ArticlePDF Available

Abstract and Figures

Electroencephalography (EEG) has a wide range of applications in medical diagnosis, and novel form of Human Machine Interfaces (HMI) for controlling prosthetic implants, wheelchairs, and home appliances in various forms of paralysis. However, the current EEG setups are composed of many wires hanging down from the system, and individual electrodes that must be set manually, which is time-consuming. As a result, the overall system is neither comfortable, nor aesthetically appealing. Here, we introduce for the first time, a comfortable textile-based EEG headband system that is soft, conformal to the skin, and comfortable. We present materials and methods for fabrication of multi-layer stretchable e-textile, that interfaces the human epidermis from one side through printed electrodes, and interfaces a rigid PCB island on the second layer. We as well demonstrate a method that allows creation of VIAs (vertical interconnect access) between the layers, using a CO2 laser. All Electrodes are integrated into the headband, and thus there is no need for individual electrode placement, and individual wiring. By screen printing a home-made conductive stretchable ink, patient-specific EEG headbands can be tailor made considering the optimal positioning of the electrodes for each patient. We show that these printed electrodes benefit from a very low skin-electrode impedance, comparable to gold standard Ag/AgCl, or gold cup electrodes, thanks to the high surface area silver flakes used in this work. The e-textile headband interfaces with an EEG acquisition device that captures, amplifies, and transmits the data to an external mobile phone or a PC. Furthermore, the integrated amplification in the textile and the use of an EMF rejection layer on top of the electrodes were shown to reduce the unwanted EM noise that is picked up by the system. We as well show application of the developed headband for usage in Human Machine Interfaces and Sleep Data Acquisition. Altogether, this device is step toward wider use of EEG acquisition devices for daily-use applications.
Content may be subject to copyright.
IEEE SENSORS JOURNAL, VOL. 20, NO. 24, DECEMBER 15, 2020 15107
Wearable and Comfortable e-Textile Headband
for Long-Term Acquisition of Forehead
EEG Signals
Manuel Reis Carneiro , Aníbal T. de Almeida,
Life Senior Member, IEEE
and Mahmoud Tavakoli ,
Member, IEEE
Electroencephalography (EEG) has a wide range
of applications in medical diagnosis, and novel form of Human
Machine Interfaces (HMI) for controlling prosthetic implants,
wheelchairs, and home appliances in various forms of paraly-
sis. However, the current EEG setups are composed of many
wires hanging down from the system, and individual elec-
trodes that must be set manually, which is time-consuming.
As a result, the overall system is neither comfortable, nor
aesthetically appealing. Here, we introduce for the first time,
a comfortable textile-basedEEG headband system that is soft,
conformal to the skin, and comfortable. We present materials
and methods for fabrication of multi-layer stretchablee-textile ,
that interfaces the human epidermis from one side through
printed electrodes, and interfaces a rigid PCB island on the second layer. We as well demonstrate a method that allows
creation of VIAs (vertical interconnect access) between the layers, using a CO2 laser. All Electrodes are integrated into
the headband, and thus there is no need for individual electrode placement, and individual wiring. By screen printing
a home-made conductive stretchable ink, patient-specific EEG headbands can be tailor made considering the optimal
positioning of the electrodes for each patient. We show that these printed electrodesbenefit from a very low skin-electrode
impedance, comparable to gold standard Ag/AgCl, or gold cup electrodes, thanks to the high surface area silver flakes
used in this work. The e-textile headband interfaceswith anEEG acquisitiondevice that captures, amplifies, and transmits
the data to an external mobile phone or a PC. Furthermore, the integrated amplification in the textile and the use of an
EMF rejection layer on top of the electrodes were shown to reduce the unwanted EM noise that is picked up by the
system. We as well show application of the developed headband for usage in Human Machine Interfaces and Sleep Data
Acquisition. Altogether, this device is step toward wider use of EEG acquisition devices for daily-use applications.
Index Terms
Conformable Electronics; E-textile, wearable sensor, biomonitoring, electroencephalography, epidermal
IN SPITE of the growing availability of neuroimaging
techniques (magnetic resonance imaging - MRI, com-
Manuscript received June 10, 2020; revised July 13, 2020; accepted
July 13, 2020. Date of publication July 16, 2020; date of current
version November 18, 2020. This work was supported in part by
the Foundation of Science and Technology (FCT) of Portugal through
the Carnegie Mellon University (CMU)-Portugal project Wireless biO-
monitoring stickers and smart bed architecture: toWards unthered
patients (WoW) under Grant 45913, in part by the Dermotronics under
Grant PTDC/EEIROB/31784/2017, in part by the European Union (EU)
Structural and Investment Funds (FEEI) through operational program
of the center region, in part by the Materiais e Tecnologias Indus-
triais Sustentáveis (MATIS) under Grant CENTRO-01-0145-FEDER-
000014, in part by CENTRO2020, in part by Add.Additive under
Grant POCI-01-0247-FEDER-024533, and in part by the Regional
Development Funds (FEDER), through Programa Operacional Com-
petitividade e Internacionalização (POCI). The associate editor coor-
dinating the review of this article and approving it for publication was
Dr. Edward Sazonov.
(Corresponding author: Mahmoud Tavakoli.)
The authors are with the Soft and Printed Microelectronics Lab, Institute
of Systems and Robotics, University of Coimbra, 3004-531 Coimbra,
Portugal (e-mail:
This article has supplementary downloadable material available at, provided by the authors.
Digital Object Identifier 10.1109/JSEN.2020.3009629
puted tomography - CT, single-photon emission tomogra-
phy - SPECT and positron emission tomography - PET),
electroencephalography (EEG) is still the most commonly
used sensing method for brain signal acquisition, which can
be used for detection of epileptic activity [1], [2] and non-
convulsive status epilepticus (NCSE) [3]. It also has a key
role in the investigation and research on encephalitis [4],
head trauma [5], coma [6], stroke [7] and brain death [8]
and other neurological conditions [9]–[11]. In addition to the
medical diagnosis, EEG signals have received an increasing
attention as an HMI for control of implanted prosthetics, and
wheelchairs [12]–[16], [17] and for study of emotions and
sensorial perception [18]–[20].
Despite the advances in the signal processing and clas-
sification techniques that allow real-time EEG based con-
trol of prosthetics and wheelchairs, one limiting factor for
wider application of the EEG-based HMIs is the current data
acquisition setups. Today’s EEG data acquisition systems are
bulky, have many wires, and require time-consuming prepa-
ration. They are not comfortable, nor aesthetically appealing.
In contrast to biological organs that are made of soft tissues,
1558-1748 © 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See for more information.
Authorized licensed use limited to: Carnegie Mellon Libraries. Downloaded on February 22,2022 at 14:54:24 UTC from IEEE Xplore. Restrictions apply.
15108 IEEE SENSORS JOURNAL, VOL. 20, NO. 24, DECEMBER 15, 2020
current electrophysiological monitoring devices are made of
rigid materials. The mechanical mismatch between the current
rigid electronic systems, and human tissues, is one of the
main driving forces for the field of stretchable electronics.
The main objective in this field is to develop soft and
stretchable e-skins that are able to adapt to the dynamic
morphology of the human skin [21], [22], without losing their
functionality. In the last decade, materials and methods for
the fabrication of stretchable circuits have been developed and
improved, including mask deposition [23], [24], laser pattern-
ing [25]–[27], and printing techniques [28]–[36]. Some efforts
were made in order to develop soft and stretchable e-skin
devices for conformal and comfortable electrophysiological
monitoring [29], [37]–[47]. For most applications in bio-
electronics, such circuits need to be epidermis-conformable,
i.e. they must keep their functionality while being bent, folded,
twisted and when subject to strains up to 30%, the normal
stretchability of the human skin [29], [30]. Among them, few
works presented soft electronic skins for electrophysiological
monitoring and EEG sensing [48], [49]. While promissing
as a disposable system, these ultrathin polymer-based archi-
tectures are generally fragile for long-term utilization. As a
consequence, the state of the art EEG technology can’t be
widely used for long-term research and daily applications [50]
and its utilization is mostly limited to experiments inside
laboratories. On the other hand, textile, as the common and
widely accepted “wearable” for human being, has an out-
standing potential to host electronics circuits for wearable
bio-electronics [51], [37], [52], [53]. In contrast to polymeric
e-skins, textile can be cut and easily tailor-made to the desired
In this article, we propose a plug-and-play wearable e-textile
(Fig. 1), that integrates multiple flexible electrodes and all
the necessary electrical interconnects, which can easily and
inexpensively be adapted to the specific needs of each user
in terms of electrodes quantity and positioning. By stencil
printing of a novel stretchable conductive ink developed by the
authors, and laser patterning of electrically conductive vias,
we implemented a two layer e-textile architecture, in which
electrodes and interconnects are on two opposite sides of the
textile. The conductive stretchable ink was synthesized by
mixing a high surface area Ag flake s(Technic inc), and a
Styrene-isoprene block copolymers (SIS) (Aldrich Chemistry).
