Reinforced Feedback in Virtual Environment

Reinforced Feedback in Virtual Environment Facilitates the Arm Motor Recovery in Patients after a Recent Stroke Lamberto Piron, Paolo Tombolini, Andrea Turolla, Carla Zucconi, Michela Agostini, Mauro Dam, Giovanna Santarello, Francesco Piccione and Paolo Tonin

Abstract— Previous studies have shown that the motor training in a virtual-environment with the augmented feedback promotes motor learning in normal subjects and in long-term post-stroke patients. We evaluated whether this approach could be useful also for treating patients with arm motor deficits due to a recent stroke. Thirty-eight patients were included in the study within 3 months from an ischemic stroke in the territory of the middle cerebral artery. Twenty-five subjects received training with the Reinforced Feedback in Virtual Environment (RFVE) therapy for the arm, and thirteen patients received an equal amount of a conventional rehabilitation (CR) therapy focused to the upper limb. Before and after therapy, the autonomy of daily living activities were assessed with the Functional Independence Measure (FIM) and the degree of motor impairment was measured with the Fugl-Meyer scale for the upper extremity (FM-UE). The RFVE therapy group showed significant improvements in the FM-UE and the FIM scale mean scores. The conventional therapy determined smaller and not statistically significant scores improvements. These data indicate that the recovery of arm motor function in patients after a recent stroke appear to be speeded up by an augmented feedback provided in a virtual-environment.

with enhanced feedback promotes the learning of complex multi-join motor task, better than conventional methods [7]. Preliminary reports have indicated that a virtual-reality based therapy improves upper-limb motor performance in the chronic phase after stroke [8] – [11]. Along this line, we evaluated whether this approach could be useful also for treating patients with arm motor deficits due to a recent stroke. II.

METHODS

Thirty-eight patients were included in the study; they were suffering from mild/intermediate arm motor impairments due to an ischemic stroke, occurring within three months before the study. All patients presented a mild to intermediate arm impairment. The demographic characteristics of the 38 patients are summarized in Table I. TABLE I GROUPS’ DEMOGRAPHIC CHARACTERISTICS

All patients (n.=38)

I. INTRODUCTION

P

REVIOUS evidence on healthy subjects has demonstrated that appropriate feedback on the nature of the movement (knowledge of performance) and some variables of outcome (knowledge of results) may help to temporarily or permanently improve motor performance [1] – [6]. In this regard, computer-based systems can provide subjects with artificially-enhanced feedback that may facilitate the acquisition of new motor skills. Todorov and co-workers reported that training in virtual environments

Manuscript received April 23, 2007. L. Piron is with the IRCCS San Camillo Hospital, via Alberoni 70, 30126 Lido di Venezia, Venezia, Italy (+39 041 2207214; fax: +39 041 731330; e-mail: [email protected] ). P. Tombolini as Ordinary Associate of A.I.FI. (Associazione Italiana Fisioterapisti)) (e-mail: [email protected]). A. Turolla is with the IRCCS San Camillo Hospital, via Alberoni 70, 30126 Lido di Venezia, Venezia, Italy (e-mail: [email protected]). C. Zucconi is with the IRCCS San Camillo Hospital, via Alberoni 70, 30126 Lido di Venezia, Venezia, Italy (e-mail: [email protected]). M. Agostini is with the IRCCS San Camillo Hospital, via Alberoni 70, 30126 Lido di Venezia, Venezia, Italy (e-mail: [email protected]). M. Dam is with the Dept. of Neurology and Psychiatry, University of Padova, via Giustiniani 5, 35218 Padova, Italy ([email protected] ). G. Santarello is with the IRCCS San Camillo Hospital, via Alberoni 70, 30126 Lido di Venezia, Italy (e-mail: [email protected]). F. Piccione is with the IRCCS San Camillo Hospital, via Alberoni 70, 30126 Lido di Venezia, Italy (e-mail: [email protected] ). P. Tonin is with the IRCCS San Camillo Hospital, via Alberoni 70, 30126 Lido di Venezia, Venezia, Italy (e-mail: [email protected]).

