Xu Yang, He Xianyu, Xu Guofeng, Qi Guanqiu, Yu Kun, Yin Li, Yang Pan, Yin Yuehui, Chen Hao
College of Automation, Chongqing University of Posts and Telecommunications, Chongqing, China.
Department of Computer Information Systems, Buffalo State College, Buffalo, NY, United States.
Front Neurosci. 2022 Sep 15;16:1009581. doi: 10.3389/fnins.2022.1009581. eCollection 2022.
Medical image segmentation has important auxiliary significance for clinical diagnosis and treatment. Most of existing medical image segmentation solutions adopt convolutional neural networks (CNNs). Althought these existing solutions can achieve good image segmentation performance, CNNs focus on local information and ignore global image information. Since Transformer can encode the whole image, it has good global modeling ability and is effective for the extraction of global information. Therefore, this paper proposes a hybrid feature extraction network, into which CNNs and Transformer are integrated to utilize their advantages in feature extraction. To enhance low-dimensional texture features, this paper also proposes a multi-dimensional statistical feature extraction module to fully fuse the features extracted by CNNs and Transformer and enhance the segmentation performance of medical images. The experimental results confirm that the proposed method achieves better results in brain tumor segmentation and ventricle segmentation than state-of-the-art solutions.
医学图像分割对临床诊断和治疗具有重要的辅助意义。现有的大多数医学图像分割解决方案都采用卷积神经网络（CNN）。尽管这些现有解决方案能够实现良好的图像分割性能，但CNN专注于局部信息而忽略了全局图像信息。由于Transformer可以对整个图像进行编码，因此它具有良好的全局建模能力，并且对全局信息的提取有效。因此，本文提出了一种混合特征提取网络，将CNN和Transformer集成到其中，以利用它们在特征提取方面的优势。为了增强低维纹理特征，本文还提出了一种多维统计特征提取模块，以充分融合CNN和Transformer提取的特征，并提高医学图像的分割性能。实验结果证实，所提出的方法在脑肿瘤分割和脑室分割方面比现有最先进的解决方案取得了更好的结果。
