Spatial Interaction Network Research Articles

Spatial Dynamic Interaction Effects and Formation Mechanisms of Air Pollution in the Central Plains Urban Agglomeration in China

Accurately identifying the dynamic interaction effects and network structure characteristics of air pollution is essential for effective collaborative governance. This study investigates the spatial dynamic interactions of air pollution among 30 cities in the Central Plains Urban Agglomeration using convergent cross mapping. Social network analysis is applied to assess the overall and node characteristics of the spatial interaction network, while key driving factors are analyzed using an exponential random graph model. The findings reveal that air pollution levels in the Central Plains Urban Agglomeration initially increase before they decrease, with heavily polluted cities transitioning from centralized to sporadic distribution. Among the interactions, Heze’s air pollution impact on Kaifeng was the strongest, while Xinxiang’s impact on Changzhi was the weakest. The emission and receiving effects peaked during 2010–2012. The air pollution interactions among cities exhibit significant network characteristics, with block model results indicating that emitting and receiving relationships are primarily concentrated in the bidirectional spillover plate. Natural factors such as temperature and precipitation significantly influence the spatial interaction network. Economic and social factors like economic level and industrial sector proportion also have a significant impact. However, population density does not influence the spatial interaction network. This study contributes to understanding the spatial network of air pollution, thereby enhancing strategies for optimizing regional collaborative governance efforts to address air pollution.

Coupling Hyperbolic GCN with Graph Generation for Spatial Community Detection and Dynamic Evolution Analysis

Spatial community detection is a method that divides geographic spaces into several sub-regions based on spatial interactions, reflecting the regional spatial structure against the background of human mobility. In recent years, spatial community detection has attracted extensive research in the field of geographic information science. However, mining the community structures and their evolutionary patterns from spatial interaction data remains challenging. Most existing methods for spatial community detection rely on representing spatial interaction networks in Euclidean space, which results in significant distortion when modeling spatial interaction networks; since spatial community detection has no ground truth, this results in the detection and evaluation of communities being difficult. Furthermore, most methods usually ignore the dynamics of these spatial interaction networks, resulting in the dynamic evolution of spatial communities not being discussed in depth. Therefore, this study proposes a framework for community detection and evolutionary analysis for spatial interaction networks. Specifically, we construct a spatial interaction network based on network science theory, where geographic units serve as nodes and interaction relationships serve as edges. In order to fully learn the structural features of the spatial interaction network, we introduce a hyperbolic graph convolution module in the community detection phase to learn the spatial and non-spatial attributes of the spatial interaction network, obtain vector representations of the nodes, and optimize them based on a graph generation model to achieve the final community detection results. Considering the dynamics of spatial interactions, we analyze the evolution of the spatial community over time. Finally, using taxi trajectory data as an example, we conduct relevant experiments within the fifth ring road of Beijing. The empirical results validate the community detection capabilities of the proposed method, which can effectively describe the dynamic spatial structure of cities based on human mobility and provide an effective analytical method for urban spatial planning.

Downscaling spatial interaction with socioeconomic attributes

A variety of complex socioeconomic phenomena, for example, migration, commuting, and trade can be abstracted by spatial interaction networks, where nodes represent geographic locations and weighted edges convey the interaction and its strength. However, obtaining fine-grained spatial interaction data is very challenging in practice due to limitations in collection methods and costs, so spatial interaction data such as transportation data and trade data are often only available at a coarse scale. Here, we propose a gravity downscaling (GD) method based on readily accessible socioeconomic data and the gravity law to infer fine-grained interactions from coarse-grained data. GD assumes that interactions of different spatial scales are governed by the similar gravity law and thus can transfer the parameters estimated from coarse-grained regions to fine-grained regions. Results show that GD has an average improvement of 24.6% in Mean Absolute Percentage Error over alternative downscaling methods (i.e., the areal-weighted method and machine learning models) across datasets with different spatial scales and in various regions. Using simple assumptions, GD enables accurate downscaling of spatial interactions, making it applicable to a wide range of fields, including human mobility, transportation, and trade.

