Discrimination of Type 2 diabetic patients from healthy controls by using metabonomics method based on their serum fatty acid profiles

Metabonomics, the study of metabolites and their roles in various disease states, is a novel methodology arising from the post-genomics era. This methodology has been applied in many fields, including work in cardiovascular research and drug toxicology. In this study, metabonomics method was employed to the diagnosis of Type 2 diabetes mellitus (DM2) based on serum lipid metabolites. The results suggested that serum fatty acid profiles determined by capillary gas chromatography combined with pattern recognition analysis of the data might provide an effective approach to the discrimination of Type 2 diabetic patients from healthy controls. And the applications of pattern recognition methods have improved the sensitivity and specificity greatly. (C) 2004 Elsevier B.V. All rights reserved.

Application of probabilistic neural network in the clinical diagnosis of cancers based on clinical chemistry data

As a recently developed and powerful classification tool, probabilistic neural network was used to distinguish cancer patients from healthy persons according to the levels of nucleosides in human urine. Two datasets (containing 32 and 50 patterns, respectively) were investigated and the total consistency rate obtained was 100% for dataset 1 and 94% for dataset 2. To evaluate the performance of probabilistic neural network, linear discriminant analysis and learning vector quantization network, were also applied to the classification problem. The results showed that the predictive ability of the probabilistic neural network is stronger than the others in this study. Moreover, the recognition rate for dataset 2 can achieve to 100% if combining, these three methods together, which indicated the promising potential of clinical diagnosis by combining different methods. (C) 2002 Elsevier Science B.V. All rights reserved.

Decision tree classification of land cover from remotely sensed data

Decision tree classification algorithms have significant potential for land cover mapping problems and have not been tested in detail by the remote sensing community relative to more conventional pattern recognition techniques such as maximum likelihood classification. In this paper, we present several types of decision tree classification algorithms arid evaluate them on three different remote sensing data sets. The decision tree classification algorithms tested include an univariate decision tree, a multivariate decision tree, and a hybrid decision tree capable of including several different types of classification algorithms within a single decision tree structure. Classification accuracies produced by each of these decision tree algorithms are compared with both maximum likelihood and linear discriminant function classifiers. Results from this analysis show that the decision tree algorithms consistently outperform the maximum likelihood and linear discriminant function classifiers in regard to classf — cation accuracy. In particular, the hybrid tree consistently produced the highest classification accuracies for the data sets tested. More generally, the results from this work show that decision trees have several advantages for remote sensing applications by virtue of their relatively simple, explicit, and intuitive classification structure. Further, decision tree algorithms are strictly nonparametric and, therefore, make no assumptions regarding the distribution of input data, and are flexible and robust with respect to nonlinear and noisy relations among input features and class labels.

Fast Haar Transform Based Feature Extraction for Face Representation and Recognition

Subspace learning is the process of finding a proper feature subspace and then projecting high-dimensional data onto the learned low-dimensional subspace. The projection operation requires many floating-point multiplications and additions, which makes the projection process computationally expensive. To tackle this problem, this paper proposes two simple-but-effective fast subspace learning and image projection methods, fast Haar transform (FHT) based principal component analysis and FHT based spectral regression discriminant analysis. The advantages of these two methods result from employing both the FHT for subspace learning and the integral vector for feature extraction. Experimental results on three face databases demonstrated their effectiveness and efficiency.

Supervised Gaussian Process Latent Variable Model for Dimensionality Reduction

The Gaussian process latent variable model (GP-LVM) has been identified to be an effective probabilistic approach for dimensionality reduction because it can obtain a low-dimensional manifold of a data set in an unsupervised fashion. Consequently, the GP-LVM is insufficient for supervised learning tasks (e. g., classification and regression) because it ignores the class label information for dimensionality reduction. In this paper, a supervised GP-LVM is developed for supervised learning tasks, and the maximum a posteriori algorithm is introduced to estimate positions of all samples in the latent variable space. We present experimental evidences suggesting that the supervised GP-LVM is able to use the class label information effectively, and thus, it outperforms the GP-LVM and the discriminative extension of the GP-LVM consistently. The comparison with some supervised classification methods, such as Gaussian process classification and support vector machines, is also given to illustrate the advantage of the proposed method.

