Beyond the Light Peak: AI-Assisted Whole-Waveform Electrooculography as a Multi-Dimensional Functional Biomarker of Retinal Pigment Epithelium Health

This record documents the conceptual work presented at the Winterthur Ophthalmology Symposium 2026 (Winterthur, Switzerland). The presentation consisted of a conference poster and a short oral presentation, both based on the conceptual framework described in the manuscript archived in Zenodo (DOI: https://doi.org/10.5281/zenodo.18918403).The extended manuscript included in this record provides a written summary of the ideas presented at the symposium and consolidates the associated presentation materials. For documentation purposes, the original conference poster and the opening slide of the oral presentation are incorporated in the introductory section of the manuscript.The central concept explored in this work proposes a shift in electrooculography (EOG) interpretation from reliance on the traditional Arden ratio toward analysis of the entire electrooculographic waveform. In current clinical electrophysiology practice, EOG responses are typically summarized by the light peak–to–dark trough ratio, which compresses a complex electrophysiological signal into a single scalar metric. However, the EOG waveform itself reflects temporally structured physiological processes arising from ion transport mechanisms within the retinal pigment epithelium (RPE) and its interaction with photoreceptors.The novelty of the proposed framework lies in exploring artificial intelligence–assisted extraction of physiological information from the full EOG waveform and proposing it as a multidimensional functional biomarker of Retinal Pigment Epithelial health. Rather than reducing the signal to a single parameter, the approach considers the waveform as a multidimensional physiological time series, from which features such as temporal dynamics, waveform morphology, oscillatory components, and signal variability may potentially be extracted computationally.These features may be integrated to generate a functional biomarker profile of RPE physiology, potentially providing richer information about RPE functional states than the traditional Arden ratio alone. To our knowledge, systematic AI-assisted analysis of the entire electrooculographic (EOG) waveform to derive multidimensional functional descriptors of RPE health has not been established in the ophthalmic electrophysiology literature.This record therefore serves as a conference-linked conceptual report, preserving the scientific content of the symposium presentation and outlining a theoretical framework intended to stimulate future research into waveform-level electrophysiological biomarkers in retinal disease, therapeutic monitoring, and translational ophthalmology.Files includedExtended conceptual manuscriptConference poster presented at the Winterthur Ophthalmology Symposium 2026Oral presentation materialConference participation documentationReferenceLindamulage, S. D. (2026). Artificial Intelligence Assisted Whole Waveform Electrooculography as a Functional Biomarker Profile of Retinal Pigment Epithelium Health (2.0). Beyond the Light Peak: AI-Assisted Whole-Waveform Electrooculography as a Multi-Dimensional Functional Biomarker of Retinal Pigment Epithelium Health Beyond the Light Peak: AI-Assisted Whole-Waveform Electrooculography as a Multi-Dimensional Functional Biomarker of Retinal Pigment Epithelium Health, 3rd Winterthur Ophthalmology Conference - Switzerland. Zenodo. https://doi.org/10.5281/zenodo.19201818