Skip to main content
QMSQMS
QMS
  • Welcome to your QMS
  • Quality Manual
  • Procedures
  • Records
  • Legit.Health Plus Version 1.1.0.0
    • CAPA Plan - BSI CE Mark Closeout
    • Index
    • Overview and Device Description
    • Information provided by the Manufacturer
    • Design and Manufacturing Information
    • GSPR
    • Benefit-Risk Analysis and Risk Management
    • Product Verification and Validation
      • Software
      • Artificial Intelligence
      • Cybersecurity
      • Usability and Human Factors Engineering
      • Clinical
        • Evaluation
          • Appendix
          • R-TF-015-001 Clinical Evaluation Plan
          • R-TF-015-003 Clinical Evaluation Report
          • R-TF-015-011 State of the Art Legit.Health Plus
          • R-TF-015-013 Statistical Summary of Clinical Evidence
        • Investigation
        • Legit.Health-US-version-1-1-0-0
      • Commissioning
    • Post-Market Surveillance
  • Legit.Health Plus Version 1.1.0.1
  • Legit.Health version 2.1 (Legacy MDD)
  • Legit.Health Utilities
  • Licenses and accreditations
  • Applicable Standards and Regulations
  • BSI Non-Conformities
  • Pricing
  • Public tenders
  • Legit.Health Plus Version 1.1.0.0
  • Product Verification and Validation
  • Clinical
  • Evaluation
  • R-TF-015-013 Statistical Summary of Clinical Evidence

R-TF-015-013 Statistical Summary of Clinical Evidence

Table of contents
  • Introduction
    • Scope
  • Device intended use
    • Critical clarification on diagnostic output
  • Clinical benefits
    • Benefit 7GH -- Diagnostic accuracy improvement
    • Benefit 5RB -- Objective severity assessment
    • Benefit 3KX -- Care pathway management
  • Aggregate sample demographics
    • Sex distribution
    • Age group distribution
    • Fitzpatrick phototype distribution
  • Individual study demographics
    • MC_EVCDAO_2019 -- Melanoma detection
    • COVIDX_EVCDAO_2022 -- Severity monitoring
    • DAO_Derivation_O_2022 -- Referral optimisation (Osakidetza)
    • DAO_Derivacion_PH_2022 -- Referral optimisation (Puerta de Hierro)
    • IDEI_2023 -- Clinical flow optimisation
    • BI_2024 -- GPP and skin conditions (MRMC)
    • PH_2024 -- Primary care diagnosis (MRMC)
    • SAN_2024 -- Multi-condition assessment (MRMC)
    • AIHS4_2025 -- Hidradenitis suppurativa (retrospective)
    • MAN_2025 -- Fitzpatrick V--VI validation (ongoing)

Introduction​

This document provides a statistical summary of the clinical evidence generated through the clinical investigation programme for the device. Its purpose is to serve as a self-contained reference for the external statistics consultant reviewing the Clinical Evaluation Report (CER), enabling an independent assessment of whether the evidence base is sufficient to support the claimed clinical benefits.

Scope​

This summary covers 10 clinical investigations conducted between 2019 and 2026. Nine studies are completed; one (MAN_2025) is ongoing and is included with sample composition data only.

#Study codeDesignSampleStatus
1MC_EVCDAO_2019Prospective observational, multi-centre105 patientsCompleted
2COVIDX_EVCDAO_2022Observational, single-centre160 patientsCompleted
3DAO_Derivation_O_2022Prospective observational, multi-centre117 patients (127 enrolled, 10 excluded)Completed
4DAO_Derivacion_PH_2022Prospective observational, multi-centre131 patientsCompleted
5IDEI_2023Observational, non-interventional, single-centre204 subjectsCompleted
6BI_2024MRMC, prospective, self-controlled100 imagesCompleted
7PH_2024MRMC, prospective, self-controlled30 imagesCompleted
8SAN_2024MRMC, prospective29 imagesCompleted
9AIHS4_2025Retrospective observational2 subjectsCompleted
10MAN_2025MRMC, prospective, self-controlled149 images (planned)Ongoing

Total completed sample size: 878 subjects/images across 9 completed studies (using analysed samples where applicable).

note

Studies 1--5 enrolled patients directly in clinical settings. Studies 6--8 and 10 are multi-reader multi-case (MRMC) studies using anonymised images evaluated by panels of healthcare professionals. Study 9 is a retrospective analysis of 2 subjects from a clinical trial.