Towards zero-shot object counting via deep spatial prior cross-modality fusion

Existing counting models predominantly operate on a specific category of objects, such as crowds and vehicles. The recent emergence of multi-modal foundational models, e.g., Contrastive Language-Image Pre-training (CLIP), has facilitated class-agnostic counting. This involves counting objects of any given class from a single image based on textual instructions. However, CLIP-based class-agnostic counting models face two primary challenges. Firstly, the CLIP model lacks sensitivity to location information. It generally considers global content rather than the fine-grain location of objects. Therefore, adapting the CLIP model directly is suboptimal. Secondly, these models generally freeze pre-trained vision and language encoders, while neglecting the potential misalignment in the constructed hypothesis space. In this paper, we address these two issues in a unified framework termed Deep Spatial Prior Interaction (DSPI) network. The DSPI leverages the spatial-awareness ability of large-scale pre-trained object grounding models, i.e., Grounding DINO, to incorporate spatial location as an additional prior for a specific query class. This enables the network to be more specifically focused on the precise location of the objects. Additionally, to align the feature space across different modalities, we tailor a meta adapter that extracts textual information into an object query. This serves as an instruction for cross-modality matching. These two modules collaboratively ensure the alignment of multi-modal representations while preserving their discriminative nature. Comprehensive experiments conducted on a diverse set of benchmarks verify the superiority of the proposed model. The code is available at https://github.com/jinyongch/DSPI.

HorSR: High-order spatial interactions and residual global filter for efficient image super-resolution

Recent advances in efficient image super-resolution (EISR) include convolutional neural networks, which exploit distillation and aggregation strategies with copious channel split and concatenation operations to fully exploit limited hierarchical features. In contrast, the Transformer network presents a challenge for EISR because multiheaded self-attention is a computationally demanding process. To respond to this challenge, this paper proposes replacing multiheaded self-attention in the Transformer network with global filtering and recursive gated convolution. This strategy allows us to design a high-order spatial interaction and residual global filter network for efficient image super-resolution (HorSR), which comprises three components: a shallow feature extraction module, a deep feature extraction module, and a high-quality image-reconstruction module. In particular, the deep feature extraction module comprises residual global filtering and recursive gated convolution blocks. The experimental results show that the HorSR network provides state-of-the-art performance with the lowest FLOPs of existing EISR methods.

DRSI-Net: Dual-residual spatial interaction network for multi-person pose estimation

Multi-person pose estimation (MPPE), which aims to locate the key points for all persons in the frames, is an active research branch of computer vision. Variable human poses and complex scenes make MPPE dependent on local details and global structures; their absence may cause key point feature misalignment. In this case, high-order spatial interactions that can effectively link the local and global information of features are particularly important. However, most methods do not include spatial interactions. A few methods have low-order spatial interactions, but achieving a good balance between accuracy and complexity is challenging. To address the above problems, a dual-residual spatial interaction network (DRSI-Net) for MPPE with high accuracy and low complexity is proposed herein. Compared to other methods, DRSI-Net recursively performs residual spatial information interactions on the neighbouring features so that more useful spatial information can be retained and more similarities can be obtained between shallow and deep extracted features. The channel and spatial dual attention mechanism introduced in the multi-scale feature fusion also helps the network to adaptively focus on features relevant to the target key points and further refine the generated poses. Simultaneously, by optimising the interactive channel dimensions and dividing the gradient flow, the spatial interaction module is designed to be lightweight, thus reducing the complexity of the network. According to the experimental results on the COCO dataset, the proposed DRSI-Net outperforms other state-of-the-art methods in accuracy and complexity.

Unveiling spatial complexity in solid tumor immune microenvironments through multiplexed imaging.

Deciphering cellular components and the spatial interaction network of the tumor immune microenvironment (TIME) of solid tumors is pivotal for understanding biologically relevant cross-talks and, ultimately, advancing therapies. Multiplexed tissue imaging provides a powerful tool to elucidate spatial complexity in a holistic manner. We established and cross-validated a comprehensive immunophenotyping panel comprising over 121 markers for multiplexed tissue imaging using MACSima™ imaging cyclic staining (MICS) alongside an end-to-end analysis workflow. Applying this panel and workflow to primary cancer tissues, we characterized tumor heterogeneity, investigated potential therapeutical targets, conducted in-depth profiling of cell types and states, sub-phenotyped T cells within the TIME, and scrutinized cellular neighborhoods of diverse Tcell subsets. Our findings highlight the advantage of spatial profiling, revealing immunosuppressive molecular signatures of tumor-associated myeloid cells interacting with neighboring exhausted, PD1high T cells in the TIME of hepatocellular carcinoma (HCC). This study establishes a robust framework for spatial exploration of TIMEs in solid tumors and underscores the potency of multiplexed tissue imaging and ultra-deep cell phenotyping in unraveling clinically relevant tumor components.

