GIS Open Source Application as a Support to a Hospital

Morbidity Database

Hospital GIS

Lia Duarte

1,2

, Ana Cláudia Moreira Teodoro

1,2

and Alberto Freitas

Department of Geosciences, Environment and Land Planning, Faculty of Sciences, University of Porto, Porto, Portugal

Earth Sciences Institute (ICT), Faculty of Sciences, University of Porto, Porto, Portugal

Center for Health Technology and Services Research (CINTESIS), Faculty of Medicine, University of Porto,

Porto, Portugal

Keywords: GIS, Open Source, Hospital Morbidity Database, Administrative Data, PostGIS.

Abstract: Geographical Information Systems (GIS) capabilities are increasing in health area. GIS have been used to

investigate global health studies due to the huge capabilities to manipulation, storage, management, analysis,

modelling and mapping of geographical data. A simple and intuitive graphic interface to represent spatially

an administrative database information would be a great usefulness to the health community. In this work, an

open source application was developed in Python language under a GIS open source software (QGIS). The

application, incorporated in GIS software, is composed by two tabs: Symbology and Mapping. In order to test

the developed application, a zone from Porto Metropolitan Area and a database with administrative data,

hospital morbidity database, was considered. This data was previously added to PostGIS (an open source

database) and automatically connected to the application. The difficulty of health professionals in the creation

of multiple visualizations of tabular data defined by rigorous position, and the maps creation to later analysis

and printing, can be overcome with this application. The large amount of data requires the connection to a

free database in GIS environment enhancing the practical applicability, rapid, safe and efficient data

representation.

1 INTRODUCTION

Spatial epidemiology is the study of geographical

variation in disease risk or incidence (Kelen et al.,

2012). In 1854, John Snow created the first map

relating environmental factors in order to investigate

the base of cholera deaths (Snow, 1855). Nowadays,

Geographical Information Systems (GIS) capabilities

are increasing in health area. GIS have been used to

investigate global health studies due to the huge

capabilities to manipulation, storage, management,

analysis, modelling and mapping of geographical

data. GIS presented new opportunities for researchers

providing the tools required for exploring the

geographic variation in disease risk relating the

geographically indexed health events with

demographic, environmental, behavioural,

socioeconomic and genetic risk factors (Zhang et al.,

2016). The spatial epidemiology takes advantage of

GIS tools combined with Remote Sensing (RS) to

enhance accessibility to spatial data and to measure

the spatial-temporal variation in disease risks (Kelen

et al., 2012; Jeong et al., 2016). For example, Zhang

et al. (2016) review and analyse the types of spatial

measurement errors more commonly encountered

during spatial epidemiological analysis of spatial data

combining GIS, RS, Global Positioning Systems

(GPS) and statistical methodologies. In order to

perform GIS spatial-temporal analysis, other study

conducted by Shiode et al. (2015), prepares their own

data on the estimated number of residents at each

house location along with the space-time data of the

victims. Several studies applied to health spatial-

analysis were performed using GIS tools (Gómez-

Barroso et al., 2016; Ferguson et al., 2016;

Ruktanonchai et al., 2014; Ayres-Sampaio et al.,

2014; Sadler, 2016; Jeong et al., 2016; Makanga et

al., 2016; Panciera et al., 2016). Unfortunately, all

these studies are mainly focused on applying GIS

tools under proprietary software. Other works used

GIS under web environment used PostGIS databases

Duarte, L., Teodoro, A. and Freitas, A.

GIS Open Source Application as a Support to a Hospital Morbidity Database - Hospital GIS.

DOI: 10.5220/0006268901690176

In Proceedings of the 3rd International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2017), pages 169-176

ISBN: 978-989-758-252-3

169

and open servers (Moncrieff et al., 2014; Smith and

Hayward, 2016; Bui and Pham, 2016).

Some GIS applications were developed under free

and open source software such as OpenJump (Fisher

and Myers, 2011) and a new open source tool,

openModeller was created (Muñoz et al., 2011).

There is not an open source application under

QGIS software developed specifically with the

purpose of spatially representing a huge amount of

data storage in a database, and automatically

symbolize and create a map. An open source

application allowing health experts or common users

(not familiarized with GIS environment) to represent

spatially the database information, with a simple and

intuitive graphic interface, would be a great

usefulness to the health community. This research

was addressed to health experts who handle with

spatial geographic data in scientific research field

improving the GIS analysis capabilities in health

spatial information field.

