The segmentation of unlabeled medical images is troublesome due to the high cost of annotation, and unsupervised domain adaptation is one solution to this. In this paper, an improved unsupervised domain adaptation met...The segmentation of unlabeled medical images is troublesome due to the high cost of annotation, and unsupervised domain adaptation is one solution to this. In this paper, an improved unsupervised domain adaptation method was proposed. The proposed method considered both global alignment and category-wise alignment. First, we aligned the appearance of two domains by image transformation. Second, we aligned the output maps of two domains in a global way. Then, we decomposed the semantic prediction map by category, aligning the prediction maps in a category-wise manner. Finally, we evaluated the proposed method on the 2017 Multi-Modality Whole Heart Segmentation Challenge dataset, and obtained 82.1 on the dice similarity coefficient and 4.6 on the average symmetric surface distance, demonstrating the effectiveness of the combination of global alignment and category-wise alignment.展开更多
Cardiomyopathy is one of the most serious public health threats.The precise structural and functional cardiac measurement is an essential step for clinical diagnosis and follow-up treatment planning.Cardiologists are ...Cardiomyopathy is one of the most serious public health threats.The precise structural and functional cardiac measurement is an essential step for clinical diagnosis and follow-up treatment planning.Cardiologists are often required to draw endocardial and epicardial contours of the left ventricle(LV)manually in routine clinical diagnosis or treatment planning period.This task is time-consuming and error-prone.Therefore,it is necessary to develop a fully automated end-to-end semantic segmentation method on cardiac magnetic resonance(CMR)imaging datasets.However,due to the low image quality and the deformation caused by heartbeat,there is no effective tool for fully automated end-to-end cardiac segmentation task.In this work,we propose a multi-scale segmentation network(MSSN)for left ventricle segmentation.It can effectively learn myocardium and blood pool structure representations from 2D short-axis CMR image slices in a multi-scale way.Specifically,our method employs both parallel and serial of dilated convolution layers with different dilation rates to capture multi-scale semantic features.Moreover,we design graduated up-sampling layers with subpixel layers as the decoder to reconstruct lost spatial information and produce accurate segmentation masks.We validated our method using 164 T1 Mapping CMR images and showed that it outperforms the advanced convolutional neural network(CNN)models.In validation metrics,we archived the Dice Similarity Coefficient(DSC)metric of 78.96%.展开更多
Numerous methods have been published to segment the infarct tissue in theleft ventricle, most of them either need manual work, post-processing, or suffer from poorreproducibility. We proposed an automatic segmentation...Numerous methods have been published to segment the infarct tissue in theleft ventricle, most of them either need manual work, post-processing, or suffer from poorreproducibility. We proposed an automatic segmentation method for segmenting the infarct tissue irleft ventricle with myocardial infarction. Cardiac images of a total of 60 diseased hearts (55 humanhearts and 5 porcine hearts) were used in this study. The epicardial and endocardial boundariesof the ventricles in every 2D slice of the cardiac magnetic resonance with late gadoliniumenhancement images were manually segmented. The subsequent pipeline of infarct tissuesegmentation