U.S. patent application number 11/598936 was filed with the patent office on 2007-03-22 for system and methods to overcome gravity-induced dysfunction in extremity paresis.
Invention is credited to Julius P.A. Dewald, Wilhelmus J. Lam.
Application Number | 20070066918 11/598936 |
Document ID | / |
Family ID | 36142883 |
Filed Date | 2007-03-22 |
United States Patent
Application |
20070066918 |
Kind Code |
A1 |
Dewald; Julius P.A. ; et
al. |
March 22, 2007 |
System and methods to overcome gravity-induced dysfunction in
extremity paresis
Abstract
The present invention relates to a system for use in
rehabilitation and/or physical therapy for the treatment of injury
or disease. The system can overcome gravity-induced dysfunction in
extremity paresis following stroke or other neurological
disorders.
Inventors: |
Dewald; Julius P.A.;
(Downers Grove, IL) ; Lam; Wilhelmus J.; (Orchard
Park, NY) |
Correspondence
Address: |
BELL & ASSOCIATES
416 FUNSTON ST., SUITE 100
SAN FRANCISCO
CA
94118
US
|
Family ID: |
36142883 |
Appl. No.: |
11/598936 |
Filed: |
November 13, 2006 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11239709 |
Sep 29, 2005 |
|
|
|
11598936 |
Nov 13, 2006 |
|
|
|
60614928 |
Sep 29, 2004 |
|
|
|
Current U.S.
Class: |
601/5 ; 482/901;
601/23; 601/33 |
Current CPC
Class: |
A63B 23/0355 20130101;
A61H 2201/5097 20130101; A63B 21/4019 20151001; A63B 2024/009
20130101; A63B 2220/51 20130101; A61H 1/0229 20130101; A61H 1/0266
20130101; A61H 2201/1638 20130101; A61H 2230/08 20130101; A63B
23/0482 20130101; A61H 1/0237 20130101; A61H 1/0274 20130101; A61H
2203/0431 20130101; A63B 21/00181 20130101; A63B 21/0058 20130101;
A63B 23/14 20130101; A63B 23/1209 20130101; A63B 22/0235 20130101;
A61H 1/024 20130101; A61H 1/0296 20130101; A63B 2024/0096 20130101;
A63B 21/4035 20151001; A61H 2201/1659 20130101; A63B 21/4047
20151001; A63B 21/00178 20130101; A61H 2201/1238 20130101; A63B
21/0023 20130101; A63B 2208/0233 20130101; A61H 1/02 20130101; A61H
1/0285 20130101; A61H 1/0281 20130101; A61H 2201/5069 20130101;
A63B 23/03508 20130101; A63B 23/08 20130101; A63B 21/008 20130101;
A63B 23/0494 20130101; A63B 23/16 20130101; A63B 2071/0638
20130101; A61H 2201/5058 20130101; A61H 2201/5061 20130101; A61H
2201/5064 20130101; A61H 2201/1215 20130101; A61H 1/0244 20130101;
A63B 2213/004 20130101; A63B 2071/0655 20130101; A61N 1/36003
20130101; A61H 2201/5007 20130101; A63B 2208/0204 20130101; A61H
1/0277 20130101; A63B 2022/0094 20130101; A63B 2220/16 20130101;
A63B 23/025 20130101; A61H 2201/10 20130101; A63B 2071/0018
20130101; A63B 2225/50 20130101 |
Class at
Publication: |
601/005 ;
601/023; 601/033; 482/901 |
International
Class: |
A61H 1/02 20060101
A61H001/02 |
Goverment Interests
[0002] The United States government has certain rights to this
invention pursuant to Grant No. H133G030143 from the National
Institutes of Disability and Rehabilitation Research to
Northwestern University.
Claims
1. A system for measuring, treating, and self-rehabilitating an
individual having a neurological condition, the system comprising
mechanical means, at least one computer, display means, and
interconnecting means, the mechanical means further comprising an
interacting instrumented member that interacts with the body or a
part thereof of the individual, a force sensor, a force generator,
at least one moveable non-compliant linkage, and a base, the force
sensor further being attachedly connected by the linkage to the
force generator, the linkage having at least three degrees of
freedom, the computer further comprising an interactive software
program, and the base supporting at least one of the above.
2. The system of claim 1 wherein the interconnecting means provide
radio communicating signals, electrical communicating signals,
photonic communicating signals, or a combination thereof, between
the mechanical means, the computer, and the display means.
4. The system of claim 1 wherein the force generator is an actuator
selected from the group consisting of a rotary hydraulic motor, a
linear hydraulic motor, a pneumatic motor, and an electric
motor.
5. The system of claim 1 wherein the system further comprises at
least one position measurement device, the position measurement
device being placed on a predetermined position selected from the
group consisting of an end effector, a linkage, a force sensor, a
force generator, a shoulder, a hip, a neck, and a head.
6. The system of claim 1 wherein the generator generates a force
that compensates for the force due to gravity on the body or part
thereof and wherein the generated force is equivalent in magnitude
to between about -1 times and about +4 times the force of gravity
upon the body or part thereof.
7. The system of claim 6 wherein the generated force is essentially
equivalent to a force required for manipulating joint abduction
torques of the individual, the joint selected from the group
consisting of the shoulder and the hip.
8. The system of claim 1 wherein the interacting instrumented
member further comprises a sensor selected from the group
consisting of a force sensor, a position sensor, and a motion
sensor.
9. The system of claim 1 wherein the interacting instrumented
member further comprises an electrical stimulator, the electrical
stimulator being further releasably connected to an extremity of
the body or part thereof.
10. The system of claim 9 wherein the electrical stimulator
stimulates movement in the extremity of the body, the extremity
being selected from the group consisting of a finger, a thumb, a
hand, an elbow, a shoulder, a wrist, a toe, a foot, an ankle, a
knee, and a hip.
11. The system of claim 9 wherein the interacting instrumented
member comprises a member selected from the group consisting of a
splint, a limb support, a hand support, a foot support, and a
force-sensing treadmill.
12. The system of claim 10 wherein the stimulated movement results
in a propriosensory effect in the individual.
13. The system of claim 10 wherein the stimulated movement results
in a dermal tactile sensory effect in the individual.
14. The system of claim 10 wherein the stimulated movement results
in a muscle sensory effect in the individual.
15. The system of claim 1 wherein the neurological condition is
selected from the group consisting of hemiparetic stroke, cerebral
palsy, head trauma, and multiple sclerosis.
16. The system of claim 1 wherein the neurological condition
results in a loss of independent joint control in the body or part
thereof.
17. The system of claim 1 wherein the system further comprises an
end effector articulatedly attached between an appendage attaching
member and the force generator.
18. The system of claim 1 wherein the computer further comprises
memory means for storing the force input data, the virtual
environment, a position data, and the force output data.
19. A device for measuring, treating, and self-rehabilitating an
individual having a neurological condition, the device comprising
mechanical means and interconnecting means, the mechanical means
further comprising an interacting instrumented member that
interacts with the body or a part thereof of the individual, a
force sensor, a force generator, at least one moveable
non-compliant linkage, and a base, the force sensor further being
attachedly connected by the linkage to the force generator, the
linkage having at least three degrees of freedom, and the base
supporting at least one of the above.
20. The device of claim 19 wherein the interconnecting means
provide radio communicating signals, electrical communicating
signals, photonic communicating signals, or a combination
thereof.
21. The device of claim 19 wherein the force generator is an
actuator selected from the group consisting of a rotary hydraulic
motor, a linear hydraulic motor, a pneumatic motor, and an electric
motor.
22. The device of claim 19 wherein the device further comprises at
least one position measurement device, the position measurement
device being placed on a predetermined position selected from the
group consisting of an end effector, a linkage, a force sensor, a
force generator, a shoulder, a hip, a neck, and a head.
23. The device of claim 19 wherein the generator generates a force
that compensates for the force due to gravity on the body or part
thereof and wherein the generated force is equivalent in magnitude
to between about -1 times and about +4 times the force of gravity
upon the body or part thereof.
24. The device of claim 23 wherein the generated force is
essentially equivalent to a force required for manipulating joint
abduction torques of the individual, the joint selected from the
group consisting of the shoulder and the hip.
25. The device of claim 19 wherein the interacting instrumented
member further comprises a sensor selected from the group
consisting of a force sensor, a position sensor, and a motion
sensor.
26. The device of claim 19 wherein the interacting instrumented
member further comprises an electrical stimulator, the electrical
stimulator being further releasably connected to an extremity of
the body or part thereof.
27. The device of claim 26 wherein the electrical stimulator
stimulates movement in the extremity of the body, the extremity
being selected from the group consisting of a finger, a thumb, a
hand, an elbow, a shoulder, a wrist, a toe, a foot, an ankle, a
knee, and a hip.
28. The device of claim 26 wherein the interacting instrumented
member comprises a member selected from the group consisting of a
splint, a limb support, a hand support, a foot support, and a
force-sensing treadmill.
29. The device of claim 27 wherein the stimulated movement results
in a propriosensory effect in the individual.
30. The device of claim 27 wherein the stimulated movement results
in a dermal tactile sensory effect in the individual.
31. The device of claim 29 wherein the stimulated movement results
in a muscle sensory effect in the individual.
32. The device of claim 19 wherein the neurological condition is
selected from the group consisting of hemiparetic stroke, cerebral
palsy, head trauma, and multiple sclerosis.
33. The device of claim 19 wherein the neurological condition
results in a loss of independent joint control in the body or part
thereof.
34. The device of claim 19 wherein the system further comprises an
end effector articulatedly attached between an appendage attaching
member and the force generator.
Description
[0001] The present application is a Divisional Patent Application
of U.S. Non-provisional patent application Ser. No. 11/239,709
entitled "System and Methods to Overcome Gravity-Induced
Dysfunction in Extremity Paresis", filed Sep. 29, 2005, which
claims priority to U.S. Provisional Patent Application Ser. No.
60/614,928 entitled "Devices and Methods to Overcome
Gravity-Induced Dysfunction in Upper Extremity Paresis", filed Sep.
29, 2004, which are herein incorporated by reference in their
entirety for all purposes.
FIELD OF THE INVENTION
[0003] The invention relates to the field of rehabilitation and/or
physical therapy for the treatment of injury and/or disease using a
haptic system that is used to train and/or assist an individual
having a neurological condition. In one aspect the invention
provides an assistive robotic device in combination with a
3-dimensional virtual reality workspace. More specifically, the
invention relates to a system and method to overcome
gravity-induced dysfunction in upper extremity paresis.
BACKGROUND OF THE INVENTION
[0004] Disturbances in movement coordination are the least well
understood but often the most debilitating with respect to
functional recovery following brain injury. These deficits in
coordination are expressed in the form of abnormal muscle synergies
and result in limited and stereotypic movement patterns that are
functionally disabling. The result of these constraints in muscle
synergies is an abnormal coupling between shoulder abduction and
elbow flexion in the arm, which significantly reduces a stroke
survivor's reaching space when he/she lifts up the weight of the
impaired arm against gravity. Current neurotherapeutic approaches
to mitigate these abnormal synergies have produced, at best,
limited functional recovery.
[0005] In the leg the expression of abnormal synergies results in
coupling between hip/knee extension with hip adduction. The result
of this is a reduced ability of activating hip abductor muscles in
the impaired leg during stance.
Disturbances of Voluntary Movement in Hemiparetic Stroke
[0006] A detailed qualitative description of the abnormal movement
patterns in the impaired limb, and the natural history of the
evolution of the various components of these abnormal clinical
signs was first provided by Twitchell in 1951, in which he
delineated both the major features of the movement disturbance, and
the time course of recovery from stroke (Twitchell (1951) Brain 74:
443-480). A prominent feature of the disturbed movement patterns
was the emergence of "stereotypic" movements, in which there
appeared to be a relatively tight coupling of motion at adjacent
joints in the upper and lower limbs.
