An Evaluation of the Agreement Between a Computerized Stereoscopic Game Test and the TNO Stereoacuity Test

Binocular vision measurements with a new online digital platform: comparison with conventional clinical measures

CLINICAL RELEVANCE

New digital systems are being developed for evaluating different aspects of the visual function, such as binocularity, and it is important to know their real performance in clinical practice in order to use them appropriately.

BACKGROUND

The aim was to compare binocular vision measures obtained with an online digital platform with conventional measures using prisms and printed tests.

METHODS

Prospective study enrolling 49 healthy patients (mean age: 35.5 ± 13.6 years). A complete visual examination was performed including measurement of near phoria (cover test), negative fusional vergence (NFV) and positive fusional vergence (PFV) ranges (prism bar), and stereopsis (24 patients Randot Stereo Test and 25 patients TNO Random Dot Test 19th edition). These same parameters were also measured with the Bynocs system (Kanohi Eye Pvt Ltd). Bland - Altman plots were used to analyse the agreement between methods.

RESULTS

Digital measurement of near phoria was significantly lower than that obtained with the cover test, with a median difference (MD) of 4.71 (-0.07-20.07) prism dioptres (pd) (p < 0.001). No significant differences were found between Bynocs and prism bar methods in NFV break (MD 2.00, range -21-26 pd, p = 0.584) and recovery points (MD 0.00, range -16-24 pd, p = .571). Near PFV were significantly lower with Bynocs (break: MD -9.00, range -38-12 pd; recovery: MD -14.00, range -43-20 pd; p < 0.001). Bynocs stereoacuity threshold was significantly lower than that obtained with TNO (p = 0.004), but significantly higher compared to Randot (p < 0.001). Large and clinically relevant confidence intervals for the comparison between digital and conventional measures were detected in Passing-Bablok analysis.

CONCLUSIONS

Digital measures of near phoria, NFV, PFV, and stereopsis with the Bynocs platform cannot be used interchangeably with conventional measures. The normal ranges of normality for this new tool are defined.

The eRDS v6 Stereotest and the Vivid Vision Stereo Test: Two New Tests of Stereoscopic Vision

Purpose

To describe two new stereoacuity tests: the eRDS v6 stereotest, a global dynamic random dot stereogram (dRDS) test, and the Vivid Vision Stereo Test version 2 (VV), a local or "contour" stereotest for virtual reality (VR) headsets; and to evaluate the tests' reliability, validity compared to a dRDS standard, and learning effects.

Methods

Sixty-four subjects passed a battery of stereotests, including perceiving depth from RDS. Validity was evaluated relative to a tablet-based dRDS reference test, ASTEROID. Reliability and learning effects were assessed over six sessions.

Results

eRDS v6 was effective at measuring small thresholds (<10 arcsec) and had a moderate correlation (0.48) with ASTEROID. Across the six sessions, test-retest reliability was good, varying from 0.84 to 0.91, but learning occurred across the first three sessions. VV did not measure stereoacuities below 15 arcsec. It had a weak correlation with ASTEROID (0.27), and test-retest reliability was poor to moderate, varying from 0.35 to 0.74; however, no learning occurred between sessions.

Conclusions

eRDS v6 is precise and reliable but shows learning effects. If repeated three times at baseline, this test is well suited as an outcome measure for testing interventions. VV is less precise, but it is easy and rapid and shows no learning. It may be useful for testing interventions in patients who have no global stereopsis.

Translational Relevance

eRDS v6 is well suited as an outcome measure to evaluate treatments that improve adult stereodepth perception. VV can be considered for screening patient with compromised stereovision.