GIS SUPPORT FOR EMERGING INFECTIOUS DISEASES IN EAST ... · GIS support for emerging infectious diseases in East Africa 9. Emerging infectious diseases cause significant health and

GIS SUPPORT FOR

EMERGING INFECTIOUS DISEASES

IN EAST AFRICA

A dissertation submitted to

The University of Manchester

for the degree of

Master of Science in the Faculty of

Engineering and Physical Sciences

2010

By

JOHN MUUMBI

SCHOOL OF COMPUTER SCIENCE

Table of Contents

Table of Contents

List of Tables and Figures......................................................................................................4

Abstract..................................................................................................................................5

Declaration.............................................................................................................................6

Copyright................................................................................................................................7

Acknowledgements................................................................................................................8

Chapter 1 Introduction and Background...............................................................................9

1.1 Introduction: Emerging Infectious Diseases (EIDs)........................................................9

1.2 What are emerging infectious diseases and why are they important?.............................9

1.2.1 Avian influenza. Origins and effects. .........................................................................11

1.2.2 Swine Flu. Origins and impact...................................................................................13

1.3 What is AVID?...............................................................................................................14

1.3.1 Rift Valley Fever as a case study................................................................................15

1.3.2 Prevention is better than cure (and ultimately less expensive)..................................16

1.4 Disease monitoring and outbreak prediction.................................................................19

1.4.1 Arthropod Borne Viruses............................................................................................19

1.4.2 Predictive modelling...................................................................................................24

1.5 Geographical Information Systems...............................................................................27

1.5.1 What is a Geographical Information System, (GIS)? ................................................27

1.5.2 Related Areas..............................................................................................................31

Chapter 2 Methodology.......................................................................................................35

2.1 Overview of existing application..................................................................................35

2.1.1 Design considerations.................................................................................................37

2.2 Web mapping ................................................................................................................38

2.2.1 Evaluation...................................................................................................................39

2.3 Building blocks..............................................................................................................44

2.3.1 Hypertext Pre Processor (PHP)..................................................................................44

2.3.2 MySQL Relational Database Management System...................................................45

2.3.3 Apache Web Server....................................................................................................46

2.3.4 Google Maps API.......................................................................................................47

GIS support for emerging infectious diseases in East Africa 2

2.3.5 Keyhole Markup Language........................................................................................48

2.4 Conclusion: Bringing it all together..............................................................................48

Chapter 3 Implementation...................................................................................................51

3.1 Implementation Objectives ...........................................................................................51

3.1.1 Data acquisition and preparation ...............................................................................52

3.1.2 Data Processing..........................................................................................................54

3.1.3 Data Presentation........................................................................................................59

3.2 Typical use case.............................................................................................................78

Chapter 4 Conclusion..........................................................................................................80

4.1 Review ..........................................................................................................................80

4.2 Challenges.....................................................................................................................81

4.3 Recommendations.........................................................................................................82

4.4 Future work...................................................................................................................83

References............................................................................................................................84

Appendices...........................................................................................................................89

Appendix A: H1N1 Worldwide Incidences ....................................................................89

Appendix B: Global Risk Map for Rift Valley Fever .....................................................90

Appendix C: WampServer 2.0i Installation Guide .........................................................91

Appendix D: Installing the Application...........................................................................96

Appendix E: Source Codes..............................................................................................99

Word Count: 18,544 (Main text )


List of Tables and Figures

List of Tables

1.1 Confirmed cases of HPAI.…................................................................................................. 111.2 Confirmed cases of H1N1.…................................................................................................ 143.1 Organisms table structure.…................................................................................................ 553.2 Samples table structure.….................................................................................................... 553.3 Projects table structure.…..................................................................................................... 563.4 ELISA table structure.…...................................................................................................... 563.5 Storage table structure.…..................................................................................................... 573.6. Spatial table structure.…...................................................................................................... 573.7 Trial view structure.….......................................................................................................... 603.8 Trial1 view structure.…........................................................................................................ 613.9 Avidresults1 view structure.................................................................................................. 623.10 Storage1 view structure........................................................................................................ 633.11 Storage2 view structure........................................................................................................ 63

List of Figures

1.1 GLEWS Modus Operandi .................................................................................................... 181.2 SOI against RVF activity .................................................................................................... 251.3 Dr. John Snow's map of the Cholera epidemic..................................................................... 321.4 Health Care Coverage Status................................................................................................. 341.5 Smoking Survey ..…...…...................................................................................................... 342.1 The three-tier architecture model.......................................................................................... 362.2 Market Share for Top Servers.…........................................................................................... 462.3 Process Model for the extended application......................................................................... 493.1 AVID Search form …..…..................................................................................................... 643.2 AVID Search results.....…..................................................................................................... 683.3 Points on map ….........…..................................................................................................... 723.4 Overlay choices…............................................................................................................... 733.5 Single overlay ….........….................................................................................................... 743.6. Combined overlays.…......................................................................................................... 753.7a Clear map …...….................................................................................................................. 773.7b Animation begins........…...................................................................................................... 773.7c Animation progressing.…..................................................................................................... 773.8 Login Screen..…................................................................................................................... 783.9 Populated search form ......................................................................................................... 783.10 Search results........................................................................................................................ 793.11a Marker clicked. ..........…...................................................................................................... 793.11b Sample storage details.......................................................................................................... 79


Abstract

Geographical information systems have become ubiquitous in managing spatial

information in a variety of fields ranging from navigation to urban planning. These systems

provide a way to represent and manipulate spatial data in a useful way to complement

decision making and often form part of complex decision support systems. This project

applies geographical information systems to map incidences of Rift Valley Fever and other

diseases over multiple datasets to allow for the observation of trends. Using Google Maps,

incidences are plotted and the map views combined with other spatial data via KML

overlays to create a framework to evaluate relationships between environmental factors

and disease occurrence.

At present, there is a need to provide a visualization tool for the geographical element of

the large datasets currently being collected by the Biosciences East and Central Africa,

(BECA), project known as AVID. The project is currently working on disease prediction

mechanisms for Arthropod Borne viruses in East and Central Africa. This involves

collecting biological samples, several times a year, from both human and animal sources.

The data collected includes geographical information in the form of the coordinates,

latitude and longitude, of the point of collection. The date and time is also recorded,

forming a time stamp for the geographical data.

The aim of this project is to extend the functionality of previous work carried out in

establishing an information management system. This previous work resulted in a web

based application which provides a basic information system to manage and query the vast

amounts of data being collected. The system provided a database structure to hold key data

items and establish relationships between them in an efficient manner for formulating

queries. This extension to the current system provides a visual representation of the data

using Google maps and other open source geographical information system tools.


Declaration

No portion of the work referred to in the dissertation has been submitted in support of an

application for another degree or qualification of this or any other university or other

institute of learning.


Copyright

Copyright in text of this dissertation rests with the author. Copies (by any

process) either in full, or of extracts, may be made only in accordance with

instructions given by the author. Details may be obtained from the appropriate

Graduate Office. This page must form part of any such copies made. Further

copies (by any process) of copies made in accordance with such instructions may

not be made without the permission (in writing) of the author.

The ownership of any intellectual property rights which may be described in this

dissertation is vested in the University of Manchester, subject to any prior agreement to

the contrary, and may not be made available for use by third parties without the written

permission of the University, which will prescribe the terms and conditions of any such

agreement.

Further information on the conditions under which disclosures and exploitation may take

place is available from the Head of the School of Computer Science.


Acknowledgements

I would like to firstly acknowledge Almighty God for granting me the focus and strength to

complete this project. I would like to thank my family and friends for their support and

encouragement throughout. Special thanks go to my supervisor, Dr. Richard Banach, for

his guidance, direction and faith in my abilities to complete this project.

I would also like to thank the team at AVID for their assistance in this project and last and

definitely not least I would like to thank everyone at the University of Manchester who

provided me any assistance in any small way during my studies here.


Chapter 1 Introduction and Background

1.1 Introduction: Emerging Infectious Diseases (EIDs)

Emerging infectious diseases are a constant threat to public health. With every new disease

strain or re-emergence of an old incurable foe, public health infrastructure comes under

increasing strain to deliver appropriate responses. In particular, as population densities

increase with longer life expectancies, ironically, epidemics spread faster and claim more

victims. To understand some of the complexities of battling these new diseases, let us

explore them through a series of questions.