[0007] Brunnstrom subsequently classified these abnormal
stereotypic movement patterns into so called "synergies" which were
broadly of either flexor or extensor type (Brunnstrom (1970) In:
"Movement therapy in hemiplegia: a neurophysiological approach"
Harper & Row, Publishers Inc., Hagerstown Md.). This
qualitative classification of abnormal synergies, summarized in
Table 1, has received limited modification or study by other
investigators. TABLE-US-00001 TABLE 1 Upper Limb synergies in
hemiparetic stroke after Brunnstrom (1970, supra) Extension synergy
Arm Shoulder Girdle protraction adduction internal rotation Elbow
extension pronation Leg Hip extension adduction internal rotation
Knee extension Flexion synergy Arm Shoulder Girdle Retraction
abduction to 90.degree. external rotation Elbow Flexion Supination
Leg Hip extension adduction internal rotation Knee extension
[0008] Recent treatment approaches for hemiparetic upper extremity
such as "motor relearning program", electromyographic
(EMG)-triggered/functional electrical stimulation, repeated mental
practice, constraint-induced movement therapy, robot-aided
sensory-motor training and bilateral arm training focus on
task-specific repetition, increased intensity, and/or exercise in a
real-world context. (See, for example, Langhammer and Stanghelle
(2000) Clin. Rehabil. 14: 361-369; Cauraugh et al. (2000) Stroke
31: 1360-1364; Page et al. (2001) Phys. Ther. 81: 1455-1462;
Miltner et al. (1999) Stroke 30: 586-592; van der Lee et al. (1999)
Stroke 30: 2369-2375; Volpe et al. (2000) Neurology 54: 1938-1944;
Volpe et al. (1999) Neurology 53: 1874-1876; Whitall et al. (2000)
Stroke 31: 2390-2395; and Richards and Pohl (1999) Clin Geriatr Med
15: 819-832; Woldag and Hummelsheim (2002) J. Neurol. 249: 518-528)
Despite favorable mounting evidence for the newer treatment models,
none of the current neurorehabilitation techniques directly address
the presence of abnormal synergistic patterns that constrain
functional reaching (Dewald et al. (2001) Topics in Stroke
Rehabilitation 8: 1-11). Interventions that target abnormal
synergistic movement patterns may ameliorate functional reaching
and greatly benefit individuals with chronic stroke-induced
movement discoordination.
[0009] In the lower limb recent findings from basic science provide
preliminary evidence that functional locomotor recovery is possible
after stroke or spinal cord injury when intense and accurate
afferent input is provided in a task-specific and repetitive
manner. Treadmill training is an example of a therapeutic modality
that is derived from studies of adult cats with a low thoracic
spinal transection who recovered the ability to step on a moving
treadmill belt after they were trained on the treadmill and
provided with truncal support, stimulation to recover extensor
activity, and assistance in paw placement (Barbeau and Rossignol
(1987) Brain Res. 412: 84-95; de Leon et al. (1998a) J.
Neurophysiol. 80: 83-91; de Leon et al. (1998b) J. Neurophysiol.
79: 1329-1340; Lovely, et al., (1986) Exp. Neurol. 92: 421-435).
Investigators have found that the spinal locomotor pools, which
include a central pattern generator for automatic, alternating
flexor and extensor leg muscle activity, are highly responsive to
phasic segmental sensory inputs associated with walking and
demonstrate evidence of learning during step training (Edgerton et
al. (1997a) Adv Neurol. 72: 233-247; Edgerton et al. (1997b).
Repetitive practice of the task was essential to the learning.
[0010] Barbeau and colleagues were the first investigators to
translate this paradigm to human application for re-training
walking after spinal cord injury and stroke (Barbeau et al., (1987)
Brain Res. 437: 83-96; Finch et al. (1991) Phys. Ther. 71: 842-855;
Visintin and Barbeau (1989) Can. J. Neurol. Sci. 16: 315-325;
Visintin and Barbeau (1994) Paraplegia 32: 540-553). In their
initial work, Barbeau et al. (Barbeau et al., (1987) supra)
suspended the consumer over a treadmill using an overhead lift for
body-weight support and clinician-provided assistance to the
legs.
[0011] Task-specific training appears to be critical to the success
of a locomotor training intervention post-stroke (Richards et al.
(1993) Arch. Phys. Med. Rehabil. 74: 612-620). Treadmill training
is a method of locomotor training that closely simulates the
sensory elements specific to walking such as load on the lower
extremities, upright trunk posture, proper lower limb kinematics,
and normal walking speeds to generate effective lower limb stepping
(Edgerton et al. (1997) supra; Behrman and Harkema (2000) Phys.
Ther. 80: 688-700).
[0012] Within the past 10 years, there have been many studies that
have specifically investigated the effects of treadmill training
with or without body weight support (BWS) on post-stroke locomotor
recovery. Treadmill training (with or without BWS) appears to be
more effective than conventional therapy alone in locomotor
recovery after stroke (Richards et al. (1993) supra; Hesse et al.
(1995a) Stroke 26: 976-981; Hesse et al. (1995b); Laufer et al.
(2001) J. Rehabil. Res. Dev. 38: 69-78; Pohl (2002) Stroke 33:
553-558; Sullivan et al. (2002) Arch. Phys. Med. Rehabil. 83:
683-691). While there is building evidence that this therapeutic
modality may be beneficial in improving locomotor ability after
stroke, there is little agreement or systematic study of the
optimal training parameters to maximize functional outcomes
(Tuszynski, Edgerton, and Dobkin (1999) J. Spinal Cord Med. 22:
143). None of the current studies have incorporated abnormal muscle
coactivation patterns and associated joint toques in the lower
extremity. We have quantitative evidence that abnormal coupling
between hip and knee extension and hip adduction exists.
Furthermore, we have preliminary data that this abnormal coupling
reduces the ability to generate hip abduction while standing on the
paretic leg. This results then in the inability to keep the pelvis
horizontal and could result in the stroke subject falling towards
the affected side. As in the case of the arm, interventions that
target abnormal synergistic movement patterns may ameliorate
balance and greatly benefit individuals with chronic stroke-induced
movement discoordination.
[0013] Implementation of current treatment philosophies is more
dependent on the therapist's background and training rather than
clear clinical indications or objective and quantitative measures.
Furthermore, there is no consensus in the literature to support one
approach over the other or even a gold standard objective measure
of their effectiveness in increasing functional recovery. Heinemann
et al. reported on the relationship between functional status at
discharge and intensities of therapies received during the
patient's in-patient medical rehabilitation (Heinemann et al.
(1995) Am. J. Phys. Med. Rehabil. 74: 315-326). The results for a
group of 140 patients with traumatic brain injury (TBI) identified
no significant correlation between functional outcome and the
intensity of therapies. The apparent lack of benefit related to
intensity of therapies may be due to such factors as spontaneous
recovery, lack of adequate level of intensity based on the stroke
patient's absolute tolerance, and most importantly, to inadequate
measurement tools, which are subjective and non-quantitative, and
do not possess the discrimination power required to detect
meaningful functional change. Furthermore, most current approaches
may not be effective in promoting the use of more functional
elbow-shoulder torque combinations because of the implementation of
limited, poorly controlled exercise sequences. None of the current
neurorehabilitation techniques encourage movements outside abnormal
synergic patterns in a rigorous and quantifiable way.
Evidence for Motor Learning and Strength Training Capabilities
Following Stroke
[0014] We have evidence from previous work that, depending on the
lesion location, hemiparetic stroke subjects are able to adapt to
novel force perturbations applied to their impaired arms during
reaching and retrieval movements (see Krebs et al. (1996) 18th
Annual Conference of IEEE-EMBS; Raasch et al. (1997) Society for
Neuroscience Abstracts 23). These findings demonstrate that a
considerable level of motor learning capability persists in
relation to the impaired arm. We also have evidence that chronic
stroke subjects are able to use the residual motor learning
capability to partially regain functional elbow/shoulder torque
combinations (for example, shoulder abduction/external rotation
combined with elbow extension) during an eight-week training
protocol (see Ellis et al. (2002a) Program No. 169.2 Abstract
Viewer/Itinerary Planner. Washington, D.C.: Society for
Neuroscience Abstracts; Ellis et al. (2002b) Neurology Report 26:
191, Abstract; and
[0015] Ellis et al. (2003) Program No. 714 Abstract
Viewer/Itinerary Planner. Washington, D.C.: Society for
Neuroscience Abstracts).
The Use of Robotics in Stroke-Rehabilitation
[0016] At present, very little technology exists to support the
recovery phase of stroke rehabilitation. However, there has been a
surge of academic research on this topic in recent years (see, for
example, Proceedings of the ICORR International Conference on
Rehabilitation Robotics, 2001 and 2005). Of the academic research
in progress, most research centers have elected to attempt to adapt
or re-configure industrial robots for use in this application (Lum
et al. (1995) Arch. Phys. Med. Rehabil. 83: 952-959). While this
appears to be a reasonable approach it suffers from a critical
drawback: twenty years of experience with industrial robots has
shown that low impedance comparable to the human arm cannot be
achieved with these machines. Because of their electro-mechanical
design and control architecture, commercial robots are
intrinsically position-controlled machines that do not yield easily
under the action of external forces. Active force feedback can be
used to enhance robot responsiveness but it is not sufficient to
produce the "back-drivability" (low mechanical impedance) required
to move smoothly and rapidly in compliance with a patient's actions
(Lawrence (1988) Proc. IEEE Int. Conf. Robotics & Automation
1185-1191).
[0017] In contrast to commercial robotic technology, the MIT-MANUS
robotic device was specifically designed for clinical neurological
applications (Hogan et al. (1995) J. Interactive Robotic
Therapist). The MIT-MANUS robotic device is configured for safe,
stable and compliant operation in close physical contact with
humans. Its computer impedance control (synonymous with position
control) system modulates the way the robot reacts to mechanical
perturbation from a patient or clinician and ensures a gentle
compliant behavior (technically, a low and controllable impedance)
(Hogan (1985) ASME J. Dynamic Systems Measurement and Control 107:
1-24). Operationally, a low impedance means that the robot can "get
out of the way" as needed. However, due to the impedance control
system, there is a moderate level of resistance due to inertia that
the user must overcome to produce movement. This attribute limits
the applicability of the MIT-MANUS to subjects who are able to
exert enough force to overcome the inertial resistance of the
device.
[0018] To test the feasibility of robot-aided neuro-rehabilitation,
MIT investigators have used the MIT-MANUS robotic device in pilot
studies on a daily basis for over seven years with CVA (cerebral
vasculary incidence resulting in a stroke), Parkinson's disease,
multiple sclerosis, spinal cord injury, amyotrophic lateral
sclerosis (ALS), and Guillain-Barre (GB) patients at the Burke
Rehabilitation Hospital. The key research objective in these pilot
studies was to validate the concept of robot-aided exercise therapy
and assess whether: (a) robot-aided therapy had adverse effects,
(b) patients would tolerate the procedure, and (c) manipulation of
the impaired limb influenced motor recovery. The results in these
pilot clinical trials with 96 stroke patients showed that
robot-aided neuro-rehabilitation did not impede recovery or
exacerbate joint or tendon pain, and no adverse events occurred in
an estimated 2000 hours of operation involving close contact with
patients. A questionnaire administered during the bi-weekly
standard assessment by the therapists showed that robot-assisted
therapy was well accepted and tolerated by the patients. Most
important, results indicated that patients in the experimental
group improved further and faster, outranking the control group in
the clinical assessments of the motor impairment involving shoulder
and elbow. (See, for example, Aisen et al. (1997) Arch. Neurol. 54:
443-446; Krebs et al. (1998) IEEE Transact. Rehab. Engineer. 6:
75-87; Krebs et al. (2000) VA J. Rehab. Res. Dev. 37: 639-652;
Volpe et al. (2001) Curr. Opin. Neurol. 14: 745-752; Volpe et al.
