Research & validation · iris-image analysis
ExploreIris™ is built around a research question developed over decades: whether patterns identified through structured iris-image analysis show reproducible relationships with independently documented information.
We do not consider the question answered.
We believe it is testable — and that a meaningful effect should be reproducible independently.
This page sets out what has been done, what is being done now, and how an independent partner could test it with us.
In one minute
The question
We are not asking anyone to accept historical theories about the iris.
We are posing a contemporary, testable question: does a standardised iris image contain a measurable and repeatable signal that can be identified analytically and compared against independent reference data?
If it does not, a well-designed study should show that.
If it does, the effect should be independently replicable.
Put precisely, the question is whether patterns identified by the ExploreIris™ model in standardised iris images correspond, reproducibly, to information documented independently of the model.
That is a narrower question than it may appear, and deliberately so.
It assumes nothing about disease, diagnosis or whether such a relationship exists at all.
It asks only whether an observed signal is real and repeatable — which is the only question that can sensibly be answered first.
Only if this first question is answered convincingly would broader questions — including possible future medical applications — become appropriate to investigate.
Everything else depends on the answer.
“I did not begin because I believed.
I began because I saw something I could not explain — and decided it had to be tested.”
For almost twenty years someone close to me had a markedly elevated ESR that no one could explain.
Years later, imaging finally revealed ankylosing spondylitis.
A few days after the diagnosis, a visitor with an interest in natural medicine offered to look at his eyes, looked briefly and asked: “What is happening with your spine?”
What stayed with me was not that someone had named a disease from the eye.
It was that a person who knew nothing about the medical history had asked about the very area in which a long-standing problem had just been confirmed.
My first reaction was sceptical. I do not believe in miracles — but a fact is a fact, and I have to understand what happened. Everything since has been an attempt to turn that question into something structured, measurable and open to verification by others.
Andrzej Święcicki · Founder
The result was clear, but no one could explain why it stayed high.
By the time the diagnosis came, the disease had already caused irreversible changes.
So I began to look.
To observe, compare and document — not for belief, but for repetition, structure and consistency.
Over the years individual observations became recurring patterns, patterns became a structured model, and eventually the question could be asked differently: can these patterns be measured, reproduced, and compared against information obtained independently?
I do not regard that case as proof.
It was the reason I began asking.
Not from certainty to confirmation.
From an observation that did not fit what I knew, to a question I believed deserved to be tested.
What makes it unusual
Most wellness tools return a single figure: stress, recovery, readiness, biological age.
ExploreIris™ does not.
What we are trying to establish
What keeps us researching is not a promise.
It is an observation we have seen repeatedly enough to believe it deserves a rigorous test.
The working observation motivating this research is that the correspondence appears to be regional.
Not a general signal of strain, but a relationship between specific areas of the image and specific areas of the body — one that a practitioner can identify independently, without seeing the image.
What follows are questions about the method, not claims about what the method does.
Each depends on the one before it, and none of them can be answered by us alone.
Does a standardised iris image contain a measurable and repeatable signal that can be identified analytically?
This is the only question that can sensibly be answered first
When the same person is measured again, is the result comparable across time, operators and devices?
Do the patterns identified by the model correspond with independently documented information at a rate above chance?
Can an external team, working with its own cases and its own criteria, obtain a comparable result?
Our hypothesis is testable. The profile is generated before any clinical information is disclosed, and the result can afterwards be compared against independent documentation.
If concordance does not exceed chance, the hypothesis fails.
Only if these questions are answered convincingly would broader questions become appropriate to investigate. Any such investigation would follow the appropriate scientific and regulatory pathways.
We are not there, and we do not present the current work as if we were.
What we suspect, and have not shown
The four questions above concern whether the signal is real.
These four concern what it might be. None of them has been tested against established indicators. We set them out because they define what we are looking for — and what would have to be shown before any of it could be claimed.
First
That the iris acts as an interface reflecting adrenergic activity in the body — particularly where that activity is sustained rather than momentary.
Second
That the patterns recorded there describe a predisposition to regional circulatory disturbance — a tendency within the model, not a diagnosis of anything.
Third
That these patterns describe a configuration that has become established over time, rather than a reaction to current circumstances.