1.2 What are emerging infectious diseases and why are they important?

An Emerging infectious disease is defined as, “An infectious disease that has newly

appeared in a population or that has been known for some time but is rapidly increasing in

incidence or geographic range” [1]. Examples of these that have appeared over the last four

decades include the Ebola virus, HIV / AIDS and Hepatitis C [2]. Quite often these

diseases will re-emerge with no known pattern at seemingly random intervals. As such,

emerging infectious diseases, EIDs, represent an ever changing frontier for modern

medicine. They often render existing treatments ineffective or redundant and are usually

more resilient than variants that may have been seen previously. They threaten human and

animal life and since they often have no immediate cure, have the potential to take

significant toll in this regard.

Awareness of these diseases is increasing as modern ways of life make the transmission of

these diseases far quicker than in decades past. Previous outbreaks that were largely

localized now “benefit” from the world's developed transportation infrastructure.

Commercial passenger flights mean that a person infected in South East Asia can be in

North America in 24 hours, before showing any signs of illness while spreading pathogens

all along the route. This type of scenario means that the disease can spread much farther

and quicker than the means to contain it initially. If such a disease is fatal, the potential for

disaster is evident.


Emerging infectious diseases cause significant health and economic damage. The fact that

some of them appear only periodically, or seemingly at random, makes treatments and

studies difficult. The AVID focus on Rift Valley Fever provides a good model because of

the comparatively large amount of historical data available. According to the Centre for

Disease Control, CDC, Rift Valley Fever is an “ … acute, fever-causing viral disease that

affects domestic animals (such as cattle, buffalo, sheep, goats, and camels) and humans”

[3]. Rift Valley Fever belongs to a group of diseases referred to as Zoonoses. A Zoonosis is

“An infection or an infectious disease transmissible under natural conditions from

vertebrate animal and humans” [4]. These are discussed in more detail later in this chapter.

Why Zoonoses? Zoonoses and their impact on the world

According to the World Health Organization, 75% of new diseases affecting human beings

in the last decade have been zoonotic in nature, i.e. they originated from pathogens found

in animals or animal products [5]. There are various reasons why the percentage is so large.

Some of these include environmental changes, human and animal demography, pathogen

changes and changes in farming practice [6]. Human population increases sometimes

results in increased interaction of domestic animals and wildlife as human beings encroach

on what were previously uninhabited areas and encounter naturally occurring zoonoses.

These are diseases common in wildlife populations but that would normally not occur in

domesticated animals because they would never come into contact with one another.

Examples of recently emerging zoonoses include Avian influenza, more commonly known

as “bird flu” and Swine influenza, also known as “swine flu”. More traditional zoonoses

include anthrax, rabies and the West Nile Virus. Some of these diseases have been isolated

and treated due to their causes being understood and studied long enough to develop

effective cures. Others remain endemic and form part of the justification for projects like

AVID. In order to understand the impact of these zoonoses, we need to realize that animal

health impacts human health directly with animals providing the bulk of human food and

livelihood particularly in the developing countries. To illustrate this, let us look briefly at

two of the recently emerged diseases; avian influenza and swine influenza.


1.2.1 Avian influenza. Origins and effects.

Avian influenza is defined as, “... an infectious disease of birds caused by type A strains of

the influenza virus.” [7]. It's effects range from mild illness to a fatal condition in birds.

Normally, this disease will not infect humans, however, a virulent strain called Highly

Pathogenic Avian Influenza, (HPAI), is known to affect human beings. It is identified as

H5N1 avian influenza and has spread from infected poultry to human beings, causing

deaths. It is highly virulent and mutative and differs significantly from other strains. The

most recent outbreak occurred in 2003. The main difference with the latest outbreak was

that the prior outbreaks remained largely localized. All major outbreaks though still

resulted in high poultry losses.

The WHO maintains figures of reported cases, including deaths related since the 2003

outbreak. This currently stands at 504 reported cases with deaths standing at 299 as at 12th

August 2010. See table 1.1 below.

Table 1.1 Confirmed cases of HPAI. Source: W.H.O. [8].

Country

2003 2004 2005 2006 2007 2008 2009 2010 Total

cases deaths cases deaths cases deaths cases deaths cases deaths cases deaths cases deaths cases deaths cases deaths

Azerbaijan 0 0 0 0 0 0 8 5 0 0 0 0 0 0 0 0 8 5

Bangladesh 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0

Cambodia 0 0 0 0 4 4 2 2 1 1 1 0 1 0 1 1 10 8

China 1 1 0 0 8 5 13 8 5 3 4 4 7 4 1 1 39 26

Djibouti 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0

Egypt 0 0 0 0 0 0 18 10 25 9 8 4 39 4 21 8 111 35

Indonesia 0 0 0 0 20 13 55 45 42 37 24 20 21 19 6 5 168 139

Iraq 0 0 0 0 0 0 3 2 0 0 0 0 0 0 0 0 3 2

LaoPeople'sDemocraticRepublic

0 0 0 0 0 0 0 0 2 2 0 0 0 0 0 0 2 2

Myanmar 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0

Nigeria 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 1 1

Pakistan 0 0 0 0 0 0 0 0 3 1 0 0 0 0 0 0 3 1

Thailand 0 0 17 12 5 2 3 3 0 0 0 0 0 0 0 0 25 17

Turkey 0 0 0 0 0 0 12 4 0 0 0 0 0 0 0 0 12 4

Vietnam 3 3 29 20 61 19 0 0 8 5 6 5 5 5 7 2 119 59

Total 4 4 46 32 98 43 115 79 88 59 44 33 73 32 36 17 504 299


The primary concern with H5N1 avian influenza is the possibility for mutation into a form

that will allow for pandemic human to human transmission. This has led to aggressive

preventive control measures being enacted.

Control measures

Since the primary method of transmission was through contact with infected poultry, this

became the main focus of the control efforts. To curb the spread of H5N1 avian influenza,

large numbers of poultry were culled in the affected areas. The numbers varied

significantly but in all cases, the local poultry industry was significantly affected.

Quarantines were effected and bans on poultry product exports were implemented against

those countries where outbreaks had occurred. Estimates of losses in poultry for the 2003

outbreaks according to the FAO, are put at 44 million birds in Vietnam, (approximately

17.5% of the poultry population), and 29 million in Thailand, (approximately 14.5% of the

poultry population) [9]. These losses directly affected the economies of those countries

with Thailand losing it's position as the world's 5th largest exporter of fresh poultry.

Furthermore, the long term effects on the livelihoods of the small scale producers are yet to

be taken into account. Most of those will have taken the heaviest losses, relative to their

income.

World Bank estimates on the potential economic impact of avian influenza, should it

mutate into a human pandemic, on Gross Development Product, (GDP), range between a

drop of 0.7%, (mild), to 4.8% , (severe), across the world with the most affected areas

being Europe and Central Asia where the drop is estimated to range between 2.1%, (mild),

to 9.9%, (severe), at 2006 figures [10]. Further detailed estimates translate this into a

monetary value of US$ 965.4 billion, with the bulk of that figure, US$ 774 billion, coming

from high income countries. Latin America and the Caribbean would incur the heaviest

losses by percentage, incurring an estimated loss of 4.4% in their GDP [11].

It is against this potential scenario that the research projects into the behaviour of emerging

infectious diseases, like AVID, are being undertaken. Let us consider another example, the

H1N1 pandemic, more commonly referred to as swine flu.


1.2.2 Swine Flu. Origins and impact.

Swine flu refers to the pandemic H1N1 virus that is defined by the WHO as a previously

unidentified cause of infection prior to the current H1N1 pandemic, with its origins being

traced back to animal influenza viruses [12]. This virus is different from the seasonal

human H1N1 influenza virus but is transmitted in the same way and therefore spreads

quickly among human populations. Outbreaks in North America in 2009 quickly spread to

several countries with a worldwide pandemic being declared in June of that year. To date,

the disease has been reported virtually worldwide with greatly varying numbers of

infections and fatalities, see appendix A.

The virus causes severe illness and death and as of 6th August 2010 has been responsible

for at least 18,449 human fatalities according to the WHO, see table below, with

continuing cases being reported in parts of India and New Zealand. These figures are

considered a severe underestimation of actual numbers because of the fact that many

deaths will not be tested or diagnosed for influenza and therefore not linked. Considering

the fact that there was no effective vaccine available for the disease at its discovery, its

high rate of fatalities and the speed at which it propagates through human populations, it is

clear that it poses a significant threat to public health and development.

