Random Field-based Method Research Articles

A reliability analysis framework coupled with statistical uncertainty characterization for geotechnical engineering

Reliability analysis plays an important role in the risk management of geotechnical engineering. For the random field-based method, it is expected that the uncertainty characterization of geo-material parameters and the realization of random field can be integrated effectively. Moreover, as the increase in measured data size is generally difficult in the field investigation of geotechnical engineering due to limitation of budget and time etc., the statistical uncertainty resulting from sparse data should be paid great attention. Therefore, taking the determination of hyper-parameters for Bayesian-based conditional random field as the breakthrough, this study proposed a reliability analysis framework to achieve the expectation above. In this proposed reliability analysis framework, the present characterization method of statistical uncertainty is improved by setting the lognormal distribution as the prior distribution of scale of fluctuation (SOF). Subsequently, the performance of statistical uncertainty characterization method is tested by a set of unconfined compressive strength (UCS) database about rocks. Then, a case study about the stability analysis of slope is employed to demonstrate the beneficial effect of the proposed reliability analysis framework. It is found that the uncertainty in both the realization of random field and the reliability analysis results can be significantly mitigated by the proposed reliability analysis framework.

Contingency analysis with warm starter using probabilistic graphical model

Cyberthreats are an increasingly common risk to the power grid and can thwart secure grid operations. We propose to extend contingency analysis to include cyberthreat evaluations. However, unlike the traditional N-1 or N-2 contingencies, cyberthreats (e.g., MadIoT) require simulating hard-to-solve N-k (with k ≫ 2) contingencies in a practical amount of time. Purely physics-based power flow solvers, while being accurate, are slow and may not solve N-k contingencies in a timely manner, whereas the emerging data-driven alternatives are fast but not sufficiently generalizable, interpretable, and scalable. To address these challenges, we propose a novel conditional Gaussian Random Field-based data-driven method that performs fast and accurate evaluation of cyberthreats. It achieves speedup of contingency analysis by warm-starting simulations, i.e., improving starting points, for the physical solvers. To improve the physical interpretability and generalizability, the proposed method incorporates domain knowledge by considering the graphical nature of the grid topology. To improve scalability, the method applies physics-informed regularization that reduces model complexity. Experiments validate that simulating MadIoT-induced attacks with our warm starter becomes approximately 5x faster on a realistic 2000-bus system.

Reliability analysis of unsaturated soil slope stability using spatial random field-based Bayesian method

Rainfall and water level change are two of the main factors causing failure of reservoir slopes. Thus, soil-fluid mechanics is applied with the fluctuation of groundwater table. However, many of the geotechnical parameters needed for the analysis are highly varied. Spatial random field-based Bayesian method is proposed, which can systematically assimilate prior knowledge, borehole testing data, and long-term monitoring data to obtain posterior distributions. There are three major components of the method. They include unsaturated soil-fluid coupling mechanics, spatial multivariate discretization, and subset Monte Carlo simulation of reliability analysis. The approach is applied to the Shiliushubao slope of the Three Gorges Reservoir area, which is located 9.0 km downstream of the Badong County in Hubei Province, China. Results prove that the updated key geotechnical parameters will quantitatively predict the geohazards of landslide, especially for unsaturated soil slope conditions that suffer an unprecedented heavy rainfall subject to low reservoir water level in the upcoming summer.

Non‐parametric mixture model with TV spatial regularisation and its dual expectation maximisation algorithm

An image segmentation method based on a non-parametric mixture model together with total variation (TV) regularisation is proposed. The authors use a kernel density estimator as a basic mixture model, which can better separate the non-central distributed data. To enforce its robustness, they integrate the well-known TV regularisation into the statistical method. They use the dual method to efficiently solve the TV-related energy and get a new dual expectation maximisation algorithm. Experiments on both synthetic images and real images show that the proposed algorithm can achieve good segmentation results. Compared with the parametric models and hidden Markov random field-based method, the proposed method can produce better result in some cases.

Incorporating interaction networks into the determination of functionally related hit genes in genomic experiments with Markov random fields.

