On this page
- First: Is the Demand Even There?
- The Decision Starts Before the Consult
- Does Visualization Actually Work? The Honest Evidence
- The Two Categories — and Where Each One Loses
- Try an Embeddable Simulator
- The 8 Criteria That Actually Matter
- Applying These Criteria to Mirror (By Its Own Test)
- So — Should You Use One?
TL;DR
- Patient demand is at an all-time high: ~70% of Americans are considering a cosmetic dermatologic procedure (ASDS, 2025) — but cost (24%), safety (16%), and pain (14%) are the top reasons they hesitate.
- The decision starts online. 95% of patients research before they ever consult, and ~94% say online reviews influence which provider they pick. Helping them see the result before booking is the unclaimed advantage.
- Simulators split into two categories with real trade-offs: five-figure in-office 3D hardware (VECTRA, Crisalix) wins on clinical fidelity; embeddable web AI (Perfect Corp, Mirror) wins on cost and top-of-funnel reach. This guide includes where each one loses.
- The 8 criteria that actually matter — photorealism (validated, not vendor-claimed), breadth, speed, cost structure, embeddability, lead capture, HIPAA/data handling, and independent evidence — applied honestly, including to the tool this guide is published by.
- Try a live embeddable simulator in the middle of this guide and judge it against the criteria yourself.
If you run a dermatology or aesthetic practice, you've probably watched "AI treatment simulator" go from a novelty you saw at a conference to something patients are starting to expect. The question isn't whether visualization tools are coming to aesthetics — they're here. The question every owner is now asking is the practical one: do I need one, and how do I tell a genuinely useful tool from an expensive gimmick?
This guide answers both. Full disclosure up front: it's published by ClearPath AI, which makes one of these tools (Mirror). So we've done two things to keep it useful rather than self-serving — we've included the criteria where our own category of tool is weaker, and we apply every test in this guide to Mirror itself, by name, later on. Judge accordingly.
First: Is the Demand Even There?
Yes — and it's well-documented. The American Society for Dermatologic Surgery's 2025 Consumer Survey (n=3,500+) found that roughly 70% of Americans are considering a cosmetic dermatologic procedure — a number that has stayed elevated since it doubled in the mid-2010s.
But the same survey is just as clear about why people hold back:
24%
cite cost as the top barrier to treatment
16%
cite fear of side effects / safety
14%
cite pain or recovery concerns
Look closely at those barriers. Cost, safety, and "will it be worth it?" are all, at root, uncertainty about the outcome. A patient who can clearly picture the result they're paying for evaluates cost differently, weighs the risk differently, and decides faster. That's the entire premise of a treatment simulator: convert uncertainty into a concrete, visual expectation.
The Decision Starts Before the Consult
Here's the structural shift that makes this a 2026 question and not a 2016 one: the aesthetic decision increasingly begins online, before the patient ever walks in.
Quick Takeaway
A peer-reviewed study found 95% of aesthetic patients research online and 45% use social media before even booking a consultation (PMC6084688). The separate ASDS 2025 survey adds that ~94% say online reviews and ~50% say a provider's social presence shape who they choose. By the time a patient sits in your chair, much of their research — and their shortlist — is already set.
This is why where a simulator lives matters as much as how good it is. An in-office-only tool — however impressive — can only influence the patient after they've already chosen you and shown up. A simulator embedded on your website influences the decision earlier, at the research stage where most patients start.
Does Visualization Actually Work? The Honest Evidence
You'll see vendors claim everything from "14x sales uplift" to "doubles conversion." Most of those numbers are vendor marketing with no methodology — ignore them. Here's what the independent, peer-reviewed research actually establishes:
- Visualization-enabled consults convert as well as in-person ones. A 2025 study in PRS Global Open found 3D-imaged consultations converted at 87% versus 90% in-person — statistically indistinguishable. Visualization held conversion while removing the requirement to be in the room.
- Seeing the result builds confidence and sets expectations. In a peer-reviewed breast-augmentation study, the large majority of patients said 3D simulation helped them choose their procedure, and most said the final result was close to the simulation — meaning fewer surprised patients and fewer revisions.
