G-UNeXt: a lightweight MLP-based network for reducing semantic gap in medical image segmentation

Zhang, Xin; Cao, Xiaotian; Wang, Jun; Wan, Lei

doi:10.1007/s00530-023-01173-z

Cited by 4 publications

(1 citation statement)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…UNeXt [4] introduces the sliding window idea of the Swin Transformer [30] and designs the Tok-MLP module, which effectively enhances the model's ability to capture long-distance dependencies between features and achieves advanced performance on the ISIC2018 and BUSI datasets with low consumption. The subsequent works based on this foundation include G-UNeXt [31] and Res2-UNeXt [32]. The former uses linear computation to reduce the computational cost of redundant features, and the latter enhances the fusion of multi-scale information.…”

Section: Hybrid Modelsmentioning

confidence: 99%

PIS-Net: Efficient Medical Image Segmentation Network with Multivariate Downsampling for Point-of-Care

Zhang,

Wang

2024

Entropy

View full text Add to dashboard Cite

Recently, with more portable diagnostic devices being moved to people anywhere, point-of-care (PoC) imaging has become more convenient and more popular than the traditional “bed imaging”. Instant image segmentation, as an important technology of computer vision, is receiving more and more attention in PoC diagnosis. However, the image distortion caused by image preprocessing and the low resolution of medical images extracted by PoC devices are urgent problems that need to be solved. Moreover, more efficient feature representation is necessary in the design of instant image segmentation. In this paper, a new feature representation considering the relationships among local features with minimal parameters and a lower computational complexity is proposed. Since a feature window sliding along a diagonal can capture more pluralistic features, a Diagonal-Axial Multi-Layer Perceptron is designed to obtain the global correlation among local features for a more comprehensive feature representation. Additionally, a new multi-scale feature fusion is proposed to integrate nonlinear features with linear ones to obtain a more precise feature representation. Richer features are figured out. In order to improve the generalization of the models, a dynamic residual spatial pyramid pooling based on various receptive fields is constructed according to different sizes of images, which alleviates the influence of image distortion. The experimental results show that the proposed strategy has better performance on instant image segmentation. Notably, it yields an average improvement of 1.31% in Dice than existing strategies on the BUSI, ISIC2018 and MoNuSeg datasets.

show abstract

Section: Hybrid Modelsmentioning

confidence: 99%