The objective of this work is the development of

a GIS open source application which allows: (i) to

automatically connects to a database in order to

access the (huge amount of) data; (ii) to create maps

for printing with information provided by a database;

(iii) to automatize the vector files symbology:

graduated or categorized; (iv) to connects to Bing

Aerial Maps in GIS environment in order to overlap

the database information; (v) to create a polygon file

with the extension of the study area, and; (vi) to

convert a shapefile to Keyhole Markup Language

(KML) format. This application can be easily used to

improve the studies referred creating the required

maps with an easy connection to an open source

database.

This manuscript is divided into 5 sections:

Introduction, Methodology, Results, Discussion and

Conclusions. The Methodology section is divided

into PostGIS database subsection and Hospital GIS

subsection where are explained the API libraries

(QGIS and Python) used and the application

development.

2 METHODOLOGY

The application was developed under the GIS open

source software QGIS (QGIS, 2016). QGIS is an

open source software developed by Gary Sherman in

2002 which respects the Stallman four freedoms:

freedom to run the program for any purpose, to study

how the program works and modify it, to redistribute

copies and to distribute copies of modified versions

(Stallman, 2007). It is developed in C++ and

complemented with Python extensions or plugins.

QGIS presents several advantages in the plugins

development using Python language, as it had several

libraries to use. Python language is also an open

source language, interpreted, high level, and object

oriented language (Python, 2016). Several libraries

and Application Programming Interfaces (API’s)

were used. QGIS has its own API’s such as QGIS

API, Geospatial Data Abstraction Library

(GDAL)/OGR API and PyQt4 API (PyQt4, 2016;

QGIS API, 2016; GDAL, 2016) and it supports vector

files, raster formats and spatial databases (e.g.,

PotgreSQL and PostGIS; QGIS, 2016). The APIs are

composed by modules, classes and functions which

help to connect the graphic interfaces with the spatial

information manipulation. With the implementation

and use of the algorithms provided by QGIS software,

this application improves the using of GIS

functionalities and connection to databases with

multiple data using a simple and intuitive interface.

2.1 PostGIS Database

PostgreSQL is a Relational Database Management

System (RDBMS) which manages data stored with

relationships. PostGIS is a GIS spatial database which

adds spatial support to PostgreSQL and follows the

interoperability standards from Open Geospatial

Consortium (OGC; PostGIS, 2016). It is open source

and is available through GNU General Public License

(GLP) license. PostGIS supports specified geometries

from OGC: point and polygon, and supports

geographic objects allowing their location through

SQL (SQL, 2016). SQL language is a standard

universal language used in the database manipulation

through RDBMS and allowing several tasks such as

insertion, modification and object creation, user

management, information query, among others. The

most common operation in SQL, the query, makes

use of the declarative SELECT statement which

retrieves data from one or more tables or expressions.

In the work presented, the SELECT command was

used to database queries (SQL language, 2016).

2.2 Hospital GIS

The aim of this work was the creation of HospitalGIS

application which allows to relate the database

information in an intuitive and simple way. The

graphic interface was created through Qt Designer

and all the configurations were applied automatically.

The application is composed by a button added to

QGIS tools.

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

170

Through the mouse click in the button, a dock

widget opened as a panel incorporated in QGIS

environment in a way that the user can obtain

information from the data visualization. Figure 1

presents the button and the HospitalGIS graphic

interface.

Figure 1: HospitalGIS graphical button and graphic

interface.

HospitalGIS is composed by two tabs with

different functionalities: Symbology and Mapping.

The first one is composed by the symbology

parameters regarding the maps and presents four

combo boxes: the information level (village,

municipality, district, Nomenclatura das Unidades

Territoriais para Fins Estatísticos (NUTS II)

residential, NUTS II hospital, NUTS III residential,

NUTS III hospital); the attribute to visualize; the

symbology style according to the information type (it

can be Categorized when the attribute is discrete or

Graduated when the attribute is numeric); and finally

the ramp color. In Symbology tab, some parameters

from graduated classes are also presented, such as:

number of classes which the user can choose and the

statistical method which can be applied to the study

variable (minimum, maximum, average, mode and

standard deviation). The labels (text) defined in the

combo box Attribute are automatically defined from

the village shapefile attribute table. This shapefile is

automatically incorporated in the application. The

other combo boxes were created with predefined

styles according to the symbology type (Categorized

or Graduated). The color ramp was assigned with a

set of colors and added to the respective color ramp.