Device intended use​

The device is a computational software-only medical device (Class IIb, Rule 11, MDR 2017/745) that processes images of the skin and its structures using computer vision algorithms. It provides two categories of clinical output:

  1. Diagnosis support -- The device generates an interpretative probability distribution across International Classification of Diseases (ICD-11) categories that may be represented in the image. This output represents the relative likelihood of each visible condition, enabling the healthcare professional to consider a ranked differential diagnosis.

  2. Objective severity assessment -- The device quantifies the intensity, count, and extent of clinical signs (e.g. erythema, desquamation, induration, papules, pustules, and others), providing a reproducible, objective measurement of disease involvement.

Critical clarification on diagnostic output​

The device does not function as a binary positive/negative diagnostic test. It does not confirm or rule out a single condition. Instead, for every image processed, the device emits the complete probability distribution of all visible ICD-11 categories, each with an associated confidence level. This is fundamentally different from a classical diagnostic test, and it has important implications for the interpretation of performance metrics:

  • Accuracy is measured as the proportion of cases in which the correct condition appears as the highest-ranked category (Top-1 accuracy) or within the top N categories (Top-N accuracy).
  • The device always provides information -- it does not produce "negative" results in the traditional sense. Every output is an ordered list of possible conditions with associated probabilities.

This design ensures that the healthcare professional retains full clinical decision-making authority while benefiting from a comprehensive computational analysis of the image.

Clinical benefits​

The device's clinical evidence programme is designed to demonstrate three intended clinical benefits:

Benefit 7GH -- Diagnostic accuracy improvement​

The device improves the accuracy of healthcare professionals in the diagnosis of dermatological conditions across a broad spectrum of clinical presentations, including rare diseases and lesions suspicious for skin cancer. This has a positive impact on patient management and health outcomes related to diagnosis, enabling more appropriate clinical decision-making, earlier identification of rare conditions, and, in cases of suspected malignancy, reducing the risk of delayed diagnosis and the need for unnecessary invasive procedures.

Supporting studies: MC_EVCDAO_2019 (melanoma detection), BI_2024 (rare diseases and multiple conditions), PH_2024 (primary care diagnosis), SAN_2024 (multiple conditions), MAN_2025 (dark phototypes, ongoing).

Benefit 5RB -- Objective severity assessment​

The device measures the degree of involvement of disease objectively, quantitatively, and reproducibly. This increases the precision of healthcare providers during the monitoring of patients. This has a positive impact on patient management and outcomes related to the monitoring of patients and treatment.

Supporting studies: COVIDX_EVCDAO_2022 (remote severity monitoring of chronic pathologies), AIHS4_2025 (hidradenitis suppurativa severity scoring).

Benefit 3KX -- Care pathway management​

The device improves the precision of healthcare professionals in managing dermatological care pathways, encompassing referral decisions, resource allocation, and clinical assessment in remote care settings. This has a positive impact on patient management and outcomes related to the diagnosis and monitoring of patients, resulting in reduced waiting times for specialist consultation, improved adequacy of referrals, and expanded access to dermatological assessment across in-person and remote care settings.

Supporting studies: DAO_Derivation_O_2022 (referral optimisation, Osakidetza), DAO_Derivacion_PH_2022 (referral optimisation, Puerta de Hierro), IDEI_2023 (clinical flow optimisation), SAN_2024 (referral decision-making), PH_2024 (primary care referral decision-making).

Aggregate sample demographics​

The tables below aggregate demographic data across all completed studies where each data category is available. Not all studies report all demographic breakdowns; footnotes indicate which studies contribute to each aggregation.

Sex distribution​

SexCountPercentage
Male31341.5%
Female44258.5%
Total755100.0%

Based on 7 studies: MC_EVCDAO_2019, COVIDX_EVCDAO_2022, DAO_Derivation_O_2022, IDEI_2023, BI_2024, PH_2024, SAN_2024. Not available for DAO_Derivacion_PH_2022 (131 subjects), AIHS4_2025 (2 subjects), or MAN_2025 (ongoing).

