The Assessment and Reduction of Motion Artifact in Dry Contact Biopotential Electrodes
Research output: Book/Report › Doctoral thesis › Collection of Articles
|Place of Publication||Tampere|
|Publisher||Tampere University of Technology|
|Number of pages||76|
|Publication status||Published - 10 Sep 2015|
|Publication type||G5 Doctoral dissertation (article)|
|Name||Tampere University of Technology. Publication|
|Publisher||Tampere University of Technology|
Solving the issue of motion artifact will improve signal quality and reliability for wearable systems and, due to integration and reusability, would reduce costs. These two factors, when combined, would enable the widespread use of wearable monitoring systems in both the medical context and the consumer-user context. One effect of this will be a reduction in load and costs on health care systems due to improved preventive monitoring and better monitoring of patients in the recovery and rehabilitation phase. A second effect, combined with the information exchanging channels between individuals, will be unforeseen developments in health science due to what can be called the crowdsourcing of some aspect of physical and mental health and fitness.
This thesis aims to further state-of-the art wearable physiological monitoring by aiding motion artifact research and electrode design. To accomplish this aim, investigations into the programmable and repeatable generation of electrode movement in order to generate motion artifact, the effect of impedance current frequency on the relationship between skin-electrode interface impedance and electrode movement and motion artifact, the effect of using an electrode support structure and how its design affects the motion artifact, and the effects of garment parameters such as tightness are presented in this thesis.
A system that generates known and programmable motion of the electrode under controlled circumstances was designed, tested, and after the verification of system functionality, used in subsequent investigations. The presented system generates accurate motion of the electrode and the electrode motion can be observed as both motion artifact and skin-electrode impedance changes.
A real time impedance spectroscopy study of 24 impedance current frequencies between 25 Hz and 1 MHz was done on electrodes subject to accurately known motion generated by the designed system in order to find the impedance current frequencies most suited to motion artifact studies.
During this research, a hypothesis was formed that states that an electrode with a structural design that restricts epidermis deformation by trapping the epidermis under the electrode area can reduce motion artifact. Different electrode support structures were designed in order to test this hypothesis. The electrodes with support structures were subjected to system-generated motion and the resulting data were analyzed for the verification of support structure functionality and the hypothesis.
Electrodes that were supported by a tight garment-mimicking elastic straps were studied under subjectgenerated movement and at various clothing tightness levels. The same study was used to understand the effect of using padding between the garment and the electrode.
The motion artifact generation system was seen to be successful in accurately generating electrode motion, thus motion artifact, which was programmable and repeatable. The electrode mounting force monitoring proved to be an important functionality as the mounting force was seen to affect the motion artifact.
Skin-electrode impedance was found to correlate well with electrode motion in current frequencies between 17 kHz and 1 MHz. While the correlation between impedance and motion artifact was lower than the correlation between impedance and electrode motion, it was also highest in this frequency band.
Electrode support structure design is seen to be an important factor to consider when designing the electrode, and the electrodes that came closest to fulfilling the design criteria of the hypothesis were the best functioning electrodes. The hypothesis is seen to be promising and electrodes that distributed skin deformation over a large area and/or restrict epidermis deformation were found to reduce motion artifact.
In the presented studies, the pressures under those electrodes that were found to be the most effective in reducing motion artifact differed between experiments yet stayed in a range between 5 mmHg -36 mmHg (0.66 kPa – 4.80 kPa). A simple guideline is that the electrode should be attached firmly but not so firmly that it becomes uncomfortable. This guideline fitted well with the pressure levels found for each experiment.
The presented Motion Artifact Generation and Assessment System can be used for research or commercial purposes, furthering the research on motion artifact and aiding in the successful design of motion artifact resilient electrodes. The issue of which are the best current frequencies to use to measure skin-electrode interface impedance in motion artifact research has been clarified. Possible means of reducing motion artifact at its origin by using structural electrode designs that restrict epidermis deformation is hypothesized and proven worthy of further research. The importance of garment design and guidelines for use are given and tightness recommendations presented. The thesis presents methodology for the furthering of the understanding of motion artifact and electrode design that will eventually make wearable monitoring systems widespread over a large range of applications and a large number of users.