Towards semantic enrichment for spatial interactions

ABSTRACT Various big geo-data provide a social sensing approach to measure spatial interactions. Existing studies often aggregate individual-level movement trajectories or social ties to obtain the interaction intensity between places, neglecting the detailed meanings (i.e. the semantics) behind spatial interactions. However, such meanings help to understand the relationship between two places, and consequently, the characteristics of both places. We argue that semantics can be extracted from spatial interactions through features of space, time, symmetry, and individual-based statistics. Whereafter the calculation and applications of the features are given. Furthermore, we discuss the construction of spatial interaction networks with semantics, as well as approaches to representing places according to spatial interactions. Finally, we illustrate the potential value of spatial interaction semantics in facilitating decision-making through an example in the context of tourism planning.

Evaluation of the Development Level of Integration of Culture and Tourism in China’s Provinces under the Perspective of Spatial Interaction and Study of Influence Paths

In the new stage of the deep integration of culture and tourism, fully exploring its mechanism of action, spatial connection, and path of influence will be of great significance to the sustainable development of the integration of culture and tourism. In this paper, we analysed the mechanism of integration of culture and tourism with the help of systems theory, introduced the concept of spatial interaction to construct a new measurement model of integration of culture and tourism, and conducted empirical analyses with Chinese provinces as the target. We found that in the last decade, the tourism industry has been dominant, and most of the industrial elements have been clustered towards it. The centre of the spatial interaction network of culture and tourism is located in the area of the “North China Plain—Middle and Lower Yangtze River Plain” and moves towards the southwest over time. The northwestern and northeastern zones have become “desert zones” in the spatial interaction network of culture and tourism. The level of integration of culture and tourism peaked in 2018, with East China having a chronically higher level of integration. In addition, the deep integration of culture and tourism can be promoted through the combination of new development philosophies.

Detection of peach soluble solids based on near-infrared spectroscopy with High Order Spatial Interaction network.

Due to the scalability of deep learning technology, researchers have applied it to the non-destructive testing of peach internal quality. In addition, the soluble solids content (SSC) is an important internal quality indicator that determines the quality of peaches. Peaches with high SSC have a sweeter taste and better texture, making them popular in the market. Therefore, SSC is an important indicator for measuring peach internal quality and making harvesting decisions. This article presents the High Order Spatial Interaction Network (HOSINet), which combines the Position Attention Module (PAM) and Channel Attention Module (CAM). Additionally, a feature wavelength selection algorithm similar to the Group-based Clustering Subspace Representation (GCSR-C) is used to establish the Position and Channel Attention Module-High Order Spatial Interaction (PC-HOSI) model for peach SSC prediction. The accuracy of this model is compared with traditional machine learning and traditional deep learning models. Finally, the permutation algorithm is combined with deep learning models to visually evaluate the importance of feature wavelengths. Increasing the order of the PC-HOSI model enhances its ability to learn spatial correlations in the dataset, thus improving its predictive performance. The optimal model, PC-HOSI model, performed well with an order of 3 (PC-HOSI-3), with a root mean square error of 0.421 °Brix and a coefficient of determination of 0.864. Compared with traditional machine learning and deep learning algorithms, the coefficient of determination for the prediction set was improved by 0.07 and 0.39, respectively. The permutation algorithm also provided interpretability analysis for the predictions of the deep learning model, offering insights into the importance of spectral bands. These results contribute to the accurate prediction of SSC in peaches and support research on interpretability of neural network models for prediction. © 2024 Society of Chemical Industry.