In order to verify the suitability of the printed electrodes
for recording of electrophysiological data, we analyzed the
impedance of the skin-electrode interface over a frequency
range of 0-100kHz, and compared it with the gold-standard
Ag/AgCl and Goldcup electrodes. Results showed that the
printed electrodes using the mixture of high surface area
Ag flake sand (SIS) have an impedance in the same order
of gold standard electrodes, and extremely lower than the
composite made by a blend from the same polymer and
Carbon particles. Moreover, a modular electronic hardware
architecture was designed and implemented. This architecture
allows acquisition, amplification, processing and real-time
transmission of 8 EEG channels. Different setups of the
system in terms of the distance between the amplification
module and the electrodes were compared and the effect of a
Fig. 1. Textile EEG acquisition headband proposed in this work. Top
detail: Connection between rigid PCB and textile printed conductive lines
by means of a ink drop joint. Right detail: Contact between the printed
electrodes and the skin in the forehead.
EMF rejection layer on the acquired signal was also studied.
We show that integration of the amplification circuit near
the electrodes, as well as the implemented Faraday Cage, lead
to a decrease in the noise picked up by the system. Finally,
as a case study, examples of application of the developed
e-textile architecture are demonstrated as an HMI, and as
well for a comfortable data recording system during the
The wearable EEG system is composed of two main compo-
nents. The skin interfacing e-textile and the electronics system,
required for signal acquisition, amplification and communica-
tion. It is capable of recording electrical brain activity and
streaming it wirelessly for further processing.
A. E-Textile Headband
A conductive paste was developed by mixing 0.62wt% Ag
flakes (Technic inc), with a 0,17wt% SIS (Sigma-Aldrich)
solution. SIS is first dissolved in Toluene in a magnetic
mixer and heater (Agimatic-N). Ag flakes are then added to
the solution and mixed using a planetary mixer (THINKY
ARE-250) (3 minutes, 2000rpm). In contrast to common
Authorized licensed use limited to: Carnegie Mellon Libraries. Downloaded on February 22,2022 at 14:54:24 UTC from IEEE Xplore. Restrictions apply.
et al.
Fig. 2. Schematic of the patch fabrication process. A:Thelatex
impregnated fabric is cured on top of a glass in order to cure uniformly.
B: The vias on the fabric are cut using a Co2 laser. C: A previously cut
stencil is placed on the fabric and the ink is spread using a sharp blade or
spatula. D: The stencil is carefully removed, and the printed electrodes
are let do dry. E: The fabric is flipped over and the process is repeated
to create the circuit interconnects.
Fig. 3. Schematic of the textile patch and respective dimensions, with the
position of each printed electrode in terms of the international 10-20 EEG
electrode positioning system.
Ag based conductive inks and pastes that demand for a
post-baking (thermal sintering) procedure, the resulting com-
posite is surfactant-free and does not require a baking step. The
whole printing process is performed at the room temperature,
which is especially relevant and important for textile-based
electronic, as commercial Ag based pastes for screen printing,
require a thermal sintering step (100-250C), and therefore are
not compatible with heat sensitive substrates, including most
textiles. Fig. 2 summarizes the process for fabrication of the
EEG headband.
First, a lycra-mesh fabric is impregnated in latex (PVS
Elastica, FORMX). Then using a CO2 laser machine (Uni-
versal Laser Systems VLS 3.50), electrical vias are cut on the
fabric (Fig. 2B). Then a stencil (LineafixAdhesive, Lineafix
Hogar) is placed over the fabric and is patterned by the same
CO2 laser to desired pattern (size, geometry and position) of
the electrodes. The conductive ink is screen printed over the
fabric (Fig. 2C). While printing the first layer of the circuit,
the previously patterned VIAs are also filled. Then the stencil
is removed, and the fabric is turned over (Fig. 2D). Finally,
the process is repeated on the other side of fabric, this time
to print the electrical interconnect.
In Fig. 3 the shape of the textile patch, printed lines,
vias and printed electrodes is shown. Furthermore, each of
the 8 recording electrodes is identified by its chosen location
in terms of the 10-20 system (the internaltionaly accepted EEG
electrodes positioning system), alongside eith the reference
and ground electrodes.
In order to attach the rigid amplification PCB (described
later) to the printed lines, a simple joint is created by a drop
of conductive ink that bonds both to the line and the contacts
of the rigid PCB, as seen in Fig. 1 (top insertion).
B. Signal Amplification and Transmission
Fig. 4A illustrates the high level system block diagram
describing the architecture of the electronics system. The
electrodes patch is worn by the subject’s on his/her forehead
and the raw data is then amplified, digitized and transmited
through WiFi to a device connected to the same network for
The electronics system is composed of two main cir-
cuits. Amplification circuit, and processing/transmission cir-
cuit. Amplification circuits are placed directly on the e-textile,
to be near to the acquisition electrodes. Placing the amplifier
near the signal source is generally a good practice, since it
eliminates amplification of the noise acquired by cables in
a tethered system. As can be seen in Fig. 4B, the system
architecture is designed in order to allow integration of several
amplifiers that can be all wired to the main processing and
transmission board. In this way, one can customize the number
and location of amplifiers on the headband.
Brain electrical signals acquired by surface electrodes are of
very low amplitude (between 10 μV and 100 μV) [54]–[56]
and therefore, require high gain amplification prior to being
sent to the Analog-to-Digital converter. The ADS1299 module
(ADS1299, Texas Instruments) was selected as the signal
amplifier and ADC for the final design. This 8-channel
amplifier with an integrated 24-bit sigma-delta ADC enables
prolonged biopotential monitoring with sufficient precision.
The amplifier sits directly on the electrodes’ textile patch.
An ESP8266 microcontroller (Expressif Systems) was cho-
sen, due to its low power consumption and WiFi capabil-
ities. The communication between the ADS1299 and the
ESP8266 is via an SPI interface, which allows for a modular
design, where up to 3 amplification boards can be connected
to the same microcontroller, allowing for simultaneous reading
of up to 24 EEG signals
The system is powered by 2 li-Po cells, totalizing 7.4V and
1600mAh, allowing for 24 hours operation time.
A. EEG Data Acquisition
The data was acquired by a working prototype (Fig. 4C)
of the proposed device. It was transferred to a computer and
processed in Matlab.
Processing includes removal of the DC frequency and
removal of any frequency above 100Hz, as well as removal of
power line interference at 50Hz.
B. Electrode-Skin Impedance Measurements
In order to analyse the suitability of the printed electrodes
for acquisition of electrophysiological signals, we analysed the
Authorized licensed use limited to: Carnegie Mellon Libraries. Downloaded on February 22,2022 at 14:54:24 UTC from IEEE Xplore. Restrictions apply.
15110 IEEE SENSORS JOURNAL, VOL. 20, NO. 24, DECEMBER 15, 2020
Fig. 4. A: Block diagram illustrating the main components of the device, comprising data collection and data transmission blocks, power circuit
and data processing unit. B: The modularity of the system allows for several amplifiers to be connected simultaneously to the same processing and
transmission board. C: Working prototype of the EEG acquisition system proposed in this work.
impedance of the electrode-skin interface over a wide range
of frequencies and compared the results with gold standard
electrodes. Generally, it is accepted in the literature that a
lower skin-electrode impedance is desired, since it leads to a
higher signal to noise ratio [29].
A bioimpedance measurement system (PalmSens4, Fig. 5A)
connected to PSTrace Software was used to measure the
electrode-skin impedance in different frequencies (between
0.1Hz and 10000Hz), by using an alternating sinusoidal exci-
tation electrical current. Relying on the equivalent model
proposed by Albulbul [57], we are able to estimate each of the
electrode-skin impedance parameters. For the measurements,
3 electrodes [58] are applied in the ventral side of the subject’s
forearm and the excitation signal is applied between 2 of the
3 electrodes placed in the subject.
The ventral side of the forearm was selected to perform
the electrode-skin impedance measurements due to accessi-
bility for long-term measurement. A careful skin preparation
procedure was performed before every measurement. This
consists of cleaning the skin with alcohol swabs in order
to remove dead skin cells, sweat and oil. We then let the
skin to air-dry for 5 minutes and placed the electrodes over
the forearm. We then wait for 5 minutes before taking the
measurement. This period is considered to allow stabilization
of the temperature, and humidity. All tests were taken in
ambient humidity ranging from 52-70%. This preparation
procedure allowed to decrease the influence of external factors
(other than the electrode materials) in the taken measurements.
All the electrodes had the same displacement of 3cm
between them. The Ag/AgCl electrodes were fixed to the skin
by their adhesion, while the Goldcup and AgSIS ink electrodes
were fix by means of a textile (as seen in Fig. 5E) secured
to the forearm, maintaining the pressure constant in both
Different combinations of materials were tested and
compared– Ag/AgCl (Fig. 5C); Goldcup (with Ten20 con-
ductive paste) (Fig. 5D) Ten20; AgSIS ink (Fig. 5E)and
CSIS [59].
Using the Levenberg-Marquadt [60] fitting method, imple-
mented in PSTrace5 software, we are able to estimate the val-
ues of each component of the electro-skin interface equivalent
model in Fig. 5F, according to equation (1):
where Rdand Cdrepresent the impedance associated with the
electrode-skin interface and the polarization at this interface,
while Rsis the series resistance associated with the type of
material of the electrodes.
Regarding the Electrode-Skin equivalent model, our team
has previously shown that the Rsvalue is not a relevant
parameter for signal to noise ratio measurement [29]. On the
other hand, lower Rdand higher Cddirectly contribute to a
better signal quality, since they lead directly to a lower value
of Ze, as shown by the previous equation. Parameters such
as the electrode “softness”, and conformable skin-electrode
interface were parameters found to be directly contributing to
a better electrode-skin interface [29].
