Pupillary Complexity for the Screening of Glaucoma

Abstract

Glaucoma is a leading cause of irreversible visual field loss. The early detection and diagnosis of the disease are therefore necessary to prevent blindness. Pupillary light responses are an interesting new technique for the detection of glaucoma. However, the analysis of pupillary signals has been associated with manual supervision or involved high computational costs. The present paper is to propose an analysis framework to automatically investigate changes in the complexity of pupillary signals under ambient light conditions for the screening of glaucoma. In this work, pupillary data of 13 glaucoma patients, 13 age-matched controls, and 11 young controls were recorded at the light intensity of 100 cd/m2 using a commercial eye tracker. The pupillary complexity of the participants was analysed using Higuchi’s fractal dimension, permutation entropy, and conditional entropy. We found that there was a statistically significant difference in the pupillary complexity between glaucoma patients and control groups (P < 0.0001). Specifically, the difference was more pronounced when using the fractal dimension measure. These results confirm the potential of using pupillary complexity for the screening of glaucoma using commercial devices.