TEXTILE CAPACITIVE ELECTROCARDIOGRAPHY FOR AN

AUTOMOTIVE ENVIRONMENT

Bhavin Chamadiya, Stephan Heuer

*Research & Development, Daimler AG, Hanns-klemm str. 45, Boeblingen, Germany

FZI Forschungszentrum Informatik, Karlsruhe, Germany

Manfred Wagner*, Ulrich G. Hofmann

Institute for Signal Processing, University of Lübeck, Lübeck, Germany

Keywords: Non-contact electrocardiography (CCECG), Textile electrode, Realistic driving situation, In-car vital

parameter monitoring.

Abstract: Growing mobility demand in the western world intensifies the concern for automotive safety and health

monitoring in daily life. This study shows one solution to integrate textile based, capacitively coupled

electrocardiography into a car considering a real automotive environment. Electrodes based on conductive

textile were integrated into a car seat. Contact ECG and non-contact CCECG measurements were done in

different situations like driving high speed on highways, on surface streets, with various clothes and others.

The influences of various car functionalities on the measurement were detected. Feasibility to measure

Electrocardiography is discussed to evaluate continuous non-contact monitoring for safety, healthcare and

comfort as well.

1 INTRODUCTION

Automotive safety is a crucial topic, with human

mobility difficult to imagine without individualised

automobiles. Even though a growing number of

passive car safety solutions have been implemented

over the past decades and led to a decrease in

fatalities the human driver is still the primary cause

for accidents.

A number of biomedical and monitoring systems

have been incorporated in automobiles for

healthcare (D’Angelo et al., 2010) as well as safety

(Lee et al., 2007) improvement. However,

physiological monitoring is still a nascent tool for

real automotive healthcare and safety, mainly due to

difficult handling of monitoring equipment outside a

laboratory.

As such, non-contact measurements of driver’s

vital parameters by unobtrusive monitoring might

improve general traffic safety immensely.

Consequently the focus of this study was to

integrate capacitive electrodes for

electrocardiography in a real automotive

environment. Vital sign recordings have been

performed while driving in several real world

situations.

2 MATERIALS AND METHODS

2.1 Capacitively Coupled

Electrocardiography (CCECG)

The seemingly only customer acceptable method to

acquire ECG signals from the driver of a car would

be a non-contact one, since it avoids the bodily

contact and limitations of a regular monitoring

device. We chose, due to their simpler

implementation, a capacitive coupling (CC) to the

drivers bio-potentials, thus avoiding any type of

galvanic connection between the driver and the

measurement system (Harland et al., 2002).

Unlike conventional ECG systems that use low

impedance electrical contacts to acquire electrical

signals from the body‘s surface, CCECG systems

require a capacitive (very high impedance) contact.

Hence body and electrode form a plate capacitor to

carry a signal from the body to a very high input

422

Chamadiya B., Heuer S., Wagner M. and Hofmann U..

TEXTILE CAPACITIVE ELECTROCARDIOGRAPHY FOR AN AUTOMOTIVE ENVIRONMENT.

DOI: 10.5220/0003194504220425

In Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES-2011), pages 422-425

ISBN: 978-989-8425-37-9

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

impedance preamplifier (Lim et al., 2006). Signals

from two preamplifiers are fed to a differential

amplifier and thus remove most of the common

mode signal. To further improve common mode

rejection, a driven seat circuit is used (Kim et al.,

2005). Additionally the signal is algorithmically

processed in real time in order to remove other

unwanted signals from it. A sketch of the potential

integration of sensors is depicted in figure 1,

including steering wheel integrated biosignal

measurement (Heuer et al., 2010).

Figure 1: Textile CCECG integration concept in INSITEX

project.

Figure 2: Structure of the textile electrode (top),

Implemented electrode with preamplifier module

(bottom).

2.2 Textile Electrode

Intense testing of rigid PCB electrodes led to the

conclusion, that a flexible and soft electrode

structure is essential for a useful CCECG. Thus our

electrode is made up of three layers of conductive

textile for electrode (sensor area), guard and ground

respectively as depicted in figure 2 (top). The

conductive layers are isolated against each other

with insulating PU films. The textile is washable and

breathable Eblocker from Novonic (E-Blocker from

novonic

Novonic, 2010).

The three conductive layers of the assembly are

attached to the preamplifier module by their

respective connection through three conductive snap

fasteners.

2.3 System Integration

To make the integration suitable for both a seat in

lab and car, electrodes are fastened into a seat cover

of the Mercedes Benz C-class (W204 series).

Figure 3: The electrodes and incised seat cover with

Velcro (top left), the electrodes when fastened into the seat

cover (top right), the seat cover applied on the car seat.

To hold the electrodes tight to the seat cover,

Velcro ribbons are attached around their rims and

the seat cover incision as shown in figure 3. Hence it

maintains a stable, robust and yet removable

configuration. Signals from the electrodes are taken

through the visible blue ribbon cables to commercial

9 pin D-sub connectors (figure 3).

2.4 Measurement Setup

The seat cover is applied to a C class car seat both in

lab and in a real car as well; figure 3 shows the

arrangement inside the car.

The outputs of the electrodes are fed to a signal

Velcros

Fastened

electrodes

TEXTILE CAPACITIVE ELECTROCARDIOGRAPHY FOR AN AUTOMOTIVE ENVIRONMENT

423

processing box. Both inputs to the box are filtered

with high pass filter (0.8 Hz) to remove baseline

drift and DC offset. The differential of the signal is

further filtered with a band pass and 50Hz notch

filter before any amplification.

The pre-processed analog signal is digitised with

a data acquisition card (NI U-9162) for further

processing and final display in LabVIEW 2009.