1-4244-1204-8/07/$25.00 ©2007 IEEE

121

RFVE Group Age, years ± SD Sex, M/F Stroke to rehabilitation interval, months ± SD Control Group Age, years ± SD Sex, M/F Stroke to rehabilitation interval, months ± SD

61.5 ± 9.4 17/8 2.5 ± 1.5

61.2 ± 6.6 8/5 2.6 ± 1.6

Twenty-five subjects (RFVE Group) received Reinforced Feedback in Virtual Environment (RFVE) therapy for the arm, and thirteen patients (Control Group) received an equal amount of a conventional rehabilitation (CR) therapy focused to the upper limb. Before and after therapy, the autonomy of daily living activities were assessed with the Functional Independence Measure (FIM) and the degree of motor impairment was measured with the Fugl-Meyer scale for the upper extremity (FM-UE). For both groups the therapy lasted from 5 to 7 weeks, 1 hour daily for five days a week. The equipment used for the RFVE therapy consisted of a PC workstation, a high-resolution LCD projector, a 3D motion-capture system (Polhemus 3Space Fastrack, Vermont, U.S.A.), and a dedicated software, developed at Massachusetts Institute of Technology U.S.A., for processing data and creating the virtual environment.

The movement-recording system consisted of up to three Table II, fig.2 and fig.3 show the clinical effects of RFVE magnetic receivers positioned on the arm, forearm, and on therapy and of CR. the object handed by the subject (end-effector). The latter sensor continuously measured the motion of the end effector in the arm workspace. The static accuracy of the position signal was 0.76 mm RMS and 0.15 degrees RMS for orientation. Resolution was 0.0005 cms/cm and 0.025 degrees/degree of range. Latency was 4 msec unfiltered from the center of receiver measurement period to the beginning of transfer from the output port. The sampling rate was 120 Hz, divided by the number of the receivers: using one receiver the sampling rate was 120 Hz. During RFVE therapy, the subject was seated in front of the wall screen grasping a real object, and the physical therapist is beside him. The software created the virtual environment consisting of a number of handling and target objects, such as an envelope and a mailbox, a hammer and a nail, a glass and a carafe, respectively. Virtual handling object matched the real object held by the subject. While Fig.1. The setup for RFVE therapy. The patient holds an object in performing virtual tasks, such as putting the envelope in the which is applied the magnetic receiver. mailbox, hitting the nail, or pouring the glass in the carafe, the subject moves the real envelope, hammer, or glass and see on the screen the trajectory of the corresponding virtual TABLE II CLINICAL EFFECTS OF RFVE THERAPY AND OF CR IN POST-STROKE object toward the virtual mailbox, nail, or carafe (fig. 1). PATIENTS During performance, patients could see on the screen not only their own movement but also the correct trajectory that Before Therapy After Therapy they had to execute, prerecorded by the physical therapist. RFVE Group This setting allowed subjects to easily perceive motion Clinical parameters errors and adjust them during the task. Once subjects FM-UE score 43.7 ± 8.3 51.4 ± 9.8* completed adequately the required task, the system provided FIM score 93.4 ± 28.6 110.2 ± 13.9* Control Group a rewarding signal. The physical therapist can increase or Clinical parameters decrease the complexity of the exercise by modifying the 45.4 ± 9.3 42.0 ± 12.3 FM-UE score position of the target object or the pathway to reach it. For 95.9 ± 28.3 88.2 ± 30.0 FIM score instance the motor procedure required posting the envelope All values are indicated as, mean ± SD; (*) p”0.05. changed with the virtual position of the mailbox in the workspace (close or far) and/or with the orientation of the mailbox slot (vertical, diagonal, or horizontal). This strategy allowed patients to take advantage of executing a large variety of goal oriented movements tailored to their individual motor deficits. Both groups underwent a CR program for lower limb and balance impairments, aphasia or others cognitive deficits. Anamnestic or clinical evidence of cognitive impairments, neglect, apraxia and aphasia interfering with verbal Fig. 2. The graphics show the mean differences pre and post treatment in comprehension were all considered exclusion criteria. RFVE group, respectively in the FM–UE and FIM scale; (*) p”0.05. The Wilcoxon test was used to determine the statistical significance of the differences in the FM-UE [12] and FIM [13] scale mean score before and after therapy in each treatment group; the Mann-Whitney U test was used to evaluate the significance of the differences in the outcome measures between CR and RFVE treated patient group, at the beginning and at the end of the therapy. Statistical significance was considered at p ≤ 0.05. III.