Classification of valence changes of trivalent rare earth ions in alkaline earth borates using artificial neural networks

The investigations of classification on the valence changes from RE3+ to RE2+ (RE = Eu, Sm, Yb, Tm) in host compounds of alkaline earth berate were performed using artificial neural networks (ANNs). For comparison, the common methods of pattern recognition, such as SIMCA, KNN, Fisher discriminant analysis and stepwise discriminant analysis were adopted. A learning set consisting of 24 host compounds and a test set consisting of 12 host compounds were characterized by eight crystal structure parameters. These parameters were reduced from 8 to 4 by leaps and bounds algorithm. The recognition rates from 87.5 to 95.8% and prediction capabilities from 75.0 to 91.7% were obtained. The results provided by ANN method were better than that achieved by the other four methods. (C) 1999 Elsevier Science B.V. All rights reserved.

MULTIVARIATE CLASSIFICATION BASED ON METAL CONTENTS IN HUMAN SENILE CATARACT LENSES

Multivariate classification methods were used to evaluate data on the concentrations of eight metals in human senile lenses measured by atomic absorption spectrometry. Principal components analysis and hierarchical clustering separated senile cataract lenses, nuclei from cataract lenses, and normal lenses into three classes on the basis of the eight elements. Stepwise discriminant analysis was applied to give discriminant functions with five selected variables. Results provided by the linear learning machine method were also satisfactory; the k-nearest neighbour method was less useful.

Geochemical character and material source of sediments in the eastern Philippine Sea

Based upon analyses of grain-size, rare earth element (REE) compositions, elemental occurrence phases of REE, and U-series isotopic dating, the sediment characteristics and material sources of the study area were examined for the recently formed deep-sea clays in the eastern Philippine Sea. The analytical results are summarized as follows. (1) Low accumulation rate, poor sorting and roundness, and high contents of grains coarser than fine silt indicate relatively low sediment input, with localized material source without long distance transport. (2) The REE Contents are relatively high. Shale-normalized patterns of REE indicate weak enrichment in heavy REE (HREE), Ce-passive anomaly, and Eu-positive anomaly. (3) Elemental occurrence phases of REE between the sediments with and without crust are similar. REE mainly concentrate in residual phase and then in ferromanganese oxide phase. The light REE (LREE) enrichment, Ce-positive anomaly, and Eu-positive anomaly occur in residual phase. Ferromanganese oxide phase shows the characteristics of relatively high HREE content and Ce-passive anomaly. (4) There are differences in each above mentioned aspect between the sediments with and without ferromanganese crust. (5) Synthesizing the above characteristics and source discriminant analysis, the study sediments are deduced to mainly result from the alteration of local and nearby volcanic materials. Continental materials transported by wind and/or river (ocean) flows also have minor contributions.

Mineral chemistry and its genetic significance of olivine in Cenozoic basalts from the South China Sea

We dredged lots of Cenozoic basalts from areas covered from the northern sub-slope to the southern sub- slope of the South China Sea. Based on the study on mineral chemistry of clinopyroxenes in these Cenozoic hasalts, this paper indicates that pyroxenes are mostly enstatite and a few of augite, sahlite and Ca-rich pyroxene. Pyroxene microlite has higher content in, Ca, Ti and Fe than pyroxene phenocryst, it may reflect that the evolution trend of host magma of pyroxene is coincidence with that of alkali rock series. The depth of magma chambers which calculated from equilibrium temperatures and pressures between clinopyroxene and melt are as follows, that of magma of tephrite is about 49km, that of magma of trachybasalt is about 25km, and that of magma of basalt is about 15km. Correspondingly, Equilibrium temperatures( K) of three types rocks mentioned above gradually decrease from 1535 1498 to 1429 to 1369. By using discriminant plot which developed from pyroxene and alkali discriminant diagram of host rock, Cenozoic basalt from the South China Sea belongs to intraplate alkali basalt. The results suggest that alkali basalt series in the study area may be the products of continuous evolution of mantle plume, which result from some physical and chemistry process including partial melting and fractional crystallization of mantle plume during the course of its ascent to the surface.

