Cooperative dry-electrode sensors for multi-lead biopotential and bioimpedance monitoring
M Rapin1,2, M Proença1,3, F Braun1,3, C Meier1, J Solà1,
D Ferrario1, O Grossenbacher1, J-A. Porchet1 and
O Chételat1 1 Swiss Centre for Electronics and Microtechnology, CSEM, Neuchâtel, Switzerland 2 ETHZ, D-HEST, Zürich, Switzerland 3 EPFL, LTS5, Lausanne, Switzerland
E-mail: firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com, olivier. firstname.lastname@example.org, email@example.com and olivier.chetelat@ csem.ch
Received 6 October 2014, revised 20 December 2014
Accepted for publication 7 January 2015
Published 23 March 2015
Cooperative sensors is a novel measurement architecture that allows the acquiring of biopotential signals on patients in a comfortable and easy-tointegrate manner. The novel sensors are defined as cooperative in the sense that at least two of them work in concert to measure a target physiological signal, such as a multi-lead electrocardiogram or a thoracic bioimpedance.
This paper starts by analysing the state-of-the-art methods to simultaneously measure biopotential and bioimpedance signals, and justifies why currently (1) passive electrodes require the use of shielded or double-shielded cables, and (2) active electrodes require the use of multi-wired cabled technologies, when aiming at high quality physiological measurements.
In order to overcome the limitations of the state-of-the-art, a new method for biopotential and bioimpedance measurement using the cooperative sensor is then presented. The novel architecture allows the acquisition of the aforementioned biosignals without the need of shielded or multi-wire cables by splitting the electronics into separate electronic sensors comprising each of two electrodes, one for voltage measurement and one for current injection.
The sensors are directly in contact with the skin and connected together by only one unshielded wire. This new configuration requires one power supply per sensor and all sensors need to be synchronized together to allow them to work in concert.
M Rapin et al
Printed in the UK 767
PMea © 2015 Institute of Physics and engineering in Medicine 2015 36
PMea 0967-3334 10.1088/0967-3334/36/4/767
Special issue papers (internally/externally peer-reviewed) 4 767 783
Institute of Physics and Engineering in Medicine
IOP 0967-3334/15/040767+17$33.00 © 2015 Institute of Physics and Engineering in Medicine Printed in the UK
Physiol. Meas. 36 (2015) 767–783 doi:10.1088/0967-3334/36/4/767 768
After presenting the working principle of the cooperative sensor architecture, this paper reports first experimental results on the use of the technology when applied to measuring multi-lead ECG signals on patients.
Measurements performed on a healthy patient demonstrate the feasibility of using this novel cooperative sensor architecture to measure biopotential signals and compliance with common mode rejection specification accordingly to international standard (IEC 60601-2-47) has also been assessed.
By reducing the need of using complex wiring setups, and by eliminating the presence of central recording devices (cooperative sensors directly sense and store the measured biosignals on the site), the depicted novel technology is a candidate to a novel generation of highly-integrated, comfortable and reliable technologies that measure physiological signals in real-life scenarios.
Keywords: wearable sensor, ECG monitoring, bioimpedance (Some figures may appear in colour only in the online journal) 1. Introduction
In this paper we present a new system architecture based on innovative cooperative dry-electrode sensors (Chételat et al 2011) that are comfortable to wear while showing high signal quality. The sensors are defined as cooperative in the sense that at least two independent devices working in concert are required to measure a physiological signal, in particular biopotentials and bioimpedances. These sensors benefit from simplified connection requirements (no shielded or multi-wire cable, no central electronic recording device), which significantly eases their connection and integration in a garment (Chételat et al 2008).
This section describes the state-of-the-art methods used to acquire biopotentials and bioimpedances and explains why shielded (coaxial) or double-shielded (triaxial) cables with passive electrodes or multi-wire cables with active electrodes are classically required to acquire signals of high quality.
In the second section, the link is made from the state-of-the-art methods to the new system built with cooperative sensors. This architecture, which is the main innovation presented in this paper, is detailed with conceptual schematics blocks and actual electronics implementation examples. Due to the fact that the electronics functions are split in separate cooperative sensors, one of the challenges of the proposed method is to synchronize all sensors together.
Therefore, the second section also covers the method applied to allow the synchronization of the sensors.
Then, the third section shows the first practical results of the synchronized cooperative sensors. The chosen example shows the measurement of electrocardiogram (ECG) leads with cooperative sensors.
Finally, the last section includes a discussion and a conclusion underlining the advantages and drawbacks of the new proposed architecture compared to the state-of-the-art methods. 1.1. State-of-the-art biopotentials measurement 1.1.1. Simplest circuit. Although biopotentials could originate from different sources (e.g.
ECG, electromyogram, electroencephalogram, electrooculogram, etc), this article focuses on
ECG. Of course, any other biopotential would be measured in the same way.
M Rapin et alPhysiol. Meas. 36 (2015) 767 769
The simplest circuit for ECG measurement is shown in figure 1. Basically, the ECG can be modelled as a voltage source located in the body while series impedances model the impedances of the body-skin and electrode-skin interface. Other impedances, like the impedance of the tissues inside the body or the impedance of the wires, are negligible in this context. From this simple circuit, one can see that the measured voltage u1 directly corresponds to the ECG voltage.