💡 What we will NOT claim
We're not going to tell you a simulator will lift your conversion by a specific percentage — no credible study has measured that for dermatology specifically, and the precise figures vendors quote are unverifiable. The defensible, research-backed claim is narrower and still powerful: visualization holds conversion while expanding reach, builds patient confidence, and aligns expectations — which is exactly what reduces the cost/safety/"worth it?" hesitation the ASDS data identifies.
The Two Categories — and Where Each One Loses
For most of the last decade, "treatment simulator" meant dedicated 3D-imaging hardware — a capture booth, a trained operator, and a five-figure capital outlay. Canfield's VECTRA is the category leader; even used systems resell for $10,000–$14,500, and new systems are quote-only and substantially higher. Crisalix offers 3D simulation as well. The newer generation is software: web/photo-based AI (Perfect Corp / YouCam, Crisalix's web tools, and Mirror) that runs on any phone or laptop and embeds on your site.
Here's the honest head-to-head — including the rows where an embeddable web tool is worse:
| Factor | In-office 3D hardware (VECTRA / Crisalix) | Embeddable web AI (e.g., Mirror) |
|---|---|---|
| Capture consistency | Controlled lighting, angle, distance | Varies with the uploaded selfie |
| Clinical color / pigment accuracy | Calibrated | Not calibrated for clinical color |
| True 3D volumetrics (filler, rhino, breast) | Yes | No — a 2D photo preview |
| Standardized medico-legal documentation | Yes | No — not a records system |
| Cost structure | Five-figure capital purchase | Low monthly subscription |
| Reaches prospects before they book | No — in-office only | Yes — on your website |
| Captures the lead at peak intent | No | Yes — gated at the result |
Read that honestly: hardware wins the top four rows — the clinical ones. If you do precise volumetric planning for breast, rhinoplasty, or structural filler, or you need calibrated, repeatable, documentation-grade captures, dedicated 3D hardware has real value a web tool can't match. A web tool eats whatever selfie the patient uploads under whatever lighting they have — that's a genuine limitation for clinical accuracy.
Where web tools win is the bottom three rows — reach and conversion. They live where 95% of patients start (online), and they capture the lead at the moment of peak intent. So the real question isn't "which is better?" — it's "which problem am I solving: clinical planning, or turning online researchers into booked consults?" Many practices end up wanting both, for different stages.
Try an Embeddable Simulator
Rather than describe it, here's a live one. Below is Mirror by ClearPath AI — the same embeddable simulator a practice would put on its own site. Upload a photo, choose a treatment, and watch it render. Photorealism is the make-or-break axis, so judge it yourself rather than trusting the word:
As you use it, run it against the criteria below — and notice its limits, too. The result is a believable, conservative simulation framed to the patient as a discussion starter, not a clinical guarantee — not a calibrated clinical image. That framing is deliberate: an over-promising render is a liability, and the right job for a web preview is to start the conversation, not finish the diagnosis.
The 8 Criteria That Actually Matter
Use this as your evaluation checklist for any AI treatment simulator — Mirror or otherwise.
1. Photorealism & accuracy
The make-or-break axis. An over-promising, cartoonish result creates liability and disappointed patients. Demand evidence that is independently validated or testable by you — not a vendor adjective. Peer-reviewed work shows simulation accuracy directly correlates with post-op satisfaction, so this is clinical, not cosmetic.
2. Breadth of treatments
Match the procedures the tool covers to your actual service menu — injectables, skin/laser, surgical. A tool that only simulates one category limits its usefulness across your patient mix.
3. In-consult speed & workflow fit
Can it produce a result in real time, on a phone or tablet, without a dedicated room or operator? Friction in the consult flow kills adoption by your staff.
4. Cost structure: capex vs subscription
Five-figure hardware vs a low monthly SaaS fee is the core financial decision. Match it to whether you need clinical-grade 3D or top-of-funnel conversion.
5. Embeddability & top-of-funnel reach
Can it live on your website so prospects visualize results before booking — where most research happens — or is it trapped in the office? This is where web tools earn their keep.
6. Lead capture at peak intent
A web simulator should capture the lead (email or booking) right when the patient sees their result. Hardware cannot do this; it is a structural advantage of an embeddable tool.
7. HIPAA & image data handling
Patient face photos are sensitive. Ask the exact questions: Are photos stored or discarded? Used to train models? Is any PHI collected? Get specific, binding answers — vagueness here is a fast disqualifier.
