Leveraging virtual reality for vestibular testing: Clinical outcomes from tests of dynamic visual acuity

Abstract

BACKGROUND:

Virtual reality (VR) use as a platform for vestibular rehabilitation is widespread. However, the utility of VR based vestibular assessments remains unknown.

OBJECTIVE:

To compare dynamic visual acuity (DVA) scores, perceived balance, and perceived dizziness when using traditional versus VR environments for DVA testing among healthy individuals.

METHODS:

DVA testing occurred for both a traditional clinical protocol and in a VR variant. Horizontal, vertical, and no head motion conditions were conducted for both clinical and VR test protocols. DVA scores, balance ratings, and dizziness ratings were obtained per condition. Two-way ANOVAs with repeated measures were used to assess differences in DVA scores, balance, and dizziness ratings.

RESULTS:

No differences in DVA results, balance or dizziness ratings were observed when comparing traditional clinical protocol versus the VR variant. Differences across head motion conditions were observed, with no motion trials exhibiting significantly higher DVA scores and perceived balance, and lower perceived dizziness compared to vertical and horizontal head motion. Vertical head motion exhibited this same trend compared to horizontal.

CONCLUSION:

DVA testing conducted in VR demonstrated clinical utility for each measure. Effects of head motion were similar across test variants, indicating DVA testing in VR produces similar effects on vestibular function than traditional clinical testing. Additional research should be conducted to assess the feasibility of VR assessment in individuals with vestibular disorder.

Keywords

Vestibular testing dynamic visual acuity virtual reality

1 Introduction

According to data from the National Health and Nutrition Examination Survey taken from 2001–2004, nearly 70 million adults (40 years or older) were afflicted with vestibular (inner ear/balance) disorder [1]. Vestibular disorders are commonly associated with symptoms of dizziness, and visual blurring or gaze instability [3], which can negatively impact postural stability. These symptoms contribute to 12 times greater risk of falling in people with vestibular dysfunction, contributing to heightened morbidity, particularly in older populations [20].

Oscillopsia can be described as disruption of dynamic visual acuity (DVA), or the ability of an observer to perceive the sharpness or clarity of a visual target while the observer is moving and occurs as a result of diminished gain associated with the vestibule-ocular reflex. Tests of DVA are frequently employed as a means to assess vestibule-ocular reflex functioning in individuals with vestibular disorders [3, 8]. Generally, DVA tests are administered using an optometric chart (Snellen) physically displayed in front of the patient [3] or portrayed on a computer screen [9, 15]. Common variations of the DVA testing involve patients identifying a series of optotypes [9, 15], or the orientation of optotypes [8, 16] during conditions involving volitional head motion, or head motion at a set frequency. Head motion conditions are designed to assess the integrity of the vestibule-ocular reflex, which directly affects visual acuity, and provides an indication of the level of vestibular functioning [14]. The format of traditional clinical DVA testing lends itself to adaptation for virtual reality (VR) environments, which could allow for more efficient and cost-effective means for conducting vestibular assessment in conjunction with VR based interventions for vestibular disorders, particularly in home-based therapy [2].

With advancements in technology, VR applications have become an intriguing intervention and assessment tools for a variety of neurophysiological conditions, including vestibular disorders [2]. Several studies have shown that VR interventions can improve measures of balance and gait in people with vestibular disorders [5, 6]. Yet, fewer studies have focused on the feasibility and practicality of adapting traditional vestibular assessment tests for VR. Recently, Lubetzky & Hujsak [11] determined that head-mounted VR systems had the clinical utility to measure differences in head stability between individuals with vestibular dysfunction and healthy controls. Additionally, Uloziene et al. [19] used a mobile VR system to assess vestibular function in healthy individuals and reported similar scores compared to traditional clinical test variants. Similar to the current study, Marozas et al. [17] adapted DVA testing into a VR environment using the Oculus Rift (Oculus VR Inc., USA), demonstrating feasibility of implementation, but did not report outcomes of VR testing compared to traditional clinical testing. These studies represent promising initial efforts to integrate vestibular assessment into VR environments. Yet, despite these efforts, there remains a void in understanding the utility of VR based DVA testing compared to traditional clinical testing. In addition, it is important to consider how VR might impact the user’s perception of balance as well as propensity for dizziness during VR DVA testing.

