Parkinson’s Disease Tremor Suppression
A Double Approach Study - Part 1
Wellington C. Pinheiro, Bruno E. Bittencourt, Lucas B. Luiz, Lucas A. Marcello, Vinicius F. Antonio,
Paulo Henrique A. de Lira, Ricardo G. Stolf and Maria Claudia F. Castro
Department of Electrical Engineering, University Center of FEI, S
˜
ao Bernardo do Campo, Brazil
Keywords:
Parkinson’s Disease, Tremor Suppression, Orthosis, Functional Electric Stimulation.
Abstract:
Parkinson’s Disease (PD) is a neurodegenerative disorder that affects mostly elderly people. Approximately
2% of world population, over 60 years old, lives with PD. This pathology is recognized not only by motor
symptoms such as tremor, postural gait and rigidity, but also, nonmotor symptoms as depression and sleep
abnormalities may be developed as well. In Brazil, according to the Ministry of Health, 200,000 people face
the challenge to develop day-by-day activities due to PD. More than just a disease causing motor disturbances,
PD brings to patients uncertainties about their ability to take care of themselves independently. In this con-
text, assistive technologies assume an important position in order to bring back life quality and self-trust to
PD patients. This work aims to study techniques, develop hardware and software for a better approach in
tremor suppression in order to bring back life quality to PD patients. This study approaches the problem of
flexion/extension carpi radialis tremor suppression using two different strategies. The first is a mechanical sup-
pression based on a servomotor opposing to tremor movement. The second strategy is a functional electrical
stimulator. Both systems are triggered by electromyogram (EMG).
1 INTRODUCTION
Parkinson’s disease is the second most common neu-
rodegenerative disorder in the world, loosing only to
Alzheimer’s. According to Dexter and Jenner (2013)
2% of the world population, over 60 years old, lives
with PD. In Brazil, numbers from Ministry of Health
estimates that 200,000 Brazilians are living with PD,
among those 75% have tremors as a main symptom
(Helmich et al., 2012). As a matter of fact, the
increasing life expectancy rates, also increases PD
prevalence. Projections from United Nations (UN) in-
dicate that in six decades the elderly population will
be 3 times bigger than now, achieving 2 billion peo-
ple world wide. The Brazilian Institute of Geography
and Statistics (IBGE) estimates that in 2050 Brazilian
aged population will be around 54.8 million people.
Economic related aspects of Parkinson’s disease are
also relevant. According to Findley (2007), just in the
United States, PD moves the economy in 23 billion
dollars yearly, considering direct and indirect costs.
A PD patient costs on average US$ 10,000 every year
in the United States (Dexter and Jenner, 2013).
In this context, assistive technology has a crucial
role, not only because market growing projections,
but also to help an increasing portion of world popu-
lation to live without all impairments imposed by PD.
This work focus on action tremor and presents two
methods of tremor suppression based on a mechanical
orthosis controlled by a servomotor and a functional
electrical stimulation (FES) system, both of them trig-
gered by eletromyography (EMG). It is organized as
follow: Section 1 presents an introductory view and
clinical aspects of PD. Section 2 presents alternatives
for PD symptoms control. Section 3 presents materi-
als and methods. Section 4 introduces hardware de-
velopment. Section 5 shows signal processing strate-
gies applied to EMG signal of PD. Finally, section
6 presents results of hardware and software develop-
ment and tests.
1.1 Clinical Aspects
The central nervous system (CNS) consists of neu-
rons interconnected as a network through synapses
1
Corresponding author: Maria Claudia F. Castro, PhD
(mclaudia@fei.edu.br). University Center of FEI, Depar-
tament of Electrical Engineering. Avenida Humberto de
Alencar Castelo Branco, 3972, S
˜
ao Bernardo do Campo -
SP, Zip: 09850-901
Pinheiro W., Bittencourt B., Luiz L., Marcello L., Antonio V., de Lira P., Stolf R. and Castro M.
Parkinsonâ
˘
A
´
Zs Disease Tremor Suppression - A Double Approach Study - Part 1.
DOI: 10.5220/0006152501490155
In Proceedings of the 10th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2017), pages 149-155
ISBN: 978-989-758-216-5
Copyright
c
2017 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved
149
which includes dendrites and axons. Communication
neuron to neuron takes place using neurotransmitters.
