Two dimensional CCA via penalized matrix decomposition for structure preserved fMRI data analysis

Two dimensional canonical correlation analysis (2DCCA) is a data driven method that has been used to preserve the local spatial structure of functional magnetic resonance (fMR) images and to detect brain activation patterns. 2DCCA finds pairs of left and right linear transforms by directly operating on two dimensional data (i.e., image data) such that the correlation between their projections is maximized without neglecting the local spatial structure of the data. However, in the context of high dimensional data, the performance of 2DCCA suffers from interpretability of learned projection variables. In this study, to improve the interpretability of projection variables while preserving the local spatial structure of fMR images, we propose two new 2DCCA approaches, sparse 2DCCA and regularized 2DCCA. The proposed algorithms aim at improving the activation detection performance in terms of specificity of activated voxels by directly operating on image data without rearranging fMRI slices in 1D-vectors. The validity of the proposed algorithms has been evaluated on synthetic and real fMRI datasets and it has been shown that the proposed algorithms produce activation maps with higher specificity of activated voxels compared with CCA, 2DCCA, and existing sparse 2DCCA (S2DCCA).     

A novel method for spatio-temporal pattern analysis of brain fMRI data

1 Introduction Functional magnetic resonance imaging (fMRI) is a non-invasive brain imaging technique which has been utilized in brain function researches since the early 1990s[1]. However, it is often difficult to do analysis in fMRI data because of the low signal to noise ratio (SNR) (about 2%—4% with 1.5T magnetic field strength) and the delay within the true neural activity and the stimuli-induced signal dynamic responses. The prevalent methods applied to fMRI data could be divided i…     

基于GPU加速的定向凝固相场模拟计算研究

相场法作为一种极具优势的微观组织数值模拟方法,已经在凝固微观组织演化机制的研究中得到了广泛应用。然而无论是从计算尺度还是微观组织演化时间上考虑,相场模拟计算量均非常大,对计算机有着非常高的要求。相对于传统的中央处理器(CPU)计算,图形处理器(GPU)计算是最近发展的一种高效计算手段。提出了一种基于GPU加速的定向凝固相场模拟计算策略,实现了大尺度条件下的定向凝固界面形态演化的加速计算。计算结果表明,对于单个计算机,GPU计算与CPU计算的加速比可以高达30余倍。GPU加速将为相场模拟的发展及应用带来新的契机。     

基于一般线性模型的功能磁共振成像回归量正交化

针对功能磁共振成像（fMRI）模型回归量之间存在共线性的问题,提出了一种正交化的方法。首先,确定感兴趣以及待正交的回归量;其次,从待正交回归量中减去与感兴趣回归量相关的部分,使模型中共线的回归量正交分解为相互独立的部分,以此来消除共线性的影响。此外,还讨论和分析了正交化对一般线性模型的影响。最后,分别使用一些合成数据和当前一个流行的fMRI数据分析软件包——脑功能磁共振图像软件包（FSL）进行实验。实验结果表明,正交化方法可以消除模型中的共线性,并且提高感兴趣回归量的显著性,从而实现准确的脑功能定位,可以应用于对脑的基础研究和临床治疗。     

基于CUDA的汇流分析并行算法的研究与实现*

针对基于数字高程模型（DEM）生成流域等流时线的快速运算问题,提出了一种基于统一设备计算架构（CUDA）平台同时可发挥图形处理器（GPU）并行运算特性的汇流分析的快速并行算法。采用改进后的归并排序算法进行数据排序及新的内存分配策略和改进的并行算法进行汇流分析。用该并行算法和CPU上的串行算法, 对生成基于DEM的等流时线运算时间和矩阵乘法运算时间进行分析验证。实验结果表明,基于CUDA的汇流分析并行算法能提高系统的计算效率,具有较好的效果。     

自适应阈值化功能磁共振统计参数映射图

为了准确检测及定位功能激发信息,需选择一个客观、有效的阈值来阈值化功能磁共振统计参数映射图.为此提出了一种组合控制错误发现率及分析皮层血流动力学响应的空间相关性阈值化功能磁共振统计参数映射图的方法.该方法首先采用基于控制错误发现率的方法确定阈值,进行激发体素判别,然后分析已判别为激发的体素与其三维空间26-邻域体素的血流动力学响应的相关性,并进行空间相关检验.该方法不仅能够自适应地选取阈值,而且能够识别由于随机因素而导致的伪激发体素,具有更好脑功能激发信息检测及空间定位能力.     

