Comparative Evaluation of Global, Attention-Guided, and Modern Deep Representations for Multiclass Skin Lesion Classification

Automated multiclass skin lesion classification remains challenging because of visual overlap and heterogeneous acquisition conditions. This study compared eight representation strategies for basal cell carcinoma, squamous cell carcinoma, cutaneous leishmaniasis, pressure ulcer, and venous ulcer. Global, attention-guided regional, fusion-based, and modern deep representations from DeiT-Small, EfficientNet-B0, EfficientNetV2-S, ConvNeXt-Tiny, and Swin-Tiny were evaluated with support vector machines under a group-aware protocol including duplicate auditing, repeated cross-validation, multiseed fine-tuning, source-aware analyses, and a held-out test. Swin-Tiny achieved the highest OOF Macro-F1 (0.9466), a repeated-cross-validation mean of 0.9463 ± 0.0015, and a test Macro-F1 of 0.9333, without significant superiority over ConvNeXt-Tiny. Regional and fusion strategies did not outperform the global representation. Source information remained strongly encoded, and SCC-to-BCC errors were concentrated in PAD-UFES-20. Thus, the results demonstrate strong internal discrimination but not source-invariant or external clinical generalization, supporting the need for independent multicenter validation.

Authors

Institutions

Publication Details

Journal
BioMedInformatics
Published
2026-11-19
DOI
https://doi.org/10.3390/biomedinformatics6060113
Primary Topic
Cutaneous Melanoma Detection and Management
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Comparative Evaluation of Global, Attention-Guided, and Modern Deep Representations for Multiclass Skin Lesion Classification

Jorge Raúl Navarro-Cabrera, Miguel Angel Valles-Coral, Lloy Pinedo, Pierre Vidaurre-Rojas et al.
BioMedInformatics
Cutaneous Melanoma Detection and Management
article

Comparative Evaluation of Global, Attention-Guided, and Modern Deep Representations for Multiclass Skin Lesion Classification

Jorge Raúl Navarro-Cabrera, Miguel Angel Valles-Coral, Lloy Pinedo, Pierre Vidaurre-Rojas, Jhosep Sánchez-Flores, Monica Gabriela Leveau-Paredes, Alberto Alva-Arevalo, Augusto Ricardo Llontop-Reátegui
article en

Abstract

Automated multiclass skin lesion classification remains challenging because of visual overlap and heterogeneous acquisition conditions. This study compared eight representation strategies for basal cell carcinoma, squamous cell carcinoma, cutaneous leishmaniasis, pressure ulcer, and venous ulcer. Global, attention-guided regional, fusion-based, and modern deep representations from DeiT-Small, EfficientNet-B0, EfficientNetV2-S, ConvNeXt-Tiny, and Swin-Tiny were evaluated with support vector machines under a group-aware protocol including duplicate auditing, repeated cross-validation, multiseed fine-tuning, source-aware analyses, and a held-out test. Swin-Tiny achieved the highest OOF Macro-F1 (0.9466), a repeated-cross-validation mean of 0.9463 ± 0.0015, and a test Macro-F1 of 0.9333, without significant superiority over ConvNeXt-Tiny. Regional and fusion strategies did not outperform the global representation. Source information remained strongly encoded, and SCC-to-BCC errors were concentrated in PAD-UFES-20. Thus, the results demonstrate strong internal discrimination but not source-invariant or external clinical generalization, supporting the need for independent multicenter validation.

BioMedInformaticsVol. 6(6)
Universidad Nacional de San Martín (PE), Universidad Norbert Wiener (PE)
Cutaneous Melanoma Detection and Management
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon