Geometric Network Research Articles

Evaluation of Bus Rapid Transit System: Riyadh city, Saudi Arabia

Al Riyadh city, the capital of the Kingdom of Saudi Arabia, is experiencing significant economic and population growth, it is also a major metropolitan area in Saudi Arabia and in the Gulf regions. This has led to a high level of car ownership and use of vehicles, and this has led to the construction of a series of slow-moving roads and roads around the city. Traffic congestion is a major problem, where road behavior has an economic and psychological impact on people. The Riyadh Bus stations and Bus Rapid Transit (BRT) System Project is an integral part of the current Riyadh metropolitan area including the new Riyadh Metro. The main objective of this paper’s is to study the “Bus Rapid Transit” network in the city of Al Riyadh and to assess the functioning of this network using geographic information systems and the theory of graphs. A centralized geodatabase has been built for BRT lines and stations. A geometric network was conducted for BRT routes to facilitate moving from any location to another within Riyadh city. The assessment and evaluation of BRT routes focused on routes and station distribution regarding Riyadh city and how BRT solved population needs in Riyadh city. Using the powerful GIS network analysis functions, a navigation system was developed to assist passengers in planning their trips from one station to another giving detailed information about route directions and different BRT lines used. The study reinforces the value of good design techniques for accessible public transportation systems. The utilization of accessible vehicles, such as low-floor buses, is needed but lacking for such an accessible system, planners and operators should be aware.

Geometric learning and statistical modeling for surgical outcomes evaluation in craniosynostosis using 3D photogrammetry

Background and objectiveAccurate and repeatable detection of craniofacial landmarks is crucial for automated quantitative evaluation of head development anomalies. Since traditional imaging modalities are discouraged in pediatric patients, 3D photogrammetry has emerged as a popular and safe imaging alternative to evaluate craniofacial anomalies. However, traditional image analysis methods are not designed to operate on unstructured image data representations such as 3D photogrammetry. MethodsWe present a fully automated pipeline to identify craniofacial landmarks in real time, and we use it to assess the head shape of patients with craniosynostosis using 3D photogrammetry. To detect craniofacial landmarks, we propose a novel geometric convolutional neural network based on Chebyshev polynomials to exploit the point connectivity information in 3D photogrammetry and quantify multi-resolution spatial features. We propose a landmark-specific trainable scheme that aggregates the multi-resolution geometric and texture features quantified at every vertex of a 3D photogram. Then, we embed a new probabilistic distance regressor module that leverages the integrated features at every point to predict landmark locations without assuming correspondences with specific vertices in the original 3D photogram. Finally, we use the detected landmarks to segment the calvaria from the 3D photograms of children with craniosynostosis, and we derive a new statistical index of head shape anomaly to quantify head shape improvements after surgical treatment. ResultsWe achieved an average error of 2.74 ± 2.70 mm identifying Bookstein Type I craniofacial landmarks, which is a significant improvement compared to other state-of-the-art methods. Our experiments also demonstrated a high robustness to spatial resolution variability in the 3D photograms. Finally, our head shape anomaly index quantified a significant reduction of head shape anomalies as a consequence of surgical treatment. ConclusionOur fully automated framework provides real-time craniofacial landmark detection from 3D photogrammetry with state-of-the-art accuracy. In addition, our new head shape anomaly index can quantify significant head phenotype changes and can be used to quantitatively evaluate surgical treatment in patients with craniosynostosis.

Thrust formation using a numerical granular rock box experiment

A numerical granular rock box experiment based on the Discrete Element Method (DEM) was performed in thin 3D geometry to investigate the role of granularity in thrust formation. This rock box experiment is an extension of the numerical sandbox test, which is considered an analog experiment for accretionary prism formation with cohesive contact force. The parameters of contact-based interaction are adjusted to account for the failure envelope of the host rock in the triaxial test. The developed model allows the numerical simulation of the horizontal shortening of a granular layer on geologically relevant scales. The rock box test successfully reproduces the characteristic structures of accretionary wedges developed during sequential thrust formation. In contrast to the non-cohesive case, the cohesive models generated a surface geometry with steeper angles, surface vertical faults, and enhanced bifurcation of the shear bands. The geometric network of fault planes depended on the healing of the cohesive force within the faulted region. The increasing complexity of the network was demonstrated using finer elements until its maximum radius reached 12.5 m. We also evaluated the granular nature of the thrust thickness, which was found to depend on the number of frictional elements rather than physical length. A decrease in the growth rate of the fault damage zone thickness with increasing fault displacement was found to be consistent with observational data. Our results suggest an important role for granularity in thrust evolution.