基于fMRI的思维数据分析方法研究

利用功能磁共振成像(functional magnetic resonance imaging, fMRI)技术解读思维数据,已经实现大脑活动的功能定位,但是大脑的思维过程具体如何运行还不得而知;利用何种分析方法更能揭示思维过程也待进一步研究.采用解决4×4数独问题和图像情感反应的两种刺激任务获取思维过程数据,来分别解读不同的思维状态,探索适用于不同思维数据的分析方法.实验数据证明t-test的特征选择方法、峰值所在时间点的特征提取的方法和SVM分类算法较适用于分析这两种不同思维状态的fMRI数据,揭示正确的思维状态.     

A theorem on the generalized canonical projective vectors

This paper proposes a kind of generalized canonical projective vectors (GCPV), based on the framework of canonical correlation analysis (CCA) applying image recognition. Apart from canonical projective vectors (CPV), the process of obtaining GCPV contains the class information of samples, such that the combined features extracted according to the basis of GCPV can give a better classification performance. The experimental result based on the Concordia University CENPARMI handwritten Arabian numeral database has proved that our method is superior to the method based on CPV.     

Orthogonal canonical correlation analysis and applications

Canonical correlation analysis (CCA) is a cornerstone of linear dimensionality reduction techniques that jointly maps two datasets to achieve maximal correlation. CCA has been widely used in applications for capturing data features of interest. In this paper, we establish a range constrained orthogonal CCA (OCCA) model and its variant and apply them for three data analysis tasks of datasets in real-life applications, namely unsupervised feature fusion, multi-target regression and multi-label classification. Numerical experiments show that the OCCA and its variant produce superior accuracy compared to the traditional CCA.     

基于CUDA的并行加速渲染算法

GPU可以快速有效的处理海量数据,因此在近些年成为图形图像数据处理领域的研究热点。针对现有GPU渲染中在处理含有大量相同或相似模型场景时存在资源利用率低下和带宽消耗过大的问题,在原有GPU渲染架构的基础上提出了一种基于CUDA的加速渲染方法。在该方法中,根据现有的GPU渲染模式构建对应的模型,通过模型找出其不足,从而引申出常量内存的概念;然后分析常量内存的特性以及对渲染产生的作用,从而引入基于常量内存控制的方法来实现渲染的加速,整个渲染过程可以通过渲染算法进行控制。实验结果表明,该方法对解决上述问题具有较好的效果,最终实现加速渲染。     

基于栈式循环神经网络的血液动力学状态估计方法

利用fMRI数据准确地估计血液动力学状态, 能得到一种更接近神经元层面的大脑活动的客观表示, 这将促进人们对大脑运行机理的深刻理解, 推动脑认知的进一步发展.迄今为止, 人们已经提出了许多血液动力学状态估计方法.然而, 这些方法大都只考虑了相邻时刻血液动力学状态之间的关系, 忽视了更深层次的时序特征.而对模型参数先验信息的需求也使一些方法在实际应用中受到了限制.为此, 本文提出了一种基于循环神经网络的血液动力学状态估计新方法.首先, 利用血液动力学模型中非线性函数的反函数建立BOLD信号与血液动力学状态之间的映射关系, 并构建模型的反演过程.然后, 采用一种堆叠三个RNN模块的栈式神经网络结构来拟合这种映射关系, 使其能够以BOLD信号作为输入, 得到血液动力学状态的估计值.最后, 在仿真数据上验证新方法的性能.实验结果表明:与一些代表算法相比, 新方法能够更合理地提取fMRI数据中的时间特性, 有效地拟合BOLD信号与血液动力学状态之间的动态非线性关系.     