TMIC-20. A SPATIALLY RESOLVED HUMAN GLIOBLASTOMA ATLAS REVEALS DISTINCT CELLULAR AND MOLECULAR PATTERNS OF ANATOMICAL NICHES

Abstract Glioblastoma (GBM) is a highly plastic ecosystem where the complex interplay between different cellular components contributes to disease progression. Although single-cell RNA (scRNA)-seq has revealed remarkable cellular heterogeneity of GBM, our knowledge regarding the spatial organization of its cellular components is currently lacking. Here we created a comprehensive dataset of 115,914 spatial transcriptomes across 32 tissue sections with matched multi-omics profiling on a set of genotyped glioma samples. We present spatial maps of 56 fine-grained cellular components, including previously unrecognized subtypes of oligodendrocytes and stromal cells, and their spatial interaction networks in each GBM- associated anatomical niche. Furthermore, the deconvolution of bulk RNA-seq data using the integrated spatial and single-cell atlas revealed clinically relevant GBM ecotypes. Our data provides comprehensive insights into the cellular architecture of GBM at high spatial resolution. It will be a valuable resource to develop effective combinatorial therapies to target all tumor-fostering niches simultaneously.

Quantitative and qualitative interpretation of community partitions by map overlaying and calculating the distribution of related geographical features

Abstract Applying community detection algorithms in spatial interaction networks constructed from modern human communication records is an essential means of evaluating urban territorial subdivisions. Previous studies have usually involved qualitative rather than quantitative interpretations of community detection results. This article proposes a method of quantitatively and qualitatively interpreting community partition results by map overlaying the spatial regions corresponding to the detected communities with the related geographical features and by calculating the distribution of the geographical features contained in the regions and the entropy value of each distribution. The interpretation of the communities detected from the spatial interaction networks is carried out from the perspective of multi-temporal and multi-spatial scales and multi-geographical features. Extensive experiments were conducted with Milan, Italy, as the study area. The spatial interaction records reflected by telephone calls, land use, and point of interest (POI) data were used as the experimental data. Experimental results demonstrated the effectiveness of our method, and the specific results include: (1) Qualitative interpretation of multi-spatial resolution scale communities detected from the long-term aggregated spatial interaction network. The cohesiveness, homogeneity, and heterogeneity of the detected communities were qualitatively interpreted by the spatial distribution patterns of the land use dataset and the POI dataset. (2) Quantitative interpretation of multi-spatial resolution scale communities detected from the long-term aggregated spatial interaction network. The low spatial resolution scale community partitions and the high spatial resolution scale community partitions were interpreted through the statistical distribution of the land use dataset and the POI dataset, respectively. (3) Qualitative interpretation of the stable and active regions discovered from the community time series. Regardless of the community partitions’ spatial resolution scales, the stable and active regions were distinguished with the statistical distributions of the land use dataset and the POI dataset.

Analyzing the Spatial Interaction Characteristics of Urban Area Shared Bicycle Systems: A Case Study of Beijing’s Central Area

Shared bicycle systems play a crucial role in promoting sustainable urban transportation, addressing challenges such as traffic congestion and air pollution. Understanding the spatiotemporal patterns of shared bike usage is essential for optimizing bike-sharing infrastructure and improving transportation planning. In this study, we analyzed 2.4 million records of shared bicycle data to explore the spatial distribution, interaction patterns, and flow dynamics within Beijing’s urban central area. We found that bike distribution peaks during commuting hours, particularly in central regions with employment centers. Complex networks are an important method for studying travel flows. Through a spatial interaction network, we identified key streets with high node strength and popularity, often concentrated in central areas. They experience heavy shared bicycle use during peak hours due to their employment-centric location. Conversely, peripheral areas see increased usage in the evenings, reflecting distinct commuting patterns. The morning exhibits higher positive central values compared to the evening, while negative values show the opposite trend. Based on these findings, we recommend enhancing bike infrastructure in high-density areas with bike lanes and ample shared bikes during peak hours. Implementing mixed-use zoning policies in the central region can reduce traffic congestion. Expanding shared bike services to peripheral regions can promote equitable access. This research underscores the importance of considering spatial and temporal factors in urban transportation planning. Future work should incorporate additional data sources, explore environmental impacts, and analyze usage in different seasons and special events, further contributing to sustainable urban mobility development.

