线性映射函数对无损颅内压估计的影响

提出了一种基于截断奇异值分解的线性映射函数颅内压无损估计方法。在基于时间序列的数据挖掘框架基础上,分别推导了基于总体最小二乘法的线性映射函数和基于截断奇异值分解法的线性映射函数,通过实验仿真比较证明:不同的线性映射函数对颅内压的预测确实有较为明显的影响,在所采用的方法中,基于截断奇异值分解法的线性映射函数效果明显优于基于总体最小二乘法的线性映射函数。因此,在该数据挖掘框架中,选择合适的线性映射函数对无损颅内压预测精确性的提高有着非常重要的意义。     

An application of cartographic area interpolation to German administrative data

In many situations the applied researcher wishes to combine different data sources without knowing the exact link and merging rule. This paper considers different cartographic interpolation methods for interpolating attributes from German employment office districts to German counties and vice versa. In particular, we apply dasymetric weighting as an alternative to simple area weighting, both of which are based on estimated intersection areas. We also present conditions under which the choice of interpolation method does not matter and confirm the theoretical results with a simulation study. Our application to German administrative data suggests robustness of estimation results of interpolated attributes with respect to the choice of interpolation method. We provide weighting matrices for regional data sources of the two largest German data producers.     

一种无接缝纹理映射算法的并行实现

随着多核处理器的问世,并行编程变得简单和必要。通过对一种纹理映射算法的研究,对其耗时部分进行并行化,以加快算法运行速度。该纹理映射算法先建立三维网格顶点和视点之间的对应关系,然后利用重心坐标权重插值的颜色填充方法构造出一个无接缝的纹理图像。通过利用算法过程中存在的并行性,对算法进行并行实现。实验结果显示,能够得到一个无接缝的纹理图像,而且在双核处理器上的平均加速比达到了1.70。     

基于MPI的进程拓扑感知映射研究

本文首先介绍了获取应用程序通信拓扑和物理拓扑的方法,结合进程间通信量和通信频次定义了进程间亲和度,并针对具有分层网络、NUMA效应的硬件系统,设计了相应的进程映射算法.该算法根据进程间的亲和度对进程集合进行划分,在划分过程中构造出和物理拓扑树同构的树状结构,随后在两棵同构树的对应节点上做映射来解决应用程序进程到物理节点、处理核心的映射问题.在一个多核集群系统上的实验表明,使用本文提出的进程映射方法可以有效提高并行应用程序的通信效率,显著减少程序的总体执行时间.     

Soft error susceptibility mapping of DRAMs using a high-energy nuclear microprobe

Soft errors in 16 Mbit dynamic random access memories (DRAMs) have been investigated using proton microprobes at 400 keV with a spot size of 1 × 1 μm². The newly developed susceptibility mapping can reveal the correlation between the particle hit-position position and the susceptibility to soft errors in a DRAM. The cell-mode soft-errors were found to take place by the incidence of ions within 6 μm around a monitored cell. These errors would be induced by minority carrier diffusion in a lateral direction. This result manifests the possibility of multiple-bit errors by the incidence of an energetic particle.     

The association of genes for purple coleoptile with chromosomes of the wheat variety Mironovskaya 808

Summary The association of genes for purple pigment in the coleoptile with the chromosomes of the winter wheat variety Mironovskaya 808 was investigated using monosomic F₂ analysis. The segregation ratio for F₂ hybrids of Chinese Spring monosomics x Mironovskya 808 seems to indicate that the purple colour of the coleoptile is determined by two dominant genes, Rc₃ and Rc₄, which are located on the chromosomes 7D and 6B respectively, and which reinforce each other. Apart from these two genes, suppressors found on the chromosomes 2A, 2B, 2D, 4B and 6A also play a role in the intensity of the purple colour.With the aid of a Chinese Spring telocentric chromosome marker it was observed that the Rc₃ gene is located on the chromosome arm 7DS, at a distance of 16±4.23 crossover units from the centromere.     

