Crosslink-Net: Double-branch Encoder Segmentation Network via Fusing Vertical and Horizontal Convolutions
This work addresses medical image segmentation challenges like varied shapes and blurry boundaries, but it appears incremental as it builds on existing encoder-decoder architectures with specific kernel and attention modifications.
The paper tackles the problem of accurate medical image segmentation by proposing Crosslink-Net, a double-branch encoder architecture that fuses vertical and horizontal convolutions and uses an attention loss, achieving improved performance validated on four datasets.
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 large-sized 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.