Super-resolution reconstruction algorithm produces a high-resolution imagefrom a low-resolution image sequence. The accuracy and the stability of the motion estimation (ME)are essential for the whole restoration. In t...Super-resolution reconstruction algorithm produces a high-resolution imagefrom a low-resolution image sequence. The accuracy and the stability of the motion estimation (ME)are essential for the whole restoration. In this paper, a new super-resolution reconstructionalgorithm is developed using a robust ME method, which fuses multiple estimated motion vectorswithin the sequence. The new algorithm has two major improvements compared with the previousresearch. First, instead of only two frames, the whole sequence is used to obtain a more accurateand stable estimation of the motion vector of each frame; second, the reliability of the ME isquantitatively measured and introduced into the cost function of the reconstruction algorithm. Thealgorithm is applied to both synthetic and real sequences, and the results are presented in thepaper.展开更多
Although previous studies have made some clear leap in learning latent dynamics from high-dimensional representations,the performances in terms of accuracy and inference time of long-term model prediction still need t...Although previous studies have made some clear leap in learning latent dynamics from high-dimensional representations,the performances in terms of accuracy and inference time of long-term model prediction still need to be improved.In this study,a deep convolutional network based on the Koopman operator(CKNet)is proposed to model non-linear systems with pixel-level measurements for long-term prediction.CKNet adopts an autoencoder network architecture,consisting of an encoder to generate latent states and a linear dynamical model(i.e.,the Koopman operator)which evolves in the latent state space spanned by the encoder.The decoder is used to recover images from latent states.According to a multi-step ahead prediction loss function,the system matrices for approximating the Koopman operator are trained synchronously with the autoencoder in a mini-batch manner.In this manner,gradients can be synchronously transmitted to both the system matrices and the autoencoder to help the encoder self-adaptively tune the latent state space in the training process,and the resulting model is time-invariant in the latent space.Therefore,the proposed CKNet has the advantages of less inference time and high accuracy for long-term prediction.Experiments are per-formed on OpenAI Gym and Mujoco environments,including two and four non-linear forced dynamical systems with continuous action spaces.The experimental results show that CKNet has strong long-term prediction capabilities with sufficient precision.展开更多
目的:探索磁共振成像多点非对称回波采集与迭代最小二乘法水脂分离(iterative decomposition of water and fat with echo asymmetry and the least squares estimation quantification sequence,IDEAL-IQ)和体素内不相干运动(intravoxe...目的:探索磁共振成像多点非对称回波采集与迭代最小二乘法水脂分离(iterative decomposition of water and fat with echo asymmetry and the least squares estimation quantification sequence,IDEAL-IQ)和体素内不相干运动(intravoxel incoherent motion,IVIM)在子宫内膜癌(endometrial carcinoma,EC)细胞增殖状态评估中的价值。方法:回顾并分析24例Ki-67增殖指数低(≤50%)和19例Ki-67增殖指数高(>50%)的EC患者的资料,分别测量病灶IDEAL-IQ成像的脂肪分数(fat fraction,FF)、R_(2)^(*)弛豫率(R_(2)^(*))值和IVIM成像的慢速表观弥散系数(slow apparent diffusion coefficient,ADCslow)、快速表观弥散系数(fast apparent diffusion coefficient,ADC-fast)和灌注分数(perfusion fraction,f)并进行对比。受试者工作特征(receiver operating characteristic,ROC)曲线的曲线下面积(area under curve,AUC)被用于确定各参数的诊断效能,logistic回归和DeLong检验分别被用于多参数联合诊断和不同AUC间的差异分析。Spearman相关被用于评估各参数值与Ki-67增殖指数的相关性。结果:Ki-67增殖指数高的组ADC-slow、ADC-fast和R_(2)^(*)值均显著低于Ki-67增殖指数低的组(P分别为<0.001、0.004、<0.001)。ADC-slow、ADC-fast、R_(2)^(*)以及三者联合鉴别Ki-67增殖指数高、低组EC的AUC分别为0.860、0.748、0.862和0.978。DeLong分析显示,ADC-slow+ADC-fast+R_(2)^(*)与ADC-slow、ADC-fast及R_(2)^(*)之间的AUC差异均有统计学意义(Z分别为2.109、3.134、2.227;P分别为0.035、0.002、0.023)。ADC-slow和R_(2)^(*)值均与Ki-67增殖指数呈中度负相关(r分别为-0.547、-0.711,P<0.001),ADC-fast与Ki-67增殖指数呈轻度负相关(r分别为-0.324,P=0.034)。结论:ADC-slow、ADC-fast和R_(2)^(*)均有助于评估EC患者的细胞增殖状态,且三者联合能够对Ki-67增殖指数高、低组EC进行更有效的鉴别。展开更多
文摘Super-resolution reconstruction algorithm produces a high-resolution imagefrom a low-resolution image sequence. The accuracy and the stability of the motion estimation (ME)are essential for the whole restoration. In this paper, a new super-resolution reconstructionalgorithm is developed using a robust ME method, which fuses multiple estimated motion vectorswithin the sequence. The new algorithm has two major improvements compared with the previousresearch. First, instead of only two frames, the whole sequence is used to obtain a more accurateand stable estimation of the motion vector of each frame; second, the reliability of the ME isquantitatively measured and introduced into the cost function of the reconstruction algorithm. Thealgorithm is applied to both synthetic and real sequences, and the results are presented in thepaper.