RESULTS

No patient complained of any discomfort caused by the interaction with the “virtual world” in our setup.

122

Fig. 3. The graphics show the mean differences pre and post treatment in CR group, respectively in the FM–UE and FIM scale.

RFVE therapy showed statistical significant improvements in the Fugl-Meyer upper extremity and the Functional Independence Measure scale mean scores over baseline values: 17.6% and 17.9% respectively (p≤0.001; p≤0.001). The CR determined smaller and not statistically significant scores improvements: 8.1% and 8.7% respectively. There wasn’t statistical significance in the differences of the outcome measures between CR and RFVE therapy before and after the therapy. IV.

These data indicate that the recovery of arm motor function in patients after a recent stroke is promoted by an augmented feedback strategy, which is applied through a virtual-environment in our setting [9]. Our preliminary data must be confirmed with a larger subject group. REFERENCES [1] [2]

DISCUSSION

In this study, we have evaluated the safety and effectiveness of the RFVE therapy in subjects suffering from mild to intermediate motor impairment of the arm enrolled within three months after a cerebro-vascular injury. The fact that none of our patients complained of any discomfort due to interaction with the virtual word indicate that such a “non immersive” type of Virtual Reality is devoid of side effects that may hamper the process of learning. In the first period after a brain injury some neurological mechanisms intervene for the attempt of restoring spontaneously the brain functions. In the immediate postacute phase of stroke, indeed, motor function improvement is mostly attributed to the phenomenon of the “spontaneous recovery” and, only in part, to rehabilitation therapies. However, in our study, the process of motor recovery after a recent stroke seems to take advantage from the rehabilitation techniques based on the augmented information for the Central Nervous System (CNS). On the other hand, the CR group shows a little, not statistically significant, improvement of the motor function that may be attributed to the combination of spontaneous recovery and conventional physical therapy. It could be the case that some rehabilitation strategies are not so synergic with the spontaneous recovery in the early stage after the CNS vascular lesion.

123

[3] [4] [5] [6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

Gentile AM. A working model of skill acquisition with application to teaching. Quest. 1972; 17: 3-23. Khan MA, Franks IM. The effect of practice on component submovements is dependent on the availability of visual feedback. J Motor Behav. 2000; 32: 227-240. Newell KM, Carlton MJ. Augmented information and the acquisition of isometric task. J Motor Behav. 1987; 19: 4-12. Young DE, Schmidt RA. Augmented kinematics feedback for motor learning. J Motor Behav. 1992; 24: 261-273. Winstein CJ. Knowledge of results and motor learning implications for physical therapy. Phys Ther. 1991; 71: 140-149. Woldag H, Hummelsheim H. Evidence-based physiotherapy concepts for improving arm and hand function in stroke patients. J Neurol. 2002; 249(5): 518-528. Todorov E., Shadmer R., Bizzi E. Augmented feedback presented in a virtual environment accelerates learning of a difficult motor task. J Motor Behav. 1997; 29: 147-158. Holden M, Todorov E, Callahan J, Bizzi E. Virtual Environment training improves motor performance in two patients with stroke: case report. Neurology Report. 1999; 23: 57-67. L. Piron, M. Dam, E. Trivello et al. Virtual Environment Training Ameliorates Motor Deficits in Post-Stroke Patients. Neurology 1999;52(2):A138. Merians AS, Jack D, Boian R, Tremaine M, Burdea GC, Adamovich SV, Recce M, Poizner H. Virtual reality-augmented rehabilitation for patients following stroke. Phys Ther. 2002; 82: 898-914. Piron L., Tonin P., Piccione F., Iaia V., Trivello E. and Dam M. “Virtual environment training therapy for arm motor rehabilitation” Presence 2005; 6 (14):732-740. Fugl-Meyer A.R., Jaasko L., Leyman I., Olsson S., Steglind S. The post-stroke hemiplegic patient: a method for evaluation of physical performance. Scandinavian Journal of Rehabilitation Medicine. 1975; 7: 13-31. Keiith R., Granger C., Hamilton B., Sherwin. The functional independence measure: a new tool for rehabilitation. In: Eisenburg M.G., Grzesiak R.C., eds. Advances in Clinical Rehabilitation. New York: Springer Publishing Co. 1987: 6-18.