This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Serious Games, is properly cited. The complete bibliographic information, a link to the original publication on http://games.jmir.org, as well as this copyright and license information must be included.
Use of exergames can complement conventional therapy and increase the amount and intensity of visuospatial neglect (VSN) training. A series of 9 exergames—games based on therapeutic principles—aimed at improving exploration of the neglected space for patients with VSN symptoms poststroke was developed and tested for its feasibility.
The goal was to determine the feasibility of the exergames with minimal supervision in terms of (1) implementation of the intervention, including adherence, attrition and safety, and (2) limited efficacy testing, aiming to document possible effects on VSN symptoms in a case series of patients early poststroke.
A total of 7 patients attended the 3-week exergames training program on a daily basis. Adherence of the patients was documented in a training diary. For attrition, the number of participants lost during the intervention was registered. Any adverse events related to the exergames intervention were noted to document safety. Changes in cognitive and spatial exploration skills were measured with the Zürich Maxi Mental Status Inventory and the Neglect Test. Additionally, we developed an Eye Tracker Neglect Test (ETNT) using an infrared camera to detect and measure neglect symptoms pre- and postintervention.
The median was 14 out of 15 (93%) attended sessions, indicating that the adherence to the exergames training sessions was high. There were no adverse events and no drop-outs during the exergame intervention. The individual cognitive and spatial exploration skills slightly improved postintervention (
The high adherence rate and absence of adverse events showed that these exergames were feasible and safe for the participants. The results of the amount of exergames use is promising for future applications and warrants further investigations—for example, in the home setting of patients to augment training frequency and intensity. The preliminary results indicate the potential of these exergames to cause improvements in cognitive and spatial exploration skills over the course of training for stroke patients with VSN symptoms. Thus, these exergames are proposed as a motivating training tool to complement usual care. The ETNT showed to be a promising assessment for quantifying spatial exploration skills. However, further adaptations are needed, especially regarding calibration issues, before its use can be justified in a larger study sample.
Unilateral spatial neglect (USN) is characterized by the inability to detect, respond, or orient toward stimuli presented on the contralateral side of a brain lesion [
A variety of accepted and proven traditional methods exist to treat USN [
Virtual reality (VR), defined as “an advanced form of human-computer interface that allows the user to ‘interact’ with and become ‘immersed’ in a computer-generated environment in a naturalistic fashion” [
The European research consortium Rehabilitative Wayout in Responsive Home Environments (REWIRE) developed a nonimmersive VR system for stroke patients using portable devices with good performance and affordable equipment [
Mainetti et al [
We adopted a quasi-experimental pretest-posttest design with a subsequent follow-up to test the feasibility of the exergames in a case series of stroke patients with VSN symptoms. As we aimed to assess implementation of the exergames, thus testing if our intervention can be fully implemented as planned and proposed, an uncontrolled pretest-posttest design is appropriate [
Identification of potential patients for this project was carried out by staff neuropsychologists and occupational therapists in 2 collaborating rehabilitation clinics (Klinik Bethesda Neurorehabilitation, Parkinson-Zentrum, Epileptologie, Tschugg, Bern, and Zürcher RehaZentrum Wald, Faltigberg-Wald, Zurich). They screened all incoming stroke patients with a diagnosis of VSN for eligibility in this study. We aimed for at least 5 participants, as this amount is considered the minimum reasonable number of independent subjects in a group to combine their data [
All patients signed written informed consent before study entry. Ethical approval for the study was received from the local ethics committees (Zurich No. 2014-0543 and Bern No. 389/2014) as well as from the Swiss agency for the authorisation and supervision of therapeutic products (Swissmedic, 2015-MD-0003). The latter approval was required as the software was not yet certified with the European Community marking for medical devices. The study is registered at ClinicalTrials.gov [NCT02353962].
Patients were seated at a table in front of a 21-inch computer monitor at a distance of 60 to 65 cm in order to provide optimal eye tracking (
Setup of the exergames training station.