The number of classes is defined with value 5, by

default, but can be modified by the user.

The Mapping tab incorporates the auxiliary

cartography and contains several functionalities to

add other types of maps or creation of maps for

printing. The tab is composed by two check boxes and

three buttons. The check boxes allow the addition of

an interactive map from Bing Aerial Maps (Open

Web Map) and the possibility of adding the extension

zone shapefile, so the user can verify the delimited

extension (Open study area limit). These options can

be performed after the map symbology. The user can

create a map for printing (Layout) with the

symbolized map and export the shapefile to kml

format to verify the result through Google Earth or

Google Maps (Export to KML).

2.2.1 API libraries (QGIS and Python)

QGIS API or PyQGIS allows to virtually control the

QGIS graphic environment and it is based on QGIS

C++ API divided by five categories: Core, GUI,

Analysis, Map Composer and Network Analysis

(QGIS API, 2016). The PyQt4 API is composed by

several Python modules developed for Qt framework.

This is a multiplatform framework composed by a set

of C++ libraries built to the development of graphic

interfaces, Structured Query Language (SQL)

databases, Scalable Vector Graphics (SVG), Open

Graphics Library (OpenGL), eXtensible Markup

Language (XML) and other configurations (PyQt4

API, 2016).

Processing Toolbox algorithms were also used in

the developed application. Processing Toolbox

belongs to QGIS and it is a framework composed by

a set of algorithms disposed in a tree (Sextante, 2016).

The algorithms belong to external applications with

geoprocessing capacities, such as System for

Automated Geoscientific Analyses (SAGA),

Geographic Resources Analysis Support System

(GRASS) or R (SAGA, 2016; GRASS, 2016; R,

2016). These algorithms were crucial to the

application development. The plugins development

follows a specified structure provided by QGIS

official page (QGIS, 2016): have an idea, create the

files required, write the code and test while writing,

and in the end publish the plugin in the official page.

Several scripts were developed according to the rules

defined. The graphic interface was created through

the Plugin Builder/Qt Designer extension which

configures all the files needed (Plugin Builder 2015;

Plugin Builder Documentation 2015). Widgets from

Qt API were created, such as combo boxes, check

boxes, push buttons, labels, edit lines, among others.

2.2.2 Application Development

The application architectureis presented in Fig.2. The

application was created based in two classes, the main

class (composed by 15 functions) and a class referred

to the kml conversion. Through the initGui function,

an automatic reading of hospital database is

performed to verify the existent attributes and add

them to the Attribute combo box. In this function the

GIS Open Source Application as a Support to a Hospital Morbidity Database - Hospital GIS

171

Figure 2: HospitalGIS architecture.

strings to the Level, Style and Color Ramp combo

boxes are also added, such as:

self.dockwidget.comboBox_2.addItems(['C

ategorized', 'Graduated'])

The Color Ramp combo box incorporates an

image. This image is considered as an icon, so QIcon

class was used. The buttons connections through the

clique action are also performed in this function. The

Classify function is called through the clique action

and incorporates the Categorized and Graduated

types of symbology. After that, if the check box Open

Web Map is activated, Bing Map is added to the QGIS

area and then the map is symbolized. If the

symbolization is Categorized,

symbology_categorized is called, and if is

Graduated, symbology_graduated function is called.

Symbology_categorized function connects to

PostGIS database through QGIS API classes which

stored components from the Uniform Resource

Identifier (URI) data source from

PostgreSQL/RDBMS. The structure stored the

information related to the database connection

including server, database, username and password,

scheme and, if exist, the SQL conditions. Then the

database connection is performed and consequently

opens. Through SQL language a code line was

created to join the shapefile and the database through

village code, municipality and district (dicofre), and

the attribute variable mode estimation to each village.

Figure 3 presents a schema with the connections

established.

In order to execute the SQL condition, the

pgsql2shp.exe application from PostgreSQL was

used. This application applies a SQL condition saving

the result into a new shapefile. The new shapefile

should contain two fields, the village name and the

statistical value (estimated) of the chosen variable to

Figure 3: HospitalGIS workflow presenting the

connections.

each village. QGIS functions were used to read these

functions. The second column (statistical variable) is

acquired to use in the symbology.

