Virtual reality combined with robot-assisted gait training to improve walking ability of children with cerebral palsy: A randomized controlled trial

Abstract

BACKGROUND:

Children with cerebral palsy (CP) have disorders of posture and movement and which can limit physical activities such as walking

OBJECTIVE:

This study aims to investigate the effectiveness of virtual reality (VR) combined with robot-assisted gait training (RAGT) on walking ability in children with CP and clarify the most effective degree of weight reduction.

METHODS:

Sixty CP children were recruited and randomly allocated into four different groups. The control group received conventional physical therapy ( $n=$ 15), and task groups performed VR combined with RAGT with 15% (Group A, $n=$ 15) /30% (Group B, $n=$ 15) /45% (Group C, $n=$ 15) weight loss. All participants were given 50 min of therapy per session four times a week for 12 weeks and were assessed pre-and post-test with the surface electromyography (EMG), the Modified Ashworth Scale, the Gross Motor Function Measure (GMFM) dimension E and D, and Six-Minute Walking Test (6-MWT).

RESULTS:

All indicators had improved significantly in each group after the intervention ( $P<$ 0.05). The result of our study demonstrated that the more effective impacts of VR combined with RAGT on walking ability compared to the control group ( $P<$ 0.05), and 30% of weight loss had the best improvement in CP children ( $P<$ 0.01).

CONCLUSIONS:

VR combined RAGT can effectively improve walking ability in children with CP, especially when the weight loss is 30%.

Keywords

Virtual reality robot-assisted gait cerebral palsy walking ability

1. Introduction

Cerebral palsy (CP) is the most prevalent neurodevelopmental disorder caused by non-progressive brain injury in fetuses and infants [1]. Children with CP have disorders of posture and movement and which can limit physical activities such as walking [2]. Since the limitation of lower limb function directly affects the activities of daily living (ADL), how to improve the gait ability in children with CP is of great significance.

Currently, the conventional methods for CP include physical therapy, occupational therapy [3], as well as traditional Chinese therapy like acupuncture and massage [4, 5]. These interventions are often concurrently used and consume numerous amounts of energy, cost, and time [6]. More importantly, few studies have proven that these conventional therapies are effective in gait recovery in CP children [7].

Besides the conventional therapies, technology-based treatments have been recently proposed for improving walking ability [8, 9]. Among others, robot-assisted gait training (RAGT) has been widely used in gait rehabilitation of adults with different diseases such as stroke, spinal cord injury, as well as children with CP [10]. RAGT is composed of a body weight support (BWS) system and two actuated orthoses that drive the patients’ legs on a treadmill. Based on the principles of motor learning and neuroplasticity, the effectiveness of RAGT relies on repeated input of the correct gait pattern and goal-oriented tasks. Jin et al. showed that RAGT was effective in improving gross motor function in CP children [11]. Similarly, Cherni et al. also confirmed the role of RAGT in enhancing the walking speed and endurance in children with CP [12]. The rationale of VR is also grounded on some theories of neuroplasticity, including goal-oriented, repetitive, and intensive training [13, 14, 15]. It can provide real-time feedback by allowing users to interact with virtual objects, which makes VR a funnier and playful approach especially for children [16] Moreover, children have higher neuroplasticity and better motor learning ability than adults, so it is more effective to improve motor function in children’s training [17].

Nevertheless, VR combined with RAGT is often applied in the treatment of stroke patients rather than in pediatric CP [18, 19]. Although it has been shown that the level of BWS can alter gait parameters such as stride length, stride speed, and standing time, thus affecting the recovery of gait in patients [20, 21], there have been no reports to determine the optimal level of BWS for children with CP. Therefore, the present work aimed to investigate the effect of VR combined GRAT with different (15%/30%/45%/) levels of BWS in improving the walking ability of CP children and to determine the most valid level of BWS.

2. Methods

2.1 Participants

We recruited 60 children (28 males and 32 females, 6 to 11 years old) from the Rehabilitation Department of Shanxi Provincial Rehabilitation Research Center who met the diagnosis of spastic CP. Participants were selected in line with the inclusion and exclusion criteria.

