Contextual feature aggregation pyramid vision transformer for polyp segmentation

Ziwei Lin; Kaiyue Zhang; Shanshan Li; Qing Qin

doi:10.1117/12.3039730

11 September 2024 Contextual feature aggregation pyramid vision transformer for polyp segmentation

Ziwei Lin, Kaiyue Zhang, Shanshan Li, Qing Qin

Author Affiliations +

Proceedings Volume 13270, International Conference on Future of Medicine and Biological Information Engineering (MBIE 2024); 1327004 (2024) https://doi.org/10.1117/12.3039730
Event: 2024 International Conference on Future of Medicine and Biological Information Engineering (MBIE 2024), 2024, Shenyang, China

Abstract

Polyp segmentation has consistently been a difficult task because of the varying sizes of polyps and the significant intrinsic similarity between polyps and the surrounding tissues. To address the above problems, a contextual feature aggregation polyp segmentation algorithm combining Pyramid Vision Transformer and convolution (CFA-PVT) is proposed. Firstly, the Pyramid Vision Transformer is used to extract image global features, and the stage bridging module(SBM) is employed to enhance the ability of the network to handle polyp details and aggregate high-level polyp features. Subsequently, a feature enhancement module (FEM) is used to explore shallow polyp information. Finally, cross-layer feature fusion is performed by a global adaptive module (GAM) to realize feature interaction. This algorithm is evaluated on the CVC-ClinicDB and Kvasir-SEG datasets and further tested for generalization capability on the CVC-ColonDB dataset. The results demonstrate that the proposed method effectively segments colorectal polyp images, offering a new approach to diagnosing colorectal polyps.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Ziwei Lin, Kaiyue Zhang, Shanshan Li, and Qing Qin "Contextual feature aggregation pyramid vision transformer for polyp segmentation", Proc. SPIE 13270, International Conference on Future of Medicine and Biological Information Engineering (MBIE 2024), 1327004 (11 September 2024); https://doi.org/10.1117/12.3039730

ACCESS THE FULL ARTICLE

PROCEEDINGS
6 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY