Studying pupillary light reflex in young children with autism spectrum disorders

Abstract

[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI AT AUTHOR'S REQUEST.] Pupillary light reflex (PLR) is a phenomenon wherein the pupil size changes according to retina illumination. PLR of adults and children has been investigated in various studies. However, its significance in the medical care of young children and infants has been limited due to the lack of reliable instrumentation. Traditional PLR devices need to restrain subject or have to be worn on the head, which is not suitable for young children. A remote PLR measuring device was developed which allows the subject to sit far from the device during the PLR test. The device can capture high resolution images at a high sample rate similar to what can be found in a desktop system, but without the need for a restraint. The device follows the subject by tracking their eye movements during the PLR test. The device has been used to test PLR in typical developing (TD) children and children with autism spectrum disorders (ASDs), who were between ages of two to six years old. It was found that PLR latency in children with ASDs was longer than in typical developing children. The PLR constriction of the ASD subjects was slightly smaller than the TD subjects, but that difference was not statistically significant in our test. Moreover, the base pupil radius increased and latency decreased with age in the TD children but not in the ASD group. We found that PLR parameters are associated with autonomic nervous system (ANS) dysfunction. TD children with abnormal sweating issues showed a larger PLR constriction and smaller base pupil radius. We found that children with ASD did not show such differences. However, constriction time is negatively correlated with the normalized high frequency component of heart rate variability power spectrum in the ASD group, but not in the TD group. Nine TD children and eight ASD children were tested again 1-1.5 years from their first test date. The PLR parameter differences between the first and second tests were calculated, and we found that latency difference was negatively correlated to base pupil radius difference. Moreover, constriction difference was found to be positively correlated to the base pupil radius difference in TD children, but not in the ASD group. Remote PLR device camera setup was changed to capture the test subject's motion by a one camera while the other camera captures a high resolution pupil image. Because the two cameras are off axis, 3D location of the eye position is directly calculated using stereo vision base method, which helped to remove the need of physical measurement. The entire space occupied by the new rPLR device is about 50% of the previous device.