The increased use of VR for clinical and home-based vestibular rehabilitation interventions warrants the development of assessment tools adapted for virtual environments as a way of efficiently measuring patient progress in conjunction with VR interventions. Here, we aim to leverage VR to administer the DVA test. The aim of this study was to assess the utility of VR based DVA testing by comparing DVA scores, perceived balance, and perceived dizziness outcomes obtained from both traditional and VR administered DVA tests. We hypothesize that a virtual environment designed to mimic traditional clinic DVA testing would yield similar DVA test score outcomes in healthy individuals, with no vestibular impairment. Additionally, we hypothesize that participant’s perceived balance and dizziness during DVA testing in the virtual environment would be unaffected when compared to traditional clinical DVA testing.

2 Methods

2.1 Design and population

The study design was a repeated-measures experimental examination of 26 healthy adults. Participants were recruited by word of mouth and flyer, as a sample of convenience, within the School of Kinesiology and Recreation at Illinois State University. Participants were considered eligible if they were: (1) aged 18 years or more; (2) a current student, staff, or faculty member of Illinois State University; and (3) not diagnosed with a balance disorder, neuromuscular disorder, or other similar impairment in the past 18 months. Participants with corrected vision were included only if they had contacts available for data collection.

All patients provided written informed consent before participating in the study and the study (#IRB-2018-348) was approved by the institutional review board at Illinois State University. Participants reported to the laboratory on one occasion for data collection.

2.2 Instrumentation

The DVA test instrument (The American Institute of Balance®, Largo, FL, USA) was implemented in VR and traditional clinical (TC) environments as an indirect measure of vestibular function. The structure of DVA testing for VR and traditional environments was similar to previous research [15], requiring participants to read numerical optotypes presented on screen from left to right under a no head motion condition, followed by a vertical head motion condition, and then a horizontal head motion condition. Optotypes appeared in sets of five random numbers, ranging from 0 to 9. Ten sets of numbers were presented for each trial. For each set, optotypes were displayed for 3 seconds in white Arial font against a black background. Font size for each of the 10 sets of optotypes varied between 12 to 20 points with two-point increments, in accordance with previous research [15]. For head motion conditions, a metronome (2 Hz) was implemented to guide the rate of volitional vertical or horizontal head oscillations produced by the participant.

The TC variant of the DVA was conducted on a 15.6” screen Dell Inspiron 15 laptop (Dell®, Intel® i7-7700HQ CPU, Round Rock, TX, USA), with a screen resolution of 1920×1080, placed 1.2 meters in front of the participant at eye level. The Oculus Rift head-mounted device (Oculus®, Menlo Park, CA, USA) was used for the VR component of this study. The VR display has a resolution of 1080×1200 for each eye, and a field of view of 110 degrees. Participants wore the headset during the VR trials. The VR application that replicates the DVA test is designed and tested using Unity (Unity Software Inc., San Francisco, CA, USA). The VR variant of the DVA placed an individual in a fully black environment with letters simulated at approximately 1.2 meters.

Additionally, a two-item questionnaire was implemented to gauge participant’s perceived balance and dizziness during DVA testing. To assess participant’s perceived ability to maintain balance (perceived balance) during DVA testing protocol, participants were asked: “from 0–100%, how confident were you that you would not lose balance or become unsteady while performing the DVA test?”. To assess perceived dizziness during DVA testing, participants were asked: “on a scale of 0–5, 0 being not dizzy at all and 5 being the worst dizziness you have ever experienced, how dizzy are you?”.

2.3 Data collection

Upon arrival to the lab, participants were asked to fill out a consent form and provide demographics information. Then, participants were asked to wear the head-mounted device for as much time as needed to express self-reported comfort and familiarity with the device. Once the participant felt comfortable in the head-mounted device, they were asked to rest at minimum of two minutes before beginning the test.

Data collection involved assessing participants across two DVA test variant conditions and three head motion conditions. The DVA test variants consisted of: 1. Traditional clinical (TC) test performed while standing. 2. VR variant performed while standing. The order of DVA test variant was randomized for each participant. During the TC condition, overhead lights were turned off in the lab space.

