Signal processing for wearable sensing of heart rhythm and function

We are developing a novel wearable sensor based on wearable sensors for monitoring of vital life signs such as heart rate and breathing. One of our aims is to constantly monitor mechanical heart function and stroke volume in a home setting for patients with heart failure that are receiving therapy such as medication. We recently demonstrated a proof-of-principle of our sensing method in a laboratory environment and in the coming months we aim to develop our proof-of-principle device into a wearable prototype. To enable detection of heart rate we need to separate signal modulations originating from breathing and heart motion. We also want to derive the stroke volume (pumped out blood bolus per cycle) from the signal modulation. For this purpose machine learning based regression techniques will be applied to simultaneous Magnetic Resonance Imaging (MRI) dynamic heart scans and sensor measurements on human volunteers.

Assignment

We are looking for a student with a strong background in signal processing to develop a robust algorithm for detection and quantification of heart contraction from a wearable sensor that can integrated in clothing to be used in a home setting or even during exercise. To test the validity of the measurement outcomes, the student will also design and perform experiments in healthy controls to compare physiology measurements with existing methods such as ECG or MRI.

Requirements

- Strong background in signal processing, electromagnetics, and an interest in biophysics. - Experience and affinity with Matlab or Python - Strong experimental skills

Contact

dr.ir. Rob Remis

Terahertz Sensing Group

Department of Microelectronics

Last modified: 2020-11-07