Neurotransmitters are brain chemical signaling, with
inhibitory and excitatory function, which is released
at synaptic gap. In this context, substantia nigra pars
compacta (SNpc) has as its role dopamine production,
a inhibitory neurotransmitter. SNpc is connected with
other neuronal structures such as striatum (caudate
and putamen), globus pallidus internal and external
and subthalamic nucleus, together called basal gan-
glia as shown in figure 1 (Jankovic and Tolosa, 2006).
Figure 1: Sagittal and Coronal view. Adapted from Purves
et al. (2011).
For PD these structures are quite important be-
cause when a ”movement command” is started at
motor cortex, the electrical impulses are transmitted
through the striatum, which works like a filter. In the
case of an voluntary movement, the striatum increases
thalamic and cortical neuronal activity making the
movement more likely to happen. However, if it is an
involuntary movement, SNpc neurons are activated,
that in turn use dopamine as inhibitory neurotrans-
mitter, diminishing thalamic and cortical activity sup-
pressing undesired movements. Parkinson’s disease,
among many other problems, affects dopamine pro-
duction causing its diminishment, consequently mak-
ing striatum hyperactivated and difficulting move-
ment control (Jankovic and Tolosa, 2006).
Some motor symptoms are classically used to di-
agnose PD such as bradykinesia, tremor, rigidity, pos-
tural problems, difficult to speak or swallow. How-
ever, there are also some important nonmotor symp-
toms, even though it is hard to associate them with
PD at the first sight, symptoms such as sleep depri-
vation, depression and cognitive impairment may be
observed as well (Helmich et al., 2012).
1.2 Classification and Definitions of
Tremor
Tremor can be defined as an oscillatory rhythmic
activity, even though many pathologies may cause
tremor as a symptom, it is important to emphasize that
voluntary movements is accompanied with tremors,
and the boundary between normal and pathological
tremor is hard to define (Jankovic and Tolosa, 2006).
According to Jankovic and Tolosa (2006) tremors
can be largely classified into three categories as rest-
ing tremor, action tremor and postural tremor. Rest-
ing tremor is defined as the oscillatory manifestation
when a limb is not voluntarily activated and com-
pletely supported against gravity action. The postu-
ral variant is defined when tremor happen on an indi-
vidual maintaining a position opposed to gravity ac-
tion. Action tremor is the one which occur concomi-
tantly to muscular contraction for voluntary move-
ment. This work focus on action tremor because it is
more disabling than rest and postural tremors (Rahimi
et al., 2009).
Another important factor is tremor frequency, also
in agreement with Jankovic and Tolosa (2006) it is
classified as low, medium and high frequency tremor.
Low frequency are tremors with oscillation frequency
lower than 4Hz, medium frequency 4 to 10Hz, and
high frequency bigger than 10Hz.
2 ALTERNATIVES FOR PD
TREMOR SUPPRESSION
Regular treatment for PD has not change too much
over the last decades. It is based on dopaminergic
drugs intake, due to inefficiency in some cases, or
even side effect, new technologies have been devel-
oped to reduce Parkinson’s disease effects on daily
life. Some initiatives are extremely invasive, such as
Deep Brain Stimulation (DBS), and presents inher-
ent risks of a surgical procedures and promising re-
sults. Others are non-invasive technologies some for
tremor suppression on the limbs such as FES and or-
thoses, or for tremor handling devices such as Lyft-
ware and Gyroglove, trying to avoid tremor related
inconveniences. It is important to emphasize that un-
til present moment all approaches, pharmaceuticals or
not, treat just PD symptoms and do not avoid disease
progression, and consequently worsen of associated
symptoms.
2.1 Deep Brain Stimulation (DBS)
Deep Brain Stimulation (DBS) is one of the most so-
phisticated neuromodulation techniques in the world
today. It is analogous to a pacemaker as presented in
figure 2, but differently from a pacemaker it applies
electrical impulses to deep structures into the brain,
with the objective of change or modulate electrical
activity in important structures for motor disorders
BIODEVICES 2017 - 10th International Conference on Biomedical Electronics and Devices
150
Figure 2: Deep Brain Stimulation Device - Illustrative fig-
ure adapted from Medtronic Catalog.
(Oluigbo et al., 2012). It is an alternative for drug
resistant patients.
Although DBS has been considered a highly ef-
ficient method (70-90% successful rate), a large list
of cons must be considered. First of all surgery
for DBS implantation are highly invasive, and also
two surgical procedures are needed for implantation.
Its surgery risks, may include infection, hemorrhage
or even implant rejection. These risks may be in-
creased with patient’s age and previous health con-
ditions. Secondly, DBS surgery is not yet an world
wide affordable treatment.
