Abstract
Purpose:
An adequate visual field (VF) is a requisite for safe driving and effects of glaucoma on motor vehicle accident (MVA) involvement have been vigorously studied. A driving simulator (DS) is often used to create controlled conditions for evaluation of driving performance, but most DSs are not equipped with gaze monitoring system which is indispensable for correlating VF abnormalities to MVA involvement. We studied effects of glaucomatous VF on collision using a DS systen equipped with a gaze-tracking system.
Methods:
A modified version of Honda safety Navi DS System (Honda Motor Co., Tokyo) which reproduces the driver's view from a right-hand-drive car on the screen 114 cm apart was equipped with a gaze-tracking system (NAC EMR-9, nac Image Technology, Tokyo), and the driver's gaze point (DGP) was monitored in real-time on a personal computer display which also reproduced the driver's view. The instant when the driver first detected the hazard was determined by a saccadic movement of the DGP toward the hazard. The driver's binocular integrated VF (IVF) calculated by merging the 2 monocular Humphrey Field Analyzer 24-2 Swedish Interactive threshold Algorithm (Carl Zeiss Meditec., CA) test results and centered on the DGP just before detecting the hazard was projected onto the driver's view. 52 glaucoma patients (66.2+/-9.2 yrs.; mean deviation (MD) of the better eye=-8.1+/-6.3 dB; mean IVF sensitivity=22.0+/-6.0 dB) participated in the DS experiment and were required to avoid an oncoming right-turning car, the hazard, at an intersection. The Ethic Committee of Gifu Prefecture Medical Association approved the study.
Results:
15 of the 52 collided with the hazard. It took longer time for the collision-involved examinees to detect the hazard than those not (0.65+/-0.66 vs. 0.25+/-0.30 sec, P＝0.012). The decision tree method correlated 3 IVF test points to the collision and sensitivity was lower in the former than the latter at a point just inferior and right to the fixation (22.0+/-3.8 vs. 29.8+/-5.8 dB, P＝0.016). Visual acuity and MD of the dominant eye were also correlated with the collision (k=0.59, Partition, JMP Pro11.0,SAS)
Conclusions:
A DS with a gaze-tracking system was developed, which enabled us to study the location and sensitivity of the patient's IVF subfield related to MVA involvement in a given scenario of DS.