基于CUDA的GMM模型快速训练方法及应用

由于能够很好地近似描述任何分布,混合高斯模型(GMM)在模式在识别领域得到了广泛的应用.GMM模型参数通常使用迭代的期望最大化(EM)算法训练获得,当训练数据量非常庞大及模型混合数很大时,需要花费很长的训练时间.NVIDIA公司推出的统一计算设备架构(Computed unified device architecture,CUDA)技术通过在图形处理单元(GPU)并发执行多个线程能够实现大规模并行快速计算.本文提出一种基于CUDA,适用于特大数据量的GMM模型快速训练方法,包括用于模型初始化的K-means算法的快速实现方法,以及用于模型参数估计的EM算法的快速实现方法.文中还将这种训练方法应用到语种GMM模型训练中.实验结果表明,与Intel DualCore PentiumⅣ3.0 GHz CPU的一个单核相比,在NVIDIA GTS250 GPU上语种GMM模型训练速度提高了26倍左右.     

基于同步EEG-fMRI采集的情绪认知重评数据特征融合分析研究

脑电(Electroencephalography, EEG)与功能磁共振成像(Functional magnetic resonance imaging, fMRI)为脑科学研究提供了互补的时空信息. 为研究大脑在对情绪图片采取认知重评策略时的神经活动, 基于同步采集的EEG-fMRI数据, 应用典型相关分析、经验模态分解及k-均值聚类等算法对融合情绪数据进行交叉关联和盲源分离, 得到空间上的fMRI图像和与之对应的EEG时间演变信号. 结果表明: 时域上, CCA分离出的脑电成分在认知重评状态下有明显的晚期正电位(Late positive potential, LPP) (潜伏期200ms～900ms)出现, 而且认知重评策略诱发下的LPP 波幅明显小于观看负性诱发的LPP波幅(F(1, 224)= 28.72, P<0.01), 而大于观看中性诱发的LPP波幅(F(1, 224)= 63.32, P<0.01); 与之对应的空域上, 可以明显地看出和情绪调节相关的扣带回, 额叶、颞叶等区域有明显激活区, 采用情绪认知重评策略时的脑区激活强度明显小于观看负性状态, 而大于观看中性, 且观看中性状态下被激活的与情绪相关的区域相对较少. 研究表明, 这种融合数据分析技术通过计算两种模态数据之间潜在的线性相关性, 可以有效地分离出大脑在时空上神经活动情况, 达到了同时描绘出大脑神经活动的时间信息与空间信息的效果.     

基于CUDA的并行粒子群优化算法研究及实现

应用图形处理器(GPU)来加速粒子群优化(PSO)算法并行计算时,为突出其加速性能,经常有文献以恶化CPU端PSO算法性能为代价。为了科学比较GPU-PSO算法和CPU-PSO算法的性能,提出用"有效加速比"作为算法的性能指标。文中给出的评价方法不需要CPU和GPU端粒子数相同,将GPU并行算法与最优CPU串行算法的性能作比较,以加速收敛到目标精度为准则,在统一计算设备架构(CUDA)下对多个基准测试函数进行了数值仿真实验。结果表明,在GPU上大幅增加粒子数能够加速PSO算法收敛到目标精度,与CPU-PSO相比,获得了10倍以上的"有效加速比"。     

A framework for joint estimation of age,gender and ethnicity on a large database

Human age, gender and ethnicity are valuable demographic characteristics. They are also important soft biometric traits useful for human identification or verification. We present a framework that can estimate the three traits jointly. The joint estimation framework could deal with the mutual influence of age, gender, and ethnicity implicitly. Under this joint estimation framework, we explore different methods for simultaneous estimation of age, gender, and ethnicity. The canonical correlation analysis (CCA) based methods, and partial least squares (PLS) models are explored under our joint estimation framework. Both the linear and nonlinear methods are investigated to measure the performance. We also validate some extensions of these methods, such as the least squares formulations of the CCA methods. We found some consistent ranking of these methods under our joint estimation framework. More importantly, we found that the CCA based methods can derive an extremely low dimensionality in estimating age, gender and ethnicity. An analysis of this property is given based on the rank theory. The experiments are conducted on a very large database containing more than 55,000 face images.     

