UPPER-EXTREMITY
Dejan Popović1,3 and Mirjana Popović1,2
1Faculty
of Electrical Engineering, 11000 Belgrade, Bulevar revolucije 73, Yugoslavia
2Institute for Medical Research, 11000 Belgrade, Dr
Subotića 5, Yugoslavia,
3Center
for SMI, Aalborg University, Frederik Bajers Vej 7, D-3, 9220 Aalborg, Denmark
Abstract: Functional electrical stimulation (
1. INTRODUCTION
In 1996 a workshop was organized to assess the state of the art in the
field of grasping neuroprosthesis (NP). This meeting concluded [29] that
grasping systems [24,25] have been developed to the
stage of being approved by the very strict guidelines (
Evaluation of three different grasping and reaching systems provides some interesting data [15-17]. The functional independence measure (FIM), the quadriplegia index of function (QIF), upper extremity functioning test (UEFT), range of movements, and strength of grasping have been followed for at least six months. It was found that all three evaluated systems improve the functioning. FIM and QIF were increased for almost 100 percents in subjects who started with a low value, but only a little for subjects who retained some level of independence. UEFT scores improved for all functions tested, but the handling of small objects (e.g., finger food) remained a problem. Range of movements and the strength of grasp were greatly increased in all subjects. The changes started to be substantial about eight weeks after starting to use NPs. The final scores for all measures show that even with the system working properly, the independence remained limited. The frequency of using of NP, for all 27 subjects evaluated, shows variability, but in general implies to relatively short periods of its usage. An issue for assessing the relative importance of NP to the functioning is that only seven subjects continued to use it for daily activities once the study was terminated, even though the NPs were given at no cost. Four other subjects continued to use NP for therapy, 30 minutes in the morning, claiming reduced spasticity after being stimulated. Only one subject, from the eight, who participated in evaluation of the reaching system, continued to use it on a daily basis [17].
Stopping to use of a NP relates most probably to the substantial improvements of functioning without NP. Prolonged, daily, externally elicited activity of neuro-musculo-skeletal structures is responsible for those improvements [15-17]. This suggests that therapeutical effects of electrical stimulation of paralyzed muscles are substantial affecting both the peripheral system (e.g., stronger muscles, increased range of movements, decreased spasticity) and the functional organization of CNS.
2. NATURAL CONTROL OF
GOAL-DIRECTED MOVEMENTS
The difficulties in using NPs are: 1) command interface is not intuitive
enough, and it requires conscious actions; and 2) control is not flexible and
adaptive. Eliminating the said problems was the main motivation for improving
control, the bottleneck for better NPs.
Some common principles of motor control organization in organisms [1,9] may serve as a useful guideline for the design of control algorithms. The fact that goal-directed movements are reproducible for a given task in different living species implies that optimization of some sort is taking place during learning of such motions. This optimization is inherent to the self-organization.
When activated to perform a self-paced, non-constrained functional motion, the neuro-muscular system prefers some trajectories from other. The constraint imposed by the task determines the trajectory (e.g., straight line movements are adequate for fast pointing movements; starting the movement with the proximal joints and adjusting the position of the hand while moving distal segments is effective for precise positioning, etc.).
Skill acquisition processes are optimized in such a way that the open loop control is preferred to the closed-loop control as long as it does not affect the performance of a functional motion. Invariant of functional motions, motor patterns, automatic movements and reflexes are the means by which the use of the open-loop control mode is extended in the execution of motor acts. Open-loop control sets free the vision system and the conscious level to monitor the environment and prepare the organism for new activities. The execution of skilled movements is organized that the flow of control information from the higher levels is minimized. Lower control level dispose, thus, with maximum autonomy while performing a skilled motor act [23].
