What is the purpose of the L-OAT 16-zone grid?

It is an exploratory spatial sampling scheme intended to preserve regional iris heterogeneity. It is not a validated mandatory grid and may be simplified if reproducibility is better with fewer sectors.

MyLumineyes™ Research · Quantitative Iris Phenotyping · Multimodal Imaging

L-OAT Research Hub: Quantitative Iris Phenotyping Beyond Color Labels

Q: Is L-OAT a validated iris classification system?

No. L-OAT is currently a structured multimodal feature framework. Stable phenotype classes, if any, would need to emerge empirically and be externally replicated.

Q: Why is the Research Archive page different from the Zenodo and journal versions?

The academic manuscripts preserve the formal argument. The Research Archive page is an original synthesis providing version mapping, evidence boundaries, a dedicated web infographic, cross-framework context, and a research roadmap without reproducing either manuscript in full.

An original Research Archive synthesis of the Lumineyes Optical Architectural Typology (L-OAT), its publication versions, measurement architecture, technical limits, and future validation pathway

Author: Mustafa Mete, MD Framework: Lumineyes Optical Architectural Typology (L-OAT) Research status: Conceptual / hypothesis-generating Updated: October 2026

Journal version

Perspective Manuscript

Title: Lumineyes Optical Architectural Typology (L-OAT): a multimodal hypothesis for quantitative iris phenotyping

Article type: Perspective

Version role: Journal-form manuscript with expanded technical qualification, acquisition logic, OCT signal constraints, validation pathway, declarations, and figure legends.

Status note: This Research Archive page does not infer a publication DOI or final journal status beyond the manuscript metadata available to the site editor.

Zenodo preprint

Preprint v1.0

Date: 30 September 2026

DOI: 10.5281/zenodo.23066772

Version role: Citable public record of the L-OAT conceptual framework and its proposed visible, architectural, and OCT-derived measurement domains.

Evidence status: Preprint / conceptual framework; no clinical cohort or predictive performance is reported.

Research Archive synthesis

This Web Page

Purpose: A site-native scholarly overview that explains the framework without reproducing either academic manuscript as full text.

Duplicate-control strategy: New structure, new language, new tables, and a dedicated web infographic; the academic versions remain the primary manuscript records.

Reader role: Navigation, interpretation, publication status, evidence boundaries, and links to related MyLumineyes™ research.

Version policy. The journal manuscript, Zenodo preprint, and this Research Archive page are intentionally not identical documents. The manuscripts preserve the formal academic argument; this page is an original explanatory and bibliographic layer. It summarizes the framework, highlights its measurement logic and limitations, and links the academic record without mirroring the manuscripts section-for-section.

Quick Answer

Lumineyes Optical Architectural Typology (L-OAT) is a proposed multimodal measurement framework for describing the living iris more quantitatively than conventional labels such as brown, hazel, green, grey, or blue. It combines three candidate domains: P, visible appearance and spatial distribution; A, image-derived iris geometry and architecture; and O, OCT-derived signal features.

L-OAT is deliberately presented as a measurement language and hypothesis-generating research framework, not as a validated prognostic instrument, a diagnostic test, or a method for prescribing a future eye color. Its immediate scientific task is to define reproducible features, acquisition conditions, quality-control rules, and validation questions before any stable phenotype classes or predictive claims are accepted.

1. Why L-OAT Was Proposed

Conventional iris color categories are useful in ordinary clinical communication, but they compress a continuous and spatially heterogeneous phenotype into a small number of names. A single iris may contain a darker peripupillary ring, lighter mid-stromal zones, sectoral variation, crypts, radial texture, and illumination-dependent shifts that are not captured by one categorical label.

Quantitative photography has already shown that continuous and spatial measures preserve information that can be lost when iris color is reduced to a single word.^[1,2,9] Separate work has measured three-dimensional iris morphology with anterior-segment OCT and explored melanin-related optical signals using polarization-sensitive AS-OCT.^[10,11] L-OAT does not claim to have invented those measurement domains. Its proposed contribution is to place visible phenotype, architecture, and OCT-derived signal inside one explicitly qualified feature framework, together with rules for uncertainty, missing data, repeatability, and future validation.

Core research definition

L-OAT is an operational description of a baseline iris phenotype in which calibrated visible appearance, image-derived architecture, and qualified OCT signal features are represented as a multimodal feature vector rather than collapsed into a single subjective color label.