The driven seat textile electrode is spread

beneath the cushion cover of the car seat to reduce

common mode noise.

3 MEASUREMENT RESULTS

The CCECG measurement system has successfully

gathered data for various real driving situations.

Measurements in conditions like driving on

highways, surface street driving, enabling car

functions while driving, by putting on different

clothes on the subject etc. have been performed.

Non-contact ECG recordings under different road

conditions are displayed in figure 4.

Figure 4: The CCECG result on highway (top), city street

in good condition (middle), uneven street in city (bottom).

During all the measurements, the driving subject

wore a t-shirt (100% cotton) with 0.68 mm thickness

and a pant (wool and polyester) with 0.29 mm

thickness. The subject was 56 years old with height

and weight, 182cm and 92kg respectively and was

driving at speed of 100-120 km/h in normal

condition on highway and in the city street 40-60

km/h.

Due to the massive amount of electrical

appliances in contemporary cars, it was crucial to

test the influence of those functionalities on the

measurement during real driving. Figure 5 shows the

low influence of telephone usage (top), automatic

seat adjustment by DC motors (middle), and

electrical seat heating (bottom).

Complete of many more functionalities will be

presented elsewhere.

Figure 5: CCECG results from different driving activities:

hands-free telephoning (top), adjusting the driver's seat

(middle), with seat heating level 1 & 2 enabled (bottom).

Figure 6 shows a clear and strong effect of the

drivers clothing on the signal quality, since this will

strongly influence the capacitive coupling and thus

the monitoring result. The clothes consisted of a

winter jacket made up of 85% polyester and 15%

polyamide with 0.35 mm thickness, a rain jacket

made up of 100% nylon lined with 0.44 mm PU and

a sportcoat made up of 55% linen and 45% viscose

with 0.85 mm thickness.

Figure 6: Signal to noise ration of the CCECG signal with

various clothes and respective CCECG Signals.

BIODEVICES 2011 - International Conference on Biomedical Electronics and Devices

424

4 DISCUSSION AND

CONCLUSIONS

A practical approach to integrate a textile CCECG

system has been implemented in our experiments.

Measurements during common driving situations

could be demonstrated. It can be observed that only

driving on bad and bumpy roads (hence strong car

and body movement) did interfere with the

monitoring by causing rapid base line drifting.

Some of the seat functions also have influences

from mild to intense on the signal as depicted in

figure 6. Interference from automatic seat

adjustment while driving was minor as the function

is enabled by DC motors driven with 50Hz pulses

(Chamadiya et al., 2008) and the body maintained a

stable contact with the electrodes in the seat. Low

frequency base line drift and 50Hz hum noise from

the DC motors were filtered by the monitoring

system. Seat heating with level 1 and 2 showed

major effects on the monitoring results as they had a

PWM signal of 24 Hz (Chamadiya et al., 2008).

However, we speculate that interfacing our

monitoring setup with the car‘s own controls and

sensors could alleviate the severity of these

distortions.

Clothes in general did have an impact in the

signal-to-noise ratio, but did not prohibit heart rate

monitoring and require further investigation.

Summing up, this work demonstrates promising

non-contact monitoring results with a textile

CCECG system in real world driving situations. The

system shows a strong potential to be incorporated

in a car for long term ECG measurement for safety

and healthcare.

ACKNOWLEDGEMENTS

We would like to thank the BMBF (German

Ministry for Education and Research) for funding

the work of INISTEX project.

REFERENCES

D’Angelo, L., Parlow, J., Spiessl, W., Hoch, S., Lüth, T.,

2010. A system for unobstrusive In-Car Vital

Parameter Acquisition and processing. 4

International Conference on Pervasive Computing

Technologies for Healthcare. Garching, Germany 22-

25 march 2010.

Lee, H., Kim, J., Kim, Y., Baek, H., Ryu, M., Park, K.,

2007. The relationship between HRV parameter and

stressful driving situation in the real road. 6

International Special Topic Conference on ITAB.

Tokyo, Japan 2007.

Harland, C. J., Clark, T. D., Prance, R. J., 2002. Electric

potential probes-new directions in the remote sensing

of the human body. Measurement Science Technology

13, 2002, pp.163-169.

Lim, Y., Kim, K. and Park, K.,. (2006) ECG measurement

on a chair without conductive contact. IEEE

Transactions on Biomedical Engineering, Volume 53

(No.5/May), pp.956-959.

Chamadiya, B., Heuer, S., Hofmann, U., Wagner, M.,

2008. Toward a capacitively coupled

Electrocardiography system for car seat integration.

European conference of the international

Federation for Medical and Biological Engineering.

Antwerp, Belgium 23-27 November 2008. Springer-

Verlag Berlin Heidelberg 2008.

Heuer, S., Chamadiya, B., Gharbi, A., Kunze, C., Wagner,

M., 2010. Unobtrusive in-vehicle biosignal

instrumentation for advance driver assistance and

active safety. IEEE EMBS Conference on Biomedical

Engineering and Science. Kuala Lumpur, Malaysia 30

November-2 December 2010.

Kim, K., Lim, Y., Park, K., 2005. Common mode noise

cancellation for electrically non-contact ECG

measurement system on a chair. In TEMPLATE’06,

Annual Conference on Engineering in Medicine

and Biology. Shanghai, China 1-4 September 2005.

IEEE.

E-Blocker, 2010. E-Blocker product information. [online]

Available at: <http://www.novonic.de/

web/NovNeu_FDI.nsf/id/pa_novonic_e_blocker_e.ht

ml> [Accessed 22 August 2010].

TEXTILE CAPACITIVE ELECTROCARDIOGRAPHY FOR AN AUTOMOTIVE ENVIRONMENT

425