Age group distribution​

Age groupCountPercentage
Newborn (birth to 1 month)10.2%
1 month to 2 years50.8%
2 to 12 years152.5%
12 to 21 years325.4%
22 to <65 years34658.2%
≥65 years19632.9%
Total595100.0%

Based on 6 studies: MC_EVCDAO_2019, DAO_Derivation_O_2022, IDEI_2023, BI_2024, PH_2024, SAN_2024. Not available for COVIDX_EVCDAO_2022 (160 subjects), DAO_Derivacion_PH_2022 (131 subjects), AIHS4_2025 (2 subjects), or MAN_2025 (ongoing).

Fitzpatrick phototype distribution​

PhototypeCountPercentage
I49055.4%
II26329.7%
III10211.5%
IV212.4%
V91.0%
VI00.0%
Total885100.0%

Based on 8 studies: MC_EVCDAO_2019, COVIDX_EVCDAO_2022, DAO_Derivation_O_2022, DAO_Derivacion_PH_2022, IDEI_2023, BI_2024, PH_2024, SAN_2024. Not available for AIHS4_2025 (2 subjects). MAN_2025 (ongoing) specifically targets Fitzpatrick V--VI phototypes and will add 149 images to the darker phototype representation upon completion.

Note on phototype representation

The predominance of Fitzpatrick phototypes I--II reflects the geographic context of the clinical programme (Southern European populations). The ongoing MAN_2025 study was specifically designed to address this gap by evaluating the device exclusively on Fitzpatrick V--VI skin presentations.

Individual study demographics​

MC_EVCDAO_2019 -- Melanoma detection​

Full title: Clinical validation study of a CAD system with artificial intelligence algorithms for early noninvasive in vivo cutaneous melanoma detection

  • Design: Prospective, observational, multi-centre study
  • Sites: Hospital Universitario Cruces, Hospital Universitario Basurto (Osakidetza, Basque Country)
  • Principal investigators: Dr. Jesus Gardeazabal Garcia, Dr. Rosa Ma Izu Belloso
  • Sample: 105 patients
  • Duration: February 2020 -- November 2023
  • Primary objective: To assess the diagnostic accuracy of the device in differentiating between benign and malignant melanocytic skin lesions
SexCountPercentage
Male5350.5%
Female5249.5%
Age groupCountPercentage
Newborn (birth to 1 month)00.0%
1 month to 2 years00.0%
2 to 12 years00.0%
12 to 21 years21.9%
22 to <65 years5653.3%
≥65 years4744.8%
Fitzpatrick phototypeCountPercentage
I9187.1%
II109.8%
III32.5%
IV10.6%
V00.0%
VI00.0%

The high proportion of phototype I reflects the melanoma-focused study population, as cutaneous melanoma incidence is highest in fair-skinned individuals.


COVIDX_EVCDAO_2022 -- Severity monitoring​

Full title: Clinical Validation of a Computer-Aided Diagnosis (CAD) System Utilizing Artificial Intelligence Algorithms for Continuous and Remote Monitoring of Patient Condition Severity in an Objective and Stable Manner

  • Design: Observational, prospective, single-centre study
  • Site: Torrejón University Hospital
  • Principal investigator: Dra. Marta Andreu
  • Sample: 160 patients (screened from 400 candidates)
  • Duration: April 2022 -- October 2023 (18 months)
  • Primary objective: To validate the use of AI algorithms for continuous and remote monitoring of patient condition severity
SexCountPercentage
Male6339.6%
Female9760.4%

Age group distribution: Not available. The study did not place specific emphasis on age as a primary factor of investigation.

Fitzpatrick phototypeCountPercentage
I7949.3%
II6238.5%
III1710.9%
IV21.3%
V00.0%
VI00.0%

Pathology distribution: 47 different dermatological conditions were represented. The most frequent were acne (67 cases), haemangioma (14), hidradenitis suppurativa (8), actinic keratosis (5), rosacea (4), nevus (4), contact dermatitis (4), and psoriasis (4).