Fourth
That the information covers a broad spectrum of vascular, neural and muscular-tension responses — so that the image would reflect not only where a process sits, but its wider context, connected reflexively across the body.
Established measures of autonomic activity already exist. Comparing our measurements against them is a study we intend to run — and one a research partner could run independently of us.
If the mechanism is confirmed, it would open a further research question: whether these patterns relate to processes that later become clinically apparent. That question would require its own study, its own evidence and its own regulatory pathway.
Current programme
21
cases in the series
Across twenty-one cases, reports were prepared without any access to clinical documentation. Each report was locked before any reference material was reviewed.
Adequate independent reference information was available for 58.5% of the principal functional-regional units. Within that evaluable subset, functional-regional concordance was 95.6% — 72 full and 7 partial concordances across 79 units, with no clear discordances recorded under the predefined criteria.
Both figures belong together. The concordance percentage describes only the part of each profile that could be checked at all; the coverage figure describes how much that was.
Where the model indicated a region for which no relevant investigation existed, that observation was set aside rather than counted either way — it could be neither confirmed nor refuted.
Each case records how much of the profile could be assessed at all.
A report is generated without access to the participant’s reference documentation, then locked.
Only afterwards is the documentation — provided by the participant — reviewed and compared against criteria defined in advance.
A blinded observational comparison
Standardised iris images, provided without associated health documentation
ExploreIris™ produces the model-derived profile
The result is fixed before reference information is disclosed
Pre-existing reference information is then made available
The two independently obtained sources are compared
Twenty-one cases are not sufficient to establish clinical performance or population-level conclusions. This is a set of observations consistent enough to justify asking for proper independent validation — which is what we are now seeking.
Not sufficient to claim clinical performance.
Consistent enough to justify larger, independently conducted testing.
The full evaluation protocol and per-case breakdown are available to research partners and clinical reviewers on request.
What concordance means here
Functional-regional concordance is the degree to which a model-derived pattern and independently established information refer to the same general functional region, body area or system-level context.
It does not mean that ExploreIris™ identified a disease, reproduced a diagnosis or predicted a laboratory or imaging result. It is not diagnostic accuracy.
How units were classified
Full concordance. Same functional region or axis, compatible overall direction. Scored 1.0.
Partial concordance. Broader region corresponded, but the match was incomplete — scope, side or hierarchy. Scored 0.5.
Clear discordance. Adequate reference information showed a genuinely contradictory region or direction. Scored 0.
Not currently evaluable. No adequate reference assessment existed. Kept outside the denominator entirely.
Why only part of each report could be checked
Existing medical documentation rarely covers every region described in a profile. Laboratory work, imaging and specialist assessments are performed selectively, according to symptoms and clinical indication.
Where no corresponding assessment existed, the unit was neither counted as concordant nor as discordant. A missing reference means the relationship cannot be assessed — not that it was confirmed, and not that it failed.
What this does not establish
Diagnostic accuracy, sensitivity or specificity. The diagnosis of any disease. The prediction of any laboratory or imaging result. Clinical effectiveness. Any ability to replace conventional medical assessment.
Twenty-one cases cannot establish clinical performance. The next stage is prospective and independent replication, using predefined criteria and broader reference assessment.
Earlier research · 2011–2012
In 2011 we ran a study designed so that neither side could influence the other.
It remains the most rigorous piece of work behind ExploreIris™.
The 2011 work forms part of the historical research background from which the current ExploreIris™ model developed.
We do not present this study as validation of the current system.
Images were taken at the start of a rehabilitation programme, without any discussion of the participants’ complaints.
Markers were derived from those images.
Independently, a practitioner recorded responses using his own method, with no access to the images or to the analyses.
Reference records were opened only after the programme ended.
Treatment points were graded on a three-point scale at the time of treatment: 1 — weak local response; 2 — intermediate; 3 — pronounced systemic response.
Only points graded 3 were included.
The two datasets were produced independently and compared only afterwards.
57
Participants
p < 0.001
On the primary comparisons
Blinded
On both sides
Analysed parameters were described using arithmetic mean, standard deviation, and minimum and maximum values.
Because the parameters were not normally distributed, as shown by the Shapiro–Wilk test, groups were compared using the Mann–Whitney U test.