MotivationIncorporating gene interaction data into the identification of ‘hit’ genes in genomic experiments is a well-established approach leveraging the ‘guilt by association’ assumption to obtain a network based hit list of functionally related genes. We aim to develop a method to allow for multivariate gene scores and multiple hit labels in order to extend the analysis of genomic screening data within such an approach.ResultsWe propose a Markov random field-based method to achieve our aim and show that the particular advantages of our method compared with those currently used lead to new insights in previously analysed data as well as for our own motivating data. Our method additionally achieves the best performance in an independent simulation experiment. The real data applications we consider comprise of a survival analysis and differential expression experiment and a cell-based RNA interference functional screen.Availability and implementationWe provide all of the data and code related to the results in the paper.Supplementary information Supplementary data are available at Bioinformatics online.

A hidden Markov random field-based Bayesian method for the detection of long-range chromosomal interactions in Hi-C data

A hidden Markov random field-based Bayesian method for the detection of long-range chromosomal interactions in Hi-C data

A hidden Markov random field-based Bayesian method for the detection of long-range chromosomal interactions in Hi-C data.

Advances in chromosome conformation capture and next-generation sequencing technologies are enabling genome-wide investigation of dynamic chromatin interactions. For example, Hi-C experiments generate genome-wide contact frequencies between pairs of loci by sequencing DNA segments ligated from loci in close spatial proximity. One essential task in such studies is peak calling, that is, detecting non-random interactions between loci from the two-dimensional contact frequency matrix. Successful fulfillment of this task has many important implications including identifying long-range interactions that assist interpreting a sizable fraction of the results from genome-wide association studies. The task - distinguishing biologically meaningful chromatin interactions from massive numbers of random interactions - poses great challenges both statistically and computationally. Model-based methods to address this challenge are still lacking. In particular, no statistical model exists that takes the underlying dependency structure into consideration. In this paper, we propose a hidden Markov random field (HMRF) based Bayesian method to rigorously model interaction probabilities in the two-dimensional space based on the contact frequency matrix. By borrowing information from neighboring loci pairs, our method demonstrates superior reproducibility and statistical power in both simulation studies and real data analysis. The Source codes can be downloaded at: http://www.unc.edu/∼yunmli/HMRFBayesHiC CONTACT: ming.hu@nyumc.org or yunli@med.unc.edu Supplementary data are available at Bioinformatics online.

An efficient space–time based simulation approach of wind velocity field with embedded conditional interpolation for unevenly spaced locations

The classical spectral representation method (SRM) has been extensively used in the simulation of multivariate stationary Gaussian random processes. Due to the application of fast Fourier transform (FFT), the simulation is usually efficient. However, for processes with a large number of simulation points, it becomes necessary to enhance the simulation efficiency. One example is the wind velocity field along a large-span bridge, where hundreds of wind velocity fluctuations are required. In the case of bridges built over a homogeneous terrain such as coastal area or flat plain, the wind velocity field can be modeled as a multivariate homogeneous random process i.e., the auto power spectral densities (PSDs) at evenly-spaced simulation points are same and the cross PSD is a function of separation distance between two simulation points. Furthermore, in some applications, additional simulation points need to be included to a set of uniformly distributed points in order to make the wind velocity field consistent to the structural dynamic analysis requirement.In this paper, a hybrid approach of space–time random-field based SRM and proper orthogonal decomposition (POD)-based interpolation is developed for simulating the above wind velocity process. In this approach, the random-field based SRM is used to simulate the multivariate homogeneous random process composed of a set of uniformly distributed simulation locations while POD-based interpolation is used to conditionally generate the wind velocities at a few unevenly distributed points using the previously simulated wind velocities. The idea of the former is based on transforming the simulation of the homogeneous random process into that of the corresponding space–time random field where the phase angle is assumed to be zero and the coherence function must be an even function in terms of separation distance. Through this procedure, customary requirement for spectral matrix decomposition is eliminated and application of two dimensional FFT can improve the simulation efficiency dramatically. The shortcomings of this method include a slight approximation regarding the simulated sample and the non-ergodicity for the correlation function. The numerical example of a homogeneous wind velocity field along a bridge deck shows that the proposed random field-based method is very efficient in terms of accuracy and efficiency when the number of simulation locations is large and the POD-based interpolation also has good performance.

