Evolutionary level set method for structural topology optimization

This paper proposes an evolutionary accelerated computational level set algorithm for structure topology optimization. It integrates the merits of evolutionary structure optimization (ESO) and level set method (LSM). Traditional LSM algorithm is largely dependent on the initial guess topology. The proposed method combines the merits of ESO techniques with those of LSM algorithm, while allowing new holes to be automatically generated in low strain energy within the nodal neighboring region during optimization. The validity and robustness of the new algorithm are supported by some widely used benchmark examples in topology optimization. Numerical computations show that optimization convergence is accelerated effectively.     

Sensitivity analysis and optimization of vehicle–bridge systems based on combined PEM–PIM strategy

This paper presents a new optimization method for coupled vehicle–bridge systems subjected to uneven road surface excitation. The vehicle system is simplified as a multiple rigid-body model and the single-span bridge is modeled as a simply supported Bernoulli–Euler beam. The pseudo-excitation method transforms the random surface roughness into the superposition of a series of deterministic pseudo-harmonic excitations, which enables convenient and accurate computation of first and second order sensitivity information. The precise integration method is used to compute the vertical random vibrations for both the vehicle and the bridge. The sensitivities are used to find the optimal solution, with vehicle ride comfort taken as the objective function. Optimization efficiency and computational accuracy are demonstrated numerically.     

Motion regularization for matting motion blurred objects

This paper addresses the problem of matting motion blurred objects from a single image. Existing single image matting methods are designed to extract static objects that have fractional pixel occupancy. This arises because the physical scene object has a finer resolution than the discrete image pixel and therefore only occupies a fraction of the pixel. For a motion blurred object, however, fractional pixel occupancy is attributed to the object’s motion over the exposure period. While conventional matting techniques can be used to matte motion blurred objects, they are not formulated in a manner that considers the object’s motion and tend to work only when the object is on a homogeneous background. We show how to obtain better alpha mattes by introducing a regularization term in the matting formulation to account for the object’s motion. In addition, we outline a method for estimating local object motion based on local gradient statistics from the original image. For the sake of completeness, we also discuss how user markup can be used to denote the local direction in lieu of motion estimation. Improvements to alpha mattes computed with our regularization are demonstrated on a variety of examples.     

CCS-MAC: Exploiting the overheard data for compression in wireless sensor networks

Both the overhearing and overhearing avoidance in a densely distributed sensor network may inevitably incur considerable power consumption. In this paper we propose a so-called CCS-MAC (collaborative compression strategy-based MAC) MAC protocol which facilitates to exploit those overheard data that is treated useless in traditional MAC protocols for the purpose of cost and energy savings. Particularly the CCS-MAC enables different sensor nodes to perform data compression cooperatively with regard to those overheard data, so that the redundancy of data prepared for the link layer transmission can be totally eliminated at the earliest. The problem of collaborative compression is analyzed and discussed along with a corresponding linear programming model formulated. Based on it a heuristic node-selection algorithm with a time complexity of (O(N²)) is proposed to the solve the linear programming problem. The node-selection algorithm is implemented in CCS-MAC at each sensor node in a distributed manner. The experiment results verify that the proposed CCS-MAC scheme can achieve a significant energy savings so as to prolong the lifetime of the sensor networks so far.     

Shrinkage-based regularization tests for high-dimensional data with application to gene set analysis

Traditional multivariate tests such as Hotelling’s test or Wilk’s test are designed for classical problems, where the number of observations is much larger than the dimension of the variables. For high-dimensional data, however, this assumption cannot be met any longer. In this article, we consider testing problems in high-dimensional MANOVA where the number of variables exceeds the sample size. To overcome the challenges with high dimensionality, we propose a new approach called a shrinkage-based regularization test, which is suitable for a variety of data structures including the one-sample problem and one-way MANOVA. Our approach uses a ridge regularization to overcome the singularity of the sample covariance matrix and applies a soft-thresholding technique to reduce random noise and improve the testing power. An appealing property of this approach is its ability to select relevant variables that provide evidence against the hypothesis. We compare the performance of our approach with some competing approaches via real microarray data and simulation studies. The results illustrate that the proposed statistics maintains relatively high power in detecting a wide family of alternatives.     

Modeling the glucose regulatory system in extreme preterm infants

Background

Premature infants represent a significant proportion of the neonatal intensive care population. Blood glucose homeostasis in this group is often disturbed by immaturity of endogenous regulatory systems and the stress of their condition. Hypo- and hyperglycemia are frequently reported in very low birth weight infants, and more mature infants often experience low levels of glycemia. A model capturing the unique fundamental dynamics of the neonatal glucose regulatory system could be used to develop better blood glucose control methods.

Methods

A metabolic system model is adapted from adult critical care to the unique physiological case of the neonate. Integral-based fitting methods were used to identify time-varying insulin sensitivity and non-insulin mediated glucose uptake profiles. The clinically important predictive ability of the model was assessed by assuming insulin sensitivity was constant over prediction intervals of 1, 2 and 4 h forward and comparing model-simulated versus actual clinical glucose values for all recorded interventions. The clinical data included 1091 glucose measurements over 3567 total patient hours, along with all associated insulin and nutritional infusion data, for N = 25 total cases. Ethics approval was obtained from the Upper South A Regional Ethics Committee for this study.

