Artificial Intelligence‐Based Identification of Common Canine Skin Lesions From Clinical Images
Veterinary Dermatology · 2026-08 · 37 (4) · 471-477
DOI: 10.1111/vde.70083 ↗Kurz gefasst
Fachartikel in der Zeitschrift Veterinary Dermatology, online erschienen am 01.05.2026. Verfasst von Soh-Yoon Kang und 8 weiteren Autor:innen. Beteiligte Einrichtung: Laboratory of Veterinary Dermatology and the Research Institute for Veterinary Science, College of Veterinary Medicine, Seoul National University, Se…. Hauptthemen laut MEDLINE: Dog Diseases, Skin Diseases, Artificial Intelligence. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: AIFORPET Corp.
Abstract (Originalsprache)
Background
Accurate evaluation of skin lesions is an essential component of dermatological examination, yet it can be time‐consuming and subject to interobserver variability. While artificial intelligence (AI) models have shown reliability in diagnosing specific skin diseases, lesion‐level identification remains underexplored in veterinary dermatology.
Objectives
To develop and evaluate a convolutional neural network (CNN)‐based AI model for the automated identification of four skin lesion types in dogs: erythema, lichenification, alopecia and erosion/ulcer. Animals Clinical skin images were collected from dogs presented to a veterinary medical teaching hospital. Materials and Methods Four EfficientNet models were independently trained, one for each lesion type. Model performance was evaluated by comparing the prediction results with veterinary surgeon‐labelled data, using six metrics: accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and F1 score.
Results
All four CNN models achieved an accuracy of > 90%, indicating a reliable performance across all lesion types. The alopecia model yielded the highest accuracy (98.12%) and F1 score (98.18%). The erythema and erosion/ulcer models exhibited balanced performance across all metrics. For the lichenification model, the sensitivity (87.02%) and F1 score (89.88%) were the lowest among the four lesion types. Conclusions and Clinical Relevance The CNN‐based AI models developed in this study demonstrated validity in identifying common canine skin lesions from clinical images. These models may facilitate rapid and objective dermatological evaluation, supporting clinical diagnosis and lesion monitoring throughout treatment.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Laboratory of Veterinary Dermatology and the Research Institute for Veterinary Science, College of Veterinary Medicine, Seoul National University, Seoul, Korea.
- Western Animal Medical Center, Seoul, Korea.
- AIFORPET Corporation, Seoul, Korea.
- Suvidha Pet Care and Derma Centre, Jammu, Jammu and Kashmir, India.
ORCID-Profile: Soh-Yoon Kang, Yeong-Hun Kang, Min-Sun Kim, Anisha Tiwari, Cheol-Yong Hwang
Förderung
- AIFORPET Corp
Zitieren
APA
Kang, S. Y., Kang, Y. H., Kim, H. J., Huh, E. A., Jeon, J. S., Kim, H. S., Kim, M. S., Tiwari, A., & Hwang, C. Y. (2026). Artificial Intelligence‐Based Identification of Common Canine Skin Lesions From Clinical Images. Veterinary Dermatology, 37(4), 471-477. https://doi.org/10.1111/vde.70083
Vancouver
Kang SY, Kang YH, Kim HJ, Huh EA, Jeon JS, Kim HS, et al. Artificial Intelligence‐Based Identification of Common Canine Skin Lesions From Clinical Images. Veterinary Dermatology. 2026;37(4):471-477. doi:10.1111/vde.70083
BibTeX
@article{Kang2026,
author = {Kang, Soh-Yoon and Kang, Yeong-Hun and Kim, Hyun-Jung and Huh, Eun-A and Jeon, Jong-Soo and Kim, Hyun-Su and Kim, Min-Sun and Tiwari, Anisha and Hwang, Cheol-Yong},
title = {Artificial Intelligence‐Based Identification of Common Canine Skin Lesions From Clinical Images},
journal = {Veterinary Dermatology},
year = {2026},
volume = {37},
number = {4},
pages = {471--477},
doi = {10.1111/vde.70083}
}
EUVIMED