The head motion conditions consisted of: 1. No head motion. 2. Vertical head motion. 3. Horizontal head motion. Head motion conditions were performed for each DVA test variant condition (VR and TC) in a set order of no head motion, vertical head motion, then horizontal head motion. The rate of head motion was guided by a metronome set at 2 Hz, and the arc of motion was approximately 60 degrees. Trials consisted of 10 individual sets of 5 number strings, displayed for 3 seconds per set, and were performed continuously with a mandatory minimum 2-minute rest between each condition. Three trials were performed for each head motion condition, for a total of 18 trials across test variant conditions. Performance during DVA testing was measured by dividing the number of correct read responses by the total number of optotypes presented during the trial (50 total optotypes) and multiplying by 100 to yield a percentage score [15]. Following each trial, research assistants administered the perceived balance and perceived dizziness questions and recorded participant’s responses. Assistants conducting the scoring were trained by a practicing physical therapist with experience in vestibular rehabilitation. Finally, there were no reports of optotype blurredness during the no head motion condition in VR, signifying that participants clearly recognized the optotypes without blurredness in the static VR environment.

2.4 Statistical analysis

Independent variables consisted of the two factors (DVA test variant and head motion). Dependent variables included scores from the DVA, and ratings from perceived balance and dizziness questions. Assessment scores were transcribed into SPSS software (version 20.0, SPSS Inc, Chicago, IL, USA) and an alpha level was set a priori at α= 0.05. Two-way repeated measures ANOVAs were conducted to test for interactions between factors and within-subject differences for each factor. Post-hoc comparisons were performed with Bonferroni adjustment if significant main effects or interactions were observed. Cohen’s d effect sizes were calculated for statistically significant pairwise comparisons. Effect sizes were categorized as small (< 0.5), medium (0.5 to 0.79) and large (≥0.8).

3 Results

Sixteen males and 10 females (age: 24±5.5 yrs.; height: 1.68±0.12 m; mass: 73.3±17.7 kg) completed this study. Average outcome data for each condition are presented in Table 1. No significant interactions were observed for each of the dependent variables across conditions. Additionally, no main effects were observed for test variant (TC vs VR) for each of the dependent variables.

Table 1
Average clinical outcome data across conditions (Mean±SD)

Clinical Variant VR Variant

No Motion Vertical Horizontal No Motion Vertical Horizontal

DVA 99.2±1.8% 94.8±5.9% 90.2±11.4% 99.6±1.13% 95.9±5.61% 89.7±16.2%

Balance Rating 99.4±2.2% 96.1±10.5% 92.0±15.6% 98.6±4.4% 96.9±5.8% 94.7±8.5%

Dizziness Rating 0.08±0.39 0.54±0.86 1.1±1.3 0.23±0.65 0.40±0.57 0.87±1.2

	Clinical Variant	VR Variant
DVA	99.2±1.8%	94.8±5.9%	90.2±11.4%	99.6±1.13%	95.9±5.61%	89.7±16.2%
Balance Rating	99.4±2.2%	96.1±10.5%	92.0±15.6%	98.6±4.4%	96.9±5.8%	94.7±8.5%
Dizziness Rating	0.08±0.39	0.54±0.86	1.1±1.3	0.23±0.65	0.40±0.57	0.87±1.2

DVA –Dynamic visual acuity.

The DVA scores showed a significant main effect for head motion condition (p < 0.01) (Fig. 1). Post hoc analysis showed that DVA scores during no head motion trials (99.4±0.1%) were significantly higher compared to vertical motion (95.9±1.1%) (p < 0.01; d = 0.64) and horizontal head motion (89.7±2.4%) (p < 0.01; d = 0.81). Additionally, DVA scores during vertical head motion trials were significantly higher compared to horizontal head motion (p < 0.01; d = 0.52).

Fig. 1

Differences in DVA score due to Head Motion Condition.

The perceived balance results showed a significant main effect for head motion condition (p = 0.011) (Fig. 2). Post hoc analysis showed that no head motion trials (99.2±0.5%) led to significantly higher balance confidence scores compared to vertical head motion (97.2±1.3%) (p < 0.01; d = 0.31) and horizontal head motion (93.9±1.9%) (p < 0.01; d = 0.56). Vertical head motion scores were also significantly higher than horizontal head motion scores (p = 0.024; d = 0.35).

Fig. 2

Differences in perceived balance rating due to Head Motion Condition.

The perceived dizziness results showed a significant main effect for head motion condition (p < 0.001) (Fig. 3). Post hoc analysis showed that no motion trials (0.16±0.07) led to significantly lower perceived dizziness ratings than both vertical motion (0.51±0.14) (p = 0.01; d = 0.50) and horizontal motion (1.10±0.21) (p < 0.01; d = 0.90) trials. Additionally, vertical motion perceived dizziness ratings were significantly lower than horizontal motion ratings (p < 0.01; d = 0.60).

