Spatiotemporal Clustering Patterns Research Articles

Epidemiological Characteristics and Spatiotemporal Clustering of Symptomatic Hepatitis E Virus Reinfection in Zhejiang Province, 2005–2023

Hepatitis E virus (HEV) reinfection is prevalent among the population, posing a significant burden on prevention and control efforts. In this study, we conducted a comprehensive analysis of data from China’s Disease Prevention and Control Information System’s infectious disease surveillance system to identify the epidemiological characteristics, spatiotemporal clustering, and high-risk populations of HEV reinfection. From 2005 to 2023, HEV reinfection in Zhejiang Province exhibited a fluctuating trend, peaking in 2020, with a 3–5-year lag compared to the pattern of HEV incidence. The Cox model indicated that individuals aged 40–50 and females are at higher risk of reinfection. Spatial autocorrelation was observed in reinfection cases from 2011 to 2016, with high–high clustering areas concentrated in downtown Hangzhou. Additionally, spatiotemporal scanning revealed that the clustering of reinfection cases has shifted from Hangzhou to coastal areas in recent years. Our findings suggest that targeted prevention and control measures for HEV rein fection should be implemented based on the characteristics of high-risk populations and spatiotemporal clustering patterns.

Spatiotemporal Epidemiological Trends of Mpox in Mainland China: Spatiotemporal Ecological Comparison Study.

The World Health Organization declared mpox an international public health emergency. Since January 1, 2022, China has been ranked among the top 10 countries most affected by the mpox outbreak globally. However, there is a lack of spatial epidemiological studies on mpox, which are crucial for accurately mapping the spatial distribution and clustering of the disease. This study aims to provide geographically accurate visual evidence to determine priority areas for mpox prevention and control. Locally confirmed mpox cases were collected between June and November 2023 from 31 provinces of mainland China excluding Taiwan, Macao, and Hong Kong. Spatiotemporal epidemiological analyses, including spatial autocorrelation and regression analyses, were conducted to identify the spatiotemporal characteristics and clustering patterns of mpox attack rate and its spatial relationship with sociodemographic and socioeconomic factors. From June to November 2023, a total of 1610 locally confirmed mpox cases were reported in 30 provinces in mainland China, resulting in an attack rate of 11.40 per 10 million people. Global spatial autocorrelation analysis showed that in July (Moran I=0.0938; P=.08), August (Moran I=0.1276; P=.08), and September (Moran I=0.0934; P=.07), the attack rates of mpox exhibited a clustered pattern and positive spatial autocorrelation. The Getis-Ord Gi* statistics identified hot spots of mpox attack rates in Beijing, Tianjin, Shanghai, Jiangsu, and Hainan. Beijing and Tianjin were consistent hot spots from June to October. No cold spots with low mpox attack rates were detected by the Getis-Ord Gi* statistics. Local Moran I statistics identified a high-high (HH) clustering of mpox attack rates in Guangdong, Beijing, and Tianjin. Guangdong province consistently exhibited HH clustering from June to November, while Beijing and Tianjin were identified as HH clusters from July to September. Low-low clusters were mainly located in Inner Mongolia, Xinjiang, Xizang, Qinghai, and Gansu. Ordinary least squares regression models showed that the cumulative mpox attack rates were significantly and positively associated with the proportion of the urban population (t0.05/2,1=2.4041 P=.02), per capita gross domestic product (t0.05/2,1=2.6955; P=.01), per capita disposable income (t0.05/2,1=2.8303; P=.008), per capita consumption expenditure (PCCE; t0.05/2,1=2.7452; P=.01), and PCCE for health care (t0.05/2,1=2.5924; P=.01). The geographically weighted regression models indicated a positive association and spatial heterogeneity between cumulative mpox attack rates and the proportion of the urban population, per capita gross domestic product, per capita disposable income, and PCCE, with high R2 values in north and northeast China. Hot spots and HH clustering of mpox attack rates identified by local spatial autocorrelation analysis should be considered key areas for precision prevention and control of mpox. Specifically, Guangdong, Beijing, and Tianjin provinces should be prioritized for mpox prevention and control. These findings provide geographically precise and visualized evidence to assist in identifying key areas for targeted prevention and control.