From Fig. 5G, we see that the AgSIS printed electrodes
have a Rd value similar to the gold standard electrodes (Gold-
cup (with conductive paste) and Ag/AgCl). This is a very desir-
able result, since lower resistance paths to the ionic current
biological signals lead to higher quality measurements [29].
As a comparison, we can see that the Rd Value for CSIS
printed electrodes is much higher than the previously referred
electrodes. This might be due to the high aspect ratio of silver
flakes, that allow a better percolating network compared to the
spherical type conductors.
It is known, by previous works, that higher Cd values
are related to the ability of the electrodes to accumulate
charges at theelectrode-skin interface and are thus expected in
polarizable electrodes, which are likely to behave capacitively
at the electrode-skin interface. The highest values of this
parameter are observed in the Ag/AgCl and AgSIS electrodes,
followed by Goldcup (with Ten20 conductive paste) and CSIS
Authorized licensed use limited to: Carnegie Mellon Libraries. Downloaded on February 22,2022 at 14:54:24 UTC from IEEE Xplore. Restrictions apply.
et al.
Fig. 5. Electrode-skin impedance analysis and electrode characterization. A: experimental setup - two electrodes placed over the subject’s forearm
are connected to the impedance analyzer and a frequency sweep is performed. B: single electrode/skin interface simplification circuit for the system’s
equivalent model. C: Gold-standard commercially available pre-gelled Ag/AgCl and D: Goldcup electrodes. E: Novel conductive ink electrodes.
F: Bode plots of the electrode-skin impedance of each of the tested electrodes (Ag/AgCl, Goldcup – interfaced to the skin via a layer of Ten20
conductive paste, CSIS and AgSIS). G: Components of the equivalent electrode-skin model (Rd, Cd and Rs) for each of the tested electrode
We are able to infer, from the previously presented results,
that this AgSIS printed electrodes benefit from a low electrode-
skin impedance, comparable to the Gold standard electrode
types. This means that this conductive ink, contrary to what
is observed in other printable conductive inks (Carbon-SIS for
example), can be reliably used for biomonitoring applications,
as we demonstrate later in this work for the case of EEG
Regarding the Electrode-Skin Impedance shown in the
Bode plots from Fig. 8A, we observe, as expected, that the
electrode-skin impedance of AgSIS electrodes are similar to
the values of the Goldcup (with Ten20 conductive paste) and
Ag/AgCl electrodes and much lower than the CSIS electrodes.
C. Signal-to-Noise Ratio Measurements
SNR is a good indicator of the quality of the acquired
data and thus a good indicator of the quality of the overall
system. Different setups of the system were implemented,
tested and compared, in an effort to reduce the Signal-to-noise
ratio (SNR) in similar EEG readings acquired by the proposed
system. The SNR value is calculated by computing the ratio of
the filtered signal (x) summed squared magnitude to that of the
noise (y), using the snr(x,y) matlab function. First, the ampli-
fier was connected to the patch through wires, like usually
seen in similar devices used in a clinical setting or research
Fig. 6. A: Connection through long wires between the amplification
board and the electrodes’ patch. B: Amplification connected directly in
the electrodes’ patch. C: Tin foil faraday cage covering the system.
labs (Fig. 6A). In another setup, the amplification board was
fixed directly in the fabric patch, near the electrodes (Fig. 6B).
In an effort to reduce further the noise levels in the system,
a Faraday cage was implemented, by covering the fabric patch
with tin foil, as it can be seen in Fig. 6C.
In Table I, we can see that the usage of a faraday cage
covering the exposed printed lines is effective in reducing
noise levels acquired by the system, increasing the SNR from
39.8139dB to 36.0465dB.
Authorized licensed use limited to: Carnegie Mellon Libraries. Downloaded on February 22,2022 at 14:54:24 UTC from IEEE Xplore. Restrictions apply.
15112 IEEE SENSORS JOURNAL, VOL. 20, NO. 24, DECEMBER 15, 2020
Furthermore, as seen in Ta b le I I, there is an improvement
in the SNR value (from 35.5489dB to 30.1996dB), when
the amplification is integrated near the electrodes, compared
to the case of tethered headband.A parallel experiment was
conducted to compare the SNR between EEG data acquired
by goldstandard electrodes (Ag/AgCl and goldcup) and the
custom printed electrodes presented in this work. After careful
skin preparation, all 3 electrodes were attached around the FP1
location (10-20 electrode positioning system) so they would
acquire the same EEG signal and connected to the developed
EEG acquisition board through 1-meter wires, and EEG was
recorded simultaneously from the 3 electrodes.
8 independent EEG datasets were acquired, and the results
are presented in Table III. We can observe that, although
Ag/AgCl electrodes (SNR=−32.1874dB), followed by gold-
cup electrodes (SNR=−32.5452dB), perform slightly bet-
ter than AgSIS printed electrodes (SNR=−32.7113dB),
the results are comparable.
D. Mechanical Characterization
A latex-impregnated lycra dogbone (width 10mm, length
50mm, thickness 0.55m) with an AgSIS printed track
Fig. 7. Variation of electrical resistance of an AgSIS track with tensile
(with 1.5mm, length 50mm) was stretched while measuring
the track’s electrical resistance. This test was performed in a
Instron 5943 and the results are presented in Fig. 7. We can
observe that between 0 and 20% tensile strain, the resistivity
doubles, which is not a relevant variation since the initial
resistance was 11.93and a normal everyday usage of the
developed headband would never subject the printed tracks
to strains bigger that 20%. We can also observe that at 32%
strain there is a ramp up in electrical resistance leading us to
conclude that that is the tensile strain limit for the developed
E. Comparison With Commercially Available Devices
Table IV (adapted from [61]), summarizes the characteris-
tics of current EEG acquisition headsets that are both available
in the market and presented in state-of-the-art literature.
The system developed in this work is low weight- only 115g
regarding the 24 channels version (including textile electrodes,
Authorized licensed use limited to: Carnegie Mellon Libraries. Downloaded on February 22,2022 at 14:54:24 UTC from IEEE Xplore. Restrictions apply.
et al.
electronic circuits and batteries), and low cost - only
$225.33 BOM. The maximum operation time was found to be
24 hours and 42 minutes, more than any of the other presented
Regarding the used electrodes, in contrast with most of
the presented commercially available devices (which rely on
disposable, wet contact electrodes), the headband proposed in
this work relies on dry ink-based stretchable electrodes that are
comfortable and can be used for long periods. Furthermore,
this headband can accommodate more electrodes than any of
the other forehead EEG headbands presented and it is fully
customizable in terms of electrode positioning.
F. Case Studies
1) Human-Machine Interface:
Since the eyes act as electrical
dipoles, with the cornea being positive and the retina negative,
when the eyes move, changes in electrical potentials can be
observed [71]–[73]. Using the 2 outermost electrodes of the
fabric patch, and observing one of them relative to the other,
we can see a potential that changes as follows: When the eyes
move to the right, the potential gets positive and in contrast,
gets negative when the eyes move to the left. The results
from this simple application can be observed in Fig. 8 and
Multimedia extension 1.
2) Sleep Data Acquisition:
The system developed in this
work was used to acquire and record sleep data over a night
of sleep (Fig. 9D). Although the data was recorded but not
analyzed in depth, a simple visual inspection of the EEG data
plots allows us to see clear differences in brain activity along
Fig. 8. Human-Machine Interface by eye position detection. A: subject
looking left (blue box activated). B: subject looking right (red box acti-
vated). In each figure we can also see the measured potential, in the
white text box, in uV.
the night, as seen in Fig. 9. During deep sleep (Fig. 9 A,B,E)
slow waves are predominant and high frequency activity is
almost non-existing. For instance, in Fig. 9 B, one can easily
observe predominant 7Hz brain activity usually related to
stage 1 non-REM sleep in healthy subjects [74]Furthermore,
in Fig 9 E, it can be observed that there is no relevant
brain activity above 40 Hz and low frequency waves are
predominant (below 15Hz, with peaks around 2Hz and 9Hz),
which correspond to either sleep stage 3 (predominant Delta
waves in the 1-4Hz range) or a drowsy-like non-awake state
(predominant alpha waves in the 9,10Hz range)[74].
In contrast, we can observe in Fig. 9 C a greater range of
frequencies coexisting in the plot, as well as some artifacts
related to eye-blinks (2s and 3.5s), which indicate that the
subject is awake [74]. The same happens in Fig. 9 F where
an even distribution of amplitudes between all frequencies can
be observed.
Authorized licensed use limited to: Carnegie Mellon Libraries. Downloaded on February 22,2022 at 14:54:24 UTC from IEEE Xplore. Restrictions apply.
15114 IEEE SENSORS JOURNAL, VOL. 20, NO. 24, DECEMBER 15, 2020
Fig. 9. Sleep data acquisition. A: subject wearing the proposed EEG acquisition headband during sleep. B,C: EEG activity during sleep. Low
frequency waves can be observed D: Normal awake EEG activity. E: Predominant frequencies in EEG during sleep. High frequencies in EEG during
sleep. High frequency activity is almost non-existing during sleep F: Existing frequencies in a normal awake EEG.
The portability and comfort of the system bring important
advantages for long-term recordings. It must be noted that the
tested subject reported no discomfort during sleep and showed
no signs of skin-damage or irritation after waking up.
A comfortable and reusable e-textile with integrated flexible
and stretchable electrodes and amplification circuitry for EEG
biosignals’ acquisition was successfully developed in this
work. The main goal of creating this flexible, fabric-based
headband was to overcome the existing problems in state-of-
the-art EEG setups: The bulkiness of the systems, the dis-
comfort caused by rigid electrodes and the time consuming,
laborious task of placing individual electrodes in a subject,
using conductive pastes.
