摘要
目的:针对食管早癌图像分割过程中病灶边缘等细节信息丢失的问题,在U-net基础上提出一种基于上下文特征感知和双频上采样的食管早癌图像分割网络。方法:利用注意力机制和可分离空洞卷积改进上下文特征感知模块,获取全文上下文信息,提取更多特征细节。提出双频上采样模块,分别从高频和低频进行上采样,有效减少单一上采样因像素插值产生的锯齿效应和转置卷积造成的棋盘效应,减少细节信息的丢失。结果:本文方法的平均交并比、敏感度和特异性分别达到80.34%、87.47%和91.53%。结论:本文模型优于nnU-Net等主流语义分割模型,保留更多的细节信息,提高食管早癌图像分割精度。
Objective To propose a network for early esophageal cancer image segmentation using U-net with contextual feature awareness module and dual frequency upsampling module which solves the problem of loss of detailed information such as lesion edges during image segmentation.Methods The contextual feature awareness module improved with the attention mechanism and separable dilated convolution was used to obtain full-text contextual information and extract more feature details.The dual frequency upsampling module was adopted for upsampling from high frequency and low frequency,thereby effectively reducing the aliasing effect caused by pixel interpolation,minimizing the checkerboard effect caused by transposed convolution during single upsampling,and avoiding the loss of detail information.Results The mean intersection over union,sensitivity and specificity of the proposed method reached 80.34%,87.47%,and 91.53%,respectively.Conclusion The proposed model is superior to mainstream semantic segmentation models such as nnU-Net for it can retain more detailed information and improve the accuracy of early esophageal cancer image segmentation.
作者
孟延宗
李小霞
周颖玥
文黎明
秦佳敏
刘爽利
MENG Yanzong;LI Xiaoxia;ZHOU Yingyue;WEN Liming;QIN Jiamin;LIU Shuangli(School of Information Engineering,Southwest University of Science and Technology,Mianyang 621000,China;Robot Technology Used for Special Environment Key Laboratory of Sichuan Province,Mianyang 621000,China;Department of Gastroenterology,Sichuan Mianyang 404 Hospital,Mianyang 621000,China)
出处
《中国医学物理学杂志》
CSCD
2023年第8期957-963,共7页
Chinese Journal of Medical Physics
基金
国家自然科学基金(62071399)
四川省科技计划重点研发项目(2021YFG0383,2023YFG0262)。
关键词
食管早癌
上下文特征感知
注意力机制
空洞卷积
双频上采样
early esophageal cancer
contextual feature awareness
attention mechanism
dilated convolution
dual frequency upsampling
