Pupil examination is a routine part of ophthalmological and neurological assessment. The authors evaluated whether automated pupillometry implemented in a virtual reality headset could reliably detect relative afferent pupillary defect (RAPD) and anisocoria and how its performance compared with standard clinical pupil testing.
The device used was the HP Reverb G2 Omnicept VR headset. Two VR-based protocols were applied: a Pupil symmetry test and a Swinging light test.
Twenty-nine patients and 19 healthy controls underwent VR pupillometry and a routine clinical pupil examination. Pathological thresholds for the VR measures were defined using the upper limits observed in the control group. Results from VR testing were compared directly with findings from the conventional clinical exam. Correlations between pupillary metrics and functional ophthalmological tests (visual acuity and visual field mean deviation, MD) were analyzed.
The VR assessment included measurements of inter-eye pupillary size differences and dynamic responses during the Swinging light test. The authors report using control-derived thresholds to classify RAPD, relative afferent pupillary asymmetry (RAPA), and pathological anisocoria.
Participants were divided as follows:
Control measurements established reference values: RAPA within 0.1–0.2 mm was considered within control range. Anisocoria values observed in controls ranged from 0.4–0.8 mm and were on average greater in darkness by a mean of 0.2 mm.
In Group I (asymmetric optic neuropathy), RAPD was detected in 16 of 21 patients by both VR pupillometry and routine clinical exam. The remaining five patients showed relative afferent pupillary asymmetry (RAPA) values that fell within the control range (0.1–0.2 mm), and thus were not classified as pathological by the study thresholds.
These findings indicate that VR pupillometry matched the conventional exam in identifying clinically significant RAPD in this cohort.
Quantitative measures of RAPA/RAPD derived from VR pupillometry correlated with inter-eye differences in functional ophthalmological tests. Specifically, correlations with inter-eye differences in visual acuity and visual field mean deviation (MD) were reported as R=0.58 and R=0.68 respectively, with p<0.01 for both correlations.
The authors note that all patients with an inter-eye MD difference greater than 5.8 dB had RAPD by testing, suggesting a relationship between the magnitude of visual field asymmetry and detectability of RAPD with these methods.
In Group II (suspected pathological anisocoria), VR pupillometry detected pathological anisocoria defined as greater than 0.8 mm. By comparison, routine clinical exam identified pathological anisocoria in only 1 of the 8 patients in this group. The VR method also revealed qualitative differences in pupillary constriction patterns not specified in numerical detail in the abstract.
Among controls, anisocoria in the 0.4–0.8 mm range was present in 37% (7/19). The authors observed that these control anisocoria values were greater in dark conditions by a mean of 0.2 mm, illustrating physiological variability with lighting.
Overall, the study reports that VR pupillometry matched the conventional clinical exam for RAPD detection and surpassed it for detection of pathological anisocoria in the tested cohorts. The VR platform provided objective, quantifiable inter-eye size differences and dynamic response data from the Swinging light test implemented in the headset.
The abstract highlights the potential advantage of VR-based longitudinal recordings of pupillary constriction characteristics for diagnosis and monitoring, although specific longitudinal results were not detailed in the abstract.
The authors suggest that automated VR pupillometry could complement or enhance routine pupil assessment by providing objective, quantitative detection of RAPD and improved sensitivity for pathological anisocoria. Longitudinal pupillary response data may offer additional diagnostic or monitoring value in clinical studies and could aid in understanding physiological variability of pupil size under different lighting conditions.
Ethical approval was obtained from the Republic of Slovenia National Medical Ethics Committee (decision No. 0120-644/2025-2711-3). The authors confirm that patient consent and other relevant ethical procedures were followed. All study data are available upon reasonable request to the authors.
Declared funders include the Slovenian Research and Innovation Agency (J3-1750) and Ljubljana University Medical Centre (20220038). The authors declared no competing interests.
This summary is based on the medRxiv preprint abstract and associated metadata. Detailed methodological parameters (for example the exact VR stimulus protocol timing, measurement precision, reproducibility metrics, and full numerical breakdown of anisocoria detection by subject) were not reported in the abstract. The study is a preprint and has not been peer reviewed; the authors note that the findings should not be used to guide clinical practice without further validation. For full methodological and result details, readers should consult the full preprint or contact the authors for data.