The developed textile patch with integrated electrodes is
easy to place in the subject, allowing the immediate placement
of multiple electrodes in the forehead in a matter of seconds,
without the need for time-consuming individual electrode
placement, skin abrasion, or complex skin preparation. Due to
its comfort, it can be used on a daily basis for Human Machine
Interfaces, continuous brain signal monitoring, or sleep-data
Regarding the screen printed electrodes, the Electrode-Skin
impedance was measured over a wide range of frequencies
(0.1Hz - 10000Hz) and it was found that these electrodes have
a behavior comparable to state of the art electrodes used in
medical environment (Ag/AgCl and Goldcup (with Ten20 con-
ductive 8paste) electrodes). Also, in terms of acquired noise,
the printed electrodes are comparable to Ag/AgCl and Gold-
cup electrodes. The fabrication technique is simple, fast and
is performed entirely at ambient conditions. Therefore, this
headband can be easily adapted to each patient in terms of
quantity and positioning of electrodes in the system.
The proposed architecture allows approximation of the
amplification circuit to the electrodes, which is an efficient
measure in reducing the noise, as longer wires in traditional
systems are a known cause of induced noise. The current
electronic circuit developed in this work, is a preliminary
proof-of-concept version and should be still analyzed and
improved in terms of internal contribution to the noise, which
the subject of future works.
It was also observed that the use of an electromagnetic
radiation rejection layer (Faraday cage) covering the patch
interconnects proved to be an effective way of further reducing
noise levels in the system.
The wireless capabilities of the headband make the proposed
EEG acquisition system ideal for outpatient EEG studies and
use in remote sites, since it is self-contained and there are no
wires holding the patient under study. Also, the operating time
of more than 24 hours is an outstanding advantage allowing
for long term brain activity recordings, alongside with the
advantages of the device being comfortable and low weight
The before presented characteristics of the proposed EEG
acquisition system, is a major development in the field of
EEG acquisition hardware, that can have strong impact in the
way brain-activity studies and medical exams are performed,
allowing for long-term patient monitoring. Furthermore, this
work is a major step toward application of wearable EEGs in
HMI for control of appliances.
The authors declare no conflict of interest.
[1] M. Sundaram, R. M. Sadler, G. B. Young, and N. Pillay, “EEG in
epilepsy: Current perspectives,” Can. J. Neurol. Sci., vol. 26, no. 4,
pp. 255–262, Nov. 1999.
Authorized licensed use limited to: Carnegie Mellon Libraries. Downloaded on February 22,2022 at 14:54:24 UTC from IEEE Xplore. Restrictions apply.
et al.
[2] S. Noachtar and J. Rémi, “The role of EEG in epilepsy: A critical
review,” Epilepsy Behav., vol. 15, no. 1, pp. 22–33, May 2009, doi:
[3] B. J. Kolls and A. M. Husain, “Assessment of hairline EEG as a
screening tool for nonconvulsive status epilepticus,” Epilepsia, vol. 48,
no. 5, pp. 959–965, May 2007, doi: 10.1111/j.1528-1167.2007.01078.x.
[4] R. Sutter et al., “Electroencephalography for diagnosis and prognosis of
acute encephalitis,” Clin. Neurophysiol., vol. 126, no. 8, pp. 1524–1531,
Aug. 2015, doi: 10.1016/j.clinph.2014.11.006.
[5] J. N. Ianof and R. Anghinah, “Lesão cerebral traumática: Um ponto de
vista EEG,” Dementia Neuropsychologia, vol. 11, no. 1, pp. 3–5, 2017,
doi: 10.1590/1980-57642016dn11-010002.
[6] B. Young, “The EEG in coma,J. Clin. Neurophysiol., vol. 17, no. 5,
pp. 473–485, 2000.
[7] K. G. Jordan, “Emergency EEG and continuous EEG monitoring in acute
ischemic stroke,” J. Clin. Neurophysiol., vol. 21, no. 5, pp. 341–352,
2004, doi: 10.1097/01.WNP.0000145005.59766.D2.
[8] Z. Chen et al., “An empirical EEG analysis in brain death diagnosis for
adults,” Cognit. Neurodyn., vol. 2, no. 3, pp. 257–271, Sep. 2008, doi:
[9] S. F. Khan, R. Ashalatha, S. V. Thomas, and P. S. Sarma, “Emer-
gent EEG is helpful in neurology critical care practice,” Clin.
Neurophysiol., vol. 116, no. 10, pp. 2454–2459, Oct. 2005, doi:
[10] J. Praline et al., “Emergent EEG in clinical practice,” Clin.
Neurophysiol., vol. 118, no. 10, pp. 2149–2155, Oct. 2007, doi:
[11] P. N. Varelas. (2003). Emergent EEG Indications and Diagnostic Yield
Hours 2 or Any EEG Done to Exclude NCSE With 24-Hours-a-Day,
7-Days-a-Week. [Online]. Available:
[12] A. Ferreira, W. C. Celeste, F. A. Cheein, T. F. Bastos-Filho, and
M. Sarcinelli-Filho, “Human-machine interfaces based on EMG and
EEG applied to robotic systems,” in Proc. 2nd Int. Workshop Biosignal
Process. Classification (BPC), 2006, pp. 116–125, doi: 10.1186/1743-
[13] J. Zhang, B. Wang, C. Zhang, Y. Xiao, and M. Y. Wang,
“An EEG/EMG/EOG-based multimodal human-machine interface to
real-time control of a soft robot hand,” Frontiers Neurorobotics, vol. 13,
p. 7, Mar. 2019, doi: 10.3389/fnbot.2019.00007.
[14] M. Tariq, P. M. Trivailo, and M. Simic, “EEG-based BCI control
schemes for lower-limb assistive-robots,” Frontiers Hum. Neurosci.,
vol. 12, p. 312, Aug. 2018, doi: 10.3389/fnhum.2018.00312.
[15] Z. T. Al-qaysi, B. B. Zaidan, A. A. Zaidan, and M. S. Suzani,
“A review of disability EEG based wheelchair control system: Coher-
ent taxonomy, open challenges and recommendations,” Comput. Meth-
ods Programs Biomed., vol. 164, pp. 221–237, Oct. 2018, doi:
[16] S. K. Swee, K. Desmond, T. Kiang, and L. Z. You, “EEG controlled
wheelchair,” in Proc. MATEC Web Conf., Apr. 3016, Art. no. 02011,
doi: 10.1051/matecconf/20165102011.
[17] F. Babiloni et al., “On the use of brain–computer interfaces outside sci-
entific laboratories: Toward an application in domotic environments,Int.
Rev. Neurobiol., vol. 86, pp. 133–146, Jan. 2009, doi: 10.1016/S0074-
[18] G. Cartocci et al., “Gender and age related effects while watching TV
advertisements: An EEG study,” Comput. Intell. Neurosci., vol. 2016,
pp. 1–10, Oct. 2016, doi: 10.1155/2016/3795325.
[19] A. G. Maglione et al., “A neuroelectrical brain imaging study on the
perception of figurative paintings against only their color or shape
contents,” Frontiers Hum. Neurosci., vol. 11, p. 378, Jul. 2017, doi:
[20] G. Cartocci et al., “Electroencephalographic, heart rate, and galvanic
skin response assessment for an advertising perception study: Applica-
tion to antismoking public service announcements,” J. Vis. E xp. , no. 126,
Aug. 2017, Art. no. e55872, doi: 10.3791/55872.
[21] A. Nag, S. C. Mukhopadhyay, and J. Kosel, “Wearable flexible sensors:
Areview,IEEE Sensors J., vol. 17, no. 13, pp. 3949–3960, Jul. 2017,
doi: 10.1109/JSEN.2017.2705700.
[22] Y. Shao, H. Hu, and Y. Visell, “A wearable tactile sensor array for
large area remote vibration sensing in the hand,” IEEE Sensors J.,
vol. 20, no. 12, pp. 6612–6623, Jun. 2020, doi: 10.1109/JSEN.2020.
[23] R. K. Kramer, C. Majidi, and R. J. Wood, “Masked deposition
of gallium-indium alloys for liquid-embedded elastomer conductors,”
Adv. Funct. Mater., vol. 23, no. 42, pp. 5292–5296, Nov. 2013, doi:
[24] Y.-L. Park, B.-R. Chen, and R. J. Wood, “Design and fabrication of soft
artificial skin using embedded microchannels and liquid conductors,”
IEEE Sensors J., vol. 12, no. 8, pp. 2711–2718, Aug. 2012, doi:
[25] I. D. Joshipura, H. R. Ayers, C. Majidi, and M. D. Dickey, “Methods to
pattern liquid metals,” J. Mater. Chem. C, vol. 3, no. 16, pp. 3834–3841,
2015, doi: 10.1039/c5tc00330j.
[26] T. Lu, L. Finkenauer, J. Wissman, and C. Majidi, “Rapid prototyp-
ing for soft-matter electronics,” Adv. Funct. Mater., vol. 24, no. 22,
pp. 3351–3356, Jun. 2014, doi: 10.1002/adfm.201303732.