Various treatments and a vaccine are now available and the logistics of producing and

delivering these treatments to the affected areas will be the focus of mitigation efforts.

Without research into the behaviour and cause of the disease, the relatively quick

availability of such treatments would have been impossible. The impact of the disease

beyond the immediate loss of life is yet to be studied.


Table 1.2 Confirmed cases of H1N1. Source: WHO [13]

The human cost of swine flu is far more than that of avian influenza. While it is technically

not a zoonosis, it is an emerging infectious disease and a prime example of what can

happen during an epidemic. The study of the behaviour of such diseases is the premise of

the AVID project.

1.3 What is AVID?

The Arbovirus Incidence and Disease (AVID) project is a three year project coordinated by

icipe - African Insect Science for Food and Health, to undertake a project titled “An

integrated response system for emerging infectious diseases in East Africa”. It brings

together an array of partners from the fields of veterinary health, wildlife services, public

health, agricultural research, medical research and livestock ministries in the target area.

Specifics of AVID research. What is AVID looking at?

AVID is an initiative that is focused on increasing the understanding of the emergence / re-

emergence of such diseases by developing detection platforms for Arthropod Borne

Vectors in Kenya. By studying the model of Rift Valley Fever epidemics, it is hoped that

insights into how these types of disease re-emerge will be gained and the factors

influencing their prevalence shall be documented.


The stated objective of the project is to “ … demonstrate the feasibility of developing a

multidisciplinary surveillance, research and response system to enhance the prediction and

prevention of emerging infectious diseases, particularly arboviruses, using Rift Valley

Fever virus (RVF) as an initial model” [14]. Lessons learned and methods derived can then

be applied to other scenarios.

1.3.1 Rift Valley Fever as a case study

Rift Valley Fever has been recorded for approximately a century, with the first incidences

reported in 1900s by veterinary officers in Kenya [15]. The virus was identified in 1937

and is mostly associated with periods of excessive rainfall, also known as the El

Niño/Southern Oscillation (ENSO) weather phenomenon and has been documented several

times in the last century with the outbreaks spreading beyond Africa in the recent past [16].

A detailed listing of recent outbreaks is found later in this chapter.

The effects of Rift Valley Fever can be devastating. An outbreak in 1950 - 1951 led to the

loss of an estimated 100,000 head of sheep [17]. In human terms, Rift Valley Fever has led

to fatalities in less than 1% of those infected and occurs in two forms; a mild form with

feverish symptoms that pass in between four and seven days from the onset of symptoms

and a severe form with symptoms that range from haemorrhagic fever to

meningoencephalitis. The impact of the outbreaks on the livestock sector of the economy

of the affected country is usually significant. Besides the loss of animals and livelihood for

the cattle owners, bans on exports usually follow. The 2007 outbreak of Rift Valley Fever

led to Kenya's loss of a beef export quota to the European Union of 4,000 metric tonnes

[18]. The reason given was the failure to control animal diseases with the quota being

awarded to a competing country, Botswana.

The unpredictability of Rift Valley Fever outbreaks has long been a hindrance to effective

management. Despite the fact that it is associated with the ENSO phenomenon, the factors

surrounding its emergence are still not fully comprehended. The infrequent nature of the

outbreaks and the comparative localization of the virus has meant that inadequate resources

have been spent in developing effective countermeasures. Only in more recent outbreaks


has the virus began to emerge outside its traditional confines, representing new challenges.

Appendix B contains a map of the countries in the world at risk of Rift Valley Fever

outbreaks as well as those where actual cases have been reported. Understanding how the

virus sustains itself between outbreaks and how it is influenced by environmental factors

leading up to an outbreak is at the core of modern Rift Valley Fever research. Further

details on the nature of the disease and the research being carried out by AVID are

contained in the section on disease monitoring further in this chapter.

Understanding the way in which Rift Valley Fever sustains itself in the periods between

epidemics as well as why similar environmental factors do not always trigger outbreaks

will make the prevention and managing of outbreaks much more of an exact science. It will

also allow for the fine tuning of practices that can be disseminated to cattle rearers to

mitigate the spread of the disease in the event of an epizootic. In order to prevent an

outbreak, it has to be known that one is expected and where. There are ongoing initiatives

in this regard with the biggest one being coordinated by the world's three major agencies

concerned with human and animal health; the W.H.O., the F.A.O. and the O.I.E.

1.3.2 Prevention is better than cure (and ultimately less expensive)

The World Health Organization, (WHO), the Food and Agriculture Organization of the

United Nations, (FAO), and the World Organization for Animal Health,(OIE), have

developed the Global Early Warning System for Major Animal Diseases, including

Zoonoses, (GLEWS), monitoring system. GLEWS lists its objective as, “... to improve the

early warning and response capacity to animal disease threats of the three sister

organizations for the benefit of the international community” [19].

GLEWS allows the three organizations to channel information from a variety of sources

into a shared space and to coordinate their responses in a way that uses their disparate

resources and strengths to complement one other. GLEWS currently prioritizes a total of

25 diseases, both zoonotic in nature and otherwise, including Rift Valley Fever. GLEWS

focuses on early warning and response capabilities for the international community with

regards to infectious disease outbreaks and is charged with the dissemination of


information as deemed relevant to any country which may be currently or potentially

affected. The coordination of the three bodies allows for a mining of a larger pool of

information, including informal networks and other reporting resources, to deliver

information on incidences of disease.

GLEWS is also placing emphasis on trend analyses and forecasting on four diseases,

namely Contagious Bovine Pleuropneumonia, (CBPP), Foot and Mouth Disease, (FMD),

Rift Valley Fever and Sheep Pox. This is key to fulfilling the early warning mandate and

shows just how important an activity it is. Early warning allows authorities to prepare

medical resources and take preventive action where possible to reduce the impact or even

occurrence of an outbreak, avoiding potentially heavy human and animal losses.

The establishment of GLEWS is a natural progression of the separate early warning and

response systems that existed previously among the three organizations, particularly as

disease outbreaks often do not respect any geographic or political boundaries and so often

require cross border coordination in response to them. The scale of such tasks becomes

much easier when efforts can be unified to make use of the combined resources. Figure 1.1

below shows GLEWS operational model.


The relatively long period that the Rift Valley Fever virus has been known to exist and the

amount of data collected has allowed for the development of vaccines and more recently,

predictive models. These predictive models are of immense interest in providing ample

warning time for the outbreak of infectious disease. There are clear humanitarian reasons

for conducting this research as well as economic ones. We shall investigate these predictive

models in detail in the following section covering disease monitoring and outbreak

prediction.


Figure 1.1 GLEWS Modus Operandi Source: GLEWS [20]

1.4 Disease monitoring and outbreak prediction

Disease monitoring can be defined as the observance and recording of disease incidence

and prevalence in a population of any kind within a specific environment. This activity is

not carried out merely to create a historical record but to provide information that could be

used to combat the disease in the future.

We shall define outbreak prediction in this context as the process of evaluating

environmental and biological factors with a view to forecast incidences disease in the near

future. This process relies on accurate current and historical data to predict an outcome

with a high degree of confidence, based on trends observed in the past and any correlations

that have been established as a result of the investigation of those trends.

The AVID project seeks to perform both of these activities by performing disease

monitoring in East and Central Africa and is targeting Arthropod Borne Viruses. In

particular, the Rift Valley Fever, (RVF), disease is being monitored with samples being

collected from various subjects at regular intervals and epidemiological tests being carried

out to determine prevalence. The amount of data captured in these samples is vast, though

most of the parameters do not reference the spatial domain and as such the subset required

for geographical representation remains small.

1.4.1 Arthropod Borne Viruses

Arthropod Borne Viruses or arboviruses are, as the name suggests, a class of viruses

transmitted to humans by arthropods [21] i.e. the arthropod is the vector. An arthropod is

an invertebrate animal characterised by jointed legs and a hard outer exoskeleton and

includes some insects [22]. A vector is defined as any organism that transmits a pathogen

from an infected host to an uninfected one [23] and in this context it is an arthropod.

Several species of mosquito are known vectors of Rift Valley Fever as well as other

bloodsucking insects such as ticks.


Arboviruses fall into three main families:-

• Togaviruses – examples include western equine encephalomyelitis (WEE) and

Eastern equine encephalitis virus (EEE).

• Bunyaviruses – examples include Rift Valley Fever (RVF) and Crimean-Congo

Haemorrhagic Fever.

