Empirical Comparisons of Deep Learning Networks on Liver Segmentation

作     者：Yi Shen Victor S.Sheng Lei Wang Jie Duan Xuefeng Xi Dengyong Zhang Ziming Cui 

作者机构：School of Electronic and Information EngineeringSuzhou University of Science and TechnologySuzhouChina Department of Computer ScienceUniversity of Central ArkansasConwayArkansasUSA Huan Provincial Key Laboratory of Intelligent Processing of Big Data on TransportationSchool of Computer and Communication EngineeringChangsha University of Science and TechnologyChangshaChina 

出 版 物：《Computers, Materials & Continua》 (计算机、材料和连续体（英文）)

年 卷 期：2020年第62卷第3期

页      面：1233-1247页

核心收录：

学科分类：1002[医学-临床医学] 100201[医学-内科学(含：心血管病、血液病、呼吸系病、消化系病、内分泌与代谢病、肾病、风湿病、传染病)] 10[医学] 

基　　金：This research has been partially supported by National Science Foundation under grant IIS-1115417 the National Natural Science Foundation of China under grant 61728205,61876217 the“double first-class”international cooperation and development scientific research project of Changsha University of Science and Technology(No.2018IC25) the Science and Technology Development Project of Suzhou under grant SZS201609 and SYG201707 

主　　题：Liver segmentation deep learning FCN U-Net Segnet Resnet Densenet 

摘      要：Accurate segmentation of CT images of liver tumors is an important adjunct for the liver diagnosis and treatment of liver *** recent years,due to the great improvement of hard device,many deep learning based methods have been proposed for automatic liver *** them,there are the plain neural network headed by FCN and the residual neural network headed by Resnet,both of which have many *** have achieved certain achievements in medical image *** this paper,we firstly select five representative structures,i.e.,FCN,U-Net,Segnet,Resnet and Densenet,to investigate their performance on liver *** original Resnet and Densenet could not perform image segmentation directly,we make some adjustments for them to perform live *** experimental results show that Densenet performs the best on liver segmentation,followed by *** perform much better than Segnet,U-Net,and *** Segnet,U-Net,and FCN,U-Net performs the best,followed by *** performs the worst.