[27] R. Rocha, P. Lopes, A. T. de Almeida, M. Tavakoli, and C. Majidi,
“Soft-matter sensor for proximity, tactile and pressure detection,” in
Proc. IEEE/RSJ Int. Conf. Intell. Robots Syst. (IROS), Sep. 2017,
pp. 3734–3738, doi: 10.1109/IROS.2017.8206222.
[28] A. Eshkeiti et al., “Screen printing of multilayered hybrid printed
circuit boards on different substrates,” IEEE Trans. Compon., Packag.,
Manuf. Technol., vol. 5, no. 3, pp. 415–421, Mar. 2015, doi:
[29] P. A. Lopes et al., “Soft bioelectronic stickers: Selection and evaluation
of skin-interfacing electrodes,” Adv. Healthcare Mater., vol. 8, no. 15,
Aug. 2019, Art. no. 1900234, doi: 10.1002/adhm.201900234.
[30] P. Filipe, A. Lopes, H. Paisana, A. T. de Almeida, C. Majidi, and
M. Tavakoli. (2018). Hydroprinted Electronics: Ultrathin Stretchable
Ag-In-Ga E-Skin for Bioelectronics and Human-Machine Interaction.
[Online]. Available:
[31] M. Tavakoli et al., “EGaIn-assisted room-temperature sintering of
silver nanoparticles for stretchable, inkjet-printed, thin-film electron-
ics,” Adv. Mater., vol. 30, no. 29, Jul. 2018, Art. no. 1801852, doi:
[32] M. Tavakoli et al., “Carbon doped PDMS: Conductance stability
over time and implications for additive manufacturing of stretchable
electronics,” J. Micromech. Microeng., vol. 27, no. 3, Feb. 2017,
Art. no. 035010, doi: 10.1088/1361-6439/aa5ab1.
[33] J. Krzeminski, A. Kanthamneni, D. Wagner, M. Detert, B. Schmidt, and
M. Jakubowska, “Microscale hybrid flexible circuit printed with aerosol
jet technique,” IEEE Trans. Nanotechnol., vol. 17, no. 5, pp. 979–984,
Sep. 2018, doi: 10.1109/TNANO.2018.2841639.
[34] C. Votzke, U. Daalkhaijav, Y. Mengüç, and M. L. Johnston,
“3D-printed liquid metal interconnects for stretchable electronics,”
IEEE Sensors J., vol. 19, no. 10, pp. 3832–3840, May 2019, doi:
[35] D. F. Fernandes, C. Majidi, and M. Tavakoli, “Digitally printed
stretchable electronics: A review,J. Mater. Chem . C, vol. 7, no. 45,
pp. 14035–14068, Nov. 2019, doi: 10.1039/c9tc04246f.
[36] S. Khan, L. Lorenzelli, and R. S. Dahiya, “Technologies for print-
ing sensors and electronics over large flexible substrates: A review,”
IEEE Sensors J., vol. 15, no. 6, pp. 3164–3185, Jun. 2015, doi:
[37] M. A. Yokus and J. S. Jur, “Fabric-based wearable dry electrodes for
body surface biopotential recording,” IEEE Trans. Biomed. Eng., vol. 63,
no. 2, pp. 423–430, Feb. 2016, doi: 10.1109/TBME.2015.2462312.
[38] D.-H. Kim et al. Epidermal Electronics. Accessed: Oct. 26, 2019.
[Online]. Available:
[39] M. Mehrali et al., “Blending electronics with the human body: A path-
way toward a cybernetic future,” Adv. Sci., vol. 5, no. 10, Oct. 2018,
Art. no. 1700931, doi: 10.1002/advs.201700931.
[40] Kenry, J. C. Yeo, and C. T. Lim, “Emerging flexible and wearable
physical sensing platforms for healthcare and biomedical applica-
tions,” Microsyst. Nanoeng., vol. 2, no. 1, pp. 1–19, Dec. 2016, doi:
[41] D. G. Marques, P. A. Lopes, A. T. de Almeida, C. Majidi, and
M. Tavakoli, “Reliable interfaces for EGaIn multi-layer stretchable
circuits and microelectronics,” Lab Chip, vol. 19, no. 5, pp. 897–906,
Feb. 2019, doi: 10.1039/c8lc01093e.
[42] C. Leal, P. A. Lopes, A. Serra, J. F. J. Coelho, A. T. de Almeida,
and M. Tavakoli, “Untethered disposable health monitoring electronic
patches with an integrated Ag2O–Zn battery, a AgInGa current collector,
and hydrogel electrodes,” ACS Appl. Mater. Interfaces, vol. 12, no. 3,
pp. 3407–3414, Jan. 2020, doi: 10.1021/acsami.9b18462.
[43] J. Alberto et al., “Fully untethered battery-free biomonitoring electronic
tattoo with wireless energy harvesting,” Sci. Rep., vol. 10, no. 1,
Dec. 2020, Art. no. 5539, doi: 10.1038/s41598-020-62097-6.
[44] V. Goverdovsky, D. Looney, P. Kidmose, C. Papavassiliou, and
D. P. Mandic, “Co-located multimodal sensing: A next generation solu-
tion for wearable health,” IEEE Sensors J., vol. 15, no. 1, pp. 138–145,
Jan. 2015, doi: 10.1109/JSEN.2014.2338612.
Authorized licensed use limited to: Carnegie Mellon Libraries. Downloaded on February 22,2022 at 14:54:24 UTC from IEEE Xplore. Restrictions apply.
15116 IEEE SENSORS JOURNAL, VOL. 20, NO. 24, DECEMBER 15, 2020
[45] Y.-L. Zheng et al., “Unobtrusive sensing and wearable devices for health
informatics,” IEEE Trans. Biomed. Eng., vol. 61, no. 5, pp. 1538–1554,
May 2014, doi: 10.1109/TBME.2014.2309951.
[46] G. Wang et al., “Stretchable optical sensing patch system integrated
heart rate, pulse oxygen saturation, and sweat pH detection,” IEEE
Trans. Biomed. Eng., vol. 66, no. 4, pp. 1000–1005, Apr. 2019, doi:
[47] G. S. Guvanasen et al., “A stretchable microneedle electrode array for
stimulating and measuring intramuscular electromyographic activity,
IEEE Trans. Neural Syst. Rehabil. Eng., vol. 25, no. 9, pp. 1440–1452,
Sep. 2017, doi: 10.1109/TNSRE.2016.2629461.
[48] L. M. Ferrari et al., “Ultraconformable temporary tattoo electrodes for
electrophysiology,Adv. Sci., vol. 5, no. 3, Mar. 2018, Art. no. 1700771 ,
doi: 10.1002/advs.201700771.
[49] P. Lepola et al., “Screen-printed EEG electrode set for emergency
use,” Sens. Actuators A, Phys., vol. 213, pp. 19–26, Jul. 2014, doi:
[50] X. Chen and Z. J. Wang, “Design and implementation of a wearable,
wireless EEG recording system,” in Proc. 5th Int. Conf. Bioinf. Biomed.
Eng., Wuhan, China, 2011, pp. 1–4, doi: 10.1109/icbbe.2011.5781501.
[51] D. Patron et al., “On the use of knitted antennas and inductively
coupled RFID tags for wearable applications,” IEEE Trans. Bio-
med. Circuits Syst., vol. 10, no. 6, pp. 1047–1057, Dec. 2016, doi:
[52] W. Wu, “Stretchable electronics: Functional materials, fabrication strate-
gies and applications,” Sci. Technol. Adv. Mater., vol. 20, no. 1,
pp. 187–224, Dec. 2019, doi: 10.1080/14686996.2018.1549460.
[53] T.-H. Kang, C. R. Merritt, E. Grant, B. Pourdeyhimi, and H. T. Nagle,
“Nonwoven fabric active electrodes for biopotential measurement during
normal daily activity,IEEE Trans. Biomed. Eng., vol. 55, no. 1,
pp. 188–195, Jan. 2008, doi: 10.1109/TBME.2007.910678.
[54] J. Malmivuo and R. Plonsey, Bioelectromagnetism,1sted.NewYork,
NY, USA: Oxford Univ. Press, 1995.
[55] W. O. Tatum and A. M. Husain, Handbook of EEG Interpretation,1sted.
New York, NY, USA: Demos Medical Publishing, 2007.
[56] M. C. Fields, J. J. Yoo, and L. V. Marcuse, Rowan’s Primer of EEG,
2nd ed. Amsterdam, The Netherlands: Elsevier, 2003.
[57] A. Albulbul, “Evaluating major electrode types for idle biological signal
measurements for modern medical technology,” Bioengineering,vol.3,
no. 3, p. 20, Aug. 2016, doi: 10.3390/bioengineering3030020.
[58] S. H. Nasrollaholhosseini, J. Mercier, G. F is cher, a nd W. G. B esio,
“Electrode–electrolyte interface modeling and impedance characterizing
of tripolar concentric ring electrode,IEEE Trans. Biomed. Eng., vol. 66,
no. 10, pp. 2897–2905, Oct. 2019, doi: 10.1109/tbme.2019.2897935.
[59] R. Tchoudakov, O. Breuer, M. Narkis, and A. Siegmann, “Conduc-
tive polymer blends with low carbon black loading: High impact
polystyrene/thermoplastic elastomer (styrene-isoprene-styrene),” Polym .
Eng. Sci., vol. 37, no. 12, pp. 1928–1935, Dec. 1997.