• Flaviviruses – examples include Yellow fever and St. Louis Encephalitis (SLE).

The differences between them are beyond the scope of this project but are useful in

separating from amongst them the types which may not affect human beings. Rift Valley

fever affects both animals and humans and is the subject of much research as the reasons

behind its periodical re-emergence are still unclear.

The AVID project is carrying out research to this end as well as other arbovirus research

and is collecting samples from various organisms. Various epidemiological tests are

performed on the samples to determine the presence and nature of any pathogens. The

primary detection test for the arbovirus in question is serology.

Arbovirus Transmission

Arboviruses have two defined transmission cycles. The WHO definition of an arbovirus

describes the transmission by stating, in part, “... through biological transmission between

susceptible vertebrate hosts by haematophagous arthropods ...” [24]. The role of the

arthropod plays as the disease vector is the key difference in the two cycles.

1. Man – arthropod – man cycle: in this instance, the arthropod will transmit the virus

from an infected person, the host, to another uninfected person. This is the mode of

transmission for arboviruses such as dengue fever. The reservoir is the infected

person although it can also be the arthropod itself in some cases.

2. Animal – arthropod – man cycle: in this case, the reservoir is in an animal. The

virus is transmitted by the arthropod in between animals and also from animal to

human. This is the case with Rift Valley Fever.


Some arboviruses, such as yellow fever, can be transmitted through both cycles. Rift Valley

Fever may also be transmitted through contact with the blood of infected animals and

through the air in a laboratory situation [25]. These other methods do not involve

arthropods and so we shall not detail them any further.

The fact that Rift Valley Fever is transmitted via the animal – arthropod – man cycle means

that there are several possible reservoirs of the disease to be monitored in addition to the

vector. As the disease primarily affects domestic animals and humans, there are a large

variety of samples to be collected.

Let us look at the process of sample collection and its integration with GIS and other

technologies to enable outbreak prediction.

Sample Collection

The AVID project collects samples from various field locations from a variety of subjects.

The samples are tested for the presence of arboviruses and the results processed. Sample

collection is a constant process with subjects being sampled regularly as part of the

monitoring aspect of the project. The screening of the samples for arboviruses provides a

picture of the prevalence of the disease even when it may not be causing immediate harm

to the host.

The AVID project collects samples from various animal sources, ranging from cows and

sheep to wild animals such as warthog [26]. Humans are also sampled as are the arthropods

themselves, in this case the mosquitoes. This is necessary to monitor the presence of the

virus in all the stages of the cycle. Various methods are employed to gain sufficient

numbers of vectors with traps being used to collect mosquitoes. All the samples collected

in the field will have their location recorded. This adds the spatial component to the data

instance. The date is recorded as well providing a time reference as well. An overview of

arboviruses follows, together with the most prevalent methods of virus transmission.


Environmental Factors

The Rift Valley Fever virus occurs naturally in the eggs of the Aedes mosquito [27]. The

mosquito larvae hatch and are infected with the virus. As adults, they go on to infect the

animals they feed on. Changes in the environment, flooding in particular, allows for large

increases in vector population. Mosquito eggs that have lain dormant will have an

opportunity to hatch. It has been observed that excessive rainfall that causes localized

flooding, often precedes widespread incidences of the disease and this is attributed to the

increased population of the vector.

As a result, weather prediction has become a key component in the larger process of

outbreak prediction. The monitoring of environmental factors in general is considered to be

a key part of arbovirus management because of the effect of these factors on vector

population. Remote sensing plays a big role in the prediction of weather as the land and sea

temperatures are key indicators in predicting the expected rainfall. The advance warning of

environmental conditions that may favour a significant increase in vector populations can

allow for preventive action to take place and for better preparations to be made to manage

the anticipated increase in disease prevalence.

Previous outbreaks

Outbreaks of Rift Valley Fever have been documented over several decades with the first

known instances being reported in the early 1900s. The virus itself was identified in 1931

and cases have been reported in sub-Saharan and North Africa as well as in Saudi Arabia

and Yemen. The major outbreaks, (also known as epizootics), of Rift Valley Fever recorded

in the second half of the twentieth century up to present day are as follows [28] [29]:

1. 1950 – 1951, Kenya.

2. 1967 – 1970, Nigeria.

3. 1969, Central African Republic.

4. 1976 – 1977, Sudan.

5. 1977 – 1978, Egypt.

6. 1987, Mauritania.

7. 1993, Egypt and Senegal.


8. 1997 - 1998, Kenya and Somalia.

9. 2000 – 2001, Saudi Arabia and Yemen.

10. 2006 – 2007, Kenya, Tanzania and Somalia.

11. 2008 – Madagascar.

12. 2010, Republic of South Africa.

In addition to any human losses, the economic costs of Rift Valley Fever are quite

significant with large numbers of animals lost in the outbreaks. In the outbreak of 1950 –

1951 in Kenya, an estimated 100,000 animals were lost. Apart from the direct impact on

animals, there is a knock on effect on the farming industry of the affected country as well

as the meat exports which can be suspended for the duration. All of these result in a strong

motivation for the management and mitigation of Rift Valley Fever.

The documentation of the past outbreaks provide useful statistical data for researchers to

study. Environmental data in particular has proven to be most useful, allowing researchers

to study the conditions in the months preceding an epizootic. The analysis of weather

conditions prior to and during an epizootic have been of great interest and has led to the

association of certain environmental indicators, primarily above average rainfall and

localized flooding, with Rift Valley Fever [30].

The understanding of the conditions that precede an epizootic are critical to the forecasting

process. The retrospective application of new techniques allows for trends to be observed

and correlations drawn that can then be used to inform the establishment early warning

systems. The continued data collection of the AVID project and other Rift Valley Fever

monitoring activities like it, are essential in providing this information.

Arbovirus detection

Technology plays a key role in the detection of arboviruses. The continued improvements

in computing power and new techniques in the field of medicine have led to better

detection and treatment of arboviruses. Serology is the method or process used to detect

Rift Valley Fever in a serum sample. Serology is defined as “ … the scientific study or

diagnostic examination of blood serum, especially with regard to the response of the


immune system to pathogens or introduced substances” [31]. Serological tests are used to

detect arboviruses and the samples collected in the AVID project are processed via the

Enzyme-linked immunosorbent assay, (ELISA), test. This test detects the presence of

particular antigens or antibodies [32].

The outcome of this test on a sample gives an indication of the presence (or absence) of

Rift Valley Fever and other pathogens depending on other specifics that may be employed

during testing. The volume of samples collected means that results take some time to be

collated, however this period appears consistent with the resources available to the project.

Serology thus remains a reliable method for determining the presence of Rift Valley Fever

and it is anticipated that improvements in technology will continue to increase the speed

and volume of processing.

1.4.2 Predictive modelling

In 1999, a report was published that detailed a method for accurately forecasting outbreaks

of Rift Valley Fever [33]. This method made extensive use of remote sensing techniques

and when applied retrospectively, was able to detect previous outbreaks with great

accuracy, based on the data provided by satellites. This method utilized the Southern

Oscillation Index, (SOI), commonly used in the monitoring of the El Nino –Southern

Oscillation , (ENSO), that is associated with above average rainfall patterns in East Africa

[34].

This technological breakthrough came about as a result of applying new technologies to

the large body of data available on Rift Valley Fever and the corresponding weather

information. All of this data was available through unrelated information gathering efforts

but none the less proved instrumental in the development of the model as without it, there

would be no possibility of generating proofs.

The figure 1.2 below shows the apparent relationship between SOI anomalies and Rift

Valley Fever outbreaks. It should be noted that not all negative SOI anomalies resulted in

an outbreak, however, all outbreaks were preceded by negative SOI anomalies suggesting

that SOI has a role to play in conjunction with other factors.


Also investigated were the equatorial Pacific Sea Surface Temperatures (SSTs), Indian

Ocean Sea Surface Temperatures, and Normalized Difference Vegetation Index, (NDVI),

collected from remote sensing satellites. These factors were input to a computer simulation

with the aim of determining which combinations would assist in the reliable prediction of a

Rift Valley Fever epizootic. The report suggested the most promising factors based on their

ability to accurately predict precious outbreaks based on historical data as well as avoiding

false predictions when similar weather conditions were observed. The report proposed the

set of factors that were found to have the most accurate detection rate; equatorial Pacific

and Indian Ocean SST and NDVI anomaly data.


