Monitoring physiological signals during physical activity

OData support
Dr. Jobbágy Ákos Andor
Department of Measurement and Information Systems

This BSc thesis is about a design, creat and test a chest-strap construction, which is able to monitoring physiological signals during physical activity. During my work I used the ArguScan Ltd.’s ArguStress Sport-Pro Kayak chest-strap, which, likes a Holter, able to measure ECG, to implement my own design PPG sensor, for PWTT (Pulse Wave Transit Time) measurement.

The PPG sensor contains 3 LED, 1 photodiode and an ADPD104 PPG AFE (analog front-end) microcontroller. During the creation of my sensor, I tried different combina-tion of parts’ parameters and sensor layouts, to improve the measured PPG signal. To analyze the measured ECG and PPG signals, I used the ArguScan Ltd.’s arrhythmia analyzer software. I implemented my own software modul into the analyzer software to be able to calculate PWTT during the offline and online monitoring. Because the PPG sensor is sensitive for movement, it’s hard to detect the PPG waves’ maximum, which needs to calculate the PWTT. In order to PPG waves’ maximum detection, I split-up the PPG signal by using the ECG signal’s R peaks, calculate average of the last N pieces of PPG section and in the calculated average I detected the maximum by using the least squares method.

I implemented an inspector feature into the chest-strap’s control software, which helps the user with vibration alerts to stay in a given heart rate interval.

I tested the chest-strap operation with step test on bike-ergometer. During the tests I examined the PPG signal’s quality in aspect of sensor placement and movement sensibi-lity, and the change of the PWTT as a function of pulse rate change.

I created a program in Matlab, which can helps to doctors to compare his patient’s pre-vious ECG records and detect the changes what can be occured in years. I tried diffe-rent methods to split- up the ECG by cardiac cycle to find the better way, which defor-mates less the averaged ECG signal.


Please sign in to download the files of this thesis.