EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

IMPROVED DEEP LEARNING MODEL-BASED SKIN DISEASE IMAGE SEGMENTATION ALGORITHM

Tao Song

World Journal of Information Technology · 26.03.2026 · 4 (2) · 27-30

DOI: 10.61784/wjit3088 ↗

Kurz gefasst

Fachartikel in der Zeitschrift World Journal of Information Technology, online erschienen am 26.03.2026. Verfasst von Tao Song. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Aiming at the problems of complex background interference, low segmentation accuracy of small targets, blurred lesion boundaries and large scale differences in skin disease image segmentation, this paper proposes two improved deep learning models, namely the CAU-Net model fused with the coordinate attention mechanism and the CSwinTransU-Net model integrated with the spatial perception module. Experimental results on the ISIC2018 skin disease dataset show that the mean Intersection over Union (mIoU), mean Dice coefficient (mDice) and average accuracy (aAcc) of the CAU-Net model reach 85.50%, 92.06% and 93.69% respectively, and the corresponding indicators of the CSwinTransU-Net model are 84.84%, 91.66% and 93.49%. Both models outperform the basic models, effectively improving the accuracy and robustness of skin disease image segmentation, and providing technical support for the early diagnosis of skin cancer.

Zitieren

APA

Song, T. (2026). IMPROVED DEEP LEARNING MODEL-BASED SKIN DISEASE IMAGE SEGMENTATION ALGORITHM. World Journal of Information Technology, 4(2), 27-30. https://doi.org/10.61784/wjit3088

Vancouver

Song T. IMPROVED DEEP LEARNING MODEL-BASED SKIN DISEASE IMAGE SEGMENTATION ALGORITHM. World Journal of Information Technology. 2026;4(2):27-30. doi:10.61784/wjit3088

BibTeX
@article{Song2026,
  author = {Song, Tao},
  title = {IMPROVED DEEP LEARNING MODEL-BASED SKIN DISEASE IMAGE SEGMENTATION ALGORITHM},
  journal = {World Journal of Information Technology},
  year = {2026},
  volume = {4},
  number = {2},
  pages = {27--30},
  doi = {10.61784/wjit3088}
}
Kostenlos starten →