Establishing the operating conditions of Ocula AI in capturing the pupil light reflex

In recent years, rapid advances in technology have made it possible to monitor human health with personal handheld devices. While initially limited to measures of electrical skin potential for cardiac assessment, this has now expanded to include brain health by leveraging visual reflexes. In this study, we examine the effectiveness of a cutting-edge smartphone application, Ocula AI (Equinox), to capture and quantify the pupillary light reflex (PLR). In particular, the present study evaluates and compares the capability and operating range of Ocula AI, against an established clinical-standard device, the PLR-3000 pupillometer (NeurOptics). Both devices capture the PLR waveform providing estimates of key metrics such as such as latency, maximum and minimum pupils, and constriction and dilation velocities. The ability of Ocula AI to capture the PLR was assessed under different indoor illumination conditions (indicated by illuminance levels ranging from 0 to 1200 lux) in 16 healthy young adults. Our comparison and Bland-Altman analyses showed that Ocula AI effectively and reliably captured the PLR and estimated key features of the PLR waveform to a similar standard and with high agreement to the PLR-3000 device. Furthermore, Ocula AI was capable of capturing the PLR up to approximately 1000 lux, at which point the pupils are maximally constricted. These results provided preliminary evidence of the utility and potential benefit of mobile devices in providing accurate and easy estimation of the PLR.