The inclusion criteria were: (1) height suitable for walking robots; (2) rate in Level II-III on the Gross Motor Function Classification System (GMFCS); (3) able to follow instructions and cooperate with trainers.

The exclusion criteria were: (1) epilepsy, severe heart, lung, liver, and kidney disease; (2) other neuromuscular and osteoarticular diseases; (3) Botox injection or orthopedic surgery and muscle release within 6 months; (4) vision or hearing impairment.

This study was approved by the Ethics Committee of our hospital. All parents/caregivers of each child have been informed of the purpose and procedures of the study and signed informed consent. Rehabilitation programs were subsidized by the government, and all children received half-year treatment in outpatient visits.

2.2 Interventions

This research was a randomized controlled trial. Random allocation software was used to divided children into 4 groups. All participants completed 48 treatment sessions for 3 months, each session was conducted for 50 min (30 min basic physical therapy followed by 20 min different kinds of walking training), 4 times per week. The control group underwent conventional walking training, whereas the task groups underwent the VR combined RAGT at different levels of weight support:15% for group A, 30% for group B, 45% for group C. The flow diagram for experimental procedures is shown in Fig. 1.

Figure 1.

Flow diagram for experimental procedures.

The detailed procedures of basic training are as follows:

(1)

Warm-up (5 min): Stretching all the lower limb joints and major muscles to relieve muscles tension and improve the joints’ range of motion.

(2)

Strength Training (5 min): Standing on one foot, double bridge training to build the muscle strength of lower limbs.

(3)

Balance Training (10 min): Weight shift training, kneel training, throw and catch ball training in standing, balance board standing training to improve the load-bearing ability of affected lower limbs and to increase dynamic balance.

(4)

Coordination Training (5 min): Alternating the lower limb flexion and extension training, alternating step-up training to improve the coordination of the lower limb.

For 20 min different kinds of walking training: the control group was assisted with walking training in parallel bars or on a flat surface, according to the condition of the child. The task groups wore VR throughout the training process. GRAT BWS was set for each group (Group A 15%, Group B 30%, Group C 45%). Walking speed was adjusted to the patient’s ability to participate actively during therapy. The guidance force was maintained at 100% during training. A physical therapist monitored the lower limbs, prompted the experimental procedure, and provided verbal encouragement when necessary [22, 23].

2.3 Therapy device

The Lokomat (Hocoma AG, Volketswil, Switzerland) was used for RAGT in task groups. This device comprises two robotic-orthosis, each equipped with motor drives in hip and knee joints which guide the patients’ legs moved along the physiological walking pattern on a treadmill. Children’s limbs were fixed into the orthosis with suspenders tied to a BWS system to bear subjects’ body weight. The orthosis is adjusted to the width of the child’s hips and the length of the lower limbs to ensure that walking with the device is as natural and comfortable as possible. The hip and knee joints of the Lokomat orthosis are computer-controlled, moving in the sagittal plane through the gait cycle and ensuring that the walking speed is synchronized with that of a treadmill [24]. The Lokomat’s orthosis with kids legs was chosen for children whose femur lengths were between 210 and 350 mm. In both groups, patients were encouraged to behave actively during RAGT.

VR was applied in task groups while they are training on Lokomat, the scenario of VR took place in the field, where subjects had to walk along a straight path in the middle of the screen. The Lokomat can both input and output the children’s movements. It sensed the steps of the children and transferred the signal into the movement of the virtual character’s legs in real-time. Consequentially, the character in VR walked along the path according to the children’s own steps. The training was supervised by a trained therapist.

2.4 Outcomes measures

There were four pre- and post-testing outcome measures: the EMG, the Modified Ashworth Scale, GMFM88 dimension E, and D as well as the 6-MWT.

Surface electromyography (EMG) was used to evaluate the strength and tension of the gastrocnemius muscle. Its evaluation index consists of EMG median frequency(MF), integrated EMG (iEMG) and co-contraction ratio (CR) [25]. CR $=$ agonistic muscle MF/[muscle MF $+$ antagonistic muscle MF]* 100%. The lowered CR means better ankle coordination as well as improved ankle spasm in CP children [26].

The Modified Ashworth Scale scores the level of muscular tension between 0 (lack of spasticity) and ? (severe rigidity) is used to provide the evaluation of the ankle spasm. Evidence indicates that the Modified Ashworth Scale is a valid and reliable measurement tool [27].

