Space-time Cluster Detection Research Articles

Detection of Space-Time Clusters and Ambient Temperature Effects on Non-Toxigenic Vibrio Cholerae in Russia from 2005 To 2021

Introduction: The identification of climate temperature-sensitive pathogens and infectious diseases is essential in addressing health risks resulting from global warming. Such research is especially crucial in regions where climate change may have a more significant impact like Russia. Recent studies have reasoned that the abundance of non-toxigenic V. cholerae is environmentally driven and can be part of early global warming signals for Russian territory. The aim of the study is to investigate the spatial-temporal trends and thermo-climatic sensitivity of non-toxigenic V. cholerae abundance in Russia.
 Methods: This study employed Kulldorff’s space-time statistics to identify persistent clusters of the V. cholerae ctx- isolation and areas for exploring temperature-depended patterns of the vibrio distribution. Correlation analysis was used to identify regions with temperature-driven Vibrio abundance in water samples.
 Results: The spatial analysis detected 16 persistent (7-8 year) clusters of V. cholerae ctx- across the study period 2005-2021. The number of clusters with RR >1 abandoning from the south to the north and the total number of persistent clusters (9) is greater in the period of 2014(5)-2021 compared with the period 2005-2013 (7). A distinct significant thermo-climatic effect on the abundance of V. cholerae ctx- in water basins was found in three Russian regions with temperate marine (the Kaliningrad region) and sharp continental climatic conditions (the Irkutsk region and the Republic of Sakha). The temperature and Vibrio prevalence trend curves are peaky (the Kaliningrad region and the Republic of Sakha) or bell-shaped (the Irkutsk region) changed and closely followed together.
 Conclusion: The persistent clusters should become targeted areas to improve sanitation conditions. The study offers valuable outcomes to support simplified empirical evaluations of the potential hazards of vibrio abundance that might be useful locally for public health authorities and globally as a part of Russia's warning system of climate change effects.

Evidence of transgenerational effects on autism spectrum disorder using multigenerational space-time cluster detection

BackgroundTransgenerational epigenetic risks associated with complex health outcomes, such as autism spectrum disorder (ASD), have attracted increasing attention. Transgenerational environmental risk exposures with potential for epigenetic effects can be effectively identified using space-time clustering. Specifically applied to ancestors of individuals with disease outcomes, space-time clustering characterized for vulnerable developmental stages of growth can provide a measure of relative risk for disease outcomes in descendants.Objectives(1) Identify space-time clusters of ancestors with a descendent with a clinical ASD diagnosis and matched controls. (2) Identify developmental windows of ancestors with the highest relative risk for ASD in descendants. (3) Identify how the relative risk may vary through the maternal or paternal line.MethodsFamily pedigrees linked to residential locations of ASD cases in Utah have been used to identify space-time clusters of ancestors. Control family pedigrees of none-cases based on age and sex have been matched to cases 2:1. The data have been categorized by maternal or paternal lineage at birth, childhood, and adolescence. A total of 3957 children, both parents, and maternal and paternal grandparents were identified. Bernoulli space-time binomial relative risk (RR) scan statistic was used to identify clusters. Monte Carlo simulation was used for statistical significance testing.ResultsTwenty statistically significant clusters were identified. Thirteen increased RR (> 1.0) space-time clusters were identified from the maternal and paternal lines at a p-value < 0.05. The paternal grandparents carry the greatest RR (2.86–2.96) during birth and childhood in the 1950’s–1960, which represent the smallest size clusters, and occur in urban areas. Additionally, seven statistically significant clusters with RR < 1 were relatively large in area, covering more rural areas of the state.ConclusionThis study has identified statistically significant space-time clusters during critical developmental windows that are associated with ASD risk in descendants. The geographic space and time clusters family pedigrees with over 3 + generations, which we refer to as a person’s geographic legacy, is a powerful tool for studying transgenerational effects that may be epigenetic in nature. Our novel use of space-time clustering can be applied to any disease where family pedigree data is available.

Detection of space-time clusters using a topological hierarchy for geospatial data on COVID-19 in Japan.