Geometric graph neural networks on multi-omics data to predict cancer survival outcomes

The advance of sequencing technologies has enabled a thorough molecular characterization of the genome in human cancers. To improve patient prognosis predictions and subsequent treatment strategies, it is imperative to develop advanced computational methods to analyze large-scale, high-dimensional genomic data. However, traditional machine learning methods face a challenge in handling the high-dimensional, low-sample size problem that is shown in most genomic data sets. To address this, our group has developed geometric network analysis techniques on multi-omics data in connection with prior biological knowledge derived from protein-protein interactions (PPIs) or pathways. Geometric features obtained from the genomic network, such as Ollivier-Ricci curvature and the invariant measure of the associated Markov chain, have been shown to be predictive of survival outcomes in various cancers. In this study, we propose a novel supervised deep learning method called geometric graph neural network (GGNN) that incorporates such geometric features into deep learning for enhanced predictive power and interpretability. More specifically, we utilize a state-of-the-art graph neural network with sparse connections between the hidden layers based on known biology of the PPI network and pathway information. Geometric features along with multi-omics data are then incorporated into the corresponding layers. The proposed approach utilizes a local–global principle in such a manner that highly predictive features are selected at the front layers and fed directly to the last layer for multivariable Cox proportional-hazards regression modeling. The method was applied to multi-omics data from the CoMMpass study of multiple myeloma and ten major cancers in The Cancer Genome Atlas (TCGA). In most experiments, our method showed superior predictive performance compared to other alternative methods.

State Identification of Power System Based on Visual Geometric Group Network

Accurate and timely identification of power system operation states are requisite for taking control strategies and ensuring power system security. With the high grid penetration of renewable energy generation and wide application of power electronic equipment, power system is becoming more complicated and facing more uncertainties. Traditional model based power system state identification methods are faced with challenges and data driven methods have great prospects. This paper proposes a new data-driven method for identifying power system operating states based on visual geometric group network VGG16 (16-layer Visual Geometry Group (VGG16). Firstly, the Lasso algorithm is used to select the operating state features of the power system to realize the features dimensionality reduction. Then, the weights of the loss functions are adjusted according to the difficulty of model for identifying power system states. Finally, the IEEE39 node system is used as an example, the effectiveness and robustness of the proposed method is verified.

Formation of crevasse‐squeeze ridges at Trygghamna, Svalbard

AbstractCrevasse‐squeeze ridges (CSRs) are landforms that have been unequivocally linked to surge‐type glaciers. The formation of CSRs has been discussed since they were first defined in the mid‐1980s. Here, we describe geometric CSR networks from the terrestrial glacier forefields of two glaciers in Trygghamna, Western Svalbard. No glacier surges have been observed in Trygghamna; however, the presence of the CSRs signifies past surge activity. Detailed geomorphological maps were constructed, and the spatial context of these landforms described. Cross‐sections of several CSRs highlight ridge architecture, structure and relationships to surrounding landforms and sediments. Most CSRs are symmetrical in cross‐profile, orientated perpendicular or oblique to the ice‐flow direction. Like previous investigations, we observe these ridge networks on top of till and flutes. Additionally, we, for the first time, document CSRs deposited directly on non‐glaciogenic subsurfaces, namely, beach gravels and bedrock. Our findings confirm previous CSR formation theories; basal sediments are squeezed into bottom‐up crevasses during surges, which are subsequently transported englacially until surge termination and are finally released by melt out from stagnant ice. Consequently, a network of CSRs is the product of a significant reorganisation and down‐glacier transport of basal sediment, exemplifying how single surges are agents of glacial sediment redistribution. These formation processes are illustrated in a refined schematic model. The results further contemporary understanding of CSRs in terrestrial surge‐type glacier settings and may also apply to landforms and sediments in certain marine settings and palaeoglacial environments.

