Professional Documents
Culture Documents
RESEARCH PAPER
Abstract
Objective. To assess the use of a video-capture projected VR system for children with acquired brain injury (ABI), and to
compare their performance to that of matched healthy controls.
Subjects. Thirty-three children (age range: 6–11.4 years) were divided into two groups: 17 children with ABI and 16
controls matched for age, gender and maternal education.
For personal use only.
Methods. Participants experienced three video-captured virtual environments and completed the SFQ child at the end of
each specific environment. Participants with ABI experienced three virtual reality (VR) sessions over a period of 10 days.
Attention was evaluated using the TEA-Ch. Upper extremity motor abilities were evaluated with the Melbourne Assessment,
and self-care abilities were evaluated with the PEDI.
Results. The video-capture projected system differentiated between the performance of children with ABI and control
participants. There was a correlation between VR performance and some attention factors and self-care abilities. No
significant correlations were found between performance in the Melbourne assessment and performance within VR.
Conclusions. The results highlight the potential of the video-capture virtual reality as a tool in the rehabilitation process of
children with ABI.
Keywords: Children, acquired brain injury, virtual reality, attention, self-care abilities
Correspondence: Orit Bart, Department of Occupational Therapy, Tel Aviv University, Tel Aviv, Israel. E-mail: o_bart@bezeqint.net
ISSN 0963-8288 print/ISSN 1464-5165 online ª 2011 Informa UK, Ltd.
DOI: 10.3109/09638288.2010.540291
1580 O. Bart et al.
activities to client’s abilities in a manner that will user-friendly assessment tool to aid diagnosis of
enable them to be involved in the activity and to ADHD, and more acceptable to children than
improve their functioning without causing frustration the conventional Test of Variables of Attention
or fatigue [10]. Empirical evidence points to (TOVA) [28].
activities that require a higher level of interaction Although these initial results are encouraging, to
with the environment as being more successful than date, the number of VR studies in paediatric
those that require low levels of interaction [11,12]. populations is quite limited, and none of them
For children, play is a meaningful and motivating investigated the feasibility of this technology for
activity that encourages social participation and children with ABI. This study was conducted to
cooperation [13]. In environments that encourage assess the use of a video-capture VR system for
play children can develop their motor, cognitive and children with ABI, and to compare their perfor-
social skills [14]. Mortenson and Harris [7] sug- mance to that of matched healthy controls.
gested incorporating play strategies in the interven- In addition, the relationships between the chil-
tion process for children suffering from ABI. dren’s performance within the virtual environments
During the last decade, various virtual reality (VR) and their performance in well-known tests of
technologies have been used in rehabilitation for attention, upper extremity motor skills and self-care
Disabil Rehabil Downloaded from informahealthcare.com by Mcgill University on 10/27/14
both evaluation and intervention [15–17]. VR has activities were examined. This was done in order to
many advantages for intervention, such as enabling explore the potential of this VR technology as a tool
the grading of activities, obtaining precise perfor- for improving these skills, which are necessary for
mance measures, providing a safe and ecologically their participation in daily life activities.
valid environment and being enjoyable and motivat-
ing [18]. VR also provides a distracting effect that
helps the participant focus on the task rather than on Methods
painful or unpleasant medical procedures [19,20].
The familiarity of young people with computer Participants
For personal use only.
Gender
Male 7 41.2 7 43.8
Female 10 58.8 9 56.3
Maternal education
Middle school 2 11.8 1 6.3
High school 8 47.1 7 43.8
Post-secondary education 6 35.3 7 43.8
Academic 1 5.9 1 6.3
Computer experience
computer 17 100 16 100
Dance pad/play station 3 17.6 6 37.5
VR system (projected/HMD) 0 0 0 0
Disabil Rehabil Downloaded from informahealthcare.com by Mcgill University on 10/27/14
and one ependimoma) and one participant had an application does not use red glove mode since it does
infection. Fifteen participants had upper extremity not require reaching with the upper extremities).
dysfunction, six of them in the dominant limb, seven Performance was measured by the percentage of
in the non-dominant limb and two in both limbs. successfully avoided obstacles out of the total
number of obstacles.