In this paper, we detected space–time clusters using data on coronavirus disease 2019 (COVID-19) collected daily by each prefecture in Japan. COVID-19 has spread globally since the first confirmed case in China, in December 2019. Several people have to date been infected in Japan since the first confirmed case in January 2020. The outbreak of COVID-19 has had a significant impact on many people’s lives. Studies are being conducted to detect regions, called clusters, which pose a significantly higher risk of infection than their surrounding areas, based on a spatial scan statistics of COVID-19 infections. Among these studies, space–time cluster detection has to date been actively performed to gain knowledge regarding infection status. Based on the spatial scan statistic, the cylindrical scan method is a widely used space–time cluster detection method. This method enables concurrent detection of the location and time of a cluster occurrence. However, this method cannot capture spatial changes in a cluster over time. When applying the existing method to a cluster whose shape changes over time, the number of calculations required becomes extremely large, and the analysis may become difficult. In this study, we focused on the hierarchical structure of the data obtained by conducting an echelon analysis and applied the space–time cluster detection method based on this structure to enable the capture of changes in a cluster’s shape. Furthermore, we visualized the location and period of a cluster’s occurrence and considered the cause of the cluster.

Detection of Space-Time Disease Clusters Using A Matrix Factorization Method

Space-time cluster detection has important applications in public health management and epidemiology to devise disease prevention strategies and to find the causes of a particular disease outbreak in a country. This study introduced a new method to detect the potential space-time clusters with no restriction on cluster shape and size and further visualize them distinctly on the heat map. The proposed algorithm is based on matrix factorization technique to find the significant components in spatial as well as temporal dimension. Applications to malaria data in Khyber Pakhtunkhwa, Pakistan shows that the proposed method is effective in detecting the potential clusters.

Space-Time Cluster's Detection and Geographical Weighted Regression Analysis of COVID-19 Mortality on Texas Counties.

As COVID-19 run rampant in high-density housing sites, it is important to use real-time data in tracking the virus mobility. Emerging cluster detection analysis is a precise way of blunting the spread of COVID-19 as quickly as possible and save lives. To track compliable mobility of COVID-19 on a spatial-temporal scale, this research appropriately analyzed the disparities between spatial-temporal clusters, expectation maximization clustering (EM), and hierarchical clustering (HC) analysis on Texas county-level. Then, based on the outcome of clustering analysis, the sensitive counties are Cottle, Stonewall, Bexar, Tarrant, Dallas, Harris, Jim hogg, and Real, corresponding to Southeast Texas analysis in Geographically Weighted Regression (GWR) modeling. The sensitive period took place in the last two quarters in 2020 and the first quarter in 2021. We explored PostSQL application to portray tracking Covid-19 trajectory. We captured 14 social, economic, and environmental impact’s indices to perform principal component analysis (PCA) to reduce dimensionality and minimize multicollinearity. By using the PCA, we extracted five factors related to mortality of COVID-19, involved population and hospitalization, adult population, natural supply, economic condition, air quality or medical care. We established the GWR model to seek the sensitive factors. The result shows that adult population, economic condition, air quality, and medical care are the sensitive factors. Those factors also triggered high increase of COVID-19 mortality. This research provides geographical understanding and solution of controlling COVID-19, reference of implementing geographically targeted ways to track virus mobility, and satisfy for the need of emergency operations plan (EOP).

Detection of spatial and spatio-temporal Salmonella Heidelberg and Salmonella Typhimurium human case clusters focused around licensed abattoirs in Ontario in 2015, and their potential relation to known outbreaks.

Salmonellosis is one of several zoonotic diseases for which individuals with occupational animal contact, including abattoir workers, are at an increased risk. If meat is contaminated during slaughter, this can increase the risk of enteric illness for consumers. In this study, we investigated whether reported cases of Salmonella Heidelberg and Typhimurium were clustered around abattoirs in Ontario in 2015 and whether there was any evidence (laboratory/exposure) to suggest an abattoir at the centre of a cluster might be the source of exposure. Data for each reported case of S.Heidelberg and S.Typhimurium in Ontario in 2015 were collected. Multi-focused and non-focused spatial and space-time cluster detection tests were performed for each serotype, with and without cases linked to known outbreaks, using Poisson and space-time permutation models. Focused tests included the location of abattoirs operational in all or part of 2015. Laboratory data and exposure information were used to explore the relatedness of cases within identified clusters. Focused spatial tests identified clusters of S.Heidelberg and S.Typhimurium around abattoirs. Focused space-time permutation tests identified 2 significant space-time clusters of S.Heidelberg; one cluster (n=11 cases) included 8 of 9 cases associated with a known outbreak and the other cluster (n=18 cases) was not part of a previously identified outbreak. Review of laboratory and risk factor information suggested that cases within each cluster shared a common exposure. Cases were not asked about goat or sheep meat consumption. The focused cluster test, particularly with the space-time permutation model, could assist in identifying outbreaks associated with a particular physical location, such as an abattoir. Improvements to the current case investigation process, such as consistent collection and reporting of high-risk occupation information and more detailed food consumption history, could assist in outbreak identification when coupled with this statistic.