Study on Community Detection Method for Morning and Evening Peak Shared Bicycle Trips in Urban Areas: A Case Study of Six Districts in Beijing

Examining the clustering characteristics and fluctuations within urban areas during peak hours through the lens of bike-sharing is of utmost importance in the optimization of bike-sharing systems and urban transportation planning. This investigation adopts the principles of urban spatial interaction network construction and employs streets as the fundamental units of analysis to model bike-sharing activities during morning and evening peak hours within Beijing’s six central districts. Subsequent to this, a comprehensive analysis of the network’s structural attributes was carried out. A Walktrap method, rooted in modularity analysis, was introduced to discern and scrutinize the clustering patterns and characteristics of communities within the network across different temporal intervals. Empirical findings reveal a predominant usage pattern of shared bicycles for short-distance travel during both morning and evening peak hours. Notably, distinctive community structures manifest during these periods, characterized by two large communities and multiple smaller ones during the morning peak, while the evening peak showcases a single large community alongside several medium-sized and smaller ones. Moreover, the extended interaction radius points to an expanded geographic range of interactions among streets. These findings bear significant implications for the management of urban transportation, bike-sharing enterprises, and urban residents, proffering valuable insights for the optimization of bike-sharing schemes and transportation strategies. These research findings not only contribute to enhancing urban transportation planning and bike-sharing systems but also provide robust guidance for advancing more efficient and sustainable urban transportation solutions, thereby fostering the sustainable development of cities.

Multi-order spatial interaction network for human pose estimation

Recent vision Transformer has been applied to human pose estimation and has achieved excellent performance by two-order spatial interaction with self-attention. However, it is still unclear whether higher-order spatial interaction can facilitate pose estimation. In this paper, we propose a novel approach based on multi-order spatial interactions and confirm that the combination of different orders is beneficial for human pose estimation task. We first build a Triple Interaction Module (TIM) by pure convolutions to make spatial information interactions three times. In contrast to Transformer, the TIM is compatible with several pure convolutions and extends two-order interaction in Transformer to triple-order without extensive additional computation, which makes it easier to explore inter-related features between keypoints in the human body. In addition, we combine TIM with traditional CNN and Transformer to form Multi-order Spatial Interaction Network (MSIN). This paper takes advantage of MSIN to extract keypoint heatmaps and certifies that the order-by-order structure can enhance the overall performance of locating human keypoints. Experimental results demonstrate that MSIN performs favorably against the most state-of-the-art CNN-based and Transformer-based counterparts on the COCO and MPII datasets, while being more lightweight.

Tumor-associated macrophages trigger MAIT cell dysfunction at the HCC invasive margin

Mucosal-associated invariant T (MAIT) cells represent an abundant innate-like Tcell subtype in the human liver. MAIT cells are assigned crucial roles in regulating immunity and inflammation, yet their role in liver cancer remains elusive. Here, we present a MAIT cell-centered profiling of hepatocellular carcinoma (HCC) using scRNA-seq, flow cytometry, and co-detection by indexing (CODEX) imaging of paired patient samples. These analyses highlight the heterogeneity and dysfunctionality of MAIT cells in HCC and their defective capacity to infiltrate liver tumors. Machine-learning tools were used to dissect the spatial cellular interaction network within the MAIT cell neighborhood. Co-localization in the adjacent liver and interaction between niche-occupying CSF1R+PD-L1+ tumor-associated macrophages (TAMs) and MAIT cells was identified as a key regulatory element of MAIT cell dysfunction. Perturbation of this cell-cell interaction in exvivo co-culture studies using patient samples and murine models reinvigorated MAIT cell cytotoxicity. These studies suggest that aPD-1/aPD-L1 therapies target MAIT cells in HCC patients.