(2000) Neurology 54: 1938-1944, and Ibid. (1999) Neurology 53:
1874-1876). However, a shortcoming of the MIT-MANUS obviating its
usefulness for application in the evaluation and rehabilitation of
gravity-induced discoordination is that it only works in a
horizontal plane unable to provide various levels of limb support
or operate in all directions of movement. In addition, impedance
control technology must use a very light structure which may
contain mechanical shortcomings such as friction or mechanical
compliance. Even though forces may be measured at the patient
interface, no compensation can be made for such non-linearities
since they occur between the force control device, the motor, and
the patient introducing errors that cannot be compensated.
[0019] Several similar US Patents have been issued to the above
technology. For example, U.S. Pat. No. 5,466,213 to Hogan et al. is
directed to an interactive robotic therapist that guides a
patient's limb along a desired path through a desired series of
exercises. The robotic therapist incorporates sensors that provide
position, velocity, and force information at the patient's hand.
The reference, however, does not teach using-force and position
information in both real and virtual environments to measure,
treat, or self-rehabilitate impaired movement performance.
[0020] U.S. Pat. No. 5,421,798 to Bond et al. is directed to an
apparatus for evaluation of a limb of a test subject. The distal
end of the limb is secured to the apparatus. The test subject moves
the limb along a linear track. At least two components of the
forces generated by the limb against the track are sensed. The
force components are used to calculate the forces applied at each
limb joint contributing to movement. The reference, however, does
not teach using force and position information in both real and
virtual environments to measure, treat, or self-rehabilitate
impaired movement performance.
[0021] U.S. Pat. No. 5,830,160 to Reinkensmeyer is directed to a
movement guiding system for quantifying, diagnosing, and treating
impaired movement performance. The guiding system guides movement
of a limb along a linear path and can quantify movement performance
by measuring constraint forces generated during the movement. The
reference does not teach force and position information in both
real and virtual environments to measure, treat, or
self-rehabilitate impaired movement performance.
[0022] U.S. Pat. No. 6,413,190 to Wood and Koval is directed to a
rehabilitation apparatus and method that monitors patient
rehabilitation therapy activity, the apparatus detecting sequential
muscle contractions thereby operating a computer game that reflects
the movement upon a screen. The patient is therefore encouraged to
ensure that two muscles move in a temporal sequence to "play" the
game. The reference does not teach using force and position
information in both real and virtual environments to measure,
treat, or self-rehabilitate impaired movement performance.
[0023] U.S. Pat. No. 4,936,299 to Erlandson discloses a
rehabilitation apparatus having a robotic arm controlled by
application software and a control board of a CPU. The patent also
discloses a viewing screen and that the rehabilitation is initiated
and under direction of a therapist. The reference does not teach
using force and position information in both real and virtual
environments to measure, treat, or self-rehabilitate impaired
movement performance.
[0024] U.S. Pat. No. 6,613,000 to Reinkensmeyer discloses a system
providing arm movement therapy for patients with sensory motor
impairments having a joystick controlled by application software of
a CPU over the World Wide Web using client-side applets. The patent
also discloses a viewing screen and that the rehabilitation is
performed without the direction or supervision of a therapist but
in response to a predetermined desired therapeutic exercise. The
reference does not teach using force and position information in
both real and virtual environments to measure, treat, or
self-rehabilitate impaired movement performance.
[0025] Admittance control technology uses a force measurement
device (loadcell) placed at the patient interface. The loadcell
functions as the force feedback device in a closed loop force
control system. Therefore the forces are always controlled at the
patient's interface, and system non-linearities mentioned before
are minimized because they occur inside the force control loop and
can therefore be compensated to a large degree.
[0026] A commercial robot that uses admittance control is the
HAPTICMASTER (HM) from FCS Control Systems. The FCS Control
Systems' control technology originated and was patented in the late
1970s in the field of flight training and simulation to generate
aircraft control forces for the pilot. It has matured over the
years from a patented to a company proprietary technology. See U.S.
Pat. No. 4,398,889, herein incorporated by reference in its
entirety.
[0027] The HAPTICMASTER, which was designed with rehabilitation
applications in mind, has low inertia such that the user doesn't
feel much resistance when attempting to move the device. The low
level inertial properties of the HM enable application to
individuals with all levels of impairment severity including
individuals with severe impairment who would otherwise be unable to
move against the inertial resistance of other robotic devices. Work
done at the University of Reading, England has shown that the robot
is safe and can assist reaching movements to various targets in the
workspace of the paretic arm following stroke (Coote and Stokes
(2003) Technol. Disabil. 15: 27-34; Harwin and Hillman (2003)
Robotica 21; Marin.ae butted.cek et al (2001) Association for the
Advancement of Assistive Technology in Europe AAATE '01. Amsterdam;
Washington, D.C.: IOS Press,). However, it has been employed to
assist subjects in reaching movements with the upper extremity
constantly supported by an external device.
[0028] Each of the robotic devices described above demonstrate the
ability to use robotics as a device for implementing therapeutic
training post-stroke. In addition, each of these devices is capable
of measuring motion and tracking progress during training. With
this current patent, we propose to generate virtual
mechanical/visual environments that can simulate weightlessness or
make the body or limb progressively heavier to beyond its actual
weight. Using these realistic simulated environments generated by a
combination of multi degree of freedom robotics and visual feedback
we can measure the effect of abnormal joint torque coupling in the
upper and lower extremities as well as train individuals to slowly
relearn to deal with the weight of their limb or body while
reaching (arm) or walking (leg).
Virtual Reality
[0029] Haptics is the science of applying tactile or force
sensation to human interaction with computers. A haptic device is
one that involves physical contact between the computer and the
user, usually through an input/output device, such as a joystick or
data gloves, that senses the body's movements. By using haptic
devices, the user can not only feed information to the computer but
can receive information from the computer in the form of a felt
sensation on some part of the body. This is referred to as a haptic
interface. For example, in a virtual reality environment, a user
can pick up a virtual tennis ball using a data glove. The computer
senses the movement and moves the virtual ball on the display.
However, because of the nature of a haptic interface, the user will
feel the tennis ball in his hand through tactile sensations that
the computer sends through the data glove, mimicking the feel of
the tennis ball in the user's hand. Typical uses a haptic interface
are disclosed in U.S. Pat. No. 6,636,161 (Rosenberg, issued Oct.
21, 2003) and U.S. Pat. No. 6,697,043 (Shahoian, issued Feb. 24,
2004).
BRIEF DESCRIPTION OF THE INVENTION
[0030] The invention provides a system and methods to measure, to
train and/or to assist and/or to rehabilitate and/or
self-rehabilitate an individual having a neurological condition
that results in loss in the ability to coactivate certain muscle
combinations in affected limb or similar extremety of the
individual.
[0031] In one embodiment the invention provides a method for
measuring, treating, and self-rehabilitating an individual having a
neurological condition, the method comprising: (i) providing a
system comprising mechanical means, at least one computer, display
means, and interconnecting means, the mechanical means further
comprising an interacting instrumented member that interacts with
the body or a part thereof of the individual, a force sensor, a
force generator, at least one moveable non-compliant linkage, and a
base, the force sensor further being attachedly connected by the
linkage to the force generator, the linkage having at least three
degrees of freedom, the computer further comprising an interactive
software program, and the base supporting at least one of the
above; (ii) securing a body or part thereof of the individual to
the interacting instrumented member; (iii) permitting the
individual to move the body or part thereof to a desired position;
(iv) sensing the force required to move the body or part thereof
using the force sensor; (v) producing force input data using the
sensed force; (vi) transmitting the force input data from the force
sensor to the computer; (vii) processing the force input data using
the interactive software program; (viii) transmitting the processed
data to the display means whereby the display shows a virtual
environment; (ix) processing the data to produce force output data;
(x) transmitting the force output data to the force generator and
the linkage thereby generating a force upon the interacting
instrumented member and the body or part thereof, the resulting
generated force upon the body or part thereof causing the muscles
and nerves in the body or part thereof to be stimulated, the
stimulation resulting in regaining muscle coactivation patterns and
associated joint torques patterns for the individual; thereby
measuring, treating, and self-rehabilitating the individual. In a
preferred embodiment the neurological condition is selected from
the group consisting of hemiparetic stroke, cerebral palsy, head
trauma, and multiple sclerosis. In a more preferred embodiment the
neurological condition results in a loss of independent joint
control in the body or part thereof.
[0032] In one embodiment the body or part thereof is selected from
the group consisting of a whole body, a trunk, a shoulder, a neck,
a head, an arm, an elbow, a wrist, a hand, a hip, a leg, a knee, an
ankle, and a foot.
[0033] In another embodiment the interconnecting means provide
radio communicating signals, electrical communicating signals,
photonic communicating signals, or a combination thereof, between
the mechanical means, the computing means, and the display
means.
[0034] In another embodiment the force generator is an actuator
selected from the group consisting of a rotary hydraulic motor, a
linear hydraulic motor, a pneumatic motor, and an electric
motor.
[0035] In yet another embodiment the method further comprises the
step of attaching at least one position measurement device, the
position measurement device being placed on a predetermined
position selected from the group consisting of an end effector, a
linkage, a force sensor, a force generator, a shoulder, a hip, a
neck, and a head.
[0036] In a still further embodiment the generated force
compensates for the force due to gravity on the body or part
thereof and wherein the generated force is equivalent in magnitude
to between about -1 times and about +4 times the force of gravity
upon the body or part thereof.
[0037] In another embodiment the generated force is essentially
equivalent to a force required for manipulating joint abduction
torques of the individual, the joint selected from the group
consisting of the shoulder and the hip.
[0038] In another embodiment the interacting instrumented member
further comprises a sensor selected from the group consisting of a
force sensor, a position sensor, and a motion sensor.
[0039] In an alternative embodiment the system further comprises an
end effector articulatedly attached between the appendage attaching
member and the force generator.
[0040] In another alternative embodiment the method further
comprises a step of determining the position of the appendage
attaching member to generate position data and providing the
position data to the computer and the display.
[0041] In still another alternative embodiment the method further
comprises the computer further comprising memory means for storing
the force input data, the virtual environment, the position data,
and the force output data.
[0042] In another embodiment the interacting instrumented member
further comprises an electrical stimulator, the electrical
stimulator being further releasably connected to an extremity of
the body or part thereof. In a preferred embodiment the electrical
stimulator stimulates movement in the extremity of the appendage,
the extremity being selected from the group consisting of a finger,
a thumb, a hand, an elbow, a shoulder, a wrist, a toe, a foot, an
ankle, a knee, and a hip. In a more preferred embodiment the
stimulated movement results in a propriosensory effect in the
individual. In a most preferred embodiment the stimulated movement
results in a dermal tactile sensory effect or muscle sensory effect
in the individual. In another alternative embodiment the
interacting instrumented member comprises a member selected from
the group consisting of a splint, a limb support, a hand support, a
foot support, and a force-sensing treadmill.
[0043] In another embodiment, the invention provides a system for
providing at least one force in at least one plane to a limb or
extremety of an individual having a neurological condition, the
force resulting in negating the force acting upon the limb or
extremety due to gravity and allowing the individual to move the
limb or extremety in a desired direction, the system comprising
means for supporting the limb or extremety, a device for detecting
the force of gravity acting upon the limb or extremety, and a
device for negating the force of gravity acting upon the limb or
extremety. In one embodiment, the system is a haptic system. In
another embodiment, the force further results in allowing the
individual to move the limb or extremety to a target site. In a
preferred embodiment, the force is provided in at least two planes.
In a more preferred embodiment, the force is provided in at least
three planes. In another preferred embodiment, the force is
provided in a plurality of planes.