Daily surveillance of COVID-19 using the prospective space-time scan statistic in the United States

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was first discovered in late 2019 in Wuhan City, China. The virus may cause novel coronavirus disease 2019 (COVID-19) in symptomatic individuals. Since December of 2019, there have been over 7,000,000 confirmed cases and over 400,000 confirmed deaths worldwide. In the United States (U.S.), there have been over 2,000,000 confirmed cases and over 110,000 confirmed deaths. COVID-19 case data in the United States has been updated daily at the county level since the first case was reported in January of 2020. There currently lacks a study that showcases the novelty of daily COVID-19 surveillance using space-time cluster detection techniques. In this paper, we utilize a prospective Poisson space-time scan statistic to detect daily clusters of COVID-19 at the county level in the contiguous 48 U.S. and Washington D.C. As the pandemic progresses, we generally find an increase of smaller clusters of remarkably steady relative risk. Daily tracking of significant space-time clusters can facilitate decision-making and public health resource allocation by evaluating and visualizing the size, relative risk, and locations that are identified as COVID-19 hotspots.

Space-time cluster detection with cross-space-time relative risk functions

ABSTRACTSpace-time kernel density estimation (STKDE) commonly is used for space-time cluster detection. But, this technique might be limited because it does not take into account an underlying population at risk for observed events. A space-time relative risk function (STRRF) can help overcome this limitation by allowing a comparison of each kernel density of observations with that of controls. This paper proposes a cross-STRRF to identify spatio-temporal locations that experience statistically significant changes in their density of events. With events organized in a space-time voxel structure, the cross-STRRF evaluates space-time patterns by comparing event occurrences at a spatial location in a previous time period with ones in its future as well as with its spatial neighbors in its contemporaneous time period. The test statistics of the cross-STRRF values in each voxel are obtained with a permutation test in which cases and controls are shuffled within each time period to maintain the space-time envelope of events. An application to assault crime incidents in the city of Plano, Texas between 2008 and 2012 illustrates that the cross-STRRF and its significance test results emphasize spatio-temporal changes in event density rather than constantly focusing on high density regions, which STKDE does.

Detecting space-time disease clusters with arbitrary shapes and sizes using a co-clustering approach.

Ability to detect potential space-time clusters in spatio-temporal data on disease occurrences is necessary for conducting surveillance and implementing disease prevention policies. Most existing techniques use geometrically shaped (circular, elliptical or square) scanning windows to discover disease clusters. In certain situations, where the disease occurrences tend to cluster in very irregularly shaped areas, these algorithms are not feasible in practise for the detection of space-time clusters. To address this problem, a new algorithm is proposed, which uses a co-clustering strategy to detect prospective and retrospective space-time disease clusters with no restriction on shape and size. The proposed method detects space-time disease clusters by tracking the changes in space-time occurrence structure instead of an in-depth search over space. This method was utilised to detect potential clusters in the annual and monthly malaria data in Khyber Pakhtunkhwa Province, Pakistan from 2012 to 2016 visualising the results on a heat map. The results of the annual data analysis showed that the most likely hotspot emerged in three sub-regions in the years 2013-2014. The most likely hotspots in monthly data appeared in the month of July to October in each year and showed a strong periodic trend.

Is the current pertussis incidence only the results of testing? A spatial and space-time analysis of pertussis surveillance data using cluster detection methods and geographically weighted regression modelling.

