EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

SHIELD-Net: A Sperm Halo Image Classification Framework via Hybrid ROI Extraction and ROI-Guided Dual-View Learning

Lesong Zheng, Yiwen Xu, Wenjun Gu, Yizhi Song, Yunbo Guo, Jingkai Wang

AI Chemistry · 02.11.2026 · 1 (4) · 19

DOI: 10.3390/aichem1040019 ↗

Kurz gefasst

Fachartikel in der Zeitschrift AI Chemistry, online erschienen am 02.11.2026. Verfasst von Lesong Zheng und 5 weiteren Autor:innen. Die Arbeit nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of Jiangsu Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The sperm chromatin dispersion (SCD) test is widely used to assess sperm DNA fragmentation by observing the halo structure around the sperm head. However, manual halo interpretation is labor-intensive and subjective, which limits its consistency and throughput in clinical practice. To address these limitations, we propose SHIELD-Net, an image-level single-sperm halo morphology classification framework for four halo categories: large halo, medium halo, small halo, and without halo. SHIELD-Net includes two modules, i.e., a hybrid ROI extraction (HyREx) module and dual-view learning with ROI-guided gate fusion (RGF). The HyREx module converts the traditional direct classification pipeline into a segmentation-classification pipeline. It uses the cell segmentation foundation model, i.e., Cellpose, to segment halo-centered regions of interest (ROIs) and proposes a fallback mechanism to ensure segmentation effectiveness. In this way, HyREx can construct a head–halo-centered ROI and reduce interference from tail and background regions. Based on the original image and the extracted ROI, SHIELD-Net performs dual-view ConvNeXt encoding. RGF further uses the ROI representation to generate a spatial gate that recalibrates the original-view feature map toward halo-morphology-related responses. Experiments were conducted on a public expert-annotated sperm microscopy dataset containing 403 individual sperm halo images using stratified five-fold cross-validation. SHIELD-Net achieved a Macro F1 of 0.9290 ± 0.0161. Compared with the original-image ConvNeXt-Small baseline, SHIELD-Net improved Macro F1 by 0.1755. These results indicate that combining head–halo-centered ROI construction with ROI-guided feature recalibration improves image-level sperm halo morphology classification.

Förderung

  • Natural Science Foundation of Jiangsu Province

Zitieren

APA

Zheng, L., Xu, Y., Gu, W., Song, Y., Guo, Y., & Wang, J. (2026). SHIELD-Net: A Sperm Halo Image Classification Framework via Hybrid ROI Extraction and ROI-Guided Dual-View Learning. AI Chemistry, 1(4), 19. https://doi.org/10.3390/aichem1040019

Vancouver

Zheng L, Xu Y, Gu W, Song Y, Guo Y, Wang J. SHIELD-Net: A Sperm Halo Image Classification Framework via Hybrid ROI Extraction and ROI-Guided Dual-View Learning. AI Chemistry. 2026;1(4):19. doi:10.3390/aichem1040019

BibTeX
@article{Zheng2026,
  author = {Zheng, Lesong and Xu, Yiwen and Gu, Wenjun and Song, Yizhi and Guo, Yunbo and Wang, Jingkai},
  title = {SHIELD-Net: A Sperm Halo Image Classification Framework via Hybrid ROI Extraction and ROI-Guided Dual-View Learning},
  journal = {AI Chemistry},
  year = {2026},
  volume = {1},
  number = {4},
  pages = {19},
  doi = {10.3390/aichem1040019}
}
Kostenlos starten →