Virtual reality system. Gesture Tek’s GX Interactive ing different components of attention: selective
Rehabilitation and Exercise system (IREX) was used attention, sustained attention, divided attention and
to provide the virtual reality experience [30,31]. In alternating attention. In this study, we used the short
this study, the user sat in front of a large screen onto screening version that includes the first four subtests
which the VR environment was projected. In two of assessing the four attention components. The test
the virtual environments, the system was pro- measures performance and completion time. The
grammed to respond only to red objects (red glove TEA-Ch has age-related norms and is reliable (test–
mode) such as a glove worn on one or both hands. In retest r’s ¼ 0.65–0.87) and valid [33]. The scores are
this mode, the system responded only to arm converted into standardised scores to each age
movements and not to movement of other body group.
parts. Three virtual environments were used in this
study: Melbourne assessment of unilateral upper limb function
[34]. This test was used to assess motor abilities. It
Birds and balls. The user views himself in a pastoral scores the quality of unilateral upper limb functional
setting. He has to touch virtual balls emerging from movements using activities involving reach, grasp,
different locations on the screen and flying towards release and manipulation. The maximal score is 122
him. Performance in this environment was measured points, which indicates full function of the upper
as the percentage of successfully touched balls out of limb. The scores are converted into percentiles. The
the total number of balls that appeared on the screen, test has high internal consistency (a ¼ 0.96) [35] and
and by response time (i.e. the elapsed time from a was validated with the Paediatric Evaluation of
ball’s first appearances on the screen until it was Disability Inventory (PEDI) and correlations were
touched). found to be high (r’s ¼ 0.72–0.94) [36].
Soccer. The user is a goalkeeper and has to prevent Paediatric evaluation of disability inventory: PEDI [37].
the balls, which emerge from different locations, This clinical assessment was used to evaluate
from entering the net. Performance in this environ- functional capabilities, performance and changes in
ment was measured in terms of percentage of functional skills in children with disabilities. In this
successfully repelled balls out of the total number study, 73 activities related to self-care, e.g. eating,
of balls. grooming, dressing and personal hygiene were
queried. The Caregiver Assistance Scale (CAS)
Snowboard. The user is required to avoid obstacles related to self-care, which measures the amount
as he skis downhill, by leaning from side to side; this and type of assistance the child needs, was also used
1582 O. Bart et al.
[37]. The PEDI has internal reliability and test– abilities, upper extremity function and self-care
retest reliability (r ¼ 0.67–1.00). It has good content capabilities were assessed within 10 days of the VR
and construct validity [38]. session.
Results
Procedure
Statistically significant differences in VR perfor-
This study was approved by the Sheba Medical mance between the ABI group and the control group
Centre Institutional Review Board. Informed con- were found in most of the measured variables. In
sent was obtained from the subjects’ legal guardians. general, participants in the control group performed
For personal use only.
Participants experienced all three virtual environ- better than the participants in the ABI group
ments: Birds and Balls, Soccer and Snowboard. At (Table II).
the start of each trial, the environment was intro- As shown in Table II, the success rate for Birds
duced, and participants were trained individually for and Balls was lower in the ABI group, but it was
a period of 1 min. The participant then experienced statistically significant only at level 3 (t ¼ 73.05,
3 min of the environment with the difficulty level p 5 0.01). The mean response time for Birds and
increasing every minute. For the Soccer task, which Balls in the ABI group was significantly longer than
is more complex, the first 2 min were maintained at in the control group at all three levels: level 1
level 1 (1a and 1b) and the third minute at level 2. (t ¼ 3.86, p 5 0.001), level 2 (t ¼ 2.80, p 5 0.09) and
Between each environment, participants had 2 min level 3 ( t ¼ 4.94, p 5 0.009).
rest. The total experience time in the three environ- Success rate in Soccer for the ABI group was
ments lasted for about 20 min. significantly lower than that of the control group, at
The SFQ-Child questionnaire was administered all levels: level 1a (t ¼ 73.04, p 5 0.01), level 1b
after experiencing each VR environment. Attention (t ¼ 72.29, p 5 0.05) and level 2 (t ¼ 73.73,
Table II. Differences between study and control groups in VR performance in the three environments (means and standard deviations).