BackgroundDespite high vaccination coverage, pertussis incidence in the Netherlands is amongst the highest in Europe with a shifting tendency towards adults and elderly. Early detection of outbreaks and preventive actions are necessary to prevent severe complications in infants. Efficient pertussis control requires additional background knowledge about the determinants of testing and possible determinants of the current pertussis incidence. Therefore, the aim of our study is to examine the possibility of locating possible pertussis outbreaks using space-time cluster detection and to examine the determinants of pertussis testing and incidence using geographically weighted regression models.MethodsWe analysed laboratory registry data including all geocoded pertussis tests in the southern area of the Netherlands between 2007 and 2013. Socio-demographic and infrastructure-related population data were matched to the geo-coded laboratory data. The spatial scan statistic was applied to detect spatial and space-time clusters of testing, incidence and test-positivity. Geographically weighted Poisson regression (GWPR) models were then constructed to model the associations between the age-specific rates of testing and incidence and possible population-based determinants.ResultsSpace-time clusters for pertussis incidence overlapped with space-time clusters for testing, reflecting a strong relationship between testing and incidence, irrespective of the examined age group. Testing for pertussis itself was overall associated with lower socio-economic status, multi-person-households, proximity to primary school and availability of healthcare. The current incidence in contradiction is mainly determined by testing and is not associated with a lower socioeconomic status.DiscussionTesting for pertussis follows to an extent the general healthcare seeking behaviour for common respiratory infections, whereas the current pertussis incidence is largely the result of testing. More testing would thus not necessarily improve pertussis control. Detecting outbreaks using space-time cluster detection is feasible but needs to adjust for the strong impact of testing on the detection of pertussis cases.

Disease cluster detection methods: recent developments and public health implications

Methods for detecting spatial and spatiotemporal clusters of health and disease have advanced significantly in the past decade. This article reviews recent advances in four areas: spatial search processes, network-based methods, statistical analysis and modelling of local clusters and space-time cluster detection. I then turn to a more critical discussion of the implications of hotspot mapping for public health policy and intervention, highlighting the need to incorporate process-based understandings that impact spatial and social inequalities in ill health for particular health issues in particular geographic contexts.

Modifiable temporal unit problem (MTUP) and its effect on space-time cluster detection.

BackgroundWhen analytical techniques are used to understand and analyse geographical events, adjustments to the datasets (e.g. aggregation, zoning, segmentation etc.) in both the spatial and temporal dimensions are often carried out for various reasons. The ‘Modifiable Areal Unit Problem’ (MAUP), which is a consequence of adjustments in the spatial dimension, has been widely researched. However, its temporal counterpart is generally ignored, especially in space-time analysis.MethodsIn analogy to MAUP, the Modifiable Temporal Unit Problem (MTUP) is defined as consisting of three temporal effects (aggregation, segmentation and boundary). The effects of MTUP on the detection of space-time clusters of crime datasets of Central London are examined using Space-Time Scan Statistics (STSS).Results and ConclusionThe case study reveals that MTUP has significant effects on the space-time clusters detected. The attributes of the clusters, i.e. temporal duration, spatial extent (size) and significance value (p-value), vary as the aggregation, segmentation and boundaries of the datasets change. Aggregation could be used to find the significant clusters much more quickly than at lower scales; segmentation could be used to understand the cyclic patterns of crime types. The consistencies of the clusters appearing at different temporal scales could help in identifying strong or ‘true’ clusters.

An evaluation of SaTScan for the prospective detection of space-timeCampylobacterclusters in the North East of England

Campylobacteriosis is the commonest cause of bacterial enteritis in England yet the epidemiology of apparently sporadic cases is not well understood. Here we evaluated the feasibility of applying a space-time cluster detection method to routine laboratory surveillance data in the North East of England by simulating prospective weekly space-time cluster detection using SaTScan as if it had been performed for 2008-2011. From the 209 simulated weekly runs using a circular window, 20 distinct clusters were found which contained a median of 30 cases (interquartile range 15-66) from a median population of ~134,000 persons. This corresponds to detection of a new cluster every 10 weeks. We found significant differences in age, sex and deprivation score distributions between areas within clusters compared to those without. The results of this study suggest that space-time detection of Campylobacter clusters could be used to find groups of cases amenable to epidemiological investigation.