The Gross Motor Function Measure 88 (GMFM88) dimension D (standing) and E (walking) were applied by a GMFM-certified therapist to evaluate subjects’ gross motor ability. Dimension D includes 13 items with a total score of 39 points, while E includes 24 items with a total score of 72 points. The GMFM88 was developed and validated by Engelen in 2007 [28].

Children’s gait ability and endurance were assessed by the Six-Minute Walking Test (6-MWT) [29]. During the test, children with CP were allowed to are required to walk as fast as possible.

All participants were allowed to wear their orthoses or footwear during the assessments of GMFM88 and 6-MWT,

2.5 Statistical analysis

Data were evaluated for normality and symmetry to ensure that the use of parametric statistics as appropriate. A paired t-test was applied to compare the changes in the data pre- and post-test. An analysis of variance (ANOVA) was used for comparisons of the values at different levels of body weight reduction between 4 groups All statistical analyses were performed using SPSS Statistical Package, version 26.0.

Table 1
Characteristics of all participants

	Gender	Age	Weight	Time of	Hight	GMFCS rating
	(M/F)	(years)	(kg)	treatment (m)	(cm)	level II III
Control group	8/7	4.53 $\pm$ 1.73	20.01 $\pm$ 6.29	21.07 $\pm$ 8.24	110 $\pm$ 8	10/5
Group B	6/9	5.00 $\pm$ 1.65	19.27 $\pm$ 4.32	24.33 $\pm$ 10.14	107 $\pm$ 13	9/6
Group C	7/8	5.33 $\pm$ 2.93	20.27 $\pm$ 5.02	27.01 $\pm$ 11.32	115 $\pm$ 11	9/6
Group D	7/8	5.67 $\pm$ 1.58	18.78 $\pm$ 6.14	25.07 $\pm$ 10.75	101 $\pm$ 9	8/7
$P$	0.91	0.42	0.71	0.49	0.39	0.9

Values are expressed as mean $\pm$ SD. GMFCS, Gross Motor Function Classification System.

3. Results

In total, all 60 participants completed the 12-week training without a drop-off. As shown in Table 2, compared to the baseline, the post-test level of iEMG, MF, CR, the Modified Ashworth, GMFM dimension D, and E, as well as 6-MWT were improved significantly in all groups ( $P<$ 0.01). Moreover, these changes were greater in task groups than in the control group ( $P<$ 0.05). Notably, all indicators besides CR in Group B ( $P>$ 0.05) registered the greatest change among other task groups ( $P<$ 0.05), while the outcomes were not obviously different between group A and C ( $P>$ 0.05) except for GMFM dimension D and 6-MWT ( $P<$ 0.05).

4. Discussion

The current study investigated the effects of VR combined with GRAT at different levels of BWS on gait ability in CP children. According to the results of the EMG, the Modified Ashworth Scale, the GMFM dimension E, and D, and 6-MWT, comparing to the control group, the greater changes were the observed in task groups, among which 30% weight reduction registered the most remarkable improvements in walking ability.

To the best of our knowledge, this is the first study to determine the most valid level of BWS when applying RAGT to improve gait ability in children with CP.

A study suggested that varying degrees of BWS may influence the size of steps and the shift of weight during walking, which in turn affects the recovery of patients’ walking ability [20, 21]. For adult patients with stroke on treadmills, the most applied level of BWS is 30% [30, 31], and the effectiveness has been proved on gait training [32, 33]. Yet the optimal level of weight support provided by walking-assisted robots for children with CP has not been clearly investigated. Current studies also vary greatly in the level of BWS taken for CP children. In the study by Jin, et al., RAGT BWS was adapted to the condition of the knee of the children with CP, and the results showed that RAGT improved the motor ability of the children [11]. Pool et al. found that, compared to conventional ground training, RAGT had no significant effect on indoor walking speed and gross motor ability in children with cerebral palsy when 60% of BWS was taken [34].