基金National Natural Science Foundation of China,Grant/Award Numbers:61825305,62003361,U21A20518China Postdoctoral Science Foundation,Grant/Award Number:47680。
文摘Although previous studies have made some clear leap in learning latent dynamics from high-dimensional representations,the performances in terms of accuracy and inference time of long-term model prediction still need to be improved.In this study,a deep convolutional network based on the Koopman operator(CKNet)is proposed to model non-linear systems with pixel-level measurements for long-term prediction.CKNet adopts an autoencoder network architecture,consisting of an encoder to generate latent states and a linear dynamical model(i.e.,the Koopman operator)which evolves in the latent state space spanned by the encoder.The decoder is used to recover images from latent states.According to a multi-step ahead prediction loss function,the system matrices for approximating the Koopman operator are trained synchronously with the autoencoder in a mini-batch manner.In this manner,gradients can be synchronously transmitted to both the system matrices and the autoencoder to help the encoder self-adaptively tune the latent state space in the training process,and the resulting model is time-invariant in the latent space.Therefore,the proposed CKNet has the advantages of less inference time and high accuracy for long-term prediction.Experiments are per-formed on OpenAI Gym and Mujoco environments,including two and four non-linear forced dynamical systems with continuous action spaces.The experimental results show that CKNet has strong long-term prediction capabilities with sufficient precision.
文摘目的:探索磁共振成像多点非对称回波采集与迭代最小二乘法水脂分离(iterative decomposition of water and fat with echo asymmetry and the least squares estimation quantification sequence,IDEAL-IQ)和体素内不相干运动(intravoxel incoherent motion,IVIM)在子宫内膜癌(endometrial carcinoma,EC)细胞增殖状态评估中的价值。方法:回顾并分析24例Ki-67增殖指数低(≤50%)和19例Ki-67增殖指数高(>50%)的EC患者的资料,分别测量病灶IDEAL-IQ成像的脂肪分数(fat fraction,FF)、R_(2)^(*)弛豫率(R_(2)^(*))值和IVIM成像的慢速表观弥散系数(slow apparent diffusion coefficient,ADCslow)、快速表观弥散系数(fast apparent diffusion coefficient,ADC-fast)和灌注分数(perfusion fraction,f)并进行对比。受试者工作特征(receiver operating characteristic,ROC)曲线的曲线下面积(area under curve,AUC)被用于确定各参数的诊断效能,logistic回归和DeLong检验分别被用于多参数联合诊断和不同AUC间的差异分析。Spearman相关被用于评估各参数值与Ki-67增殖指数的相关性。结果:Ki-67增殖指数高的组ADC-slow、ADC-fast和R_(2)^(*)值均显著低于Ki-67增殖指数低的组(P分别为<0.001、0.004、<0.001)。ADC-slow、ADC-fast、R_(2)^(*)以及三者联合鉴别Ki-67增殖指数高、低组EC的AUC分别为0.860、0.748、0.862和0.978。DeLong分析显示,ADC-slow+ADC-fast+R_(2)^(*)与ADC-slow、ADC-fast及R_(2)^(*)之间的AUC差异均有统计学意义(Z分别为2.109、3.134、2.227;P分别为0.035、0.002、0.023)。ADC-slow和R_(2)^(*)值均与Ki-67增殖指数呈中度负相关(r分别为-0.547、-0.711,P<0.001),ADC-fast与Ki-67增殖指数呈轻度负相关(r分别为-0.324,P=0.034)。结论:ADC-slow、ADC-fast和R_(2)^(*)均有助于评估EC患者的细胞增殖状态,且三者联合能够对Ki-67增殖指数高、低组EC进行更有效的鉴别。