EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep Learning‐Based Skin Lesion Classification

Isabela Rocha Veiga da Silva iD, André Gonçalves Jardim iD, Giulia Rita de Souza Faés iD, Thatiane Alves Pianoschi Alva iD, Carla Diniz Lopes Becker iD, Viviane Rodrigues Botelho iD

Journal of Ultrasound in Medicine · 2026

DOI: 10.1002/jum.70125 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Ultrasound in Medicine, 2026. Verfasst von Isabela Rocha Veiga da Silva und 5 weiteren Autor:innen. Beteiligte Einrichtung: Federal University of Health Sciences of Porto Alegre, Porto Alegre, Brazil. Hauptthemen laut MEDLINE: Deep Learning, Skin Neoplasms, Image Interpretation, Computer-Assisted, Skin Diseases. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 12 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

High‐frequency ultrasound (HFUS) is valuable for assessing skin lesions, supporting diagnosis, treatment monitoring, and surgical planning. This study evaluates deep learning models for binary classification of HFUS images acquired in B‐mode and Doppler mode. Two single‐input CNNs were trained with each modality, while Unity and Cascade architectures combined both. The HFUS‐Doppler model achieved the best performance (95.0% accuracy, AUC 0.98), followed by Unity (90.5% accuracy, AUC 0.97). Cascade showed lower accuracy but greater confidence in malignant predictions. Probability distribution analysis revealed differences in model certainty near the decision threshold. Results indicate that combining B‐mode and Doppler can enhance diagnostic performance, depending on network design and data quality, supporting the potential of customized deep learning for non‐invasive HFUS‐based skin lesion classification.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningSkin NeoplasmsImage Interpretation, Computer-AssistedSkin DiseasesHumansUltrasonographySkinReproducibility of ResultsFemaleMaleSensitivity and Specificity

Stichwörter der Autor:innen

HFUSconvolutional neural networksskin neoplasmsultrasonography

Autor:innen und Institutionen

  • Federal University of Health Sciences of Porto Alegre, Porto Alegre, Brazil.

ORCID-Profile: Isabela Rocha Veiga da Silva, André Gonçalves Jardim, Giulia Rita de Souza Faés, Thatiane Alves Pianoschi Alva, Carla Diniz Lopes Becker, Viviane Rodrigues Botelho

Förderung

  • Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES)
  • Coordenação de Aperfeiçoamento de Pessoal de Nível Superior

Zitieren

APA

da Silva, I. R. V., Jardim, A. G., de Souza Faés, G. R., Alva, T. A. P., Becker, C. D. L., & Botelho, V. R. (2026). Deep Learning‐Based Skin Lesion Classification. Journal of Ultrasound in Medicine. https://doi.org/10.1002/jum.70125

Vancouver

da Silva IRV, Jardim AG, de Souza Faés GR, Alva TAP, Becker CDL, Botelho VR. Deep Learning‐Based Skin Lesion Classification. Journal of Ultrasound in Medicine. 2026. doi:10.1002/jum.70125

BibTeX
@article{daSilva2026,
  author = {da Silva, Isabela Rocha Veiga and Jardim, André Gonçalves and de Souza Faés, Giulia Rita and Alva, Thatiane Alves Pianoschi and Becker, Carla Diniz Lopes and Botelho, Viviane Rodrigues},
  title = {Deep Learning‐Based Skin Lesion Classification},
  journal = {Journal of Ultrasound in Medicine},
  year = {2026},
  doi = {10.1002/jum.70125}
}
Kostenlos starten →