Spectral and melanopic dose calibration of consumer see-through extended-reality glasses for controlled retinal photostimulation
- 1Washington University
Abstract
Consumer extended-reality (XR) glasses could provide controlled ocular light exposure, but their settings do not specify the light delivered at the eye. We developed and validated a calibration pipeline linking display control, corneal-plane spectroradiometry, settings-to-exposure modeling, and rendered application content for a see-through binocular micro-OLED display (VITURE Luma Ultra). Automated measurements characterized color channels, digital codes, hardware brightness levels, and duty-cycle settings across repeated fittings on a phantom head. The blue primary peaked at 459 nm with no resolved spectral shift across intensity settings. Blue melanopic equivalent daylight illuminance (EDI) ranged from 14.7 to 252 lx across hardware brightness levels, while full white delivered 269 lx melanopic EDI at its highest usable setting, exceeding the 250 lx daytime recommendation for healthy adults. A model calibrated on separate sessions predicted melanopic EDI for held-out codes, brightness levels, and color mixtures with a mean absolute percentage error of 3.4% (4.5% for photopic illuminance), allowing exposure to be estimated from logged display settings without remeasuring each uniform stimulus. Rendered application frames revealed position- and illuminated-area-dependent exposure; a spatial correction derived from an independent display grid accounted for these effects and closely reproduced measured frame exposure. Together, these results establish a reusable workflow for converting see-through XR hardware into a quantitatively specified ocular-light platform.
Related
- Cross-validation of wearable light loggers across electric and natural light conditions: focus on illuminance, melanopic EDI, and CCT
- Constructing melanopic light exposure profiles from sparse daily-route measurements.
- Photographic Atlas for Cosmetic Noodling Characterization: Development, Analytical Validation, and Consumer Perception
- FDA Accelerated Approval Disclosures on Direct-to-Consumer Prescription Drug Websites: Testing Presence, Prominence, and Wording
- A modified Delphi study to determine Australian healthcare consumer priorities and consensus for the use of genomics and precision medicine approaches in pregnancy healthcare and research
BibTeX
@misc{gaidica2026spectral,
title = {{Spectral and melanopic dose calibration of consumer see-through extended-reality glasses for controlled retinal photostimulation}},
author = {M. Gaidica and M. Rosengart},
howpublished = {medRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.26.26361398},
url = {https://www.medrxiv.org/content/10.64898/2026.08.26.26361398v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Spectral and melanopic dose calibration of consumer see-through extended-reality glasses for controlled retinal photostimulation
AU - M. Gaidica
AU - M. Rosengart
PY - 2026
DA - 2026/08/26
PB - medRxiv
DO - 10.64898/2026.08.26.26361398
UR - https://www.medrxiv.org/content/10.64898/2026.08.26.26361398v1
AB - Consumer extended-reality (XR) glasses could provide controlled ocular light exposure, but their settings do not specify the light delivered at the eye. We developed and validated a calibration pipeline linking display control, corneal-plane spectroradiometry, settings-to-exposure modeling, and rendered application content for a see-through binocular micro-OLED display (VITURE Luma Ultra). Automated measurements characterized color channels, digital codes, hardware brightness levels, and duty-cycle settings across repeated fittings on a phantom head. The blue primary peaked at 459 nm with no resolved spectral shift across intensity settings. Blue melanopic equivalent daylight illuminance (EDI) ranged from 14.7 to 252 lx across hardware brightness levels, while full white delivered 269 lx melanopic EDI at its highest usable setting, exceeding the 250 lx daytime recommendation for healthy adults. A model calibrated on separate sessions predicted melanopic EDI for held-out codes, brightness levels, and color mixtures with a mean absolute percentage error of 3.4% (4.5% for photopic illuminance), allowing exposure to be estimated from logged display settings without remeasuring each uniform stimulus. Rendered application frames revealed position- and illuminated-area-dependent exposure; a spatial correction derived from an independent display grid accounted for these effects and closely reproduced measured frame exposure. Together, these results establish a reusable workflow for converting see-through XR hardware into a quantitatively specified ocular-light platform.
ER -