Genome analysis of Streptococcus spp. isolates from animals in pre-antibiotic era with respect to antibiotic susceptibility and virulence gene profiles.

Lyophilized Streptococcus spp. isolates(n = 50) from animal samples submitted to the diagnostic laboratory at the University of Connecticut in the 1940s were revivified to investigate the genetic characteristics using whole-genome sequencing (WGS). The Streptococcus spp. isolateswere identified as follows; S. agalactiae (n = 14), S. dysgalactiae subsp. dysgalactiae (n = 10), S. dysgalactiae subsp. equisimils (n = 5), S. uberis (n = 8), S. pyogenes (n = 7), S. equi subsp. zooepidemicus (n = 4), S. oralis (n = 1), and S. pseudoporcinus (n = 1). We identified sequence types (ST) of S. agalactiae, S. dysgalactiae, S. uberis, S. pyogenes, and S. equi subsp. zooepidemicus and reported ten novel sequence types of those species. WGS analysis revealed that none of Streptococcus spp. carried antibiotic resistance genes. However, tetracycline resistance was observed in four out of 15 S. dysgalactiae isolates and in one out of four S. equi subsp. zooepidemicus isolate. This data highlights that antimicrobial resistance is pre-existed in nature before the use of antibiotics. The draft genome sequences of isolates from this study and 426 complete genome sequences of Streptococcus spp. downloaded from BV-BRC and NCBI GenBank database were analyzed for virulence gene profiles and phylogenetic relationships. Different Streptococcus species demonstrated distinct virulence gene profiles, with no time-related variations observed. Phylogenetic analysis revealed high genetic diversity of Streptococcus spp. isolates from the 1940s, and no clear spatio-temporal clustering patterns were observed among Streptococcus spp. analyzed in this study. This study provides an invaluable resource for studying the evolutionary aspects of antibiotic resistance acquisition and virulence in Streptococcus spp.

Spatiotemporal cluster analysis of reputable tourist accommodation in Greater Amman Municipality, Jordan

研究目的为了测试社交旅游网站 (STNS) 的用户生成内容 (UGC) 对在线声誉管理 (ORM) 的适用性, 本研究分析了知名酒店的空间聚类（基于 TripAdvisor 最佳价值指标) 和信誉良好的户外座位 (ODS) 餐厅(基于排名指标)。研究设计/方法/途径该研究使用数据挖掘技术从 TripAdvisor 获取 UGC。 基于（HDBSCAN）算法的分层基于密度的空间聚类方法用于鲁棒聚类分析。研究发现调查结果显示, 最具价值 (BV) 酒店和信誉良好的 ODS 餐厅最有可能位于人口和经济活动较为密集的城市旅游目的地的中心区及其周边地区。 BV 酒店的时空聚类分析形成了不同大小、密度和形状模式的聚类。研究原创性目前的文献扔缺乏专注于分析利用 UGC 的酒店和餐厅 (H&R) 空间聚类的研究。 因此, 本研究首次绘制并分析了知名酒店（TripAdvisor BV 指标）和餐厅（排名指标）的时空聚类模式。 因此, 本研究通过利用数据挖掘和 HDBSCAN 算法显示 H&Rs 聚类的模式变化, 为城市旅游研究做出了重要的方法论贡献。理论意义这项研究表明, 著名的 H&R 集中在历史悠久的市中心附近的地区。 这应该是对城市投资环境的应用研究的推动力。实践意义研究结果将为企业家和潜在投资者提供最具吸引力的旅游投资环境的理性指导。

Identification of Emerging Roadkill Hotspots on Korean Expressways Using Space-Time Cubes.

