TexNet: Texture Loss Based Network for Gastric Antrum Segmentation in Ultrasound

Dong, Guangming; Zou, Yaoxian; Jiao, Jiaming; Liu, Yuxi; Liu, Shuo; Liang, Tianzhu; Liu, Chaoyue; Chen, Zhijie; Zhu, Lei; Ni, Dong; Lin, Muqing

doi:10.1007/978-3-030-59719-1_14

Cited by 3 publications

(1 citation statement)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…H-DenseUNet [15] is a hybrid U-Net model fusing 2D and 3D features for liver tumor segmentation in CT images. Dong et al [16] propose a TexNet on U-Net architecture with attention blocks on decoder to segment gastric antrum ultrasound images. Similarly, Yu et al [17] present a U-Net-like neural architecture search to overcome 3D segmentation tasks in a coarse-to-fine manner.…”

Section: A Encoder-decoder Architecturementioning

confidence: 99%

Crosslink-Net: Double-branch Encoder Segmentation Network via Fusing Vertical and Horizontal Convolutions

Yu¹,

Zhou

et al. 2021

Preprint

View full text Add to dashboard Cite

Accurate image segmentation plays a crucial role in medical image analysis, yet it faces great challenges of various shapes, diverse sizes, and blurry boundaries. To address these difficulties, square kernel-based encoder-decoder architecture has been proposed and widely used, but its performance remains still unsatisfactory. To further cope with these challenges, we present a novel double-branch encoder architecture. Our architecture is inspired by two observations: 1) Since the discrimination of features learned via square convolutional kernels needs to be further improved, we propose to utilize non-square vertical and horizontal convolutional kernels in the double-branch encoder, so features learned by the two branches can be expected to complement each other. 2) Considering that spatial attention can help models to better focus on the target region in a largesized image, we develop an attention loss to further emphasize the segmentation on small-sized targets. Together, the above two schemes give rise to a novel double-branch encoder segmentation framework for medical image segmentation, namely Crosslink-Net. The experiments validate the effectiveness of our model on four datasets. The code is released at https://github.com/ Qianyu1226/Crosslink-Net.

show abstract

Section: A Encoder-decoder Architecturementioning

confidence: 99%

Crosslink-Net: Double-branch Encoder Segmentation Network via Fusing Vertical and Horizontal Convolutions

Yu¹,

Zhou

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

A weakly supervised deep active contour model for nodule segmentation in thyroid ultrasound images

Zhou

Chang

et al. 2023

Pattern Recognition Letters

View full text Add to dashboard Cite

Crosslink-Net: Double-Branch Encoder Network via Fusing Vertical and Horizontal Convolutions for Medical Image Segmentation

Wang

et al. 2022

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

TexNet: Texture Loss Based Network for Gastric Antrum Segmentation in Ultrasound

Cited by 3 publications

References 9 publications

Crosslink-Net: Double-branch Encoder Segmentation Network via Fusing Vertical and Horizontal Convolutions

Crosslink-Net: Double-branch Encoder Segmentation Network via Fusing Vertical and Horizontal Convolutions

A weakly supervised deep active contour model for nodule segmentation in thyroid ultrasound images

Crosslink-Net: Double-Branch Encoder Network via Fusing Vertical and Horizontal Convolutions for Medical Image Segmentation

Contact Info

Product

Resources

About