用于3D器官图像分割的波随机自注意力编码器  

Wave Random Self-attention Encoder for 3D Organ Image Segmentation

作　　者：周迪 刘豪 程远志 李辉[1] 刘晓亚[1] ZHOU Di;LIU Hao;CHENG Yuan-Zhi;LI Hui;LIU Xiao-Ya(School of Data Science,Qingdao University of Science and Technology,Qingdao 266061,China;School of Information Science and Technology,Qingdao University of Science and Technology,Qingdao 266061,China;School of Computer Science and Technology,Harbin Institute of Technology,Harbin 150001,China)

机构地区：[1]青岛科技大学数据科学学院,青岛266061 [2]青岛科技大学信息科学技术学院,青岛266061 [3]哈尔滨工业大学计算机科学与技术学院,哈尔滨150001

出　　处：《计算机系统应用》2025年第2期84-91,共8页Computer Systems & Applications

基　　金：国家重点研发计划(2023YFF0612102);青岛市重点科技攻关及产业化示范项目(23-7-2-qljh-4-gx,24-1-2-qljh-19-gx)。

摘　　要：在光谱三维CT数据中,传统卷积的全局特征捕捉能力不足,而全尺度的自注意力机制则需要大量的计算资源.为了解决这一问题,本文引入一种新视觉注意力范式(wave self-attention,WSA).相比于ViT技术,该机制使用更少的资源获得同等的自注意力信息.此外,为更充分地提取器官间的相对依赖关系并提高模型的鲁棒性和执行速度,本文为WSA机制设计了一种即插即用的模块——波随机编码器(wave random encoder,WRE).该编码器能够生成一对互逆的非对称全局(局部)位置信息矩阵.其中,全局位置矩阵用来对波特征进行全局性的随机取样,局部位置矩阵则用于补充因随机取样而丢失的局部相对依赖.本文在标准数据集Synapse和COVID-19的肾脏和肺实质的分割任务上进行实验.结果表明,本文方法在精度、参数量和推理速率方面均超越了nnFormer、Swin-UNETR等现有模型,达到了SOTA水平.In spectral 3D CT data,the traditional convolution has a poor ability to capture global features,and the fullscale self-attention mechanism consumes large resources.To solve this problem,this study introduces a new visual attention paradigm,the wave self-attention(WSA).Compared with the ViT technology,this mechanism uses fewer resources to obtain the same amount of self-attention information.In addition,to more adequately extract the relative dependency among organs and to improve the robustness and execution speed of the model,a plug-and-play module,the wave random-encoder(WRE),is designed for the WSA mechanism.The encoder is capable of generating a pair of mutually inverse asymmetric global(local)position information matrices.The global position matrix is used to globally conduct random sampling of the wave features,and the local position matrix is used to complement the local relative dependency lost due to random sampling.In this study,experiments are performed on the task of segmenting the kidney and lung parenchyma in the standard datasets Synapse and COVID-19.The results show that this method outperforms existing models such as nnFormer and Swin-UNETR in terms of accuracy,the number of parameters,and inference rate,arriving at the SOTA level.

关 键 词：医学影像 图像分割 波自注意力机制 波随机编码器 

分 类 号：R318[医药卫生—生物医学工程] TP391.41[医药卫生—基础医学]