Our results indicated that the walking robot works best when the level of BWS was set at 30%. Possible reasons are as follows: On the one hand, excessive weight loss reduces the load-bearing of the lower limbs of children with CP, which is detrimental to both the improvement of lower limb muscle strength and the proactive engagement of active movement in children with CP. In response, the walking robot has to increases the driving force to ensure a certain walking speed, which will further affect the muscular endurance of the lower limb. On the other hand, insufficient weight loss will pose a strong stimulation on

Table 2
Results for outcome measures

	Mean $\pm$ SD				$p$
	Control group	Group A	Group B	Group C	Control group vs			Group A vs	Group B vs
					Group A	Group B	Group C	Group C	Group A	Group C
EMG
iEMG (mv)
Pre-test	23.09 $\pm$ 3.95	24.37 $\pm$ 4.56	25.68 $\pm$ 4.13	23.71 $\pm$ 5.02	0.002++	0.000+++	0.000+++	0.174	0.000+++	0.007++
Post-test	34.85 $\pm$ 3.46	39.58 $\pm$ 5.13	45.75 $\pm$ 4.52	41.62 $\pm$ 4.84
$p$	0.000**	0.000**	0.000**	0.000**
MF (HZ)
Pre-test	89.46 $\pm$ 9.33	92.29 $\pm$ 11.16	91.76 $\pm$ 10.33	91.92 $\pm$ 10.52	0.001++	0.000+++	0.000+++	0.585	0.001++	0.005++
Post-test	109.26 $\pm$ 12.05	122.41 $\pm$ 10.76	135.15 $\pm$ 11.93	124.38 $\pm$ 11.27
$p$	0.000**	0.000**	0.000**	0.000**
CR (%)
Pre-test	60.02 $\pm$ 10.5	57.86 $\pm$ 8.11	58.4 $\pm$ 29.75	58.55 $\pm$ 9.41	0.008++	0.000+++	0.011+	0.936	0.000+++	0.082
Post-test	43.45 $\pm$ 9.42	34.74 $\pm$ 10.02	29.3 $\pm$ 58.21	58.55 $\pm$ 9.41
$p$	0.000**	0.000**	0.000**	0.000**
Ashworth
Pre-test	3.93 $\pm$ 0.77	3.87 $\pm$ 0.72	3.87 $\pm$ 1.02	3.87 $\pm$ 0.80	0.019+	0.000+++	0.009++	0.789	0.009++	0.019+
Post-test	3.27 $\pm$ 0.57	2.67 $\pm$ 0.70	2.00 $\pm$ 0.73	2.60 $\pm$ 0.61
$p$	0.002*	0.001*	0.001*	0.001*
GMFM
D
Pre-test	14.30 $\pm$ 5. 23	15.62 $\pm$ 3.51	14.07 $\pm$ 2.78	15.49 $\pm$ 4.39	0.000+++	0.000+++	0.000+++	0.035+	0.000+++	0.000+++
Post-test	21.51 $\pm$ 4.80	26.18 $\pm$ 6.07	33.97 $\pm$ 4.24	28.36 $\pm$ 5.72
$p$	0.001*	0.000**	0.000**	0.000**
E
Pre-test	25.57 $\pm$ 4.62	26.74 $\pm$ 5.24	25.39 $\pm$ 6.15	26.51 $\pm$ 4.82	0.000+++	0.000+++	0.001++	0.74	0.000+++	0.000+++
Post-test	34.07 $\pm$ 5.38	40.25 $\pm$ 6.35	46.47 $\pm$ 4.63	39.85 $\pm$ 4.93
$p$	0.000**	0.000**	0.000**	0.000**
6-MWT (M)
Pre-test	299.5 $\pm$ 13.69	312.6 $\pm$ 15.18	305.3 $\pm$ 14.26	320.4 $\pm$ 13.76	0.000+++	0.000+++	0.000+++	0.007++	0.000+++	0.000+++
Post-test	373.16 $\pm$ 19.42	400.85 $\pm$ 12.82	442.33 $\pm$ 13.63	415.64 $\pm$ 14.55
$p$	0.000**	0.000**	0.000**	0.000**

Values are expressed as mean $\pm$ SD. *differences between pre and post-test. ${}^{*}p<$ 0.01; ${}^{**}p<$ 0.001. +Significant differences between control group, group A , B and C. + $p<$ 0.05; ++ $p<$ 0.01; +++ $p<$ 0.001.

gastrocnemius muscle strength, which will worsen the gastrocnemius muscle tension growth and in turn, restrict the growth of strength.