Collisions with wild animals on high-speed expressways not only lead to roadkill but can also cause accidents that incur considerable human and economic costs. Based on roadkill data from 2004-2019 for four common wildlife species involved in collisions with vehicles on expressways in Korea (water deer, common raccoon dog, Korean hare, and wild boar), the present study conducted optimized hotspot analysis and identified spatiotemporal patterns using a space-time cube (STC) approach. Temporal and spatial differences in the roadkill data were observed between species. Water deer were the most common roadkill species of the four studied, with hotspots in the southern region of the capital area, in the Chungnam region, and in the western Chungbuk and Gangwon-do regions. However, the instances of water deer roadkill over time differed between each region. In addition, it was found that the number of cases of wild boar roadkill has increased recently. In particular, a number of new hotspot areas were observed centered on the metropolitan area Gyeonggi-do, which contains a high population and significant infrastructure. Overall, the emerging hotspot analysis based on STCs was able to determine cold spot and hotspot trends over time, allowing for a more intuitive understanding of spatiotemporal clustering patterns and associated changes than cumulative density-oriented hotspot analysis. As a result, it becomes easier to analyze the causes of roadkill and to establish reduction measures according to priority.

Spatiotemporal clustering patterns and sociodemographic determinants of COVID-19 (SARS-CoV-2) infections in Helsinki, Finland

This study aims to elucidate the variations in spatiotemporal patterns and sociodemographic determinants of SARS-CoV-2 infections in Helsinki, Finland. Global and local spatial autocorrelation were inspected with Moran's I and LISA statistics, and Getis-Ord Gi* statistics was used to identify the hot spot areas. Space-time statistics were used to detect clusters of high relative risk and regression models were implemented to explain sociodemographic determinants for the clusters. The findings revealed the presence of spatial autocorrelation and clustering of COVID-19 cases. High–high clusters and high relative risk areas emerged primarily in Helsinki's eastern neighborhoods, which are socioeconomically vulnerable, with a few exceptions revealing local outbreaks in other areas. The variation in COVID-19 rates was largely explained by median income and the number of foreign citizens in the population. Furthermore, the use of multiple spatiotemporal analysis methods are recommended to gain deeper insights into the complex spatiotemporal clustering patterns and sociodemographic determinants of the COVID-19 cases.

1442. Spatiotemporal Clusters of Varicella and the Regional Risks through Bayesian Approach: A National Five-year Cohort Analysis

Abstract Background Since varicella epidemics repeatedly occurred in Korea, it is essential to control varicella outbreaks preemptively in the targeted region. Therefore, we aimed to reveal spatiotemporal clusters of varicella and the regional risk factor of varicella incidence at the national level. Methods All varicella cases (defined as ICD-10 codes, B01-B09) from 2013 to 2017 in Korea were extracted from National Health Insurance Service. Of the total, 566,978 cases were realigned spatially by 250 administrative districts of Korea and temporally by a week. Spatial autocorrelation was tested using the global Moran’s I statistics using Monte Carlo simulation. Kulldorff’s prospective space-time scan statistics were used to reveal space-time clusters of varicella. Possible risk factors were extracted from the Korean Statistical Information Service and Community Health Survey of Korea, including hand hygiene perceptions, alcohol and smoking status, the proportion of children under 15 years old, the number of households, and household income by regions. After selecting significant risk factors through non-spatial generalized linear models, a conditional autoregressive spatiotempoal model with Bayesian extension was applied to estimate the regional factors affecting varicella incidence. Results There was spatial autocorrelation using Global Moran’s I statistics (P< 0.01). When the maximum cluster size was limited to 10% of the population, 17 spatiotemporal clusters were detected in specific regions in Korea (figure 1). Low perception of hand hygiene, the high proportion of alcohol drinking and cigarette smoking, high children proportion, low number of familial member, and low household income were associated with varicella spatiotemporal incidence (odds ratio: 0.97, 1.01, 2.31, 1.10, 0.99, 0.99, respectively; 95% credible intervals of all risk factors did not include 1). Figure 1. Space-time prospective clusters of varicella in Korea using varicella incidence from 2013 to 2017. Relative risks ratio of each cluster is described at the point. Conclusion Varicella incidence shows spatiotemporal clustering patterns in specific regions. Since regional factors such as the perception rate of hand hygiene, child proportion, alcohol drinking, cigarette smoking, and low household income affect varicella’s spatiotemporal incidence, strategies for targeted control of high-risk regions are strongly recommended. Disclosures All Authors: No reported disclosures