is fully automatic. The segmentation results with the automatic algorithm proposed inthis paper were compared to the consensus ground truth. The median of Dice overlap between ourautomatic method and the consensus ground truth is 0.79. We also compared the automatic methodwith the consensus ground truth using different image sources from diferent centers with diferentscan parameters and different scan machines. The results showed that the Dice overlap with thepublic dataset was 0.83, and the overall Dice overlap was 0.79. The results show that our method isrobust with respect to different MRI image sources, which were scanned by different centers withdifferent image collection parameters. The segmentation accuracy we obtained is comparable toor better than that of the conventional semi-automatic methods. Our segmentation method may beuseful for processing large amount of dataset in clinic.展开更多
右心室分割对肺动脉高压等疾病的心功能分析具有重要的临床意义.然而,右心室心肌薄、易变且不规则,其传统的医学图像分割方法仍然未能取得突破性进展.本文提出基于COLLATE(Consensus Level, Labeler Accuracy and Truth Estimation)的...右心室分割对肺动脉高压等疾病的心功能分析具有重要的临床意义.然而,右心室心肌薄、易变且不规则,其传统的医学图像分割方法仍然未能取得突破性进展.本文提出基于COLLATE(Consensus Level, Labeler Accuracy and Truth Estimation)的多图谱分割方法,首先以归一化互信息为相似测度对目标图像和图谱集进行B样条配准以获取粗分割结果;然后利用COLLATE对粗分割结果进行融合;最后采用基于形状约束的区域生长算法修正出现错误的数据.10例临床心脏磁共振短轴电影图像被用于算法验证.本文还将使用基于COLLATE的多图谱分割方法得到的结果与深度学习算法及手动分割进行了比较.结果显示与深度学习算法比较,使用本文算法得到的射血分数(Ejection Fraction, EF)与手动分割更加一致和相关,表明该算法的分割结果有望辅助临床心脏功能诊断.展开更多
针对左心室外轮廓类似椭圆的特点,提出了基于椭圆约束的水平集模型,该模型在Chan and Vese模型的基础上增加椭圆形状约束项,来控制曲线的演化,将水平集的演化曲线作为对轮廓新的位置预测,并用椭圆对预测结果进行修正,把预测结果和修正...针对左心室外轮廓类似椭圆的特点,提出了基于椭圆约束的水平集模型,该模型在Chan and Vese模型的基础上增加椭圆形状约束项,来控制曲线的演化,将水平集的演化曲线作为对轮廓新的位置预测,并用椭圆对预测结果进行修正,把预测结果和修正结果分别作为新的水平集曲线和椭圆信息,直到曲线停止演化。实验表明,该方法能够有效地分割心脏外轮廓。展开更多
目的通过研究和搭建人工智能深度学习网络,实现多模态心脏磁共振(cardiac magnetic resonance,CMR)图像分割,并提升Dice系数。材料与方法回顾性分析来自2019年多序列CMR分割挑战赛的公开数据集,它包含了45例患者平衡稳态自由进动(balanc...目的通过研究和搭建人工智能深度学习网络,实现多模态心脏磁共振(cardiac magnetic resonance,CMR)图像分割,并提升Dice系数。材料与方法回顾性分析来自2019年多序列CMR分割挑战赛的公开数据集,它包含了45例患者平衡稳态自由进动(balanced-steady state free precession,bSSFP)模态,晚期钆增强(late gadolinium enhancement,LGE)模态与T2WI模态的CMR图像数据。本文构建了一种新的双流U型网络框架,实现bSSFP与LGE两种模态以及bSSFP与T2WI两种模态的CMR图像分割。在编码阶段,未配准各模态图像被交替地送入各自分支进行特征学习,所获取的特征图接着都流入共享层,实现多模态信息的交互补充,最终共享特征分开流出到各自分支进行解码输出。通过在45例患者的CMR图像数据集上进行五折交叉验证实验,分别对bSSFP与LGE模态、bSSFP与T2WI模态进行了分割,以Dice系数对提出的模型进行性能评估,Wilcoxon符号秩检验被用来检验模型差异性。结果在bSSFP与LGE模态的分割实验中,本文方法在bSSFP模态的平均Dice系数相较于传统UNet模型和最新的Swin-Unet模型都有显著提升(P<0.001);在LGE模态的平均Dice系数较传统UNet模型(P<0.001)、Swin-Unet模型(P=0.001)、双流UNet(P=0.021)均有显著提升。在bSSFP与T2WI模态的分割实验中,本文方法在bSSFP模态的平均Dice系数较UNet模型、Swin-Unet模型与双流UNet均有显著提升(P<0.001);在T2WI模态的平均Dice系数较UNet模型有显著提升(P<0.001),较Swin-Unet模型有提升(P=0.025)。结论本研究提出的双流U型网络框架为CMR图像多模态分割提供有效方法,且该网络提高了CMR图像bSSFP模态与LGE模态及bSSFP模态与T2WI模态的Dice系数,很好地解决了多模态CMR图像个体解剖学差异大和图像间存在灰度不一致问题,提升了模型的泛化能力。展开更多
文摘The segmentation of unlabeled medical images is troublesome due to the high cost of annotation, and unsupervised domain adaptation is one solution to this. In this paper, an improved unsupervised domain adaptation method was proposed. The proposed method considered both global alignment and category-wise alignment. First, we aligned the appearance of two domains by image transformation. Second, we aligned the output maps of two domains in a global way. Then, we decomposed the semantic prediction map by category, aligning the prediction maps in a category-wise manner. Finally, we evaluated the proposed method on the 2017 Multi-Modality Whole Heart Segmentation Challenge dataset, and obtained 82.1 on the dice similarity coefficient and 4.6 on the average symmetric surface distance, demonstrating the effectiveness of the combination of global alignment and category-wise alignment.
