Dermatology · Intelligent dermatology platform — visual search, teledermatology, lesion risk
DermEngine
MetaOptima Technology
DermEngine, from Vancouver's MetaOptima, is best understood as a teledermatology platform with assistive AI rather than a cleared diagnostic device — and the distinction is the whole story. The company's strongest evidence is genuine: a manufacturer-funded, multicentre, prospective clinical trial in the Lancet Digital Health found its smartphone AI matched specialists and beat junior doctors at diagnosing pigmented lesions, though clinicians remained clearly superior at the harder question of what to do next. But the regulatory reality is unambiguous and must not be dressed up: the diagnostic AI has no FDA clearance, no CE notified-body certificate and no equivalent authorisation anywhere. "FDA Registered" refers only to a Class I establishment listing of the dermatoscope hardware, and the FDA Breakthrough Device Designation the company holds for its DermDx algorithm is a review-priority status, not an approval — a pivotal US trial was still recruiting. Real, self-declared hardware registrations (CE Class I, an Australian TGA listing), a clean safety record and a prospective trial earn one mark; the absence of any diagnostic-device clearance and of independent validation, plus mixed management-decision results, hold it firmly to one.
Performance Metrics
Clinical Evidence
The evidence base for DermEngine's AI is dominated by one strong prospective study and a cluster of earlier manufacturer methods papers, and the two tiers should be weighed differently. The load-bearing study is a manufacturer-funded, multicentre, prospective diagnostic clinical trial (Lancet Digital Health 2023; PMID 37775188). Images were captured with a MoleScope II dermatoscope on a smartphone and analysed by two AI applications — a novel seven-class algorithm and an ISIC-based algorithm — embedded in DermEngine. The diagnostic arm evaluated 172 suspicious pigmented lesions from 124 patients; a separate management arm covered 5,696 pigmented lesions from 66 high-risk patients. The headline finding is that the AI performed comparably to specialists and better than junior/novice clinicians on diagnosis, while clinicians were clearly superior on management decisions — what to biopsy, monitor or reassure. The precise effect-size estimates reported by the trial are manufacturer-reported and were not independently reconfirmed against the full text in this review, so they are described directionally here rather than quoted as verified figures. Even at face value the result supports an assistive role, not autonomous diagnosis, and least of all autonomous management. The supporting literature is older, smaller and manufacturer-authored: a content-based image-retrieval pilot (Skin Research and Technology 2020; PMID 31845429), a study of retrieval accuracy with deep classification features (British Journal of Dermatology 2019; PMID 30207594), and a multimodal deep-learning classifier that reported a melanoma AUC of 0.866 versus a 0.784 baseline on retrospective data (Experimental Dermatology 2018; PMID 30187575). No independent, non-manufacturer clinical validation of DermEngine's AI was located. The honest reading is a promising assistive tool with one good prospective trial behind it, not a proven or authorised diagnostic device.
| Study | Design | n | Sensitivity | Specificity | AUC | Published |
|---|---|---|---|---|---|---|
| MetaOptima-funded multicentre trial group (Lancet Digital Health; prospective, manufacturer-funded) | ProspectiveProspective | 172 | — | — | AI comparable to specialists and superior to novices on diagnosis; clinicians superior on management. Precise effect sizes manufacturer-reported, not independently reconfirmed. | Lancet Digit Health, 2023; PMID 37775188 (diagnostic arm 172 lesions / 124 patients; management arm 5,696 lesions / 66 patients) |
| Yap J, Yolland W, Tschandl P, et al. (manufacturer; multimodal deep-learning classifier) | RetrospectiveRetrospective | 2,917 | — | — | Melanoma AUC 0.866 vs 0.784 baseline; multiclass mAP 0.729 vs 0.598 | Exp Dermatol, 2018; PMID 30187575 |
Clinical Pulse
Verified clinician sentiment
Verified clinician reviews launching soon · Apply to contribute →
Reviews are signed by clinicians verified against their regulator's public register.
Inside the algorithm
Editorial featureHow DermEngine turns a lesion photo into an assistive read.
Five stages — from raw input to verdict — drawn from manufacturer documentation and the public regulatory record.
- INGEST
- NORMALISE
- DETECT
- LOCALISE
- VERDICT
Stage 01 · INGEST
A clinician captures the lesion on a smartphone dermatoscope.
A clinician photographs a skin lesion using the paired MoleScope dermatoscope (or the OptimaScan device) on a smartphone, capturing both dermoscopic and clinical views. The images are uploaded into the DermEngine platform, where they are stored and organised for the patient's record, total-body photography and mole-mapping over time.
No specialised imaging suite is required — the capture hardware is a smartphone attachment — which is the platform's core accessibility premise.
Input
Dermoscopic / clinical images
Inference
Cloud, per-lesion
Inside the Auris+ Listing
Five more sections complete this device’s Auris+ Listing.
Decision Ledger
ProPro unlocks a private, cross-vendor log of every case you read on this device — what the AI called, what you concluded, and a one-line reason if you overrode it. Ready for the EU AI Act's deployer logging obligations when they land in 2028.
Clinical Evidence Deep Dive
ProPro unlocks the structured clinical-evidence summary — study count, target patient population, and a tabular accuracy-metrics view drawn from peer-reviewed sources.
Peer-Reviewed Publications
ProPro unlocks the curated peer-reviewed publication list with PubMed cross-links — the citation backbone of every editorial verdict.
Post-Market & Regulatory Conditions
ProPro unlocks the post-market surveillance summary, recall record, and the conditions of approval that bound real-world use.
AI Algorithm Version History
ProPro unlocks the chronological record of algorithm version changes — what changed when, drawn from manufacturer changelogs and regulatory filings.
Regulatory Approvals
Class I (self-declared)
285661
Class 1
Safety Record
No MetaOptima, DermEngine or MoleScope recall or adverse-event report was identified in publicly available sources as of July 2026. Because the diagnostic AI carries no FDA clearance there is no cleared-device MAUDE record specific to it to search, and the FDA enforcement and EU vigilance databases could not be queried directly from this environment, so this is "none found in public reporting," not an exhaustive audit. The AI is assistive rather than autonomous: it supports a clinician's read and referral decision and does not by itself diagnose, biopsy or treat, which bounds the harm surface. Two limitations matter for safe use. First, the one prospective trial found clinicians clearly superior to the AI at management decisions, so the tool must not drive what to biopsy or monitor. Second — and this is the load-bearing safety and neutrality point — the diagnostic AI is not authorised as a diagnostic device in any jurisdiction, so its outputs are decision support to be confirmed by a qualified clinician, and a reassuring AI result is never a substitute for clinical judgement or histopathology.
Intended Use & Indications
DermEngine is a cloud dermatology platform used by clinicians for imaging, documentation, teledermatology and lesion tracking, with AI features layered on top. The paired MoleScope smartphone dermatoscope (or the OptimaScan device) captures dermoscopic and clinical images, which the platform stores, sequences for total-body photography and mole-mapping, and analyses. The AI offers content-based image retrieval ("Visual Search"), returning catalogued lesions that look similar to a query image together with their diagnoses, and a lesion-classification / melanoma-risk model. In every mode the AI is a secondary, assistive aid: it supports a clinician's assessment and referral decision rather than issuing an autonomous diagnosis. Critically for a procurement reviewer, the diagnostic AI (marketed for FDA purposes as "DermDx") is not cleared, approved or notified-body-certified as a diagnostic device in any jurisdiction as of the review date; the only regulatory footprint is a US Class I establishment registration and self-declared Class I / low-class registrations for the MoleScope dermatoscope hardware.