Abstract 5367: Deep neural networks using protein-protein network information predict multiple myeloma survival

Abstract The modern development of sequencing technologies provides a comprehensive molecular portrait of human cancers. There is a strong need to develop methods to not only improve patient prognosis predictions but also to understand the driving factors for treatment. However, the high-dimension, low-sample size nature of the genomic data poses challenges for typical machine learning algorithms. The systematic understanding of genes with respect to a network (protein-protein interaction (PPI) network) is a way to handle the limit and the nonparametric analysis of geometric properties such as Ollivier-Ricci curvature and associated invariant measure developed by our group have proven to be successful for the prediction of survival in multiple cancers. In this work, we propose a novel supervised deep learning approach combining the aforementioned geometric methods, which benefit from the flexibility provided by deep learning techniques while still preserving much of the interpretability of the geometric analysis. We take advantage of a state-of-the-art graph neural network approach. Sparse connections between layers were inspired by the known biology of the PPI network from the Human Protein Reference Database (HPRD) and pathway information from the Kyoto Encyclopedia of Genes and Genomes (KEGG) database, supplemented with geometric network features which are fed into the network in corresponding layers. The prediction is based on a local-global principle, where highly predictive features are selected from early layers of the network and fed directly to the final layer to produce a multivariable Cox regression. We applied our method to RNA-Seq gene expression data from the CoMMpass study of multiple myeloma (MM). More specifically, 657 patients in the data set were randomly divided into training, validation and set-aside testing sets by a ratio of 6:2:2. We obtained an average C-index 0.66 of the prediction in the testing set from a 10-fold data split. Dichotomizing the testing set by its mean value to define high-risk vs. low-risk yielded a significant p-value of the log-rank test in the set-aside data (p-value =3e-4). We observed that geometric protein network information not only improved the outcome prediction (vs. 6% worse without geometric feature inputs), but was also more robust to fold splitting. From our model, we identified WEE1, CENPE and CENPF as top genes driving survival differences (higher expression of WEE1 increased risk and lower negative curvature between CENPE and CENPF increased risk). WEE1 is a cell cycle-related gene that regulates DNA repair and CENPE and CENPF are components of a fibrous layer of mitotic kinetochores, which have been indicated in the literature to be related to the prognosis as well as possible targets for treatment. While it is therefore logical that these genes would be implicated in the natural history of MM, they were identified entirely on the basis of network analysis. Citation Format: Jiening Zhu, Jung Hun Oh, Anish K. Simhal, Rena Elkin, Larry Norton, Joseph O. Deasy, Allen R. Tannenbaum. Deep neural networks using protein-protein network information predict multiple myeloma survival. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5367.

Abstract 2061: Protein network analysis uncovers a poor-survival subtype in multiple myeloma

Abstract Multiple myeloma (MM) prognosis incorporates a variety of metrics including treatment received, clinical factors, and genomic characteristics, with several specific genomic features predicting for shorter progression free survival (PFS). No study to date has integrated genomic data from a systems view, incorporating interactions between biomarkers in a network. We use a geometric network analysis that integrates complex interactions to characterize patterns of biological behavior not captured by individual genomic events. The methodology is mathematically well-defined and has no fitting parameters. We hypothesized that such a systems mathematical approach applied to gene interaction networks may delineate biologically relevant MM subtypes and potential new therapeutic targets. We overlaid RNA-Seq and copy number alteration data from the MMRF CoMMpass study (IA19) on a gene interactome derived from the Human Protein Reference Database using a novel graph metric of network robustness — Ollivier-Ricci curvature (ORC). Results were clustered, with the optimal number determined via silhouette score. Survival analysis for PFS was performed employing Kaplan-Meier and log-rank tests. A differential gene expression analysis between high and low risk groups was conducted. Differences in scalar ORC between the low-risk and high-risk groups were examined and contextualized using a pathway analysis. Pathway analysis was performed using the Broad Institute’s Gene Set Enrichment Analysis tool and the pathways used are from the hallmark gene set from the Human MSigDB collection. The dataset included 659 patients and the incorporated protein-protein interactions resulted in a network with 8,468 nodes and 33,695 edges. The ORC analysis discovered 6 clusters, with specific genomic features being associated with clusters predicting for long [hyperdiploidy, t(11:14)], and short [t(4;14), MAF/MAFB translocations] PFS. A differential gene expression analysis comparing the high risk and low risk groups identified 118 key genes. These genes were associated with various pathways both known and unknown to be associated with multiple myeloma, including mitotic spindle, DNA repair, inflammatory response, and the P53 pathways. Further scalar curvature analysis showed differences in the apoptosis, TGF beta signaling, and other signaling pathways. In summary, we applied the geometric network analysis tool ORC to multi-omics data in MM represented as biological networks to identify individuals at high risk of short PFS and relevant biological correlates. Decreased robustness of signaling near immune-related genes was associated with shorter survival, highlighting the plausible utility of using these methods to uncover new biological insights. Citation Format: Anish K. Simhal, Kylee H. Maclachlan, Rena Elkin, Jiening Zhu, Saad Z. Usmani, Jonathan J. Keats, Larry Norton, Joseph O. Deasy, Jung Hun Oh, Allen Tannenbaum. Protein network analysis uncovers a poor-survival subtype in multiple myeloma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 2061.