3. ARTIFICIAL CONTROL OF
GOAL-DORECTED MOVEMENTS
The control proposed for NP is based on the above stated principles. We originally synthesized control of reaching by using nonanalytical, hierarchical controller for the elbow joint movements in humans with CNS lesion [15]. The implemented strategy uses the principle of synergy found in able-bodied subjects, which allows persons with disability to plan the movement in terms of the most distal segment, and actually volitionally move the most proximal segment [1,5,9]. When reaching, the movement is planned in the external reference frame based only on the visual information of the initial position of the hand and the position of the target [5,26,27]. The visual information is automatically transferred to the universal, that is, user non-specific constraint between the shoulder and elbow joints [26]. Hence, when a person moves volitionally his/her upperarm, the elbow joint angular velocity profile is constrained. An approximation of this constraint is so called scaling law [14,16]. The final step is the usage of the mapping at the lowest, actuator level. The actuator for assisting the elbow extension is m. Triceps Brachii, and its mapping connects the joint angular velocity with the parameters of electrical stimulation. The principle of this operation follows the original ideas of extended physiological proprioception [23].
An
appealing version of the controller, which can be used for the control of
goal-directed movements, has been suggested [12]. The control that we are
developing (Fig. 1) for goal directed movements is presented.
The algorithm at the top includes a decision level (Fig. 1). At this level the user selects the task and voluntarily triggers the appropriate interface. Two activities will be activated simultaneously; one (prehension) ensuring that the hand is properly oriented for the selected task [1,9], and the second to bring the hand to the appropriate position (reaching). The reaching control will rely on synergistic behavior of neighboring joints, while the prehension has to be an open-loop type activity. At this point the prehension allows only lateral and palmar grasps.
The coordination level is a sampled data, discrete feedback system. In this system a rule-based structure is to be implemented. Machine learning can generate rules, that is the knowledge base. Once the control signals are generated, they are translated to the lower execution level. The execution level must include customized model, because of the nonlinearities and time variations of the neuro-musculo-skeletal system. These models are not existing in a ready to use form, but there are some valuable developments on the way.
Literature
1.
M.
A. Arbib, T. Iberall and D. M. Lyons, Coordinated control programs for movements
of the hand, In: A. W. Goodwin and I. Darian-Smith (Eds) Hand Function, Springer Verlag, Berlin, pp 111-129, 1985.
2.
P.E.
Crago, W. D. Membaerg, K. K. Usey, M. W. Keith, R. F. Kirsch, G. J. Chapman, M.
A. Katorgi and E. J. Perreault, An elbow extension neuroprosthesis for
individuals with tetraplegia, IEEE Trans
Rehab Eng, TRE-6:1-6, 1998.
3.
C.
J. Feng and A. F. T. Mak, Three dimensional motion analysis of the voluntary
elbow movement in subjects with spasticity, IEEE
Trans Rehab Eng, TRE-5: 253-262, 1997.
4.
T.
Fujiwara, Y. Hara, K. Akabochi and N. Chino, Relationship between shoulder
muscle strength and functional independence measure score among C6
tetraplegics, Spinal cord, 37:58-61,
1999.
5.
A.
P. Georgopoulos, On reaching, Ann Rev
Neurosci, 9:147-170, 1986.
6.
R.
L. Hart, K. L. Kilgore and P. H. Peckham, A comparison between control methods
for implanted FES hand-grasp systems, IEEE
Trans Rehab Eng, TRE-6: 208-218, 1998.
7.
A.
E. Hines, P. E. Crago and C. Billian, Hand opening by electrical stimulation in
patients with spastic hemiplegia, IEEE
Trans Rehab Eng, TRE-3:193-205., 1995.
8.
J.
Hollander and P. H. Peckham, Functional neuromuscular stimulation for combined
control of elbow extension and hand grasp in C5 and C6 quadriplegics, IEEE Trans Rehab Eng, TRE-6:190-199,
1998.
9.
T.
Iberall and C. L. MacKenzie, Opposition space and human prehension, In: S. T.
Venkataraman and T. Iberall (Eds) Dexterous
robot hand, Springer Verlag, New York, pp. 32-54, 1990
10. N. Lan, H. Q. Feng and P. E.
Crago, Neural network generation of muscle stimulation patterns for control of
arm movements, IEEE Trans Rehab Eng,
TRE-4:213-224, 1994.
11. M. A. Lemay and P. E. Crago,
Closed loop wrist stabilization in C4 and C5 tetraplegia IEEE Trans Rehab Eng, TRE-5:244-252, 1997.
12. P. Lima and G. N. Saridis, Intelligent
controllers as hierarchical stochastic automata, IEEE Trans Syst Man and Cybern, Part B, SMC-29:151-163, 1999.
