Measuring Relative Afferent Pupillary Defect (RAPD)

A Mechanism for Measurement of Relative Afferent Pupillary Defect (RAPD) Using Machine Learning

The goal of this prospective reliability study is to test the effectiveness of a commercially available, off-the-shelf virtual reality head-mounted display (VR HMD) and machine learning (ML) algorithms in detecting Relative Afferent Pupillary Defect (RAPD) in a group of subjects with known RAPD and another group with no known RAPD.

The main questions it aims to answer are:

• Does the use of the VR HMD and ML to replace the standard of care swinging flashlight test provide a more reliable and objective pupil measurement to detect RAPD?
• Can RAPD be detected by the VR HMD and ML algorithms at an earlier stage than the standard of care swinging light test?

Participants will be asked to undergo the standard of care swinging flashlight test, have their pupils manually measured, then have the test repeated using the VR HMD and ML.

Researchers will compare the measurements taken manually, following the standard of care swinging light test and those recorded by the VR HMD and ML to help answer the above questions.

Study Overview

• Status: Terminated
• Conditions: Relative Afferent Pupil Defect (RAPD)
• Intervention / Treatment: Diagnostic test: Swinging Light test; Diagnostic test: Virtual Reality Head-Mounted Display (VR HMD)

Detailed Description

Purpose of the Study/Objectives

The objective of this reliability study is to successfully develop a mechanism for diagnostics of Relative Afferent Pupillary Defect (RAPD) using pupil measurement values and pathologic response. The diagnostics of this test will later be compared with the traditional swinging flashlight test results. The physiological basis of the study is that the reaction of the pupils in the right and left eyes are linked. When a bright light is shone into one eye, it will lead to an equal constriction of both pupils. When the light source is removed from the eye, the pupils will enlarge equally.

Our hypothesis is difference between pupil measurement values of two eyes when the light source moves from one eye to the other is an indicator of RAPD. This reliability study is to use a commercially-available, off-the-shelf virtual reality (VR) head-mounted display (HMD) and machine learning (ML) algorithms to detect RAPD at an earlier stage with small differences in pupil measurement values. The RAPD is a critically important sign in patients with unexplained visual loss because it is an objective finding of afferent pupillary dysfunction. The research question is if the proposed use of VR and ML to replace the swinging flashlight test provides a more reliable and objective pupil measurement to detect RAPD.

Background

A relative afferent pupillary defect (RAPD) is a critically important ophthalmologic examination finding that defines a defect (lesion) in the pupil pathway on the afferent side. A RAPD is relative to the fellow eye and occurs because of the bilateral and equal innervation of the pupils in normal individuals. The RAPD manifests as a difference in pupillary light reaction between the two eyes. The test requires two eyes but only one working pupil.

In a normal patient without afferent pupillary disease, shining a light in either eye will produce constriction of both eyes equally. In the case of an afferent pupillary pathway lesion, the light response is tested individually in each eye and then the light swings between the two eyes to detect RAPD. The normal pupillary response is characterized by equal constriction of the pupils in both eyes when the light stimulus is applied to each eye individually and no dilation or pupillary escape when the light swings from the one eye to the fellow eye. The pathologic response that characterizes the RAPD includes the following: (1) the light reaction causes pupil constriction in both eyes when the light shines in the normal eye, and (2) dilatation of the pupils in both eyes when the light stimulus is rapidly transferred from the normal eye to the pathologic eye.

Study Design

The study design is a reliability study evaluating the performance of the HTC Vive Pro Eye, to detect RAPD. The HTC Vive Pro Eye (product name) is not an experimental device. It is a consumer-grade VR HMD available as an off-the-shelf VR Headset. To validate the ability of the VR HMD to detect RAPD in patients, a reliability study will be performed by using the device on patients with known RAPD and patients without RAPD. Participants will undergo manual pupil measurement in the eye clinic via penlight for direct and swinging light test reflexes. Patients then will undergo the VR HMD eye measurements as discussed previously.

The anticipated total number of participants for this study is 120 patients aged 18 to 85, consisting of 60 participants in the control group (no known RAPD) and 60 participants in the experimental group (known RAPD). Both groups will undergo the traditional swinging flashlight test and testing with the VR HMD.

