融合隐向量对齐和Swin Transformer的OCTA血管分割

目的 光学相干断层扫描血管造影（optical coherence tomography angiography,OCTA）是一种非侵入式的新兴技术,越来越多地应用于视网膜血管成像。与传统眼底彩照相比,OCTA技术能够显示黄斑周围的微血管信息,在视网膜血管成像邻域具有显著优势。临床实践中,医生可以通过OCTA图像观察不同层的血管结构,并通过分析血管结构的变化来判断是否存在相关疾病。大量研究表明,血管结构的任何异常变化通常都意味着存在某种眼科疾病。因此,对OCTA图像中的视网膜血管结构进行自动分割提取,对众多眼部相关疾病量化分析和临床决策具有重大意义。然而,OCTA图像存在视网膜血管结构复杂、图像整体对比度低等问题,给自动分割带来极大挑战。为此,提出了一种新颖的融合隐向量对齐和Swin Transformer的视网膜血管结构的分割方法,能够实现血管结构的精准分割。方法 以ResU-Net为主干网络,通过Swin Transformer编码器获取丰富的血管特征信息。此外,设计了一种基于隐向量的特征对齐损失函数,能够在隐空间层次对网络进行优化,提升分割性能。结果 在3个OCTA图像数据集上的...     

EMT 系统在磁性催化物分布检测中的应用

研究了电磁层析成像技术(EMT)在生物柴油制备过程中磁性催化物的分布检测上的应用。为了能够有效应用磁性催化物的导磁特性,在一种8线圈TMR-线圈结构的传感器阵列的基础上设计了仿真实验,比较了不同粒径大小的催化剂分布的成像结果,研究表明,该线圈结构适用于磁性催化物的分布检测。为了改进传统成像算法成像质量差、成像速度慢的缺点,研究了匹配追踪算法(CoSaMP)在图像重建上的应用,并与传统的Tikhonov算法和Landweber算法进行比较,结果表明CoSaMP算法有更好的边缘处理能力和更快的成像速度,与其他两种算法相比,图像误差平均降低了30.4%,成像速度提高了46%,有效提高了重建图像质量与速度。     

自适应加权的多特征融合ECT图像重建算法

为了解决电容层析成像(electrical cpacitance tomography, ECT)图像重建中电容值与介电常数这一非线性病态问题,提出一种自适应加权的多特征融合(adaptive weighted multi-feature fusion, AWMF) ECT图像重建算法,利用网络模型拟合电容张量和介电常数的非线性映射关系。首先,在网络模型中采用密集卷积网络(DenseNet),不仅缓解梯度消失现象,还融合不同通道的特征信息;添加挤压激励网络(squeeze excitation network, SENet)自适应调整特征通道的权重,用以提取不同通道的关键特征,提高重建图像的精度。其次,构建树形聚合结构（tree aggregation structure network, TASN）网络模块,扩大感受野并提取丰富的多尺度特征信息,消除普通卷积所带来的伪影现象。在COMSOL5.3软件上建模仿真后,通过MATLAB2014a对图像进行重建。实验结果表明,重建图像误差系数降低至0.0256,相关系数提高至0.9717,与传统算法和CNN算法相比,具有更高的图像重建质量。     

InPBi薄膜材料的结构性能和光学性能研究

室温下,InPBi表现出强而宽的光致发光光谱,其宽光谱特性来自于材料中的PIn反位深能级和与Bi相关的深能级。该特性使得InPBi有希望应用于制备光学相干层析扫描系统中的超辐射光源。文章利用透射电子显微镜和三维原子探针研究了InPBi薄膜材料的结构性能,发现Bi原子在InPBi薄膜中的分布极不均匀,在InPBi/InP界面出现了Bi的富集区,从该区域沿[001]方向出现了Bi的纳米面,此纳米面位于(110)平面上。这种Bi原子的富集分布阻碍了PIn反位参与的载流子复合过程,对InPBi的光学性能有显著的影响。研究结果可为制造光学相干层析扫描系统的超辐射发光二极管提供一定的理论基础。     

基于高密度子网格光线追迹的三维发射层析重建算法

三维(three- dimensional,3D) 发射层析技术(emission computerized tomography,ECT) 是一种简单、高效且准确的燃 烧场3D成像与检测技术,其中权重矩阵的计算 精度决定了层析重建的精度和质量。本文研究了一种基于高密度子网格光线追迹的权重矩阵 计算方法, 将被测区域划分为密度更高的子网格,并根据相机成像模型实现光线追迹,以确定离散网格 对投影像素 的权重因子。数值模拟和燃烧火焰重建实验表明该算法具有较高的精度和计算效率。该研究 对于3D发射层析技术的实用化具有重要的理论意义。     

Dynamic estimation of the modeling error statistics in Diffuse Optical Tomography

Diffuse Optical Tomography (DOT) is a non-invasive imaging technique that suffers from a typical large-scale and ill-posed inverse problem with low spatial resolution. In DOT, the inverse problem is computationally intensive and decreasing the computation complexity and making it well-posed is the one of the most challenging research areas. More precisely, one of the well-known complexity reduction techniques is defined as applying modelling error originated from discretization of forward problem. Applying the discretization error in Bayesian inference has already been discussed; the method in which the likelihood is modified by an off-line prior density estimation. This paper implements a new method to enhance the modelling error approach using an iterative scheme to update statistical parameters of modelling discrepancy in DOT. The algorithm is very similar to Ensemble Kalman Filter. Moreover, the reconstruction process in the applied method is conducted by a small sample size rather than off-line method. Hence, the computation complexity is decreased and the algorithm converges in few iterations. The efficiency of the proposed method is illustrated by simulations.     

Bragg Edge Neutron Transmission Strain Tomography in Granular Systems

The advent of pixelated detectors for time‐of‐flight neutron transmission experiments has raised significant interest in terms of the potential for tomographic reconstructions of triaxial strain distributions. A recent publication by Lionheart and Withers [WRB Lionheart and PJ Withers, “Diffraction tomography of strain”, Inverse Problems, v31:045005, 2015] has demonstrated that reconstruction is not possible in the general sense; however, various special cases may exist. In this paper, we outline a process by which it is possible to tomographically reconstruct average triaxial elastic strains within individual particles in a granular assembly from a series of Bragg edge strain measurements. This algorithm is tested on simulated data in two and three dimensions and is shown to be capable of rejecting Gaussian measurement noise. Sources of systematic error that may present problems in an experimental implementation are briefly discussed.     

Elemental partitioning,lattice misfit and creep behaviour of Cr containing γ′ strengthened Co base superalloys

Novel Cr containing Co-Al-W base superalloys were studied by atom probe tomography and neutron diffraction. Cr is found to predominantly partition to the γ matrix and decrease partitioning of W to γ′. Furthermore, Cr significantly enhances the γ′ volume fraction, decreases the γ/γ′ lattice misfit and deteriorates the creep resistance. Addition of Ni to the Cr containing alloys affects partitioning of W and Al, further decreases the lattice misfit and results in the formation of irregularly shaped precipitates. Al, W and Cr tend to occupy the ‘B'sublattice in the γ′-A₃B phase (L1₂ type), while Co and Ni reside in the ‘A' sublattice.