[60] C. Kanzow, N. Yamashita, and M. Fukushima, “Levenberg–marquardt
methods with strong local convergence properties for solving nonlinear
equations with convex constraints,” J. Comput. Appl. Math., vol. 172,
no. 2, pp. 375–397, Dec. 2004, doi: 10.1016/
[61] B. Farnsworth. (2017). EEG Headset Comparison—A Technical
Overview of 10 Headsets. Accessed: Oct. 26, 2019. [Online]. Available:
[62] J. Xu et al., “A wearable 8-channel active-electrode EEG/ETI acqui-
sition system for body area networks,” IEEE J. Solid-State Cir-
cuits, vol. 49, no. 9, pp. 2005–2016, Sep. 2014, doi: 10.1109/JSSC.
[63] BrainLink Specs. Accessed: Nov. 12, 2019. [Online]. Available:
[64] Dreem. Accessed: Nov. 12, 2019. [Online]. Available: https://shop.
[65] iBand Specs. Accessed: Nov. 12, 2019. [Online]. Available:
[66] BrainBit. Accessed: Nov. 12, 2019. [Online]. Available: https://brainbit.
[67] MindWave Specs. Accessed: Nov. 12, 2019. [Online]. Available:
[68] L. Inzelberg, M. D. Pur, S. Steinberg, D. Rand, M. Farah, and Y. Hanein,
“Wireless electronic-tattoo for long-term high fidelity facial muscle
recordings,” Proc. SPIE, vol. 10194, May 2017, Art. no. 101940U, doi:
[69] L. Inzelberg and Y. Hanein, “Electrophysiology meets printed elec-
tronics: The beginning of a beautiful friendship,” Frontiers Neurosci.,
vol. 12, p. 992, Jan. 2019, doi: 10.3389/fnins.2018.00992.
[70] A. Dabbaghian, T. Yousefi, S. Z. Fatmi, P. Shafia, and H. Kassiri,
“A 9.2-g fully-flexible wireless ambulatory EEG monitoring and diag-
nostics headband with analog motion artifact detection and compensa-
tion,” IEEE Trans. Biomed. Circuits Syst., vol. 13, no. 6, pp. 1141–1151,
Dec. 2019, doi: 10.1109/TBCAS.2019.2936327.
[71] R. Zhang et al., “An EOG-based human–machine interface to control a
smart home environment for patients with severe spinal cord injuries,”
IEEE Trans. Biomed. Eng., vol. 66, no. 1, pp. 89–100, Jan. 2019, doi:
[72] Q. Huang et al., “An EOG-based human–machine interface for wheel-
chair control,” IEEE Trans. Biomed. Eng., vol. 65, no. 9, pp. 2023–2032,
Sep. 2018, doi: 10.1109/TBME.2017.2732479.
[73] G. M. Paul, F. Cao, R. Torah, K. Yang, S. Beeby, and J. Tudor,
“A smart textile based facial EMG and EOG computer interface,”
IEEE Sensors J., vol. 14, no. 2, pp. 393–400, Feb. 2014, doi:
[74] G. Torres, M. P. Cinelli, A. T. Hynes, I. S. Kaplan, and J. R. Leheste,
“Electroencephalogram mapping of brain states,” J. Neurosci. Neuroeng.,
vol. 3, no. 2, pp. 73–77, Dec. 2014, doi: 10.1166/jnsne.2014.1098.
Manuel Reis Carneiro received the M.Sc.
degree in electrical and computer engineer-
ing with a specialization in automation from
the University of Coimbra, Portugal, in 2019.
He is currently a Researcher with the Institute
of Systems and Robotics, University of Coim-
bra. His research interests include stretchable
electronics, printed electronics, biosignals acqui-
sition, human machine interfaces, and wearable
Aníbal T. De Almeida (Life Senior Member,
IEEE) received the degree in electrical engi-
neering from the Faculty of Engineering, Porto
University, in 1972, the Ph.D. degree in com-
puter control from London University in 1977,
and the Aggregation degree in electrical engi-
neering from the University of Coimbra in 1981.
He is currently a Full Professor with the Electrical
Engineering Department, University of Coimbra,
and the Director of the Institute for Systems and
Robotics, Coimbra.
Mahmoud Tavakoli (Member, IEEE) is currently
an Assistant Professor with the Department
of Electrical Engineering with the University of
Coimbra. He is the author of over 70 publications.
His current research interests are wearable com-
puting, soft and stretchable electronics, printed
electronics, and human-machine interfaces.
Authorized licensed use limited to: Carnegie Mellon Libraries. Downloaded on February 22,2022 at 14:54:24 UTC from IEEE Xplore. Restrictions apply.
... [40]. (c) The e-textile headband system overview: (left) system block diagram; (middle) processing and transmission board for the e-textile EEG system connected to multiple amplifiers; and (right) e-textile headband in use [41]. [40]. ...
... [40]. (c) The e-textile headband system overview: (left) system block diagram; (middle) processing and transmission board for the e-textile EEG system connected to multiple amplifiers; and (right) e-textile headband in use [41]. To validate this system, sleep data were collected from 29 healthy subjects using both PSG and the designed system in a sleep laboratory setting over the course of one night. ...
... In 2020, Carneiro and his collaborators developed a wearable and comfortable e-textile headband designed for long-term forehead EEG signal acquisition [41]. Their aim was to address the challenges posed by the bulky size and complex wiring of current EEG monitoring systems, which often require significant time for set-up. ...
Full-text available
Sleep is a fundamental aspect of daily life, profoundly impacting mental and emotional well-being. Optimal sleep quality is vital for overall health and quality of life, yet many individuals struggle with sleep-related difficulties. In the past, polysomnography (PSG) has served as the gold standard for assessing sleep, but its bulky nature, cost, and the need for expertise has made it cumbersome for widespread use. By recognizing the need for a more accessible and user-friendly approach, wearable home monitoring systems have emerged. EEG technology plays a pivotal role in sleep monitoring, as it captures crucial brain activity data during sleep and serves as a primary indicator of sleep stages and disorders. This review provides an overview of the most recent advancements in wearable sleep monitoring leveraging EEG technology. We summarize the latest EEG devices and systems available in the scientific literature, highlighting their design, form factors, materials, and methods of sleep assessment. By exploring these developments, we aim to offer insights into cutting-edge technologies, shedding light on wearable EEG sensors for advanced at-home sleep monitoring and assessment. This comprehensive review contributes to a broader perspective on enhancing sleep quality and overall health using wearable EEG sensors.
... Subsequent research by the same team led to the integration of these sensors into the fabric, culminating in the creation of wireless, non-contact EEG-monitoring headbands [75]. Carneiro (2020) further refined this technology, introducing a second layer of contact electrodes to mitigate noise from extended wires and electromagnetic interference (EMI) [76]. They combined this with the Internet of Things (IoT) technology, resulting in an EEG headband equipped with printed silver-based conductive fabric electrodes. ...
Full-text available
This study aimed to systematically review the application and research progress of flexible microfluidic wearable devices in the field of sports. The research team thoroughly investigated the use of life signal-monitoring technology for flexible wearable devices in the domain of sports. In addition, the classification of applications, the current status, and the developmental trends of similar products and equipment were evaluated. Scholars expect the provision of valuable references and guidance for related research and the development of the sports industry. The use of microfluidic detection for collecting biomarkers can mitigate the impact of sweat on movements that are common in sports and can also address the issue of discomfort after prolonged use. Flexible wearable gadgets are normally utilized to monitor athletic performance, rehabilitation, and training. Nevertheless, the research and development of such devices is limited, mostly catering to professional athletes. Devices for those who are inexperienced in sports and disabled populations are lacking. Conclusions: Upgrading microfluidic chip technology can lead to accurate and safe sports monitoring. Moreover, the development of multi-functional and multi-site devices can provide technical support to athletes during their training and competitions while also fostering technological innovation in the field of sports science.
... According to the different signal acquisition methods, BCIs can be divided into invasive (Rapeaux and Constandinou, 2021) and non-invasive (Jo and Choi, 2018;Zhuang et al., 2020). In non-invasive BCIs, electroencephalography (EEG) signals are acquired by external sensors with multiple channels (Carneiro et al., 2020;Singh et al., 2021). In EEG systems, the greater number of the channels, the more comprehensive the signal obtained. ...
Full-text available
Background For non-invasive brain-computer interface systems (BCIs) with multiple electroencephalogram (EEG) channels, the key factor limiting their convenient application in the real world is how to perform reasonable channel selection while ensuring task accuracy, which can be modeled as a multi-objective optimization problem. Therefore, this paper proposed a two-objective problem model for the channel selection problem and introduced a domain knowledge-assisted multi-objective optimization algorithm (DK-MOEA) to solve the aforementioned problem. Methods The multi-objective optimization problem model was designed based on the channel connectivity matrix and comprises two objectives: one is the task accuracy and the other one can sensitively indicate the removal status of channels in BCIs. The proposed DK-MOEA adopted a two-space framework, consisting of the population space and the knowledge space. Furthermore, a knowledge-assisted update operator was introduced to enhance the search efficiency of the population space by leveraging the domain knowledge stored in the knowledge space. Results The proposed two-objective problem model and DK-MOEA were tested on a fatigue detection task and four state-of-the-art multi-objective evolutionary algorithms were used for comparison. The experimental results indicated that the proposed algorithm achieved the best results among all the comparative algorithms for most cases by the Wilcoxon rank sum test at a significance level of 0.05. DK-MOEA was also compared with a version without the utilization of domain knowledge and the experimental results validated the effectiveness of the knowledge-assisted mutation operator. Moreover, the comparison between DK-MOEA and a traditional classification algorithm using all channels demonstrated that DK-MOEA can strike the balance between task accuracy and the number of selected channels. Conclusion The formulated two-objective optimization model enabled the selection of a minimal number of channels without compromising classification accuracy. The utilization of domain knowledge improved the performance of DK-MOEA. By adopting the proposed two-objective problem model and DK-MOEA, a balance can be achieved between the number of the selected channels and the accuracy of the fatigue detection task. The methods proposed in this paper can reduce the complexity of subsequent data processing and enhance the convenience of practical applications.