Figure 1.2: SOI against RVF activity, a time series plot of SOI anomalies between January 1950 and May

1998 [35]

Advances in computing power since the report was published, as well as increased access

to satellite data means that prediction of this nature can now take place in near real time,

allowing for stakeholders to take preventive action. The successful development of early

warnings systems using satellite imagery and weather forecasting data can be used to

support this [36]. The aversion of an epizootic and the associated human, animal and

economic losses would greatly reduce the impact of Rift Valley Fever.

Geographical Information Systems are playing a key role in the forecasting process by

making spatial data available to a wider set of stakeholders. The satellite imagery used in

the 1999 report was taken from satellites operated by the National Oceanic and

Atmospheric Administration (NOAA) satellites using the advanced very high resolution

radiometer (AVHRR) instrument. It is noted that such imagery and associated data may not

be easily accessible to all stakeholders, whether due to cost or other factors, and yet it is a

key part of accurate disease forecasting. It becomes clear therefore that any other GIS that

can provide complimentary functions to the sampling and monitoring aspects of the AVID

project and others like it will be useful in enhancing the understanding of the behaviour of

arboviruses like Rift Valley Fever.

The development of support tools that are more accessible would go a long way in

assisting projects like AVID in their research. The ideal GIS tool would be easy to use,

widely compatible with a range of technology platforms and be easily maintained by

trained staff. This tool would need to incorporate the relevant GIS capabilities necessary

for the work of forecasting and trend analyses. In the next section we expound on the field

of GIS and the technologies that make it such a powerful observation tool.


1.5 Geographical Information Systems

Spatial or location data is increasingly being used to provide perspective in performing

data analyses. Historically, spatial data has been used only in the context of providing the

boundaries within which sample collection takes place and nothing more. This is now

changing as newer technology makes it easier to process spatial data. Geographical

information systems allow spatial data to be integrated into sample collection by

processing the data and turning it into useful perspectives. Providing a geographical

context to the data that is being collected allows the analyst to view it against other

demographic data in a simplified, visual manner that can provide the basis for further study

or drawing correlations.

1.5.1 What is a Geographical Information System, (GIS)?

“A geographic information system (GIS) integrates hardware, software, and data for

capturing, managing, analysing, and displaying all forms of geographically referenced

information” [37]. Historically, the basic form of a GIS was a series of maps. With newer

technology, it is possible to create digital maps of the same quality as paper maps and even

higher. Digital maps have the benefit of being manipulated easily to represent different

viewpoints, perspectives or locales.

“Cartography (in Greek chartis = map and graphein = write) is the study and practice of

making maps (also can be called mapping). Combining science, aesthetics, and technique,

cartography builds on the premise that reality can be modelled in ways that communicate

spatial information effectively” [38]. Traditional cartography allows for very rich, detailed

maps to be produced using modern printing techniques. A variety of materials may be used,

resulting in rugged, durable maps. These are practical and portable and very useful in field

operations, especially in areas where other types of maps will be non-existent. They can

also be more accurate since they are likely to be vetted for accuracy by third parties before

publishing and may be produced to higher scales than other types of maps that may be

commercially available.


Traditional cartography involves working with physical media. The skills of cartographic

artists in collecting, classifying, indexing and finally portraying these geographic features

on the physical media available to them are very specialized and take years of training to

acquire [39]. These advanced skill sets and the principles behind them have a long history

and are no less important in modern cartography. It is however important to note that the

field has evolved, primarily in the tools and media being used. Maps are no longer

confined to physical media such as paper or canvas. Data collection has moved forward

and incorporated many technologies. Aerial photography has provided much of the raw

data for the basis of modern cartography ever since the advent of powered flight in the

beginning of the twentieth century. Advances in camera technology and imaging also

allowed for better resolution photographs and better tolerance to varying light conditions.

Wartime events in the first half of the twentieth century were to bring the most significant

improvements to the field of cartography at the time with the need for up to date

assessment of battle damage and troop movements resulting in the application of new

technologies to the field of imaging, (both still photography and video) [40]. The quality of

picture and resolution led to the provision of high quality raw material for cartographers to

compile detailed maps of any surveyed area [41].

In the second half of the twentieth century, satellite technologies made an appearance,

redefining the work of the modern cartographer. While the principles of classifying and

representing geographical data remained the same, it was now possible to have images that

were not months or days but just hours old. Truly up to date imagery of the globe became a

reality. The technology available allowed for the photographing of any part of the globe

and the transmission of those images back to the ground instantaneously or within minutes.

Access to high resolution imagery from these satellites has been expensive. However, the

advent of consumer grade devices and applications for street navigation and outdoor

recreational pastimes such as hiking and trekking have opened new avenues for access to

this geographical data.


Technological advances in electronics has allowed portable devices to have sufficient

communication and processing power to become a platform for delivering mapping and

location information. Small, light, energy efficient displays have been developed to

provide rich colour screen output on hand held of vehicle mounted devices that can display

maps as well as the user's location in real time. Improvements in battery technology means

that these devices can be truly portable, working for several hours or days on small battery

packs. These devices rely on global positioning systems, GPS, as well as calculations of

user orientation and movement from electronic sensors, gyroscopes and compass. GPS is

reliant on a signal from orbiting satellites and can vary in accuracy by a few metres with

modern systems being accurate to within a few feet [42].

Digital photography has also improved due to the same technological advances. Advances

in optics and electronics means that it is possible to take extremely high resolution pictures

with equipment that is far less expensive than was possible before and without the need for

a separate process to convert the images from their physical form to a digital copy. This

shortens the process of image acquisition and allows cartographers to produce maps much

more quickly. Digital cameras can be controlled to take pictures that are geographically

tagged as they are taken so that the image is referenced to an exact geographical point. This

allows the alignment of aerial photos to take place with better accuracy and is cheaper than

using satellite platforms to take high resolution photos of smaller areas. This type of digital

photography is used by cartographers to produce high scale maps of areas for both digital

and paper media.

In the last two decades, the internet has expanded massively as an information sharing

platform. Since its origins in the 1970s to the widespread availability in the 1990s and the

continued propagation in the present day, the ability to collaborate and share information

across disparate geographic locations has impacted all areas of life. Cartography is no

exception. The ability to share geographical information without committing it to a

physical medium and sending it to the recipient has allowed for new forms of collaboration

in assembling geographical data.


The propagation of the internet and the advent of consumer digital photography has led to

a phenomenon known as geotagging. Geotagging is defined as “… the process of adding

geographical identification metadata to various media such as photographs, video,

websites, or RSS feeds and is a form of geospatial metadata. These data usually consist of

latitude and longitude coordinates, though they can also include altitude, bearing, distance,

accuracy data, and place names”, (Miller, 2009) [43]. We can also use the term social

mapping to refer to this practice. This is because the spatial features that are being captured

are usually collected by the general public and not specialized, trained cartographers. The

collaboration afforded by the internet allows for this spatial data to be combined into

digital map interfaces to produce maps with additional information that would not be

possible by other means or that would be very costly to produce in such fashion. An

example of this is the ability of users to suggest landmarks and building names that may be

new or overlooked in previous maps by forwarding their modifications via mapping

websites. This is very useful for modern cartographers because they provide an alternative

source of updated content for the large maps that they have to maintain. In addition, it

allows for custom maps to be produced by a user for their personal use, without the

cartographer having to go through the entire process of producing a custom map for a

single user. Moreover, this can be maintained digitally separate such that it does not

interfere with other users views of the same map i.e. each user can have their own

customised digital map without affecting any others who are using the mapping resource.

This custom map can then be shared with anyone who may need to use it and at they can

further modify it, making it truly a social map.

With regards to the area in question, eastern Kenya, detailed paper maps would need to be

acquired or commissioned from government or private agencies. There are private agencies

that can provide digital and paper maps of the area but that will be outside our scope at the

present time. We shall instead make use of web mapping tools that shall be discussed in the

following section. The base application we are integrating with is a web based information

system that holds data on biological samples collected by the AVID project in East Africa.

Location data is collected as part of the data set. This data consists of a set of coordinates

that provides the location at which the data was collected as well as several fields that

relate to the biological sample itself.


We would like to use this information to build a Geographical Information System, GIS,

component to the information system that is able to display these locations on a map and

provide additional functionality to overlay other geographical datasets for comparison and

decision making purposes.

