4/27/18

Contents
Bioinstrumentation 1
Other Biosignals — Electromyography (EMG)
— Magnetoencephalography (MEG)
— Mechanomyography (MMG)
— Electrodermal Activity (EDA)
— Electrooculography (EOG)
— Phonocardiography (PCG)

Prepared by: Mohd Hanafi bin Mat Som, D.Eng

Biopotentials Voltage and Frequency Electromyogram (EMG)

EMG freq spans from 20 Hz to several kHz.
— The EMG amplitude depends on:
Researchers focus between 20 to 450 Hz only. i. Ranges from µV to mV, typically from 0.1 to 0.5 mV
ii. Types of electrodes.
iii. Placement of electrodes.
ECG: 0.05 Hz to 100 Hz
iv. Degree of muscular exertions.

EEG: 0.5 Hz to 100 Hz

1
 4/27/18

Electromyogram (EMG) Electromyogram (EMG)

— The following are the properties of an evoked — Disadvantage of using surface electrode:
extracellular potentials of SMU: i. Only used with superficial muscle.
i. Triphasic waveform. ii. Sensitive to electrical activity over too wide an area.
ii. Duration of 3-15ms. — Types of electrode for deep tissue recording:
iii. Amplitude of 20-2000µV. i. Monopolar
iv. Discharge frequency of 6-30 per second. ii. Bipolar
iii. Multipolar insertion-type

EMG Recording EMG Preamplifier

— Differential type, INA
— High input impedance, greater than several orders of
electrode impendace
— Mounted near the subject
— High CMRR, greater than 90 dB up to 5 kHz

— A calibrating square wave signal of 100 µVpp at 100 Hz

2
 4/27/18

EMG Preamplifier EMG Preamplifier

— Very small electrode leads to avoid
i. undesirable effects of stray capacitance between
connecting cables and the earth.
ii. significant noise in the cable, which feed into the
subsequent amplifier.
— The larger the surface of the electrode, the less input
resistance is allowable.
i. a needle electrode with a surface of 15,000 μm may
need an amplifier with input impedance of 5 MΩ.
ii. A needle electrode with a surface of 500 μm will
ensure a record with acceptable distortion by means of
an amplifier with minimum input impedance of 100 MΩ.

EMG Artefacts EMG Artefacts

— Power line interference, 50/60 Hz — DC offsets:- due to differences in the electrical
o Notch filter, but might remove the signal of interest impedances between the skin and electrodes.
— ECG signals — Muscle crosstalk:- due to electrical signals generated by
o HPF at 100 Hz, but might remove the signal of interest other muscles.
— Movement whenever patient moves and the electrodes o Appropriate placement of electrodes.
are disturbed or the cables are pulled. o Recommended inter-electrode distance is about 2 cm.
o Firm contact between electrodes-skin
o Fasten the electrodes cables

3
 4/27/18

EMG: References Magnetoencephalography (MEG)

— Khandpur, R.S. (2014). Handbook of Biomedical — MEG measures the magnetic fields generated
Instrumentation, 3rd Ed., McGraw-Hill. by electric currents in the brain.
— The magnetic field are detected by a
sophisticated technology based on
superconducting quantum interference device
(SQUID)

Magnetoencephalography (MEG) Magnetoencephalography (MEG)

— The magnetic field measurements are in the
range of femto-tesla (10−15 T) to pico-tesla
(10−12 T).
— Provides a very accurate resolution of the
timing of neuronal activity
— MEG vs EEG?
— Two approved indications in the United
States, one is for pre-operative Brain Mapping
and the other is for use in epilepsy surgery.

4
 4/27/18

Magnetoencephalography (MEG) Magnetoencephalography (MEG)

MEG: References Mechanomyography (MMG)

— Magnetoencephalography: Basic principles — MMG detect the mechanical signal generated
— Basic Principles of Magnetoencephalography by lateral oscillations of the muscle fibres
during contraction or relaxation.
— MMG is considered the mechanical
counterpart of EMG and it is rich in
information on the activity and physical
properties of the muscle.

5
 4/27/18

Mechanomyography (MMG) Mechanomyography (MMG)

— MMG may be measured by microphones,
piezoelectric contact sensors, accelerometers
or laser distance sensors on the surface of
the skin.
— MMG has several applications, both in BMI as
a control signal for externally-powered
prosthetic devices and in neuroscience,
medical and biomechanics research, e.g. as a Fig. 1. (A) Microphone biasing circuit to provide reference zero-point voltage for microphone signal. (B1) Microphone sound chamber with (B2) top

diagnostic tool. view and (B3) lateral view. All length dimensions shown are in mm. Instrumentation circuit with microphone, (C1) gain stage and (C2) 5th order low-pass
Butterworth filter. The gain stage is connected to the biasing circuit via the Vbias connection and provides zero-signal at 2.5 V. (D) The bode plot of the
instrumentation circuit showing a clear cut-off at 200Hz.