... Recently, there has been increasing interest in soft electronics in various fields, such as electronic skin [1][2][3], high-performance transistors [4,5], wearable skin patch biosensors [6][7][8], implantable healthcare devices [9][10][11], and human-machine interface applications [12]. In particular, research focused on the accurate control of prosthetics using wearable bioelectronics [13,14] was conducted frequently, which required stable and precise physiological signals from the body, especially electrocardiography (ECG), electromyography (EMG), and electroencephalography signals [15][16][17]. ...
Full-text available
In wearable bioelectronics, various studies have focused on enhancing prosthetic control accuracy by improving the quality of physiological signals. The fabrication of conductive composites through the addition of metal fillers is one way to achieve stretchability, conductivity, and biocompatibility. However, it is difficult to measure stable biological signals using these soft electronics during physical activities because of the slipping issues of the devices, which results in the inaccurate placement of the device at the target part of the body. To address these limitations, it is necessary to reduce the stiffness of the conductive materials and enhance the adhesion between the device and the skin. In this study, we measured the electromyography (EMG) signals by applying a three-layered hydrogel structure composed of chitosan–alginate–chitosan (CAC) to a stretchable electrode fabricated using a composite of styrene–ethylene–butylene–styrene and eutectic gallium-indium. We observed stable adhesion of the CAC hydrogel to the skin, which aided in keeping the electrode attached to the skin during the subject movement. Finally, we fabricated a multichannel array of CAC-coated composite electrodes (CACCE) to demonstrate the accurate classification of the EMG signals based on hand movements and channel placement, which was followed by the movement of the robot arm.
... The longitudinal monitoring of biopotential signals like electrocardiogram (ECG), electromyogram (EMG), and electroencephalogram (EEG) enables a wide range of use for cases such as the remote assessment of medical disorders [1], athletic performance [2], and brain-computer interfaces [3]. Contact-based biopotential-sensing electrodes such as Ag/Ag-Cl electrodes typically use gel or adhesive which provides minimal skin-electrode impedance [4]. ...
Full-text available
Biopotential electrodes play an integral role within smart wearables and clothing in capturing vital signals like electrocardiogram (ECG), electromyogram (EMG), and electroencephalogram (EEG). This study focuses on dry e-textile electrodes (E1–E6) and a laser-cut knit electrode (E7), to assess their impedance characteristics under varying contact forces and moisture conditions. Synthetic perspiration was applied using a moisture management tester and impedance was measured before and after exposure, followed by a 24 h controlled drying period. Concurrently, the signal-to-noise ratio (SNR) of the dry electrode was evaluated during ECG data collection on a healthy participant. Our findings revealed that, prior to moisture exposure, the impedance of electrodes E7, E5, and E2 was below 200 ohm, dropping to below 120 ohm post-exposure. Embroidered electrodes E6 and E4 exhibited an over 25% decrease in mean impedance after moisture exposure, indicating the impact of stitch design and moisture on impedance. Following the controlled drying, certain electrodes (E1, E2, E3, and E4) experienced an over 30% increase in mean impedance. Overall, knit electrode E7, and embroidered electrodes E2 and E6, demonstrated superior performance in terms of impedance, moisture retention, and ECG signal quality, revealing promising avenues for future biopotential electrode designs.
... Forehead EEG recording techniques make wearability more accessible. Golparvar et al. developed a graphene-based e-textile interface that can record brain waves [84] . This graphene-based textile EEG interface is made with a Dip-Dry-Reduce method. ...
Full-text available
Human interaction with machines can be made easy, comfortable, and accessible by introducing user-friendly interfaces. In the case of wearable devices, their sensors and other interfacing elements are very well within the proximity of users. Since biopotential signals can be accessed from the surface of the human skin, users can have seamless interaction with wearable human-computer interactive devices. Rigid interfaces can hinder the user experience, and therefore, the need for soft biopotential interfaces is important. Imperceptible and unobtrusive soft biopotential interfaces will drastically enhance many aspects of human-computer interaction. This paper reviews the use of soft, flexible, and stretchable biopotential interfaces in wearable human-machine interactive devices. Additionally, attention is brought to the scope of other possible applications of soft biopotential interfaces in wearable devices.
... Smartwatches [7], [8] are the most diffused typology, given their similarity to common watches, making them particularly user-friendly. But there are also T-shirts [9], smart glasses [10], headbands [11], inear sensors [12], and smart rings [13]. Most of them can be worn with no trouble in indoor environments during activities of daily living, like at the workplace [14], at home [15], or even in sports facilities [16]. ...
Conference Paper
The development of personal comfort models (PCMs) is pivotal for the optimization of both the occupants’ comfort and the building energy consumption; wearable sensors assessing physiological parameters can be exploited for this aim. This paper considers commercial and laboratory prototypes of wearable sensors and aims at evaluating the propagation of the input measurement uncertainties on the features computed for the development of PCMs. Different types of wearables, suitable for physiological monitoring in a living environment, are considered: smartwatch, smartband, smart garment, smart ring, patch, and headband. The Monte Carlo simulation method is exploited for the uncertainty analysis. The results show that some features are more robust than others and this is relevant when selecting a feature subset for a model creation. The findings are useful for the design of PCMs aiming at improving the quality of life in indoor living environments, developed with a human-centric view.
... For this setup the operating time of the system will be increased more than 24 hours. It is the low-cost EEG system [5]. EEG signals can be acquired ultra-shielded capsule. ...
Full-text available
Brain computer interfaces (BCI) are the fledgling field to rehabilitate the immobilized people. This BCI technology can succour paralyzed patients to operate wheelchairs independently for locomotion, also to lift and carry the objects based on brain-neuronal activity with robotic control. EEG (Electroencephalography) is a device used to provide information immeasurably identifying about brain conditions and disabilities with an effective stimulus using graphite or noble metals. It indicates extremely herculean and targeted EEG applications to guide devices utilizing brain activity. This study gives the survey of (ML) machine learning and (DL) deep learning associated with MI (Motor Imagery), MeI (Mental Imagery) and ME (Motor Execution) gesture classifications applicable for BCI. There are two public domain datasets (PhysioNet, BCIC-Motor Execution) and self-collected datasets were utilized for computerized process since inception.
Full-text available
Soft and stretchable electronics have diverse applications in the fields of compliant bioelectronics, textile‐integrated wearables, novel forms of mechanical sensors, electronics skins, and soft robotics. In recent years, multiple material architectures have been proposed for highly deformable circuits that can undergo large tensile strains without losing electronic functionality. Among them, gallium‐based liquid metals benefit from fluidic deformability, high electrical conductivity, and self‐healing property. However, their deposition and patterning is challenging. Biphasic material architectures are recently proposed as a method to address this problem, by combining advantages of solid‐phase materials and composites, with liquid deformability and self‐healing of liquid phase conductors, thus moving toward scalable fabrication of reliable stretchable circuits. This article reviews recent biphasic conductor architectures that combine gallium‐based liquid‐phase conductors, with solid‐phase particles and polymers, and their application in fabrication of soft electronic systems. In particular, various material combinations for the solid and liquid phases in the biphasic conductor, as well as methods used to print and pattern biphasic conductive compounds, are discussed. Finally, some applications that benefit from biphasic architectures are reviewed.
Sleep recordings are increasingly being conducted in patients' homes where patients apply the sensors themselves according to instructions. However, certain sensor types such as cup electrodes used in conventional polysomnography are unfeasible for self-application. To overcome this, self-applied forehead montages with electroencephalography and electro-oculography sensors have been developed. We evaluated the technical feasibility of a self-applied electrode set from Nox Medical (Reykjavik, Iceland) through home sleep recordings of healthy and suspected sleep-disordered adults (n = 174) in the context of sleep staging. Subjects slept with a double setup of conventional type II polysomnography sensors and self-applied forehead sensors. We found that the self-applied electroencephalography and electro-oculography electrodes had acceptable impedance levels but were more prone to losing proper skin-electrode contact than the conventional cup electrodes. Moreover, the forehead electroencephalography signals recorded using the self-applied electrodes expressed lower amplitudes (difference 25.3%-43.9%, p < 0.001) and less absolute power (at 1-40 Hz, p < 0.001) than the polysomnography electroencephalography signals in all sleep stages. However, the signals recorded with the self-applied electroencephalography electrodes expressed more relative power (p < 0.001) at very low frequencies (0.3-1.0 Hz) in all sleep stages. The electro-oculography signals recorded with the self-applied electrodes expressed comparable characteristics with standard electro-oculography. In conclusion, the results support the technical feasibility of the self-applied electroencephalography and electro-oculography for sleep staging in home sleep recordings, after adjustment for amplitude differences, especially for scoring Stage N3 sleep.