潮波模型的优化开边界方法研究

数值模式是潮波研究的一种有利手段，但在研究中会面临各种具体问题，包括开边界条件的确定、底摩擦系数和耗散系数的选取等。数据同化是解决这些问题的一种途径，即利用有限数量的潮汐观测资料对潮波进行最优估计，其根本目的是迫使模型预报值逼近观测值，使模式不要偏离实际情况太远。本文采用了一种优化开边界方法，沿着数值模型的开边界优化潮汐水位信息，目的是设法使数值解在动力约束的意义下接近观测值，获得研究区域的潮汐结果。边界值由指定优化问题的解来定，以提高模拟区域的潮汐精度，最优问题的解是基于通过开边界的能量通量的变化，处理开边界处的观测值与计算值之差的最小化。这里提供了辐射型边界条件，由Reid 和Bodine（本文简称为RB）推导，我们将采用的优化后的RB方法（称为ORB）是优化开边界的特殊情况。 本文对理想矩形海域（ E- E， N- N， 分辨率 ）进行了潮波模拟，有东部开边界，模式采用ECOM3D模式。对数据结果的误差分析采用，振幅平均偏差，平均绝对偏差，平均相对误差和均方根偏差四个值来衡量模拟结果的好坏程度。 需要优化入开边界的解析潮汐值本文采用的解析解由方国洪《海湾的潮汐与潮流》（1966年）方法提供，为验证本文所做的解析解和方文的一致，本文做了其第一个例子的关键值a,b,z,结果与其结果吻合的相当好。但略有差别，分析的可能原因是两法在具体迭代方案和计算机保留小数上有区别造成微小误差。另外，我们取m=20，得到更精确的数值，我们发现对前十项的各项参数值，取m=10,m=20各项参数略有改进。当然我们可以获得m更大的各项参数值。 同时为了检验解析解的正确性讨论m和l变化对边界值的影响，结果指出，增大m，m=20时，u的模最大在本身u1或u2的模的6%；m=100时，u的模最大在本身u1或u2的模的4%；m再增大，m=1000时，u的模最大在本身u1或u2的模的4%，改变不大。当l<1时， =0处u的模最大为2。当l=1时， =0处u的模最大为0.1，当l>1时，l越大，u的模越小，当l=10时，u的模最大为0.001，可以认为为0。 为检验该优化方法的应用情况，我们对理想矩形区域进行模拟，首先将本文所采用的优化开边界方法应用于30m的情况，在开边界优化入开边界得出模式解，所得模拟结果与解析解吻合得相当好，该模式解和解析解在整个区域上，振幅平均绝对偏差为9.9cm,相位平均绝对偏差只有4.0 ，均方根偏差只有13.3cm，说明该优化方法在潮波模型中有效。 为验证该优化方法在各种条件下的模拟结果情况，在下面我们做了三类敏感性试验： 第一类试验：为证明在开边界上使用优化方法相比于没有采用优化方法的模拟解更接近于解析解，我们来比较ORB条件与RB条件的优劣，我们模拟用了两个不同的摩擦系数，k分别为：0,0.00006。 结果显示，针对不同摩擦系数，显示在开边界上使用ORB条件的解比使用RB条件的解无论是振幅还是相位都有显著改善，两个试验均方根偏差优化程度分别为84.3%，83.7%。说明在开边界上使用优化方法相比于没有采用优化方法的模拟解更接近于解析解，大大提高了模拟水平。上述的两个试验得出， k=0.00006优化结果比k=0的好。 第二类试验，使用ORB条件确定优化开边界情况下，在东西边界加入出入流的情况，流考虑线性和非线性情况，结果显示，加入流的情况，潮汐模拟的效果降低不少，流为1Sv的情况要比5Sv的情况均方根偏差相差20cm,而不加流的情况只有0.2cm。线性流和非线性流情况两者模式解相差不大，振幅，相位各项指数都相近, 说明流的线性与否对结果影响不大。 第三类试验，不仅在开边界使用ORB条件，在模式内部也使用ORB条件，比较了内部优化和不优化情况与解析解的偏差。结果显示，选用不同的k,振幅都能得到很好的模拟，而相位相对较差。另外，在内部优化的情况下，考虑不同的k的模式解， 我们选用了与解析解相近的6个模式解的k,结果显示，不同的k,振幅都能得到很好的模拟，而相位较差。 总之，在开边界使用ORB条件比使用RB条件好，振幅相位都有大幅度改进，在加入出入流情况下，流的大小对模拟结果有影响，但线形流和非线性流差别不大。内部优化的结果显示，模式采用不同的k都能很好模拟解析解的振幅。