different subjects without needing a per subject calibration. to bias the signal by the half supplied voltage, since the
Finally, we apply a pattern-recognition algorithm on our A/D converter has a single voltage reference with precision
MMG signal in order prove its effectiveness as a control of 5 mV. Our analog filter provides a frequency cut-off at
signal in BMI applications. 200Hz, which preserves the MMG frequency components.
Additionally, the 5th order Butterworth filter guarantees a
II. METHODS good compromise between high attenuation rate and flat-
band pass. We designed the Butterworth filter in Filterlab
Our MMG sensor and instrumentation circuit (see Fig. 1)
(Microchip Technology Inc., Chandler, AZ) and the whole
are based on novel designs aiming at improving signal
circuit was simulated in LTspice (Linear Technology, Milpi-
quality and stability while reducing the manufacturing cost.
tas, CA) prior to production.
MMG sensor design: We use electret condenser mi-
crophones for our MMG signal detector, as they have been
proven to be robust to body motion artefacts [9]. To get even
better mechanical response via the air column, we confined

MMG: References Electrodermal Activity (EDA)

the air chamber between the muscle and the microphone
diaphragm (Fig. 1.B2) as to concentrate the pressure waves
and transfer the forces generated to the diaphragm more
efficiently. The microphone chamber (Fig. 1.B1) has been
designed to guarantee a good response to low frequency
— Robust, ultra low-cost MMG system with — EDA is a term used to describe changes in
signals, as the MMG signal frequency band has been shown
to lie in the 5 100 Hz interval[16]. The bottom conical

Brain-Machine-Interface applications the electrical properties of the skin resulting

section (Fig. 1.B3) of the chamber ensures that sensor col-
lects mechanical signal from the wider surface. The straight
cylindrical section, above the conical section, stabilises the
from autonomic nervous system functions
concentrated pressure wave and enhances the sensitivity of
the microphone by reducing the effective area. Finally, the

[Dawson et al., 2000]

rubber section around the chamber area (yellow annular in
Fig. 1.B1), seals the chamber to avoid ambient noise whereas
the overall tiny sensor size provides higher precision when

— Due to activation of sweat glands that are

used on the small muscles.
MMG instrumentation: The MMG instrumentation Fig. 2. Experimental setup arrangement showing a healthy subject sitting
circuit consists of a microphone with its pre-amplifier on a chair, having the elbow at right angle position while resting the arm

controlled by the sympathetic nervous

(Fig. 1.C1), a low- pass Butterworth filter (Fig. 1.C2) and
a data acquisition unit. The pre-amplifier part amplifies the
on an aluminium platform. The force applied is measured using a force
transducer and a visual feedback is provided on the monitor. The MMG
sensor and EMG electrodes are attached on the belly of the right hand

system, which autonomously regulates the

signal collected by the microphone and provides an offset
724
biceps brachii.

mobilisation of the human body for action.

6
 Section 1.2: Terms & Definitions
4/27/18
Units

The typical units of electrodermal activity are (i) the microsiemens (µS) or (ii) the micromho
(µmho). Both units are equivalent. So 1µS is equal to 1µmho.

Electrodermal Activity (EDA) Electrodermal Activity (EDA)

Table 1. Basic definitions for Electrodermal components (adapted from Dawson et al,
— EDA has been closely linked to autonomic 2001). — EDA measurement registers the inverse of
emotional and cognitive processing, and EDA the electrical resistance ‘ohm’
Measure between two
Definition
Skin conductance level Tonic level of electrical conductivity of skin
is a widely used as a sensitive index of points on the skin.
(SCL)
emotional processing and sympathetic activity. Skin conductance response
— The typical unitsPhasic change in electrical conductivity of skin
of electrodermal activity are
(SCR)
— Clinical application such as assessment of i. the microsiemens (μS) or
Non-specific SCR SCRs that occur in the absence of an identifiable eliciting
(NS-SCRs) stimuli
pain, schizophrenia, and peripheral Frequency ii. the micromhoRate
of NS-SCRs (μmho).
of NS-SCRs that occur in the absence of
neuropathy. BothSCR
Event-related units are SCRs
equivalent.
that canSo
identifiable stimuli
1μS is equal
be attributed to 1μmho.
to a specific eliciting stimuli
— (ER-SCR)

There are two main components to the overall complex referred to as EDA. One component
is the general tonic-level EDA which relates to the slower acting components and
background characteristics of the signal (the overall level, slow climbing, slow declinations
over time). The most common measure of this component is the Skin Conductance Level
(SCL) and changes in the SCL are thought to reflect general changes in autonomic arousal.
The other component is the phasic component and this refers to the faster changing
Section 1.2: Terms & Definitions
elements of the signal - the Skin Conductance Response (SCR). Recent evidence suggests
that both components are important and may rely on different neural mechanisms (Dawson

Electrodermal Activity (EDA)

Units

The typical units of electrodermal activity are (i) the microsiemens (µS) or (ii) the micromho
Electrodermal Activity (EDA)
et al., 2001; Nagai et al., 2004). Crucially, it is important to be aware that the phasic SCR,
which often receives the most interest, only makes up a small proportion of the overall EDA
(µmho). Both units are equivalent. So 1µS is equal to 1µmho. complex.