2.2 Lyftware and Gyroglove
Fortunately not all tremor handling technologies are
invasive or risky. Utensils for daily routine improve-
ment are also being developed. Technologies such as
Liftware 3(b) or Gyroglove 3(a) may be one step in di-
rection to PD patient life quality. According to Mertz
(2016), the idea behind lyftware was not only create
a technology which compensate tremors while people
are eating, but also that patients had the experience of
using fork and spoon as any other person. All sen-
sors and processing are placed at the spoon handle,
and the implemented algorithm compensates tremor
through mechanical actuators continuosly.
Another initiative is Gyroglove. The project con-
sists of a glove equipped with gyroscope, and through
the angular momentum conservation principle oppose
to tremor. It is achieved varying gyroscope spin ve-
locity. After tests in July 2015 in an essential tremor
patient with positive results, Gyroglove’s team be-
lieve clinical trials will begin at the end of 2016.
(a) GyroGlove (b) Liftware
Figure 3: Assistive new technologies for tremor handling
(Mertz, 2016).
2.3 Functional Electrical Stimulation -
FES
Functional electrical stimulation has been defined by
Sujith (2008) as the electrical stimulation of mus-
cles deprived of nervous control in order to improve
impaired motor function. According to Lynch and
Popovic (2008) contractions generated using FES can
be coordinated to actuate joints by stimulating mus-
cles that exert torque about the joint. In this tech-
nology, electrical stimuli is delivered to the patient
through surface or implanted electrodes. Other stud-
ies such as Maneski et al. (2011) and Zhang and Ang
(2007) have used FES as a tremor suppression alter-
native.
2.4 Orthoses
Orthoses are devices used to imobilize, mobilize, ad-
just, alleviate or stabilize limbs affected by motor
disorders or accidents with motor impairment. As
any other medical treatment, they must be prescribed
by an specialized physician (Ottobock, 2016). As
biomedical engineering advances, orthoses perform
an important role in rehabilitation, being applied for
a large number of health problems. According to Ro-
con et al. (2005), upper limbs rehabilitation is based
on affected joint stabilization. Orthoses can be con-
sidered rehab robots used to apply forces or mechan-
ical charges to determined limb. Studies developed
by Hogan (1984), and clinical observations made in
the following years have shown that tremor answers
in different ways to biomechanical charges, acting
as a parallel impedance to the affected limb. This
impedance change has direct effect on tremor. This
property of biomechanical systems brings the possi-
bility to satisfactory handle upper limb tremors us-
ing an orthoses approach (Hogan, 1984; Rocon et al.,
2005). Some works such as those from Taheri et al.
(2014), Seki et al. (2011), Rahimi et al. (2009) and
Rocon et al. (2005) were developed for tremor han-
dling with application of orthoses.
3 MATERIALS AND METHODS
This work was divided into three parts consisting of
candidates selection, electromyography (EMG) sig-
nal acquisition for prototyping purposes and clinical
tests. All procedures described in this section were
submitted and approved by Plataforma Brasil ethical
committee. Medical aspects such as patient selection
were performed by neurologists from Federal Univer-
sity of S
˜
ao Paulo (UNIFESP).
Parkinsonâ
˘
A
´
Zs Disease Tremor Suppression - A Double Approach Study - Part 1
151
As part of the first stage of this work three PD
patients, with tremor as main symptom, were se-
lected in accordance with Hughes et al. (1992) and
United Kingdom Parkinson’s Disease Society Brain
Bank criteria, and one healthy volunteer was selected
for control purposes. Initially, all four volunteers
were clarified about this work risks, tests and pos-
sible achievements. After their agreement and sig-
nature of control documentation a questionnaire was
applied to each one individually, in order to assess
any other medical conditions that might be relevant
to this work. After that, all patients were familiarized
with the procedure for EMG acquisition and aspects
related to both systems in development, FES and me-
chanical orthosis based suppression.
At the second section, which occurred later at
the same day, four surface electrodes were positioned
in agreement with protocols established by ”Surface
EMG for Non-Invasive Assessment of Muscles” (SE-
NIAM), two on wrist flexor muscle, and two on wrist
extensor muscle, to record EMG activity in PD pa-
tients and the control volunteer. The protocol fol-
lowed the sequence of movements previewed in fig-
ure 4.
(a) Isometric (b) Grabbing a
cup
(c) Pinch
Figure 4: Applied protocol for signal acquisition.