The significance level was set at p < 0.05.
Calculations were performed using the SPSS statistical package.
1,710 sector measurements from 57 patients: 628 sectors identified by treatment, compared against 1,082 remaining sectors.
Nine parameters were examined; six reached p < 0.001.
Sectors identified by treatment showed lower luminance than the remaining sectors.
The analysis was commissioned externally and carried out independently of the team.
The study was conducted in 2011; the methodology and results were presented at a manual-therapy conference in March 2012.
Limitations: a single centre, a single assessor, and a selected group — only patients who obtained significant relief after treatment were included.
The finding concerns sectors mapped to metameric levels, not to organs.
Why. Material from this study contributed to the development of the current analytical model.
Data used to build a model cannot also confirm it.
That is why a separate programme exists, and why independent replication with a different practitioner is the step we are taking now.
The technical lineage reaches further back: a peer-reviewed paper published in 1996 described a method for measuring the luminance of light reflected from the human iris.
It described a measurement method, not a relationship to physiological state. See the full timeline →
Four kinds of evidence, kept separate
We distinguish clearly between historical clinical observation, earlier scientific work, current exploratory evaluation and future independent validation.
None of these are presented as equivalent forms of evidence.
Validation partnerships
If you work in research, clinical science or health technology and would like to test the method independently, we are open to structured evaluation.
A pilot can be structured so that the partner controls participant selection and reference documentation.
ExploreIris™ receives only the material required to perform its analysis.
Reports are locked before any reference information is disclosed.
Comparison follows criteria agreed in advance.
The question is not whether you should believe us.
The question is whether the effect can be reproduced.
Cases and reference documentation are chosen and held by the partner, or by an organization the partner appoints
ExploreIris receives only the material needed for analysis.
Reference information stays with the partner
Reports are completed and locked before anything is disclosed
Reference information is revealed and compared against predefined criteria.
All cases are included — concordant and discordant alike
The appropriate first step is a controlled pilot built around predefined criteria and a limited number of documented cases.
It does not require a large programme or a significant commitment.
If the results justify further investigation, the collaboration can progress toward a larger independent validation programme.
We are open to jointly supported validation programmes and strategic collaborations where there is a shared interest in evaluating the technology further.
Terms are discussed individually.
This page is an invitation to test, not a commercial proposal.
The larger opportunity
ExploreIris™ has a defined role today within professional wellness and longevity.
At the same time, the underlying analytical technology is being investigated for a potentially broader future role.
If prospective independent validation confirms reproducible associations between predefined image-derived patterns and independently established clinical information, this could provide a scientific basis for exploring future regulated applications in healthcare.
The current non-diagnostic offering may therefore represent the first application of a broader analytical platform — not necessarily the limit of its future potential.
Any future regulated application would require its own evidence, a clearly defined intended purpose and the appropriate scientific and regulatory review.
Non-diagnostic functional orientation
Professional wellness & longevity
Predefined image-derived patterns
Independent reference information
Reproducibility
Regulated healthcare applications
only if supported by evidence
Selected partner access
We welcome conversations with selected partners interested in independent validation, research collaboration and the long-term development of the ExploreIris™ platform.
Supported validation
Independent evaluation requires dedicated funding.
We are seeking a partner with a genuine interest in obtaining an independent answer, and willing to support that evaluation financially.
Our preference is for the study to be conducted or supervised by an independent research or clinical team, with endpoints defined in advance and results reported transparently — including negative ones.
The partner would fund the evaluation and provide access to an appropriate setting and reference material.
ExploreIris™ provides the technology, the protocol and analytical support.
We are not raising capital or offering equity.
This is a research partnership, not an investment proposal.
A negotiated first-mover position — for example a time-limited right of first negotiation, or priority access within an agreed field or territory.
Scope is defined contractually.
Scientific independence comes first. Terms are structured so that neither party is committed to commercial deployment before the results are known, and so that the outcome of the evaluation is not influenced by the commercial arrangement.
Get in touch
No belief required.
Just data.
Whether you would like the study protocol and blinding procedure, a conversation about a validation pilot, or simply to understand the current state of the evidence — write to us and we will reply personally.
Would you rather discuss it first? Get in touch and we’ll arrange a conversation.