基金This work was supported by the Project of Sichuan Outstanding Young Scientific and Technological Talents(19JCQN0003)the major Project of Education Department in Sichuan(17ZA0063 and 2017JQ0030)+1 种基金in part by the Natural Science Foundation for Young Scientists of CUIT(J201704)the Sichuan Science and Technology Program(2019JDRC0077).
文摘Cardiomyopathy is one of the most serious public health threats.The precise structural and functional cardiac measurement is an essential step for clinical diagnosis and follow-up treatment planning.Cardiologists are often required to draw endocardial and epicardial contours of the left ventricle(LV)manually in routine clinical diagnosis or treatment planning period.This task is time-consuming and error-prone.Therefore,it is necessary to develop a fully automated end-to-end semantic segmentation method on cardiac magnetic resonance(CMR)imaging datasets.However,due to the low image quality and the deformation caused by heartbeat,there is no effective tool for fully automated end-to-end cardiac segmentation task.In this work,we propose a multi-scale segmentation network(MSSN)for left ventricle segmentation.It can effectively learn myocardium and blood pool structure representations from 2D short-axis CMR image slices in a multi-scale way.Specifically,our method employs both parallel and serial of dilated convolution layers with different dilation rates to capture multi-scale semantic features.Moreover,we design graduated up-sampling layers with subpixel layers as the decoder to reconstruct lost spatial information and produce accurate segmentation masks.We validated our method using 164 T1 Mapping CMR images and showed that it outperforms the advanced convolutional neural network(CNN)models.In validation metrics,we archived the Dice Similarity Coefficient(DSC)metric of 78.96%.
基金supported by the National Key Researchand Development Program of China(No.2016YFC1301002 to Jianzeng Dong)the National Natural Science Foundation of China(No.81901841 to Dongdong Deng,No.81671650 and No.81971569 to Yi He,No.61527811 to Ling Xia)+1 种基金the Key Research and Development Program of Zhejiang Province(No.2020C03016 to Ling Xia)Dongdong Deng also acknowledges support from Dalian University of Technology(No.DUT18RC(3)068)。
文摘Numerous methods have been published to segment the infarct tissue in theleft ventricle, most of them either need manual work, post-processing, or suffer from poorreproducibility. We proposed an automatic segmentation method for segmenting the infarct tissue irleft ventricle with myocardial infarction. Cardiac images of a total of 60 diseased hearts (55 humanhearts and 5 porcine hearts) were used in this study. The epicardial and endocardial boundariesof the ventricles in every 2D slice of the cardiac magnetic resonance with late gadoliniumenhancement images were manually segmented. The subsequent pipeline of infarct tissuesegmentation is fully automatic. The segmentation results with the automatic algorithm proposed inthis paper were compared to the consensus ground truth. The median of Dice overlap between ourautomatic method and the consensus ground truth is 0.79. We also compared the automatic methodwith the consensus ground truth using different image sources from diferent centers with diferentscan parameters and different scan machines. The results showed that the Dice overlap with thepublic dataset was 0.83, and the overall Dice overlap was 0.79. The results show that our method isrobust with respect to different MRI image sources, which were scanned by different centers withdifferent image collection parameters. The segmentation accuracy we obtained is comparable toor better than that of the conventional semi-automatic methods. Our segmentation method may beuseful for processing large amount of dataset in clinic.