L-OAT(T₀) = [ P(T₀), A(T₀), O(T₀) ]

The three domains are not assumed to be independent, equally weighted, or reducible to one score. L-OAT also does not assume in advance that stable discrete iris “types” exist. If future clustering is weak, continuous axes may be scientifically more appropriate than forcing the phenotype into predefined classes.

2. The Three Candidate Domains

Visible Appearance

Captured phenotype under controlled visible-light imaging.

CIE L*, a*, b* coordinates
Colorimetric difference measures
Global and sectoral summaries
Spatial heterogeneity
Documented illumination and white balance
Visible color labels retained only as secondary descriptors

Architecture

Image-derived geometry and structural descriptors.

Iris thickness at reproducible landmarks
Curvature or cross-sectional area where supported
Sectoral geometric heterogeneity
Pupil diameter and acquisition state
Coordinate / registration scheme
3-D volume only when the acquisition genuinely supports volume

OCT Signal

Qualified signal features at the OCT instrument wavelength.

Repeatable stromal signal summaries
Depth distribution
Candidate anterior-to-posterior ratios
Sectoral signal heterogeneity
Apparent attenuation only under defensible calibration
Signal floor, processing, focus and device metadata recorded

3. A Distinct Research-Archive Infographic

L-OAT Web Research Map — from multimodal capture to testable phenotype hypotheses

P · Visible phenotype Calibrated photography, CIELAB coordinates, sectoral distribution, spatial heterogeneity, and repeat imaging under documented capture conditions.

A · Architecture AS-OCT geometry, iris thickness, curvature, pupil state, spatial landmarks, and reproducible registration constraints.

O · OCT signal Device-qualified stromal signal features, depth behavior, ratios or apparent attenuation only where export quality and calibration permit defensible interpretation.

↓

Multimodal Quality Gate · repeatability · signal floor · segmentation · registration · missingness · export feasibility

↓

Baseline Feature Vector · L-OAT(T₀) = [P(T₀), A(T₀), O(T₀)]

↓

Future empirical questions · continuous axes? stable clusters? longitudinal association? added information beyond baseline color?

Figure 1. Original Research Archive infographic. Unlike the manuscript figures, this web-specific map emphasizes the quality gate between data acquisition and phenotype interpretation. It contains no observed patient data, validated class boundaries, or predictive result.

This web infographic is intentionally different from the academic manuscript’s figure architecture. The manuscript figure presents the formal P–A–O feature structure and optional exploratory stratification; the Research Archive version emphasizes a point that is central to interpretation but easy to overlook: multimodal measurements should pass technical and quality-control gates before they are allowed to acquire biological meaning.

4. What L-OAT Measures—and What It Does Not

Table 1. Measurement boundaries inside L-OAT
Candidate measure	Permitted interpretation	Interpretation that should be avoided
CIELAB color coordinates	Captured visible appearance under a documented imaging protocol.	Direct measurement of eumelanin, pheomelanin, collagen composition, or histology.
Iris thickness / geometry	Image-derived anatomical geometry under defined acquisition conditions.	A guaranteed predictor of a particular visible hue or future color transition.
OCT display grey value	A device- and processing-dependent display signal.	Equivalent to calibrated linear OCT intensity or a tissue optical coefficient.
Within-image OCT signal ratio	A candidate normalized measurement when both regions are acquired comparably and remain above the signal floor.	Proof that the reference tissue is biologically invariant or that ratios are interchangeable across devices.
μ_app	An apparent OCT signal attenuation parameter only when the signal model, export type, noise, confocal effects and sensitivity roll-off are appropriately handled.	Direct visible-light scattering, direct melanin concentration, or histological tissue attenuation.
Sectoral heterogeneity	Spatial variation within valid measured regions.	A histological map when sector coverage, signal quality, or multimodal registration is incomplete.

OCT caution. Conventional intensity-only AS-OCT should not be equated with polarization-sensitive measurement of melanin. Instrument wavelength, gain, focus, scan protocol, averaging, post-processing, segmentation, signal floor, and export format can all affect measured signal. L-OAT therefore treats OCT intensity features as device-qualified imaging measurements, not automatic tissue-composition biomarkers.

5. Spatial Mapping and the Exploratory 16-Zone Concept

L-OAT proposes a coordinate system capable of preserving regional iris heterogeneity. One exploratory photographic scheme uses two radial bands crossed with eight angular sectors, creating a 16-zone map. This is not a validated requirement and should not be treated as fixed simply because it is visually convenient.

