Local Clustering Method Research Articles

Compressing and Recovering Short-Range MEMS-Based LiDAR Point Clouds Based on Adaptive Clustered Compressive Sensing and Application to 3D Rock Fragment Surface Point Clouds.

Short-range MEMS-based (Micro Electronical Mechanical System) LiDAR provides precise point cloud datasets for rock fragment surfaces. However, there is more vibrational noise in MEMS-based LiDAR signals, which cannot guarantee that the reconstructed point cloud data are not distorted with a high compression ratio. Many studies have illustrated that wavelet-based clustered compressive sensing can improve reconstruction precision. The k-means clustering algorithm can be conveniently employed to obtain clusters; however, estimating a meaningful k value (i.e., the number of clusters) is challenging. An excessive quantity of clusters is not necessary for dense point clouds, as this leads to elevated consumption of memory and CPU resources. For sparser point clouds, fewer clusters lead to more distortions, while excessive clusters lead to more voids in reconstructed point clouds. This study proposes a local clustering method to determine a number of clusters closer to the actual number based on GMM (Gaussian Mixture Model) observation distances and density peaks. Experimental results illustrate that the estimated number of clusters is closer to the actual number in four datasets from the KEEL public repository. In point cloud compression and recovery experiments, our proposed approach compresses and recovers the Bunny and Armadillo datasets in the Stanford 3D repository; the experimental results illustrate that our proposed approach improves reconstructed point clouds' geometry and curvature similarity. Furthermore, the geometric similarity increases to 0.9 above in our complete rock fragment surface datasets after selecting a better wavelet basis for each dimension of MEMS-based LiDAR signals. In both experiments, the sparsity of signals was 0.8 and the sampling ratio was 0.4. Finally, a rock outcrop point cloud data experiment is utilized to verify that the proposed approach is applicable for large-scale research objects. All of our experiments illustrate that the proposed adaptive clustered compressive sensing approach can better reconstruct MEMS-based LiDAR point clouds with a lower sampling ratio.

Spatial and temporal clustering of anti-SARS-CoV-2 antibodies in Illinois household cats, 2021-2023.

This study aimed to evaluate the seroprevalence and spatial and temporal clustering of SARS-CoV-2 antibodies in household cats within 63 counties in Illinois from October 2021 to May 2023. The analysis followed a stepwise approach. First, in a choropleth point map, we illustrated the distribution of county-level seroprevalence of SARS-CoV-2 antibodies. Next, spatial interpolation was used to predict the seroprevalence in counties without recorded data. Global and local clustering methods were used to identify the extent of clustering and the counties with high or low seroprevalence, respectively. Next, temporal, spatial, and space-time scan statistic was used to identify periods and counties with higher-than-expected seroprevalence. In the last step, to identify more distinct areas in counties with high seroprevalence, city-level analysis was conducted to identify temporal and space-time clusters. Among 1,715 samples tested by serological assays, 244 samples (14%) tested positive. Young cats had higher seropositivity than older cats, and the third quarter of the year had the highest odds of seropositivity. Three county-level space-time clusters with higher-than-expected seroprevalence were identified in the northeastern, central-east, and southwest regions of Illinois, occurring between June and October 2022. In the city-level analysis, 2 space-time clusters were identified in Chicago's downtown and the southwestern suburbs of Chicago between June and September 2022. Our results suggest that the high density of humans and cats in large cities such as Chicago, might play a role in the transmission and clustering of SARS-CoV-2. Our study provides an in-depth analysis of SARS-CoV-2 epidemiology in Illinois household cats, which will aid in COVID-19 control and prevention.

Gradient-driven unsupervised video segmentation using deep learning techniques

We propose a three-dimensional video segmentation method using deep learning convolutional neural nets. The algorithm utilizes the local gradient computed at each pixel location together with the global boundary map acquired through deep learning methods to generate initial pixel groups by traversing from low to high gradient regions. A local clustering method is then employed to refine these initial pixel groups. The refined subvolumes in the homogeneous regions of video are selected as initial seeds and iteratively combined with adjacent groups based on intensity similarities. The volume growth is terminated at the color boundaries of the video. The oversegments obtained from the above steps are then merged hierarchically by a multivariate approach yielding a final segmentation map for each frame. The results show that our proposed methodology compares favorably well, on a qualitative and quantitative level, in segmentation quality and computational efficiency, with the latest state-of-the-art techniques utilizing the video segmentation benchmark dataset.

Weighted network community division based on co-neighbor nodes and similarity

In order to realize the accuracy of weighted network community division, based on the analysis of existing weighted network community dividing algorithm, a community dividing algorithm based on co-neighbor nodes and similarity is proposed. Firstly, the similarity between nodes is defined, and the weight of the edge is added to the local modularity function. Secondly, the Burt structural hole principle is used to select the important nodes. Then, the local clustering method is used to continuously expand the community. Finally, the judgment is made. The change of the Local modularity determines whether to continue the clustering, so as to obtain a more accurate result. Experiments on the simulation dataset and Zachary’s Karate club dataset show that compared with the CRMA algorithm, the algorithm is effectively improved in weighted network division time and accuracy.