Results

The identified model had a median absolute percentage error of 2.4% [IQR: 0.9-4.8%] between model-fitted and clinical glucose values. Median absolute prediction errors at 1-, 2- and 4-h intervals were 5.2% [IQR: 2.5-10.3%], 9.4% [IQR: 4.5-18.4%] and 13.6% [IQR: 6.3-27.6%] respectively.

Conclusions

The model accurately captures and predicts the fundamental dynamic behaviors of the neonatal metabolism well enough for effective clinical decision support in glycemic control. The adaptation from adult to a neonatal case is based on the data from the literature. Low prediction errors and very low fitting errors indicate that the fundamental dynamics of glucose metabolism in both premature neonates and critical care adults can be described by similar mathematical models.     

Comparison of multiple linear regression and artificial neural network in developing the objective functions of the orthopaedic screws

Optimizing the orthopaedic screws can greatly improve their biomechanical performances. However, a methodical design optimization approach requires a long time to search the best design. Thus, the surrogate objective functions of the orthopaedic screws should be accurately developed. To our knowledge, there is no study to evaluate the strengths and limitations of the surrogate methods in developing the objective functions of the orthopaedic screws. Three-dimensional finite element models for both the tibial locking screws and the spinal pedicle screws were constructed and analyzed. Then, the learning data were prepared according to the arrangement of the Taguchi orthogonal array, and the verification data were selected with use of a randomized selection. Finally, the surrogate objective functions were developed by using either the multiple linear regression or the artificial neural network. The applicability and accuracy of those surrogate methods were evaluated and discussed. The multiple linear regression method could successfully construct the objective function of the tibial locking screws, but it failed to develop the objective function of the spinal pedicle screws. The artificial neural network method showed a greater capacity of prediction in developing the objective functions for the tibial locking screws and the spinal pedicle screws than the multiple linear regression method. The artificial neural network method may be a useful option for developing the objective functions of the orthopaedic screws with a greater structural complexity. The surrogate objective functions of the orthopaedic screws could effectively decrease the time and effort required for the design optimization process.     

Compression and aggregation of Bayesian estimates for data intensive computing

Bayesian estimation is a major and robust estimator for many advanced statistical models. Being able to incorporate prior knowledge in statistical inference, Bayesian methods have been successfully applied in many different fields such as business, computer science, economics, epidemiology, genetics, imaging, and political science. However, due to its high computational complexity, Bayesian estimation has been deemed difficult, if not impractical, for large-scale databases, stream data, data warehouses, and data in the cloud. In this paper, we propose a novel compression and aggregation schemes (C&A) that enables distributed, parallel, or incremental computation of Bayesian estimates. Assuming partitioning of a large dataset, the C&A scheme compresses each partition into a synopsis and aggregates the synopsis into an overall Bayesian estimate without accessing the raw data. Such a C&A scheme can find applications in OLAP for data cubes, stream data mining, and cloud computing. It saves tremendous computing time since it processes each partition only once, enabling fast incremental update, and allows parallel processing. We prove that the compression is asymptotically lossless in the sense that the aggregated estimator deviates from the true model by an error that is bounded and approaches to zero when the data size increases. The results show that the proposed C&A scheme can make feasible OLAP of Bayesian estimates in a data cube. Further, it supports real-time Bayesian analysis of stream data, which can only be scanned once and cannot be permanently retained. Experimental results validate our theoretical analysis and demonstrate that our method can dramatically save time and space costs with almost no degradation of the modeling accuracy.     

移动机器人闭环检测的视觉字典树金字塔TF-IDF得分匹配方法

针对移动机器人视觉闭环检测中,基于视觉字典本的场景外观表征性能受制于有限单词个数以及算法效率低的不足,本文对机器人视觉特征分层量化,构建视觉字典树, 计算树节点的TF-IDF熵作为对应视觉单词的权重,生成图像--单词逆向文档索引.为消除视觉字典本的单尺度量化误差,并克服基于字典树投影路径的平面匹配模式中不 区分不同层次节点的区分度对闭环检测的影响,本文融合字典树低层单词的强表征性和高层单词的强鲁棒性,提出由下而上逐层计算图像间相似性增量的金字塔得分匹 配方法.将不同时刻相似性大于阈值的图像位置提取为候选闭环,通过后验确认操作剔除误正闭环.在移动机器人视觉闭环检测实验中,本文算法提高了图像相似性计算 的效率和准确性,提高了闭环检测的准确率和召回率.     

一种13bit 40MS/s采样保持电路设计

设计了一个用于13bit40MS/s流水线ADC中的采样保持电路。该电路采用电容翻转结构,主运算放大器采用增益提高型折叠式共源共栅结构,以满足高速和高精度的要求。为减小与输入信号相关的非线性失真以获得良好的线性度,采用栅压自举开关。采用电源电压为3.3V的TSMC0.18μm工艺对电路进行设计和仿真,仿真结果表明,在40MHz的采样频率下,采用保持电路的SNDR达到84.8dB,SFDR达到92dB。