INTRODUCTION

For many individuals with physical impairments, manual wheelchair offers the means for independent mobility, improved self-care, increased enjoyment of social activities and empowerment.  Using wheelchairs is not always easy for newly disabled people: they need time to understand and learn how to be confident on how to use them properly for activities of daily living.  Wheelchair training could offer solutions to increase independent wheeled mobility.  There is a growing body of evidence to support the use of a validated manual wheelchair skills training to improve the wheelchair skills capacity of manual wheelchair users [1,2]. Validated manual wheelchair skills training programs have also been effective at improving the wheelchair skills of health professional students and clinicians [3,4].

However, due to the short period of treatment time, a limited number of basic wheelchair skills were taught upon completion of rehabilitation.  Moreover, wheelchair navigation may become tiring and stressful during indoor activities when the user needs to avoid many obstacles, and during outdoor travel due to inaccessible byways.  In real life, wheelchair users face several common challenges: articulate steering, spatial relationships, and negotiating obstacles.  These could all be confronted in the process of negotiating a single intersection.  Therefore, a novel approach to teaching these wheelchair skills required for daily living is thus urgently needed.

Virtual reality (VR) is a technology which allows a user to interact with a computer-simulated virtual environment, in which multisensory stimuli may be delivered in a controlled way.  Over the past decade, VR-based systems have been developed to address cognitive, motor, and behavioral disabilities in the areas of assessment, rehabilitation and training [5,6].  VR has already demonstrated its potential for vehicular manipulation training (e.g. car, bicycle, or wheelchair [7,8].  VR allows a user to perceive a virtual environment via sensor interfaces (e.g. Head Mounted Display, and mobile platforms), and to control it via motor interfaces. These interfaces can be those which are used in daily tasks. For example, a virtual environment may be intuitively navigated via a steering wheel, handlebars, or handrims. Moreover, VR training applications have well-known assets: they are safe, cost-effective and highly motivating.

Therefore, the purpose of this study was to develop and validate a wheelchair maneuverability simulator in conjunction with a VR facility, could be used to address issues of wheelchair training program.   It was anticipate that this VR wheelchair maneuverability simulator would help wheelchair users to learn wheelchair maneuvering, obstacle-negotiating, and basic daily living skills in a setting where the user and clinicians could feel safe that they would not do any real harm to themselves or others.

METHOD

Participants

So far, we recruited five manual wheelchair users to engage this user validation study. Participants were included in the study if they met the following criteria: (1) 20 years of age or older; and (2) used a manual wheelchair as a primary mode of mobility; Participants were excluded if they (1) had previous history of upper-extremity pain interfering with wheelchair propulsion, (2) had previous history of severe motion sickness, or (3) had recent injuries related to the eyes, face, or neck that prevented comfortable use of VR hardware or software.  Participants were informed of the nature of the experiment and signed informed consent in accordance with the procedures approved by our local Institutional Review Board prior to participation in the study. Afterwards, all participants were asked to wear the head mounted display (HTC Vive), and sit on a wheelchair which is placed and tied down on a motion platform with rollers.  They went through 2-mintues wheelchair basic maneuvers such as moving the wheelchair forward a short distance, crossing intersection, and propelling uphill in a virtual environment.  They then related their experience in VR wheelchair maneuvers through the user feedback questionnaire, and semi-structured interviews. 

Wheelchair maneuverability simulator

The simulator consists of two independent rollers with encoders which connected with an Arduino board in order to convert its movements into digital inputs.  As they pushed the wheels which tied down on the roller, the encoders would feed the actions to the simulator to update the virtual environment in real-time.  A customized VR wheelchair maneuvers software, running with the Unity engine, read the inputs, computed them and allowed the user to move in the desired direction. 

User feedback questionnaire

To subjective measure the acceptance and feasibility of this prototyped wheelchair maneuverability simulator, a user feedback questionnaire, Presence Questionnaire (PQ), was used. The PQ was composed of 19 questions in which participants use a 7-point scale to rate various experience within the VR environment.  Participants were asked to place an “X” on the scale to represent their level of satisfaction.  The maximum total score is 133 points indicating a high level of presence [9].  The PQ asked participants to report their perceptions of the experiences provided by the virtual environment sensory and control interfaces, their involvement in the virtual environment task, the nature and quality of their interactions with the virtual environment, and how quickly they adjusted to the virtual environment experience.  The items assessed different aspects of presence: involvement/control, sensory fidelity, adaptation/immersion, and interface quality.