In the next step, all the occurrences of attribute

column, through a for cycle, were recorded in a new

list. Repeated elements were eliminated through the

set Python function. The categorized symbology

should be built through a specific dictionary with the

column values, the label and the associated color.

Finally the shapefile is added to QGIS environment.

If the Open study area limit check box is checked,

the extent function is used to zoom in the extension

area. Dissolve function from Processing Toolbox was

used to dissolve the extension zone in a unique

polygon and overlap the shapefile. Another condition

was added to verify if interactive map was selected.

If true, the shapefile (European Terrestrial Reference

System 1989 Portugal Transversal Mercator 2006 -

ETRS89 PTTM06; European Petroleum Survey

Group (EPSG): 3763) is projected to World Geodetic

System 1984 (WGS84) Pseudo Mercator

(EPSG:3857) through the QGIS reprojectlayer

algorithm. ETRS89 PTTM06 system is actually

mandatory in Portugal according to Infrastructure for

Spatial InfoRmation in Europe (INSPIRE) directive

(Inspire, 2015). Inspire directive has been in effect

since May 15th 2007. It is composed by several areas:

metadata, geographic data interoperability, network

services, data share, and coordination and monitoring

(Inspire, 2015). The WGS84 Pseudo Mercator

system, also nominated by Web Mercator, Google

Web Mercator, Spherical Mercator or WGS84 Web

Mercator, is the coordinate system used in web

mapping applications and it is associated to Google

Maps, Bing Maps, OpenStreetMap, Mapquest,

Mapbox among others (Web Mercator, 2015; Spatial

Reference, 2015; Spherical Mercator, 2015).

The symbology_graduated function is connected

to PostGIS database where the hospital information is

stored. Several if statements were defined with a SQL

condition according to the attribute and the statistical

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

172

method chosen by the user. Table 1 presents an

example of a condition which relate the hospital

database with the study area shapefile through the

join operation. The statistic is saved and applied to the

chosen variable into a new shapefile. The following

command line present an example of a SQL

condition.

SELECT freguesia,

sum(totdias)/count(totdias), geom FROM

grande_porto LEFT OUTER JOIN

sample_reside_join ON

grande_porto.dicofre =

sample_reside_join.reside GROUP BY

freguesia, geom ORDER BY freguesia

The Layout button allows the automatic creation

of a map for printing and is connected to the layout

function. In this function, the following layout

elements are defined: title, legend, graphic scale and

north arrow. The shapefile projection system is

verified and if it is ETRS89 PT-TM06, the layout is

created. Other way, the shapefile is projected through

reprojectlayer algorithm. The legend and title are

composed through QgsComposerLegend and

QgsComposerLabel classes, respectively. The north

arrow is added as an image and corresponds to the

cartographic north. Finally, a graphic scale is

automatically created according to the coordinate

system defined. The result is saved in tif format

through the printPageAsRaster function.

The addMap function allows to add the Bing

interactive map to the QGIS environment. The user

can symbolize the shapefile after or before adding the

map.

A new graphic interface to the kml conversion

was created. This action is performed when the user

clicks in Export to KML button. The QGIS

environment could have several shapefiles open, so

this graphic interface was created in order to give the

possibility to choose which file would be converted.

Figure 4 presents the graphic interface.

Figure 4: KML button graphic interface.

The new dialog box (Fig. 4) was created through

a new class Window. The function handleButton was

created, allowing to read all the vector files open in

QGIS environment, and to list the names adding them

to the combo box (Select layer to export). Also, in this

function, the Browse and Convert buttons

connections were defined, for the new kml file

directory and to convert the shapefile to kml format,

respectively. This process uses ogr2ogr algorithm

from GDAL/OGR library. This connection is

performed through the Python function subprocess.

3 RESULTS

In order to test the developed application, a zone from

Porto Metropolitan Area (PMA) and a database with

administrative data were used. These data were

previously added to PostGIS. The coordinate system

used was ETRS89 PT-TM06.

The national hospital morbidity database (an

administrative database, in PT, Morbilidade

Hospitalar, previously designated as Grupos de

Diagnóstico Homogéneo – GDH [in EN, Diagnosis

Related Groups]) is very often referred in the hospital

information system context. These databases are

composed by clinical-administrative data related to

hospital discharges (inpatient episodes, ambulatory

surgery and medic ambulatory) and other variables.