The important scientific point is spatial correspondence. A photographic sector is not automatically sampled by an OCT B-scan. OCT measurements should therefore be associated only with regions for which a defensible anatomical correspondence exists, with registration uncertainty and missing sectors reported rather than silently filled in.

Table 2. Spatial-design principles in L-OAT
Design issue	L-OAT approach	Reason
Photography grid	Two radial bands × eight angular sectors may be explored.	Preserves regional information that a single global color average can hide.
OCT sampling	Only acquired B-scan meridians should be treated as measured.	Prevents a photographic map from being falsely represented as complete OCT coverage.
Pupil state	Measure and document pupil diameter; aim for matched acquisition conditions in longitudinal work.	Iris geometry changes with pupil state.
Registration	Use anatomical landmarks and report uncertainty.	Iris deformation prevents exact pixel-to-pixel equivalence across states and modalities.
Missingness	Report non-measurable sectors explicitly.	Missing OCT signal may itself be related to phenotype or acquisition physics.

6. OCT Export Feasibility Comes Before Attenuation Modeling

One of the most consequential technical issues in the framework is whether the OCT system actually exports data suitable for quantitative signal analysis. A display-rendered image is not equivalent to linear OCT intensity, and a linear intensity signal is not equivalent to amplitude data. Consequently, an apparent attenuation model should only be explored when the export type and processing chain are sufficiently characterized.

I_meas(z) = B(z) + K · h(z;z_f) · T(z) · exp(−2μ_appz)

In this simplified candidate model, B(z) represents depth-dependent background, h the confocal response around focal depth, T sensitivity roll-off, and K a scale term that includes local backscatter and instrument effects. Without defensible correction or control of these components, the fitted slope combines tissue and instrument behavior.

Hard boundary: if the export pathway does not support calibrated or sufficiently characterized linear signal analysis, μ_app should not be reported as though it were a validated tissue coefficient. The scientifically correct outcome may be to omit the parameter rather than manufacture precision from display-processed data.

7. From Feature Space to Typology: No Classes Are Preassigned

The word Typology does not mean that L-OAT already contains validated Types I, II, III, or any other fixed class system. At the current stage, L-OAT is a structured feature space. A future feature dictionary should define each variable, units, image source, region of interest, quality rule, repeatability statistic, and missingness rule before clustering is attempted.

Exploratory principal-component analysis or clustering could later ask whether reproducible groups emerge. However, the number of groups should not be privileged in advance, and continuous axes should be retained if the data do not support stable clusters. Any classes would require internal stability assessment and external replication.

A visible pigmentation scale such as LIPS may be recorded as an external clinical descriptor, but it should not be silently incorporated into the OCT feature definition. This separation protects the distinction between a clinically useful visible grade and a multimodal quantitative phenotype.

Table 3. Current status of key L-OAT components
Component	Status	Interpretation
P / A / O domain structure	Proposed framework	Defines the candidate feature architecture.
Calibrated photography and colorimetry	Established methods	Existing methods are incorporated into the framework; L-OAT does not claim to have invented colorimetry.
AS-OCT geometry	Established modality	Existing anatomical measurements can supply candidate architecture features.
OCT signal ratios	Candidate measurement	Require technical reliability and construct validation.
μ_app	Conditional candidate	Requires appropriate export and calibration; not automatically available from a display image.
16-zone map	Exploratory design	May be simplified if reproducibility is better with fewer sectors.
Fixed phenotype classes	Not established	No Type I–III boundaries are prespecified.
Prediction of final eye color	Not established	L-OAT is not a color-prescription or prognostic instrument.

8. The Validation Sequence

L-OAT separates technical validation from biological or longitudinal interpretation. This is essential because a sophisticated statistical model cannot rescue an unstable measurement pipeline.

Stage 1 · Export feasibility Determine what the imaging system truly exports: raw / linear signal, calibrated intensity, or display-processed images.

Stage 2 · Technical repeatability Quantify repeated-acquisition error, segmentation agreement, signal floor, registration uncertainty, and missingness.

Stage 3 · Construct validation Test whether candidate ratios, geometric measures, and signal summaries behave consistently with the construct they are intended to represent.

Stage 4 · Longitudinal study Only after technical validation, assess whether baseline features add information about later phenotype change while accounting for intervention exposure.

A future longitudinal analysis could quantify visible change using repeated CIELAB measurements and CIEDE2000 ΔE₀₀, but ΔE₀₀ represents magnitude rather than direction. Directional components should therefore be examined separately. No universal threshold for a clinically meaningful iris ΔE₀₀ is assumed by L-OAT.