DAO_Derivation_O_2022 -- Referral optimisation (Osakidetza)​

Full title: Pilot study for the clinical validation of an artificial intelligence algorithm to optimize the appropriateness of dermatology referrals

  • Design: Multi-centre, prospective, observational study
  • Sites: Health Centre Sodupe-Güeñes, Health Centre Balmaseda, Health Centre Buruaga, Health Centre Zurbaran (Osakidetza, Basque Country)
  • Principal investigators: Dr. Jesus Gardeazabal Garcia, Dr. Rosa Ma Izu Belloso
  • Sample: 127 patients enrolled, 117 analysed (10 excluded)
  • Duration: November 2022 -- April 2025
  • Primary objective: To clinically validate an AI algorithm designed to optimize the appropriateness and efficiency of dermatology referrals
SexCountPercentage
Male4636.2%
Female8163.8%
Age groupCountPercentage
12 to 21 years32.4%
22 to <65 years5845.7%
≥65 years6652.0%

No patients under 12 years of age were enrolled.

Fitzpatrick phototypeCountPercentage
I8667.7%
II2922.9%
III97.5%
IV21.5%
V10.5%
VI00.0%

Pathology distribution: 48 different conditions. The most frequent were seborrheic keratosis (33, 17.9%), actinic keratosis (27, 14.7%), melanocytic nevus (24, 13.0%), psoriasis (9, 4.9%), and basal cell carcinoma (8, 4.4%).


DAO_Derivacion_PH_2022 -- Referral optimisation (Puerta de Hierro)​

Full title: Project to enhance Dermatology E-Consultations in Primary Care Centres using Artificial Intelligence Tools

  • Design: Prospective, analytical, observational case series, multi-centre
  • Sites: Pozuelo and Majadahonda Health Centers, Puerta del Hierro Majadahonda University Hospital
  • Principal investigator: Dr. Gastón Roustan Gullón
  • Sample: 131 patients, 180 diagnostic reports
  • Duration: June 2022 -- January 2024
  • Primary objective: To enhance dermatology e-consultations in primary care using AI-driven referral optimisation tools

Physician participants: 15 primary care practitioners and 2 dermatologists completed questionnaires on the device's utility.

Sex distribution: Not available in the study report.

Age group distribution: Not available in the study report.

Fitzpatrick phototypeCountPercentage
I6348.3%
II4836.7%
III1612.2%
IV32.2%
V10.6%
VI00.0%

IDEI_2023 -- Clinical flow optimisation​

Full title: Optimisation of clinical flow in patients with dermatological conditions

  • Design: Observational, non-interventional, single-centre study
  • Site: Instituto de Dermatología Integral (IDEI), Madrid
  • Principal investigator: Dr. Miguel Sánchez Viera
  • Sample: 204 subjects (76 retrospective pigmented lesions, 32 prospective pigmented lesions, 62 retrospective alopecia, 34 prospective alopecia)
  • Duration: January 2024 -- August 2024 (7 months)
  • Primary objective: To optimise clinical workflow for patients with dermatological conditions through AI integration
  • Evaluators: Board-certified dermatologists with minimum 10 years of experience
SexCountPercentage
Male5627.5%
Female14872.5%
Age groupCountPercentage
12 to 21 years31.5%
22 to <65 years14169.1%
≥65 years6029.4%

No patients under 12 years of age were enrolled.

Fitzpatrick phototypeCountPercentage
I12963.4%
II4723.2%
III2612.5%
IV20.9%
V00.0%
VI00.0%

BI_2024 -- GPP and skin conditions (MRMC)​

Full title: Multi-Reader Multi-Case Study for Assessing the Impact of the Device on the Clinical Assessment of Generalised Pustular Psoriasis and Other Skin Conditions by Healthcare Professionals

  • Design: Multi-reader multi-case (MRMC), prospective, self-controlled
  • Conducted remotely: Images distributed to participating healthcare professionals
  • Coordinating investigator: Dr. Antonio Martorell Calatayud
  • Sample: 100 anonymised images (101 with demographics)
  • Duration: June 2024 -- September 2024
  • Primary objective: To improve the diagnostic accuracy for generalised pustular psoriasis and other skin conditions

Reader panel: 15 healthcare professionals (11 primary care physicians, 4 dermatologists). 9 completed the full 100-image protocol; 6 completed partial reviews. Total evaluations: 1,449 out of 1,500 planned (96.6%).