[0044] In another embodiment, the system comprises a device for
negating the force of gravity acting upon the limb or extremety
having at least one degree of freedom. In another embodiment the
device for negating the force of gravity acting upon the limb or
extremety has at least one degree of freedom. In preferred
embodiment the device for negating the force of gravity acting upon
the limb or extremety has at least two degrees of freedom. In a
more preferred embodiment, the device for negating the force of
gravity acting upon the limb or extremety has at least three
degrees of freedom. In a most preferred embodiment, the device for
negating the force of gravity acting upon the limb or extremety has
at least four degrees of freedom.
[0045] The invention also provides a system as recited above
further comprising means for supporting the individual, the means
selected from the group consisting of a chair, a bed, a back
support, and a trunk support.
[0046] The invention also provides a system as recited above
wherein the device for detecting the force of gravity acting upon
the limb or extremety further comprises a power transfer medium
coupled at one end to the device for detecting the force of gravity
acting upon the limb or extremety and extending away from the
system to a second end coupled to a computer processor.
[0047] The invention also provides a system as recited above
wherein the device for negating the force of gravity acting upon
the limb or extremety further comprises a power transfer medium
coupled at one end to the device for negating the force of gravity
acting upon the limb or extremety and extending away from the
system to a second end coupled to a computer processor.
[0048] In one other embodiment, the system as recited above further
comprises a device for detecting the force of gravity acting upon
the limb or extremety and/or a device for negating the force of
gravity acting upon the limb or extremety wherein the device is
automated.
[0049] The invention also provides a method of training an
individual having a neurological condition using the system as
recited above. In one embodiment, the training results in the
individual having improved motor neuron activity compared with the
motor neuron activity prior to the training, the method of training
comprising the steps of: (i) providing a system that provides at
least one force in at least one plane to a limb or extremety of an
individual having a neurological condition, the force resulting in
negating the force acting upon the limb or extremety due to gravity
and allowing the individual to move the limb or extremety in a
desired direction, the system comprising means for supporting the
limb or extremety, a device for detecting the force of gravity
acting upon the limb or extremety, and a device for negating the
force of gravity acting upon the limb or extremety; (ii) supporting
a limb or extremety of the individual using the means for
supporting; (iii) requiring the individual to move the limb or
extremety to a target site; (iv) repeating step (iii) ten times;
(v) repeating step (iv) three times; (vi) repeating step (v) to a
different target; (vii) repeating step (vi) three times; (viii)
allowing the individual to rest for a predetermined time period;
(ix) repeating the cycle of steps (ii) through (viii) but
concomitantly increasing the negating force acting upon the limb or
extremety by about 10% of that used in the previous cycle of steps;
(x) repeating step (ix) at least twenty two times; the steps
resulting in training the individual having improved motor neuron
activity when the negating force acting upon the limb or extremety
is equivalent to about 300% of the weight of the individual's limb
or extremety.
[0050] In a further embodiment, the invention provides a robotic
device, the robotic device comprising a support splint, the support
splint communicably attached to a force sensor, the force sensor
communicably attached to a transducer capable of transducing a
force input into a corresponding electrical or optical output; a
central processing unit communicably attached to and receiving an
input signal from the transducer and further communicably attached
to and sending an output signal to a motor, wherein the motor is
functionally attached to the robot arm.
[0051] In one embodiment, the system is a haptic system. In another
embodiment, the force further results in allowing the individual to
move the limb or extremety to a target site. In a preferred
embodiment, the force is provided in at least two planes. In a more
preferred embodiment, the force is provided in at least three
planes. In another preferred embodiment, the force is provided in a
plurality of planes.
[0052] In another embodiment, the invention provides a robotic
device, the robotic device comprising a support splint, the support
splint comprising an arm rest, an arm cuff, a hand splint; a
gimbal, a strain gauge force sensor, an end effector, a robot arm,
a robot stand, and a servo motor; wherein the support splint is
fixedly attached to the gimbal, the gimbal is fixedly attached to
the strain gauge force sensor, the strain gauge force sensor is
attached to the end effector, and the end effector is fixedly
attached to the robot arm; the robot arm is fixedly attached to the
servo motor, and the servo motor is fixedly attached to the robot
base; wherein in use, when a first force is applied to the support
splint the force is then transmitted to the gimbal, then
transmitted to the strain gauge force sensor, the strain gauge
force sensor converts the first force into a proportional
electronic signal, the electronic signal is transmitted to a
control unit comprising a central processing unit, the central
processing unit processes the electronic signal, the central
processing unit transmits the processed electronic signal to the
servo motor, and the servo motor responds to the electronic signal
by exerting a second force upon the robot arm. Preferably, the
second force acting upon the robot arm results in an apparent
inertial mass of the robot arm at the gimbal of not more than 2 kg.
More preferably, the apparent inertial mass of the robot arm at the
gimbal is not more than 1 kg.
[0053] In an additional embodiment, the gimbal has at least two
degrees of freedom. More preferably, the gimbal has at least three
degrees of freedom. In a further additional embodiment, the load
cell has at least three degrees of freedom. More preferably, the
load cell has six degrees of freedom.
[0054] In a yet further embodiment, the gimbal comprises at least
one position sensor. The position sensor is used to measure the
elbow and shoulder rotation angles. Preferably the gimbal has a
three position sensor, a position sensor for each degree of
freedom.
[0055] In a further embodiment of the invention, the robotic device
further comprises supporting means, the supporting means selected
from the group consisting of a chair, a bed, a back support, and a
trunk support. In one preferred embodiment of the invention, the
robotic device comprises a chair. In a more preferred embodiment,
the chair is a Biodex chair. In a still further embodiment the
robotic device further comprises a T-support track, wherein the
robotic device and the chair are positioned proximally on the
T-support track.
[0056] In an additional embodiment of the invention, the robotic
device comprises a visual display unit (VDU) screen in interactive
communication with the control unit. The interactive communication
is preferably a signal, the signal being selected from the group
consisting of an electrical signal, a photonic signal, and a radio
signal.
[0057] In another embodiment, the robotic device comprises a data
acquisition computer with screen and printer that is in
communication with and regulates the robotic device and the
three-dimensional (3-D) visual display screen. In a preferred
embodiment, the computer comprises the user interface to allow data
collection during the evaluation and training of the stroke subject
and subsequent data access for creating standardized clinical
progress reports.
[0058] In one embodiment of the invention the robotic device, the
control unit comprises a computer with a real-time operating
system, motor amplifiers, and an emergency circuit.
[0059] In another preferred embodiment, the robotic device
comprises safety and protection mechanisms to safeguard against
equipment malfunction. In an additional embodiment, the robotic
device comprises hardware safeguards to physically limit the
available robot travel range. In a preferred embodiment, a
disconnecting switch is activated when a predetermined and safe
force level at the subject's arm is exceeded, while simultaneously
turning electrical power off to the robotic device.
[0060] The invention provides a method for diagnosing a
neurological condition in a subject, the method comprising
measuring a subject's joint torques using the robotic device as
described herein. In a preferred embodiment, the neurological
condition is selected from the group consisting of hemiparetic
stroke, cerebral palsy, head trauma, and multiple sclerosis.
[0061] The invention also provides a method for measuring the
degree of rehabilitation of a subject with a neurological
condition, the method comprising measuring a subject's joint
torques at a first time starting rehabilitation and then
periodically during rehabilitation using the robotic device as
described herein. In a preferred embodiment, the neurological
condition is selected from the group consisting of hemiparetic
stroke, cerebral palsy, head trauma, and multiple sclerosis.
[0062] The invention also provides a method for measuring the
temporal change in severity of a neurological condition in a
subject, the method comprising measuring a subject's joint torques
at a first time and then periodically over time using the robotic
device as described herein. In a preferred embodiment, the
neurological condition is selected from the group consisting of
hemiparetic stroke, cerebral palsy, head trauma, and multiple
sclerosis.
[0063] The invention provides a method for treating a neurological
condition in a subject, the method comprising training a subject
using the robotic device as described herein, the treatment being
tracked and assessed by measuring the subject's isometric joint
torques. In a preferred embodiment, the neurological condition is
selected from the group consisting of hemiparetic stroke, cerebral
palsy, head trauma, and multiple sclerosis.
BRIEF DESCRIPTION OF THE DRAWINGS
[0064] The file of this patent contains at least one drawing
executed in color. Copies of this patent with color drawing(s) will
be provided by the Patent and Trademark Office upon request and
payment of the necessary fee.
[0065] FIG. 1 illustrates exemplary elements of the system.
[0066] FIG. 2 illustrates an exemplary embodiment of the system
further comprising computing means that can interact with the
gravity sensing means and the gravity negating means.
[0067] FIG. 3 illustrates an alternative exemplary embodiment of
the system.
[0068] FIG. 4 shows a schematic diagram of a common model for the
admittance control algorithm of the system.
[0069] FIG. 5 shows exemplary dimensions of the volumetric
workspace accommodated by the system in use.
[0070] FIG. 6 illustrates an exemplary prototypic embodiment of the
haptic system, the haptic system being a robotic device.
[0071] FIG. 7 shows a partial view of a part of the prototypic
robotic device comprising the support splint, the gimbal, position
sensors, the strain gauge force sensor, and the robot arm.
[0072] FIG. 8 illustrates an alternative embodiment of the gravity
negating means, a Stewart platform.
[0073] FIG. 9 shows a 3-D visual feedback display.
[0074] FIG. 10 illustrates exemplary results from an experiment
that tested the work area of a left hemiparetic subject.
[0075] FIG. 11 shows the cumulative data for experiments that
tested the work area of hemiparetic subjects. (*p<0.05;
**p<0.001).
[0076] FIG. 12 shows the results for experiments that tested the
work area before (12A) and after (12B) training.
[0077] FIG. 13 illustrates data for relative finger flexion forces
for increasing levels of active limb support.
[0078] FIG. 14 illustrates the percent of finger extension force
that subjects were able to generate with and without assistance of
functional electrical stimulation (FES) of wrist and finger
extensor muscles.
[0079] FIG. 15 illustrates one embodiment of the system of the
invention showing the robotic device, the chair, and the visual
display unit. In particular the robotic device illustrated is a
3-DOF haptic device with independent degrees of freedom.
[0080] FIG. 16 illustrates one embodiment of the system of the
invention showing the robotic device and the chair.
[0081] FIG. 17 illustrates another embodiment of the system of the
invention showing the robotic device and the chair, the robotic
device comprising a scissor mechanism to provide the degrees of
freedom. FIG. 17A shows the device from a three-quarter view; FIG.
17B shows the device from a side view; FIG. 17C shows the device
from a top view.
[0082] FIG. 18 illustrates a schematic diagram of elements of the
invention.
[0083] FIG. 19 illustrates a schematic diagram of alternative
elements of the invention having at least three degrees of
freedom.
DESCRIPTION OF THE INVENTION AND PREFERRED EMBODIMENTS
[0084] The invention provides provides a method and a system having
means for allowing a subject having a neurological condition, such
as hemiparetic stroke, cerebral palsy, head trauma, a motor neuron
disorder, or multiple sclerosis, to use a limb or other appendage,
such as an arm, a leg, a hand, a finger, a thumb, or the like, and
to provide a training and rehabilitation system and method that can
regenerate the subject's ability to perform tasks unaided or with
little additional energy input.
[0085] The method and system can be used to support such an
individual's whole body or part thereof, such as the trunk, the
shoulder, the neck, the head, the arm, the elbow, the wrist, the
hand, the hip, the leg, the knee, the ankle, and the foot. One
object of the invention is to controllably apply a force to the
body or body part such that the body or body part has a sensation
of weightlessness. Under these force conditions the individual
responds more quickly to training and self-rehabilitation than an
individual using a system from the prior art. Such an improvement
is of particular benefit in relation to time spent in
rehabilitation, thereby reducing clinical and health-related costs,
and from a physiological perspective to a more rapid rehabilitation
of tissue that might otherwise become wasted through lack of use or
lack of stimulation. In particular, the method and system help the
joint abduction responses that are of consequence for arm
extension, walking, and the like.
