Purchase this article with an account.
elaine tran, Ling Wan, William Yan, Yang Sun, Robert Chang; Comparison of Virtual Reality (PalmScan VF2000) Visual Fields Analyzer with Humphrey Visual Field in Glaucoma Patients. Invest. Ophthalmol. Vis. Sci. 2020;61(7):3893.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
This study examines the correlation of a commercially available, smartphone-based virtual reality (VR) automated perimetry test called the PalmScan VF2000 visual fields analyzer (Micro Medical Devices, Calabasas) in glaucoma patients undergoing traditional Humphrey Visual Field (HFA). The PalmScan VF2000 administers a 24-2 full threshold test within the VR headset, with the highest sensitivity detectable by patients recorded for 54 pre-designed points within a central 24 degrees of visual field.
Glaucoma and glaucoma suspects were recruited by consecutive series at the Byers Eye Clinic at Stanford. One or both eyes were tested using the PalmScan VF2000 device twice, within two weeks of Humphrey Field Analyzer (HFA) testing. Eyes with most reliable parameters and most severe perimetric defect were selected for comparison. Primary outcome data collected included mean deviation (MD) and pattern standard deviation (PSD) from the PalmScan software. Mean Deviation (MD) was selected for comparison against the HFA reference standard.
A total of 41 patients were recruited (M 42%, F 58%, Mean Age 62), including eyes with 28 known glaucoma and 13 glaucoma suspects among experienced field takers. Mean deviation on VF2000 ranged from -25.4 to +1.1 with average -5.9 dB. The average difference between VF2000 and HFA MD was found to be -0.17 db. A paired t-test for zero bias was not statistically significant (mean -0.17; CI -1.02 to 0.67; p=0.68), nor was correlation between the bias and the magnitude of the mean of both measurement types (r=0.10, p=0.53). A Bland-Altman plot is included to illustrate assessment of the magnitude of disagreement.
The Palmscan VF2000 mean deviation were found to be comparable with HFA mean deviation results. We found less than 0.2 db difference, with no suggested correlation with increasing magnitude of measurement, indicating that results may be consistent even with increasingly severe disease.
This is a 2020 ARVO Annual Meeting abstract.
Figure 1. Bland-Altman plot of the difference between HFA and PalmScan measurements, vs. the mean of the two methods for each patient
This PDF is available to Subscribers Only