Reliability assessment of iPhone-based pupillometry: A test-retest analysis

INTRODUCTION

Pupillometry serves as a valuable tool in various fields such as neuroscience, psychology, and ophthalmology.¹ It provides insights into autonomic nervous system activity, cognitive processing, and visual function.¹ Traditional pupillometry methods often involve specialized equipment and controlled settings, limiting their widespread application outside of research environments. However, recent advancements in technology, particularly the integration of pupillometry into smartphones, have shown promise in enhancing accessibility and convenience for both researchers and clinicians.²

The test-retest reliability of pupillometry measures is a crucial aspect to consider when evaluating the effectiveness and validity of smartphone-based pupillometers. Test-retest reliability refers to the consistency of measurements obtained from the same individual under identical conditions across multiple testing sessions over time. A high level of reliability indicates that the measurements are stable and reproducible, which is essential for drawing valid conclusions from pupillary data.

Studies investigating the test-retest reliability of smartphone-based pupillometers, particularly those integrated with iPhones, are emerging in the literature.^3,4 These studies typically employ various experimental designs, protocols, and statistical analyses to assess the consistency of pupillary measurements over time. For instance, a study by Gramkow et al.⁴ assessed the test-retest reliability and short-term variability of quantitative light reflex pupillometry in a mixed memory clinic cohort. Results showed high reliability across multiple sessions and minimal short-term variability in pupillary parameters, supporting the potential use of pupillometry in cognitive assessment within memory clinic settings.⁴ Similarly, in a study by Herbst et al., a good repeatability of a chromatic pupillometer was found using two selected wavelengths (470 and 660 nm) for maximal pupillary contraction amplitude and sustained contraction amplitude.⁵

In this study, we aim to assess the test-retest reliability of an iPhone-based pupillometer. We hypothesize that the iPhone-based pupillometer will demonstrate high test-retest reliability. This evaluation will provide essential insights into the potential of smartphone-based tools in routine clinical assessments.

MATERIAL AND METHODS

This study employed a test-retest design to evaluate the reliability of an iPhone-based pupillometer. Participants were tested twice with the same device under consistent conditions to assess the repeatability of the measurements. A total of 40 participants were recruited for the study. Participants included subjects aged between 18 and 35 years who had no significant ocular or neurological disorders affecting pupillary function and who had normal visual acuity (corrected or uncorrected). Exclusion criteria encompassed individuals with a history of eye surgery within the past year, known pupillary abnormalities such as anisocoria exceeding 0.5 mm, significant neurological conditions like epilepsy or Parkinson’s disease, current use of medications influencing pupillary dynamics (e.g., opioids and anticholinergics), and inability to provide informed consent or cooperate due to cognitive impairment or language barriers. The study was conducted in accordance with the principles of the Declaration of Helsinki. The Institutional Ethics Committee reviewed the protocol and determined that formal ethical approval was not required, as the study involved minimal risk to participants. Protecting participants’ rights, privacy, and confidentiality was paramount throughout all stages of the research process. Informed consent was obtained from each participant, outlining the study’s purpose, procedures, potential risks, and benefits.

RESULTS

A total of 40 subjects (80 eyes) were included, of whom 22 were males. The mean ± SD age of the subjects was 28 ± 5 years. No significant differences in the pupillary parameters were noted between genders (p > 0.78). All participants had normal or corrected-to-normal vision and no history of ocular or neurological disorders. Table 1 provides the descriptive statistics for the various pupillary parameters measured in the study. These include the mean, standard deviation, minimum, and maximum values for each parameter.

Bland-Altman analysis

Bland-Altman plots were used to assess the agreement between the two test sessions. The limits of agreement for all the pupillary parameters are as follows: ACS: 0.08, CT: 0.09, MCS: 0.03, Amplitude: 0.02, Average diameter: 0.02, Latency: 0.01, Maximum diameter: 0.04, Minimum diameter: 0.03. The plots indicated that most differences fell within the acceptable range, demonstrating good agreement between the test and retest measurements [Figure 1].