High-performance forward error correction: Enabling multi-gigabit flows and beyond on commodity GPU and CPU hardware in presence of packet loss

In demanding real-time multimedia transmissions, even a small packet loss might significantly degrade the visual quality. As retransmission is not an option in real-time transfers especially when transmitting the data over long distances, it is necessary to employ mechanisms of Forward Error Correction (FEC). Low-Density Generator Matrix (LDGM) codes are known to be suitable for coding on large block sizes, however, high bitrates of currently used video formats (FullHD, 4K) also require high throughput of FEC coding and decoding. We propose a parallel design of LDGM encoding and decoding algorithms suitable for off-the-shelf, (massively) parallel platforms, such CPUs with vector units or GPUs, and evaluate our approach in real-world scenarios of high-definition and 4K video transmissions. Our results show that offloading FEC computation to such platform is beneficial for low-latency, high-quality multimedia transmissions and may even enable transmissions beyond 10Gbps once the commodity network interfaces reach this speed.     

基于模糊CCA的图像特征提取和识别

利用典型相关分析(CCA)和隶属度的思想,提出一种基于模糊典型相关分析的图像特征提取新方法。通过分析图像样本的分布特点,定义合适的隶属度函数描述图像空间的样本分布。利用CCA进行多信息源特征提取,得到同时包含图像灰度信息和分布信息的有效判别特征。可证明Fisher线性判别分析是该算法的一种极限情形。在ORL标准人脸数据库上的实验结果表明新特征具有良好的分类能力,证实了该方法的有效性。     

基于高阶最小生成树的脑网络分析及对阿兹海默氏症患者的分类

利用静息态功能磁共振成像技术来研究大脑的功能连接网络是当前脑疾病研究的重要方法之一。这种方法能准确地检测包括阿兹海默氏症在内的多种脑疾病。然而,传统的网络只是研究两个脑区之间相关程度,而且缺乏对大脑区域之间更深层次的交互信息和功能连接之间关联程度的研究。为了解决这些问题,提出了一种构建高阶最小生成树功能连接网络的方法,该方法不仅保证了功能连接网络的生理学意义,而且研究了网络中更复杂的交互信息,提高了分类的准确率。分类结果显示,基于高阶最小生成树功能连接网络的静息态功能磁共振成像分类方法大幅提高了阿兹海默氏症检测的准确率。     

Probe test yield optimization based on canonical correlation analysis between process control monitoring variables and probe bin variables

Process control monitoring (PCM) data provide information that is used to track abnormal processes and estimate various probe bin yields. However, multi-dimensional information has not yet been fully utilized from both PCM data and probe bins. In this paper, we proposed a canonical correlation analysis in order to investigate the relationship between multiple PCM variables and various probe bin variables. Polynomial regression was also employed as a methodology for maximizing the performance yield based on the results of the canonical correlation analysis. Two conclusions were drawn from the results of this research. First, the PCM variables that affected the probe bins were contact resistance, sheet resistance, and Isat_P4H as well as threshold voltage (Vt) during the process tuning step. Second, the typical values of Vtl_P4H and Isat_P4H should be changed in order to maximize the performance yield. The proposed method can be used for yield improvement and as a problem-solving approach for optimizing the IC process.     

深度学习与生物医学图像分析2020年综述

医学大数据主要包括电子健康档案数据（electronic health record,EHR）、医学影像数据和基因信息数据等,其中医学影像数据占现阶段医学数据的绝大部分。如何将医学大数据应用于临床实践?这是计算机科学研究人员非常关注的问题,医学人工智能提供了一个很好的答案。通过结合医学图像大数据分析方向截至2020年的最新研究进展,以及医学图像大数据分析领域最近的工作,梳理了当前在医学图像领域以核磁共振影像、超声影像、病理和电信号为代表的4个子领域以及部分其他方向使用深度学习进行图像分析的方法理论和主要流程,对不同算法进行结果评价。本文分析了现有算法的优缺点以及医学影像领域的重难点,介绍了智能成像和深度学习在大数据分析以及疾病早期诊断领域的应用,同时展望了本领域未来的发展热点。深度学习在医学影像领域发展迅速,发展前景广阔,对疾病的早期诊断有重要作用,能有效提高医生工作效率并减轻负担,具有重要的理论研究和实际应用价值。