PerEye: Co-Designing Extended Reality Rendering Attributes for Vision Health Diagnosis and Education

Paper Title

PerEye: Co-Designing Extended Reality Rendering Attributes for Vision Health Diagnosis and Education

Publication Info

• Topic area: Extended Reality (XR) applications in vision health diagnosis, rehabilitation, and education.
• Keywords: Vision health, XR, co-design, immersive simulation, visual field testing, empathy, education, monocular rendering, gaze-contingent effects, telehealth.

Background and Problem

• Problem / challenge: Current XR systems rely on symmetrical binocular rendering, which fails to address asymmetric and dynamic visual impairments common in clinical practice. XR tools lack integration into clinical workflows and educational contexts, limiting their utility in vision health.
• Significance: Vision impairments affect over 2.2 billion people globally, with disparities in access to diagnosis and education. XR offers potential for portable and accessible solutions, addressing gaps in clinical and educational contexts.
• Motivation and related work: Prior XR applications in vision health include diagnostic tools like Olleyes VisuALL VR and RadiusXR, but these systems face limitations such as restricted rendering flexibility and lack of clinician involvement in design. Co-design approaches have shown promise in health HCI but remain underexplored in XR for vision care.

Solution

• Proposed approach: PerEye, a co-designed XR toolkit enabling gaze-contingent control of visual attributes for each eye, tailored for diagnostic, educational, and rehabilitative applications.
• Novelty:
  1. A sustained co-design trajectory involving clinicians to shape XR rendering attributes for vision health.
  2. Development of monocular rendering functions for asymmetric and dynamic vision simulation, addressing clinical gaps.
  3. Empirical validation through applied studies in visual field testing and orthoptic education.
• Procedure and key techniques:
  • Co-design with clinicians to define rendering attributes (e.g., occlusion, blur, contrast, scale, positional/rotational offsets).
  • Development of a VR-based visual field prototype and comparison with the Humphrey Field Analyzer.
  • Application of gaze-contingent simulations in orthoptic training to enhance empathy and understanding of vision loss.

Results

• Concrete findings:
  • Study 1: Clinicians identified rendering attributes critical for simulating and correcting asymmetric vision impairments.
  • Study 2: VR-based visual field testing reproduced spatial sensitivity patterns but showed limited agreement with gold-standard perimetry due to luminance constraints.
  • Study 3: Immersive scotoma simulation significantly increased empathy scores in orthoptic students (VSEQ: pre-simulation mean = 31.92, post-simulation mean = 35.46, p < 0.001).
• Advantage over baselines:
  • PerEye enables fine-grained monocular rendering, addressing asymmetric vision conditions that conventional XR systems overlook.
  • Combines diagnostic and educational applications within a single XR framework.
• Experiments / evaluation:
  • Study 1: Participatory design with 11 clinicians to parameterise rendering attributes.
  • Study 2: Comparison of VR-based visual field testing (N=32) with Humphrey Field Analyzer.
  • Study 3: Empathy-focused training with 13 orthoptic students using gaze-contingent scotoma simulation.
• Limitations and future work:
  • Limited luminance range in VR displays restricts clinical equivalence for visual field testing.
  • Challenges in standardising rendering across headsets due to hardware variability.
  • Future work includes broader patient validation, telehealth applications, and improving display luminance and calibration.

Summary

PerEye introduces a co-designed XR toolkit to address gaps in vision health diagnosis and education, enabling asymmetric and dynamic rendering for each eye. Empirical studies demonstrated its feasibility in visual field testing and its educational impact in fostering empathy among orthoptic students. While limitations in luminance and hardware standardisation constrain clinical equivalence, the findings highlight XR's potential as a multi-purpose platform for diagnosis, consultation, and training. Future research will focus on telehealth deployment, broader validation, and improving technical precision for clinical adoption.