[0086] The system comprises means to support the limb or appendage
of an individual, the means being, for example, a splint, a plate,
a glove, a sock, a boot, or the like; means for detecting the force
of gravity upon the limb or appendage, the means being, for
example, a force sensor or the like; means for negating the force
of gravity upon the limb or appendage, the means being, for
example, a linear actuator, a rotary actuator, or the like;
controlling means for integrating the output from the force
detecting means and the input to the force negating means, the
means being, for example, an integrated circuit, a switch, a
microprocessor, a linear actuator, a lever arm coupling, or the
like.
[0087] The means for negating the force of gravity can have at
least one degree of freedom; for example, a solenoid has one degree
of freedom, the Arm Coordination Training 3D device disclosed below
has at least three degrees of freedom in which the force is
controlled and three degrees of freedom in which the position is
measured. The invention does not limit the number of such means
that can be used in combination with one another, nor limit the use
of combining different such means in one system.
[0088] The system of the invention solves the problem that when the
arm is extended away from the shoulder and body of a user, the
force of gravity acting upon the extremity in proportion to the
distance of the extremety (wrist or hand) to the axis of the
shoulder, thereby activating the abnormal synergy between shoulder
abduction and elbow flexion. The user therefore experiences less
resistance to extending his or her arm and is more able to reach a
distant object. The system therefore has the novelty of
manipulating the force of gravity upon a limb or appendage of a
subject.
[0089] The method and system can also have incorporated parameters
that determine velocity of movement and range of force necessary to
aid treatment and self-rehabilitation using the disclosed elements
of the invention. Such parameters are, for example, but not limited
to, a subject's condition, such as being adult or child, being a
victim of stroke, a victim of head trauma, a child having cerebral
palsy, degree of disability, size, age, time elapsed since onset of
condition, or the like. Additional parameters can include, but are
not limited to, configuration of the system, such as being mounted
upon a floor, wherein the subject sits or stands in proximity to
the system, whether the system is mounted as an exoskeleton that
the subject wears, sits or stands in, or is mounted on a wheelchair
or the like, or a chair, bed, stool, or the like, or is mounted on
a separate object, such as a workstation, a control panel, a
cockpit, a capsule, or the like, or is mounted in a garb that a
subject can wear, such as a diving suit, a space suit, or the
like.
[0090] Two exemplary schematic illustrations of the system are
shown in FIG. 18 and FIG. 19.
[0091] The development of the Arm Coordination Training 3D device
(ACT.sup.3D) to measure gravity-induced discoordination and to
deliver highly controlled patient-specific upper extremity
rehabilitation provides a superior training and quantification tool
that will benefit individuals who have had a stroke. Likewise,
administering novel functionally relevant training protocols to a
chronic stroke population whom has reached a functional plateau in
formal physical therapy may demonstrate that these individuals
retain untapped potential for recovery not reached prior to
discharge from outpatient therapies.
Isometric Training to Reduce Gravity-Induced Discoordination
[0092] We have trained stroke subjects to generate isometric joint
torques that explore combinations away from the abnormal torque
patterns they exhibit (Ellis et al. (2002a) supra; Ellis et al.
(2002b) supra; Ellis et al. (2003) supra). More specifically we
combined shoulder abduction and elbow extension torques as one of
the isometric strengthening tasks that stroke subjects performed.
After an eight-week training protocol (three 1.5 hour sessions per
week) the ability to concurrently generate shoulder abduction with
elbow extension and shoulder flexion improved. In all chronic
stroke subjects (n=8) trained to date we saw marked improvements in
the ability to combine elbow extension torques (normalized to max
abilities) and shoulder abduction torques (expressed as a function
of arm weight). To demonstrate the effect of these improvements on
reaching movements, we compared the pre-versus post-training
kinematic results for our most impaired subject trained to date.
Before the training period this stroke subject was not able to
reach a gray target when actively supporting the arm against
gravity. Following the eight-week isometric training protocol he
was able to reach the target. Moreover, he was able to reach the
target with a velocity profile similar to the one obtained with his
arm on an air bearing support at the onset of the training. Thus
our subject significantly increased his reaching ability both with
regards to reaching area and movement velocity.
[0093] In the less impaired subjects the improvements during
reaching movements were not always as consistent. We believe that
this is due to the fact that these subjects were already able to
extend their elbow when lifting up their arm against gravity.
However, this ability may be challenged when subjects attempt to
move objects with a certain weight especially during reaching
movements. This expectation is based on our observation that even
mildly impaired subjects exhibit abnormal isometric torque patterns
when generating maximum shoulder abduction torques (Dewald and Beer
(2001b) Topics Stroke Rehab. 8: 1-11). The use of the robotic
device of the invention for the administration of a selective
dynamic strengthening protocol tests this hypothesis by having
subjects lift loads greater than the weight of their arm,
simulating tasks such as picking up a book. Under these conditions
we saw similar reductions in work area occurring in these more
mildly impaired stroke subjects.
[0094] The lower extremeties have also been evaluated using the
disclosed isometric protocols and have produced similar results
thereby validating the method and system for the whole body.
Therefore all references herein to upper extremeties can also be
taken to apply to the lower extremeties.
[0095] Thus, there is a current deficiency in rehabilitation
science calling for a device that can both accurately evaluate and
deliver highly controlled patient-specific upper extremity
rehabilitation while taking into account disabling nature of
shoulder elbow joint torque coupling when attempting to lift the
limb against the force due to gravity. The present invention
realizes this need by providing high-resolution measurements of
physiological (strength and coordination) and functional (reaching
workspace) performance at various levels of limb support and in
various virtual environments. The device, accessories, and methods
described herein allow practitioners to evaluate and train
movements with partial limb support, and hence vastly increase the
ability to successfully train subjects across all levels of
impairment. Furthermore, the invention is based upon results from
an isometric training study. The advantages of the invention is
that it incorporates the ability to control the level of limb
support and to move in a three-dimensional (3-D) workspace,
features that are unavailable in the isometric training protocol or
in other robotic technologies. Therefore, it is expected that the
device and method of the invention have clear potential to improve
the functional abilities in the upper limb by reducing the
gravity-induced discoordination during real-world reaching efforts
beyond that of isometric training.
[0096] The invention is designed to deliver novel interventions
that train and rehabilitate individuals with a broad spectrum of
upper extremity reaching impairment to progressively overcome the
negative effects of gravity by providing various levels of limb
support. Furthermore, the device and methods of the invention
emulate real life scenarios such as reaching and retrieval of
objects of different weight in space. The device interfaces with
the user in a safe and comfortable fashion quickly setup by a
healthcare provider. It may be implemented with the large number of
people who currently suffer from the disabling effects of
stroke.
[0097] The major focus of the invention is that it implements a
virtual mechanical environment both to evaluate synergistic
movement constraints and to deliver patient-specific rehabilitation
for the impaired limb with gravity-induced discoordination.
Furthermore, in an effort to accommodate patients with various
severities of impairment, the invention is able to simulate: 1)
movements of objects with increasing mass for mildly impaired
subjects, 2) movements with partial support of the impaired limb
for the more severely impaired subjects and 3) movements in both
horizontal and inclined planes. The overall goal of the invention
is to increase the paretic arm's functional workspace by increasing
levels of shoulder abduction while reaching in various directions.
This is accomplished by having subjects perform reaching movements
to targets on virtual planes centered through the shoulder
progressively from down to horizontal to up. It is important to
note that it is simpler to measure the forces acting upon and
position of a subject's limb having the system mounted on an
exo-skeleton than if the subject is not constrained. Constraining
the shoulders is therefore necessary in situations whereby the
system would otherwise enable the subject to move the shoulder in
relation to the body and arm, thereby confounding the ability of
the system to accurately measure force, position, and
acceleration.
[0098] In one embodiment of the invention, a 3-D force controlled
robot arm was implemented to progressively overcome the negative
effects of gravity (abducting or lifting the weight of impaired
limb) during functional movements. In this embodiment, the robotic
device comprises a controlled robot arm that is a modified
HAPTICMASTER device (FCS Control Systems B.V., Schiphol, The
Netherlands). The device integrated a 3-D force controlled robot
arm with a seating system and a compact real-time two-dimensional
(2-D) or 3-D virtual reality visual system, creating an upper
extremity rehabilitation device that implements a virtual
mechanical environment for the progressive reduction of upper
extremity gravity-induced dysfunction. The seating system and the
robot arm were placed on a T-support track that allowed the seating
system and the robot to move relative to each other and to rotate
(see, for example, U.S. Pat. No. 5,209,223).
Admittance Control Robotics
[0099] Admittance control technology uses a force measurement
device (loadcell) placed at the patient interface. The loadcell
functions as the force feedback device in a closed loop force
control system. Therefore the forces are always controlled at the
patient's interface, and system non-linearities mentioned before
are minimized because they occur inside the force control loop and
can therefore be compensated to a large degree.
[0100] One feature of the robot arm is that it must have virtually
no inertia such that the user does not feel any resistance when
attempting to move the device. The inertia of the robot arm should
not be more than 2 kg. Currently, the only commercially available
robot that couples this feature with a 3-dimensional operational
capability is the HAPTICMASTER device (FCS Control Systems), but
other robots with these features may also be used. The low level
inertial properties of the HAPTICMASTER device enables application
to individuals with all levels of impairment severity including
individuals with severe impairment who would otherwise be unable to
move against the inertial resistance of other robotic devices. Work
done at the University of Reading, England has shown that the robot
is safe and can assist reaching movements to various targets in the
workspace of the paretic arm following stroke (Coote and Stokes
(2003) Technol. Disability 15: 27-34; Harwin and Hillman (2003)
Robotica 21; and in: "Assistive technology-added value to the
quality of life": AAATE '01. Amsterdam, Association for the
Advancement of Assistive Technology in Europe, vol. 10 (2001)
Marincek, Buhler, Knops and Andrich, editors, Washington, D.C.: IOS
Press, Washington D.C.). However, it has not been employed to
assist subjects in reaching movements with the upper extremity
constantly supported by an external device.
[0101] A modified HAPTICMASTER was used in the ACT.sup.3D system
prototype as described in the Examples section to demonstrate and
validate the systems and protocols envisioned, even though the
HAPTICMASTER's range of motion is limited, but adequate for
protocol validation. The robot can be redesigned such that
individuals will be able to move over virtual planes in all
dimensions of their available reaching area, albeit limited in
travel. Furthermore, the actual limb movement and level of limb
support in a virtual mechanical environment can be provided via
real-time 3-D visual feedback, such that individuals can learn to
progressively support more of the weight of their arm as they reach
for objects on various horizontal and vertical planes. To date no
center applying robotic technology has targeted the negative
effects of gravity loading on reaching with the impaired upper limb
in all planes of available movement.
[0102] The type of admittance control (force control) that is used
in the robotic device results in a device that individuals can move
freely as if they were not attached to the robot. The key features
accomplished with this type of control system are that the weight
of the robot itself and any friction in the movement are
compensated and thus not sensed by the user. These properties are
considered when designing a device for rehabilitation, especially
for individuals with stroke who will exhibit weakness and
discoordination as compared to able-bodied individuals.
[0103] In a preferred embodiment of the present invention, the
robotic device is designed to support a subject's arm and is
configured such that the subject can move their arm over virtual
planes in all dimensions of their available reaching area.
Furthermore, the actual limb movement and level of limb support in
a virtual mechanical environment are provided via real-time 3-D
visual feedback, such that subjects can learn to progressively
support more of the weight of their arm as they reach for objects
on various horizontal and vertical planes.
[0104] The robotic device can comprise a transducer that converts
the force input to an output signal comprising an electrical
signal, an optical signal, a radio signal, or combination thereof.
The output signal is transmitted to the computer unit that receives
the output signal for processing and which then sends the processed
output signal to the servo motor.