The intervention program consisted of a series of 9 exergames performed while seated. The games were designed to simulate real-world tasks, such as cooking from a recipe, going for a walk with a dog, or doing a puzzle (for detailed game information on 4 games, see Pirovano et al [
The VR-based VSN training intervention took place in the 2 collaborating rehabilitation clinics serving as an additional therapy option to the standard program, which comprised daily occupational, physical, and neuropsychological therapy. Each patient was asked to attend 5 30- to 45-minute sessions per week for 3 weeks. The supervising therapist individually adjusted the intensity of playing the exergames by changing the difficulty level or game duration in the game menu and by deciding if short breaks between each game would be necessary or not. In accordance with the training principle of individuality [
In order to measure the likelihood and extent to which our intervention can be fully implemented as planned and proposed [
In order to test limited efficacy of our intervention [
Setup of the Eye Tracker Neglect Test.
The ETNT is an adapted version of the cancellation test developed by Rabuffetti et al [
The Neglect Test (NET) consists of 7 paper-and-pencil subtests (letter and star cancellation, copying 3 figures, and line crossing and bisection with a total possible score of 70 points) and 10 behavioral subtests (representational drawing, scanning 3 pictures, menu and article reading, telling time, setting time on a digital and analog watch, and address copying with a total possible score of 100 points) designed to identify a wide variety of visual neglect behaviors [
This anosognosia index (AI) will be smaller than 0 if the patient suffers from anosognosia, indicating an overestimated self-performance to what objectively has been performed. If the patient is able to rate his or her performance realistically, thus being below or matching the external rating, the index becomes equal to or greater than 0, indicating no signs of anosognosia.
Formula for anosognosia index.
The ZüMAX is a domain-specific assessment tool measuring cognitive impairment by evaluating executive function, language, praxia, visual perception and construction, and learning and memory (see Tobler-Ammann et al [
Data analysis was carried out using SPSS for Windows version 23.0 (IBM Corp). A Shapiro-Wilk test was administered and quantile-quantile plots were drawn to test normality of the data. The results confirmed our assumption of nonnormally distributed data due to the small sample size (
The ETNT data were provided by the software described in Rabuffetti and colleagues [
In order to perform an a priori power analysis to determine the minimum sample size for a future randomized controlled trial, we calculated the effect sizes for the secondary outcome measures. We applied the Cohen formula for nonparametric tests [
Cohen formula for nonparametric tests.
From the 18 VSN patients consecutively screened for eligibility in both clinics from March 2015 to March 2016, 7 patients (39%) were eligible and consented to participate in this study, therefore taking part in the VR exergaming program including baseline, postintervention, and 3-month follow-up measures. Reasons preventing patients from participating were suffering from a right-sided VSN due to a left brain lesion, having a severe apraxia (fewer than 5 points on the TULIA (AST) screening instrument, and being in a poor health condition confining them to bed. Patient characteristics are presented in
Patient characteristics.
Patient | Age, years | Sex | Days post-stroke at study entry | RBLa stroke type | Handedness/ affected hand |
Education | Locomotion | CBSb,i | ASTc,j | |
P1 | 64 | M | 25 | ischemia | R/L |
PEd | WCe | 7 | 12 | |
P2 | 74 | M | 29 | ischemia | R/L |
PE | Wf | 17 | 12 | |
P3 | 64 | M | 114 | ischemia | R/L |
PE | WC | 5 | 12 | |
P4 | 70 | M | 32 | ischemia | R/L |
SEg | WC | 6 | 12 | |
P5 | 53 | M | 42 | ischemia | R/L |
PE | W | 5 | 12 | |
P6 | 78 | F | 35 | hemorrhage | R/L |
SE | W | 10 | 12 | |
P7 | 77 | F | 47 | hemorrhage | R/L |
PE | WC | 16 | 9 | |
25 | 64 | — | 29 | — | — | — | — | 5 | 12 | |
50 | 70 | — | 35 | — | — | — | — | 7 | 12 | |
75 | 77 | — | 47 | — | — | — | — | 16 | 12 |
aRBL: right brain lesion.
bCBS: Catherine-Bergego Scale.
cAST: Apraxia Screen of TULIA.
dPE: primary education.
eWC: wheelchair.
fW: walker.
gSE: secondary education.
hIQR: interquartile range.
iMaximum score = 30 (severe neglect); 0 points = no neglect.
jMaximum score = 12 (no apraxia); threshold for apraxia: ˂9 points; severe apraxia: ˂5 points.