Markov random field-based method for super-resolution mapping of forest encroachment from remotely sensed ASTER image

Forest encroachment (FE) is a problem in Andaman and Nicobar Islands (ANI) in India for environment and planning. Small gaps created in the forest slowly expand its periphery disturbing the biodiversity. Therefore, intrusion of poachers, slash and burn and other factors causing FE must be carefully detected and monitored. Remote sensing offers a great opportunity to accomplish this task because of its synoptic view. Conventional classification methods with remotely sensed images are problematic because of small size of FE and mixed landcover composition. This study presents an application of super-resolution mapping (SRM) based on Markov random field for detection of FE using ASTER (15 m) images. The SRM results were validated using multispectral IRS LISS-IV (5.8 m) image. Non-contiguous FE patches of various sizes and shapes are characterized using the spatial contextual information. The novelty of this approach lies in the identification and separability of small FE pockets which could not be achieved with pixel-based maximum likelihood classifier (MLC). The SRM parameters were optimized and found comparable to previous studies. Classification accuracy obtained with SRM at scale factor 3 is κ = 0.62 that is superior to accuracy of MLC (κ = 0.51). SRM is a promising tool for detection and monitoring of FE at Rutland Island in ANI, India.

BMRF-Net: a software tool for identification of protein interaction subnetworks by a bagging Markov random field-based method.

Identification of protein interaction subnetworks is an important step to help us understand complex molecular mechanisms in cancer. In this paper, we develop a BMRF-Net package, implemented in Java and C++, to identify protein interaction subnetworks based on a bagging Markov random field (BMRF) framework. By integrating gene expression data and protein-protein interaction data, this software tool can be used to identify biologically meaningful subnetworks. A user friendly graphic user interface is developed as a Cytoscape plugin for the BMRF-Net software to deal with the input/output interface. The detailed structure of the identified networks can be visualized in Cytoscape conveniently. The BMRF-Net package has been applied to breast cancer data to identify significant subnetworks related to breast cancer recurrence. The BMRF-Net package is available at http://sourceforge.net/projects/bmrfcjava/. The package is tested under Ubuntu 12.04 (64-bit), Java 7, glibc 2.15 and Cytoscape 3.1.0.

Random field-based reliability analysis of prestressed concrete bridges

A method for reliability assessment of prestressed concrete bridges accounting for the spatial variability of structural parameters is presented in this paper. The spatial variability of these parameters is modeled by means of random fields and the reliability estimates were determined by a generalized first-order reliability algorithm. The main feature of the random field-based reliability method is to consider the spatial variability of structural parameters that is disregarded in the traditional reliability method for prestressed concrete bridges. The accuracy and efficiency of the random field-based reliability method is demonstrated through two numerical examples. A software strategy for interfacing the random field-based reliability method with ANSYS is developed via a freely available MATLAB software tool (FERUM). As a practical engineering example, the reliability of a prestressed concrete bridge is illustrated. The obtained results clearly show the applicability and merits of the random field-based reliability method and the associated software strategy.

Domain-Based Approaches to Prediction and Analysis of Protein-Protein Interactions

Protein-protein interactions play various essential roles in cellular systems. Many methods have been developed for inference of protein-protein interactions from protein sequence data. In this paper, the authors focus on methods based on domain-domain interactions, where a domain is defined as a region within a protein that either performs a specific function or constitutes a stable structural unit. In these methods, the probabilities of domain-domain interactions are inferred from known protein-protein interaction data and protein domain data, and then prediction of interactions is performed based on these probabilities and contents of domains of given proteins. This paper overviews several fundamental methods, which include association method, expectation maximization-based method, support vector machine-based method, linear programming-based method, and conditional random field-based method. This paper also reviews a simple evolutionary model of protein domains, which yields a scale-free distribution of protein domains. By combining with a domain-based protein interaction model, a scale-free distribution of protein-protein interaction networks is also derived.

Suitability of Markov Random Field-Based Method for Super-Resolution Land Cover Mapping

Suitability of Markov Random Field-Based Method for Super-Resolution Land Cover Mapping Super-resolution mapping (SRM) works by dividing the coarse pixel into sub-pixels and assign the class proportion estimated by subpixel classification to each corresponding sub-pixels then the class labelling is optimized based on the principle of spatial dependency. Among the existing SRM techniques Markov random field (MRF)-based SRM is one of the most recently introduced technique. This study attempts to assess the suitability of the technique for superresolution land cover mapping.

