LESS-Net: a lightweight network for epistaxis image segmentation using similarity-based contrastive learning
Ablation studies confirmed the distinct contribution of each architectural component to the model's overall efficacy. LESS-Net provides a powerful and data-efficient framework for medical image segmentation. Its demonstrated ability to achieve superior performance with limited supervision highlights its substantial potential to enhance AI-driven diagnostic capabilities and improve patient care in real-world clinical workflows, especially in underserved settings.
Authors
- Yunhan Wu (ORCID: https://orcid.org/0000-0001-7884-2994)
- Guanyu Chen (ORCID: https://orcid.org/0000-0002-1070-0911)
- Xianyao Wang (ORCID: https://orcid.org/0000-0002-6927-5737)
- Bing Yang (ORCID: https://orcid.org/0000-0002-2441-3649)
- Xi Zhu (ORCID: https://orcid.org/0000-0002-6563-6582)
- Jun Li (ORCID: https://orcid.org/0000-0002-1336-2241)
- Junyang Chen (ORCID: https://orcid.org/0000-0002-1093-0848)
- N. Xu (ORCID: https://orcid.org/0009-0004-5989-558X)
- Mengzhen Lai
- Shengxiang Zhou
- Yutong Huang (ORCID: https://orcid.org/0009-0006-2380-9795)
Institutions
- Sichuan Agricultural University (CN)
- Ya'an Polytechnic College (CN)
- Yan'an University (CN)
- Southeast University (CN)
Publication Details
- Journal
- Frontiers in Physiology
- Published
- 2025-10-29
- DOI
- https://doi.org/10.3389/fphys.2025.1644589
- Primary Topic
- Vascular Anomalies and Treatments
- Type
- article
- Field-Weighted Citation Impact
- 0.00