1.5.2 Related Areas

The application of GIS to sample collections is not restricted to the biological research

field. Spatial information can be used to complement many other data sets in fields varying

from urban planning to resource management [44]. In the field of demographics, GIS has

been applied to provide key visual indicators of the effects of government policy

implementation such as the building of new infrastructure, population density, impact of

population growth, human migration as well as environmental factors such as forest cover

and precipitation.

This has led to the existence of complimentary data sets that allow for insightful

comparisons to be made. In implementation of the GIS, we will seek to incorporate several

of these datasets to provide overlays of useful information. The GIS will allow for the

extraction of the location data in the AVID samples to produce a visual representation that

can then be compared to a dataset of population density for example. This could provide

interesting insights into possible correlations between population densities and incidences

of disease in a very visual way. The same can be used to compare a variety of

environmental factors that may be relevant to the research taking place with regard to the

effect they may have on the propagation of arthropod borne viruses.

Similarly, any recurring patterns that would be complex to explain with numbers and

graphs in a purely statistical report, can emerge with great impact when displayed in a

visual manner using a GIS. This can greatly aid in conveying the impact that disease

prevalence has in a particular area. The graphical presentation of the sample collection data

and by extension the disease incidence locations can provide support to theories on the

spread and influences on the occurrences of the arthropod borne viruses in question. This

makes GIS very useful in disease mapping and monitoring.


Disease Mapping

While asset management, monitoring of land use and properties have been the primary

application fields of geographical information systems [45], many other fields have found

it to be an invaluable tool for visual analysis. One of the earliest recorded and most

famous uses of a GIS for disease outbreak was by Dr. John Snow, circa 1854, showing the

occurrences of Cholera in a section of London, (see figure 2.3).


A modern example of similar use of disease mapping with GIS is observed with a study in

Lincoln, Nebraska, which aimed to locate the population requiring particular health

services and those affected by particular health problems, in order to allow the health

services to target their efforts more effectively [47].

In this case, surveys were administered to 1000 respondents, and the data input to a

database. The respondents were selected using census data to determine other factors

relevant to survey such as age, ethnicity and size of household. Locational data was

compiled by recording the addresses of all the interviewees. This database was then

brought into a GIS program, ESRI Arc/Info and geocoding was used to match the

addresses against a street database with the results being plotted on a map. The address

information provided made it possible to produce useful visuals to aid in decision making

as far as providing preventative health care. By looking at behavioural trends in the area

and generating visualizations of the same, it becomes possible to predict the types of health

services that would be required going forward as well as informational campaigns that

would impact on the anticipated health concerns.

A selection of the maps produced in that study are reproduced below and give an indication

of the kind of visuals that are possible with spatial data and GIS.


The base map quality available presently is far improved and can provide much richer

detail. This is a factor that this project takes advantage of in its implementation. Let us now

look at the design of the application that allows us to achieve all of this.


Figure 1.5 Smoking Survey [49]Figure 1.4 Health Care Coverage Status

[48]

Chapter 2 Methodology

In this chapter we shall discuss the design of the application and the tools that were used to

create the application. Modern application development typically requires an integration of

various platforms to achieve the desired result. We shall focus on all those aspects here.

Bearing in mind that we are building upon an existing application we shall start with a

brief review.

2.1 Overview of existing application

The existing application makes use of the established practice of Model-View-Controller

paradigm with its associated advantages. The MVC paradigm separates application

development into three:

• The model, which is the business data.

• The view, which is the presentation logic.

• The controller, which is the application logic.

According to Gamma et al (1995), “MVC consists of three kinds of objects. The Model is

the application object, the View is its screen presentation, and the Controller defines the

way the user interface reacts to user input” [50]. In reality, there can be a little overlap

between the three.

The main advantage of this method is that it allows for a high degree of modularization.

This means that the different aspects of the application can be designed in tandem with

very little conflict. This usually results in very quick application development and very

high specialization at each level.


Web applications typically make use of the three-tier model to implement MVC. This

allows for the deployment of an application to a very wide user base with no configuration

necessary. The client needs only a web browser. The existing application is no different and

applies this by incorporating a three tier model using a website as its front end, a web

server and middle-ware as its middle tier and a database server as its back end or third tier.

This is demonstrated in figure 2.1 below.

In the context of AVID, this type of application model works particularly well as it allows

the field operations continue without any modification: they can collect data in the way

that is most efficient for them so long as they can process it into an appropriate format for

the information system. AVID have deployed a custom data collection software which is

used by field agents to record information. Employing the MVC paradigm means that the

information system will not interfere with the specialized data collection activities.The

existing application consists of a set of web pages that query a relational database

management system, RDBMS, via a web server running a scripting engine. The web pages

allow for data capture and input to the database as well as display of stored data via the

browser interface,using tables.


Figure 2.1: The three-tier architecture model of a web

database application. Source Williams et al [51]

2.1.1 Design considerations

In extending this application, we are to introduce a GIS component to the provide a

visualisation of the stored data. This application, which forms an information management

system in itself, is therefore to be extended to manipulate spatial data in addition to

retrieving and displaying the standard text information that is stored in the database.

The use of a custom data collection software has rendered the data input / capture pages of

the existing application redundant. The extended application will therefore take this into

account during the design revision. Updated user requirements have been received and

have resulted in the following set of objectives with regard to the GIS component:

• Visual representation of sample collection points within the geographical research

area and by extension incidences of any diseases detected.

• Overlay of additional datasets of information that are relating to the geographical

research area as may be made available, such as weather conditions or livestock

population densities.

• Representation of the AVID dataset within the bounds of a time period. In order to

represent events occurring over a timespan of selected weeks or months it should

be possible represent the incidences in an animated form showing the occurrence of

disease over the map with time.

The visual representation of the sample collection points will be the first level of

application for the GIS component. This plots all the sample collection points or a dynamic

subset of the same based on a user query. Each unique type of sample should be

represented by a different style of icon.

Additional overlay information will be provided by the datasets that will be sourced from

other research and data collection institutions. These datasets will cover livestock density,

average rainfall and tsetse fly prevalence.


The animation component of the GIS will be developed to provide a visualization of the

data set over time. This is dependent on the provision of adequate results data to model

trends, failing which it will be possible only to model the feature but not to have any trend

discovery utility.

The tools used have been carefully chosen to provide the most appropriate result within

the constraints present. We will be reusing the set of tools used to develop the existing

application and introducing a new tool for the GIS component: web mapping.

2.2 Web mapping

This is the area of GIS that we are concerned with in this project because we are

combining a web based information system with a GIS application.

Web mapping extends the functionality of traditional GIS by allowing map information to

be displayed in a browser, without the need for the underlying hardware and software to be

present in the same system. Web mapping allows remote systems to access maps over the

internet and have dynamic views built on the fly for display at the user end. This is

achieved by using an Application Programming Interface, (API), that generates rich

content that web browsers can interpret and display.

These APIs have provided a means to access vast amounts of mapping data without having

to invest in the hardware and software to run a full GIS. This greatly simplifies the task of

converting the location data collected in the samples into visual representations. All tasks

involving the data manipulation are performed by a remote GIS installation and only the

results are returned.

This frees the users of the system from the high cost of running a GIS system which would

require systems and mechanisms for satellite imagery and aerial photography, processing

and dissemination as well as other cartographic expertise. While there are cheaper

commercial options to running a complete GIS system, there would still be the cost of

licensing those applications and maps which for most users may be prohibitive.


2.2.1 Evaluation

There exist several applications that provide various levels of GIS functionality that are

required for this project. We shall explore of them.

• Infrastructure for Spatial Information in the European Community (INSPIRE)

• OpenLayers by Open Source Geospatial Foundation.

• ESRI web mapping API.

• Microsoft Bing maps platform.

• Google earth / maps API.

Infrastructure for Spatial Information in the Europ ean Community (INSPIRE)

INSPIRE is a platform that was established in 2007 to provide a common standard for the

representation and sharing of spatial information across the European Union (EU) by

member states. It aims to facilitate the sharing of this information amongst public sector

organizations to assist in policy making and provide better public access to the same [52].

INSPIRE intends to provide access to detailed environmental spatial data to EU by having

member states compile spatial data according to a common set of implementation rules.

Currently the member states have set a target date of the establishment of a portal offering

spatial information at community level of November 2010. Full implementation across

member states is targeted for completion in May 2019 [53].

Being a public sector initiative, it is key to note that the quality of spatial data available is

likely to be quite high, benefiting from access to government resources and data sets.

Though INSPIRE is still in its infancy, it has the possibility to provide very detailed

information services once implemented.