Linear feature projective geometric damped convolutional deep belief network for indoor floor planning

Wireless localization or positioning is essential for delivering location-based services for designing location tracking systems. Traditional indoor floor planning system employs wireless signals for accurate position estimation. But these positioning schemes failed to perform position estimation effectively and accurately through many obstacles or objects. The novel technique called Linear Features Projective Geometric Damped Convolutional Deep Belief Network (LFPGDCDBN) is introduced to improve the position estimation accuracy with minimum error. The proposed LFPGDCDBN technique includes two major processes namely dimensionality reduction and position estimation. First, the dimensionality reduction process is performed by projecting the principle features using Linear Helmert–Wolf blocked Sammon projection. After the feature selection, Geometric Levenberg–Marquardt Convolutional deep belief network is employed to estimate the position of the devices with higher accuracy and minimum error. The Convolutional deep belief network uses the triangulation geometric method to identify the position of the device in an indoor positioning system. Then the Levenberg–Marquardt function is a Damped least square method to minimize the squares of the deviations between the expected and observed results at the output unit. As a result, the LFPGDCDBN increases the positioning accuracy and minimizes the error rate. Experimental MATLAB assessment is carried out with various factors such as computational time, Computational space, positioning accuracy, and positing error. The experimental results and discussion indicate that the proposed LFPGDCDBN provides improved performance in terms of achieving higher positioning accuracy and minimum error as well as computational time when compared to the existing methods. The experimental results and discussion indicate that the proposed LFPGDCDBN increases the positioning accuracy by 47% and computational time, computational complexity, and reduces the positioning error by 45%, 29%, and 74% as compared to state-of-the-art works.

Precision estimation of 3D objects using an observation distribution model in support of terrestrial laser scanner network design

First order geometric network design is an important quality assurance process for terrestrial laser scanning of complex built environments for the construction of digital as-built models. A key design task is the determination of a set of instrument locations or viewpoints that provide complete site coverage while meeting quality criteria. Although simplified point precision measures are often used in this regard, precision measures for common geometric objects found in the built environment—planes, cylinders and spheres—are arguably more relevant indicators of as-built model quality. The computation of such measures at the design stage—which is not currently done—requires generation of artificial observations by ray casting, which can be a dissuasive factor for their adoption. This paper presents models for the rigorous computation of geometric object precision without the need for ray casting. Instead, a model for the 2D distribution of angular observations is coupled with candidate viewpoint-object geometry to derive the covariance matrix of parameters. Three-dimensional models are developed and tested for vertical cylinders, spheres and vertical, horizontal and tilted planes. Precision estimates from real experimental data were used as the reference for assessing the accuracy of the predicted precision—specifically the standard deviation—of the parameters of these objects. Results show that the mean accuracy of the model-predicted precision was 4.3% (of the read data value) or better for the planes, regardless of plane orientation. The mean accuracy of the cylinders was up to 6.2%. Larger differences were found for some datasets due to incomplete object coverage with the reference data, not due to the model. Mean precision for the spheres was similar, up to 6.1%, following adoption of a new model for deriving the angular scanning limits. The computational advantage of the proposed method over precision estimates from simulated, high-resolution point clouds is also demonstrated. The CPU time required to estimate precision can be reduced by up to three orders of magnitude. These results demonstrate the utility of the derived models for efficiently determining object precision in 3D network design in support of scanning surveys for reality capture.