Any outcome-oriented analysis should also account for correlation between two eyes of the same participant and divide training and evaluation data at the participant level, not the eye level. Treatment exposure must be modeled because a response-guided staged intervention changes exposure over time.

9. Relationship to Lumineyes™, SSMM, LIPS, and L-SAFE

L-OAT originates within the MyLumineyes™ research program, but its definitions are intended to describe iris phenotype beyond any single intervention. Application to pigment-altering treatment is therefore a separate validation question rather than part of the definition of L-OAT itself.

Table 4. How L-OAT relates to other MyLumineyes™ research frameworks
Framework	Main question	Relationship to L-OAT
Selective Stromal Melanin Modulation (SSMM)	What pigment-bearing compartment is intended to be targeted?	SSMM is a treatment-target concept; L-OAT is a phenotype-measurement framework. One should not be used as proof of the other.
LIPS	How can visible iris pigmentation be described clinically?	LIPS may be recorded as an external visible descriptor but should remain distinct from OCT feature definitions.
L-SAFE	How should treatment-related safety state and kinetics be interpreted?	L-SAFE concerns response and readiness; L-OAT concerns phenotype and measurement. A future study could examine whether baseline phenotype features are associated with response, but this is not currently established.
Response-guided staged treatment	Should another exposure occur?	Longitudinal L-OAT analysis must account for changing treatment exposure because staging alters the intervention over time.

For the broader procedural context, see the canonical Laser Eye Color Change overview and the MyLumineyes™ Research Library.

10. What the Two Academic Versions Contribute

Table 5. Version map: journal Perspective manuscript vs Zenodo preprint vs Research Archive page
Version	Primary role	Content emphasis	Duplicate-control principle
Journal Perspective manuscript	Formal journal-form academic version	Full conceptual argument, acquisition requirements, quantitative framework, OCT signal qualifications, validation path, limitations, declarations, and formal references.	Remains the formal manuscript; this web page does not reproduce it section-for-section.
Zenodo preprint v1.0	Citable public record	Publicly accessible conceptual framework with DOI and version history.	Linked as the archival source rather than copied into the site.
Research Archive owner page	Site-native interpretation and navigation layer	Version cards, evidence status, distinct infographic, domain summaries, measurement boundaries, cross-links to related frameworks, and research roadmap.	Original structure and language; avoids full-text duplication while preserving scientific depth.

11. What L-OAT Does Not Claim

No validated prognostic classes: L-OAT does not currently define stable phenotype classes.
No direct melanin concentration from ordinary AS-OCT: conventional intensity images should not be treated as polarization-sensitive melanin measurements.
No guaranteed color destination: the framework does not predict that a particular iris will become blue, green, hazel, or another named color.
No automatic histology from imaging: visible photography and OCT signal cannot recover microscopic composition without independent evidence.
No cross-device equivalence by normalization alone: ratios do not eliminate differences in wavelength, focus, sensitivity roll-off, processing, or geometry.
No substitution for clinical monitoring: phenotype measurement does not replace ophthalmic assessment or response-guided safety decisions.

Scientific boundary: L-OAT is intentionally designed to make overinterpretation harder. A feature should remain an imaging feature until technical reliability and biological meaning have been independently demonstrated.

12. Research Significance

The immediate value of L-OAT is not that it claims a new iris classification has already been discovered. Its value is methodological: it proposes a common language for describing what is seen, what is geometrically measured, what the OCT instrument records, and how uncertainty should be carried forward.

If the framework survives technical validation, it could support research questions that conventional color labels cannot answer efficiently. These include whether baseline spatial phenotype is reproducible, whether architectural features add information beyond color alone, whether continuous feature axes are more stable than named categories, and whether any baseline feature is associated with later phenotypic change after a well-characterized intervention.

Those questions remain empirical. L-OAT is designed to make them testable rather than to answer them in advance.

Research FAQ

Is L-OAT a validated iris classification system?

No. At the present stage, L-OAT is a structured multimodal feature framework. Stable classes, if any, would need to emerge empirically and be externally replicated.

Does L-OAT measure melanin concentration with conventional AS-OCT?

No. Conventional intensity-only AS-OCT signal is device- and wavelength-dependent and should not be interpreted as a direct melanin concentration measurement. Polarization-sensitive methods represent a different measurement approach.

Why does L-OAT include colorimetry if it is meant to move beyond color labels?

Because calibrated continuous color measurements preserve information that categorical labels lose. L-OAT moves beyond simple labels, not beyond visible phenotype itself.