Fig. 3

Differences in perceived dizziness rating due to Head Motion Condition.

4 Discussion

This study assessed the utility of a VR-based test of DVA compared to traditional clinical testing. In addition, effects of VR on self-reported measures of balance and dizziness were included. The VR variant of the DVA test showed clinical utility, as no differences in the clinical metrics were observed. Differences were observed across head motion conditions for DVA, perceived balance, and perceived dizziness, indicating head motion conditions had a significant impact on participant’s measured and perceived outcomes, regardless of test variant condition.

The DVA test takes advantage of vestibular anatomy to isolate the stimulation of semi-circular canals (SCC). The vertical motion should challenge the anterior and posterior SCC, while the horizontal motion targets the horizontal canal, and the no motion trial serves as a control. If vestibular function is disrupted, the vestibule-ocular reflex will be impacted and an individual will be less able to maintain visual focus [7]. In the current study, head motion condition had a clear effect on DVA test score performance, as well as perceived balance and dizziness. Specifically, participants consistently scored higher on the DVA test and reported having better balance and less dizziness during no head motion conditions, compared to vertical and horizontal head motion conditions. Further, horizontal head motion conditions consistently yielded poorer outcomes compared to no motion and vertical head motion conditions, respectively, which is in agreement with previous research [14]. These results were independent of test variant conditions, indicating statistically similar outcomes during clinical and VR test variants.

The absence of differences between clinical and VR test variants across all head motion conditions suggests that DVA testing in VR may be a suitable alternative to traditional DVA testing conducted in a clinical setting. This is particularly useful for clinicians and researchers using virtual environments for vestibular rehabilitation interventions. Specifically, incorporating DVA testing in conjunction with VR interventions may improve the efficiency and accessibility of collecting assessment data, particularly through home-based therapeutic interventions.

4.1 Limitations

Several limitations of this study should be men-tioned. First, although results from subjective single-item assessments of balance and dizziness were favorable for the VR condition, these assessments have not previously been validated. Therefore, interpretations surrounding balance and dizziness ratings from this study should be made with caution. Second, college-aged students were chosen as a sample of convenience, which introduces sampling bias. This limits the generalizability of the results to older populations and patients with vestibular dysfunction, who experience higher fall rates than younger individuals [4, 20]. However, older populations have previously demonstrated acceptance of VR technology and minimal cybersickness with use [10], and patients with vestibular loss have demonstrated lower motion sickness susceptibility compared to healthy individuals [12]. Therefore, it is feasible that VR based DVA testing may have clinical utility for older patients with vestibular disease.

4.2 Conclusions

DVA testing adapted for use in a head-mounted VR environment exhibited clinical utility, with participant’s exhibiting similar scores to traditional clinical DVA testing. Furthermore, perceived balance and dizziness ratings were unaffected during DVA testing in the VR environment. DVA testing adapted for VR environments may provide an efficient and effective means of measuring patient progress during VR based vestibular rehabilitation interventions. Future studies are needed to assess the utility of the VR based DVA testing in older populations and in individuals with vestibular disorders, and should consider use of valid tests of motion sickness (e.g. simulator sickness questionnaire) and balance (e.g. center of pressure deviation) to rule out negative impacts of VR use.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the author on reasonable request.

Declarations of interest

The authors declare that there is no conflict of interest.

Funding

This research received no specific grant(s) from any funding agency in the public, commercial, or not-for-profit sectors.

References

Agrawal

, Ward

B.K.

, Minor

L.B.

, Vestibular dysfunction: Prevalence, impact and need for targeted treatment, J Vestib Res Equilib Orientat 23(3) (2013), 113–117.

Bergeron

, Lortie

C.L.

, Guitton

M.J.

, Use of Virtual Reality Tools for Vestibular Disorders Rehabilitation: A Comprehensive Analysis, Adv Med 2015 (2015), 1–9.

Bhansali

S.A.

, Stockwell

C.W.

, Bojrab

D.I.

, Oscillopsia in patients with loss of vestibular function, Otolaryngol. - Head Neck Surg 109(1) (1993), 120–125.

Bird

M.L.

, Pittaway

J.K.

, Cuisick

, Rattray

, Ahuja

K.D.K.