Slow changes
Table 1. Basic definitions for Electrodermal components (adapted from Dawson et al, &
2001). DC components
Tonic
Measure Definition EDA
Skin conductance level Tonic level of electrical conductivity of skin Skin Conductance Level
(SCL)
(SCL)
Skin conductance response Phasic change in electrical conductivity of skin
(SCR) Electrodermal Activity
Non-specific SCR SCRs that occur in the absence of an identifiable eliciting (EDA)
(NS-SCRs) stimuli Non-specific SCRs
Frequency of NS-SCRs Rate of NS-SCRs that occur in the absence of (NS-SCRs)
identifiable stimuli
Phasic
Event-related SCR SCRs that can be attributed to a specific eliciting stimuli
(ER-SCR) SCRs
Event-related SCRs
ER-SCRs
There are two main components to the overall complex referred to as EDA. One component
is the general tonic-level EDA which relates to the slower acting components and
background characteristics of the signal (the overall level, slow climbing, slow declinations
over time). The most common measure of this component is the Skin Conductance Level 4
(SCL) and changes in the SCL are thought to reflect general changes in autonomic arousal.
The other component is the phasic component and this refers to the faster changing
elements of the signal - the Skin Conductance Response (SCR). Recent evidence suggests
that both components are important and may rely on different neural mechanisms (Dawson
et al., 2001; Nagai et al., 2004). Crucially, it is important to be aware that the phasic SCR,
which often receives the most interest, only makes up a small proportion of the overall EDA
complex.
7
Slow changes
&
DC components
Section 1.4: Methodological Issues with Quantifying Phasic ER-SCRs
4/27/18
A general friction that has to be balanced in any experiment seeking to measure EDA is
determining a-priori whether a given SCR is event-related or non-specific. If the criteria are
too loose, one risks including NS-SCRs into the analysis for ER-SCRs and erroneously
thinking that such values might be tied to the experimental manipulation. Too strict and one
risks missing many ER-SCRs to meet your criteria by wrongly discarding or misclassifying
Electrodermal Activity (EDA)
them as NS-SCRs. A graphical representation of the components of an ER-SCR is provided
below.
Electrodermal Activity (EDA)
— The most common recording sites for
electrodermal are the fingers and palms
(~240 eccrine glands per cm2).
Figure 5.23. Principle of bio­feedback
— If those sites are not possible or are
undesirable, electro- dermal activity (EDA)
It may be noted that biofeedback is not a treatment. Rather, biofeedback training is an educational process for learning specialized mind/body skills.
Through practice, one learns to recognize physiological responses and to control them rather than having them control us. The objective of can be detected on the bottom of the feet
biofeedback training is to gain self­regulatory skills which help to adjust the activity in various systems to optimal levels. Many different (~180cm2), and the forehead (~240cm2).
physiological processes have been evaluated for possible control by biofeedback methods. However, the following four neural functions are
commonly employed:

Emotions or Electrodermal Activity (Galvanic skin response measurements)
An example of the components of an ER-SCR (taken from
An exampleMuscle tension or EMG (Electromyograph measurements)
of Dawson
the components of an ER-SCR (taken from Dawson et al., 2001).
et al., 2001).

Temperature/sympathetic pattern (Thermistor readings)

Pulse (Heart rate monitoring)

Electrodermal activity is measured in two ways: BSR (basal skin response) and GSR (galvanic skin response) is a measure of the average activity of
the sweat glands and is a measure of the phasic activity (the high and low points) of these glands. BSR gives the baseline value of the skin resistance
6
whereas GSR is due to the activity of the sweat glands. The GSR is measured most conveniently at the palms of the hand, where the body has the