Qualitative outcomes

    Participants also underwent an individual semi-structured interview following the completion VR wheelchair maneuvers.  The interview questions extended on the information provided from the participant’s satisfaction and gave the opportunity for participants to express their views on the VR wheelchair maneuvers in more detail. Interviews were conducted face-to-face by an independent investigator who was not involved in the development of wheelchair maneuverability simulator or VR wheelchair maneuvers software, allowing the participants to express their opinion in a non-biased manner.  The interview was digitally recorded, and then all interview records were transcribed into documentation. Thematic analysis was used to identify, analyze, and report patterns (themes) that stemmed from the data [10]. Critical and representative patterns were extracted from the data and coded into four major themes that indicated how VR wheelchair maneuverability simulator made for immersive experiences to their daily livings.

Statistical analysis

All quantitative data, including participants’ characteristics and the Presence Questionnaire, were analyzed using SPSS software (Version 20 for Windows, IBM, USA).  Descriptive statistics were used to describe satisfaction level of PQ based on above four aspects of presence: involvement, sensory fidelity, adaptation/immersion, and interface quality. 

RESULTS

Five male with spinal cord injuries ranging from T8 to L2 participated in this study.  Their mean age and years postinjury were 40.2 + 9.5 and 12.1 + 8.3 years, respectively.  Participants in general greatly appreciated their experience with the simulator.  The mean score of four aspects of presence were: 5.35, 5.13, 5.13 and 4.8, respectively (Table 1).

Table 1.  The satisfaction of virtual environment based on Presence Questionnaire

n=5	Involvement	Sensory fidelity	Adaptation/immersion	Interface quality
Satisfaction score	5.35 + 0.49	5.13 + 0.64	5.13 + 0.50	4.88 + 0.64

During the interviews, participants made similar comments about the activities as questionnaire showed in Table 1. They also were interested in the importance of realism in the simulator activities for their use in wheelchair skill training.  Four themes were generated draw from the interview data and were described as follows:

Theme 1: Involvement is important

When participants were asked for their first impression of the VR wheelchair simulator, all thought it was appealing and attractive.  They were able to control VE events, such as propelling forward, or turning in tight spaces, responsiveness of the virtual environment to user-initiated action.  They indicated that the simulator provided them interactive experiences in a similar manner. 

 

Theme 2: Sensory fidelity is limited

Though this simulator made an impressive and appealing first impression on the participants, some critical opinions emerged afterwards, especially when the researcher asked whether they felt the VR environment could relate to their prior experience.  Participants felt they were forced to view the street or objects they were provided in VR environment without any choice.  Sometimes they criticized the quantity and quality of the VR environment as comedic game. 

Theme 3: Benefit varies in adaptation/immersion

When asked whether the simulator was a useful tool in wheelchair skill training, all participants addressed that it could be helpful sometimes, but not always.  The simulator performance in helping complicated wheelchair skill training might vary according to scenario.  One concern was that a simulator might be helpful for propelling across pavement but not for negotiating curbs. 

Theme 4: Interface triggering does not work well enough

Most participants complained about the effectiveness of the wheeling speed triggering.  Participants noticed that it lagged for first few seconds then runs smoothly in the beginning of propulsion.  Sometimes the turning direction did not correspond to the driving wheel direction.

Discussion

The results of this study indicated that this proposed VR wheelchair simulator might have potential to increase wheelchair users’ learning motivation, however, some issues arose that should be considered in order to use it more.  In this study, the simulator made an attractive, novel impression and therefore demonstrated potential as a tool for stimulating motivation.  But, participants stated that the training benefits varied according to scenario.  In order validate this proposed VR simulator, the PQ were used.  The results showed that average satisfaction score rating 5 out 7 indicated participants mostly had positive perceptions of this VR simulator performance.  But, Low performance rating on interface quality was pointed out.  This finding was consistent with qualitative results from interview.  Interface quality items addressed whether control devices or display devices interfere or distract from task performance, and the extent to which the participants felt able to concentrate on the tasks.  In order to simulate more real life situations, such uphill/downhill condition, some wheelchair simulators incorporate a mobile platform, such as a hydraulic platform composed of six jacks, which permits the user to be inclined according to the slope in the virtual environment.  Despite its many advantages in a virtual environment, these simulators with mobile platforms are too expensive, and need significant physical space to setup.  It becomes impracticable to permit this application in clinical setting.  Therefore, we built this low-cost prototype of a VR simulator.  Although this simulator could not fully satisfy all realistic needs, it offered functionalities, notably a learning pathway enabled to propose exercises that meet the user specific training needs.