Estimating Reproductive Numbers of Disease Diffusion  

R package: EpiTrans  |@Github|

Installation


EpiTrans Tutorials: 1. Dengue; 2. COVID-19
Introduction: The package provides a new method for quantifying spatially adjusted (time-varying) reproductive numbers that reflects spatial heterogeneity in transmission potential among individuals. This new method estimates individual-level effective reproductive numbers and a summarized indicator for population-level time-varying reproductive number. We provides the tutorials which include sample datasets from the dengue outbreak in Tainan City, 2007 and the COVID-19 cases in northern Taiwan, 2021, and demonstrates the major functions of the package , including (1) plotting descriptive epidemic trends in time and space; (2) estimating spatially-adjusted reproductive numbers; (3) visualizing the graphics and animations of individual-level and aggregated-level reproductive numbers.  The list of detailed functions and sample datasets can be found here.
Further Reading: Ng TC, Wen TH * (2019), Spatially Adjusted Time-varying Reproductive Numbers: Understanding the Geographical Expansion of Urban Dengue Outbreaks, Scientific Reports 9: 19172.


Epi-Pro (Epidemic Profiler): The Software for Profiling Epidemics
Language: Python 3.x + Qt
User Interface


Epi-Pro's Diffusion Analytics
| PySDA (Python Suite for Diffusion Analysis) | how to install |

PySDA Cluster Evolution: Modified Space–Time DBSCAN (MST-DBSCAN) | Tutorial
Language: Python 3.x
Introduction: Epidemic diffusion is a space–time process, and showing time-series disease maps is a common way to demonstrate an epidemic progression in time and space. Previous studies used time-series maps to demonstrate the animation of diffusion process. Epidemic diffusion patterns were determined subjectively by visual inspection, however. There currently are still methodological concerns in developing effective analytical approaches for profiling diffusion dynamics of disease clustering and epidemic propagation. The objective of this study is to develop a geo-computational algorithm, the modified space–time density-based spatial clustering of application with noise (MST-DBSCAN), for detecting, identifying, and visualizing disease cluster evolution, which takes the effect of the incubation period into account. We also map the MST-DBSCAN algorithm output to visualize the diffusion process. Our results show that compared to kernel-smoothed mapping, the MST-DBSCAN algorithm can better identify the evolution type of any cluster at any epoch. Furthermore, using only one two-dimensional map (and graphs), our approach can demonstrate the same diffusion process that time-series maps or three-dimensional space–time kernel plotting displays but in an easy-to-read manner. We conclude that our MST-DBSCAN algorithm can profile the spatial pattern of epidemic diffusion in detail by identifying disease cluster evolution.
Further Reading: Kuo FY, Wen TH *, Sabel C (2018), Characterizing Diffusion Dynamics of Disease Clustering: A Modified Space-Time DBSCAN (MST-DBSCAN) Algorithm, Annals of the American Association of Geographers 108(4):1168-1186.

 PySDA Diffusion Structure: TrAcking Progression In Time And Space (TaPiTaS) | Tutorial

Language: Python 3.x
Introduction: A diffusion process can be considered as the movement of linked events through space and time. Therefore, space-time locations of events are key to identify any diffusion process. However, previous clustering analysis methods have focused only on space-time proximity characteristics, neglecting the temporal lag of the movement of events. We argue that the temporal lag between events is a key to understand the process of diffusion movement. Using the temporal lag could help to clarify the types of close relationships. This study aims to develop a data exploration algorithm, namely the TrAcking Progression In Time And Space (TaPiTaS) algorithm, for understanding diffusion processes. Based on the spatial distance and temporal interval between cases, TaPiTaS detects sub-clusters, a group of events that have high probability of having common sources, identifies progression links, the relationships between sub-clusters, and tracks progression chains, the connected components of sub-clusters. Dengue Fever cases data was used as an illustrative case study. The location and temporal range of sub-clusters are presented, along with the progression links. TaPiTaS algorithm contributes a more detailed and in-depth understanding of the development of progression chains, namely the geographic diffusion process.
Further Reading: Chin WC, Wen TH *, Sabel C, Wang IH (2017), A Geo-Computational Algorithm for Exploring the Structure of Diffusion Progression in Time and Space, Scientific Reports 7:12565.

Assessing Personal Exposure to Disease Transmission Risk
      
Language: Microsoft Visual C# + Android app development
Introduction: Due to the complex interactions between human behaviors and the environment, it is important to quantify the association between environmental exposure and human health. Recent studies have indicated that different
behaviors by an individual may result in different levels of risk exposure. Different personal behaviors should be incorporated into a framework used to assess risk exposure. Therefore, we established a location-based client-server architecture to assess the exposure of an individual to the risk of contagious disease transmission. The framework integrates with the transmission dynamics of the disease simulation models and can reflect transmission risks due to behavioral changes in a timely manner. Apps on smart phones are an appropriate platform for individuals to effectively collect information on changes in personal mobility behaviors. Through a location-based framework, individuals can also receive timely risk alerts to assess their movement patterns and reduce their risks of exposure to infection. We successfully integrated disease transmission dynamics with mobility behaviors and developed a personalized exposure assessment framework that will broadcast the individual risks of exposure to infection in a timely manner. We were also invited to demonstrate the architecture prototype in Future Tech Expo in 2017.
Further Reading: Wen TH *, Hsu CS, Sun CH, Jiang JA, Juang JY (2018), A Location-Based Client-Server Framework for Assessing Personal Exposure to the Transmission Risks of Contagious Diseases, In Shaw SL and Sui D (Eds.). Human Dynamics Research in Smart and Connected Communities, Springer Series: GeoJournal Library, Springer.