Medical Records will be used to obtain ocular imaging, pupil measurements, past medical history and past ocular history. Demographic information will be obtained through a self-reporting survey for age, gender and ethnicity. Data will be acquired by using the VR HMD. The protocol presents light through the VR HMD displays to each eye individually and will use participants' pupil diameters via its integrated eye tracking cameras to collect information about pupillary reflex deficit. Data from visual function tests such as visual acuity, visual field, chromatic sensitivity, etiology of RAPD etc. will be taken from the EPIC (product name) electronic medical record (EMR) to compare with current study findings.

This research will include records of participants' voices or images. Video of the pupils of the participant and their gaze location may be recorded via VR device for research purpose. Audio, video or photographs may be recorded for publication purposes. Subject identity will be removed from the audio, video and photographs. Informed Consent Form will ask the subject for their permission to be photographed or audio/video recorded.

Principal Investigator (PI) and researchers will ensure that all study personnel are informed about the research plan and their research-related duties by routine meetings, regular communication via email and phone. All researchers will take the human subject protection training sanctioned by the university via Collaborative Institutional Training Initiative (CITI) Program training.

A counterbalanced repeated measures experimental design will be used to compare participant performance in two conditions: using traditional swinging flashlight test versus using pupil measurement head-mounted display. The independent variable in the study will be modes (2 conditions) and the measured dependent variables (1 condition) will be RAPD.

Study Intervention

The study intervention that is being evaluated is whether virtual reality testing through a VR HMD device can adequately diagnose RAPD.

Each participant will undergo the standard of care swinging flashlight test to diagnose RAPD by their treating ophthalmologist, an IRB approved study team member, at their regular clinic visit, prior to participation in the study. If the patient elects to participate in the study, participates in the consent process and signs the Informed Consent Form, he or she will undergo a repeat of the standard of care swinging flashlight test, then participant will undergo testing with the VR HMD. Data on each participant's pupil sizes in both light and dark simulations is measured and coded. The VR HMD pupil measurement device is a consumer-grade virtual reality headset available as an off-the-shelf gaming product. The risk of using the device is minimal and no greater than the standard of care swinging flashlight test. The device uses illumination of less than 90 cd/m2 which is below and in accordance with the typical flashlight exposure of 100 cd/m2. There is therefore no disproportionate increase in risk. The device's maximum illumination is well below the threshold which would be of potential risk of harm. Risk of loss of private health information (PHI) is minimized by the manner with which the data will be stored.

At the acquisition time, subjects will be assigned a study identification number. That number will be recorded on a subject identity master key, along with the subject's medical record number (MRN). Once the subject identity master key has been created and stored separately from study data in a password protected file on the University of Texas Medical Branch's (UTMB's) secure server behind a firewall. All identifying information will then be removed from the subject's data. The deidentified data will then be stored in a secure folder on the UTMB secure server behind a firewall and that of the collaborating site (University of Nevada, Reno) computer behind a firewall.

The total testing should take approximately 10-30 minutes. After each participant has their standard of care ophthalmology visit with their treating ophthalmologist, an Institutional Review Board- (IRB-) approved study team member, consent will be obtained by an IRB-approved study team member, who has been properly trained and designated by the PI to obtain consent. If the patient consents, they will undergo a repeat of the swinging light test and their pupil measurements recorded manually. Then the subject will undergo testing with the VR HMD during which the device will take pupillary measurements lasting approximately five minutes. No follow-up appointments for research purposes are necessary.

Subjects will be recruited from the UTMB eye clinic. Subjects' participation is limited to one visit. Study enrollment is anticipated to take six months. The duration of the study is estimated to be two years.