Close

Figure 1:

Bland–Altman analysis showing agreement between test and retest measurements of pupillary parameters obtained using the iPhone-based Reflex Pro PLR analyzer application. The solid line in each plot represents the mean difference (bias) between the two sessions, while the dashed lines indicate the 95% limits of agreement. The x-axis represents the mean of test and retest measurements, and the y-axis represents the difference between test and retest values for each parameter. (a) Average constriction speed (x-axis: mean ACS [mm/s]; y-axis: difference in ACS [mm/s]), (b) Constriction amplitude (x-axis: mean amplitude [mm]; y-axis: difference in amplitude [mm]), (c) Average diameter (x-axis: mean diameter [mm]; y-axis: difference in diameter [mm]), (d) Constriction time (x-axis: mean CT [s]; y-axis: difference in CT [s]), (e) Latency (x-axis: mean latency [s]; y-axis: difference in latency [s]), (f) Maximum diameter (x-axis: mean maximum diameter [mm]; y-axis: difference in maximum diameter [mm]), (g) Maximum constriction speed (x-axis: mean MCS [mm/s]; y-axis: difference in MCS [mm/s]), (h) Minimum diameter (x-axis: mean minimum diameter [mm]; y-axis: difference in minimum diameter [m]). PLR: Pupillary light reflex, ACS: Average constriction speed, CT: Constriction time, MCS: Maximum constriction speed.

Export to PPT

DISCUSSION

The findings from our study demonstrate that the iPhone-based pupillometer exhibits high test-retest reliability, making it a promising tool for clinical and research applications in pupillometry. The ICC values obtained from our repeated measures are consistent with those reported in similar studies, reinforcing the robustness of smartphone-based pupillometry. This aligns closely with the findings of Herbst et al., who also reported high repeatability of the pupil light response to both blue and red light stimuli using a novel pupillometer.⁵ Herbst et al.⁵ highlighted the device’s capacity for consistent measurements across different light conditions, underscoring its utility for clinical applications where accurate pupillary assessments are crucial.

Our results showed an ICC greater than 0.90 for most of the pupillary parameters, indicating excellent reliability, which aligns with findings from previous research. For instance, a study by Mariakakis et al. using the PupilScreen smartphone application reported high within-group reliability (k = 0.85) and interrater reliability (k = 0.75), showcasing the potential of smartphone applications in accurately assessing the PLR.⁶ Another study evaluating a custom-designed infrared smartphone pupillography system reported ICC values ranging from 0.982 to 0.995, further validating the high reliability of smartphone-based measurements.⁷ Likewise, Sousa et al. reported ICC values ranging from 0.982 to 0.995 for their infrared smartphone pupillometer, further validating the reliability of smartphone-based devices.⁸

The high reliability observed in these studies, including ours, suggests that smartphone-based pupillometers can serve as reliable alternatives to traditional and digital pupillometry methods. Traditional methods, such as penlight examinations, are inherently subjective and can lead to variability in measurements. Digital pupillometers, while accurate, are often expensive and require specialized training. In contrast, smartphone-based pupillometers are cost-effective, easily accessible, and user-friendly, making them ideal for a wide range of applications, from clinical practice to telemedicine and remote monitoring.

Moreover, the use of infrared technology in some smartphone pupillometers enhances measurement accuracy under various lighting conditions, as shown in studies where infrared pupillography yielded high-quality videos and photographs that closely matched those obtained with traditional desktop systems.⁷ This is particularly important for ensuring reliable measurements in diverse clinical scenarios.

The implications of these findings are significant for clinical practice. The portability and ease of use of the iPhone-based pupillometer can enhance patient care in diverse settings, including emergency departments and resource-limited environments. The high reliability of these devices ensures that clinicians can confidently use them for neurological and ophthalmological assessments, potentially improving diagnostic accuracy and patient outcomes.

Potential limitations of the study include the sample size, the fixed environmental conditions, which might not reflect real-world variability, and the reliance on a single iPhone model, which may limit generalizability to other devices. Future research should focus on further validating the iPhone-based pupillometer in various clinical populations and settings. Longitudinal studies comparing its performance with standard digital pupillometers will help establish its credibility and facilitate its broader adoption in clinical practice. Additionally, exploring its application under different lighting conditions and across various patient demographics will provide deeper insights into its practical utility and limitations.

CONCLUSION

In conclusion, our study corroborates the growing body of evidence supporting the use of smartphone-based pupillometers as reliable tools for measuring pupillary responses. The high test-retest reliability observed in our findings underscores the potential of these devices to enhance clinical practice and research in neurology, ophthalmology, and beyond. Future research should continue to explore the application of smartphone pupillometry in different patient populations and settings to further validate its utility and expand its adoption in clinical practice.