EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A YOLOv8 Implementation for Automatic Skin Pigmentation Detection: A Deep Learning Approach in Biomedical Image Analysis

Indriyani, Paula Dewanti, Yupiter HP Manurung

Engineering, Technology & Applied Science Research · 06.06.2026 · 16 (3) · 35529-35536

DOI: 10.48084/etasr.17895 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Engineering, Technology & Applied Science Research, online erschienen am 06.06.2026. Verfasst von Indriyani und 2 weiteren Autor:innen. Die Arbeit nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Skin pigmentation disorders such as melasma and Post-Inflammatory Hyperpigmentation (PIH) require objective assessment tools due to the subjectivity of clinical visual inspection. This study proposes a hybrid deep learning system that integrates YOLOv8 for real-time lesion detection and a fine-tuned Convolutional Neural Network (CNN) for quantitative severity classification, further validated by automated software quality assurance. Unlike previous works focusing on cancerous lesions, this study targets non-cancerous pigmentation and provides a complete, deployable pipeline. Using the HAM10000 dataset (9,988 dermoscopic images), the YOLOv8 detector achieved a high localization accuracy (mAP@0.5: 0.92, IoU: 0.85) with a low-latency speed of 125–168 ms per image. The CNN classifier (MobileNetV2) achieved a training accuracy of 75% and a weighted average precision of 0.60 on the validation set, with a precision of 0.70 on the Severe class. The system was deployed as a FastAPI web service, and its reliability was confirmed through automated end-to-end testing with Katalon Studio, passing all API and UI test cases. The results demonstrate that the hybrid model provides an objective tool for automated skin pigmentation screening, capable of assisting clinicians in treatment planning and monitoring. The key novelty lies in combining detection, quantitative severity grading, and rigorous software validation into a single, clinically-oriented system.

Zitieren

APA

Indriyani, Dewanti, P., & Manurung, Y. H. (2026). A YOLOv8 Implementation for Automatic Skin Pigmentation Detection: A Deep Learning Approach in Biomedical Image Analysis. Engineering, Technology & Applied Science Research, 16(3), 35529-35536. https://doi.org/10.48084/etasr.17895

Vancouver

Indriyani, Dewanti P, Manurung YH. A YOLOv8 Implementation for Automatic Skin Pigmentation Detection: A Deep Learning Approach in Biomedical Image Analysis. Engineering, Technology & Applied Science Research. 2026;16(3):35529-35536. doi:10.48084/etasr.17895

BibTeX
@article{Indriyani2026,
  author = {Indriyani and Dewanti, Paula and Manurung, Yupiter HP},
  title = {A YOLOv8 Implementation for Automatic Skin Pigmentation Detection: A Deep Learning Approach in Biomedical Image Analysis},
  journal = {Engineering, Technology & Applied Science Research},
  year = {2026},
  volume = {16},
  number = {3},
  pages = {35529--35536},
  doi = {10.48084/etasr.17895}
}
Kostenlos starten →