[0105] The visual display unit may be of any suitable type, for
example, with respect to the 3-D visual display, the Dimension
Technologies Inc. (DTI, Rochester N.Y.) 18.1'' screen is preferred
for the following reasons: it does not require consumers to wear
glasses and when the screen is centered, it provides realistic
real-time 3-D displays of the virtual arm, targets and objects.
Furthermore, the screen also permits for regular 2-D displays as
well without losing its resolution. Any display having similar
properties can be used with the invention, such as a 19'' LCD VDU
from Dell (Round Rock, Tex.).
[0106] The primary application of the invention is to test a
patient's ability to generate shoulder abduction torque while the
paretic arm is moved over virtual planes. However, it is also
feasible to monitor the ability of stroke survivors to control
wrist and finger extension using a commercially available pressure
mapping system installed on the hand splint. Individuals with
stroke would be asked to extend their fingers and wrist (zero
pressure) while performing the same reaching motions toward a
virtual object. The force exerted by the fingers and the palm of
the hand on the splint can be measured by the pressure mapping
system and displayed for feedback.
[0107] Another example of the application of the invention is the
addition of neuromuscular electrical stimulation of hand and wrist
muscles as subjects get close to objects that they would like to
grasp in their virtual world. This is particularly helpful if
volitional control of finger, wrist and/or elbow extensors is
absent or insufficient.
[0108] With respect to therapeutic techniques various approaches
can be used based under the unifying principle of multi-degree of
freedom strengthening. Although in all cases movements on virtual
planes will be encouraged, while generating increasing levels of
shoulder abduction, the strengthening of shoulder/elbow
flexion/extension torques, required especially for reaching
directions can be accomplished by different approaches: 1)
ballistic movements; 2) movements in viscous fields or 3) movements
with capped maximum velocities or isokinetic movements.
[0109] In another embodiment, the device includes a data
acquisition computer with screen and printer wherein the computer
is in communication with and regulates the robot and the 3-D DTI
screen. In a preferred embodiment, the computer houses the user
interface to allow data collection during the evaluation and
training of the stroke subject and subsequent data access for
creating standardized clinical progress reports. A large segment of
chronic and acute patients could benefit from the proposed 3-D
measurement and training system. It is important to point out that
currently there are no commercially available systems designed for
rehabilitation of the upper extremity. All systems that have been
reported in the literature are currently being tested as research
units and cannot deal with the devastating effects of abnormal
elbow/shoulder synergies on movement following stroke. As such,
this is the first unit that targets the devastating effects of
synergies when lifting the paretic arm against gravity.
[0110] In a preferred embodiment, safety and protection mechanisms
are implemented to safeguard against equipment malfunction, which
could cause an axis to move at high speed towards an extreme
position. Software limits for travel, velocity, acceleration, and
force, incorporated in the real time software continuously guard
against inadvertent signals in the robot control systems, and once
triggered, turn the electrical power off to the robot in a
controlled manner, thereby forcing the robot in a passive state.
Over and above these software safeguards, hardware safeguards can
also be included to physically limit the available robot travel
range. These travel limits can be set by the health care provider,
dependent on the available and safe range of motion of the patient
so as not to cause any harm to the patient under the most severe
equipment malfunction, an uncontrolled motor run-away. A "quick
disconnect" can also be introduced at the patient/machine
interface. In a preferred embodiment, this disconnect activates
when a predetermined and safe force level at the patient's arm is
exceeded, while simultaneously turning electrical power off to the
robot.
Hardware
[0111] The HAPTICMASTER hardware comprises two main functional
components: the robot arm (29) and the control unit (14), shown in
FIG. 6 and FIG. 15. The robot arm serves as the actual force
display unit, whereas the control unit houses the electronics such
as amplifiers, safety relay, and the haptic server.
[0112] The robot arm is built with zero backlash and minimal weight
in consideration of the human operator. Zero backlash is a
requirement because human tactile senses have a spatial resolution
of vibrations with amplitudes as small as 10-100 microns, such as
those caused by play in mechanical parts (Burdea (1996) "Force and
Touch Feedback for Virtual Reality" Wiley, New York N.Y.). Minimal
weight is necessary for safety, as both the speed and the mass of
the robot arm determine its energy content in a possible collision
with the human operator. The speed is set to the value of a normal
human arm motion (for example, about 2.2 m/s), and lightweight
aluminum tubing construction minimizes the mass of the robot arm.
The kinematic chain from the bottom up yields: base rotation, arm
up/down, arm in/out, which gives three degrees of freedom at the
end effector. A volumetric workspace is created, which is large
enough to enclose most human single-handed or double-handed tasks,
shown in FIG. 5. The dimensions of the volumetric workspace space
are approximately as follows: horizontal rotational displacement:
about 1 rad; vertical displacement: about 0.4 m (15.7''); and
horizontal extent: about 0.36 m (14.2''). In the alternative, the
dimensions can be about 1 rad, about 0.4 m, and about 0.18 m
(7.1''), respectively. In another alternative, the dimensions can
be about 0.5 rad, about 0.2 m (7.85''), and about 0.18m,
respectively. In yet another alternative, the dimensions can be
from between 0.5 rad and 1.0 rad, 0.2 m and 0.4 m, and 0.18 and
0.36 m, respectively.
[0113] The support splint can be provided in sizes suitable to
conform and fit the anatomical size of a subject's hand, wrist,
and/or arm.
[0114] Anti-backlash leadscrew spindles are mounted with a flexible
coupling to the axis of a DC brushed motor, such as the Starsys
Precision Brush Motor (Starsys Research Corporation, Boulder,
Colo.). This backlash-free solution introduces some friction in the
mechanism. However, the friction is completely eliminated by the
control loop, up to the accuracy of the force sensor. The result in
force control mode is a backlash-free, stick-free, and slip-free
smooth motion at the end effector. A sensitive strain gauge force
sensor is located right behind the end effector. Such strain gauge
force sensors, (or load cells) are well known in the art, for
example, the JR3 Sensors, the EBB Series Load Cell, or the LSP
Series Load Cell, available from Transducer Techniques (Temecula,
Calif.). By placing the force sensor adjacent to the end effector,
the interaction force is measured as close to the human hand as
possible to avoid distortion of the force signal and to optimize
system performance. Exchangeable end effectors can be mounted to
the force sensor to match the application.
[0115] Servo motors that drive the robot arm are well known to
those in the art, and include, for example but are not limited to,
the 815 BR or the 1525-BRS (Servo Dynamics Corp., Chatsworth,
Calif.).
[0116] The control box contains electronics such as the computer,
motor amplifiers, and an emergency circuit. The virtual model is
rendered by a dedicated computer with a real-time operating system,
such as VxWorks (Wind River Systems, Alameda Calif.), QNX, LynxOS,
VRTX, pSOS, Windows-CE, Nucleus RTX, RT Linux, or the like, herein
termed the haptic server. Such servers and operating systems are
well known to those in the art. It runs at a fixed 2,500 Hz refresh
rate. This frequency is high enough to guarantee a haptic quality
for a smooth and realistic experience since it is approximately ten
times higher than the maximal human discrepancy value (Burdea G.
(1996) "Force and Touch Feedback for Virtual Reality" Wiley, New
York, N.Y.). Finally, the PID motor control loop runs on the
amplifiers at a 20 kHz pulse width modulated frequency.
[0117] It must be noted that other configurations of the haptics
device are possible. FIG. 7 shows one such example of a 3-D haptic
device concept where the axes are fully independent. It is expected
that many versions of the haptics device can be developed,
accommodating the full performance envelope of force, velocity and
position.
[0118] It is understood that for someone skilled in machine design
arts many different ways can be found to drive an end effector in
three or more degrees of freedom. Another example which is far more
complex is a Stewart platform given in FIG. 8. This is used
extensively in flight simulators to provide motion sensation to the
pilots. The hexapod structure uses six actuators, and is thus
capable of providing 6-DOF motion. In the system of the present
invention the end effector and associated additional hardware may
be mounted on the top platform.
[0119] End effectors may also be of different configurations
dependent on the application and type. As mentioned before;
specific protocols integrating wrist and/or hand functions may
require unique and dedicated end effectors.
[0120] At least one position measurement device can be included in
the system, a position measurement device such as a tracker, a
potentiometer, or the like. The position measurement device can be
fixedly attached to any part of the system, however it is
preferable to attach the position measurement device to or adjacent
to a joint, a sensor, or the like so that the position of an
element of the system can be accurately determined. In addition,
several position measurement devices can be attached to different
locations on the system in order to obtain an accurate estimate of
the position of the entire system.
Extra-Skeletal Approach
[0121] In addition to end effector approaches, extra-skeletal
robots can be designed to implement the same virtual mechanical
environment to overcome gravity-induced dysfunction in upper
extremity paresis following stroke, head trauma and spastic
cerebral palsy. As long as such systems can provide the means to
alter weight of the limb in the workspace of the arm and the
ability to provide resistance to motion, if so desired, the
neurorehabilitation concept discussed as part of this application
can be realized.
Rehabilitation Software and Visual System
[0122] The rehabilitation software can allow for data collection of
evaluation and training data. The quantitative evaluation software
is developed to:
[0123] A: Determine any limitations in available torque
combinations by performing ballistic reaching/retrieval arm
movements in different directions while maintaining various levels
of shoulder abduction torque. The protocol allows an operator to
establish the severity of abnormal coupling(s) between
shoulder/elbow torques as a function of limb support activity. The
software can display sparks and/or provide a sound (indicating that
the arm is touching the table) if subjects do not maintain an
abduction level within 10% of the assigned torque value during
reaching/retrieval movements.
[0124] B: Determine the maximal reaching workspace. Similar
movements as in A) can be performed for different inclination
angles of the virtual table to determine the maximal reaching
workspace of the paretic arm. In the case that subjects are mildly
impaired, the weight of the arm can be increased by simulating
objects with increasing weight. The software can be developed to
provide a quantitative measure of workspace for different
percentages of active limb support.
[0125] The results provided by protocols A & B can be used to
establish parameters for the training software. The software for
the training protocols is similar to the software developed for the
quantification and evaluation protocols allowing visual display and
measurement of virtual planar surfaces and active support of the
limb. In addition, this software can allow for a progressive
increase of active support of the limb during subsequent training
sessions.
[0126] An additional component in the training software can be to
provide a subject the ability to strengthen the paretic limb while
subjects move on a virtual plane supporting part of the weight of
the arm. This can be realized by:
[0127] A) Ballistic reaching movements using predetermined targets
on the virtual plane;
[0128] B) Reaching motions within an imposed viscous field on the
virtual plane; and/or
[0129] C) Isokinetic reaching motions that limit the maximum
movement velocity on the virtual plane.
[0130] The training software can include these options for research
and clinical trials.
[0131] It is expected that the software may evolve into a
stimulating "gaming scenario", perhaps associated with some form of
"scoring" providing the subject with additional incentives. This is
even more important if the subjects are children reorienting the
concentration from the rehabilitation aspects to more "fun and
games".
[0132] In order to create a realistic training environment for
individuals with stroke, 3-D visual feedback is incorporated into
the system. We have used a 3-D virtual reality world displayed on a
large flat panel monitor, as is seen in FIG. 9. The display
includes an avatar of the subject's arm and hand providing a direct
link from the task to the virtual world. FIG. 9, an exemplary
display of a VDU screen image, shows a virtual environment
comprising a surface (table-top; dark grey), the area delimited by
the reach of the individual's arm (wall; red), several targets in
the vicinity if the individual's reach (hemispheres; blue), an
avatar of the individual's virtual arm consisting of movable joints
(hand, elbow, and shoulder; orange) and limb bones (forearm and
upper arm; blue-green), the starting position of the hand
(hemisphere; light grey), and the track/path of the hand to one of
the targets (lines; green). Different colors may be chosen to
represent different objects or avatars in the display according to
the subject's preference. Subjects with color-blindness may also
have a choice of color schemes.