SexCountPercentage
Male6463.4%
Female3736.6%
Age groupCountPercentage
Newborn (birth to 1 month)00.0%
1 month to 2 years33.0%
2 to 12 years1413.9%
12 to 21 years2019.8%
22 to <65 years5251.5%
≥65 years1211.9%
Fitzpatrick phototypeCountPercentage
I2020.0%
II4343.0%
III2222.0%
IV99.0%
V66.0%
VI00.0%

Pathology distribution (15 conditions):

ConditionCount
Generalised pustular psoriasis10
Eczematous dermatitis13
Acute generalised exanthematous pustulosis (AGEP)1
Acne5
Acne conglobata3
Severe inflammatory acne3
Seborrheic keratosis8
Seborrheic dermatitis6
Palmoplantar pustulosis3
Plaque psoriasis5
Pemphigus vulgaris5
Impetigo10
Hidradenitis suppurativa5
Subcorneal pustular dermatosis2
Tinea corporis2

PH_2024 -- Primary care diagnosis (MRMC)​

Full title: Multi-Reader Multi-Case Study Assessing the Impact of the Device on the Diagnostic Accuracy and Referral Decision-Making of Primary Care Physicians for Skin Lesions

  • Design: Multi-reader multi-case (MRMC), prospective, self-controlled
  • Conducted remotely: Images distributed to participating primary care physicians
  • Coordinating investigator: Dr. Gastón Roustán Gullon
  • Sample: 30 anonymised images
  • Duration: June 2024 -- September 2024
  • Primary objective: To evaluate and improve the diagnostic process for skin pathologies in primary care settings

Reader panel: 9 primary care physicians. All completed the full image assessment.

SexCountPercentage
Male1446.7%
Female1653.3%
Age groupCountPercentage
Newborn (birth to 1 month)13.3%
1 month to 2 years13.3%
2 to 12 years00.0%
12 to 21 years00.0%
22 to <65 years2170.0%
≥65 years723.3%
Fitzpatrick phototypeCountPercentage
I1033.3%
II1240.0%
III723.3%
IV13.3%
V00.0%
VI00.0%

Pathology distribution (9 conditions):

ConditionCount
Atypical melanocytic nevus2
Melanocytic nevus3
Melanoma5
Basal cell carcinoma3
Urticaria5
Pustular psoriasis2
Actinic keratosis2
Plaque psoriasis3
Hidradenitis suppurativa5

SAN_2024 -- Multi-condition assessment (MRMC)​

Full title: Multi-Reader Multi-Case Study for Evaluating the Impact of the Device on the Healthcare Practitioners' Assessment of Skin Lesions

  • Design: Multi-reader multi-case (MRMC), prospective
  • Conducted remotely: Images distributed via a web-based platform
  • Coordinating investigator: Dr. Antonio Martorell Calatayud
  • Sample: 29 anonymised images
  • Duration: June 2024 -- October 2024
  • Primary objective: To validate that the information provided by the device increases the true accuracy of healthcare professionals in the diagnosis of multiple dermatological conditions

Reader panel: 16 healthcare professionals (10 primary care practitioners, 6 dermatologists). 12 completed the full 29-image assessment; 4 completed partial reviews. Total evaluations: 401 out of 464 planned (86.4%).

SexCountPercentage
Male1760.7%
Female1139.3%
Age groupCountPercentage
Newborn (birth to 1 month)00.0%
1 month to 2 years13.6%
2 to 12 years13.6%
12 to 21 years414.3%
22 to <65 years1864.3%
≥65 years414.3%
Fitzpatrick phototypeCountPercentage
I1242.8%
II1242.8%
III27.2%
IV13.6%
V13.6%
VI00.0%

Pathology distribution (13 conditions):

ConditionCount
Dermatitis5
Melanoma3
Alopecia2
Urticaria1
Granuloma annulare1
Seborrheic keratosis1
Herpes2
Dermatophytosis (tiña)2
Psoriasis3
Onychomycosis2
Acne2
Pressure ulcer1
Nevus4

AIHS4_2025 -- Hidradenitis suppurativa (retrospective)​

Full title: Evaluation of AIHS4 Performance in the M-27134-01 Clinical Trial for Hidradenitis Suppurativa

  • Design: Retrospective, observational, longitudinal study
  • Coordinating investigator: Dr. Antonio Martorell Calatayud
  • Sample: 2 subjects with confirmed hidradenitis suppurativa, evaluated at 4 time points (Day 1, 15, 29, 43)
  • Duration: June 2024 -- July 2024
  • Primary objective: To evaluate the performance and reliability of the AIHS4 scoring system within the context of a clinical trial for hidradenitis suppurativa
  • Expert panel: 2 dermatologists (Dr. Antonio Martorell Calatayud, Dr. Gema Ochando)

Demographic breakdowns: Not available due to the minimal sample size (N=2). Both subjects had confirmed hidradenitis suppurativa.


