Mr. Onur Gilleard
Qualifications & Experience
Mr Onur Gilleard is a consultant plastic surgeon on the GMC specialist register. He previously held an NHS consultant post at St Bartholomew’s Hospital in London, specialising in skin cancer and microsurgical reconstruction. In 2021, Mr Gilleard transitioned from the NHS to dedicate himself fully to private practice, allowing him to focus exclusively on providing personalised, high-quality care to his patients. He is recognised as an expert in laser treatments, having conducted research and developed advanced treatment protocols using cutting-edge laser technology to optimise both clinical outcomes and cosmetic results.
Get started today
Book a Consultation
The FotoFinder bodystudio ATBM system does two things that a clinician cannot do unaided: it records the skin in standardised, reproducible conditions, and it compares this year’s images against last year’s at a level of precision below human perception. What it does not do is decide. Every flagged lesion is assessed clinically by a consultant plastic surgeon, and the software’s output is a shortlist for examination rather than a diagnosis. That division of labour is the whole argument for the technology, and it is also the reason a phone app is not a smaller version of it.
The System in Use
Mole mapping carried out with the FotoFinder® system at 101 Harley Street. The dermatoscope head is placed on individual lesions to capture close-up images; these sit alongside the standardised full-body photographs and form the record that next year’s images are compared against.
Two Kinds of Image, Doing Different Jobs
A mole map is not one photograph. It is two distinct image sets, and confusing them is the source of most misunderstanding about what the technology achieves.
| Total-body photography | Dermoscopic imaging | |
|---|---|---|
| What it captures | The whole skin surface from standardised angles | Individual lesions at high magnification, through the surface layer |
| What it answers | Is there a lesion here that was not here before? | What is the internal structure of this particular lesion? |
| Detects | New lesions; lesions that have grown or moved position in the overall map | Pigment network, vessel patterns, colour distribution — features invisible to the naked eye |
| Comparable year on year | Yes, provided position and lighting are standardised | Yes, and this is where subtle change shows first |
Dermoscopy is what separates this from photography. A dermatoscope uses polarised light and magnification to see beneath the stratum corneum, revealing structures that determine whether a lesion is behaving normally. The techniques and what the structures mean are covered in digital dermoscopy explained and polarised versus non-polarised dermoscopy.
What the AI Actually Does
Two separate things are happening on that screen, and they are worth separating.
The scoring systems are established dermoscopic algorithms rather than machine learning: the 3-point checklist, the 7-point checklist and the ABCD rule, each applied independently to the same lesion. A lesion has to look unremarkable under all three.
The AI panel is separate. It returns a score between 0 and 1, banded as 0.0–0.2 inconspicuous, 0.21–0.49, and 0.50–1.0, alongside a distinct classification of how melanocytic the lesion appears — 95% in the example shown. There is also a heatmap view indicating which regions of the image drove the score.
The module in use is FotoFinder Moleanalyzer pro. It scores lesions and flags those with features associated with malignancy, comparing the current image set against previous scans. Specifically it identifies:
- Changes in size, shape or colour of an existing lesion
- New lesions that were not present at the previous visit
- Asymmetry or irregular borders developing where they were not before
- Variation in colour or texture within a single lesion
The value is consistency rather than cleverness. A clinician comparing this year’s skin against a memory of last year’s is attempting something impossible; the software performs a straightforward comparison, on every catalogued lesion, at a resolution where a fraction of a millimetre registers.
“It is decision support. It does not diagnose, it is not a second opinion, and it does not override the clinician: a lesion I am worried about is excised regardless of its score, and a reassuring score never keeps a suspicious lesion in place.”— Mr Onur Gilleard, Consultant Plastic Surgeon
What It Does Not Do
It does not diagnose, and it does not decide. The final assessment is made by a consultant plastic surgeon who reviews the flagged lesions clinically and determines whether the answer is removal, biopsy or continued monitoring.
Understanding why this matters requires being clear about what a flag is. The algorithm reports a difference. A difference can mean a melanoma developing; it can equally mean a lesion that was photographed at a slightly different angle, skin that has tanned since the last visit, a benign lesion enlarging in the ordinary way, or a seborrhoeic keratosis appearing with age. Distinguishing between those requires clinical context that no image contains — the patient’s history, the behaviour of their other lesions, what the lesion feels like on palpation.
Run the other way, this also explains why the absence of a flag is not clearance. Software compares against what it was given. A lesion in a site that images poorly, or one that has not changed but was always abnormal, is not going to be flagged by a comparison algorithm. That is precisely the gap clinical examination fills.
Why a Phone App Is Not a Smaller Version of This
Consumer mole-check applications describe themselves in similar language — AI analysis, risk scoring, skin cancer detection — and the resemblance is superficial. Four differences are structural rather than incremental.
- No dermoscopy
- A phone camera photographs the skin surface. It cannot see the sub-surface pigment network or vessel structures that dermoscopy exists to reveal, and those structures are where the diagnostic information sits.