基于表面肌电信号的人体运动意图识别方法研究

提出了基于表面肌电信号进行人体运动意图识别的新方法。该方法将智能发育思想结合其中,分为离线知识库建立和在线知识库检索两个过程。首先,进行离线训练,建立IHDR(Incremental Hierarchical Discriminant Regression)知识库。采集人体表面肌电信号,对肌电信号进行小波分解,将各层小波系数的方差作为每个通道信号的特征;然后,将每个通道的特征作为输入,人体运动行为作为输出,离线建立基于智能发育思想的IHDR知识库。人体运动意图识别的过程即知识库的检索运用过程,通过每个通道的特征来在线检索IHDR知识库,搜索到合适的节点即为运动意图。实验表明,基于智能发育思想的人体运动意图识别方法可以达到满意的识别率。

改进的统计不相关最优鉴别矢量集

该文对统计不相关最优鉴别矢量集算法进行研究,在分析统计不相关最优鉴别矢量集算法的基础上提出了一种改进的方法。该方法在类内散布矩阵的特征空间中求解统计不相关最优鉴别矢量集。为了加快特征抽取速度,利用基于图像鉴别分析的维数压缩方法,对图像数据进行了压缩。在ORL和Yale人脸数据库的数值实验,验证本文所提出的方法的有效性。

广义统计不相关最优鉴别矢量集的一个理论结果

本文对统计不相关最优鉴别矢量集的理论问题进行研究 ,提出了广义统计不相关最优鉴别准则 ,并给出了广义统计不相关最佳鉴别矢量集的一个理论结果 ,研究表明 ,广义统计不相关最佳鉴别矢量集的计算公式与基于Fisher最优鉴别准则的统计不相关最佳鉴别矢量集的计算公式完全一样 ,但是以前这一点没有被认识到 .本文的研究丰富了统计不相关最优鉴别分析的特征抽取理论 .

基于模糊神经网络的人脸识别方法研究

人脸识别是模式识别研究领域的重要课题,具有广阔的应用前景。本文提出了基于模糊神 经网络的人脸识别方法。首先用最优鉴别分析方法提取人脸的最优鉴别矢量集,构成特征空间,然后在 特征空间中设计模糊神经网络分类器。在ORL人脸图象库上的实验结果表明了该方法的有效性。

统计不相关最佳鉴别矢量集的本质研究

对统计不相关最佳鉴别矢量集的本质进行研究 ,在基于总体散布矩阵特征分解的基础上 ,构造了一种白化变换 ,使得变换后的样本空间中的总体散布矩阵为单位矩阵 ,这样使得传统的最佳鉴别矢量集算法得到的均是具有统计不相关的最佳鉴别矢量集 ,从而揭示了统计不相关最佳鉴别变换的本质———白化变换加普通的线性鉴别变换。该方法的最大优点在于所获得的最优鉴别矢量同时具有正交性和统计不相关性。该方法对代数特征抽取具有普遍适用性。用ORL人脸数据库的数值实验 ,验证了该方法的有效性