However, being an EU concern, INSPIRE data collection is limited to EU member states

and this it will not have any bearing on the current BECA project as the research location

falls outside of this. Even if the INSPIRE standard can spur worldwide adoption, this will

happen in the long term, way beyond the scope of this project.

The availability of more mature products with a larger user base and better support and

documentation also provide reasons why INSPIRE would be an unsuitable technology at

this time.

OpenLayers by Open Source Geospatial Foundation

OpenLayers is a programming interface that allows a programmer to display a dynamic

map in a web page [54]. It is a product of the Open Source Geospatial Foundation and is

purely open source software.

OpenLayers can display map and markers from any source and can use a number of

industry standards such as Web Mapping Service (WMS), Web Feature Service (WFS),

Time Map Service (TMS), WorldWind, (an SDK developed by NASA for representing

environmental data) and GeoRSS, (a method for geographically tagging RSS feeds). This

is useful because it makes it possible to import other geographical datasets compiled in or

exported from other GIS applications and overlay them on top of a map layer. A map layer

is the equivalent of a paper map represented in a browser window. This is especially useful

in providing additional information. Road networks for example can be represented in this

way. Power lines and other forms of infrastructure as well can be overlaid in the same

manner.

The open source nature of OpenLayers is attractive because it is constantly being evaluated

and improved by third parties. This makes it easier to find and build third party add-ons to

provide functionality that may be absent in other products.

OpenLayers makes use of Javascript which is supported by all desktop browsers, making it

cross platform. The ability to integrate numerous data sources makes it attractive for use in

our intended application.


ESRI web mapping API

Environmental Systems Research Institute, (ESRI), is a company that produces GIS

software for commercial use and is well known for its stand alone products in the field.

The company provides a broad range of products that cater to different industries and

different expertise levels under the branding ArcGIS.

ESRI also makes available a free web mapping API to extend this functionality to the

internet. This gives access to the mapping capabilities of its proprietary software and

allows web designers to include maps in their web applications. The API provides the

ability to overlay selected ESRI mapping resources and allows for the creation of “mash-

ups”. A “mash-up” is the integration of two seemingly unconnected applications, for

example, combining a map and a social networking application to display the locations of

user activity in real time. The API allows for the creation of such combinations.

There are some restrictions on the usage of the maps and functionality provided and the

proprietary nature of the software means that it may not always be possible to customize it.

Some content and functions require payment before usage and deployment may be subject

to some restrictions while using the free API.

However, ESRI brings a rich feature set to the web mapping API as follows:

• support for GeoRSS

• third party map overlays

• compatibility with content created by commercial stand alone ESRI products.

• A selection of reference maps for use with the mapping API

ESRI web mapping API provides an established platform to incorporate dynamic maps

into the web information system. Its proprietary system is built on years of experience in

the field and leverages the stand alone suite of products which are considered industry

standard. On the other hand, the web mapping API is relatively new and does not have a

wide user base despite the desktop and server suite of GIS products holding a commanding

market share, quoted at up to 30% in 2009 [55].


Microsoft Bing maps platform

The Microsoft Bing maps platform is another free web mapping API that seeks to provide

mapping functionality for web pages. It is also a proprietary platform, produced by the

Microsoft Corporation. The Bing maps platform was known previously as Microsoft

Virtual Earth. The Bing platform builds on the expertise of Microsoft Virtual Earth

technologies and is designed to provide tools to layer information on top of Bing maps.

Bing maps simply refer to the map images provided by the underlying GIS.

The Bing maps platform provides an API with the following functionality :

• Overlays.

• Bird's eye and street views.

• Mobile device support.

• GeoRSS feeds.

• Routing and navigation.

• Geocoding i.e. the translation of a geographical address to a set of geographical

coordinates on a map.

The satellite map imagery provided by the Bing maps platform is the most up to date of the

East Africa region out of all the web mapping options that are being evaluated in this

report. Having said that, other supporting data for the region remains unavailable. There is

no geocoding or navigation support for the region and landmark information is minimal.

This means that without the additional supporting information, the map imagery remains

without sufficient context to make it a good candidate for selection. It is anticipated that in

future, upgrades to this supplementary information will remedy this.

Alternatively, a solution that will be able to merge the satellite imagery provided by the

Bing platform with extensive geocoding and navigation support for the East and Central

Africa region, from a third party, would be able to generate a high quality hybrid which

would satisfy all the necessary requirements. This is however out of the current scope of

this project.


Google earth / maps API

Google earth / Google maps is yet another proprietary mapping platform that provides

functionality to display and manipulate map data within a web page. The platform is

provided free of charge by the Google company, subject to terms and conditions of use and

deployment. The Google maps API takes the functionality present in the Google Earth

suite of programs and provides the interface to display similar imagery in web pages.

The Google earth API provides all the functionality found in the Bing mapping platform

and adds facilities for time and animation. These are useful components as they allow us to

make use of the date and time component of the sample collection. The time and animation

functionalities will allow visual representations which are able to take into account when

the events occurred and not just where. This is especially useful when trying to observe the

formation of any patterns or trends.

The satellite map imagery of the East Africa region provided by the Google maps API is

not as up to date as that of the Bing mapping platform. However, the API provides

significantly more underlying infrastructure data, both roads and landmarks, and higher

zoom levels. Furthermore, the mapping data is currently open for user contribution,

increasing the underlying content regularly and map updates will see the imagery improve.

The Google maps API is considered a mature platform, based on the Google Earth product

which itself was known prior as Earthviewer 3D, a product available from Keyhole

Incorporated since 2001 [56].

While all the desired features of the mapping API found in Google earth may not be

implemented in Google maps, virtually all the necessary functions are available. The

Google earth plug-in also requires to be downloaded for the mapping display to work in

the browser and is not supported on all major operating systems as yet. Therefore the

choice of Google maps will provide a better cross platform experience. That the AVID

project is sponsored by Google.org means that this will also be a natural fit since there will

be opportunities for further support and collaboration that may not be possible with other

platforms. The Google maps API has undergone many refinements since it's inception,

with the current API version, version 3, supporting new devices and technology platforms.


2.3 Building blocks

The existing application was built using the following tools,

• PHP

• MySQL Relational Database

• Apache web server

The extended application uses the same tools and adds the following to the tool set.

• Google Maps API

• Keyhole Markup Language

We shall briefly describe each of these tools.

2.3.1 Hypertext Pre Processor (PHP)

PHP, originally named Personal Home Page tools, is a general purpose scripting language

developed in 1995 by Rasmus Lerdorf [57] and currently developed and maintained by The

PHP Group. There is more than one implementation of the language but it is open source

with the most popular implementation produced by the Zend company. It was designed for

creating HTML content although it has since extended beyond that domain.

It is suited for web development purposes due to the fact that it can be embedded within

HTML and executed on the fly by the web server. This is achieved by the web server

calling a command line executable to process the code, with the results being returned to

the server and forwarded to the browser as HTML.

PHP provides methods to access and manipulate data in SQL compliant databases. SQL

stands for Structured Query Language and is a standard through which relational databases

can be accessed and managed. The application makes use of PHP for connecting to the

database and performing searches as well as formatting and displaying the results.


It is an open source tool which is now in its fifth iteration of development with the current

version being PHP 5.3.3. The original application was developed with a previous version

while the extended application makes use of version 5.3.0.

2.3.2 MySQL Relational Database Management System

MySQL is a relational database management software, (RDBMS), that stores and

manipulates data stored within a data set, (relational database). A relational database stores

data within tables as well as the relationship between that data with tables as well. This

allows for various representations of that data to be derived and manipulated without

changing the tables themselves. This is a big advantage and allows for complex

associations to be created that would otherwise not be possible.

The data collected by the AVID project is stored a MySQL database from which it is

managed and manipulated. The MySQL RDBMS provides the means to retrieve all or

subsets of that data based on search parameters structured in query statements. These

statements act as filters on the dataset to allow users to sift through large amounts of data

to find specific information that matches their search criteria.

The current iteration of MySQL is the MySQL Community Server version 5.1.50 with the

version used in our application being version 5.1.36. The MySQL RDBMS is open source

software, multi-platform and is also available in other enterprise class forms with paid

support and other benefits for users that require that level of service. MySQL provides a

fast, optimized data store for the application to utilize.