- No standardisation
- Distance, angle, lighting and white balance vary with every photograph. Two images of the same unchanged lesion taken a year apart on a phone may differ more than two images of a genuinely changing lesion captured in standardised conditions.
- No baseline of record
- Comparison requires a controlled previous image. Casual photographs stored in a camera roll do not provide one.
- No clinician
- The output is a score presented to the person it concerns, with no examination, no history and no accountability. A reassuring score on an unexamined lesion is worse than no score at all, because it supplies a reason not to attend.
The evidence supports that scepticism. A systematic review in the BMJ of algorithm-based smartphone apps for skin cancer found sensitivity that was poor, inconsistent, and below what would be needed to safely rule out melanoma (Freeman et al., 2020); several apps have since been withdrawn or reclassified. The problem is not that the apps do nothing — it is the direction in which they fail.
“A false alarm costs an appointment. False reassurance costs months, and months change Breslow thickness.”— Mr Onur Gilleard
“An app or a photograph can tell you that a mole looks unusual today, but it cannot tell you that it looked different six months ago, cannot feel that it has become firm, and cannot see the structures under the surface that decide whether it comes off.”— Mr Onur Gilleard
None of this means self-monitoring is pointless — it is valuable, and the practical method is in how to photograph a mole properly. It means self-monitoring is a prompt to be seen, not a substitute for being seen.
Where the Consultant Adds What the Machine Cannot
- Palpation. Firmness, thickness and whether a lesion is fixed to underlying tissue are not visible in any photograph. A firm lesion is treated differently from a soft one that looks identical.
- Context. The same dermoscopic pattern carries different weight in a patient with a family history of melanoma than in one without.
- The ugly duckling judgement. Recognising that one lesion does not belong among a patient’s others is a comparison across the whole person, not between two images of one spot — see the ugly duckling sign.
- Deciding what to do. Whether a lesion is removed, watched or biopsied — and if removed, by which technique so that histology remains possible — is a clinical decision with consequences the software does not carry.
“The real danger is complacency. Occasionally something that looks harmless is not, which is why every lesion is examined before anything is done to it and anything atypical is sent for histology.”— Mr Onur Gilleard, Consultant Plastic Surgeon
Your Images
A mole map is a set of full-body clinical photographs, and that raises reasonable questions about storage and access. They are held as medical records under the same duties of confidentiality as any other clinical information, retained because comparison over time is the entire point of holding them, and accessed by the clinicians involved in your care. The specific questions patients ask about consent, retention and security are addressed in AI mole mapping, privacy and image security.
Cost
| Item | Price |
|---|---|
| Mole mapping — complete package | £300 |
| Initial consultation | £100 |
| Mole removal | £350 |
| Each additional mole removed | £200 |
| Histological analysis | £180 |
Appointments at 101 Harley Street: mole mapping or contact the clinic. What the appointment involves is set out in your mole mapping appointment.
Frequently Asked Questions
- What is FotoFinder mole mapping?
- A digital dermoscopy and total-body photography system that records the skin in standardised conditions and compares images against previous visits. It produces two image sets — whole-body photographs and high-magnification dermoscopic images of individual lesions — which together form the record used for year-on-year comparison.
- Does the AI diagnose melanoma?
- No. It compares images and flags lesions that are new or have changed. The clinical assessment and the decision about what to do are made by a consultant plastic surgeon. A flag is a prompt to examine something closely, not a finding.
- How accurate is it?
- Its strength is detecting difference between two standardised images, which it does at a resolution below human perception. Its limitation is that a difference has many possible causes, most of them benign, and separating them requires clinical context the images do not contain.
- Is this the same as a mole-check app?
- No, and the differences are structural. Apps have no dermoscopy, no standardised conditions, no clinical baseline and no clinician. A BMJ systematic review of algorithm-based skin cancer apps found sensitivity that was poor, inconsistent and below what would be needed to safely rule out melanoma (Freeman et al., 2020), and several have since been withdrawn or reclassified.
- Is the imaging harmful in any way?
- No. It is photography and magnified optical imaging using visible light. There is no radiation, nothing is injected, and nothing touches the skin except the dermatoscope lens.
- What happens if the system flags one of my moles?
- The consultant assesses it clinically and decides between removal, biopsy or documented monitoring at the next visit. Most flagged lesions prove benign; the flag identifies which lesions deserve close attention rather than declaring what they are.
- Is it useful on a first visit if there is nothing to compare?
- The comparison function has nothing to work with at baseline, so its contribution is limited on a first map. Anything already abnormal is picked up clinically, and the record created becomes the reference for every subsequent visit.
- Who can see my images and how long are they kept?
- They are held as confidential medical records, accessible to the clinicians involved in your care, and retained because comparison over time is the purpose of taking them. Retention and consent are covered in detail in our guide to mole mapping privacy.
Share this Post