图像重建的统计自适应子集算法

有序子集算法大大提高了图像重建迭代算法的收敛速度.增加子集的数目可以加速收敛,但图像质量会由于子集内缺少统计信息而下降.提出了一种基于假设检验的子集划分方法,建立了检验统计量,给出了算法的迭代公式.该算法可以根据用户定义的显著性水平,在每次迭代时自动调节子集的个数,生成含有相同统计信息量的子集.实验结果表明:该方法可以在少数次迭代后得到较高质量的重建图像.     

Changes in percentage of lymphocyte subsets after ^131I treatment in patients with differentiated thyroid cancer

To monitor the extent and the duration of lymphocyte subset changes in patients with thyroid carcinoma undergoing therapeutic ^131I administration, the percentage of lymphocyte subsets were serially analyzed before and after ^131I treatment. In patients who received 1850 MBq of ^131I for ablation of thyroid remnants, only tor NK cells and B cells showed a significant reduction. In patients received 3700 MBq of ^131I for treatment of local lymph node metastases, NK cells, B cells and CD4＋ were found decreased. In patients received 7400 MBq of ^131I for treatment of distant metastases, NK cells, B cells and CD4＋ and CD8＋ were all affected. However, there is no significant reduction compared to the baseline in the percentage of all lymphocyte subsets three months after ^131I treatment. The results show that the sensitivity of lymphocytes to ^131I internal radiation depends upon lymphocyte phenotype and ^131I activity. The immunosuppression effects are temporary and reversible.     

知识表达系统函数规则简化的理论基础

0 引言粗集理论是波兰的Z.Pawlak等一批科学家提出来的,被用来研究不完整数据、不精确知识的表达、学习、归纳等。有些学者认为它无论在理论和应用上都是一种新的、最重要的并且是迅速发展的研充领域,为机器学习、知识获取、决策分析、数据库知识发现、专家系统、决策支持系统、归纳推理、模式识别、模糊控制等提供了一种很有效的数学方法。我们的研     

Classification: Methods for the Exploratory Analythi of Multivariate Data

A fuzzy clustering strategy is used to identify subsets of influential observations in regression. As part of the fuzzy clustering strategy, the analyst can explore the uniqueness of selected subsets and the degree of belonging of observations to selected subsets. This is accomplished by either varying a fuzzy parameter or the number of clusters. Once the observations or subsets have been identified, the analyst can then compute regression diagnostics to confirm their degree of influence in regression. Bootstrapping and high-breakdown procedures were used to confirm the influence of the previously identified subsets. This fuzzy clustering strategy is applied to the modified data on wood-specific gravity and an augmented production dataset. Both datasets have been previously presented in the literature.     

The parallel odd-even merge algorithm

This paper briefly describes the implementation of the odd-even merge algorithm on a parallel MIMD computer and discusses its computational complexity.     

CD34⁺ CD38^- subpopulation without CD123 and CD44 is responsible for LSC and correlated with imbalance of immune cell subsets in AML

Acute myeloid leukemia (AML) is regarded as a stem cell disease. However, no one unique marker is expressed on leukemia stem cells (LSC) but not on leukemic blasts nor normal hematopoietic stem cells (HSC). CD34⁺ CD38- with or without CD123 or CD44 subpopulations are immunophenotypically defined as putative LSC fractions in AML. Nevertheless, markers that can be effectively and simply held responsible for the intrinsical heterogeneity of LSC is still unclear. In the present study, we examined the frequency of three different LSC subtypes (CD34⁺ CD38^-, CD34⁺ CD38^- CD123⁺ , CD34⁺ CD38^- CD44⁺ ) in AML at diagnosis. We then validated their prognostic significance on the relevance of spectral features for diagnostic stratification, immune status, induction therapy response, treatment effect maintenance, and long-term survival. In our findings, high proportions of the above three different LSC subtypes were all significantly characterized with low complete remission (CR) rate, high relapse/refractory rate, poor overall survival (OS), frequent FLT3-ITD mutation, the high level of regulatory T cells (Treg) and monocytic myeloid-derived suppressor cells (M-MDSC). However, there was no significant statistical difference in all kinds of other clinical performance among the three different LSC groups. It was demonstrated that CD34⁺ CD38^- subpopulation without CD123 and CD44 might be held responsible for LSC and correlated with an imbalance of immune cell subsets in AML.     

Actions,finite state sets and applications to trees

Let M be a monoid acting on a set X; for any x?X and A?X we put x ^?1 A = {m?M/xm?A}. Call A?X finite state if card {x ^?1 A/x?X} <∞.

The finite state subsets of T _Σvia the action T _Σ × P _Σ→T _Σare the recognizable forests (P _Σis the monoid of all Σ-trees with just one leaf labeled by a variable x).

Next we prove that the recognizability of forests is equivalent to the finiteness of a certain “syntactic” monoid A Mezei's-like theorem for trees is established: the finite state subsets of T _Σ × T_г are exactly the finite unions of sets of sets of the form B × C B?Rec(T _Σ) and C?Rec(T _г) Another characterization of such relations is given using bimorphisms.     

基于二进制的关联规则挖掘算法

将二进制引入关联规则求解中,充分利用二进制操作方便、运算速度快、节省空间的优势.在求解事务项集真子集和支持度时,对事务数据库中相同事务只求解一次,并给出了真子集的具体求解算法.本算法一次扫描数据库可以挖掘出所有频繁集,而且可以根据需求对最小支持度和最小置信度进行修改,修改后不需要再次扫描数据库即可求出频繁项集,大大提高了挖掘效率.     

凸算法在TCT统计重建中的应用

TCT(Transmission Computed Tomograhy)统计迭代重建算法是目前研究的一个热点,相比ECT统计重建算法来讲,TCT统计重建算法的似然函数比ECT的形式更加复杂、更不易求解.为了简化该问题的求解,本文从优化转化的思想出发,结合凸不等式的性质,寻找了一个易于求解、分离变量的代理函数代替原来的似...     

All‐subsets regression under effect heredity restrictions for experimental designs with complex aliasing

We consider an all‐subsets regression method for models under effect heredity restrictions for experimental designs with complex aliasing, whose number of potential main effects and two‐factor interactions exceed the number of runs. In this paper, we present an algorithm that systematically attempts to fit all such models. We illustrate the algorithm with two published experiments. Copyright © 2009 John Wiley & Sons, Ltd.