Neoplasms

All Conditions

Skin and Connective Tissue Diseases

Carcinoma

Carcinoma, Squamous Cell

Melanoma

Carcinoma, Basal Cell

Skin Neoplasms

Melanoma, Cutaneous Malignant

Neoplasms, Glandular and Epithelial

Neoplasms by Histologic Type

Neoplasms, Basal Cell

Deep Ensemble for the Recognition of Malignancy (DERM)

This study aims establish the effectiveness of Image Analysing Algorithm (DERM) to identify melanoma, Squamous Cell Carcinoma (SCC) and Basal Cell Carcinoma (BCC) when used to analyse dermoscopic images of skin lesions within the US and European population.

DERM US and EU Validation Study

A Clinical Validation Study to Demonstrate the Effectiveness of an Artificial Intelligence Algorithm (DERM) to Identify Skin Cancer in Patients Undergoing a Skin Biopsy

AUROC, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) etc. of DERM to detect melanoma, SCC, BCC, premalignant, and benign conditions grouped and individually

Probability that the alternative classification label DERM that returns, matches the lesion diagnosis

Such as sex, age, and Fitzpatrick skin type

Such as location, size, growth, stage and sub-type

AUROC of DERM to identify malignant conditions for each individual camera / lens type

Concordance of DERM results by each individual camera / lens type

Strength of association between correct classification and acceptance/rejection status of images

AUROC of DERM when macro images are used both to train the algorithm and as test images

Correlation between clinician assessment of likelihood of skin cancer with histopathology diagnosis

Percentage of images taken that are rejected by the IQ check (MoleNotMole + Image Quality), where a second image is successfully taken

Sensitivity of DERM to detect Melanoma, SCC and BCC combined

Specificity of DERM to detect Melanoma, SCC and BCC combined

Sensitivity of DERM to correctly classify SCC

Specificity of DERM to correctly classify SCC

Sensitivity of DERM to correctly classify BCC

Specificity of DERM to correctly classify BCC

Skin Analytics Limited

Skin Diseases

Recruited participants will be attending a dermatology clinic with at least one skin lesion where there is a suspicion of skin cancer. All suspicious lesions suitable for photographing will be photographed six times in a single visit. A macro and dermoscopic image of each lesion will be captured by three different mobile phones: an iPhone, a Samsung and a Nokia smart phone, without (macro image) or with (dermoscopic image) a Dermlite DL1 lens attached. Dermoscopic images of healthy skin will also be captured by each camera. Images of the lesions will be analysed by DERM. The DERM results for lesions biopsied will be compared to the biopsy result; the DERM results for lesions not biopsied will be compared to the clinical assessment.

This study aims to establish the effectiveness of an Artificial Intelligence (AI) algorithm (DERM) to determine the presence of Basal Cell Carcinoma (BCC) and Squamous Cell Carcinoma (SCC) and frequently observed benign conditions, when used to analyse images of skin lesions taken by commonly available smart phone cameras.

DERM, an Artificial Intelligence (AI)-based diagnosis support tool, has been shown to be able to accurately identify Non-melanoma skin cancers (NMSC) and other conditions from historical images of suspicious skin lesions (moles). This study aims to establish how well DERM determines the presence of these conditions in images of skin lesions collected in a clinical setting.

Suspicious skin lesions that are due to be assessed by a dermatologist and a patch of healthy skin will be photographed using three commonly available smart phone cameras with a specific lens attachment. The images will be analysed by DERM, and the results compared to the clinician's diagnosis (all lesions) and histologically-conformed diagnosis (any lesion that is biopsied).

DERM NMSC Validation Study

Effectiveness of an Image Analysing Algorithm (DERM) to Diagnose Non-melanoma Skin Cancer (NMSC) and Benign Skin Lesions Compared to Gold Standard Clinical and Histological Diagnosis

Research to improve or test the performance of DERM only allowed in consent

The impact of patient characteristics (such as sex, age, location of lesion, total body lesion count, Fitzpatrick skin type, past medical history of skin cancer) on the diagnostic accuracy of DERM and clinician assessment;

The impact of lesions characteristic (such as growth over last 6 months, stage and sub-type) on the diagnostic accuracy of DERM and clinician assessment

The impact of image variables (such as macro and dermoscopic images) on the diagnostic accuracy of DERM assessment

Exploration of whether macro images can be used as part of DERM's assessment

Area Under the Receiver Operating Characteristic Curve (AUROC) of the DERM result of biopsied lesions, using histopathological-confirmed diagnosis as gold standard

Area Under the Receiver Operating Characteristic Curve (AUROC) of the DERM result of biopsied lesions, using clinical diagnosis as gold standard

The sensitivity of DERM when used to assess biopsied lesions

The specificity of DERM when used to assess biopsied lesions

The false positive rate of DERM when used to assess biopsied lesions

The false negative rate of DERM when used to assess biopsied lesions

The positive predictive value of DERM when used to assess biopsied lesions

The negative predictive value of DERM when used to assess biopsied lesions

The sensitivity of DERM when used to assess non-biopsied lesions

The specificity of DERM when used to assess non-biopsied lesions

The false positive rate of DERM when used to assess non-biopsied lesions

The false negative rate of DERM when used to assess non-biopsied lesions

The positive predictive value of DERM when used to assess non-biopsied lesions

The negative predictive value of DERM when used to assess non-biopsied lesions

Concordance of clinician assessment with histologically confirmed diagnosis

The concordance of DERM result generated using images from each camera

The proportion of skin lesions with 3 images that can be analysed by DERM;