First of all, PD patients were oriented to keep both
arms extended in front of the body in isometric posi-
tion as shown in figure 4(a) then the first EMG data set
was recorded. The second EMG acquisition was per-
formed based on figure 4(b) where PD patients were
challenged to develop a small movement of grab a
cup and move it to the mouth. The third and last
set of EMG data was recorded with the patient do-
ing repeatedly the pinch movement as represented in
figure 4(c). The equipment used for all acquisition
during this stage was DELSYS Trigno EMG system.
4 HARDWARE DEVELOPMENT
Following the acquisition stage, the hardware and
software development stage had started. The goal
here was to develop an EMG acquisition and stimu-
lation hardware and a software responsible for tremor
detection and suppression.
The hardware was developed inspired by ar-
duino’s shields. In order to achieve a modular project,
each one of projects represented by 5(a) and 5(b)
were built by modules. The first module is a prepro-
cessing board which integrates a differential amplifier
(INA121P) for EMG signal acquisition, a low pass fil-
ter with cut-off frequency of 500Hz in order to atten-
uate signals other than EMG, and a notch filter tuned
in 60Hz to attenuate power-line noise. Both imple-
mented with UAF42 IC. As a second step EMG signal
is digitalized by the microprocessor A/D converter,
buffered and processed to assess tremor frequency.
The microprocessor chosen was STM32F407. The H-
bridge circuit is used in both alternatives FES and me-
chanical. In the FES system it is responsible for ap-
ply a biphasic current-controlled impulse to patient’s
wrist extensor or flexor muscle. For the orthosis sys-
tem, H-bridge is responsible for change direction of
motor torque. The boost circuit is used only for FES
system, because its need for a higher voltage to apply
stimuli to the arm.
(a) Functional Electrial Stimulator
(b) Orthosis
Figure 5: Block Diagram of both approaches studied.
An orthosis was also developed using 3D printing
technology as shown in figure 6.
Figure 6: Mechanical prototype for wrist flexor/extensor
tremor suppression.
5 SIGNAL PROCESSING
STRATEGY
EMG raw signal from volunteers was first treated at
BIODEVICES 2017 - 10th International Conference on Biomedical Electronics and Devices
152
at a preprocessing board. EMG signal was passed
through a low pass filter with cut-off frequency of
500Hz, at this step the goal was attenuate frequen-
cies higher than those that characteristically compose
EMG signals. After that EMG signal was treated with
a notch filter in order to attenuate 60Hz component
from power line. Convert analog EMG into digital
was next step as shown in figure 7 flowchart. The dig-
italized EMG signal was then stored into a buffer in
order to be offline processed. The choice of an offline
processing was to avoid stimulus related artifacts and
muscle response artifacts mentioned by Widjaja et al.
(2009).
As the flowchart indicates, signal was normalized
and passed through a Fast Fourier Transform (FFT)
in order to assess tremor frequency. According to
Jankovic and Tolosa (2006), Ruonala et al. (2013),
and Helmich et al. (2012) tremor frequency may vary
from patient to patient, but the most common range is
4-10Hz. FFT results were analyzed in the interval 4-
10Hz to find the maximum value that FFT assumed,
and for consequence tremor frequency.
Figure 7: EMG signal processing flow chart.
Using the EMG signal in time domain, the max-
imum point was identified, and using tremor period
calculated through FFT previous steps, impulses were
synthesized based on maximum position and tremor
period. After that the goal was to synchronize either
FES or orthoses actuator with tremor manifestation.
It was achieved using an interruption port and timers
from STM32F407 discovery microprocessor board.
For FES system, circuits were designed to apply
electrical stimulus to wrist flexor muscle, a count-
down timer was implemented and stimulus was ap-
plied to wrist extensor muscle 90ms later. To orthosis
system, circuits implemented change rotation direc-
tion of an electrical motor to oppose wrist movement
cause by flexor contraction and 90ms later due to ex-
tensor contraction.
6 RESULTS
During the signal analysis developed to better under-
stand PD dynamics, EMG signals from wrist exten-
sor and flexor muscles were acquired with DELSYS
Trigno EMG system. Considering first a time domain
approach, as presented in figure 8, exists a delay in
activation between flexor and extensor muscles which
for our three volunteers is approximately 90ms.
Figure 8: Delay between extensor and flexor electrical ac-
tivity in the right arm (RA).