Quantification of contrast agent materials using a new image- domain multi material decomposition algorithm based on dual energy CT.

Objective:Dual-Energy CT (DECT) is an imaging modality in which the objects are scanned by two different energy spectra. Using these two measurements, two type of materials can be separated and density image pairs can be generated as well. Decomposing more than two materials is necessary in both clinical and industrial CT applications.Methods:In our MMD, barycentric coordinates were chosen using an innovative local clustering method. Local clustering increases precision in the barycentric coordinates assignment by decreasing search domain. Therefore the algorithm can be run in parallel. For optimizing coordinates selection, a fast bi-directional Hausdorff distance measurement is used. To deal with the significant obstacle of noise, we used Doubly Local Wiener Filter Directional Window (DLWFDW) algorithm.Results:Briefly, the proposed algorithm separates blood and fat ROIs with errors of less than 2 and 9 % respectively on the clinical images. Also, the ability to decompose different materials with different concentrations is evaluated employing the phantom data. The highest accuracy obtained in separating different materials with different concentrations was 93 % (for calcium plaque) and 97.1 % (for iodine contrast agent) respectively. The obtained results discussed in detail in the following results section.Conclusion:In this study, we propose a new material decomposition algorithm. It improves the MMD work flow by employing tools which are easy to implement. Furthermore, in this study, an effort has been made to turn the MMD algorithm into a semi-automatic algorithm by employing clustering concept in material coordinate’s assignment. The performance of the proposed method is comparable to existing methods from qualitative and quantitative aspects.Advances in knowledge:All decomposition methods have their own specific problems. Image- domain decomposition also has barriers and problems, including the need for a predetermined table for the separation of different materials with specified coordinates. In the present study, it attempts to solve this problem by using clustering methods and relying on the intervals between different materials in the attenuation domain.

Superpixel generation for synthetic aperture radar imagery using edge-dominated local clustering

Recently, superpixel-based methods have shown promising performance for synthetic aperture radar (SAR) image interpretation. In these methods, the statistical model-based local iterative clustering represents the mainstream of superpixel generation for SAR images. However, errors in the model parameter estimation degrade the accuracy of the model-based distance measure between a pixel and a cluster, which directly affects the performance of superpixel segmentation results. Further, the relative weight between statistical similarity and spatial proximity should be carefully selected to control the balance between boundary adherence and regularity of superpixels. An edge-dominated local clustering method is proposed to overcome these limitations. Edge information is introduced not only to define the dissimilarity of a pixel and a cluster but also to provide an adaptive grid with multiple layers for the initialization of cluster centers. Experiments on simulated and real datasets show that, compared with the previous algorithms using the statistical model-based dissimilarity, the proposed method produces superpixels, which have better edge adherence and stable performance.

Geographical clustering of incident acute myocardial infarction in Denmark: A spatial analysis approach

ObjectivesTo examine the geographical patterns in AMI and characterize individual and neighborhood sociodemographic factors for persons living inside versus outside AMI clusters. MethodsThe study population comprised 3,515,670 adults out of whom 74,126 persons experienced an incident AMI (2005–2011). Kernel density estimation and global and local clustering methods were used to examine the geographical patterns in AMI. Median differences and frequency distributions of sociodemographic factors were calculated for persons living inside versus outside AMI clusters. ResultsGlobal clustering of AMI occurred in Denmark. Throughout the country, 112 significant clusters with high risk of incident AMI were identified. The relative risk of AMI in significant clusters ranged from 1.45 to 47.43 (median=4.84). Individual and neighborhood socioeconomic position was markedly lower for persons living inside versus outside AMI clusters. ConclusionsAMI is geographically unequally distributed throughout Denmark and determinants of these geographical patterns might include individual- and neighborhood-level sociodemographic factors.

