Dermoscopy Basic Algorithms

Dermoscopy basic algorithms are systematic evaluation methodologies developed for the standardized interpretation of dermoscopic (epiluminescence microscopy) images of pigmented and non-pigmented skin lesions. These algorithms provide structured frameworks for identifying specific dermoscopic features, calculating diagnostic scores, and reaching evidence-based decisions about lesion management including reassurance, monitoring, biopsy, or excision. The development of dermoscopic algorithms has revolutionized skin cancer screening by significantly improving diagnostic accuracy compared to naked-eye examination, with sensitivity for melanoma detection increasing from approximately 60 percent (naked eye) to 80-90 percent (dermoscopy) when used by trained clinicians.

History and evolution of dermoscopic algorithms: 1) Pattern analysis (Pehamberger, Steiner, and Wolff, 1987) — first comprehensive method developed at University of Vienna; based on identification of overall pattern, individual structures, and colors; remains gold standard for expert dermoscopists but requires extensive training; 2) ABCD rule of dermoscopy (Stolz, 1994) — modification of clinical ABCD criteria adapted for dermoscopic features with semi-quantitative scoring; designed for use by less experienced clinicians; 3) Menzies method (Menzies, 1996) — alternative scoring with negative and positive features; 4) 7-point checklist (Argenziano, 1998) — simplified algorithm with major (2 points) and minor (1 point) criteria; widely used in clinical practice; 5) CASH algorithm (Henning, 2007) — Color-Architecture-Symmetry-Homogeneity approach; 6) 3-point checklist (Soyer, Argenziano et al., 2004) — simplified screening tool with high sensitivity for melanoma detection by primary care physicians; 7) Modified pattern analysis approaches; 8) Chaos and clues method (Rosendahl, 2011) — designed for non-pigmented and pigmented lesions, integrating modern dermoscopic understanding; 9) BLINCK algorithm for non-pigmented lesions; 10) Recent algorithms integrating AI and digital analysis (commercial systems including FotoFinder Bodystudio, MoleMax, Vectra WB360 with AI-assisted analysis tools).

Pattern analysis (Pehamberger): foundational algorithm requiring identification of three main components: 1) Pattern — overall arrangement of structures within lesion (reticular pattern with regular pigment network typical of acquired nevi, globular pattern with brown globules typical of compound nevi, cobblestone pattern, homogeneous pattern with diffuse pigmentation typical of dermal nevi, parallel pattern on palms and soles — parallel furrow benign acral nevus versus parallel ridge melanoma, multicomponent pattern with three or more distinct components suspicious for melanoma, atypical pattern); 2) Specific structures — pigment network (regular fine reticular versus atypical irregular thickened lines), dots and globules (regular distribution versus atypical irregular), streaks (regular versus atypical), blue-white veil (irregular structureless blue area with overlying whitish ground-glass haze suggestive of melanoma), regression structures (white scar-like areas, peppering of gray-blue dots), vessels (point vessels, comma vessels, hairpin vessels, dotted vessels, glomerular vessels, polymorphous vessels), ulceration; 3) Colors — light brown, dark brown, black, blue-gray, white, red, yellow; presence of multiple colors (5-6) suspicious for melanoma.

ABCD rule of dermoscopy (Stolz): semi-quantitative scoring system based on four criteria: 1) Asymmetry (A) — assessed in two perpendicular axes; 0 points (symmetric in both axes), 1 point (asymmetric in one axis), 2 points (asymmetric in both axes); 2) Border (B) — assessment of abrupt cutoff at lesion edge in 8 segments; score 0-8 points (one point per segment with abrupt cutoff); 3) Color (C) — number of colors present from list of 6 (white, red, light brown, dark brown, blue-gray, black); 1-6 points; 4) Differential structures (D) — number of dermoscopic structures present from list of 5 (pigment network, structureless areas, branched streaks, dots, globules); 1-5 points; total dermoscopy score (TDS) = A × 1.3 + B × 0.1 + C × 0.5 + D × 0.5; interpretation: TDS < 4.75 benign, 4.76-5.45 suspicious, > 5.45 highly suspicious for melanoma; sensitivity 70-95 percent, specificity 70-95 percent depending on observer; advantages include reproducibility, semi-quantitative score; limitations include requiring training, may underdiagnose featureless melanomas.