PulseFi: A Low Cost Robust Machine Learning System for Accurate Cardiopulmonary and Apnea Monitoring Using Channel State Information
Pranay Kocheta, Nayan Sanjay Bhatia, Katia Obraczka
TL;DR
PulseFi tackles non-intrusive vital-sign monitoring by combining amplitude-only Wi-Fi CSI with a compact LSTM engine to estimate heart rate, breathing rate, and apnea events on low-cost hardware. The approach uses a five-stage CSI processing pipeline and a lightweight LSTM architecture to achieve state-of-the-art accuracy on two diverse CSI datasets, including 118 participants with 234-subcarrier resolution. It demonstrates robust performance across distances and postures, with clinically relevant accuracy thresholds (e.g., $97.95\%$ of HR estimates within $1.5$ BPM on ESP-HR-CSI and $99.65\%$ on E-Health; BR MAE $0.09$ breaths/min at $20$ s). The work emphasizes accessibility and efficiency, showing real-time monitoring on devices like the ESP32 and Raspberry Pi, and discusses implications for wider adoption with upcoming Wi-Fi sensing standards such as IEEE 802.11bf.
Abstract
Non-intrusive monitoring of vital signs has become increasingly important in a variety of healthcare settings. In this paper, we present PulseFi, a novel low-cost non-intrusive system that uses Wi-Fi sensing and artificial intelligence to accurately and continuously monitor heart rate and breathing rate, as well as detect apnea events. PulseFi operates using low-cost commodity devices, making it more accessible and cost-effective. It uses a signal processing pipeline to process Wi-Fi telemetry data, specifically Channel State Information (CSI), that is fed into a custom low-compute Long Short-Term Memory (LSTM) neural network model. We evaluate PulseFi using two datasets: one that we collected locally using ESP32 devices and another that contains recordings of 118 participants collected using the Raspberry Pi 4B, making the latter the most comprehensive data set of its kind. Our results show that PulseFi can effectively estimate heart rate and breathing rate in a seemless non-intrusive way with comparable or better accuracy than multiple antenna systems that can be expensive and less accessible.