These administrative databases can contain incorrect

data and also data with some quality issues, but are

composed by data easily available, inexpensive and

frequently used. In some situations, it can be the only

available data to study a specified clinic question. It

can, for instance, be used as a quality indicators

production to the study and comparison of hospital

activities or in the study of relations between hospital

and environmental variables. In this context, GIS

tools can assume a focus to easily provide the

visualization and comparison of different outcomes

(hospitalization taxes, hospital morbidity) to certain

pathologies in time or in specified geographic areas

with adjustment to population data.

The categorized symbology was tested based on

the variable ADM_TIP attribute which corresponds to

the patients’ admission type (strings 1, 2 or 6). The

string 2 corresponds to urgent admissions and the

remaining strings are related to programmed

admissions. Figure 5 presents the assigned

characteristics in this specific case and the result

obtained.

In this case, the variable is discrete so the

application considers the mode and assign the value

to each village. The second part of the graphic

interface is blocked, so it can be used only with

graduated symbology. In the future, the application

will be improved to estimate the occurrences

percentage beyond the mode. The symbols without

GIS Open Source Application as a Support to a Hospital Morbidity Database - Hospital GIS

173

Figure 5: HospitalGIS graphic interface with the ADM_TIP

attribute characteristics.

legend corresponds to village without information.

The map for printing was created in A3 size and the

elements related to Grande Porto zone. A numeric

attribute was also tested, the variable totdias which

corresponds to the total number of inpatient days. The

study was also applied to the villages. The style

chosen was Graduated and the classes number were

defined by default as 5. The statistical method applied

was the average. Figure 6 presents the categorized

and graduated symbology with the respective maps.

Figure 6: a) Categorized ADM_TIP attribute symbology; b)

Categorized ADM_TIP attribute map for print; c)

Graduated tot_dias attribute symbology; d) Graduated

tot_dias attribute map for print.

Figure 7 presents examples using minimum value,

standard deviation and mode related to the inpatient

days variable.

In Figure 8 is presented the symbolized shapefile

overlapped with the limit of study zone and the

interactive map, in order to test the other

functionalities.

Figure 7: Examples considering a) minimum value, b)

standard deviation and c) mode related to the inpatient days

variable.

Figure 8: a) Symbolyzed shapefile overlapped with the limit

of study zone and the b) interactive map from Bing Aerial

Maps.

4 DISCUSSION

The database information was spatially distributed.

Different combinations and possibilities were tested

with the developed application and was concluded

that, in health area, this application could be very

valuable, improving the spatial analysis for the health

professionals. For instance, Figure 6b shows an

example of ADM_TIP attribute representation by

village which corresponds to patients’ admission type

(strings 1, 2 or 6). From the results obtained it can be

concluded that approximately 49% of the villages

don’t have information in the database (and they are

not presented in Figure 6). 12% of the villages of

PMA were classified as containing patients with

programmed admissions (string 1 or 6) and 39% of

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

174

the villages presented patients urgently admitted in

the hospital of Porto. In this particular study case the

mode of patients in each village was evaluated. If we

consider a quantitative variable, such as the average

total number of inpatient days (Figure 6d), we can

conclude that 42% of the villages contains 1 to 13

patients in the hospital, 8% with an average of 13-25

patients and 1 village has more than 43 patients in the

hospital (Vila Boa de Quires e Maureles (accessed

taking advantage of QGIS tools)). Other type of

relations can be performed with the tool. Also, this

application is based on an open source software, easy

to install and use. The database relationship with a

shapefile allows to locate and visualize

geographically and spatially the data, improving the

data analysis and interpretation. The interaction with

different information layers was also a great

advantage to a user non comfortable with GIS

software, or GIS operations such as PostGIS

connection, shapefiles creation and symbolization,

connections and calculations between several

variables and statistics applications under spatial

information. The application is free, open source,

available for any user that needs a tool easy to use and

understand, allows to create maps with associated

data, and also perform geographic and statistical

analysis from database information.