MAN_2025 -- Fitzpatrick V--VI validation (ongoing)​

Full title: Multi-Reader Multi-Case Study for Evaluating the Diagnostic Performance of Healthcare Professionals Assisted by the Device on Fitzpatrick Phototype V--VI Skin Presentations

  • Design: Multi-reader multi-case (MRMC), prospective, self-controlled
  • Conducted remotely: Through a centralised web-based platform
  • Coordinating investigator: Dr. Antonio Martorell Calatayud
  • Planned sample: 149 anonymised images, all exclusively Fitzpatrick phototype V--VI
  • Start date: April 2026
  • Status: Ongoing -- awaiting completion of reader evaluations
Ongoing study

This study has not yet been completed. The data below describes the planned sample composition based on the Clinical Investigation Plan. No results are available at this time.

Reader panel: Minimum of 5 dermatologists (recruitment ongoing).

Study rationale: This study was specifically designed to evaluate the device's diagnostic performance on darker skin tones (Fitzpatrick V--VI), addressing a recognised gap in the existing evidence base where the majority of samples are phototype I--III.

Planned phototype distribution: All 149 images are Fitzpatrick phototype V or VI (exact split between V and VI not specified in the protocol).

Planned pathology distribution (28 conditions, 149 images):

ConditionCount
Generalised eczematous dermatitis13
Hidradenitis suppurativa11
Nodular acne10
Impetigo9
Generalised pustular psoriasis8
Dermatophytosis8
Plaque psoriasis7
Acne (unspecified)7
Melanocytic naevus7
Melanoma7
Acute generalised exanthematous pustulosis (AGEP)6
Basal cell carcinoma6
Pemphigus vulgaris5
Peeling skin disorder4
Dyshidrosis3
Bullous drug eruption3
Erythroderma3
Pityriasis rubra pilaris3
Rosacea3
Porphyria cutanea tarda2
Urticaria2
Seborrheic keratosis2
Folliculitis2
Tinea corporis2
Acanthosis nigricans1
Mycosis fungoides1
Secondary syphilis1
Necrobiosis lipoidica1

Signature meaning

The signatures for the approval process of this document can be found in the verified commits at the repository for the QMS. As a reference, the team members who are expected to participate in this document and their roles in the approval process, as defined in Annex I Responsibility Matrix of the GP-001, are:

  • Author: Team members involved
  • Reviewer: JD-003 Design & Development Manager, JD-004 Quality Manager & PRRC
  • Approver: JD-001 General Manager
Previous
R-TF-015-011 State of the Art Legit.Health Plus
Next
Investigation
  • Introduction
    • Scope
  • Device intended use
    • Critical clarification on diagnostic output
  • Clinical benefits
    • Benefit 7GH -- Diagnostic accuracy improvement
    • Benefit 5RB -- Objective severity assessment
    • Benefit 3KX -- Care pathway management
  • Aggregate sample demographics
    • Sex distribution
    • Age group distribution
    • Fitzpatrick phototype distribution
  • Individual study demographics
    • MC_EVCDAO_2019 -- Melanoma detection
    • COVIDX_EVCDAO_2022 -- Severity monitoring
    • DAO_Derivation_O_2022 -- Referral optimisation (Osakidetza)
    • DAO_Derivacion_PH_2022 -- Referral optimisation (Puerta de Hierro)
    • IDEI_2023 -- Clinical flow optimisation
    • BI_2024 -- GPP and skin conditions (MRMC)
    • PH_2024 -- Primary care diagnosis (MRMC)
    • SAN_2024 -- Multi-condition assessment (MRMC)
    • AIHS4_2025 -- Hidradenitis suppurativa (retrospective)
    • MAN_2025 -- Fitzpatrick V--VI validation (ongoing)
All the information contained in this QMS is confidential. The recipient agrees not to transmit or reproduce the information, neither by himself nor by third parties, through whichever means, without obtaining the prior written permission of Legit.Health (AI Labs Group S.L.)