For the development of an orthosis or a FES sys-
tem was important to figure out tremor oscillation fre-
quency which was assessed through a FFT. Figure 9
represents FFT analysis of an volunteer with PD and
a control volunteer. Although signal on the time do-
main (bottom) analysis is unclear, on the frequency
domain is possible to verify the existence of a 4.55
Hz which was called oscillation frequency. The fre-
quency found is also supported by other studies pre-
sented before.
EMG signal was then squared in order to study
energy features and to establish tremor zones as pre-
sented in figure 10. One important aspect consid-
ered was how consistent signal was in time, keep-
ing through relatively long periods of 20 to 30 sec-
onds tremor pattern. To test graphically this consis-
tence, impulses were synthesized and placed based
on the period of oscillation, in this case calculated by
T
osc
=
1
f
osc
. The result can be seen in figure 11.
In this analysis was possible to identify that syn-
thesized impulses placed every T
osc
interval were con-
sistently into tremor zone, which is an indicator of
Parkinsonâ
˘
A
´
Zs Disease Tremor Suppression - A Double Approach Study - Part 1
153
Figure 9: Fast Fourier Transform Analysis wrist extensor
right arm (RA).
Figure 10: Tremor zones definition based on EMG energy.
Figure 11: EMG signal power and predictability analysis.
Figure 12: Simulated tremor EMG acquired with developed
hardware.
consistence and predictability. Based on this result the
synchronization strategy chosen was a peak detection
of EMG signal. This peak detection start all timers
in order to apply stimulus on FES case, or apply op-
posed torque for orthosis case. Figure 12 represents
a simulated tremor EMG acquired with hardware de-
veloped by our team and it presents all requisites for
the following stages of this work.
The project is now entering in final stage with clin-
ical tests in partnership with neurology department of
Federal University of S
˜
ao Paulo. In this stage the goal
is to test the prototype in real PD patients with med-
ical supervision. Fine adjust are expected to be nec-
essary due to each volunteer characteristics and sensi-
bility to electrical stimulus. Variables such as stimu-
lus frequency and applied torque to wrist joint can be
easily adapted to each patient through software cod-
ing.
It is also important to emphasize some differences
among this work and approaches studied before. Dif-
ferently from Zhang and Ang (2007) on FES, this
work is focused on wrist flexion/extension tremor in-
stead of elbow joint related tremors, and uses an of-
fline processing in order to avoid EMG stimulation ar-
tifacts. Filters for artifact suppression as proposed by
Zhang and Ang (2007) are planned for a next stage.
Some works such as Maneski et al. (2011) consider
essential tremor (ET) and PD as targeted diseases.
However, this work focus only in PD in order to keep
the boundary conditions well established, once ac-
cording to Thenganatt and Louis (2012) tremors am-
plitude may be 2.7 times higher in PD. Diferently
from all cited works, our team is also developing an
electrical stimulator and that will bring more flexibil-
ity to develop stimuation protocols.
Considering the mechanical approach for tremor
suppression some differences are also important. Ro-
con et al. (2005) in his study did not use EMG as
signal for controlling orthosis. Later Maneski et al.
(2011) discussed that only motion sensors may not be
enough to assess frequency and phase of a tremor dur-
ing tremor suppression stage. Therefore, our work in
the first stage uses EMG as a control signal, and for
modeling purposes.
In conclusion, completely customized hardware
and software were developed to bring flexibility to PD
tremor suppression protocols. A processing algorithm
was implemented to control FES and mechanical or-
thosis. The clinical tests will begin soon, then fine
adjustments may be needed for each patient and all
hypothesis related to PD dynamics can be validated.
Future work comprehends insertion of an ac-
celerometer for both approaches (orthosis and FES)
to close the control loop and possibly achieve better
results through tremor energy analysis.
ACKNOWLEDGEMENTS
Authors would like to thank University Center of FEI
for funding and project support, and Federal Univer-
sity of S
˜
ao Paulo by its support to clinical portions of
this work.
BIODEVICES 2017 - 10th International Conference on Biomedical Electronics and Devices
154
REFERENCES
Dexter, D. T. and Jenner, P. (2013). Parkinson disease: from
pathology to molecular disease mechanisms. Free
Radical Biology and Medicine, 62:132–144.
Findley, L. J. (2007). The economic impact of parkinson’s
disease. Parkinsonism & related disorders, 13:S8–
S12.
Helmich, R. C., Hallett, M., Deuschl, G., Toni, I., and
Bloem, B. R. (2012). Cerebral causes and conse-
quences of parkinsonian resting tremor: a tale of two
circuits? Brain, 135(11):3206–3226.