User Interface Software
[0133] With an application programming interface (API), the user
creates the virtual model on the haptic server. The real-time
operating system on the haptic server interprets the virtual model
and generates the trajectories for the robot, based on the force
sensor input. The haptic server also incorporates issues like
safety guards, communication protocols, and collision detection
with virtual objects. In one embodiment, the HAPTICAPI, which is a
C++ programming interface, is used by the programmer to define or
modify the virtual haptic world via an Ethernet connection to the
robot that controls the internal state machine (FCS Control
Systems). Through the robotic control, haptic effects can be
created (like dampers and springs), and spatial geometric
primitives can be defined (like spheres, cones and cubes). Simple
virtual worlds can be created using these effects and primitives.
When more complex virtual worlds are required, e.g. with meshed
surfaces or deformation, another rendering method needs to be
applied. A local mass model will be rendered on the haptic server,
and the forces acting on this mass due to interaction with the
virtual world are rendered from a host computer. When the end
effector collides with a virtual object, an appropriate force and
displacement are presented to the user. The relationship between
force and displacement is given by the object properties of the
virtual model (for example, stiffness, damping, friction, or the
like). With a penalty-based method, the appropriate relation
between force and displacement is calculated by the real-time
operating system and incorporated in the position, velocity, and
acceleration signal.
Control Algorithm
[0134] The robot uses an admittance control algorithm. A force
sensor measures the interaction force between the user and the
system. From these forces, a virtual mass model calculates
position, velocity, and acceleration (PVA), which an object touched
in the virtual world would obtain as a result of the applied force.
An example of such an admittance control algorithm is currently
used and available in FCS Control Systems' devices (see U.S. Pat.
No. 4,398,889 herein incorporated by reference in its
entirety).
[0135] The virtual world defines the space in which the object
lives (for example, gravity, environmental friction, position of
the object, etc.) and the object properties (for example, mass,
stiffness, damping, friction, etc.). The virtual mass model will
typically contain a mass larger than zero, to avoid commanding
infinite accelerations and causing system instabilities. The
PVA-vector serves as a reference signal for the robot, realized by
a PID servo control servo loop.
[0136] With proper feedback gain settings, this control loop will
compensate the real mass of the manipulator up to a factor of six,
and terminates its internal friction up to the accuracy of the
force sensor. So, if the mass of the manipulator behind the
actuator is 15 kg, the operator feels only 2.5 kg at the end
effector.
[0137] Since gravity can also be almost eliminated, the perception
of the user is that they are moving a mass much less than 2.5 kg.
The admittance control algorithm is shown in FIG. 4.
Robot Control Software
[0138] The robot control software allows specification of the
following features from the user interface:
[0139] a) Inclination angle of the virtual table imposed by the
robot arm, which can be varied while exercising the paretic
limb.
[0140] b) Virtual environment imposed by the robot: the environment
can be made to feel like a hard constraint onto a plane, allowing
no movement above or below it with free or guided motion on that
plane; or like a one-sided upper or lower constraint. These
abilities will allow individuals with stroke to lift their arm from
the virtual table with or without a downward "virtual gravity"
force and away from the virtual surface.
[0141] c) Virtual objects: generation of mechanical effects when
manipulating a virtual object on the plane or addition of the
"virtual gravity" force when a subject attempts to lift and move
objects of different weights. See U.S. Pat. No. 4,398,889
incorporated herein by reference in its entirety.
Training
[0142] Following an injury to the brain, gravity introduces
limitations to a compromised system and restricts arm movements in
stereotypical ways. These limitations are associated with the
activation of anti-gravity shoulder muscles used to lift the arm,
and associated overflow into elbow/wrist and hand flexor muscle
activation. This in turn reduces elbow extension capabilities,
because it is necessary to overpower the flexion activation in
order to reach outward away from the body. Furthermore, flexor
activity in the wrist and hand are expected to increase as well
resulting in very disabled upper limb. The phenomena of synergies
(shoulder abduction involuntarily coupled with elbow/wrist and
finger flexion) is seen in a variety of cases of brain injury; it
has been most extensively studied in adult hemiparetic stroke, but
it is also seen in cerebral palsy (CP) and head trauma. Using a
state-of-the-art haptic system such as, but not limited to, the
current ACT.sup.3D robot prototype or the like, can investigate how
subjects are able to interact in an environment of reduced or
eliminated gravity, as well as enhanced gravity. Such a system can
be used to provide infinite support via haptically rendered rigid
objects, or to provide forces along the vertical axis scaled to a
subject's limb weight. In this way, the system is able to reduce or
increase the amount of abduction torque a subject is required to
create at the shoulder in order to work in the environment. Using
an instrumented arm rest and adjustable gimbal on the end effector
of such a system, finger/wrist/elbow/shoulder angles (kinematics)
and forces/moments (kinetics) can be monitored during movement.
Finally, the system may also include a supporting means, such as,
but not limited to, a chair, a bed, a back support, and a trunk
support, that will constrain movements of the trunk during the
monitoring/therapy of arm movements. Furthermore, such a supporting
means should have various adjustable and marked degrees of freedom
that will allow the experimenter/therapist to place a test subject
in the same position for subsequent measurement/therapy
sessions.
[0143] We disclose herein that even highly impaired subjects are
able to increase available workspace of the hand when gravity and
the need to activate anti-gravity muscles are eliminated.
Similarly, mildly impaired subjects may revert to similar patterns
when required to work in environments of enhanced gravity, where
they are required to generate abduction torques greater than that
to lift the weight of their own limb. These patterns and movement
characteristics as well as their treatment are elucidated using a
variety of protocols, as described by example below.
[0144] Reaching in a plane can be explored with the system of the
invention while concurrently measuring shoulder/elbow/wrist and
finger forces and torques. This allows back calculation of shoulder
and elbow torques as well as wrist and finger forces/torques, which
is a highly relevant output measure of the system. The system
should also allow for the monitoring of higher speed movements
during which subjects will be able to perform ballistic reaches in
several conditions: fully supported by a rigid plane passing though
the center of rotation of the shoulder, and with various levels of
support that virtually reduce or enhance gravity.
[0145] To reduce the impact that hyperexcitable stretch reflexes
(or spasticity) may also have on the compromised arm, a second
protocol emphasizes slow movements. Subjects are asked to slowly
make the largest circle they can with their arm. Like the ballistic
reaching protocol, this is done under rigid support from a haptic
table, or while the subject is concurrently producing particular
levels of active limb support. By reducing the impact of
spacticity, the best picture of available work area can be
constructed.
[0146] A third way that the device can be used is to characterize
free reaching that is not confined to a plane. By unlocking the
position of the gimbal and removing rigid haptic constraints, a
subject's freely selected path can be studied. The effects of
gravity can still be studied in the same manner by applying bias
forces at the end effecter, as in previous protocol descriptions.
This method removes external constraints and restrictions and
provides a potentially more relevant look at movement following
brain injury.
[0147] A fourth ability that the haptic system should have is to
allow for the introduction of particular fields and external
perturbations to the arm. Such a system can then be used to
investigate the effect anti-gravity muscle activity on the
expression of spasticity. The system can be used to provide a quick
constant acceleration perturbation to the elbow joint, shoulder
joint, or a combination of the two. Using system analysis
techniques and information obtained from EMGs, excitability of the
reflex can be elicited and compared to the level of active limb
support requirements.
[0148] A fifth property of the system is that it should be able to
measure kinetics of the wrist and fingers during the monitoring of
the effect of gravity on reaching and retrieval motions with the
arm.
[0149] Finally, viscous fields can be implemented to the system
during all or parts of the movement to induce velocity dependent
resistance against movement. This may allow us to more fully
characterize the movement patterns of more mildly impaired
subjects.
[0150] These protocols, or combinations thereof, can be used as
outcome measures for the intervention protocols described in the
next section, or as a means of more specifically quantifying
impairments that result from the presence of gravity and need to
drive activation of shoulder abduction/external rotation
muscles.
[0151] Under the guidance of our gravity-induced discoordination
concept as disclosed above, several therapeutic intervention
protocols can be realized. A protocol in which gravity is
progressively re-introduced during outward reaching over a period
of 8 weeks can be used. Training can be further extended or
shortened according to a subject's clinical needs. More
specifically, subjects can be trained to reach outward towards
virtually displayed targets that are situated on a horizontal plane
associated with the height of the shoulder. Progressive gravity
re-introduction training can be expanded beyond reaching on a
shoulder-height horizontal plane to three-dimensional or free
reaching. Such a protocol would include reaching toward outward
targets positioned on the three dimensional surface defined by the
most distant reaching points throughout the entire volume of the
arm's range of motion. Gravity could be progressively re-introduced
as in the previous protocol over a period of time as a subject is
trained to reach outward toward randomized targets in their
three-dimensional work area.
[0152] The acquisition of targets during progressive gravity
re-introduction training can also be expanded upon by integrating
functional electrical stimulation (FES) of the elbow, wrist, and
finger extensors during antigravity reaching and subsequent
grasping of virtual objects. In the absence of FES subjects
progressive generate greater finger and wrist flexion forces when
lifting up more of the weight of their arm (see FIG. 13).
Artificial stimulation of extensors would assist in counteracting
abnormal flexor activity that occurs during reaching toward targets
in the presence of gravity (FIG. 14). This assistance may
facilitate the changes resultant from progressive gravity
re-introduction by triggering appropriate muscle activity.
[0153] Progressive gravity re-introduction training during
horizontal and free reaching with and without functional electrical
stimulation can also be employed in the presence of horizontal
viscous fields, inertial fields simulating the transport of
objects, or controlled joint angular velocities (isokinetic
reaching). The implementation of these various forms of horizontal
resistance against the direction of reaching would have a
strengthening effect that may facilitate the changes resultant from
progressive gravity re-introduction.
LIST OF REFERENCE NUMERALS
[0154] 1. Means for supporting limb [0155] 2. Device for detecting
force of gravity (force sensor; load cell) [0156] 3. Device for
negating force of gravity (force generator; actuator) [0157] 4.
Means for interdispositioning and communicating between detecting
device and negating device [0158] 5. Support [0159] 6. Means for
articulating and attaching system to a support [0160] 7. Basal
substrate (floor, table, etc.) [0161] 8. Means for computer
processing [0162] 9. Power transfer medium [0163] 10. Power
transfer medium [0164] 11. User, subject, or individual [0165] 12.
Force transducer [0166] 13. Force sensor [0167] 14. Controller
[0168] 15. Power transducer [0169] 16. Integrated circuit [0170]
17. Power transducer [0171] 18. Virtual mass [0172] 19. Power
transducer to servo motor [0173] 20. Power transducer to virtual
world (virtual environment) [0174] 21. Servo motor [0175] 22. Force
transducer [0176] 23. Virtual world (virtual environment) [0177]
24. Power transducer [0178] 25. Vertical axis of motion [0179] 26.
Horizontal axis of motion [0180] 27. Horizontal axis of motion,
unused [0181] 28. Rotational arc of motion [0182] 29. Robot arm
[0183] 30. Splint [0184] 31. Gimbal [0185] 32. JR3 force sensor
[0186] 33. Dampers [0187] 34. Spline [0188] 35. Robot base [0189]
36. Bottom plate [0190] 37. Position sensor [0191] 38. Top or
mobile plate [0192] 39. Upper leg [0193] 40. Cylindrical joint
[0194] 41. Lower leg [0195] 42. Lower universal joint [0196] 43.
Fixed base plate [0197] 44. Chair [0198] 45. Actuator [0199] 46.
End effector [0200] 47. Y-axis hinge [0201] 48. Linkage [0202] 49.
Display [0203] 50. Virtual world display signals [0204] 51. Object
display (FVP) [0205] 52. Subject force, velocity, and position for
data acquisition [0206] 53. Real time control system (dedicated
computer; real time operating system) [0207] 54. Motor drive,
X-axis [0208] 55. Motor drive, Y-axis [0209] 56. Motor drive,
Z-axis [0210] 57. Axis position and velocity signals [0211] 58.
