Visual Field Testing on a Personal Smartphone

Abstract

Purpose : Standard visual field testing is performed on expensive equipment with limited portability and requires dedicated facilities. The goal of this study is to develop and evaluate a low-cost Portable Perimeter (PP) that can be used anywhere (e.g., bedside, home).

Methods : The PP consists of a standard personal smartphone (Nexus 6P), a virtual reality viewer (Google Cardboard 2015) and a mobile app that presents visual stimuli and determines differential light sensitivity thresholds by using the 4-2 double reversal staircase algorithm. The app is loaded from a central server with pre-calibrated parameters for the smartphone’s display and VR viewer. At the end of the test, the recorded thresholds are sent to the server to compute the Mean Deviation (MD), Pattern Standard Deviation (PSD), Visual Field Index (VFI) and standard reliability indices, and to generate standard reports that are sent wirelessly. The PP provides a user-friendly interface with an introductory program and real-time voice feedback on fixation losses. To evaluate the performance of the PP, we tested 19 consenting patients from the Toronto Western Hospital glaucoma unit. Each patient underwent the Humphrey Field Analyzer (HFA) 24-2 SITA standard test followed by the PP test. In addition seven patients underwent a second PP test to determine inter-test variability. For each patient one eye was tested.

Results : We found a good correlation between the HFA and PP for thresholds (R²=0.64) and excellent agreement for MD (R²=0.87), PSD (R²=0.89) and VFI (R²=0.92). The reliability indices were similar for false positives and negatives, but there were significantly more fixation losses with the PP (22% vs 7.2%, p=0.01). Further analysis revealed that four patients had abnormally high percentages of fixation losses on the PP (fixation loss>56%). In the patients who repeated the PP we found good agreement between the tests for thresholds (R²=0.61), MD (R²=0.98), PSD (R²=0.96), VFI (R²=0.97) and reliability indices. Test duration was significantly longer with PP (8.6±1.0 mins) compared to HFA (5.7±1.1, p<0.001).

Conclusions : Preliminary results show good agreement between the HFA and the PP. Test duration can be improved by using a more sophisticated thresholding technique and fixation losses can be improved by better set-up procedures. The performance,affordability and portability of the PP could make it a valuable tool for visual field testing.

This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.