Related Experiment Video
Updated: May 25, 2026
06:58
An Application for Pairing with Wearable Devices to Monitor Personal Health Status
Published on: February 3, 2022
Distilling clinically interpretable information from data collected on next-generation wearable sensors
Bryan Haslam1, Ankit Gordhandas, Catherine Ricciardi
1Computational Physiology and Clinical Inference Group, ResearchLaboratory of Electronics, Massachusetts Institute of Technology, CambridgeMA, USA. bhaslam@mit.edu
Summary
Wearable sensors collect extensive physiological data, necessitating on-chip processing. This study demonstrates deriving cardiac output and total peripheral resistance from electrocardiogram and blood pressure signals for ambulatory monitoring.
Area of Science:
- Biomedical Engineering
- Cardiovascular Physiology
- Wearable Technology
Background:
- Medical electronic systems generate large datasets from diverse sensors, often in wearable formats.
- Low-power, wireless, and extended-duration monitoring requires on-chip processing to transmit essential clinical information, not raw data.
Purpose of the Study:
- To present an information processing method for continuous, high-sampling-rate data from wearable devices.
- To derive key cardiovascular metrics like cardiac output and total peripheral resistance from ambulatory monitoring data.
Main Methods:
- Utilized a wearable cardiac and motion monitor (ECG, 3-axis acceleration) and a Portapres continuous blood pressure monitor.
- Developed algorithms to estimate physical activity, generate robust heart rate from noisy ECG and arterial blood pressure (ABP) waveforms, and calculate a signal abnormality index.
- Derived cardiac output (CO) and total peripheral resistance (TPR) from heart rate, pulse pressure, and mean arterial blood pressure.
Main Results:
- Collected data from 10 healthy subjects using wearable sensors.
- Successfully derived cardiovascular quantities (CO, TPR) that exhibited physiologically consistent variations with physical activity.
- Demonstrated the feasibility of estimating CO and TPR from ambulatory ECG and ABP signals.
Conclusions:
- On-chip processing of wearable sensor data enables the derivation of clinically relevant cardiovascular parameters.
- The developed methods show promise for robust ambulatory monitoring and interpretation of cardiovascular health.
- Further research is needed to correlate derived values with specific cardiac health states.