Comparison of covariate adjustment methods using space-time scan statistics for food animal syndromic surveillance

BackgroundAbattoir condemnation data show promise as a rich source of data for syndromic surveillance of both animal and zoonotic diseases. However, inherent characteristics of abattoir condemnation data can bias results from space-time cluster detection methods for disease surveillance, and may need to be accounted for using various adjustment methods. The objective of this study was to compare the space-time scan statistics with different abilities to control for covariates and to assess their suitability for food animal syndromic surveillance. Four space-time scan statistic models were used including: animal class adjusted Poisson, space-time permutation, multi-level model adjusted Poisson, and a weighted normal scan statistic using model residuals. The scan statistics were applied to monthly bovine pneumonic lung and “parasitic liver” condemnation data from Ontario provincial abattoirs from 2001–2007.ResultsThe number and space-time characteristics of identified clusters often varied between space-time scan tests for both “parasitic liver” and pneumonic lung condemnation data. While there were some similarities between isolated clusters in space, time and/or space-time, overall the results from space-time scan statistics differed substantially depending on the covariate adjustment approach used.ConclusionsVariability in results among methods suggests that caution should be used in selecting space-time scan methods for abattoir surveillance. Furthermore, validation of different approaches with simulated or real outbreaks is required before conclusive decisions can be made concerning the best approach for conducting surveillance with these data.

Alternative Scan-Based Approaches to Identify Space-Time Clusters of Highly Pathogenic Avian Influenza Virus H5N1 in Wild Birds in Denmark and Sweden in 2006

The objective of this study was to demonstrate the effects of the nature of the information collected through passive surveillance on the detection of space-time clusters of highly pathogenic avian influenza virus (HPAIV) H5N1 cases reported among dead wild birds in Denmark and Sweden in 2006. Data included 1469 records (109 cases, 1360 controls) collected during the regional epidemic between February and June by passive surveillance of dead wild birds. Laboratory diagnoses were obtained by PCR methods and/or virus isolation. The nature of available information influences both the type of model suitable for analysis and its parameterization. Here, we explored four alternative scan-based methods, suitable for detection of clusters only when case data (univariate permutation model), case and hypothesized epidemiological variables (multivariate permutation model), case and control data (univariate Bernoulli model), and case, control, and hypothesized epidemiological variables (multivariate Bernoulli model) are available. Tufted ducks were particularly common among infected wild bird species detected in Denmark and Sweden during the initial phases of this epidemic, and species group (tufted ducks [62 cases, 57 controls] vs. other wild bird species [47 cases, 1303 controls]) was considered in the multivariate models as a covariate potentially associated with clustering. Bernoulli and permutation scan analyses both detected multiple significant (P < 0.01) clusters with similar locations, but with certain differences in their numbers and sizes. The observed-to-expected case ratios in the two clusters detected by the multivariate Bernoulli scan model were substantially heterogeneous. However, the permutation model detected only one of the Swedish clusters and only pinpointed the heterogeneity between species on clustering in the same Danish cluster as detected by the Bernoulli model. The output of the methods described here were shown to be highly sensitive to the choice of the probability model for cases and the choice of plausible assumptions to parameterize the scan statistic tests. The results of the multivariate Bernoulli suggest that with noncase information regarding a potential risk factor, such as species of birds, this method is sensitive and efficient in identifying high-risk areas and time periods for regional occurrence of HPAIV and potentially for similar infectious diseases. Results here demonstrate the impact that the nature of the collected information has on the epidemiological investigation of outbreaks. Results show the importance of collecting information on control data and on variables hypothesized to influence disease risk on the identification of periods of time and locations at high risk for the disease and risk factors associated with clustering as part of the national and international surveillance systems.

Influence of spatial resolution on space-time disease cluster detection.

BackgroundUtilizing highly precise spatial resolutions within disease outbreak detection, such as the patients’ address, is most desirable as this provides the actual residential location of the infected individual(s). However, this level of precision is not always readily available or only available for purchase, and when utilized, increases the risk of exposing protected health information. Aggregating data to less precise scales (e.g., ZIP code or county centroids) may mitigate this risk but at the expense of potentially masking smaller isolated high risk areas.MethodsTo experimentally examine the effect of spatial data resolution on space-time cluster detection, we extracted administrative medical claims data for 122500 viral lung episodes occurring during 2007–2010 in Tennessee. We generated 10000 spatial datasets with varying cluster location, size and intensity at the address-level. To represent spatial data aggregation (i.e., reduced resolution), we then created 10000 corresponding datasets both at the ZIP code and county level for a total of 30000 datasets. Using the space-time permutation scan statistic and the SaTScan™ cluster software, we evaluated statistical power, sensitivity and positive predictive values of outbreak detection when using exact address locations compared to ZIP code and county level aggregations.ResultsThe power to detect disease outbreaks did not largely diminish when using spatially aggregated data compared to more precise address information. However, aggregations negatively impacted the ability to more accurately determine the exact spatial location of the outbreak, particularly in smaller clusters (<800 km2).ConclusionsSpatial aggregations do not necessitate a loss of power or sensitivity; rather, the relationship is more complex and involves simultaneously considering relative risk within the cluster and cluster size. The likelihood of spatially over-estimating outbreaks by including geographical areas outside the actual disease cluster increases with aggregated data.