An overview of individual (P1-P7) and group (interquartile range [IQR], mean) results in the training protocol is shown in
An overview of the individual ETNT scores and group changes post- to preintervention is shown in
Examples of pre- and postintervention results of the Eye Tracker Neglect Test search paths and fixation points of P3 and P4.
Heat maps of the preintervention, postintervention, and difference post-pre results of the Eye Tracker Neglect Test.
Graphical overview of preintervention, postintervention, and follow-up Zürich Maxi Mental Status Inventory, Neglect Test, and anosognosia index results.
The pre-post assessment showed a median group trend of slight improvement in the total located targets from 15 to 16 (+6%) pre-post, which was concomitant with an increasing median total test duration (+33.9 seconds pre-post) (
P4 improved from 5 to 15 total located targets pre-post intervention (
The NET showed statistically significant improvements pre- to postintervention (
P1 and P2 declined at follow-up in the NET total scores, while P3 to P7 showed further improvement. P1, P5, and P7 significantly changed in their NET total scores, P1 post–follow-up (
The ZüMAX showed improvements in the total scores from pre- to postintervention to follow-up (
On an individual level, P1, 2, 5, and 6 improved from pre- to postintervention in the ZüMAX total scores, while P3 and 4 declined and P7 remained unchanged (
This study evaluated the feasibility of an exergames intervention aimed to affect VSN symptoms in patients early poststroke in terms of implementation (adherence, attrition, and safety) and limited efficacy testing by documenting changes in VSN symptoms. The exergames intervention was tolerated well by all participants and was mainly performed without major difficulties, showing that its implementation in the clinical setting was feasible. With 0 out of 7 (0%) attrition, no adverse events, and a median adherence rate of 14 out of 15 sessions (93%), the compliance of the patients to the exergames was excellent. Such a result was possibly due to the clinic staff’s commitment, as the VR intervention was smoothly integrated into the daily therapy schedule of the clinic. However, as we aimed to test the fit of our intervention in a real-world setting, we prioritized clinic constraints over optimal conditions and settings. As a consequence, this priority reduced potential omissions of training sessions as described in purely home-based VR interventions [
Other studies testing novel VR systems for upper limb stroke rehabilitation have also shown high levels of adherence to the training intervention [
Regarding the exergames training, the median duration per session was 30 (IQR 23-30) minutes, which fell short of the planned 30 to 45 minutes of training time. For our study, we decided to set a timeframe rather than an exact exposure time, because little is known about the optimum duration and patterning of training exposure to virtual environments [
The fact that there were no adverse events during the training period in our case series was encouraging. Besides being lucky that no recurrent stroke or other medical emergency happened during the intervention, the design of the games might have contributed as well to the safety of our participants. For example, implementing both stationary and in-motion conditions of the virtual scenario together with the option of choosing between intertrial variability and no intertrial variability while gaming allowed the patient to choose the optimal virtual environment to be challenged on the one hand but not be overwhelmed on the other hand (see
Our limited efficacy testing showed a group trend of improvements in cognitive and spatial exploration skills. However, these changes cannot be exclusively attributed to the exergames intervention. One reason is the ongoing VSN treatment in the rehabilitation clinic that might also explain some of the improvements. A further possible reason is spontaneous recovery of VSN symptoms not only during the acute phase after stroke but also during the following few weeks. Paolucci et al [
The median AI values in our sample ranged from 0.08 preintervention to 0.0 at follow-up. Comparing those results to bigger RBL stroke samples with USN—mean (SD) lower AI –0.28 (0.5) for n=34 and mean (SD) higher AI –0.47 (0.5) for n=22; Vossel et al [
The ETNT results are to be considered preliminary and should be interpreted with caution. Calibration difficulties with the Eye Tribe Tracker system (eg, most patients only reached 3 out of 5 points in calibration quality scores) may have influenced the reliability of the setup and the accuracy of the results. The calibration consisted of eye-tracking a circle that moved around the whole display. The difficulties experienced by patients in following the rapidly moving circle and the requirement to look at calibration points at the very left of the computer screen were the main reasons for the rather poor calibration results. Such difficulties may produce visible effects (in
The individual search strategy (
We performed an a priori power analysis to determine the minimum sample size for such a future trial [
The length of the training phase was rather short (ie, 3 weeks). We deliberately did not choose a longer training period, as we primarily wanted to test the exergames’ feasibility and not their effect on VSN symptoms. On the other hand, the rather short training phase allowed us to keep the drop-out risk relatively low (eg, due to discharge home during the training phase). In a next step, it would be important to test the exergame system’s feasibility in patient homes to evaluate adherence, safety, and attrition to using the system in this setting, as the provision of novel home-based rehabilitation options was the main goal of REWIRE. In this setting, a longer training phase could be tested. Furthermore, a progression as measured by the game scores should be implemented together with an immediate graphical feedback after each training session to enhance motivation for playing the exergames. For this implementation, ideas from the rehabilitation method of shaping [
The ETNT could further be developed regarding the following:
Calibration procedure of the Eye Tribe Tracker by reducing the speed of the circle to be followed, for example
Software indexes, which were initially designed for the touchscreen (hand-eye coordination) test. Indexes important for eye-tracking would be, for example, the cumulative fixation duration, spatial distribution of fixations in the horizontal and vertical plane, or the number and amplitude of exploratory saccades as explored by Müri et al [
Collection of the search strategy patterns of age-matched controls
Additionally, future work could correlate ETNT measures to scores in standardized clinical scales, such as the NET scores, in order to validate the derived ETNT measures of recovery after VSN.
