ArticlePDF Available

P25.20 Robot-aided therapy for upper limbs in patient with chronic stroke-related lesions. Brief report of a clinical experience

Authors:
14th ECCN / 4th ICTMS/DCS S177
Objectives: In the current randomized controlled trial (RCT) we used a
form of repetitive TMS, theta burst stimulation (TBS), to study whether
such stimulation could induce long-lasting modifications of LH circuits,
and improve the functional recovery and clinical outcome of sub-acute
stroke patients with neglect.
Methods: We applied real or sham continuous TBS (cTBS) over the left
PPC, for two weeks. Severity of neglect was assessed by means of the
standardized behavioural inattention test (B.I.T.). In addition, we also
measured, by means of bifocal transcranial magnetic stimulation (TMS),
how cTBS modified the excitability of the parieto-frontal functional
connections in the intact LH.
Results: We found that two weeks of cTBS were effective in improving
neglect syndrome as measured by B.I.T. Moreover, we also found that
hyper-excitability of LH parieto-frontal circuits was reduced following
treatment with real but not cTBS. All these effects were evident at the
end of the two weeks of treatment and persisted even after one month.
Conclusions: These findings suggest that a two-week course of cTBS of the
PPC of the intact hemisphere may be effective in reducing visuospatial
neglect in stroke patients.
P25.20
Robot-aided therapy for upper limbs in patient with chronic
stroke-related lesions. Brief report of a clinical experience
E. Palma1, L. Rossini1, C. Del Percio1, G. Assenza2, R. Lizio1,
F. Infarinato1, A. Pucello1, C. Cosimo Quattrocchi3, F. Vernieri2,
L. Zollo4, E. Guglielmelli4, F. Bovolenta1, P.M. Rossini2, C. Babiloni5,
M. Franceschini1, P. Sale1
1IRCCS San Raffaele Pisana, Rome, Italy, 2Dept of Clinical Neurology,
Universit`a Campus Bio-Medico, Rome, Italy, 3Dept of Radiology,
Universit`a Campus Bio-Medico, Rome, Italy, 4Center for Integrated
Research (CIR) Laboratory of Biomedical Robotics and Biomicrosystems,
Universit`a Campus Bio-Medico, Rome, Italy, 5Universit`a di Foggia,
Foggia, Italy
Introduction: A 56 years old patient, who presented chronic stroke-
related lesions, PACI right hemisphere (August 2009), left hemiparesis
with a moderate/severe upper limb impairment and moderate disability
(Barthel Index 69/100), was admitted to upper limb robotic rehabilitation
treatment.
Objective: The goal of the present study is to verify whether upper limb
motor improvement in a chronic stroke patient treated with robot-aided
rehabilitation correlates with cortical changes, and if the effects persist
for at least 2 months.
Methods: The treatment consisted of 30 sessions with the InMotion2
robotic platform, lasting 45 minutes each, 5 days a week, for a
total period of 6 weeks. The rehabilitative protocol was designed
for the improvement of movement type (i.e., the joints involved,
with a proximal-distal progression) and mode of execution (with
progression from passive to free movement). Simultaneously to the
robotic treatment, patient’s kinematic, EEG, EOG, and EMG data were
recorded. EEG data were spatially enhanced by surface Laplacian
estimation, and cortical activity was evaluated by computing ERD/ERS
events. Furthermore, corico-muscolar compuling was evaluated by
computing EEG-EMG spectral coherence and directed tranfer function
(DTF). The patient underwent assessments prior the start (T0), at the end
of the treatment (T1) and after two months from the end (T2). Several
tests were administered, such as: Fugl-Meyer (FM), Strength Evaluation
(MRC), and Ashworth scale (AS) for upper limb; Visual Analogue Scale
(VAS) for pain; Box and Block Test (BBT); Frenchay Arm test (FAT); Timed
Up and Go (TUG) test; Functional Infependence Measure (FIMTM) and
Barthel Index (BI). Additionally, the Euro-QoL questionnaire ad a VAS for
the treatment satisfaction were administered to the subject.
Results and Conclusions: Results showed a reorganization of cortical
activity and cortico-muscular coupling after 2 months of robotic
rehabilitation tasks. Kinematic data showed a distinct change in the
movement patterns (such as smoothness and trajectory approach),
tending to those measured in healthy subjects.
P25.21
A computational model of the effects of training schedules in
neurorehabilitation
L. Lonini1, L. Dipietro2, L. Zollo3, E. Guglielmelli3, H.I. Krebs2
1Frankfurt Institute for Advanced Studies Goethe University, Frankfurt
am Main, Germany, 2Newman Lab for Biomechanics and Human
Rehabilitation, Department of Mechanical Engineering, Massachusetts
Institute of Technology, Cambridge, MA, United States, 3Centre for
Integrated Research Biomedical Robotics and Biomicrosystems Lab,
Universita’ Campus Bio-Medico di Roma, Rome, Italy
Introduction: Retention of motor tasks can be affected by practice
schedule both in healthy and stroke subjects. Specifically, interleaving
tasks during training produces higher retention compared to a blocked
practice, where all the training on one task is completed before a new
task is trained. On the other hand, motor learning studies suggest that
acquisition of motor skills leads to the formation of an internal model.