• Study Type: Interventional
• Enrollment (Actual): 71
• Phase: Not Applicable

Contacts and Locations

Study Locations

• United States
  • Texas
    • Galveston, Texas, United States, 77555
      • University of Texas Medical Branch

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years to 85 years (Adult, Older Adult)
• Accepts Healthy Volunteers: Yes

Description

Inclusion Criteria:

• Age 18 - 85
• 60 patients with known RAPD.
• 60 patients with no known RAPD
• May be diagnosed with a vision disorder such as glaucoma, diabetic retinopathy, unilateral or asymmetrical disease of the retina or other degenerative ocular diseases
• Referred to study by treating eye doctor
• Volunteered to participate in study
• UTMB clinic patients
• Not belonging to an identified vulnerable population

Exclusion Criteria:

• Younger than 18; older than 85 years of age
• Not a UTMB clinic patient
• Belonging to an identified vulnerable population
• Participants with discomfort using glasses
• Experience discomfort using VR HMD device
• Participants with history of congenital pupil defects, traumatic pupils
• Those whose pupils were dilated during the standard of care examination deemed by treating eye doctor not to be eligible to participate in study
• Participants who do not follow study directions and/or safety procedures given by investigators or lab assistants
• Participants who do not speak English to prevent a communication error during the informed consent process

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Screening
• Allocation: Non-Randomized
• Interventional Model: Parallel Assignment
• Masking: None (Open Label)

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: Known RAPD; Participants with known RAPD will be administered the standard of care swinging light test and pupil measurements taken manually, then administered the light test using the VR HMD and ML.	Diagnostic test: Swinging Light test; The swinging light or pen light test with manual pupil measurements is the standard of care test for RAPD. This test consists of a light being shone in one eye for 3 to 4 seconds then shown in the other eye for 3 to 4 seconds, repeating 2 to 3 times. Pupils are measured manually immediately after the light test. Measurements are recorded in patient's medical record.; Other Names: Pen Light test; Diagnostic test: Virtual Reality Head-Mounted Display (VR HMD); A light is shone from within the VR HMD, pupillary measurements are taken and recorded by the machine learning (ML) algorithm.
Experimental: No Known RAPD; Participants with no known RAPD will be administered the standard of care swinging light test and pupil measurements taken manually, then administered the light test using the VR HMD and ML.	Diagnostic test: Swinging Light test; The swinging light or pen light test with manual pupil measurements is the standard of care test for RAPD. This test consists of a light being shone in one eye for 3 to 4 seconds then shown in the other eye for 3 to 4 seconds, repeating 2 to 3 times. Pupils are measured manually immediately after the light test. Measurements are recorded in patient's medical record.; Other Names: Pen Light test; Diagnostic test: Virtual Reality Head-Mounted Display (VR HMD); A light is shone from within the VR HMD, pupillary measurements are taken and recorded by the machine learning (ML) algorithm.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Pupillary measurements	Measure of pupil size in each eye, measured manually or via machine learning algorithms	Immediately following the swinging light test or VR HMD light test

Collaborators and Investigators

• Sponsor: The University of Texas Medical Branch, Galveston
• Collaborators: University of Nevada, Reno
• Investigators: Principal Investigator: Kevin H Merkley, MD, MBA, University of Texas

Publications and helpful links

General Publications

• Vejdani MD., Al-Zubidi N. Relative Afferent Pupillary Defect. Eyewiki 2020. https://eyewiki.org/Relative_Afferent_Pupillary_Defect
• Simakurthy S, Tripathy K. Marcus Gunn Pupil. 2023 Aug 25. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2026 Jan-. Available from http://www.ncbi.nlm.nih.gov/books/NBK557675/

Study record dates

Study Major Dates

• Study Start (Actual): May 4, 2023
• Primary Completion (Actual): March 6, 2026
• Study Completion (Actual): March 6, 2026

Study Registration Dates

• First Submitted: March 22, 2023
• First Submitted That Met QC Criteria: March 22, 2023
• First Posted (Actual): April 5, 2023

Study Record Updates

• Last Update Posted (Actual): June 3, 2026
• Last Update Submitted That Met QC Criteria: May 29, 2026
• Last Verified: May 1, 2026

More Information

Terms related to this study

• Keywords: Vision loss; Visual loss; Relative Afferent Pupil Defect (RAPD); Pupil measurements; Swinging flashlight test; Pen light test; Virtual reality head mounted display (VR HMD); Afferent pupillary dysfunction; Pupillary light reaction; HTC Vive Pro Eye
• Additional Relevant MeSH Terms: Neurologic Manifestations; Nervous System Diseases; Eye Diseases; Sensation Disorders; Pathological Conditions, Signs and Symptoms; Signs and Symptoms; Vision Disorders; Pupil Disorders
• Other Study ID Numbers: 23-0023

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No