Study author Other target conditions also assessed	Study type Country Setting Participants/lesions	Inclusion criteria	Index tests (algorithm) Diagnostic approach	Threshold		Observer qualifications (number) Experience	Reference standard Final diagnoses Prevalence (invasive melanoma)	Exclusions Comments
In‐person evaluations
Ascierto 2010	WPC P‐CS Specialist clinic Italy 54/54	Clinically relevant cutaneous pigmented lesions, undergoing dermoscopy and excision	Dermoscopy (risk stratification; modified Kenet 2001) In‐person (Also evaluates CAD Spectroscopy)	Very high risk; high or very high risk; correct diagnosis of MM		Dermatologist (n = NR; exp High) Unclear observer interpretation	Histology MM 12 'Benign' 42 12/42; 22%	‐
Coras 2003	WPC NR‐CS Private Germany NR/45	PSLs undergoing excision due to diagnosis of melanoma or atypical nevus, to rule out melanoma or at the patient's request	Dermoscopy (no details; diagnosis based on clinical exam, dermoscopy, medical history ) In‐person (Also evaluates teledermatology assessment of clinical/dermoscopic images)	NR; Correct diagnosis of MM		Dermatologist (n = 3; exp High) participating experts with great experience in dermatoscopy Single observer	Histology MM 16; 'Benign': 29 16/45; 36%	10 excluded due to poor image quality; 45 did not undergo excision
Feldmann 1998 Invasive melanoma or atypical intraepidermal melanocytic variants	NC P‐CS Secondary Austria NR/500	Melanocytic lesions examined by dermatoscopy prior to excision	Dermoscopy (ABCD) In‐person	> 5.45; > 4.2		NR (n = NR; exp NR) Unclear observer interpretation	Histology MM 25; MiS 5 BN 272; dysplasia 190; lentigines 7; lentigo naevi 1 30/500; 6%	NR
Krahn 1998	WPC P‐CS Secondary Germany 80/80	Excised PSLs	1. VI (no algorithm) 2. Dermoscopy (no algorithm) In‐person	NR; clinical diagnosis of MM		Dermatologist (assumed) (n = 1; exp NR) Single observer	Histology MM 39 BN 37; dysplastic 2; SN 1 39/80; 49%	None
Piccolo 2000	NC NR‐CS Multicentre Austria 40/43	PSLs selected because of their diagnostic difficulty	Dermoscopy (no algorithm) In‐person observer (Also evaluates teledermatology assessment of clin/dermoscopic images)	NR; correct diagnosis of MM		Dermatologists (n = 1; exp High) Single observer	Histology MM 11; BCC 3 SK 2; BN 23; other 4 11/43; 26%	NR; poor‐quality index test image. The digital images were assigned an image‐quality rating (1, excellent; 2, good; 3, sufficient; 4, poor). All images scoring 4 were excluded from the study
Viglizzo 2004	WPC NR‐CS Specialist clinic Italy NR/79	PSLs examined at the Dermoscopy Service and undergoing excisions; high and medium risk on dermoscopy were selected for excision and 2x2 can be estimated only for melanocytic subgroup	1. VI (no algorithm) 2. Dermoscopy (no algorithm) In‐person	NR; correct diagnosis of MM		Dermatologist (assumed) (n = NR; exp NR) Single observer	Histology MM 12 MN: 67 12/67; 18%	None
Image‐based
Arevalo 2008	NC RP‐CS Specialist clinic Australia NR/3367	Melanocytic lesions imaged at the Sydney Melanoma Unit with a histopathologic diagnosis or that remained unchanged following short‐term (2.5‐4.5 months) digital monitoring (diagnosed as benign)	Dermoscopy (Menzies criteria) Image‐based (blinded)		Absence of negative characteristics + ≥ 1 positive characteristic present; correct diagnosis of MM	Dermatologist (assumed) (n = 2; exp NR) Consensus of 2; referral to a 3rd observer if disagreement	Histology or FU MM 341 'Benign' 3026 341/3367; 10%	None