Objectives: Current models do not address the issue of practice struc-
ture. This is fundamental both in neuroscience and neurorehabilitation
to understand how motor tasks are learned and consolidated. We
implemented a computational model of the human arm, based on an
internal model that is able to reproduce the effects of training schedule
on retention.
Methods: The internal model is implemented with a Locally Weighted
Projection Regression neural network that dynamically allocates local
linear models depending on the input data distribution. The network
is trained on reaching movements towards 8 targets while a velocity-
dependent force field perturbs the hand. Each target corresponds to a
different task. Three training schedules are used, in which tasks are
presented in a blocked, inter-mixed and random fashion. The amount
of motor retention is quantified with the distance between the actual
and desired trajectory.
Results: Motor retention is affected by two main factors, namely
forgetting of previous observations (the weight of previous data points
in the current update of the network parameters) and training schedule.
When forgetting is low, the 3 training schedules produce similar results.
Increasing forgetting of previous observations increases retention in inter-
mixed and random training.
Conclusions: Proper values of forgetting allow the model to reproduce
the benefits of interleaved versus blocked training. We envisage the
possibility to extend this model to design neurorehabilitation treatments
schedules for hemiparetic subjects as well as to study the mechanisms
of motor recovery from stroke.
P25.22
Increases in the excitability of spinal inhibitory pathways from
intensive locomotor training after incomplete spinal cord injury
E. Zewdie1,F.Roy
2, J. Yang3, M. Gorassini4
1Department of Biomedical Engineering, University of Alberta, Alberta,
Canada, 2Department of Surgery, University of Alberta Hospital,
Edmonton, Canada, 3Department of Physical Therapy, University of
Alberta, Edmonton, Canada, 4Department of Biomedical Engineering,
University of Alberta, Edmonton, Canada
Introduction: After incomplete Spinal Cord Injury (iSCI), the excitability
of spinal inhibitory networks is reduced, contributing to poor muscle
control (excessive co-activation) during tasks such as walking.
Objective: We examined if intensive locomotor training can improve the
excitability of spinal inhibitory networks activated by both descending
and sensory inputs.
Methods: Spinal circuits to the tibialis anterior (TA) muscle were
activated by the corticospinal tract (CST) from transcranial magnetic
stimulation (TMS). These spinal circuits were then conditioned by low-
intensity stimulation of the common peroneal nerve (CPN: interstimulus
interval of 40, 50 and 60 ms prior to TMS) to activate spinal inhibitory
pathways. The resulting suppression of the evoked motor potential (MEP)
was considered to be a measure of spinal inhibition given that a similar
suppression of evoked responses occurred from direct stimulation of
the CST. In addition, the magnitude of cutaneomuscular reflexes in the
TA evoked from tibial nerve stimulation at the ankle, which has been
shown to activate spinal inhibitory circuits, was also examined before
and after training. Subjects received 2 months of endurance (treadmill)
and precision (walking over obstacles) training in random order.
Results: Training increased the amount of MEP suppression from CPN
stimulation from 15 to 30% MEP inhibition (n = 4 subjects). Skill training
... Physical therapy is still among the most effective techniques to deal with disability problems. Various works have been focused on the upper limb rehabilitation [9], [3], [12], [13], [14], [11], [6], [15]. Further researches have been specialized only in the rehabilitation of the elbow [18], [17], [19], [25], [20], [21], [26], [27] , [28], [29], [30] or hand [5], [1], [7], [2], [8]. ...
... The Kinect sensor has shown great efficiency and great accuracy in measuring of the human joints. Also in relation to previous works, including the most recent [4], [22], [1], [3], [6], [15], [12], [13], [14], [9], we present in this work a wireless remote controlled rehabilitation robot. We can summarize the major contributions of this work in the following points: ...
Conference Paper
In this paper, the design of a new upper limb rehabilitation system is presented. The developed system is an exoskeleton with two degrees of freedom that can be used for diagnosis, physical therapy, and outcome evaluation. The system is dedicated to the patients with paraplegia of their upper extremities due to stroke or any disorders of the central or peripheral nervous system. The designed robot-aided therapy actuates both movements: flexion/extension for the elbow and pronation/supination for the forearm. The robotic system used Kinects skeletal tracking for an upper limb rehabilitation. Position control is communicated via a wireless network. Indeed a ZigBee protocol using xbee communication modules has been installed to ensure remote control of rehabilitation exercise. A kinematic model has been developed based on Denavit-Hartenberg approach to make first tests. A sliding mode robust law control has been implemented. The Lyapunov-based approach has been used to establish the system asymptotic stability. Experimental results are provided to demonstrate performances of the developed robot of upper limb remote rehabilitation.
ResearchGate has not been able to resolve any references for this publication.