Full-text available
Bioelectronics stickers that interface the human epidermis and collect electrophysiological data will constitute important tools in the future of healthcare. Rapid progress is enabled by novel fabrication methods for adhesive electronics patches that are soft, stretchable and conform to the human skin. Yet, the ultimate functionality of such systems still depends on rigid components such as silicon chips and the largest rigid component on these systems is usually the battery. In this work, we demonstrate a quickly deployable, untethered, battery-free, ultrathin (~5 μm) passive “electronic tattoo” that interfaces with the human skin for acquisition and transmission of physiological data. We show that the ultrathin film adapts well with the human skin, and allows an excellent signal to noise ratio, better than the gold-standard Ag/AgCl electrodes. To supply the required energy, we rely on a wireless power transfer (WPT) system, using a printed stretchable Ag-In-Ga coil, as well as printed biopotential acquisition electrodes. The tag is interfaced with data acquisition and communication electronics. This constitutes a “data-by-request” system. By approaching the scanning device to the applied tattoo, the patient’s electrophysiological data is read and stored to the caregiver device. The WPT device can provide more than 300 mW of measured power if it is transferred over the skin or 100 mW if it is implanted under the skin. As a case study, we transferred this temporary tattoo to the human skin and interfaced it with an electrocardiogram (ECG) device, which could send the volunteer’s heartbeat rate in real-time via Bluetooth.
Full-text available
Stretchable electronics are becoming an important branch in the field of electronics, with a rapidly growing number of studies on highly deformable circuits that can handle large tensile strains without losing electronic functionality. However, as the field continues to advance, a variety of manufacturing approaches must be explored so that these circuits can be produced in a rapid, precise, repeatable, inexpensive, and high volume manner. Even in cases where methods from conventional printed circuit board (PCB) manufacturing are used, these techniques must be significantly modified in order to process the soft materials and fluids that are unique to stretchable electronic architectures. Digital printing methods such as Inkjet printing, 3D printing and laser ablation are especially attractive since they can be automated and eliminate the need for a stencil, mask or clean-room lithography. This article reviews recent advances in digital fabrication of stretchable circuits using either additive techniques, that allow direct printing of certain materials with a determined shape, or subtractive methods, in which the desired pattern is obtained by selectively removing parts of a greater film. For most of the materials used in stretchable electronics, these fabrication methods can be simple to implement, low cost, scalable over large areas, compatible with a broad variety of materials and have a size resolution that is comparable to conventional PCB manufacturing.
Full-text available
The primary developing trends in flexible and stretchable electronics involve the innovation of material synthesis, mechanical design, and fabrication strategies that employ soft substrates. The biggest challenge is that the entire electronic system must allow not only bending but also stretching. Therefore, stretchable conductors become a crucial construction unit for the connection of working circuits of various stretchable devices. Owing to the success of stretchable conductors, various stretchable electronic devices are fabricated with the help of multiple manufacturing strategies, including stretchable heaters, stretchable energy conversion and storage devices, stretchable transistors, sensors and artificial skin. The continuous development of stretchable electronics has led to the new functionality of transparency, and the fabrication of transparent stretchable electronic devices has gained a lot of interest due to the potential of wearable electronic systems. This review presents technology developments in the preparation of related materials, fabrication strategies and various applications of stretchable electronics. It focuses on the fundamental structural design, mechanisms, and tactics, as well as on challenges and opportunities in the manufacture of stretchable electronic devices and their various applications. © 2019, © 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
Full-text available
Brain-computer interface (BCI) technology shows potential for application to motor rehabilitation therapies that use neural plasticity to restore motor function and improve quality of life of stroke survivors. However, it is often difficult for BCI systems to provide the variety of control commands necessary for multi-task real-time control of soft robot naturally. In this study, a novel multimodal human-machine interface system (mHMI) is developed using combinations of electrooculography (EOG), electroencephalography (EEG), and electromyogram (EMG) to generate numerous control instructions. Moreover, we also explore subject acceptance of an affordable wearable soft robot to move basic hand actions during robot-assisted movement. Six healthy subjects separately perform left and right hand motor imagery, looking-left and looking-right eye movements, and different hand gestures in different modes to control a soft robot in a variety of actions. The results indicate that the number of mHMI control instructions is significantly greater than achievable with any individual mode. Furthermore, the mHMI can achieve an average classification accuracy of 93.83% with the average information transfer rate of 47.41 bits/min, which is entirely equivalent to a control speed of 17 actions per minute. The study is expected to construct a more user-friendly mHMI for real-time control of soft robot to help healthy or disabled persons perform basic hand movements in friendly and convenient way.
Tactile sensing is essential for skilled manipulation and object perception. Existing sensing devices cannot capture the full range of tactile information in the naturally behaving hand, and are unable to match human abilities of perception and action. Human touch sensing is mediated via contact-generated mechanical signals in the skin. Time varying contacts elicit propagating mechanical waves that are captured via numerous vibration-sensitive neurons distributed throughout the hand, yielding a wealth of sensory information. Little engineering attention has been given to important biological sensing system. Inspired by human sensing abilities, we present a wearable system based on a 126 channel sensor array capable of capturing high resolution tactile signals throughout the hand during natural manual activities. It employs a network of miniature threeaxis sensors mounted on a flexible circuit whose geometry is adapted to the anatomy of the hand, allowing tactile data to be captured during natural manual interactions. Each sensor possesses a frequency bandwidth overlapping the entire human tactile frequency range. Data is acquired in real time via a custom FPGA and an I2C network. We also present physiologically informed signal processing methods for reconstructing whole hand tactile signals. We report experiments that demonstrate the utility of this system for collecting rich tactile signals during manual interactions.
Stretchable electronics stickers that adhere to the human skin and collect biopotentials are becoming increasingly popular for biomonitoring applications. Such stickers should include electrodes, stretchable interconnects, silicon chips for processing and communication, and batteries. Here, we demonstrate a material architecture and fabrication technique for a multi-layer, stretchable, low-cost, rapidly deployable and disposable sticker that integrates skin interfacing hydrogel electrodes, stretchable interconnects, and Ag2O-Zn (Silver Oxide – Zinc) battery. In addition, the application of a printed biphasic current collector (AgInGa) for the Ag2O-Zn battery is reported for the first time. Surprisingly, and unlike previously reported batteries, the battery capacity increases after being subject to strain cycles and reaches to a record-breaking areal capacity of 6.88 mAh cm-2 post stretch. As a proof of concept, an application of heart rate monitoring is presented. The disposable patch is interfaced with a miniature battery-free electronics circuit for data acquisition, processing, and wireless transmission. A version of the patch partially covering the patient´s chest can supply enough energy for continuous operation for ~6 days.
An 8-channel wearable wireless device for ambulatory surface EEG monitoring and analysis is presented. The entire multi-channel recording, quantization, and motion artifact removal circuitries are implemented on a 4-layer polyimide flexible substrate. The recording electrodes and active shielding are also integrated on the same substrate, yielding the smallest form factor compared to the state of the art. Thanks to the dry non-contact electrodes, the system is quickly mountable with minimal assistance required, making it an ideal ambulatory front and temporal-lobe EEG monitoring device. The flexible main board is connected to a rechargeable battery on one end and to a $13 \times 17 \text{mm}^2$ rigid board on the other end. The mini rigid board hosts a low-power programmable FPGA and a BLE 5.0 transceiver, which add diagnostic capability and wireless connectivity features to the device, respectively. Design considerations for a wearable EEG monitoring and diagnostic device are discussed in details. The theory of the novel fully-analog method for motion artifact detection and removal is described and the detailed circuit implementation is presented. The device performance in terms of voltage gain (260 V/V), bandwidth (DC-300 Hz), motion artifact removal, and wireless communication throughput (up to 1Mbps) is experimentally validated. The entire wearable solution with the battery weighs 9.2 grams.
We tackle two well-known problems in the fabrication of stretchable electronics: interfacing soft circuit wiring with silicon chips and fabrication of multi-layer circuits. We demonstrate techniques that allow integration of embedded flexible printed circuit boards (FPCBs) populated with microelectronics into soft circuits composed of liquid metal (LM) interconnects. These methods utilize vertical interconnect accesses (VIAs) that are produced by filling LM alloy into cavities formed by laser ablation. The introduced technique is versatile, easy to perform, clean-room free, and results in reliable multi-layer stretchable hybrid circuits that can withstand over 80% of strain. We characterize the fabrication parameters of such VIAs and demonstrated several applications, including a stretchable touchpad and pressure detection film, and an all-integrated multi-layer electromyography (EMG) circuit patch with five active layers including acquisition electrodes, on-board processing and Bluetooth communication modules.
Electrodes are used to convert ionic currents to electrical currents in biological systems. Modeling the electrode electrolyte interface and characterizing the impedance of the interface could help to optimize the performance of the electrode interface to achieve higher signal to noise ratios. Previous work has yielded accurate models for single-element biomedical electrodes. This paper introduces a model for a tripolar concentric ring electrode (TCRE) derived from impedance measurements using electrochemical impedance spectroscopy (EIS) with a Ten20 electrode impedance matching paste. It is shown that the model serves well to predict the performance of the electrode-electrolyte interface for TCREs as well as standard cup electrodes. In this paper we also discuss the comparison between the TCRE and the standard cup electrode regarding their impedance characterization and demonstrate the benefit of using TCREs in biomedical applications. We have also conducted auditory evoked potential experiments using both TCRE and standard cup electrodes. The results show that EEG recorded from tripolar concentric ring electrodes (tEEG) is beneficial, acquiring the auditory brainstem response (ABR) with less stimuli with respect to recording EEG using standard cup electrodes.