Learning Sparse Geometric Features for Building Segmentation from Low-Resolution Remote-Sensing Images

High-resolution remote-sensing imagery has proven useful for building extraction. Unfortunately, due to the high acquisition costs and infrequent availability of high-resolution imagery, low-resolution images are more practical for large-scale mapping or change tracking of buildings. However, extracting buildings from low-resolution images is a challenging task. Compared with high-resolution images, low-resolution images pose two critical challenges in terms of building segmentation: the effects of fuzzy boundary details on buildings and the lack of local textures. In this study, we propose a sparse geometric feature attention network (SGFANet) based on multi-level feature fusion to address the aforementioned issues. From the perspective of the fuzzy effect, SGFANet enhances the representative boundary features by calculating the point-wise affinity of the selected feature points in a top-down manner. From the perspective of lacking local textures, we convert the top-down propagation from local to non-local by introducing the grounding transformer harvesting the global attention of the input image. SGFANet outperforms competing baselines on remote-sensing images collected worldwide and multiple sensors at 4 and 10 m resolution, thereby, improving the IoU by at least 0.66%. Notably, our method is robust and generalizable, which makes it useful for extending the accessibility and scalability of building dynamic tracking across developing areas (e.g., the Xiong’an New Area in China) by using low-resolution images.

EEG-based image classification using an efficient geometric deep network based on functional connectivity

EEG-based image classification using an efficient geometric deep network based on functional connectivity

Cycle optimization metric learning for few-shot classification

Metric learning methods are widely used in few-shot learning due to their simplicity and effectiveness. Most existing methods directly predict query labels by comparing the similarity between support and query samples. In this paper, we design a cycle optimization metric network for few-shot classification task that optimizes model performance based on loop-prediction of the labels of query samples and support samples. Specifically, we construct a forward network and reverse network based on a geometric algebra Graph Neural Network (GA-GNN). These two networks form the loop prediction from support samples to query samples and then back to support samples, guided by a cycle-consistency loss. We also introduce an optimization module that is able to correct the predicted results of query samples to further improve the network performance. Our extensive experimental results demonstrate that the proposed cycle optimization metric network outperforms existing state-of-the-art few-shot learning methods on classification tasks.

MsgeCNN: Multiscale geometric embedded convolutional neural network for ONFH segmentation and grading.

The incidence of osteonecrosis of the femoral head (ONFH) is increasing gradually, rapid and accurate grading of ONFH is critical. The existing Steinberg staging criteria grades ONFH according to the proportion of necrosis area to femoral head area. In the clinical practice, the necrosis region and femoral head region are mainly estimated by the observation and experience of doctor. This paper proposes a two-stage segmentation and grading framework, which can be used to segment the femoral head and necrosis, as well as todiagnosis. The core of the proposed two-stage framework is the multiscale geometric embedded convolutional neural network (MsgeCNN), which integrates geometric information into the training process and accurately segments the femoral head region. Then, the necrosis regions are segmented by the adaptive threshold method taking femoral head as the background. The area and proportion of the two are calculated to determine thegrade. The accuracy of the proposed MsgeCNN for femoral head segmentation is 97.73%, sensitivity is 91.17%, specificity is 99.40%, dice score is 93.34%. And the segmentation performance is better than the existing five segmentation algorithms. The diagnostic accuracy of the overall framework is 90.80%. The proposed framework can accurately segment the femoral head region and the necrosis region. The area, proportion, and other pathological information of the framework output provide auxiliary strategies for subsequent clinicaltreatment.

Geometric Interaction Graph Neural Network for Predicting Protein-Ligand Binding Affinities from 3D Structures (GIGN).

Predicting protein-ligand binding affinities (PLAs) is a core problem in drug discovery. Recent advances have shown great potential in applying machine learning (ML) for PLA prediction. However, most of them omit the 3D structures of complexes and physical interactions between proteins and ligands, which are considered essential to understanding the binding mechanism. This paper proposes a geometric interaction graph neural network (GIGN) that incorporates 3D structures and physical interactions for predicting protein-ligand binding affinities. Specifically, we design a heterogeneous interaction layer that unifies covalent and noncovalent interactions into the message passing phase to learn node representations more effectively. The heterogeneous interaction layer also follows fundamental biological laws, including invariance to translations and rotations of the complexes, thus avoiding expensive data augmentation strategies. GIGN achieves state-of-the-art performance on three external test sets. Moreover, by visualizing learned representations of protein-ligand complexes, we show that the predictions of GIGN are biologically meaningful.

