EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

HHBSNet: a global channel–spatial attention and multi‐scale dilated convolution network for automatic melasma segmentation

Shange Wang, Lin Xu, Linshuai Zhang, Yujie Zhang, Chen Li, Marcin Grzegorzek, Jing Guo, Tao Jiang

Frontiers in Physiology · 05.11.2025 · 16

DOI: 10.3389/fphys.2025.1665138 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Physiology, online erschienen am 05.11.2025. Verfasst von Shange Wang und 7 weiteren Autor:innen. Beteiligte Einrichtung: School of Intelligent Medicine, Chengdu University of Traditional Chinese Medicine, Chengdu, China. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

Melasma is a common acquired facial hyperpigmentation disorder characterized by symmetrical brown patches, often occurring in the zygomatic region, forehead, and upper lip. Its blurred boundaries, color similarity to normal skin, and irregular morphology—combined with lighting variability and skin reflections—pose significant challenges for automated lesion segmentation. This study aims to develop an effective and lightweight deep learning model tailored for accurate melasma segmentation.

Methods

We propose a novel lightweight segmentation network, HHBSNet, specifically designed for melasma lesion analysis. The model incorporates a Global Channel-Spatial Attention (GCSA) module that jointly leverages channel and spatial attention to suppress lighting interference and enhance feature discrimination in low-contrast, irregular boundaries. In addition, a Multiscale Cavity Fusion (MCF) module is introduced to extend the receptive field via multi-dilation rates, enabling effective capture of lesions at various scales without reducing resolution. The network further integrates local-global semantic fusion and adopts a combined loss strategy of cross-entropy and focal loss to address class imbalance.

Results

HHBSNet was evaluated on a self-constructed dataset comprising 501 practical facial melasma images. Quantitative results demonstrate that HHBSNet outperforms existing mainstream segmentation methods, achieving a mean Intersection over Union (Miou) of 79.69%, accuracy (ACC) of 96.68%, F-score of 88.10%, recall of 88.18%, and precision of 87.80%.

Conclusion

The proposed HHBSNet demonstrates superior segmentation performance and robustness in handling melasma’s challenging visual characteristics. Its lightweight structure and strong generalization ability suggest promising potential for application in computer-aided diagnosis and large-scale clinical screening of facial pigmentary disorders.

Schlagwörter

Stichwörter der Autor:innen

attention mechanismautomatic segmentationdeep learningimage segmentationmelasma

Autor:innen und Institutionen

  • School of Intelligent Medicine, Chengdu University of Traditional Chinese Medicine, Chengdu, China.
  • College of Medicine and Biological Information Engineering, Northeastern University, Shenyang, China.
  • Institute of Medical Informatics, University of Lübeck, Lübeck, Germany.
  • Chengdu University of Traditional Chinese Medicine, Chengdu, China.

Zitieren

APA

Wang, S., Xu, L., Zhang, L., Zhang, Y., Li, C., Grzegorzek, M., Guo, J., & Jiang, T. (2025). HHBSNet: a global channel–spatial attention and multi‐scale dilated convolution network for automatic melasma segmentation. Frontiers in Physiology, 16. https://doi.org/10.3389/fphys.2025.1665138

Vancouver

Wang S, Xu L, Zhang L, Zhang Y, Li C, Grzegorzek M, et al. HHBSNet: a global channel–spatial attention and multi‐scale dilated convolution network for automatic melasma segmentation. Frontiers in Physiology. 2025;16. doi:10.3389/fphys.2025.1665138

BibTeX
@article{Wang2025,
  author = {Wang, Shange and Xu, Lin and Zhang, Linshuai and Zhang, Yujie and Li, Chen and Grzegorzek, Marcin and Guo, Jing and Jiang, Tao},
  title = {HHBSNet: a global channel–spatial attention and multi‐scale dilated convolution network for automatic melasma segmentation},
  journal = {Frontiers in Physiology},
  year = {2025},
  volume = {16},
  doi = {10.3389/fphys.2025.1665138}
}
Kostenlos starten →