Birds and Balls 1 Success rate (%) 98.72 (3.63) 100.0 (0.00) 71.39 1.73
Response time* 6.12 (1.28) 4.61 (0.87) 3.87 0.001
2 Success rate (%) 97.54 (5.11) 99.79 (0.80) 71.74 0.092
Response time* 5.84 (0.89) 4.89 (1.00) 2.81 0.009
3 Success rate (%) 94.59 (6.15) 99.39 (1.31) 73.06 0.005
Response time* 6.81 (1.00) 5.27 (0.74) 4.94 0.000
Soccer 1a Success rate (%) 55.70 (18.91) 74.36 (16.03) 73.05 0.005
1b Success rate (%) 57.41 (21.01) 73.75 (19.84) 72.29 0.029
2 Success rate (%) 42.23 (15.55) 61.62 (14.19) 73.73 0.001
Snowboard 1 Success rate (%) 85.71 (18.89) 94.64 ((8.84 71.72 0.095
2 Success rate (%) 80.14 (19.29) 95.31 (7.73) 72.93 0.006
3 Success rate (%) 85.29 (12.30) 98.12 (4.03) 3.97 0.000
Table III. Spearman’s correlation between performances within virtual environments and TEA-Ch test in the study group (n ¼ 16).
Focussed atten-
tion Alternating attention
VR environment Level Measure Score Time Sustained attention Score Time Divided ttention
Birds & Balls 1 Success rate (%) 0.22 0.29 0.49* 0.34 0.11 70.12
Response time 70.48 70.40 70.65** 70.55* 70.53* 70.25
2 Success rate (%) 0.22 0.34 0.45 0.16 0.13 0.02
Response time 70.32 70.33 70.40 70.41 70.36 70.11
3 Success rate (%) 0.42 0.38 0.22 0.34 0.37 70.08
Response time 70.50* 70.42 70.48 70.52* 70.52* 70.91
Soccer 1a Success rate (%) 0.49* 0.47 0.63** 0.45 0.47 0.30
1b Success rate (%) 0.39 0.36 0.43 0.39 0.53* 0.29
2 Success rate (%) 0.34 0.26 0.37 0.50* 0.49* 0.29
Snowboard 1 Success rate (%) 0.38 0.39 0.45 0.51* 0.21 0.00
2 Success rate (%) 0.16 0.15 0.53* 70.04 0.05 0.49*
3 Success rate (%) 70.05 0.04 70.10 70.24 70.20 0.06
Disabil Rehabil Downloaded from informahealthcare.com by Mcgill University on 10/27/14
Table IV. Spearman correlation between performance within the TEA-Ch scores and performance in the Soccer
virtual environments and PEDI test in the study group (n ¼ 17). environment (Soccer percent of success in level 1a
PEDI and general alternate attention rs ¼ 0.54, p 5 0.03;
VR (therapist Soccer percent of success in level 1b and general
environment Level report) Caregiver alternate attention rs ¼ 0.53, p 5 0.034). No signifi-
cant correlations were found between the TEA-Ch
Birds & Balls 1 Success rate (%) 0.11 0.42
For personal use only.
‘Soccer’ environment (z ¼ 72.15, p 5 0.05). These and motivation that reflect the components required
results show that the children with ABI enjoyed ‘Bird to engage in the VR environment. This may account
& Balls’ more than the children in the control group, for the stronger correlations between the two.
but had less understanding of either environment. In general, high levels of satisfaction following the
VR experiences were reported by both groups with
no statistical differences between them with regard to
Discussion the sense of presence and satisfaction, or for most
indices of the SFQ-Child questionnaire. These
The findings of this study indicate that on selected findings are in accordance with other IREX VR
tasks, the IREX video-capture VR system distin- studies [23,24,26,44]. In the ABI group, the highest
guished between children with and without brain sense of presence was reported for the snowboard
injury; the participants in the control group achieved environment, which was perceived as the easiest. The
higher success rates than did the participants in the sense of control in this environment was high for
ABI group. Statistically significant differences were both groups; 28 of 30 children preferred it over the
mostly obtained at the higher levels, where faster other two environments, asking to repeat it at the end
responses are required. For example, response times of the VR session. Note that Snowboard differs from
Disabil Rehabil Downloaded from informahealthcare.com by Mcgill University on 10/27/14
in the Birds and Balls environment were significantly the other two environments since its motor demands
longer for the ABI group. These findings are were lighter, focussing more on trunk tilt and less on
consistent with previous studies which demonstrated upper extremity movements. It included competitive
that children with ABI have decreased motor speed and challenging elements, yet allowed for a sense of
and processing speed skills, especially only 3 months control and of success. These are important motiva-
after injury [42,43]. tion and volition enhancers and contribute to
The findings also revealed significant moderate participation and activity engagement [23]. Similar
correlations between some VR performance variables to previous results from adult participants with stroke
in virtual environments and attention outcomes. and spinal cord injury [44,45], the lowest scores were
For personal use only.