Electrodermal Activity (EDA) EDA: References

highest concentration of sweat glands. The measurement is made using a dc current source. Silver­silver electrodes are used to measure and record
the BSR and GSR. Fig. 5.24 shows the arrangement for measuring these parameters. The BSR output is connected to an RC network with a time
constant of 3 to 5 seconds which enables the measurement of GSR as a change of the skin resistance.
A Guide for Analysing Electrodermal Activity
—
(EDA) & Skin Conductance Responses
(SCRs) for Psychological Experiments.
— A short review and primer on electrodermal
activity in human computer interaction
applications
— A continuous measure of phasic
electrodermal activity
Figure 5.24. Block diagram for measurement and record of basal skin resistance (BSR) and galvanic skin response (GSR)
Block diagram for measurement and record of basal skin resistance (BSR) and — Khandpur, R.S. (2014). Handbook of Biomedical
galvanic skin response (GSR)
Biofeedback instrumentation for the measurement of EMG, temperature and pulse/heart rate is not different from other instruments used for the Instrumentation, 3rd Ed., McGraw-Hill.
measurement of physiological variables. Transducers and amplifiers are employed to measure the variable that is to be controlled by the feedback
process. The magnitude of the measured variable or changes in the magnitude is converted into a suitable visual or auditory stimulus that is presented
to the subject. Based on the stimulus, the subject learns to control the abnormal conditions. Reports have appeared in literature regarding applications
of biofeedback to control migraine headaches, to slow down heart rate, etc. Biofeedback techniques have been greatly refined and computerized
biofeedback training and psychological computer­assisted guidance programs in the privacy of one's home are now a reality.

8
5.9. MODEL QUESTIONS
 4/27/18

Electrooculagraphy (EOG) Electrooculagraphy (EOG)

— Recording of the biopotentials generated by — One pair of electrodes is placed above and
the movement of the eye ball. below the eye to pick up voltages
— EOG potentials are picked up by small corresponding to vertical movements of the
surface electrodes placed on the skin near eye ball.
the eye. — Another pair of electrodes is positioned to
the left and right of the eye to measure
horizontal movement.
— EOG has applications mostly for research and
is not widely used for clinical purposes.

Phonocardiography (PCG) PCG: Microphones

— PCG is used for recording the sounds — 2 types:
connected with the pumping action of the ◦ Contact microphone
heart. ◦ Air coupled microphone
— Indication of the heart rate and its — Further categorized into 2 types:
rhythmicity. ◦ Crystal type
— Information regarding effectiveness of blood ◦ Dynamic type
pumping and valve action.

9
 4/27/18

PCG: Microphones PCG: PVDF

— Crystal type: contains a wafer of piezo- — It is a piezoelectric polymer.
electric material, which generates potentials — Produces charges of equal magnitude and
when subjected to mechanical stresses due to opposite polarity on opposite surfaces when
heart sounds. They are smaller in size and a mechanical strain is imposed on the
more sensitive than the dynamic microphone. material.
— Dynamic type: consists of a moving coil — The voltage generated in the sensor due to
having a fixed magnetic core inside it. The coil the flexing motion forms the basis of
moves with the heart sounds and produces a electronic stethoscopes, and real time digital
voltage because of its interaction with the acoustic spectral analysis of heart sounds.
magnetic flux.

PCG: Block Diagram

comprises an amplification and filtering circuit. The output of the signal processing unit is given to the digital interface unit which converts thePCG: Block Diagram
analog signal into a proportional digital signal. The digital signal is then connected to the PC for display on the PC monitor in real time. The signal is
analysed in time and frequency domains with the help of software algorithms developed by Singh and Anand (2007).
Frequency range: 20 to 2000 Hz.
—
— The choice of different filters does not have
to be very critical but in general, sets of four
or five high-pass filters with different cut-off
frequencies and slopes are used in the
commercially available instruments.
— Gain compensation circuits to increase the
Figure 5.14. Block diagram of the phonocardiograph (Adapted from Singh and Anand, 2007) amplification of high frequency signals, which
The amplifier used for a phonocardiograph (PCG) has wide bandwidth with a frequency range of about 20 to 2000 Hz. Filters permit selection of are usually of low intensity.
suitable frequency bands, so that particular heart sound frequencies can be recorded. In general, the high frequency components of cardiovascular
sound have a much smaller intensity than the low frequency components and that much information of medical interest is contained in the relatively
high frequency part of this spectrum (Bekkering and Vollenhoven, 1967). Therefore, high­pass filters are used to separate the louder low frequency
components from the soft and interesting high frequency murmurs. Experiments have shown that the choice of different filters does not have to be
very critical but in general, sets of four or five high­pass filters with different cut­off frequencies and slopes are used in the commercially available
instruments. PCG amplifiers usually have gain compensation circuits to increase the amplification of high frequency signals, which are usually of
low intensity. The frequencies at the higher end of the range are of particular significance in research applications.
10
The appropriate filter characteristics may be selected to attenuate the unwanted frequencies at filter slopes of 12 dB/octave or 24 dB/octave. This is
based on the fact that cardiac vibrations follow the inverse square law which is 12 dB/octave, i.e. as the frequency of the sound is increased, the
 4/27/18

EOG & PCG: References

— Khandpur, R.S. (2014). Handbook of Biomedical
Instrumentation, 3rd Ed., McGraw-Hill.

“Experiments should be reproducible. They should all

fail in the same way…”

The End…

11