VR headset for pupillometry may allow screening for relative afferent pupillary defects
Click Here to Manage Email Alerts
SEATTLE — A virtual reality headset for pupillometry may be able to screen for and detect relative afferent pupillary defects, according to a poster presented at the Association for Research in Vision and Ophthalmology meeting.
Rashed Kashem and colleagues conducted a pupillometry test on 73 participants, 40 with no relative afferent pupillary defect (RAPD) and 33 with a RAPD, 13 in the right eye and 20 in the left eye. The test utilized Heru software on a Pico N3P Eye virtual reality device (ByteDance).
After pupil sizes were normalized by a dim screen, participants underwent a swinging light test in which one eye was lit with a maximum intensity display followed by a period of rest, a process that was repeated in the fellow eye. This established 28 pupillary response parameters, 14 for each eye, that were used to identify the existence and cardinality of RAPD using a histogram gradient boosting classifier, according to the poster.
Outcomes were compared with the ground truth diagnoses to determine the model’s efficiency in classifying RAPD.
The parameters established from the virtual swinging light test demonstrated statistical significance in distinguishing between no RAPD, RAPD in the left eye and RAPD in the right eye (P < .0001). The histogram gradient boosting classifier correctly diagnosed 19 of the 20 participants with RAPD in the left eye, 12 of the 13 participants with RAPD in the right eye and all 40 participants with no RAPD.
The tool’s accuracy was 0.97, sensitivity was 0.94 and specificity was 1.00.
“Our novel diagnostic model utilized a portable solution that may allow clinicians to screen for RAPD,” the authors wrote. “The statistically significant difference between the parameters shows that this test can detect the existence and cardinality of RAPD with relatively high accuracy.”