基于Transformer和卷积神经网络双并行分支编码器神经网络的冠状动脉分割  

Coronary artery segmentation based on Transformer and convolutional neural networks dual parallel branch encoder neural network

作　　者：潘丹 骆根强 曾安[3] PAN Dan;LUO Genqiang;ZENG An(School of Electronics and Information Engineering,Guangdong University of Technology and Education,Guangzhou 510665,P.R.China;School of Computer and Information Engineering,Guangdong Songshan Polytechnic,Shaoguan,Guangdong 512126,P.R.China;School of Computers,Guangdong University of Technology,Guangzhou 510006,P.R.China)

机构地区：[1]广东技术师范大学电子与信息学院,广州510665 [2]广东松山职业技术学院计算机与信息工程学院,广东韶关512126 [3]广东工业大学计算机学院,广州510006

出　　处：《生物医学工程学杂志》2024年第6期1195-1203,1212,共10页Journal of Biomedical Engineering

基　　金：国家自然科学基金项目(61976058,92267107);广东省科技计划项目(2021B0101220006,2021A1515012300,2019A050510041);广州市科技计划项目(202206010007,202103000034,202002020090)。

摘　　要：针对计算机断层扫描血管造影(CTA)图像的冠状动脉人工手动分割效率低下,而现有深度学习分割模型在冠状动脉图像上分割准确率较低的问题,受Transformer的启发,本文提出了一种双并行分支编码器的分割模型——DUNETR。该网络以Transformer和卷积神经网络(CNN)作为双编码器,Transformer编码器负责将三维(3D)冠状动脉数据转变成一维(1D)序列问题进行学习并捕获其有效的全局多尺度特征信息,CNN编码器则提取3D冠状动脉的局部特征,二者所提取到的不同特征信息通过噪声降低的特征融合(NRFF)模块的拼接融合后连接到解码器。在公开数据集上的实验结果表明,提出的DUNETR网络结构模型在Dice相似性系数方面达到了81.19%,召回率达到了80.18%,相比对比实验中次好结果模型有0.49%和0.46%的提升,超越了其他常规深度学习方法。将Transformer和CNN作为双编码器而共同提取到的丰富特征信息,会有助于进一步提升3D冠状动脉分割的效果。同时,该模型也为其他血管状器官分割提供了新思路。Manual segmentation of coronary arteries in computed tomography angiography(CTA)images is inefficient,and existing deep learning segmentation models often exhibit low accuracy on coronary artery images.Inspired by the Transformer architecture,this paper proposes a novel segmentation model,the double parallel encoder unet with transformers(DUNETR).This network employed a dual-encoder design integrating Transformers and convolutional neural networks(CNNs).The Transformer encoder transformed three-dimensional(3D)coronary artery data into a one-dimensional(1D)sequential problem,effectively capturing global multi-scale feature information.Meanwhile,the CNN encoder extracted local features of the 3D coronary arteries.The complementary features extracted by the two encoders were fused through the noise reduction feature fusion(NRFF)module and passed to the decoder.Experimental results on a public dataset demonstrated that the proposed DUNETR model achieved a Dice similarity coefficient of 81.19%and a recall rate of 80.18%,representing improvements of 0.49%and 0.46%,respectively,over the next best model in comparative experiments.These results surpassed those of other conventional deep learning methods.The integration of Transformers and CNNs as dual encoders enables the extraction of rich feature information,significantly enhancing the effectiveness of 3D coronary artery segmentation.Additionally,this model provides a novel approach for segmenting other vascular structures.

关 键 词：卷积神经网络 TRANSFORMER 计算机断层扫描血管造影图像 冠状动脉 注意力机制 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP391.41[自动化与计算机技术—控制科学与工程] R318[医药卫生—生物医学工程]