Familial steroid-sensitive nephrotic syndrome in Southern Israel: clinical and genetic observations

Reports on genetically informative steroid-responsive (sensitive) idiopathic nephrotic syndrome (SSNS) families are lacking. We studied an extended SSNS Bedouin (B) family with a high rate of consanguinity. The clinical presentation and steroid response of its 11 affected individuals were similar to those of sporadic SSNS (spontaneous remission towards puberty and minimal change disease by kidney biopsy). Genome-wide linkage analysis, using a 382 microsatellite-markers mapping set and additional markers adjacent to 80 candidate genes of the index family, did not support linkage to any chromosomal locus. Retrospective analysis of all additional children with SSNS treated by our institution in the past 20 years (n = 96, 50% of them of Jewish origin) revealed another five non-related B families with 2–3 first-degree cousins affected with SSNS in each. The overall familial SSNS rate among the B population (excluding the index family) was 28%, compared with 4% among Jews (Js) (OR 1.8–64, P < 0.005). There were more Bs with simple SSNS than there were Js (71% and 40%, respectively; OR 3.58, 95% CI 1.41–9.23, P < 0.01). In summary, SSNS in this index family was not linked to any of the presently known chromosomal loci nor predicted to be caused by mutation in any one of a list of genes associated with nephrotic syndrome (NS). The presence of other B families affected by SSNS supports the role for susceptibility genes enrichment, exposing highly consanguineous populations to an increased incidence of SSNS. An erratum to this article can be found at     

A peculiar Maxwell’s Demon observed in a time-dependent stadium-like billiard

The dynamics of a driven stadium-like billiard is considered using the formalism of discrete mappings. The model presents a resonant velocity that depends on the rotation number around fixed points and external boundary perturbation which plays an important separation rule in the model. We show that particles exhibiting Fermi acceleration (initial velocity is above the resonant one) are scaling invariant with respect to the initial velocity and external perturbation. However, initial velocities below the resonant one lead the particles to decelerate therefore unlimited energy growth is not observed. This phenomenon may be interpreted as a specific Maxwell’s Demon which may separate fast and slow billiard particles.     

Conditional dissipation of scalars in homogeneous turbulence: Closure for MMC modelling

While the mean and unconditional variance are to be predicted well by any reasonable turbulent combustion model, these are generally not sufficient for the accurate modelling of complex phenomena such as extinction/reignition. An additional criterion has been recently introduced: accurate modelling of the dissipation timescales associated with fluctuations of scalars about their conditional mean (conditional dissipation timescales). Analysis of Direct Numerical Simulation (DNS) results for a passive scalar shows that the conditional dissipation timescale is of the order of the integral timescale and smaller than the unconditional dissipation timescale. A model is proposed: the conditional dissipation timescale is proportional to the integral timescale. This model is used in Multiple Mapping Conditioning (MMC) modelling for a passive scalar case and a reactive scalar case, comparing to DNS results for both. The results show that this model improves the accuracy of MMC predictions so as to match the DNS results more closely using a relatively-coarse spatial resolution compared to other turbulent combustion models.     

GESFIDE-PROPELLER Approach for Simultaneous R2 and R2* Measurements in the Abdomen

Purpose

To investigate the feasibility of combining GESFIDE with PROPELLER sampling approaches for simultaneous abdominal R2 and R2* mapping.

Materials and Methods

R2 and R2* measurements were performed in 9 healthy volunteers and phantoms using the GESFIDE-PROPELLER and the conventional Cartesian-sampling GESFIDE approaches.

Results

Images acquired with the GESFIDE-PROPELLER sequence effectively mitigated the respiratory motion artifacts, which were clearly evident in the images acquired using the conventional GESFIDE approach. There was no significant difference between GESFIDE-PROPELLER and reference MGRE R2* measurements (p = 0.162) whereas the Cartesian-sampling based GESFIDE methods significantly overestimated R2* values compared to MGRE measurements (p < 0.001).

Conclusion

The GESFIDE-PROPELLER sequence provided high quality images and accurate abdominal R2 and R2* maps while avoiding the motion artifacts common to the conventional Cartesian-sampling GESFIDE approaches.