Visualizing Spatiotemporal Epidemic Clusters on a Map-based Dashboard: A case study of early COVID-19 cases in Singapore

Abstract. Spatiotemporal distribution of the epidemic data plays an important role in its understanding and prediction. In order to understand the transmission patterns of infectious diseases in a more intuitive way, many works applied various visualizations to show the epidemic datasets. However, most of them focus on visualizing the epidemic information at the overall level such as the confirmed counts each country, while spending less effort on powering user to effectively understand and reason the very large and complex epidemic datasets through flexible interactions. In this paper, the authors proposed a novel map-based dashboard for visualizing and analyzing spatiotemporal clustering patterns and transmission chains of epidemic data. We used 102 confirmed cases officially reported by the Ministry of Health in Singapore as the test dataset. This experiment shown that the well-designed and interactive map-based dashboard is effective in shorten the time that users required to mine the spatiotemporal characteristics and transmission chains behind the textual and numerical epidemic data.

Spatiotemporal Characteristics and Risk Factors of the COVID-19 Pandemic in New York State: Implication of Future Policies

The Coronavirus disease 2019 (COVID-19) has been spreading in New York State since March 2020, posing health and socioeconomic threats to many areas. Statistics of daily confirmed cases and deaths in New York State have been growing and declining amid changing policies and environmental factors. Based on the county-level COVID-19 cases and environmental factors in the state from March to December 2020, this study investigates spatiotemporal clustering patterns using spatial autocorrelation and space-time scan analysis. Environmental factors influencing the COVID-19 spread were analyzed based on the Geodetector model. Infection clusters first appeared in southern New York State and then moved to the central western parts as the epidemic developed. The statistical results of space-time scan analysis are consistent with those of spatial autocorrelation analysis. The analysis results of Geodetector showed that both temperature and population density were strong indications of the monthly incidence of COVID-19, especially in March and April 2020. There is a trend of increasing interactions between various risk factors. This study explores the spatiotemporal pattern of COVID-19 in New York State over ten months and explains the relationship between the disease transmission and influencing factors.

A global perspective on the sub-seasonal clustering of precipitation extremes

The occurrence of several precipitation extremes over sub-seasonal time windows can have major impacts on human societies, leading for instance to floods. Here, we apply a simple statistical framework based on Ripley’s K function, at a global scale and for each season separately, to identify regions where precipitation extremes tend to cluster in time over timescales of a few days to a few weeks. We analyze several observational and reanalysis datasets, as well as output from CMIP6 Global Climate Models (GCMs). Good agreement is found on the spatio-temporal clustering patterns across datasets. Sub-seasonal temporal clustering is largely concentrated over the tropical oceans, where it can be detected year-round. It is also significant over certain tropical lands, like Eastern Africa, and seasonally outside the tropics in several regions, most notably around the eastern subtropical oceans (Iberian Peninsula and Western North America during the DJF and MAM seasons) Southwest Asia (especially during JJA and SON) and Australia (in SON). We also find that CMIP6 models generally correctly reproduce clustering patterns, paving the way for an assessment of trends in sub-seasonal clustering under climate change. Clustering of present-day extremes increases in many areas under climate change. Changes diagnosed by comparing present day and future extreme percentiles are positive and negative and strongest in the tropical areas.