Friedman 2008 Invasive melanoma or atypical intraepidermal melanocytic variants	WPC CCS Secondary/private US 94/99	An industry database of images of PSL ≤ 6 mm was used to sample images of melanoma and non‐melanoma lesions; high‐grade dysplastic naevi were excluded	Dermoscopy (no algorithm) Image‐based (site, age, gender) (Also evaluates CAD‐Spectroscopy)		Correct diagnosis; excise decision	Mixed ‐ secondary (n = 10; exp High) Average result (reports mean and median; mean used)	Histology MM 21; MiS 28; BCC: BN 34; SK 2; 14 other benign 21/99; 21%	NR
Hauschild 2014 Invasive melanoma; Invasive melanoma or atypical intraepidermal melanocytic variants	WPC CCS Secondary/private US 130/130	Subset of PSL evaluated in a MelaFind study (Monheit 2011); 65 melanoma and 65 non‐melanoma randomly selected. Excluded ulcerated, non‐pigmented, or located on excluded anatomic sites.	Dermoscopy (no algorithm) (Also evaluates CAD spectroscopy) Image‐based (clinical image, pt history)		NR; excise decision	Dermatologist (n = 101; exp High) Single observer	Histology MM 36; MiS 29 'Benign' diagnoses: 65 36/130; 28%	‐
Kreusch 1992	NC RP‐CS Secondary Germany Full sample: 858/1506 (265 melanocytic included)	Pigmented lesions suspected to be malignant melanoma with adequate photo‐documentation and histology results	Dermoscopy (from Kreusch 1991) Image‐based (slides labelled only with patient code and lesion localisation)		≥ 9; correct diagnosis of MM	Dermatologist (assumed) (n = 1; ‘experienced') (Also presents results for inexperienced student – data not included) Single observer	Histology MM 96; BN 169	52 NML excluded from second‐step evaluation
Lorentzen 1999a	WPC P‐CS Specialist clinic Denmark 232/232	Patients with lesions suspicious for CMM referred to outpatients clinic	1. VI (no algorithm) 2. Dermoscopy (no algorithm) Image‐based (clinical image)		Subjective impression; correct diagnosis of MM	Dermatologist (n = 4; exp High) Average	Histology MM 49; BCC 16 SK 12; BN 137 other: 18 (SN, BD + others) 49/232; 21%	Poor‐quality index test image 10 cases excluded
Lorentzen 2000	WPC‐alg RP‐CS Specialist clinic Denmark 258/258	PSL from patients consecutively referred to the skin cancer outpatient clinic with available clinical photographs, dermatophotographs and a subsequent excision biopsy were included	Dermoscopy (ABCD; Kenet risk stratification) Image‐based (clinical image)		> 4.75; Kenet – probable melanoma; possible/probable melanoma	Dermatologist (n = 3; exp High; 3 senior dermatologists with > 5 years' daily experience in dermatoscopy Single observer (reported per observer)	Histology MM 64; BCC 25 SK 14; BN 135; dysplastic 3; other: 16 64/258; 25%	‐
Lorentzen 2008	WPC NR‐CS Specialist clinic Denmark 119/119	Patients referred to the specialist naevus clinic; compared classic dermoscopy to acrylic globe magnifier	Dermoscopy (Kenet risk stratification) Image‐based (blinded)		NR	Dermatologist (n = NR) Average	Histology MM 24; BCC 13 BN 69; mild/moderate dysplasia 2; SK 9; other 2 24/119; 20%	1 DF
Menzies 1996	NC RP‐Unclear Image libraries Multicentre NR/385	PSL from the Sydney Melanoma Unit with dermoscopic images and histological diagnoses; melanomas and randomly selected clinically atypical non‐melanoma lesions were included	Dermoscopy (Menzies criteria) Image‐based (blinded)		2 characteristics absent and ≥ 1 characteristic present; correct diagnosis of MM	Dermatologist (assumed) (n = NR; exp NR) NR	Histology MM 107; BCC: 18 SK 23; acquired BN 58; dysplastic 105; blue naevi 11; ephelis/lentigo 17; SN 6; spindle cell nevus 2; DF 2; hemangioma 13; solar keratosis 9; other 14 107/385; 28%	‐