Identifying protein interaction subnetworks by a bagging Markov random field-based method

Identification of differentially expressed subnetworks from protein–protein interaction (PPI) networks has become increasingly important to our global understanding of the molecular mechanisms that drive cancer. Several methods have been proposed for PPI subnetwork identification, but the dependency among network member genes is not explicitly considered, leaving many important hub genes largely unidentified. We present a new method, based on a bagging Markov random field (BMRF) framework, to improve subnetwork identification for mechanistic studies of breast cancer. The method follows a maximum a posteriori principle to form a novel network score that explicitly considers pairwise gene interactions in PPI networks, and it searches for subnetworks with maximal network scores. To improve their robustness across data sets, a bagging scheme based on bootstrapping samples is implemented to statistically select high confidence subnetworks. We first compared the BMRF-based method with existing methods on simulation data to demonstrate its improved performance. We then applied our method to breast cancer data to identify PPI subnetworks associated with breast cancer progression and/or tamoxifen resistance. The experimental results show that not only an improved prediction performance can be achieved by the BMRF approach when tested on independent data sets, but biologically meaningful subnetworks can also be revealed that are relevant to breast cancer and tamoxifen resistance.

Markov Random Field-based Fitting of a Subdivision-based Geometric Atlas.

An accurate labeling of a multi-part, complex anatomical structure (e.g., brain) is required in order to compare data across images for spatial analysis. It can be achieved by fitting an object-specific geometric atlas that is constructed using a partitioned, high-resolution deformable mesh and tagging each of its polygons with a region label. Subdivision meshes have been used to construct such an atlas because they can provide a compact representation of a partitioned, multi-resolution, object-specific mesh structure using only a few control points. However, automated fitting of a subdivision mesh-based geometric atlas to an anatomical structure in an image is a difficult problem and has not been sufficiently addressed. In this paper, we propose a novel Markov Random Field-based method for fitting a planar, multi-part subdivision mesh to anatomical data. The optimal fitting of the atlas is obtained by determining the optimal locations of the control points. We also tackle the problem of landmark matching in tandem with atlas fitting by constructing a single graphical model to impose pose-invariant, landmark-based geometric constraints on atlas deformation. The atlas deformation is also governed by additional constraints imposed by the mesh's geometric properties and the object boundary. We demonstrate the potential of the proposed method on the difficult problem of segmenting a mouse brain and its interior regions in gene expression images which exhibit large intensity and shape variability. We obtain promising results when compared with manual annotations and prior methods.

New automatic defect classification algorithm based on a classification-after-segmentation framework

We propose a new method that classifies wafer images according to their defect types for automatic defect classification in semiconductor fabrication processes. Conventional image classifi- ers using global properties cannot be used in this problem, because the defects usually occupy very small regions in the images. Hence, the defects should first be segmented, and the shape of the seg- ment and the features extracted from the region are used for clas- sification. In other words, we need to develop a classification-after- segmentation approach for the use of features from the small regions corresponding to the defects. However, the segmentation of scratch defects is not easy due to the shrinking bias problem when using conventional methods. We propose a new Markov random field-based method for the segmentation of wafer images. Then we design an AdaBoost-based classifier that uses the features ex- tracted from the segmented local regions. © 2010 SPIE and

Effects ofm-Chlorophenylpiperazine on Regional Brain Glucose Utilization: A Positron Emission Tomographic Comparison of Alcoholic and Control Subjects

m-Chlorophenylpiperazine (mCPP) is a mixed serotonin agonist/antagonist used extensively in psychiatric research. Alcoholics show blunted neuroendocrine responses to mCPP, and in some settings mCPP can induce craving for alcohol, particularly among early onset alcoholics. We used 2-[18F]-2-deoxy-D-glucose positron emission tomography to examine the effects of intravenously administered mCPP (0.08 mg/kg) on brain glucose utilization in a group of 18 male alcoholics and 12 healthy male control subjects. Differences between two sequential scans (the first followed placebo and the second followed mCPP) were evaluated statistically with a Gaussian random field-based method. Among healthy volunteers mCPP significantly increased brain glucose metabolism in the right medial and posterior orbital gyrus, the cerebellar hemispheres bilaterally, the left nucleus accumbens, the head of the caudate nucleus bilaterally, the anterior and medial-dorsal nuclei of the thalamus bilaterally, the middle frontal gyrus, the left insular cortex, the left middle temporal gyrus, and the posterior cingulate gyrus. Among alcoholic subjects mCPP significantly increased brain glucose metabolism in larger areas of the cerebellum and posterior cingulate than it did in healthy volunteers, but compared with the healthy volunteers, alcoholics showed a smaller area of mCPP-induced activation in the thalamus, almost no activation in the orbital cortices, and no activation at all in the head of the caudate nucleus or the middle frontal gyrus. These results suggest that a serotoninergic challenge activates basal ganglia circuits involving orbital and prefrontal cortices among healthy volunteers but that the response of these circuits is blunted among alcoholics.