SI-Net: spatial interaction network for deepfake detection

SI-Net: spatial interaction network for deepfake detection

Verification of Geographic Laws Hidden in Textual Space and Analysis of Spatial Interaction Patterns of Information Flow

The rapid development of Internet technology has formed a huge virtual information space. In the information space, information flow has become a link of communication between objects. Information flow is an alternative or supplement to the traditional physical flow for the study of the spatial interaction of geographical entities. The research uses toponym co-occurrence and search index as information flow data, verifies the geographical laws hidden in the information space by spatial autocorrelation analysis and gravity model fitting, and analyzes the spatial interaction patterns of provinces in China in the information space by complex network analysis methods. The results show that: (1) information flow in the information space obeys Tobler’s first law of geography and Goodchild’s second law of geography. The spatial interaction represented by information flow has a distance decay effect. The best distance decay coefficients for toponym co-occurrence and the search index are 0.189 and 0.186, respectively. (2) The inter-provincial spatial interaction network of China shows a hierarchical pattern of the triangular primary network and diamond secondary network, and the ranking of provinces in the centrality analysis is basically stable, but the network hierarchy is deepening. The gravity center of spatial interaction is located in the east-central region of China. (3) The information flow-based interaction network is of higher asymmetry than the population mobility network, and its spatial structure is also obvious. This research provides a new idea for studying the spatial interaction of geographical entities in the physical world from the perspective of information flow.

Exploring Public Transportation Supply–Demand Structure of Beijing from the Perspective of Spatial Interaction Network

A comprehensive understanding of the relationship between public transportation supply and demand is crucial for the construction and sustainable development of urban transportation. Due to the spatial and networked nature of public transportation, revealing the spatial configuration and structural disparities between public transportation supply and demand networks (TSN and TDN) can provide significant insights into complex urban systems. In this study, we explored the spatial configuration and structural disparities between TSN and TDN in the complex urban environment of Beijing. By constructing subdistrict-scale TSN and TDN using urban public transportation operation data and mobile phone data, we analyzed the spatial characteristics and structural disparities of these networks from various dimensions, including global indicators, three centralities, and community structure, and measured the current public transportation supply and demand matching pattern in Beijing. Our findings revealed strong structural and geographic heterogeneities of TSN and TDN, with significant traffic supply–demand mismatch being observed in urban areas within the Sixth Ring Road. Moreover, based on the percentage results of supply–demand matching patterns, we identified that the current public transportation supply–demand balance in Beijing is approximately 64%, with around 18% of both excess and shortage of traffic supply. These results provide valuable insights into the structure and functioning of public transportation supply–demand networks for policymakers and urban planners; these can be used to facilitate the development of a sustainable urban transportation system.

Spatiotemporal trends in floral visitation and interaction networks reveal shifting niches for bats in a Neotropical savanna.

Flower-vising bats are important components of tropical pollinator communities, yet little is known about the structure of their pollination networks and how resource availability through time (seasons) and space (habitat heterogeneity) affects the extent to which bats interact with plants within a community-wide context. This information is key for the conservation of threatened nectarivore species, such as the Cerrado-endemic Lonchophylla dekeyseri, for which data on its specialization on floral-resources is scarce. Within a seasonal and heterogeneous savanna in the central Brazilian Cerrado, we performed a year-round assessment of an inclusive assemblage of flower-visiting bats (both nectarivores and other guilds that can also feed on nectar) within a savanna-edge-forest gradient, the phenological trends and spatial distribution of bat and their resource plants, and the resultant temporal and spatial interaction networks between bats and plants in order to associate network structure to resource availability. Clear spatiotemporal trends emerged in the community. Nectarivores dominated the flower-visiting niche outside forests and were prolific floral visitors, resulting in networks with lower specialization and modularity. These bats diverged into savanna foragers active during the wet season and the wet-dry transition, and edge foragers active mostly during the dry season. The latter group encompassed L. dekeyseri, which visited mostly Bauhinia species. Frugivores took over as main floral visitors within forests, as well as during peak dry season, when fewer fruits were available, resulting in more specialized and modular networks. Our work shows that the turnover of floral resources across seasons and vegetation types has a defining role in bat-plant interactions and relates to network structure, as bat trophic guilds interact with plants in distinct habitats and times of the year. Frugivores dominate the flower-visiting niche in certain temporal and spatial subsets of the network, which calls for the inclusion of this guild in future studies. Moreover, the high visitation to Bauhinia species by L. dekeyseri during the dry season might reduce competition with other nectarivores and is relevant to the management of the species, although more data is needed on its resource consumption on a larger time frame and across its geographic range.