A previous study has reported that lower limb muscular effort measured by surface electromyography in children with CP was significantly improved after RAGT combined with VR [16]. The repetitive and symmetrical movements provided by the walking robot relieve muscular tension in children with CP and helps them to better control the lower limb joints, improving the stability and coordination of joint movements [20]. In the study by Kenichi Yoshikawa et al. walking assist device can increase the ankle range of motion and musculus gastrocnemius activity function [35]. Also, Cherng et al. reported that the gait and gross motor function was significantly improved after gait training with a BWS in CP children [36]. Specifically, during RAGT, the ambulation ability in stroke patients could be improved more significantly when VR is applied. This may attribute to the fact that patients in the VR augmented RAGT group have higher acceptability and motivation to undergo therapy [19]. Besides that, VR increases the fun of training, thus activates brain regions related to motor learning, and promote functional reconstruction [37]. Consistent with the above studies, our work shows that VR combined with RAGT was effective in relieving ankle spasm, balance, gross motor ability, as well as gait function in CP children, and this may be attributed to the fact that combination therapy is more effective than RAAT alone.

Some limitations in the current study should be considered in the future study, first, the insufficient samples in this study may reduce the generalization of the outcome. Second objective tests are required to confirm the underlying mechanism of these improvements. Thus, a larger sample size as well as laboratory examination, brain functional imaging, and biomechanical analysis is required in future studies.

5. Conclusion

Our results suggested that when weight loss is at 30%, the VR combined with RAGT can be most effective in reducing gastrocnemius and ankle spasm, improving gastrocnemius strength, and can further improve balance and gait in children with CP.

Footnotes

Acknowledgments

This study was supported by Shanxi University and Shanxi Rehabilitation Research Center.

Conflict of interest

The authors have no conflict of interest to report.

References

Bax

Goldstein

Rosenbaum

Leviton

Paneth

Dan

, et al. Proposed definition and classification of cerebral palsy, April 2005. Developmental medicine and child neurology. 2005; 47(8): 571-6. doi: 10.1017/s001216220500112x.

Beckung

Hagberg

. Neuroimpairments, activity limitations, and participation restrictions in children with cerebral palsy. Developmental Medicine and Child Neurology. 2002; 44(5): 309-16. doi: 10.1017/s0012162201002134.

Saleh

Korner-Bitensky

Snider

Malouin

Mazer

Kennedy

, et al. Actual vs. best practices for young children with cerebral palsy: a survey of paediatric occupational therapists and physical therapists in Quebec, Canada. Dev Neurorehabil. 2008; 11(1): 60-80. doi: 10.1080/17518420701544230.

Liu

. Acupuncture combined with music therapy for treatment of 30 cases of cerebral palsy. J Tradit Chin Med. 2009; 29(4): 243-8. doi: 10.1016/s0254-6272(09)60074-1.

Novak

McIntyre

Morgan

Campbell

Dark

Morton

, et al. A systematic review of interventions for children with cerebral palsy: state of the evidence. Developmental Medicine and Child Neurology. 2013; 55(10): 885-910. doi: 10.1111/dmcn.12246.

Cada

O’Shea

. Identifying barriers to occupational and physical therapy services for children with cerebral palsy. J Pediatr Rehabil Med. 2008; 1(2): 127-35.

Kim

Yoo

. Effects of live and video form action observation training on upper limb function in children with hemiparetic cerebral palsy. Technol Health Care. 2018; 26(3): 437-43. doi: 10.3233/thc-181220.

Wallard

Dietrich

Kerlirzin

Bredin

. Robotic-assisted gait training improves walking abilities in diplegic children with cerebral palsy. European Journal of Paediatric Neurology: EJPN: Official Journal of the European Paediatric Neurology Society. 2017; 21(3): 557-64. doi: 10.1016/j.ejpn.2017.01.012.

Gagliardi

Turconi

Biffi

Maghini

Marelli

Cesareo

, et al. Immersive Virtual Reality to Improve Walking Abilities in Cerebral Palsy: A Pilot Study. Ann Biomed Eng. 2018; 46(9): 1376-84. doi: 10.1007/s10439-018-2039-1.