Hogan, N. (1984). Impedance control: An approach to ma-
nipulation. In American Control Conference, 1984,
pages 304–313. IEEE.
Hughes, A. J., Daniel, S. E., Kilford, L., and Lees, A. J.
(1992). Accuracy of clinical diagnosis of idiopathic
parkinson’s disease: a clinico-pathological study of
100 cases. Journal of Neurology, Neurosurgery & Psy-
chiatry, 55(3):181–184.
Jankovic, J. and Tolosa, E. (2006). The Parkinson’s Disease
and Movement Disorders. LWW.
Lynch, C. and Popovic, M. (2008). Functional electri-
cal stimulation. IEEE Control Systems Magazine,
28(2):40–50.
Maneski, L. P., Jorgovanovi
´
c, N., Ili
´
c, V., Do
ˇ
sen, S., Keller,
T., Popovi
´
c, M. B., and Popovi
´
c, D. B. (2011). Elec-
trical stimulation for the suppression of pathological
tremor. Medical & biological engineering & comput-
ing, 49(10):1187–1193.
Mertz, L. (2016). Taking on essential tremor: New tools and
approaches offer patients increased treatment options.
IEEE pulse, 7(3):20–25.
Oluigbo, C. O., Salma, A., and Rezai, A. R. (2012). Deep
brain stimulation for neurological disorders. IEEE re-
views in biomedical engineering, 5:88–99.
Ottobock (2016). O que
´
e uma
´
ortese?
Purves, D., Augustine, G. J., Fitzpatrick, D., Hall, W. C.,
LaMantia, A.-S., and White, L. E. (2011). Neuro-
science, Fifth Edition. Sinauer Associates, Inc.
Rahimi, F., Almeida, Q. J., Wang, D., and Janabi-Sharifi,
F. (2009). Tremor suppression orthoses for parkin-
sons patients: a frequency range perspective. In 2009
Annual International Conference of the IEEE Engi-
neering in Medicine and Biology Society, pages 1565–
1568. IEEE.
Rocon, E., Ruiz, A., Pons, J., Belda-Lois, J. M., and
S
´
anchez-Lacuesta, J. (2005). Rehabilitation robotics:
a wearable exo-skeleton for tremor assessment and
suppression. In Proceedings of the 2005 IEEE In-
ternational Conference on Robotics and Automation,
pages 2271–2276. IEEE.
Ruonala, V., Meigal, A., Rissanen, S., Airaksinen, O.,
Kankaanp
¨
a
¨
a, M., and Karjalainen, P. (2013). Emg
signal morphology in essential tremor and parkinson’s
disease. In 2013 35th Annual International Confer-
ence of the IEEE Engineering in Medicine and Biol-
ogy Society (EMBC), pages 5765–5768. IEEE.
Seki, M., Matsumoto, Y., Ando, T., Kobayashi, Y., Fujie,
M. G., Iijima, H., and Nagaoka, M. (2011). Develop-
ment of robotic upper limb orthosis with tremor sup-
pressiblity and elbow joint movability. In Systems,
Man, and Cybernetics (SMC), 2011 IEEE Interna-
tional Conference on, pages 729–735. IEEE.
Sujith, O. (2008). Functional electrical stimulation in neu-
rological disorders. European journal of neurology,
15(5):437–444.
Taheri, B., Case, D., and Richer, E. (2014). Robust con-
troller for tremor suppression at musculoskeletal level
in human wrist. IEEE Transactions on Neural Systems
and Rehabilitation Engineering, 22(2):379–388.
Thenganatt, M. A. and Louis, E. D. (2012). Distinguish-
ing essential tremor from parkinson's disease: bed-
side tests and laboratory evaluations. Expert Review
of Neurotherapeutics, 12(6):687–696.
Widjaja, F., Shee, C. Y., Poignet, P., and Ang, W. T. (2009).
Fes artifact suppression for real-time tremor compen-
sation. In 2009 IEEE International Conference on Re-
habilitation Robotics, pages 53–58. IEEE.
Zhang, D. and Ang, W. T. (2007). Reciprocal emg con-
trolled fes for pathological tremor suppression of fore-
arm. In 2007 29th Annual International Conference of
the IEEE Engineering in Medicine and Biology Soci-
ety, pages 4810–4813. IEEE.
Parkinsonâ
˘
A
´
Zs Disease Tremor Suppression - A Double Approach Study - Part 1
155