文摘右心室分割对肺动脉高压等疾病的心功能分析具有重要的临床意义.然而,右心室心肌薄、易变且不规则,其传统的医学图像分割方法仍然未能取得突破性进展.本文提出基于COLLATE(Consensus Level, Labeler Accuracy and Truth Estimation)的多图谱分割方法,首先以归一化互信息为相似测度对目标图像和图谱集进行B样条配准以获取粗分割结果;然后利用COLLATE对粗分割结果进行融合;最后采用基于形状约束的区域生长算法修正出现错误的数据.10例临床心脏磁共振短轴电影图像被用于算法验证.本文还将使用基于COLLATE的多图谱分割方法得到的结果与深度学习算法及手动分割进行了比较.结果显示与深度学习算法比较,使用本文算法得到的射血分数(Ejection Fraction, EF)与手动分割更加一致和相关,表明该算法的分割结果有望辅助临床心脏功能诊断.
文摘针对左心室外轮廓类似椭圆的特点,提出了基于椭圆约束的水平集模型,该模型在Chan and Vese模型的基础上增加椭圆形状约束项,来控制曲线的演化,将水平集的演化曲线作为对轮廓新的位置预测,并用椭圆对预测结果进行修正,把预测结果和修正结果分别作为新的水平集曲线和椭圆信息,直到曲线停止演化。实验表明,该方法能够有效地分割心脏外轮廓。
文摘目的通过研究和搭建人工智能深度学习网络,实现多模态心脏磁共振(cardiac magnetic resonance,CMR)图像分割,并提升Dice系数。材料与方法回顾性分析来自2019年多序列CMR分割挑战赛的公开数据集,它包含了45例患者平衡稳态自由进动(balanced-steady state free precession,bSSFP)模态,晚期钆增强(late gadolinium enhancement,LGE)模态与T2WI模态的CMR图像数据。本文构建了一种新的双流U型网络框架,实现bSSFP与LGE两种模态以及bSSFP与T2WI两种模态的CMR图像分割。在编码阶段,未配准各模态图像被交替地送入各自分支进行特征学习,所获取的特征图接着都流入共享层,实现多模态信息的交互补充,最终共享特征分开流出到各自分支进行解码输出。通过在45例患者的CMR图像数据集上进行五折交叉验证实验,分别对bSSFP与LGE模态、bSSFP与T2WI模态进行了分割,以Dice系数对提出的模型进行性能评估,Wilcoxon符号秩检验被用来检验模型差异性。结果在bSSFP与LGE模态的分割实验中,本文方法在bSSFP模态的平均Dice系数相较于传统UNet模型和最新的Swin-Unet模型都有显著提升(P<0.001);在LGE模态的平均Dice系数较传统UNet模型(P<0.001)、Swin-Unet模型(P=0.001)、双流UNet(P=0.021)均有显著提升。在bSSFP与T2WI模态的分割实验中,本文方法在bSSFP模态的平均Dice系数较UNet模型、Swin-Unet模型与双流UNet均有显著提升(P<0.001);在T2WI模态的平均Dice系数较UNet模型有显著提升(P<0.001),较Swin-Unet模型有提升(P=0.025)。结论本研究提出的双流U型网络框架为CMR图像多模态分割提供有效方法,且该网络提高了CMR图像bSSFP模态与LGE模态及bSSFP模态与T2WI模态的Dice系数,很好地解决了多模态CMR图像个体解剖学差异大和图像间存在灰度不一致问题,提升了模型的泛化能力。