13. P. S. Lum, C. G. Burgar, D.
E. Kennay and H. F. M. V. der Loos, Quantification of force abnormalities
during passive and active assisted upper-limb reaching movements in post-stroke
hemiparesis, IEEE Trans Biomed Eng,
BME-46:652-662, 1999.
14. D. Popović and M.
Popović, Tuning of a nonanalytical hierarchical control system for
reaching with FES, IEEE Trans. Biomed.
Eng., BMI-45:203-212, 1998.
15. D. Popović, A.
Stojanović, A. Pjanović, S. Radosavljević, M. Popović, S.
Jović and D. Vulović, Clinical
evaluation of the Bionic Glove, Arch Phys Med Rehabil, 80:299-304, 1999.
16. M. Popović and D.
Popović, A new control for reaching in tetraplegics, J Electromyog Kinesiol, 4:242-253, 1994.
17. M. Popović and D.
Popović, Belgrade grasping system, Electronics, 2:21-28, Banja Luka, 1998.
18. D. J. Reinkensmayer, J. P.
A. Dewald and W. Z. Rymer, Guidance-based quantification of arm impairment
following brain injury: A pilot study, IEEE
Trans Rehab Eng, TRE-7:1-11, 1999.
19. S. Saxena, S. Nikolić
and D. Popović, An EMG-controlled grasping system for tetraplegics, J Rehab Res Develop, 32:17-24, 1995.
20. J. J. Sarver, B. T. Smith,
R. Selitkar M. J. Mulcahey and R. R. Betz, A study of shoulder motions to
control source for adolescents with C4 level SCI, IEEE Trans Rehab Eng, TRE-7:27-34, 1999.
21. T. R. D. Scott, P. H.
Peckham and K. L. Kilgore, Tri-state myoelectric control of bilateral upper
extremity neuroprosthesis for tetraplegic individuals, IEEE Trans Rehab Eng, TRE-4:251-263, 1996.
22. S. Sennels, F.
Biering-Soerensen, O. T. Anderson and S. D. Hansen, Functional neuromuscular
stimulation control by surface electromyographic signals produced by volitional
activation of the same muscle: adaptive removal of the muscle response from the
recorded EMG-signal, IEEE Trans Rehab Eng,
TRE-5:195:206, 1997.
23. D. C. Simpson, The choice of
control system for the multi-movement prosthesis: extended physiological
proprioception (EPP), In P. Herberts (Ed.) The
control of upper-extremity prostheses and orthoses, C. Thomas, Chapter 15,
1974.
24. B. T. Smith, M. J. Mulcahey
and R. R. Betz, Development of an upper extremity FES system for individuals
with T4 tetraplegia, IEEE Trans Rehab Eng,
TRE-4:264-270, 1996.
25. B. T. Smith, Z. Tang and M.
W. Johnson, An externally powered multichannel, implantable stimulator-telemeter
for control of muscles, IEEE Trans Biomed
Eng, BME-45:463-475, 1998.
26. J. F. Soechting and M.
Flanders, Moving in three dimensional space: frames of reference. vectors and
coordinate systems, Annu Rev Neurosci.
5:167-191, 1992.
27. J. F. Soechting and M.
Flanders, Evaluation an integrated musculoskeletal model of the human arm, J Biomech Eng, 119:93-102, 1997.
28. C. A. Stanger, C. Anglin, W.
S. Harwin and D. P. Romily, Devices for assisting manipulation: A summary of
user task priorities, IEEE Trans Rehab
Eng, TRE-2:256-265, 1994.
29. R. Triolo, R. Nathan, Y.
Handa, M. W. Keith, R. R. Betz, S. Carroll and C. Kantor, Challenges to
clinical deployment of upper limb neuroprostheses, J Rehab Res Develop, 33:111-122, 1996.
30. X. Wang, M. Maurin, F. Mazet, N. C. Maia, K. Voinot, J. P. Verriest and M. Fayet, Three-dimensional modeling of the motion range of axial rotation of the upper arm, J Biomech, 31:899-908, 1998.
Acknowledgement: The work on this project was partly supported by the Ministry for Science of Serbia, Belgrade, and the Danish National Research Foundation, Copenhagen, Denmark