5 CONCLUSIONS

The developed GIS open source application is a

valuable tool to health area where the user must deal

with geospatial data with future applicability in the

data access and hospital episodes data manipulation

and analysis. The difficulty of health professionals in

the creation of multiple visualizations of tabular data

defined by rigorous position and the maps creation to

later analysis and print can be overcome with this

application. The large amount of data requires the

connection to a free database in GIS environment

enhancing the practical applicability, rapid, safe and

efficient data representation. The presented tool is a

preliminary version of a useful and efficient scientific

tool to produce the maps required to study health

variables, helping in the decision support. Some

improvements will be done in the future automatizing

even more the developed application such as (i)

extending the data visualization to a WebGIS,

through the creation of a web page that allows the

hospital administrative data visualization using

Google Maps or Google Earth as interactive maps, (ii)

performing spatio-temporal statistical analysis,

evaluating the trend prediction of the data with the

creation of plots, (iii) developing functionalities to

compare parameters in different regions, (iv) relating

the information of the database with environmental

risk factors such as temperature, precipitation or even

vegetation indices and (v) the possibility of inserting

satellite imagery will be implemented The application

is easy to install and use in GIS environment and is

available in http://www.fc.up.pt/pessoas/liaduarte

/HospitalGIS.rar. The present application can only be

tested under inside server.

ACKNOWLEDGEMENTS

Project "NORTE-01-0145-FEDER-000016"

(NanoSTIMA) is financed by the North Portugal

Regional Operational Programme (NORTE 2020),

under the PORTUGAL 2020 Partnership Agreement,

and through the European Regional Development

Fund (ERDF).

REFERENCES

A linguagem SQL. Documentação do PostgreSQL 8.0.0.

2015. http://pgdocptbr.sourceforge.net/pg80/tutorial-

sql.html. Accessed October 2015.

Ayres-Sampaio, D., Teodoro, A.C., Sillero, N., Freitas, A.,

2014. An investigation of the environmental

determinants of asthma hospitalizations: An applied

spatial approach. Applied Geography, 47:10–19.

Bui, T.Q., Pahm, H.M., 2016. Webbased GIS for spatial

pattern detection: application to malaria incidence in

Vietnam. SpringerPlus, 5, 1014.

Duarte, L., Teodoro, A.C., Gonçalves, J.A., Guerner Dias,

A.J., Espinha Marques, J., 2015. A dynamic map

application for the assessment of groundwater

vulnerability to pollution. Environmental Earth

Sciences, 74(3), 2315-2327.

Ferguson, W.J., Kemp, K., Kost, G., 2016. Using a

geographical information system to enhance patient

access to point-of-care diagnostics in a limited resource

setting. International Journal of Health Geographics,

15:10.

Fisher, R.P., Myers, B.A., 2011. Free and simple GIS as

appropriate for health mapping in a low resource

setting: a case study in eastern Indonesia. International

Journal of Health Geographics, 10:15.

GDAL. Geospatial Data Abstraction Library. 2015.

http://www.gdal.org/. Accessed September 2015.

Gómez-Barroso, D., García-Pérez, J., López-Abente, G.,

Tamayo-Uria, I., Morales-Pia, A., Romaguera, E.P.,

Ramis, R., 2016. Agricultural crop exposure and risk of

childhood cancer: new findings from a case-control

study in Spain. International Journal of Health

Geographics, 15:18.

GIS Open Source Application as a Support to a Hospital Morbidity Database - Hospital GIS

175

GRASS. The world’s leading Free GIS software. 2015.

http://grass.osgeo.org/. Accessed September 2015.

Inspire. Infrastructure for Spatial Information in the

European Community. 2015. http://inspire.ec.europa

.eu/index.cfm/pageid/48. Accessed October 2015.

Jeong, B., Joo, H.-T., Shin, H.-S., Lim, M.-H., Park, J.-C.,

2016. Geographic information system analysis on the

distribution of patients visiting the periodontology

department at a dental college hospital. Journal of

Periodontal and Implant Science, 46(3), 207-217.

Kelen, P.T.V., Downs, J.A., Stark, L.M., Loraam, R.W.,

Anderson, J.H., Unnasch, T.R., 2012. Spatial

epidemiology of eastern equine encephalitis in Florida.

International Journal of Health Geographics, 11:47.

Luan, H., Law, J., 2014. Web GIS-Based Public Health

Surveillance Systems: A Systematic Review. ISPRS

International Journal Geo-Information, 3, 481-506;

Makanga, P.T., Schuurman, N., von Dadelszen, P., Firoz,

T., 2016. A scoping review of geographic information

systems in maternal health. International Journal of

Gynecology and Obstetrics, 134, 13-17.