Comparison of spatiotemporal characteristics of the COVID-19 and SARS outbreaks in mainland China

BackgroundBoth coronavirus disease 2019 (COVID-19) and severe acute respiratory syndrome (SARS) are caused by coronaviruses and have infected people in China and worldwide. We aimed to investigate whether COVID-19 and SARS exhibited similar spatial and temporal features at provincial level in mainland China.MethodsThe number of people infected by COVID-19 and SARS were extracted from daily briefings on newly confirmed cases during the epidemics, as of Mar. 4, 2020 and Aug. 3, 2003, respectively. We depicted spatiotemporal patterns of the COVID-19 and SARS epidemics using spatial statistics such as Moran’s I and the local indicators of spatial association (LISA).ResultsCompared to SARS, COVID-19 had a higher overall incidence. We identified 3 clusters (predominantly located in south-central China; the highest RR = 135.08, 95% CI: 128.36–142.08) for COVID-19 and 4 clusters (mainly in Northern China; the highest RR = 423.51, 95% CI: 240.96–722.32) for SARS. Fewer secondary clusters were identified after the “Wuhan lockdown”. The LISA cluster map detected a significantly high-low (Hubei) and low-high spatial clustering (Anhui, Hunan, and Jiangxi, in Central China) for COVID-19. Two significant high-high (Beijing and Tianjin) and low-high (Hebei) clusters were detected for SARS.ConclusionsCOVID-19 and SARS outbreaks exhibited distinct spatiotemporal clustering patterns at the provincial levels in mainland China, which may be attributable to changes in social and demographic factors, local government containment strategies or differences in transmission mechanisms.

Toward Measuring the Level of Spatiotemporal Clustering of Multi-Categorical Geographic Events

Human activity events are often recorded with their geographic locations and temporal stamps, which form spatial patterns of the events during individual time periods. Temporal attributes of these events help us understand the evolution of spatial processes over time. A challenge that researchers still face is that existing methods tend to treat all events as the same when evaluating the spatiotemporal pattern of events that have different properties. This article suggests a method for assessing the level of spatiotemporal clustering or spatiotemporal autocorrelation that may exist in a set of human activity events when they are associated with different categorical attributes. This method extends the Voronoi structure from 2D to 3D and integrates a sliding-window model as an approach to spatiotemporal tessellations of a space-time volume defined by a study area and time period. Furthermore, an index was developed to evaluate the partial spatiotemporal clustering level of one of the two event categories against the other category. The proposed method was applied to simulated data and a real-world dataset as a case study. Experimental results show that the method effectively measures the level of spatiotemporal clustering patterns among human activity events of multiple categories. The method can be applied to the analysis of large volumes of human activity events because of its computational efficiency.

Spatio-temporal variations of licensed doctor distribution in China: measuring and mapping disparities

BackgroundThe licensed doctor misdistribution is one of the major challenges faced by China. However, this subject remains underexplored as spatial distribution characteristics (such as spatial clustering patterns) have not been fully mapped out by existing studies. To fill the void, this study aims to explore the spatio-temporal dynamics and spatial clustering patterns of different subtypes of licensed doctors (i.e., clinicians, traditional Chinese medicine doctors, dentists, public health doctors, general practitioners) in China.MethodsData on the licensed doctor quantity and population during 2012–2016 was obtained from the National Health (and Family Planning) Yearbook. Functional boxplots were used to visualize and compare the temporal trends of densities of different subtypes of licensed doctors. This study adopted two complementary spatial statistics (space-time scan statistics and Moran’s I statistics) to explore the spatio-temporal dynamics and spatial clustering patterns of licensed doctor distribution in China. The former was used to explore the spatial variations in the temporal trends of licensed doctor density during 2012–2016, and the latter was adopted to explore the spatial changing patterns of licensed doctor distribution during the research period.ResultsThe results show that the densities of almost all subtypes of licensed doctors displayed upward trends during 2012–2016, though some provincial units were left behind. Besides, spatial distribution characteristics varied across different subtypes of licensed doctors, with the low-low cluster area of general practitioners being the largest.ConclusionsThe misdistribution of licensed doctors is a global problem and China is no exception. In order to achieve a balanced distribution of licensed doctors, the government is suggested to introduce a series of measures, such as deliberative policy design and effective human resource management initiatives to educate, recruit, and retain licensed doctors and prevent a brain drain of licensed doctors from disadvantaged units.