Defining and evaluating network communities based on ground-truth

Nodes in real-world networks organize into densely linked communities where edges appear with high concentration among the members of the community. Identifying such communities of nodes has proven to be a challenging task due to a plethora of definitions of network communities, intractability of methods for detecting them, and the issues with evaluation which stem from the lack of a reliable gold-standard ground-truth. In this paper, we distinguish between structural and functional definitions of network communities. Structural definitions of communities are based on connectivity patterns, like the density of connections between the community members, while functional definitions are based on (often unobserved) common function or role of the community members in the network. We argue that the goal of network community detection is to extract functional communities based on the connectivity structure of the nodes in the network. We then identify networks with explicitly labeled functional communities to which we refer as ground-truth communities. In particular, we study a set of 230 large real-world social, collaboration, and information networks where nodes explicitly state their community memberships. For example, in social networks, nodes explicitly join various interest-based social groups. We use such social groups to define a reliable and robust notion of ground-truth communities. We then propose a methodology, which allows us to compare and quantitatively evaluate how different structural definitions of communities correspond to ground-truth functional communities. We study 13 commonly used structural definitions of communities and examine their sensitivity, robustness and performance in identifying the ground-truth. We show that the 13 structural definitions are heavily correlated and naturally group into four classes. We find that two of these definitions, Conductance and Triad participation ratio, consistently give the best performance in identifying ground-truth communities. We also investigate a task of detecting communities given a single seed node. We extend the local spectral clustering algorithm into a heuristic parameter-free community detection method that easily scales to networks with more than 100 million nodes. The proposed method achieves 30 % relative improvement over current local clustering methods.

A global clustering algorithm to identify long intergenic non-coding RNA--with applications in mouse macrophages.

Identification of diffuse signals from the chromatin immunoprecipitation and high-throughput massively parallel sequencing (ChIP-Seq) technology poses significant computational challenges, and there are few methods currently available. We present a novel global clustering approach to enrich diffuse CHIP-Seq signals of RNA polymerase II and histone 3 lysine 4 trimethylation (H3K4Me3) and apply it to identify putative long intergenic non-coding RNAs (lincRNAs) in macrophage cells. Our global clustering method compares favorably to the local clustering method SICER that was also designed to identify diffuse CHIP-Seq signals. The validity of the algorithm is confirmed at several levels. First, 8 out of a total of 11 selected putative lincRNA regions in primary macrophages respond to lipopolysaccharides (LPS) treatment as predicted by our computational method. Second, the genes nearest to lincRNAs are enriched with biological functions related to metabolic processes under resting conditions but with developmental and immune-related functions under LPS treatment. Third, the putative lincRNAs have conserved promoters, modestly conserved exons, and expected secondary structures by prediction. Last, they are enriched with motifs of transcription factors such as PU.1 and AP.1, previously shown to be important lineage determining factors in macrophages, and 83% of them overlap with distal enhancers markers. In summary, GCLS based on RNA polymerase II and H3K4Me3 CHIP-Seq method can effectively detect putative lincRNAs that exhibit expected characteristics, as exemplified by macrophages in the study.

S142012 大規模物流センターにおける集配計画の研究([S14201]生産システムの新展開(応用・実践) (1))

To efficiently control the logistic center, several problems occur to solve them. One of them is an order-picking problem. This problem is to assign picking items to workers and to determine picking navigation route. However, some schedule often makes a queue at the specific shelves. This study proposes a new scheduling method whose solution avoid such a queue as possible as it can. For this purpose, the order-picking problem is transformed into the job-shop problem (JSP) and Local Clustering Method (LCO) is applied to the JSP. The new method is verified to be efficient by comparing its solution with the preference rule's one.

LEARNING STRUCTURED BAYESIAN NETWORKS: COMBINING ABSTRACTION HIERARCHIES AND TREE‐STRUCTURED CONDITIONAL PROBABILITY TABLES

Context‐specific independence representations, such as tree‐structured conditional probability distributions, capture local independence relationships among the random variables in a Bayesian network (BN). Local independence relationships among the random variables can also be captured by using attribute‐value hierarchies to find an appropriate abstraction level for the values used to describe the conditional probability distributions. Capturing this local structure is important because it reduces the number of parameters required to represent the distribution. This can lead to more robust parameter estimation and structure selection, more efficient inference algorithms, and more interpretable models. In this paper, we introduce Tree‐Abstraction‐Based Search (TABS), an approach for learning a data distribution by inducing the graph structure and parameters of a BN from training data. TABS combines tree structure and attribute‐value hierarchies to compactly represent conditional probability tables. To construct the attribute‐value hierarchies, we investigate two data‐driven techniques: a global clustering method, which uses all of the training data to build the attribute‐value hierarchies, and can be performed as a preprocessing step; and a local clustering method, which uses only the local network structure to learn attribute‐value hierarchies. We present empirical results for three real‐world domains, finding that (1) combining tree structure and attribute‐value hierarchies improves the accuracy of generalization, while providing a significant reduction in the number of parameters in the learned networks, and (2) data‐derived hierarchies perform as well or better than expert‐provided hierarchies.

Entropy of phase-separated structures

The morphology of phase-separated structures can often be accurately modelled by level-cut Gaussian random fields. We consider the statistical ensemble of level-cut states and its relation to the Ising model with a long-range interaction. The entropy of the ensemble can be numerically calculated using the local cluster method. The method is tested on the exact solution of one-dimensional Ising model with the long-range Kac potential and illustrated using examples of level-cut Gaussian fields. At least in one-dimensional systems, the cluster method is easy to apply and produces accurate results.