Although success in the tested VR environments obtained in the ‘Soccer’ environment where per-
entails a number of cognitive abilities (e.g. divided ceived difficulty was highest.
attention) and visual motor abilities (e.g. visual No participants from the control group reported
perception, sitting balance), there did not appear to discomfort while using the video-capture environ-
be any systematic association between VR perfor- ments, whereas four from the study group did; one
mance and perceptual-behavioural variables. participant reported general discomfort during the
No significant correlations were found between ‘Birds and Balls’ environment and three reported it
upper extremity function and performance within the during the ‘Soccer’ environment. These ‘symptoms’
virtual environments, but positive correlations were appeared to be related to general fatigue or high
demonstrated in the reaching component. The exertion rather than to optical or vestibular cyber-
disassociations may be due to differences between sickness side effects such as dizziness or nausea as
the items of the Melbourne Assessment that dissim- video capture VR systems do not typically generate
ilar from the motor function that is needed to cyber-sickness side effects in users [46]. The results
perform in the virtual environment [30]. For of this study reinforce the general usability of the
example, in the Melbourne Assessment, a single IREX VR system for a wide variety of ages and
movement is tested, while repetitive upper extremity disabilities [30].
movements are required in the virtual environments,
entailing muscle endurance.
Several significant correlations were found be- Conclusions and limitations
tween self care functional abilities measured in the
PEDI and VR performance outcomes. Those corre- This study highlights the potential of video-capture
lations indicate that the more independent the child virtual reality as a tool in the rehabilitation process of
(i.e. higher achievements in PEDI), the better the children with ABI. The findings show its ability to
performance in VR. More and stronger correlations distinguish between performance of children with
were found between the caregiver assistance scales ABI and those of matched healthy controls. These
and performance in VR. In contrast to the PEDI differences may be due to reduced processing speed
score which is based on a child’s potential (as and reduced response time. Correlations were found
determined by the examiner, usually an occupational between performance within virtual environments
therapist), the caregiver assistance score is deter- and different attention capacities, and, to some
mined by the actual function of the child. The extent, between self-care functions and upper ex-
caregiver assistance score reflects motor, cognitive tremities functionality. In addition, high levels of
and emotional components, such as self-regulation satisfaction were reported following the VR experi-
Virtual reality intervention for children with ABI 1585
ence. This study has several limitations deriving from 10. Steultjens EM, Dekker J, Bouter LM, van de Nes JC, Cup
the small size of the sample and relatively large range EH, van den Ende CH. Occupational therapy for stroke
patients: a systematic review. Stroke 2003;34:676–687.
of ages. It seems warranted to conduct a large sample 11. Grealy M, Haffernan D. The rehabilitation of brain injured
study, in the future, that includes a variety of virtual children: the case for including physical exercise and virtual
environments among different children populations. reality. Pediatr Rehabil 2001;4:41–49.
The playful characteristics and ability to enhance 12. Will B, Kelche, C. Environment approaches to recovery of
function from brain damage: a review of animal studies. Adv
motivation by this type of VR technology points to its
Exp Med Biol 1992;325:79–103.
potential success for rehabilitation of children with 13. Missiuna C, Pollock N. Play deprivation in children with
ABI. The results of this study are thus consistent physical disabilities: the role of the occupational therapist in
with previous literature [17,25–29,35] in indicating prevention secondary disability. Am J Occup Ther 1991;45:882–
that usage of VR projected systems may constitute an 888.
innovative and meaningful tool for rehabilitation. 14. Royeen CB. Play as occupation and as indicator of health. In:
Chandler BE, editor. The essence of play: a child’s occupa-
Taken together, the results of this study suggest tion. Bethesda, MD: American Occupational Therapy Asso-
that children with ABI may benefit from the use of ciation; 1997. pp 1–14.
VR technologies in promoting children’s function 15. Holden M. Virtual environments for motor rehabilitation:
and health. By using this user-friendly, motivating review. Cyberpsychol Behav 2005;8:187–211.
