Purpose To assess CT perfusion characteristics of focal pancreatic lesions including pancreatic adenocarcinoma(PCA),pancreatic endocrine tumors(PET),and to examine the variability in the perfusion parameters generated...
详细信息
Purpose To assess CT perfusion characteristics of focal pancreatic lesions including pancreatic adenocarcinoma(PCA),pancreatic endocrine tumors(PET),and to examine the variability in the perfusion parameters generated from patlak method(method1)anddeconvolution method(method2),with the identical source data.
Millions of parcels and pieces of luggage are scanned daily for threat detection at border control or high-security buildings. Currently, the process is manually operated by security agents and it is slow and time-con...
详细信息
Millions of parcels and pieces of luggage are scanned daily for threat detection at border control or high-security buildings. Currently, the process is manually operated by security agents and it is slow and time-consuming. The automation of this process will lift the burden from the security agents and will allow larger volumes of items to be scanned. The authors consider the problem of automatic threat detection, in particular firearms, in X-rays. To achieve this goal, they propose a hybrid algorithm that combines two well-established image segmentation algorithms into a two-step clustering method. The first step is a semi-supervised spectral clustering algorithm at the image level, which classifies whole images into benign or containing a threat. The images classified as threatening from the first step proceed to the second stage, where a variational image segmentation algorithm performs clustering at the pixel level to locate the threat if it exists. The hybrid algorithm is designed to scale-up the processing of hundreds of images, in comparison to the academic literature where only a handful images are used for demonstration. Numerical experiments establish that the combination of two different algorithms produces better results than using individual algorithms.
In this study, novel iterative algorithms based on optimisation are developed to solve the continuous-time and discrete-time Sylvester matrix equations. The great difference of the proposed algorithms is that solution...
详细信息
In this study, novel iterative algorithms based on optimisation are developed to solve the continuous-time and discrete-time Sylvester matrix equations. The great difference of the proposed algorithms is that solutions of the equations are updated by two different sequences generated by the proposed algorithms. Convergence rates of the proposed algorithms can be markedly improved by choosing appropriate tuning parameters. Convergence conditions of the proposed algorithms are provided for different cases. Moreover, efficient numerical methods are presented to find the appropriate tuning parameters. Finally, three examples are given to illustrate the effectiveness of the proposed algorithms, and to compare the convergence performance of different algorithms.
We present a family of margin based online learning algorithms for various prediction tasks. In particular we derive and analyze algorithms for binary and multiclass categorization, regression, uniclass prediction and...
详细信息
We present a family of margin based online learning algorithms for various prediction tasks. In particular we derive and analyze algorithms for binary and multiclass categorization, regression, uniclass prediction and sequence prediction. The update steps of our different algorithms are all based on analytical solutions to simple constrained optimization problems. This unified view allows us to prove worst-case loss bounds for the different algorithms and for the various decision problems based on a single lemma. Our bounds on the cumulative loss of the algorithms are relative to the smallest loss that can be attained by any fixed hypothesis, and as such are applicable to both realizable and unrealizable settings. We demonstrate some of the merits of the proposed algorithms in a series of experiments with synthetic and real data sets.
SNPAnalyzer is a software that performs four essential statistical analyses of SNPs in a common computational environment. It is composed of three main modules: (i) data manipulation, (ii) analysis and (iii) visualiza...
详细信息
SNPAnalyzer is a software that performs four essential statistical analyses of SNPs in a common computational environment. It is composed of three main modules: (i) data manipulation, (ii) analysis and (iii) visualization. The data manipulation module is responsible for data input and output, and handles genotype, phenotype and genetic distance data. To ensure user convenience, the data format is simple. The analysis module performs statistical calculations and consists of four subcomponents: (i) Hardy-Weinberg equilibrium, (ii) Haplotype Estimation, (iii) linkage disequilibrium (LD) and (iv) quantitative trait locus analysis. The main feature of the analysis module is multiple implementations of different algorithms and indices for haplotype estimation and for LD analysis. This enables users to compare separate results generated by different algorithms, which help to avoid biased results acquired by applying a single statistical algorithm. The performance of all implemented algorithms has been validated using experimentally proven datasets. The visualization module presents most of the analyzed results as figures, rather than as simple text, which aids in the intuitive understanding of complex data. The SNPAnalyzer has been developed using C and C++ and is available at http://***/istech/board/login_***.
暂无评论