Cluster-based bagging of constrained mixed-effects models for high spatiotemporal resolution nitrogen oxides prediction over large regions

BackgroundAccurate estimation of nitrogen dioxide (NO2) and nitrogen oxide (NOx) concentrations at high spatiotemporal resolutions is crucial for improving evaluation of their health effects, particularly with respect to short-term exposures and acute health outcomes. For estimation over large regions like California, high spatial density field campaign measurements can be combined with more sparse routine monitoring network measurements to capture spatiotemporal variability of NO2 and NOx concentrations. However, monitors in spatially dense field sampling are often highly clustered and their uneven distribution creates a challenge for such combined use. Furthermore, heterogeneities due to seasonal patterns of meteorology and source mixtures between sub-regions (e.g. southern vs. northern California) need to be addressed. ObjectivesIn this study, we aim to develop highly accurate and adaptive machine learning models to predict high-resolution NO2 and NOx concentrations over large geographic regions using measurements from different sources that contain samples with heterogeneous spatiotemporal distributions and clustering patterns. MethodsWe used a comprehensive Kruskal-K-means method to cluster the measurement samples from multiple heterogeneous sources. Spatiotemporal cluster-based bootstrap aggregating (bagging) of the base mixed-effects models was then applied, leveraging the clusters to obtain balanced and less correlated training samples for less bias and improvement in generalization. Further, we used the machine learning technique of grid search to find the optimal interaction of temporal basis functions and the scale of spatial effects, which, together with spatiotemporal covariates, adequately captured spatiotemporal variability in NO2 and NOx at the state and local levels. ResultsWe found an optimal combination of four temporal basis functions and 200 m scale spatial effects for the base mixed-effects models. With the cluster-based bagging of the base models, we obtained robust predictions with an ensemble cross validation R2 of 0.88 for both NO2 and NOx [RMSE (RMSEIQR): 3.62 ppb (0.28) and 9.63 ppb (0.37) respectively]. In independent tests of random sampling, our models achieved similarly strong performance (R2 of 0.87–0.90; RMSE of 3.97–9.69 ppb; RMSEIQR of 0.21–0.27), illustrating minimal over-fitting. ConclusionsOur approach has important implications for fusing data from highly clustered and heterogeneous measurement samples from multiple data sources to produce highly accurate concentration estimates of air pollutants such as NO2 and NOx at high resolution over a large region.

Dual-Constraint Spatiotemporal Clustering Approach for Exploring Marine Anomaly Patterns Using Remote Sensing Products

Spatiotemporal clustering patterns of marine anomaly variations are the focus of much current global climate change research. Marine anomaly variations have multidimensional attributes and are spatiotemporally continuous; existing methods for clustering face challenges in mining effectively their spatiotemporal clustering patterns. Using long-term marine remote sensing products, we present the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">dual-constraint spatiotemporal clustering approach</i> (DcSTCA) for exploring marine clustering patterns. The DcSTCA includes three steps. The first step constructs a spatiotemporal grid cube based on the spatial connectivity and time evolution process of marine anomaly variations, which is used to search the spatiotemporal neighborhood. The second step calculates the proximities in space, time, and thematic attribute to obtain the spatiotemporal clustering cores and spatiotemporal density of each grid cell. The final step derives spatiotemporal clustering patterns by connecting the clustering cores and their spatiotemporal neighbors according to their density connectivity. Experiments on simulated datasets are used to demonstrate the effectiveness and the advantages of the DcSTCA compared with spatial–temporal density-based spatial clustering of applications with noise (ST-DBSCAN). The applications on sea surface temperature in the Pacific Ocean show that the DcSTCA can effectively explore marine clustering patterns from remote sensing products, and these mined clustering patterns may provide new references for global change research.

Network-constrained spatio-temporal clustering analysis of traffic collisions in Jianghan District of Wuhan, China.