8. Skin-tone performance & independent evidence
AI face models have historically underperformed on darker skin (Fitzpatrick IV–VI) and pigmentary conditions. Test the tool on photos that match YOUR patient population. And discount vendor-only stats — weight what you can verify yourself.
🎯 The fastest disqualifier
If a vendor can't give you a straight answer on #7 (are photos stored? used to train models? any PHI?), stop there. You're putting patient faces into this system. A serious tool treats that as the first conversation, not a footnote — so let's hold our own tool to it.
Applying These Criteria to Mirror (By Its Own Test)
It would be hypocritical to publish those eight criteria and exempt our own product. So here are the concrete answers for Mirror:
- #7 Data handling (the disqualifier): An uploaded photo is sent to the AI model, processed in memory for ~20–30 seconds, and discarded immediately — never stored, cached, logged, or retained, and not used to train AI models. Mirror collects no medical history or diagnosis data, so it doesn't handle PHI and isn't a HIPAA covered entity. (We say "no PHI, photos not stored" — not "HIPAA-compliant," because the honest distinction matters.) Details: privacy page.
- #1 Photorealism: there's no published study on Mirror specifically — the peer-reviewed evidence in this guide is on 3D simulation/DSD as a category. By criterion #1's own logic, don't take our word: the 30-second demo above is the test. Run it on your patients' faces.
- #8 Skin tone: Mirror runs on Google's current image model, and customers have specifically noted accuracy on South Asian skin — but treat that as a prompt to test it yourself across Fitzpatrick IV–VI, not a guarantee.
- #4 Cost: free for 14 days, then $347/month (no setup fee), or a one-time $2,497 license — versus a five-figure hardware capital purchase. Plus an 8-week consult guarantee: if it doesn't help produce 3+ qualified consults in your first 8 weeks, full refund.
- #5/#6 Reach & capture: one embed snippet, live in ~24 hours on any website platform, with lead capture at the result and Cal.com booking built in.
Where Mirror is not the answer: if you need calibrated 3D volumetrics or documentation-grade clinical imaging, that's hardware's job (see the comparison table). Mirror is built for the top-of-funnel conversion problem, not clinical planning.
So — Should You Use One?
Here's the honest decision framework:
⚠️ You should probably wait if…
- Your real bottleneck is upstream — no traffic and no leads at all yet
- You can't commit to actually using it in the consult flow
- What you actually need is clinical-grade 3D planning (that's hardware, not a web tool)
✅ You should adopt one if…
- You get consults but lose them to "let me think about it"
- Prospects compare you to competitors online before they ever book
- Your high-value treatments need the patient to see it to commit
For most aesthetic and dermatology practices in 2026, the answer for a web simulator is yes — if it's photorealistic enough on your own patients, captures leads, and is straight about data handling. The demand is documented (70% considering treatment), the research happens online (95% research first), and the barriers are uncertainty-based (cost, safety, "is it worth it?") — exactly what a credible visualization removes.
70%
of Americans are considering a cosmetic dermatologic procedure
ASDS 2025. The demand is there. The practices that help patients see the result — at the online stage where research starts — are the ones who'll convert it.
See an embeddable simulator built for aesthetic practices
Mirror is the web-based AI before/after tool for dermatology & aesthetics — embeds in ~24 hours, photos never stored, no PHI collected. Free for 14 days, with an 8-week 3-consults-or-refund guarantee.
See Mirror for Dermatology →Want to score it against the 8 criteria one more time? Here's the live demo again — try a different treatment and judge the photorealism, speed, and skin-tone handling for yourself:
When you're ready, see how Mirror works for dermatology practices — or explore Mirror across every aesthetic vertical.
Sources: ASDS 2025 Consumer Survey; peer-reviewed patient online-behavior study (PMC6084688); PRS Global Open, 2025; breast-augmentation 3D simulation study (Annales de Chirurgie Plastique Esthétique, 2017); VECTRA resale pricing via medical-equipment marketplaces. Vendor-reported figures (e.g., "14x uplift," "doubles conversion") are excluded as unverifiable. No published study has tested Mirror specifically; cited evidence is on 3D simulation/DSD as a category.