What is the purpose of the 16-zone grid?

It is an exploratory spatial sampling scheme intended to preserve regional heterogeneity. It is not a validated mandatory grid and may be simplified if reproducibility is better with fewer sectors.

Can L-OAT predict a patient's final eye color after pigment reduction?

No. Prediction of a named final color is not established. Any future prognostic application would require separate prospective validation and explicit uncertainty.

Why is the Research Archive page different from the Zenodo and journal versions?

To avoid unnecessary duplication and give the website a distinct scholarly role. The academic manuscripts preserve the formal argument; this page provides version mapping, evidence boundaries, a dedicated web infographic, cross-framework context, and a concise research roadmap.

References

Edwards M, Gozdzik A, Ross K, Miles J, Parra EJ. Technical note: Quantitative measures of iris color using high resolution photographs. Am J Phys Anthropol. 2012;147:141–149. doi:10.1002/ajpa.21637.
Wollstein A, Walsh S, Liu F, et al. Novel quantitative pigmentation phenotyping enhances genetic association, epistasis, and prediction of human eye colour. Sci Rep. 2017;7:43359. doi:10.1038/srep43359.
Mackey DA. What colour are your eyes? Teaching the genetics of eye colour & colour vision. Eye. 2022;36:704–715. doi:10.1038/s41433-021-01749-x.
Gong P, Almasian M, van Soest G, et al. Parametric imaging of attenuation by optical coherence tomography: review of models, methods, and clinical translation. J Biomed Opt. 2020;25:040901. doi:10.1117/1.JBO.25.4.040901.
International Commission on Illumination. ISO/CIE 11664-4:2019. Colorimetry — Part 4: CIE 1976 L*a*b* Colour Space.
Lee RY, Lin SC, Chen RI, Barbosa DT, Lin SC. Association between light-to-dark changes in angle width and iris parameters in light, dark and changes from light-to-dark conditions. Br J Ophthalmol. 2016;100:1274–1279. doi:10.1136/bjophthalmol-2015-307393.
International Commission on Illumination. ISO/CIE 11664-6:2022. Colorimetry — Part 6: CIEDE2000 Colour-Difference Formula.
Chang S, Bowden AK. Review of methods and applications of attenuation coefficient measurements with optical coherence tomography. J Biomed Opt. 2019;24:090901. doi:10.1117/1.JBO.24.9.090901.
Edwards M, Cha D, Krithika S, et al. Iris pigmentation as a quantitative trait: variation in populations of European, East Asian and South Asian ancestry and association with candidate gene polymorphisms. Pigment Cell Melanoma Res. 2016;29:141–162. doi:10.1111/pcmr.12435.
Invernizzi A, Giardini P, Cigada M, Viola F, Staurenghi G. Three-dimensional morphometric analysis of the iris by swept-source anterior segment optical coherence tomography in a Caucasian population. Invest Ophthalmol Vis Sci. 2015;56:4796–4801. doi:10.1167/iovs.15-16483.
Li X, Hu L, Yang B, et al. Quantitative measurement of iris melanin concentration by polarization-sensitive anterior segment optical coherence tomography. Eur J Ophthalmol. 2026;36:872–880. doi:10.1177/11206721251407025.
Mete M. Lumineyes Optical Architectural Typology (L-OAT): a multimodal hypothesis for quantitative iris phenotyping. Zenodo Preprint. Version 1.0. 30 September 2026. doi:10.5281/zenodo.23066772.

Publication and evidence disclosure. Mustafa Mete, MD, is the proposer of the Lumineyes Optical Architectural Typology (L-OAT) framework and developer of the Lumineyes™ methodology. The journal-form Perspective manuscript and the Zenodo preprint are academic versions of the framework. This Research Archive page is an original web synthesis and is not intended to replace or reproduce either manuscript in full.

Evidence status. L-OAT has not yet established validated phenotype classes, clinical predictive accuracy, or a method for prescribing a final iris color. The framework is presented as a measurement language and hypothesis-generating research architecture whose technical and clinical value must be tested empirically.

Medical information notice. L-OAT is not a substitute for ophthalmic examination, treatment monitoring, or individual clinical decision-making.

About the author. Mustafa Mete, MD, is an ophthalmologist and the proposer of L-OAT. His research interests include iris pigmentation biology, quantitative phenotyping, anterior-segment imaging, OCT-derived signal analysis, laser–tissue interaction, and response-guided treatment frameworks. See the MyLumineyes™ Research Hub and Research Library for related records.