Incidence analyses and space-time cluster detection of hepatitis C in Fujian Province of China from 2006 to 2010.

BackgroundThere is limited epidemiologic information about the incidence of hepatitis C in China, and few studies have applied space-time scan statistic to detect clusters of hepatitis C and made adjustment for temporal trend and relative risk of regions.Methodology and Principal FindingsWe analyzed the temporal changes and characteristics of incidence of hepatitis C in Fujian Province from 2006 through 2010. The discrete Poisson model of space-time scan statistic was chosen for cluster detection. Data on new cases of hepatitis C were obtained from the Center for Disease Control and Prevention of Fujian Province. Between 2006 and 2010, there was an annualized increase in the incidence of hepatitis C of 23.0 percent, from 928 cases (2.63 per 100,000 persons) to 2,180 cases (6.01 per 100,000 persons). The incidence among women increased more rapidly. The cumulative incidence showed that people who were over 60 years had the highest risk to suffer hepatitis C (52.51 per 100,000 persons), and women had lower risk compared to men (OR = 0.69). Putian had the highest cumulative incidence among all the regions (86.95 per 100,000 persons). The most likely cluster was identified in Putian during March to August in 2009 without adjustment, but it shifted to three contiguous cities with a two-month duration after adjustment for temporal trend and relative risk of regions.Conclusions/SignificanceThe incidence of hepatitis C is increasing in Fujian Province, and women are at a more rapid pace. The space-time scan statistic is useful as a screening tool for clusters of hepatitis C, with adjustment for temporal trend and relative risk of regions recommended.

Adaptive likelihood ratio methods for the detection of space-time disease clusters

Adaptive likelihood ratio methods for the detection of space-time disease clusters

A model for space–time cluster detection using spatial clusters with flexible temporal risk patterns

Maps of estimated disease rates over multiple time periods are useful tools for gaining etiologic insights regarding potential exposures associated with specific locations and times. In this paper, we describe an extension of the Gangnon-Clayton model for spatial clustering to spatio-temporal data. As in the purely spatial model, a large set of circular regions of varying radii centered at observed locations are considered as potential clusters, e.g. subregions with a different pattern of risk than the remainder of the study region. Within the spatio-temporal model, no specific parametric form is imposed on the temporal pattern of risk within each cluster. In addition to the clusters, the proposed model incorporates spatial and spatio-temporal heterogeneity effects and can readily accommodate regional covariates. Inference is performed in a Bayesian framework using MCMC. Although formal inferences about the number of clusters could be obtained using a reversible jump MCMC algorithm, we use local Bayes factors from models with a fixed, but overly large, number of clusters to draw inferences about both the number and the locations of the clusters. We illustrate the approach with two applications of the model to data on female breast cancer mortality in Japan and evaluate its operating characteristics in a simulation study.

Space-time Bayesian small area disease risk models: development and evaluation with a focus on cluster detection

This paper extends the spatial local-likelihood model and the spatial mixture model to the space-time (ST) domain. For comparison, a standard random effect space-time (SREST) model is examined to allow evaluation of each model's ability in relation to cluster detection. To pursue this evaluation, we use the ST counterparts of spatial cluster detection diagnostics. The proposed criteria are based on posterior estimates (e.g., misclassification rate) and some are based on post-hoc analysis of posterior samples (e.g., exceedance probability). In addition, we examine more conventional model fit criteria including mean square error (MSE). We illustrate the methodology with a real ST dataset, Georgia throat cancer mortality data for the years 1994-2005, and a simulated dataset where different levels and shapes of clusters are embedded. Overall, it is found that conventional SREST models fair well in ST cluster detection and in goodness-of-fit, while for extreme risk detection the local likelihood ST model does best.