10.

Biffi

Beretta

Storm

Corbetta

Strazzer

Pedrocchi

, et al. The Effectiveness of Robot- vs. Virtual Reality-Based Gait Rehabilitation: A Propensity Score Matched Cohort. Life (Basel, Switzerland). 2021; 11(6). doi: 10.3390/life11060548.

11.

Jin

Yang

Choi

Sohn

. The Effect of Robot-Assisted Gait Training on Locomotor Function and Functional Capability for Daily Activities in Children with Cerebral Palsy: A Single-Blinded, Randomized Cross-Over Trial. Brain Sciences. 2020; 10(11). doi: 10.3390/brainsci10110801.

12.

Cherni

Ballaz

Lemaire

Dal Maso

Begon

. Effect of low dose robotic-gait training on walking capacity in children and adolescents with cerebral palsy. Neurophysiologie Clinique = Clinical Neurophysiology. 2020; 50(6): 507-19. doi: 10.1016/j.neucli.2020.09.005.

13.

Zimmerli

Jacky

Lünenburger

Riener

Bolliger

. Increasing patient engagement during virtual reality-based motor rehabilitation. Arch Phys Med Rehabil. 2013; 94(9): 1737-46. doi: 10.1016/j.apmr.2013.01.029.

14.

Holden

. Virtual environments for motor rehabilitation: review. Cyberpsychol Behav. 2005; 8(3): 187-211; discussion 2-9. doi: 10.1089/cpb.2005.8.187.

15.

Brütsch

Schuler

Koenig

Zimmerli

Koeneke

Lünenburger

, et al. Influence of virtual reality soccer game on walking performance in robotic assisted gait training for children. J Neuroeng Rehabil. 2010; 7: 15. doi: 10.1186/1743-0003-7-15.

16.

Schuler

Brsch

Müller

van Hedel

Meyer-Heim

. Virtual realities as motivational tools for robotic assisted gait training in children: A surface electromyography study. NeuroRehabilitation. 2011; 28(4): 401-11. doi: 10.3233/nre-2011-0670.

17.

Hösl

Böhm

Arampatzis

Keymer

Döderlein

. Contractile behavior of the medial gastrocnemius in children with bilateral spastic cerebral palsy during forward, uphill and backward-downhill gait. Clin Biomech (Bristol, Avon). 2016; 36: 32-9. doi: 10.1016/j.clinbiomech.2016.05.008.

18.

Calabrò

Naro

Russo

Leo

De Luca

Balletta

, et al. The role of virtual reality in improving motor performance as revealed by EEG: a randomized clinical trial. J Neuroeng Rehabil. 2017; 14(1): 53. doi: 10.1186/s12984-017-0268-4.

19.

Bergmann

Krewer

Bauer

Koenig

Riener

Müller

. Virtual reality to augment robot-assisted gait training in non-ambulatory patients with a subacute stroke: a pilot randomized controlled trial. Eur J Phys Rehabil Med. 2018; 54(3): 397-407. doi: 10.23736/s1973-9087.17.04735-9.

20.

Wang Lina

Gao

Zhang

Zhao

. Application of MOTOmed virtual reality training system in rehabilitation treatment of children with spastic cerebral palsy. Shandong Medical Journal. 2015; 55(2): 35-6. (in Chinese).

21.

Gama

Celestino

Barela

AMF

. Gait initiation and partial body weight unloading for functional improvement in post-stroke individuals. Gait Posture. 2019; 68: 305-10. doi: 10.1016/j.gaitpost.2018.12.008.

22.

Digiacomo

Tamburin

Tebaldi

Pezzani

Tagliafierro

Casale

, et al. Improvement of motor performance in children with cerebral palsy treated with exoskeleton robotic training: A retrospective explorative analysis. 2019; 37(3): 239-44. doi: 10.3233/rnn-180897.

23.

Weinberger

Warken

König

Vill

Gerstl

Borggraefe

, et al. Three by three weeks of robot-enhanced repetitive gait therapy within a global rehabilitation plan improves gross motor development in children with cerebral palsy – a retrospective cohort study. European Journal of Paediatric Neurology: EJPN: Official Journal of the European Paediatric Neurology Society. 2019; 23(4): 581-8. doi: 10.1016/j.ejpn.2019.05.003.