Detecting Spatial-Temporal Thresholds of Food-borne Disease Outbreaks
  
Language: Microsoft Visual C# + Power BI
We integrated Real-time Outbreak and Disease Surveillance System (RODS), Laboratory Automated Reporting System (LARS) and Sentinel Surveillance System for establishing an early-warning framework for warning possible outbreaks of food-borne diseases. A Block Maxima Model was conducted to determine epidemic thresholds and detect unusual diarrhea clusters from RODS for warning the possible outbreaks. The model results are validated from the reported cases in the LARS database and syndromic surveillance data. RODS’s Diarrhea hotspots in 2016 were analyzed to establish epidemic thresholds for warning possible outbreaks of food-borne pathogen infection for each age group in 2017. Our results showed that, for each age group, the sensitive (true positive rate) is above 0.85, and positive likelihood ratio (LR+) is also above 2.0. The specificity (true negative rate) is above 0.6 and negative likelihood ratio (LR-) is below 0.2. Our early-warning framework successfully demonstrated the performance of detecting locations and timing of possible epidemic events prior to the time local health bureau received reported cases. The early-warning framework shows the acceptable ability of warning possible outbreaks of food-borne pathogen infection. We concluded that the diarrhea hotspots in RODS could reflect the clusters of hospitals with food-borne pathogen infection in LARS.
Further Information: Taiwan Centers for Disease Control (Taiwan-CDC) Research Grant IDs: MOHW106-CDC-C-114-000301; MOHW105-CDC-C-114-000301; MOHW104-CDC-C-114-000704

Flow-based PageRank (FBPR): A Flow-based Ranking Algorithm
Language: Python 3.x
Introduction: For a growing number of developing cities, the capacities of streets cannot meet the rapidly growing demand of cars, causing traffic congestion. Understanding the spatial–temporal process of traffic flow and detecting traffic congestion are important issues associated with developing sustainable urban policies to resolve congestion. Therefore, the objective of this study is to propose a flow-based ranking algorithm for investigating traffic demands in terms of the attractiveness of street segments and flow complexity of the street network based on turning probability. Our results show that, by analyzing the topological characteristics of streets and volume data for a small fraction of street segments in Taipei City, the most congested segments of the city were identified successfully. The identified congested segments are significantly close to the potential congestion zones, including the officially announced most congested streets, the segments with slow moving speeds at rush hours, and the areas near significant landmarks. The identified congested segments also captured congestion-prone areas concentrated in the business districts and industrial areas of the city. Identifying the topological characteristics and flow complexity of traffic congestion provides network topological insights for sustainable urban planning, and these characteristics can be used to further understand congestion propagation.
Further Reading:  Wen TH *, Chin WC, Lai PC (2017), Understanding the Topological Characteristics and Flow Complexity of Urban Traffic Congestion, PHYSICA A: Statistical Mechanics and Its Applications 473:166-177

Geographically Modified PageRank Algorithms
Language: Python 2.x
Introduction:  A network approach, which simplifies geographic settings as a form of nodes and links, emphasizes the connectivity and relationships of spatial features. Topological networks of spatial features are used to explore geographical connectivity and structures. The PageRank algorithm, a network metric, is often used to help identify important locations where people or automobiles concentrate in the geographical literature. However, geographic considerations, including proximity and location attractiveness, are ignored in most network metrics. The objective of the present study is to propose two geographically modified PageRank algorithms—Distance-Decay PageRank (DDPR) and Geographical PageRank (GPR)—that incorporate geographic considerations into PageRank algorithms to identify the spatial concentration of human movement in a geospatial network. Our findings indicate that in both intercity and within-city settings the proposed algorithms more effectively capture the spatial locations where people reside than traditional commonly-used network metrics. In comparing location attractiveness and distance decay, we conclude that the concentration of human movement is largely determined by the distance decay. This implies that geographic proximity remains a key factor in human mobility.
Further Reading:  Chin WC, Wen TH * (2015), Geographically Modified PageRank Algorithms: Identifying the Spatial Concentration of Human Movement in a Geospatial Network, PLOS ONE 10(10): e0139509.

