文献类型：期刊文献

中文题名：USSL Net:Focusing on Structural Similarity with Light U-Structure for Stroke Lesion Segmentation

英文题名：USSL Net: Focusing on Structural Similarity with Light U-Structure for Stroke Lesion Segmentation

作者：Jiang, Zhiguo[1]; Chang, Qing[1]

机构：[1] School of Information Science and Engineering, East China University of Science and Technology, Shanghai, 200237, China

年份：2022

卷号：27

期号：4

起止页码：485

中文期刊名：Journal of Shanghai Jiaotong university(Science)

外文期刊名：Journal of Shanghai Jiaotong University (Science)

收录：EI(收录号：20221812057677);Scopus;PubMed

基金：the National Natural Science Foundation of China(No.61976091)。

语种：英文

中文关键词：structural similarity;medical image segmentation;deep convolution neural network;automatic data enhancement algorithm

外文关键词：Computerized tomography - Deep neural networks - Image enhancement - Image segmentation - Medical image processing - Metadata

摘要：Automatic segmentation of ischemic stroke lesions from computed tomography(CT)images is of great significance for identifying and curing this life-threatening condition.However,in addition to the problem of low image contrast,it is also challenged by the complex changes in the appearance of the stroke area and the difficulty in obtaining image data.Considering that it is difficult to obtain stroke data and labels,a data enhancement algorithm for one-shot medical image segmentation based on data augmentation using learned transformation was proposed to increase the number of data sets for more accurate segmentation.A deep convolutional neural network based algorithm for stroke lesion segmentation,called structural similarity with light U-structure(USSL)Net,was proposed.We embedded a convolution module that combines switchable normalization,multi-scale convolution and dilated convolution in the network for better segmentation performance.Besides,considering the strong structural similarity between multi-modal stroke CT images,the USSL Net uses the correlation maximized structural similarity loss(SSL)function as the loss function to learn the varying shapes of the lesions.The experimental results show that our framework has achieved results in the following aspects.First,the data obtained by adding our data enhancement algorithm is better than the data directly segmented from the multi-modal image.Second,the performance of our network model is better than that of other models for stroke segmentation tasks.Third,the way SSL functioned as a loss function is more helpful to the improvement of segmentation accuracy than the cross-entropy loss function.
Automatic segmentation of ischemic stroke lesions from computed tomography (CT) images is of great significance for identifying and curing this life-threatening condition. However, in addition to the problem of low image contrast, it is also challenged by the complex changes in the appearance of the stroke area and the difficulty in obtaining image data. Considering that it is difficult to obtain stroke data and labels, a data enhancement algorithm for one-shot medical image segmentation based on data augmentation using learned transformation was proposed to increase the number of data sets for more accurate segmentation. A deep convolutional neural network based algorithm for stroke lesion segmentation, called structural similarity with light U-structure (USSL) Net, was proposed. We embedded a convolution module that combines switchable normalization, multi-scale convolution and dilated convolution in the network for better segmentation performance. Besides, considering the strong structural similarity between multi-modal stroke CT images, the USSL Net uses the correlation maximized structural similarity loss (SSL) function as the loss function to learn the varying shapes of the lesions. The experimental results show that our framework has achieved results in the following aspects. First, the data obtained by adding our data enhancement algorithm is better than the data directly segmented from the multi-modal image. Second, the performance of our network model is better than that of other models for stroke segmentation tasks. Third, the way SSL functioned as a loss function is more helpful to the improvement of segmentation accuracy than the cross-entropy loss function. ? 2022, Shanghai Jiao Tong University.