24.

Colombo

Joerg

Schreier

Dietz

. Treadmill training of paraplegic patients using a robotic orthosis. Journal of Rehabilitation Research and Development. 2000; 37(6): 693-700.

25.

Wang

Guo

Zhao

, et al. Design of Rehabilitation Training System with Electromyography Feedback for Stroke Patients. Zhongguo Yi Liao Qi Xie Za Zhi. 2015; 39(3): 187-9, 205.

26.

Poon

Hui-Chan

. Hyperactive stretch reflexes, co-contraction, and muscle weakness in children with cerebral palsy. Developmental Medicine and Child Neurology. 2009; 51(2): 128-35. doi: 10.1111/j.1469-8749.2008.03122.x.

27.

Meseguer-Henarejos

Sánchez-Meca

López-Pina

Carles-Hernández

. Ejop, medicine r. Inter- and Intra-Rater Reliability of the Modified Ashworth Scale: A Systematic Review and Meta-Analysis. 2018; 54(4): 576-90. doi: 10.23736/s1973-9087.17.04796-7.

28.

Engelen

Ketelaar

Gorter

. Selecting the appropriate outcome in paediatric physical therapy: how individual treatment goals of children with cerebral palsy are reflected in GMFM-88 and PEDI. Journal of Rehabilitation Medicine. 2007; 39(3): 225-31. doi: 10.2340/16501977-0040.

29.

Thompson

Beath

Bell

Jacobson

Phair

Salbach

, et al. Test-retest reliability of the 10-metre fast walk test and 6-minute walk test in ambulatory school-aged children with cerebral palsy. Developmental Medicine and Child Neurology. 2008; 50(5): 370-6. doi: 10.1111/j.1469-8749.2008.02048.x.

30.

McCain

Pollo

Baum

Coleman

Baker

Smith

. Locomotor treadmill training with partial body-weight support before overground gait in adults with acute stroke: a pilot study. Arch Phys Med Rehabil. 2008; 89(4): 684-91. doi: 10.1016/j.apmr.2007.09.050.

31.

Combs

Dugan

Ozimek

Curtis

. Bilateral coordination and gait symmetry after body-weight supported treadmill training for persons with chronic stroke. Clin Biomech (Bristol, Avon). 2013; 28(4): 448-53. doi: 10.1016/j.clinbiomech.2013.02.001.

32.

Lindquist

Prado

Barros

Mattioli

da Costa

Salvini

. Gait training combining partial body-weight support, a treadmill, and functional electrical stimulation: effects on poststroke gait. Phys Ther. 2007; 87(9): 1144-54. doi: 10.2522/ptj.20050384.

33.

Lamontagne

Fung

. Faster is better: implications for speed-intensive gait training after stroke. Stroke. 2004; 35(11): 2543-8. doi: 10.1161/01.STR.0000144685.88760.d7.

34.

Pool

Valentine

Taylor

Bear

Elliott

. Locomotor and robotic assistive gait training for children with cerebral palsy. Developmental Medicine and Child Neurology. 2021; 63(3): 328-35. doi: 10.1111/dmcn.14746.

35.

Yoshikawa

Mutsuzaki

Koseki

Endo

Hashizume

Nakazawa

, et al. Gait Training Using a Wearable Robotic Device for Non-Traumatic Spinal Cord Injury: A Case Report. Geriatr Orthop Surg Rehabil. 2020; 11: 2151459320956960. doi: 10.1177/2151459320956960.

36.

Cherng

Liu

Lau

Hong

. Effect of treadmill training with body weight support on gait and gross motor function in children with spastic cerebral palsy. Am J Phys Med Rehabil. 2007; 86(7): 548-55. doi: 10.1097/PHM.0b013e31806dc302.

37.

Cano Porras

Sharon

Inzelberg

Ziv-Ner

Zeilig

Plotnik

. Advanced virtual reality-based rehabilitation of balance and gait in clinical practice. Ther Adv Chronic Dis. 2019; 10: 2040622319868379. doi: 10.1177/2040622319868379.