, Age-related changes in physical fall risk factors: Results from a 3 year follow-up of community dwelling older adults in Tasmania, Australia, Int J Environ Res Public Health 10(11) (2013), 5989–5997.

Corbetta

, Imeri

, Gatti

, Rehabilitation that incorporates virtual reality is more effective than standard rehabilitation for improving walking speed, balance and mobility after stroke: A systematic review, J Physiother 61(3) (2015), 117–124.

de Rooij

I.J.M.

, van de Port

I.G.L.

, Meijer

J.-W.G.

, Effect of Virtual Reality Training on Balance and Gait Ability in Patients With Stroke: Systematic Review and Meta-Analysis, Phys Ther 96(12) (1918), 1905–1918, Dec.

Fetter

, Vestibulo-Ocular Reflex, in Neuro-Ophthalmology, Basel: KARGER, 2007, pp. 35–51.

Herdman

S.J.

, Computerized dynamic visual acuity test in the assessment of vestibular deficits, in Vertigo and Imbalance: Clinical Neurophysiologyof the Vestibular System vol. 9, Elsevier, 2010, pp. 181–190.

Hillman

E.J.

, Bloomberg

J.J.

, McDonald

P.V.

, Cohen

H.S.

, Dynamic visual acuity while walking in normals and labyrinthine-deficient patients, J Vestib Res 9(1) (1999), 49–57.

10.

Huygelier

, Schraepen

, van Ee

, Vanden Abeele

, Gillebert

C.R.

, Acceptance of immersive head-mounted virtual reality in older adults, Sci Rep 9(1) Dec. 2019.

11.

Lubetzky

A.V.

, Hujsak

B.D.

, A virtual reality head stability test for patients with vestibular dysfunction, J Vestib Res Equilib Orientat 28(5–6) (2019), 393–400.

12.

Paillard

A.C.

, Quarck

, Paolino

, Denise

, Paolino

, Golding

J.F.

, Ghulyan-Bedikian

, Motion sickness susceptibility in healthy subjects and vestibular patients, Journal of Vestibular Research 23 (2013), 203–209, Jan.

13.

Robert

M.T.

, Ballaz

, Lemay

, The effect of viewing a virtual environment through a head-mounted display on balance, Gait Posture 48 (2016), 261–266, Jul.

14.

Roberts

R.A.

, Gans

R.E.

, Comparison of horizontal and vertical dynamic visual acuity in patients with vestibular dysfunction and nonvestibular dizziness, J Am Acad Audiol 18(3) (2007), pp. 236–244, Mar.

15.

Roberts

R.A.

, Gans

R.E.

, Johnson

E.L.

, Chisolm

T.H.

, Computerized dynamic visual acuity with volitional head movement in patients with vestibular dysfunction,pp, Ann Otol Rhinol Laryngol 115(9 I) (2006), pp. 658–666.

16.

Schubert

M.C.

, Herdman

S.J.

, Tusa

R.J.

, Vertical Dynamic Visual Acuity in Normal Subjects and Patients with Vestibular Hypofunction, Otol Neurotol 23(3), 2002.

17.

Stanaitis

, et al., Virtual reality based system for investigation of peripheral vestibular function, in IFMBE Proceedings 57 (2016), 110–113.

18.

Takahashi

, Murata

, Change of equilibrium under the influence of VR experience, in Proceedings - IEEE International Workshop on Robot and Human Interactive Communication (2001), pp. 642–647.

19.

Ulozienė

, et al., Subjective visual vertical assessment with mobile virtual reality system, Med 53(6) (2017), 394–402, Jan.

20.

Ward

B.K.

, Agrawal

, Hoffman

H.J.

, Carey

J.P.

, Della

C.C.

, Santina, Prevalence and impact of bilateral vestibular hypofunction: Results from the US national health interview survey, JAMA Otolaryngol - Head Neck Surg 139(8) (2013), 803–810, Aug.

	Clinical Variant			VR Variant
	No Motion	Vertical	Horizontal	No Motion	Vertical	Horizontal
DVA	99.2±1.8%	94.8±5.9%	90.2±11.4%	99.6±1.13%	95.9±5.61%	89.7±16.2%
Balance Rating	99.4±2.2%	96.1±10.5%	92.0±15.6%	98.6±4.4%	96.9±5.8%	94.7±8.5%
Dizziness Rating	0.08±0.39	0.54±0.86	1.1±1.3	0.23±0.65	0.40±0.57	0.87±1.2