The analysis of traffic collisions is essential for urban safety and the sustainable development of the urban environment. Reducing the road traffic injuries and the financial losses caused by collisions is the most important goal of traffic management. In addition, traffic collisions are a major cause of traffic congestion, which is a serious issue that affects everyone in the society. Therefore, traffic collision analysis is essential for all parties, including drivers, pedestrians, and traffic officers, to understand the road risks at a finer spatio-temporal scale. However, traffic collisions in the urban context are dynamic and complex. Thus, it is important to detect how the collision hotspots evolve over time through spatio-temporal clustering analysis. In addition, traffic collisions are not isolated events in space. The characteristics of the traffic collisions and their surrounding locations also present an influence of the clusters. This work tries to explore the spatio-temporal clustering patterns of traffic collisions by combining a set of network-constrained methods. These methods were tested using the traffic collision data in Jianghan District of Wuhan, China. The results demonstrated that these methods offer different perspectives of the spatio-temporal clustering patterns. The weighted network kernel density estimation provides an intuitive way to incorporate attribute information. The network cross K-function shows that there are varying clustering tendencies between traffic collisions and different types of POIs. The proposed network differential Local Moran’s I and network local indicators of mobility association provide straightforward and quantitative measures of the hotspot changes. This case study shows that these methods could help researchers, practitioners, and policy-makers to better understand the spatio-temporal clustering patterns of traffic collisions.

Investigating spatio-temporal distribution and diffusion patterns of the dengue outbreak in Swat, Pakistan.

Dengue has been endemic to Pakistan in the last two decades. There was a massive outbreak in the Swat valley in 2013. Here we demonstrate the spatio-temporal clustering and diffusion patterns of the dengue outbreak. Dengue case data were acquired from the hospital records in the Swat district of Pakistan. Ring maps visualize the distribution and diffusion of the number of cases and incidence of dengue at the level of the union council. We applied space-time scan statistics to identify spatio-temporal clusters. Ordinary least squares and geographically weighted regression models were used to evaluate the impact of elevation, population density, and distance to the river. The results show that dengue distribution is not random, but clustered in space and time in the Swat district. Males constituted 68% of the cases while females accounted for about 32%. A majority of the cases (>55%) were younger than 40 years of age. The southern part was a major hotspot affected by the dengue outbreak in 2013. There are two space-time clusters in the spatio-temporal analysis. GWR and OLS show that population density is a significant explanatory variable for the dengue outbreak, while GWR exhibits better performance in terms of 'R2=0.49 and AICc=700'. Dengue fever is clustered in the southern part of the Swat district. This region is relatively urban in character, with most of the population of the district residing here. There is a need to strengthen the surveillance system for reporting dengue cases in order to respond to future outbreaks in a robust way.

Spatio-temporal epidemic type aftershock sequence model for Tangshan aftershock sequence

Shallow earthquakes usually show obvious spatio-temporal clustering patterns. In this study, several spatio-temporal point process models are applied to investigate the clustering characteristics of the well-known Tangshan sequence based on classical empirical laws and a few assumptions. The relative fit of competing models is compared by Akaike Information Criterion. The spatial clustering pattern is well characterized by the model which gives the best fit to the data. A simulated aftershock sequence is generated by thinning algorithm and compared with the real seismicity.

A stack-based prospective spatio-temporal data analysis approach

Spatio-temporal data analysis has recently gained considerable attention from both the research and practitioner communities because of the increasing availability of datasets with prominent spatial and temporal data elements. In this paper, we develop a new spatio-temporal data analysis approach aimed at discovering abnormal spatio-temporal clustering patterns. We also propose a quantitative evaluation framework and compare our approach against a widely used space–time scan statistic-based method under this framework. Our approach is based on a robust clustering engine using support vector machines and incorporates ideas from existing online surveillance methods to track incremental changes over time. Initial experimental results using both simulated and real-world datasets indicate that our approach is able to detect abnormal areas with irregular shapes more accurately than the space–time scan statistic-based method.