Axis 1 [0212] 59. Axis 2 [0213] 60. Axis 3 [0214] 61. Additional
axes (rotational DOFs; wrist, fingers) [0215] 62. Robotic device
[0216] 63. Virtual limb bones [0217] 64. Virtual moveable joints
[0218] 65. Virtual tabletop [0219] 66. Starting position [0220] 67.
Virtual hand path [0221] 68. Virtual targets [0222] 69. Virtual
boundary
EXAMPLES
[0223] The invention will be more readily understood by reference
to the following examples, which are included merely for purposes
of illustration of certain aspects and embodiments of the present
invention and not as limitations.
Example I
Modification of the HAPTICMASTER to Meet Required
Specifications
[0224] The Required Needs for Stroke Subjects Fall Well within the
Performance Criteria of the current version of the HAPTICMASTER.
Its peak force ability of 250 N is considerably greater then the
maximum adduction forces measured in our strongest stroke subjects
(Dewald and Beer (2001a) Muscle Nerve 24: 273-283). The position
resolution of the HAPTICMASTER device was approximately 4-14 .mu.m.
This was more than sufficient for the kinematic measurements needed
for the real time visual display (see also rehabilitation software
development). The maximum deceleration (50 m/s.sup.2) and simulated
stiffness (50,000 N/m) were also more than sufficient for the
mechanical environments we intend to simulate. The only performance
criteria that was adjusted was the maximum velocity of the robot
which was originally 1.4 m/s, less than the needed 2.25 m/s to
accommodate the peak tangential velocities seen at the forearm and
upper arm during ballistic movements in mildly impaired subjects
(Beer et al. (2004) Exp. Brain Res. 156: 458-470). The HAPTICMASTER
device was redesigned to accommodate these movement velocities. The
basic drive mechanisms of the three primary axes were modified to
accommodate the new requirements. This was accomplished by
selecting a different drive motor, in this case a brush type servo
motor, or by changing the force transmission consisting of a lead
screw, or a combination thereof. The lead crews were placed such
that one screw was directly connected to a drive motor and this
combination is used to drive one axis. Since there are three axes
there are also three motor/lead screw combinations. Drive motor and
lead screw changes can affect the output torque and speed of each
axis. This makes it possible to select a combination of motor and
lead screw suitable to the requirements of each of the three
axes.
Example II
Impairment Quantification
[0225] Reaching in a plane was explored using the system while
concurrently measuring shoulder/elbow/wrist and finger forces and
torques. This allowed back calculation of shoulder and elbow
torques as well as wrist and finger forces/torques, which is a
highly relevant output measure of the system. The system also
allowed for the monitoring of higher speed movements during which
subjects were able to perform ballistic reaches in several
conditions: fully supported by a rigid plane passing though the
center of rotation of the shoulder, and with various levels of
support that virtually reduce or enhance gravity. Subjects were
asked to move to a variety of targets within a plane that passes
through the center of rotation of the shoulder. These targets
encompass both reach and retrieval directions, and the plane can be
horizontal, inclined or declined. In this way, the ACT.sup.3D
system can characterize fast point to point movements in several
aspects of the total work volume while also monitoring wrist and
finger kinetics and/or kinematics.
[0226] To reduce the impact that hyperexcitable stretch reflexes
(or spasticity) may also have on the compromised arm, another
protocol emphasizes slow movements. Subjects were asked to slowly
make the largest circle they could with their arm. Like the
ballistic reaching protocol, this was done under rigid support from
a haptic table, or while the subject was concurrently producing
particular levels of active limb support. By reducing the impact of
spacticity, the best picture of available work area was
constructed. We conducted some experiments using the above protocol
with adult hemiparetic stroke subjects; FIG. 10 shows an example of
the results obtained. FIG. 10A shows work areas a left hemiparetic
subject was able to achieve with their impaired limb (the x-axis
has been inverted for comparison with the right side). As active
limb support increases, there is a marked decrease in available
work area. FIG. 10B shows the results from their unimpaired side.
In the unimpaired side, there was no effect of limb support (FIG.
10B). However, in the impaired side there was a clear cost for
increased levels of active limb support, and therefore shoulder
abduction torques (FIG. 10A). Work area decreased significantly as
subjects were required to lift more of their limb weight; this was
a direct effect of decreases in elbow extension range as
illustrated on FIG. 11 for twelve moderate-to-severely impaired
stroke subjects, with elbow flexion capabilities preserved across
support levels. FIG. 11 compares the unimpaired (left bar, blue)
and impaired (right bar, red) limb work areas, normalized to the
table supported condition revealed a significant difference between
limbs in all levels of support (*p<0.05; **p<0.001).
[0227] The results were obtained at the horizontal plane at 90
degrees of shoulder abduction. This protocol can be extended into
inclined and declined planes in the same way as the ballistic
reaching protocol. Using the information collected by the
ACT.sup.3D, a complete characterization of available shoulder and
elbow angle and torque combinations and resulting 3-D workspace
during different levels of support and at different plane
inclination angles were constructed for each subject.
Example III
Therapeutic Intervention
[0228] Under the guidance of our gravity-induced discoordination
concept, several therapeutic intervention protocols can be
realized. A protocol in which gravity was progressively
re-introduced during outward reaching over a period of eight weeks
was used. More specifically, subjects were trained to reach outward
towards virtually displayed targets that were situated on a
horizontal plane associated with the height of the shoulder. The
targets required near full active range of motion to acquire. The
key component of the training was the progressive re-introduction
of gravity as the subject improved in their capacity to acquire
each target. Furthermore, gravity was re-introduced in an
aggressive fashion. Subjects performed three sets of ten
repetitions to each of five targets under the amount of gravity or
limb support that maximally challenged them to reach at least 50%
of the distance toward the target. Each target ultimately had its
own gravity setting and appeared to progress at its own rate in
terms of increasing the gravity setting on the following session as
the subject was able to reach farther toward the target. The effect
of gravity re-introduction training on motor performance is studied
by comparing this protocol with a similar protocol where active
support of the limb against gravity is not required. In this
comparison group, subjects reach toward the same targets but are
supported by a haptically rendered horizontal surface associated
with the height of the shoulder. Each group participates in eight
weeks of training at a frequency of three sessions per week. By
comparing group performance on the metrics mentioned in the
previous section, the effect of re-introducing gravity is
elucidated. An example of a positive effect of the gravity
reintroduction protocol is shown for a single subject in FIG. 12.
FIG. 12 shows hand path traces pre (FIG. 12A) and post (FIG. 12B)
training, with three support levels shown. Note the increase in
work area at 175% of limb weight using the work area measurement
discussed earlier.
Example IV
Interface Between the Paretic Limb and the HAPTICMASTER
[0229] The forearm of the stroke survivor was placed on a support
splint that was attached via a three degree-of-freedom (3-DOF)
gimbal axis system to the end of the robot. The elbow was lined up
with the vertical axis of the gimbal (see FIG. 6). A 6-DOF load
cell was placed between the support splint and the gimbal to
measure the shoulder abduction/adduction and internal/external
rotation torques. Furthermore, the gimbal was instrumented with
position sensors used to measure the elbow and shoulder rotation
angles. The mechanical interface was designed to constrain
movements to a plane such that the individual moves (reaching or
retrieving) his/her arm and hand in line with the shoulder. The
inclination of the plane was adjustable from the user interface to
different angles, resulting in upward or downward arm movements
with respect to the shoulder. This accommodated individuals with
limited shoulder abduction angles and allowed for a progression to
greater shoulder abduction angles over various training sessions.
Support splints were designed for the left and the right arm and
were made in a small and large size to accommodate individuals of
different sizes (see FIG. 6). The support splint can be attached
and detached from the robot using a quick release type system. The
position of the hand and wrist in the splint was such that it
reduces spastic activity in hand and wrist flexors. Finally, the
hand and forearm were secured to the splint using broad, stiff
VELCRO straps.
Example V
Interface Between HAPTICMASTER, Biodex Experimental Chair, and
Visual 3-D Display
[0230] In this Example, the robotic device was designed into a
single unit integrating three separate systems: a Biodex chair, a
HAPTICMASTER device and a standard 19'' LCD screen (Dell, Round
Rock Tex.) controlled via a computer interface. The system was a
self-contained unit that could be easily operated by therapists and
physicians both for patient testing setup and for designing and
running therapeutic sessions.
[0231] The HAPTICMASTER device was placed on a T support track that
allows a Biodex chair and 3-D robot to move relative to each other
and to rotate (FIG. 5 and FIG. 15). This enabled proper alignment
between the robot and the arm of the individual. The HAPTICMASTER
device and its controller box were consolidated into one unit to
reduce the overall size of the setup. Finally, a standard 19''
screen (Dell) was attached on a mobile multi-jointed arm that was
connected to the back of the Biodex chair. This allowed flexibility
in the positioning of the 3-D screen in the visual field of test
participants.
Example VI
User Interface and Visual Virtual Reality Software
[0232] The robotic device was controlled via a custom user
interface developed to allow simple operation of the device. The
user interface software allowed control of the robotic device
during reaching movements and specification of the training and
evaluation protocols.
[0233] In order to create a realistic environment for individuals
with stroke, 3-D visual feedback was incorporated into the system.
In this example, a standard 19'' LCD screen that provided a 3-D
image based on the principle of autostereoscopic 3D imaging is
used. The position signals of the robotic device combined with the
segment lengths of the arm were used to estimate the paretic arm
configuration. These configuration coordinates drive the position
of a virtual arm/hand on the 3D screen in real-time. When designing
a 3D model of the arm, a standard modeling package like Kinetix's
3D STUDIO MAX may be used. In this example, the model was driven in
Windows XP environment using the Win3D Library, compatible with the
DTI technology. The virtual visual environment also included the
plane of the virtual table, movement targets and realistic objects
that can be lifted and moved to targets like a video game. The
intent was to create a stimulating and realistic environment that
would motivate our stroke consumers to use their paretic arm and to
teach them how to progressively overcome the negative effect of
gravity in a functionally meaningful way.
Example VII
Additional Exemplary Embodiments
[0234] As noted above, other configurations of the haptics device
are possible. FIG. 15 shows one such example of a 3-D haptic device
concept where the axes are fully independent.
[0235] Another example that is far more complex is a Stewart
platform shown in FIG. 8. This is used extensively in Flight
Simulators to provide motion sensation to the pilots. The hexapod
structure uses six actuators, and is thus capable of providing
6-DOF motion. In our situation the end effector and associated
additional hardware could be mounted on the top platform.
[0236] FIG. 9 shows a prototype of 3-D visual feedback displaying
an avatar of the subject's arm (limb bones 63; movable joints 64)
on a table (65). The tip of the virtual hand is placed in a
starting position (66) and the subject is asked to move the hand
(hand path 67) to one of the targets (68). The red boundary (69) is
the maximum reaching distance of the subject's hand based on
measured forearm/upper arm segment lengths. This boundary is used
as visual feedback to encourage subjects to reach as far as they
can to determine the arm active work area. This can be repeated for
different planes to determine the paretic arm's workspace.
[0237] Those skilled in the art will appreciate that various
adaptations and modifications of the just-described embodiments can
be configured without departing from the scope and spirit of the
invention. Other suitable techniques and methods known in the art
can be applied in numerous specific modalities by one skilled in
the art and in light of the description of the present invention
described herein. Therefore, it is to be understood that the
invention can be practiced other than as specifically described
herein. The above description is intended to be illustrative, and
not restrictive. Many other embodiments will be apparent to those
of skill in the art upon reviewing the above description. The scope
of the invention should, therefore, be determined with reference to
the appended claims, along with the full scope of equivalents to
which such claims are entitled.
* * * * *