Disabil Rehabil Downloaded from informahealthcare.com by Mcgill University on 10/27/14
and safe tool children’s enjoyment of the rehabilita- 16. Rizzo A, Buckwalter JG, Neumann U. Virtual reality and
cognitive rehabilitation: a brief review of the future. J Head
tion process may be enhanced. In recent years, VR Trauma Rehabil 1997;12:1–15.
technologies have become increasingly available (e.g. 17. Rose D, Johnson A, Attree A. Rehabilitation of head injured
[31]) although some are still expensive and techni- child: basic research and new technology. Pediatr Rehabil
cally complex. We anticipate that continued research 1999;1:3–7.
to demonstrate its effectiveness for evaluation and 18. Weiss PL, Kizony R, Feintuch U, Katz N. Virtual reality in
neurorehabilitation. In: Selzer ME, Cohen L, Gage FH, Clarke S,
intervention as well as the commercial development Duncan PW, editors. Textbook of neural repair and rehabilita-
of low-cost systems will lead to increased use in tion. Cambridge, England: University Press; 2006: 182–197.
clinical practice. 19. Gold JI, Kim SH, Kant AJ, Joseph MH, Rizzo A. Effectiveness
For personal use only.
31. Weiss PL, Sveistrup H, Rand D, Kizony R. Video capture 39. Witmer BG, Singer MJ. Measuring presence in virtual
virtual reality: a decade of rehabilitation assessment and environment: a presence questionnaire. Presence: Teleopera-
intervention. Phys Ther Rev 2009;14:307–321. tor Virtual Environ 1998;7:225–240.
32. Manly T, Robertson I, Anderson V, Nimmo-Smith I. TEA- 40. Kizony R, Katz N, Weiss PL. Adapting an immersive virtual
ch: the everyday attention for children. Bury St Edmunds: reality system for rehabilitation. J Visual Comput Animation
Thames Valley Test Company; 1999. 2003;14:261–268.
33. Manly T, Nimmo-Smith I, Watson P, Anderson V, Turner A, 41. Weiss PL, Bialik P, Kizony R. Virtual reality provides leisure
Robertson I. The differential assessment of children’s atten- time opportunities for young adults with physical and
tion: the Test of Every Attention for Children (TEA-Ch), intellectual disabilities. Cyberpsychol Behav 2003;6:335–342.
normative sample and ADHD performance. J Child Psychol 42. Brookshire B, Levine HS, Song JX, Zhang L. Components of
Psychiatry 2001;42:1065–1081. executive function in typically developing and head injured
34. Randall M, Johnson L, Reddihough D. The Melbourne children. Dev Neuropsychol 2004;25:61–83.
Assessment of unilateral upper limb function: test adminis- 43. Hoffman N, Donders J, Thompson EH. Novel learning ability
tration manual. Melbourne: Royal Children’s Hospital; 1999. after traumatic brain injury in children. Arch Clin Neuropsy-
35. Randall M, Carlin JB, Chondros P, Reddihough D. Reliability chol 2000;15:47–58.
of the Melbourne assessment of unilateral upper limb 44. Kizony R, Raz L, Katz N, Weingarden H, Weiss PL. Video
function. Dev Med Child Neurol 2001;43:761–767. capture virtual reality system for patients with paraplegic
36. Bourke-Taylor H. The Melbourne Assessment of unilateral spinal cord injury. J Rehabil Res Dev 2005;42:595–608.
upper limb function: construct validity and correlation with 45. Kizony R, Katz N, Weiss PL. Virtual reality based interven-
Disabil Rehabil Downloaded from informahealthcare.com by Mcgill University on 10/27/14
the Pediatric Evaluation of Disability Inventory. Dev Med tion: relationship between motor and cognitive abilities and
Child Neurol 2003;45:92–96. performance within virtual environments for patient with
37. Haley SM, Coster WJ, Ludlow LH. Pediatric Evaluation of stroke. 5th International Conference of Disability, Virtual
Disability Inventory: development, standartization and ad- Reality & Associates Technology, Oxford, UK, 2004.
ministration manual. Boston: PEDI Research Group; 1992. 46. Rand D, Kizony R, Feintuch U, Katz N, Josman N, Rizzo AA,
38. Nicholas S, Case-Smith J. Reliability and validity of the Weiss PL. Comparison of two VR platforms for rehabilitation:
Pediatric Evaluation of Disability Inventory. Pediatr Phys video capture versus HMD. Presence, Teleoperators Virtual
Ther 1996;8:15–24. Environ 2005;1:147–160.
For personal use only.