DSCF-net: a dual-stream CNN-Mamba network for deep vein thrombosis segmentation
Physica Scripta · 12.06.2026 · 101 (23) · 236003
DOI: 10.1088/1402-4896/ae7774 ↗Kurz gefasst
Fachartikel in der Zeitschrift Physica Scripta, online erschienen am 15.06.2026. Verfasst von Meiyun Chen und 5 weiteren Autor:innen. Die Arbeit nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of Guangdong Province, National Research Program for Prevention and Management of Pulmonary Embolism and Deep Venous Thrombosis, Scientific Research Innovation Capability Support Project for Young Faculty.
Themen: Thromboseprophylaxe
Abstract (Originalsprache)
Deep vein thrombosis (DVT) poses a life-threatening risk due to potential pulmonary embolism, making timely and accurate diagnosis critical. However, automated segmentation of DVT in ultrasound images remains highly challenging due to multi-scale lesion features, pervasive speckle noise, and indistinct boundaries. To address these challenges, we propose dual-stream cross-branch fusion network (DSCF-Net), a dual-stream CNN-Mamba network with cross-branch fusion specifically designed for DVT segmentation. In the encoder, the CNN branch extracts fine-grained local features while the Mamba branch captures long-range dependencies to handle multi-scale lesion variations. The dual-stream spatial attention fusion (DSAF) module performs adaptive refinement to mitigate boundary ambiguity, while the cross pooling-guided fusion (CPGF) module employs multi-scale feature aggregation to suppress speckle noise. Experimental results on the clinical DVT dataset demonstrate that DSCF-Net achieves a Dice score of 78.27%, MIoU of 65.15%, F1 of 78.17%, and HD95 of 15.23 mm, outperforming state-of-the-art methods. Generalization experiments on the public TN3K dataset also achieved state-of-the-art performance, confirming the method’s strong transferability in ultrasound image segmentation.
Förderung
- Natural Science Foundation of Guangdong Province
- National Research Program for Prevention and Management of Pulmonary Embolism and Deep Venous Thrombosis
- Scientific Research Innovation Capability Support Project for Young Faculty
- Scientific Research Project of Guangdong Nurses Association
Zitieren
APA
Chen, M., Tan, Q., He, H., Huang, L., Wen, X., & Zhang, D. (2026). DSCF-net: a dual-stream CNN-Mamba network for deep vein thrombosis segmentation. Physica Scripta, 101(23), 236003. https://doi.org/10.1088/1402-4896/ae7774
Vancouver
Chen M, Tan Q, He H, Huang L, Wen X, Zhang D. DSCF-net: a dual-stream CNN-Mamba network for deep vein thrombosis segmentation. Physica Scripta. 2026;101(23):236003. doi:10.1088/1402-4896/ae7774
BibTeX
@article{Chen2026,
author = {Chen, Meiyun and Tan, Qilong and He, Haixia and Huang, Lishan and Wen, Xi and Zhang, Dawei},
title = {DSCF-net: a dual-stream CNN-Mamba network for deep vein thrombosis segmentation},
journal = {Physica Scripta},
year = {2026},
volume = {101},
number = {23},
pages = {236003},
doi = {10.1088/1402-4896/ae7774}
}
EUVIMED