abstract

The focus of this paper is a proof of concept, machine learning (ML) pipeline that extracts heart rate from pressure sensor data acquired on low-power edge devices. The ML pipeline consists an upsampler neural network, a signal quality classifier, and a 1D-convolutional neural network optimized for efficient and accurate heart rate estimation. The models were designed so the pipeline was less than 40 kB. Further, a hybrid pipeline consisting of the upsampler and classifier, followed by a peak detection algorithm was developed. The pipelines were deployed on ESP32 edge device and benchmarked against signal processing to determine the energy usage, and inference times. The results indicate that the proposed ML and hybrid pipeline reduces energy and time per inference by 82% and 28% compared to traditional algorithms. The main trade-off for ML pipeline was accuracy, with a mean absolute error (MAE) of 3.28, compared to 2.39 and 1.17 for the hybrid and signal processing pipelines. The ML models thus show promise for deployment in energy and computationally constrained devices. Further, the lower sampling rate and computational requirements for the ML pipeline could enable custom hardware solutions to reduce the cost and energy needs of wearable devices.