The proportion of skin lesions with at least 1 readable image that can be analysed by DERM

Innovate UK

Patients attending a Medical Photography facility with at least 1 suspicious skin lesion will be approached to participate in the study. Participants will have an additional macro and dermoscopic image of each suspicious skin lesions suitable for photography. Photographs will be taken by a healthcare professional using an iPhone XR smart phone camera with a DL1 dermoscopic lens attachment. The images will be encrypted and electronically transmitted to Skin Analytics' cloud servers for analysis by DERM. The suspected diagnosis determined by DERM will be compared with dermatologist review and histologically confirmed diagnosis, where obtained. Healthcare resource utilization information and patient satisfaction data will also be collected

This study aims to provide an initial assessment of the potential impact DERM could have on the number of onward referrals for a face to face dermatologist review and/or biopsy from a teledermatology-based service, and to improve the understanding of the patient pathways that exist.

DERM, an Artificial Intelligence (AI)-based diagnosis support tool, has been shown to be able to accurately identify melanoma, non-melanoma skin cancers (NMSC) and other conditions from historical images of suspicious skin lesions (moles).

This study aims to establish whether the use of DERM in the patient pathway could reduce the number of unnecessary referrals to dermatologist review and/or biopsy.

Suspicious skin lesions that are due to be photographed for a dermatologist to review, will have two additional photographs taken using a commonly available smart phone camera with and without a specific lens attachment. The images will be analysed by DERM, and the results compared to the clinician's diagnosis (all lesions) and histologically-confirmed diagnosis (any lesion that is biopsied).

DERM Health Economics Study

Impact of an Artificial Intelligence Platform (DERM) on the Healthcare Resource Utilisation (HRU) Needed to Diagnose Skin Cancer When Used as Part of a United Kingdom-based Teledermatology Service

The rate of unnecessary referrals for a face to face dermatologist review for the same detection rate between standard of care and DERM of lesions reviewed by teledermatology or DERM

Sensitivity of DERM on biopsied lesions, using histopathological confirmed diagnosis as gold-standard

Specificity of DERM on biopsied lesions, using histopathological confirmed diagnosis as gold-standard

False positive rate of DERM on biopsied lesions, using histopathological confirmed diagnosis as gold-standard

False negative rate of DERM on biopsied lesions, using histopathological confirmed diagnosis as gold-standard

Positive predictive value of DERM on biopsied lesions, using histopathological confirmed diagnosis as gold-standard

Number needed to biopsy by DERM on biopsied lesions, using histopathological confirmed diagnosis as gold-standard

Sensitivity of teledermatologists on biopsied lesions, using histopathological confirmed diagnosis as gold-standard

Specificity of teledermatologists on biopsied lesions, using histopathological confirmed diagnosis as gold-standard

False positive rate of teledermatologists on biopsied lesions, using histopathological confirmed diagnosis as gold-standard

False negative rate of teledermatologists on biopsied lesions, using histopathological confirmed diagnosis as gold-standard

Positive predictive value of teledermatologists on biopsied lesions, using histopathological confirmed diagnosis as gold-standard

Negative predictive value of teledermatologists on biopsied lesions, using histopathological confirmed diagnosis as gold-standard

Number needed to biopsy by teledermatologists on biopsied lesions, using histopathological confirmed diagnosis as gold-standard

Sensitivity of DERM to identify benign conditions, using clinical diagnosis as gold-standard

Specificity of DERM to identify benign conditions, using clinical diagnosis as gold-standard

False positive of DERM to identify benign conditions, using clinical diagnosis as gold-standard

False negative rate of DERM to identify benign conditions, using clinical diagnosis as gold-standard

Positive predictive of DERM to identify benign conditions, using clinical diagnosis as gold-standard

Negative predictive value of DERM to identify benign conditions, using clinical diagnosis as gold-standard

Number needed to refer by DERM to identify benign conditions, using clinical diagnosis as gold-standard

Concordance of DERM result with clinical diagnosis

Percentage of patients referred to teledermatology through 2-week wait referral, general referral, direct to teledermatology, routine follow-up (etc) referral routes

Time taken (days) from GP referral to either histopathology-confirmed or clinical diagnosis

The cost of the number of referrals for face to face dermatologist review and/or biopsy that would have been saved / charged if DERM had been used to decide whether to refer the patient onwards

Proportion of images submitted to DERM that cannot be analysed

Patient feedback on their experience of the service. Patients will rate whether they agree, or don't agree, with statements that assess their acceptance of having a computer involved in their diagnosis pathway

Study title	NCT ID	Conditions	Interventions	Eligibility	Organization
DERM US and EU Validation Study	NCT05126173	Malignant Skin ... Basal Cell Carc... Squamous Cell C...	Deep Ensemble f...	18 Years -	Skin Analytics Limited
DERM NMSC Validation Study	NCT04116983	Non-melanoma Sk...	Deep Ensemble f...	18 Years -	Skin Analytics Limited
DERM Health Economics Study	NCT04123678	Melanoma Non-melanoma Sk...	Deep Ensemble f...	18 Years -	Skin Analytics Limited

Spots Global Cancer Trial Database for deep ensemble for the recognition of malignancy (derm)