This study showed that patients adhered well to the REWIRE neglect exergames intervention with no drop-outs, no adverse events, and an adherence rate of 14 out of 15 sessions (93%). We therefore judged this intervention to be safe and feasible for VSN patients early poststroke and appropriate for further testing. Cognitive and spatial exploration skills, as evaluated using ETNT, NET (spatial exploration), and ZüMAX (cognition) assessments, improved in most patients from pre- to postintervention. The results of the amount of exergames use is promising for future applications and warrants further investigations, for example, in the home setting of patients as a motivating training tool to complement usual care and support augmenting training frequency and intensity in RBL stroke patients with VSN.
Overview of individual and group results in the training protocol.
Overview of individual Eye Tracker Neglect Test scores and group changes pre- to postintervention.
Pre- and postintervention results of individual Eye Tracker Neglect Test search paths.
Pre- and postintervention results of individual Eye Tracker Neglect Test fixation points.
Overview of Zürich Maxi Mental Status Inventory, Neglect Test, and anosognosia index scores and group changes preintervention, postintervention, and follow-up.
Graphical display of individual Zürich Maxi Mental Status Inventory, Neglect Test, and anosognosia index scores.
Individual changes in the Zürich Maxi Mental Status Inventory and Neglect Test assessments.
anosognosia index
Apraxia Screen of TULIA
Catherine Bergego Scale
Eye Tracker Neglect Test
interquartile range
Neglect Test
right brain lesion
Rehabilitative Wayout in Responsive Home Environments
test of upper limb apraxia
unilateral spatial neglect
virtual reality
visuospatial neglect
Zürich Maxi Mental Status Inventory
BCT-A designed the study; provided support in designing the exergames; conducted the acquisition, analysis, and interpretation of the data; and wrote the manuscript. EF and MR developed the software and exergames and contributed to the analysis and interpretation of data and writing of the manuscript. LW helped recruit the patients and edited the manuscript. EDdB, NAB, and RHK initiated the study and contributed to design, writing, and editing of the manuscript. All authors read and approved the final manuscript.
This work was partially supported by the REWIRE project (www.rewire-project.eu) and funded by the European Commission under the FP7 framework with contract 287713 and the Occupational Therapy Association of Switzerland. Author MR was supported by funding (Ricerca Corrente) from the Italian Ministry of Health.
Many thanks go to the collaborating clinics who made the patients recruitment possible, namely to Klaus Meyer, MD, Klinik Bethesda, Tschugg, Canton of Berne, Switzerland, and his team of neuropsychologists and to Andreas Sturzenegger, MD, Klinik Wald, Wald, Canton of Zurich, Switzerland, and his team of occupational therapists. Laura Wiederkehr, former master student of the ETH Zurich, is very much appreciated for her scientific and administrative support and for performing the measurements. Further thanks goes to Gabriel Baud Bovy (Istituto Italiano di Tecnologia, Genova) for his work on the Novint Falcon haptic device. Finally, many thanks to Martin J Watson, PhD, for proofreading the manuscript for English and structure.
None declared.