Menzies 2013	WPC‐algorithms CCS Secondary Mixed NR/467	Nodular malignant melanoma* and a random selection of non‐nodular invasive primary melanoma, benign nodular melanocytic lesions, and nodular nonmelanocytic lesions at a ratio of nodular melanoma to other subgroups of 1:2. Nodular benign melanocytic lesions and nodular nonmelanocytic lesions were identified by the clinical appearance of a solitary nodule and confirmed using dermoscopic examination.	Dermoscopy (ABCD; Menzies, CASH; 7PCL; 3PCL; Menzies (amelanotic)) Image‐based (NR)		ABCD > 5.45; CASH > 8; Menzies amelanotic > 0 and > 1; others at standard thresholds	Dermatologist (n = 1; exp NR). 12 scorers blinded to the lesion diagnosis scored 99 individual features in each lesion of approximately equal sample sizes, as previously described. Following the review of the article for publication, an additional feature (blue‐black structures) was scored for all lesions by one observer (E.C.). Single observer	Histology or FU ("some" benign melanocytic naevi showed no change over time compared with baseline photographs). NM 83; 134 MM BN 115; 217/332; 65%	135 NML excluded from second step evaluation *an invasive melanoma without an in situ (junctional) component beyond 3 rete ridges of the dermal invasive component
Nilles 1994	NC RP‐CS Secondary Germany NR/209	Melanocytic skin lesions that underwent excision	Dermoscopy (new algorithm) Image‐based (blinded)		Any characteristic present?; correct diagnosis of MM	Dermatologist (assumed) (n = 1; exp NR) S ingle observer	Histology MM 41 BN168 41/209; 20%	260 lesions used to identify best model; accuracy for overall diagnosis reported for 209 lesions investigated in year 1990
Rao 1997	WPC‐Obs RP‐CS Specialist clinic US 63/72	Patients with atypical melanocytic lesions or suspected early malignant melanoma	1. VI (no algorithm) 2. Dermoscopy (no algorithm) Image‐based (clinical image)		Subjective impression; diagnosis	Dermatologist (n = 2); melanoma fellow (n = 2) Single observer	Histology MM 21 Atypical MN 51 21/72; 29%	None
Troyanova 2003	WPC‐tests CCS Specialist clinic NR NR/50	Images of PSLs selected for a dermoscopy training study	1. VI (no algorithm) 2. Dermoscopy (no algorithm) Image‐based (blinded)		NR; correct diagnosis of MM	Dermatologist (n = 32; exp High) Average	Histology MM 25 'Benign' 50 25/50; 50%	NR
Westerhoff 2000	WPC‐Obs CCS Specialist clinic Australia NR/100	Clinically atypical PSLs randomly selected from PSL image database	1. VI (no algorithm) 2. Dermoscopy (no algorithm; Menzies criteria) Image‐based (blinded)		NR; diagnosis of MM	GPs (n = 74; no formal training in dermoscopy, randomised to dermoscopy education intervention (n = 37) or not (n = 37) Average reported	Histology or FU MM 50 'Benign' 50 50/100; 50%	*Diagnoses recorded for both groups of GPs at baseline (pre‐test) and after training of one arm (post‐test); post‐test data for the intervention group of GPs was used for the Visual Inspection analysis
AK: actinic keratosis; alg: algorithm; BD: Bowen’s disease; BCC: basal cell carcinoma; BN: benign naevi; BPC: between person comparison (of tests);CAD: computer‐assisted diagnosis; CCS: case‐control study; CS: case series; CMM: cutaneous malignant melanoma;cSCC: cutaneous squamous cell carcinoma; DF: dermatofibroma; FU: follow‐up; LS: lentigo simplex; MiS: melanoma in situ (or lentigo maligna); MM: malignant melanoma; NC: non‐comparative; NR: not reported; Obs: observer; P: prospective; PLC: pigmented lesion clinic; PSL: pigmented skin lesion; R: retrospective; RCM: reflectance confocal microscopy; SK: seborrhoeic keratosis; SN: Spitz naevi; WPC: within person comparison (of tests)