Moncrieff, S., West, G., Cosford, J., Mullan, N., Jardine,

A., 2014. An open source, server-side framework for

analytical web mapping and its application to health.

International Journal of Digital Earth, 7(4), 294-315.

Munoz, M.E.S., Giovanni, R., Siqueira, M.F., Sutton, T.,

Brewer, P., Pereira, R.S., Canhos, D.A.L., Canhos,

V.P., 2011. openModeller: a generic approach to

species’ potential distribution modelling.

Geoinformatica, 15, 111–135.

Panciera, R., Khan, A., Rizvi, S.J.R., Ahmed, S., Ahmed,

T., Islam, R., Adam, A.M., 2016. The influence of

travel time on emergency obstetric care seeking

behaviour in the urban poor of Bangladesh: a GIS study.

BMC Pregnancy and Childbirth, 16:240.

Plugin: Plugin Builder. Plugin: Plugin Builder, QGIS

Python Plugins Repository. 2015. https://plugins.qgis

.org/plugins/pluginbuilder/. Accessed October 2015.

Plugin Builder Documentation. QGIS Plugin Builder 2.8.1

documentation. 2015. http://g-sherman.github.io/Qgis-

Plugin-Builder/. Accessed October 2015.

PostGIS. Spatial and Geographic objects for PostgreSQL.

2015. http://postgis.net/. Accessed October 2015.

PyQt4 API. PyQt Class Reference. 2015. http://pyqt.

sourceforge.net/Docs/PyQt4/classes.html. Accessed

August 2015.

Python. Python Programming Language. 2015.

http://python.org/. Accessed August 2015.

QGIS API. Quantum GIS API Documentation. 2015.

http://www.qgis.org/api/. Accessed September 2015.

QGIS. Quantum GIS Project. 2015. http://www.qgis.org/.

Accessed August 2015.

R. The R Project for Statistical Computing. 2015.

https://www.r-project.org/. Accessed September 2015.

Ruktanonchai, C.W., Pindolia, D.K., Striley, C.W.,

Odedina, F.T., Cottler, L.B., 2014. Utilizing spatial

statistics to identify cancer hot spots: a surveillance

strategy to inform community-engaged outreach

efforts. International Journal of Health Geographics,

13:39.

Sadler, R.C., 2016. Integrating expert knowledge in a GIS

to optimize siting decisions for small-scale healthy food

retail interventions. International Journal of Health

Geographics, 15:19.

SAGA. System for Automated Geoscientific Analyses.

2015. http://www.saga-gis.org/. Accessed September

2015.

SEXTANTE. The SEXTANTE framework. 2015.

http://www.sextantegis.com/. Accessed September

2015.

Smith, C.M., Hayward, A.C., 2016. DotMapper: an open

source tool for creating interactive disease point maps.

BMC Infectious Diseases, 16:145.

Snow, J., 1855. On the mode of communication of cholera.

2nd edition. London: Jonh Churchill.

Spatial Reference. sr-org projection 7483 - epsg:3857 --

wgs84 web mercator (auxiliary sphere). 2015.

http://spatialreference.org/ref/sr-org/7483/. Accessed

October 2015.

Spherical Mercator. EPSG:3857. 2015. http://wiki.

openstreetmap.org/wiki/EPSG:3857. Accessed

October 2015.

SQL. SQL Wikipedia. 2015. https://pt.wikipedia.org/wiki

/SQL. Accessed October 2015.

Stallman P. Why ‘Open Source’ misses the point of free

software, GNU Operating System. 2007.

http://www.gnu.org/philosophy/open-sourcemisses-

the-point.html. Accessed February 2011. Accessed

September 2015.

Teodoro, A.C., Duarte, L., 2013. Forest Fire risk maps: a

GIS open source application – a case study in Norwest

of Portugal. International Journal of Geographic

Information Science, 27(4), 699-720.

Web Mercator. Web Mercator Wikipedia. 2015.

https://en.wikipedia.org/wiki/Web_Mercator.

Accessed October 2015.

Zhang, Z., Manjourides, J., Cohen, T., Hu, Y., Jiang, Q.

2016. Spatial measurement errors in the field of spatial

epidemiology. International Journal of Health

Geographics, 15:21.

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

176