TFCNs: A CNN-Transformer Hybrid Network for Medical Image Segmentation

Medical image segmentation is one of the most fundamental tasks concerning medical information analysis. Various solutions have been proposed so far, including many deep learning-based techniques, such as U-Net, FC-DenseNet, etc. However, high-precision remains a highly challenging task due to existence inherent magnification and distortion in images well presence lesions with similar density normal tissues. In this paper, we propose TFCNs (Transformers for Fully Convolutional denseNets) tackle problem by introducing ResLinear-Transformer (RL-Transformer) Linear Attention Block (CLAB) FC-DenseNet. not only able utilize more latent from CT feature extraction, but also can capture disseminate semantic features filter non-semantic effectively through CLAB module. Our experimental results show that achieve state-of-the-art performance dice scores 83.72% on Synapse dataset. addition, evaluate robustness lesion area effects COVID-19 public datasets. The Python code will be made publicly available https://github.com/HUANGLIZI/TFCNs .