Geometric algebra-based multiview interaction networks for 3D human motion prediction

3D skeleton-based human motion prediction is an essential and challenging task for human-machine interactions, which aims to forecasts future poses given a history of their previous motions. Recent works based on Graph Neural Networks (GCNs) show promising performance for motion prediction due to the powerful ability of feature aggregation of GCNs. However, with the deep and multi-stage GCN model deployment, its feature extraction mechanism tends to result in feature similarity over all joints, and degrade the prediction performance. In addition, such a graph structure in recent works was still insufficient to process the high dimensional structural data in Euclidean space when inference through multi-layer networks. To solve the problem, we propose a novel Geometric Algebra-based Multi-view Interaction network (GA-MIN), which captures and aggregates motion features from two interactions: 1) global-interaction, which refactors various spectrum dependencies using geometric algebra-based structure, and 2) self-interaction, which leverage self-attention mechanism to capture compact representations. Extensive experiments are conducted on three public datasets: Human3.6M, CMU Mocap, and 3DPW, which prove that the proposed GA-MIN outperforms state-of-the-art methods on 3D Mean Per Joint Position Error (MPJPE) and Mean Angle Error (MAE) on average.

Calcium oscillation on homogeneous and heterogeneous networks of ryanodine receptor.

Calcium oscillation is an important calcium homeostasis, imbalance of which is the key mechanism of initiation and progression of many major diseases. The formation and maintenance of calcium homeostasis are closely related to the spatial distribution of calcium channels on endoplasmic reticulum, whose complex structure was unveiled by recent observations with superresolution imaging techniques. In the current paper, a theoretical framework is established by abstracting the spatial distribution of the calcium channels as a nonlinear biological complex network with calcium channels as nodes and Ca^{2+} as edges. A dynamical model for a ryanodine receptor (RyR) is adopted to investigate the effect of spatial distribution on calcium oscillation. The mean-field model can be well reproduced from the complete graph and dense Erdös-Rényi network. The synchronization of RyRs is found important to generate a global calcium oscillation. Below a critical density of the Erdös-Rényi or BaraBási-Albert network, the amplitude and interspike interval decrease rapidly with the end of disappearance of oscillation due to the desynchronization. The clique graph with a cluster structure cannot produce a global oscillation due to the failure of synchronization between clusters. A more realistic geometric network is constructed in a two-dimensional plane based on the experimental information about the RyR arrangement of clusters and the frequency distribution of cluster sizes. Different from the clique graph, the global oscillation can be generated with reasonable parameters on the geometric network. The simulation also suggests that existence of small clusters and rogue RyRs plays an important role in the maintenance of global calcium oscillation through keeping synchronization between large clusters. Such results support the heterogeneous distribution of RyRs with different-size clusters, which is helpful to understand recent observations with superresolution nanoscale imaging techniques. The current theoretical framework can also be extent to investigate other phenomena in calcium signal transduction.

A Digital Twin Framework Embedded with POD and Neural Network for Flow Field Monitoring of Push-Plate Kiln

The push-plate kiln is a kind of kiln equipment widely used in the oxygen-free sintering of high-temperature alloy materials. Its flow field monitoring has an important application value for the manufacturing industry. However, traditional simulation methods cannot meet the requirements of real-time applications due to the high computational cost and being time-consuming. The rapid development of artificial intelligence technology will empower the traditional manufacturing industry. In this paper, we propose a data-driven digital twin framework for real-time flow field prediction by combining the CFD modeling simulation, IoT, and deep learning technologies. The framework integrates geometric, rule, physical, and neural network models to achieve the real-time simulation of physical and twin objects. The proper orthogonal decomposition (POD) and multiscale convolutional neural network (MCNN) are innovatively embedded into the framework. The POD is used to map high-dimensional data to low-dimensional features, and the MCNN is used to construct models predicting low-dimensional features for fast flow field prediction. The effectiveness of the proposed model is verified by the push-plate kiln case. The results show that the digital twin can quickly predict multi-physics fields based on the perceptual data to achieve the real-time evaluation of the operating state of the push-plate kiln.

Adaptive curvature exploration geometric graph neural network

Adaptive curvature exploration geometric graph neural network

Clustering constrained on linear networks

An unsupervised classification method for point events occurring on a geometric network is proposed. The idea relies on the distributional flexibility and practicality of random partition models to discover the clustering structure featuring observations from a particular phenomenon taking place on a given set of edges. By incorporating the spatial effect in the random partition distribution, induced by a Dirichlet process, one is able to control the distance between edges and events, thus leading to an appealing clustering method. A Gibbs sampler algorithm is proposed and evaluated with a sensitivity analysis. The proposal is motivated and illustrated by the analysis of crime and violence patterns in Mexico City.