A Stirring Genetic Algorithm (SGA) with Spatial and Temporal Weighting Schemes
Language: MATLAB
Introduction:  Out-of-hospital cardiac arrest (OHCA) occurs when the heart is deprived of oxygen without immediate medical treatment. The use of publicly accessible automated external defibrillators (AEDs) is generally considered to be an effective pre-hospital measure. Although studies have been undertaken to determine the locations of AED installations, the spatial and temporal characteristics of OHCA occurrence have not been considered comprehensively. This study attempts to assess the feasibility of using the 7-Eleven chain of convenience stores as possible locations for the installation of AEDs to capture the spatial and temporal characteristics of OHCA patients. The methodological framework was divided into two stages. The first stage involved the development of two weighting schemes, a temporally weighted model (TWM) and a spatially weighted model (SWM), to capture the temporal and spatial variations in selecting AED locations. In the second stage, we proposed a stirring genetic algorithm (SGA) to select the limited subset of 7-Elevens covering the most weighted OHCA occurrences from the first stage. We considered two modes of conveyance, human running and vehicle transportation, by setting the service range of the 7-Elevens at 100 and 300 m. We conclude that each 7-Eleven has a different role for allocating AEDs in an urban setting. The AEDs at 7-Elevens in commercial areas help to compensate for the temporal gap of emergency medical service (EMS) in nighttime occurrences and for a high incidence of OHCA patients. For convenience stores in residential areas, AEDs help to compensate for the spatial gap in areas that are far from fire stations.
Further Reading:  Tsai YS, Ko PC, Huang CY, Wen TH * (2012) Optimizing Locations for the Installation of Automated External Defibrillators in Urban Public Streets Through the Use of Spatial and Temporal Weighting Schemes, Applied Geography 35: 394-404.

Estimating the Place-of-Residence in National Health Insurance Research Databases (NHIRD) | code
Language: SAS Macro
Introduction: Understanding the geographic patterns and regional differences in health status plays an important role in public health research; however, the place-of-residence (township level) of an insured is not available in National Health Insurance Research Databases (NHIRD). The objective of this study was to propose principles for estimating the place-of-residence (township level) in NHIRD. Based on demographic characteristics, insurance classification, location of hospital visit, and insurance registration of the insured, this study compared three methods of estimating the place-of-residence (township level) from the Longitudinal Health Insurance Database of NHIRD in 2005. Official statistics of the usual resident population in each township from the Department of the Interior were used as reference data for comparisons among the three methods. The study further verified these methods by comparing the estimated numbers and official statistics for the medical treatment of lung and liver cancer patients in 2005. This study found that the method which combined insurance classification, location of hospital visit, and insurance registration provided an optimal estimate of place-of-residence in each area by different levels of urbanization and age-group. Consideration of either location of hospital visit or insurance registration may be appropriate for specific townships and age groups. This study demonstrated the feasibility of estimating place-of-residence in NHIRD and the applicability of these proposed methods.
Further Reading: Lin MH, Yang AC, Wen TH * (2011), Using Regional Differences and Demographic Characteristics to Evaluate the Principles of Estimation of the Residence of the Population in National Health Insurance Research Databases (NHIRD), Taiwan Journal of Public Health 30(4): 347-360.

Multilayer Epidemic Dynamics Simulator (MEDSim) | Demo
Language: MATLAB
Introduction: We describe an innovative simulation framework that combines daily commuting network data with a commonly used population-based transmission model to assess the impacts of various interventions on epidemic dynamics in Taiwan. Called the Multilayer Epidemic Dynamics Simulator (MEDSim), our proposed framework has four contact structures: within age group, between age groups, daily commute, and nationwide interaction. To test model flexibility and generalizability, we simulated outbreak locations and intervention scenarios for the 2009 swine-origin influenza A (H1N1) epidemic. Our results indicate that lower transmission rates and earlier intervention activation times did not reduce total numbers of infected cases, but did delay peak times. When the transmission rate was decreased by a minimum of 70%, significant epidemic peak delays were observed when interventions were activated before new case number 50; no significant effects were noted when the transmission rate was decreased by less than 30%. Observed peaks occurred more quickly when initial outbreaks took place in urban rather than rural areas. According to our results, the MEDSim provides insights that reflect the dynamic processes of epidemics under different intervention scenarios, thus clarifying the effects of complex contact structures on disease transmission dynamics.
Further Reading: Tsai YS, Huang CY, Wen TH *, Sun CT, Yen MY (2011), Integrating Epidemic Dynamics with Daily Commuting Networks: Building a Multilayer Framework to Assess Influenza A (H1N1) Intervention Policies. Simulation: Transactions of The Society for Modeling and Simulation International 87(5): 385-405.

e-Atlas: Taiwan Census Explorer 1956-2010 | link
Language: HTML + Javascript
Introduction: We built the Web-based Taiwan Social Explorer, which is an interactive map platform for exploring Taiwan census data spatially and temporally. The control panel on the right can be used to display different data layers. A map legend, figures, and datasets (.csv) can be accessed using the control panel on the left. Data for this map platform was provided by the Taiwan Census Office (1956, 1966, 1980, and 1990 data) and the Directorate-General of Budget, Accounting, and Statistics (2000 and 2010 data).