2.3.3 Apache Web Server

The Apache web server or Apache httpd is an open source web server that has been

developed and is maintained by the Apache HTTP Server Project. It is an HTTP server that

is used for hosting of web pages and provides extensions for all web server tasks that are

required by current standards. These range from secure encrypted connections to virtual

domains. Apache web server supports PHP and this makes it suitable for our application

and provides the framework for our pages to be accessed. It is robust, available on the

major operating system platforms and it is designed to operate with modest computer

hardware requirements. The extended application makes use of Apache version 2.2.11 with

the most recent stable version available being version 2.2.16.

Apache web server currently holds the largest market share for web servers, despite being

an open source product, and has been the most popular web server software since 1996.

This can be seen in figure 2.2 below which shows market share statistics as of August

2010.


2.3.4 Google Maps API

The Google Maps API is a programming interface that allows web pages to query the

mapping service provided by Google incorporated. The API is in its third iteration of

development with the current version being version three often abbreviated as Google

maps v3. Our application is web based and we are making use specifically of the Google

Maps JavaScript API V3 which is compatible with virtually any browser currently

available on any platform.

As mentioned before, the Google maps API provides to following functions: overlays,

bird's eye and street views, mobile device support, GeoRSS feeds, routing and navigation

and geocoding. These are largely dependent on the amount of mapping information,

imagery and road data available for the area of the surface of the earth that is being viewed.

The more underlying data is available, the more features are enabled and accessible for that

particular area.

The API provides for both static and dynamic maps to be integrated into any web page

with the use of Javascript coding and Google's mapping service. In particular, the API

provides the means to display custom spatial data on a map which have been passed to it

from another source, for example, a database. This is a key functionality since it allows us

to generate spatial datasets dynamically and pass them to the API for display. The Google

Maps API provides utilities for map manipulation such as zooming, panning and viewing

of several different map types; road maps, satellite views, topographical, sometimes known

as relief, views as well as a combination of road and satellite.

The mapping service is proprietary although the API is publicly available for developers to

make use of at no charge. The Google maps API is provided in both free and paid

enterprise class flavours. The extended application makes use of the free version.


2.3.5 Keyhole Markup Language

The idea of overlaying complimentary datasets is central to our GIS component. As

discussed earlier, these additional datasets allow for comparisons to be made with the other

factors that they represent.

In order to make these datasets available to our application, they need to be in a compatible

format. The format compatible with our application is known as Keyhole Markup

Language, (KML). According to Udell (2009), “KML is a standard that was originally

developed for use in Google earth and has since come into widespread use in the geoweb”

[59]. It is tag based and defines elements for describing geographical data. Google describe

KML as “... a file format used to display geographic data in an earth browser, such as

Google Earth, Google Maps, and Google Maps for mobile” [60]. The earth browsers

“read” KML and format and output geographical features in the same way that web

browsers “read” HTML and format the display accordingly.

KML provides the mechanisms to define overlays and attach them to the map view in a

browser. This allows us to take the rich complimentary spatial datasets and convert them to

second “map” that can be placed on top of our Google map that is embedded in our web

page. KML provides for the describing of points, lines, polygons and indeed irregular

shapes to represent regions or geographical features complete with associated text.

2.4 Conclusion: Bringing it all together

These building blocks allow us to assemble our extended application into the following

process model.

1. The user logs on to the system and submits a search requirement via a web form.

2. The Apache web server processes the HTML and Javascript while PHP is passed on

to the PHP executable.

3. PHP processes the script and if there are database operations to be performed,

connects and queries the database based on the parameters received.

4. The MySQL database performs the query and returns the results.


5. The PHP executable receives the result set and closes the connection. After closing

the connection it formats the result set and returns it to the web server as HTML or

XML depending on script commands.

6. The Apache web server receives the result set and continues with Javascript

processing to request map services from the Google API.

7. The web server formats the output which includes the request for the appropriate

map and send it back to the browser.

8. The Google map service processes the request and serves the browser with the

requested dynamic map with the specified components, (points or overlays).

9. The browser displays the combined output from the web server and the map from

the Google map service.

Figure 2.3 shows the steps in the process model.


Figure 2.3: Process model for the extended application

The different servers (software), with the exception of the Google maps service can be run

on a single hardware system although in practice the web server and the database server

are often run on separate hardware systems. Our implementation will run on a single

hardware system.


Chapter 3 Implementation

3.1 Implementation Objectives

The aim of this project is to develop GIS support for an emerging infectious disease

monitoring project, AVID. This project builds on earlier work that culminated in an

information system that supported the collection and management of biological sample

data. To this information system, we are to provide GIS support via mapping extensions to

manage the spatial data collected.

Earlier in this document, under our design considerations, we set out three objectives we

would like to fulfil with the GIS extension to the application. These were,

• Visual representation of sample collection points within the geographical research


• Overlay of additional datasets of information that are relating to the geographical

research area as may be made available.

• Representation of the AVID dataset within the bounds of a time period.

Using the tools outlined in the chapter on methodologies, the following is the

implementation of the stated project objectives in the extended application, divided into

three sections :

1. Data acquisition and preparation.

2. Data processing

3. Data presentation

These complement the three-tier architecture model described earlier in this document

although they do not directly correspond. They are not entirely discrete and overlap

wherever necessary.


3.1.1 Data acquisition and preparation

AVID is responsible for collecting and processing biological data and yet the application

requires further external data sources to process the spatial data effectively. Therefore, the

origins of the data for the application can be divided into two sources: AVID and others.

AVID data

The AVID project collects sample data from the field using a custom application.

Biological samples are collected from the human and the various animal populations and

analysed for pathogens. This process can be looked as occurring in two distinct phases;

sample collection and sample processing.

In the sample collection phase, data is collected about the sample origin, location and

storage. This is fed into a database together with information on which project it belongs

to. During the sample processing phase, the results of the serology tests are added to the

database as well. This completed the picture of the AVID data.

The original application was developed without receiving any actual data input from the

AVID project. For this project, AVID provided a batch of 5,000 records as sample input

data for the application. These records were in the form of a Microsoft Excel spreadsheet

with multiple worksheets which had been extracted from the AVID database.

The main worksheet containing sample data contained 18 columns with a secondary

worksheet detailing the organism types sampled that held two columns. The data required

sorting before it could be fit for purpose. The sorting process involved three tasks :-

• Verifying that there were no empty cells or cells with null values.

• Verifying that the data in the worksheets was valid i.e. that that the data was in line

with what was expected from the column heading.

• Verifying that there were no duplicate records and that there was consistency in the

various identification columns in the spreadsheet.


The excel spreadsheet was manipulated to list the rows in alphabetical order to group null

or empty values in various fields. The aim was to eliminate rows with incomplete or

missing data from the worksheets. This resulted in all rows with missing data (empty cells)

or null values to be removed from the sample data. Subsequently, the excel sheet was

manipulated to list the date and time data in order. Rows with invalid date and time values

were removed at this point.

Finally the excel sheet had duplicate columns removed for clarity. Rows with inconsistent

value formats were also removed.

Out of the 5,000 received records, the sorting process filtered the set down to 3,419. 1,581

rows were discarded for failing the verification criteria process outlined above. The

remaining rows would then move on to data processing which is covered later in this

chapter.

Other data

Other data is required for the GIS component to satisfy the stated objectives. Acquisition

of complementary datasets for overlay purposes is the task that this comprises of. The GIS

component of the application is to overlay multiple datasets over the base map to achieve

this.

A total of five datasets were considered for this purpose. The five datasets are comprised of

:-

1. Human population density.

2. Livestock density.

3. Rainfall / precipitation.

4. Wildlife population densities.

5. Tsetse fly prevalence.

These datasets were acquired from publicly available repositories namely the GIS section

of International Livestock Research Institute (ILRI) and the publications section of the

World Resources Institute websites. The Kenya Wildlife Service was also requested to

provide up to date GIS data.


3.1.2 Data Processing

The data processing is divided into three parts:-

1. Formatting.

2. Database creation and import.

3. Overlay conversion.

Formatting

The filtered AVID data required processing before it could be used by the application.

Specifically, the data and time information required reformatting and the geographical

coordinates were to be converted to a signed decimal notation. Date and time values were

formatted with an appropriate mask to emulate a timestamp, resulting in the format

yyyy:mm:dd hh:mm:ss.

The coordinate columns were converted to signed decimal notation with latitudes having

southerly values designated negative and vice versa. Longitudinal data was also

transformed to the same scheme with westerly values designated as negative and vice

versa. This was necessary to facilitate the passing of coordinate values to the Google API

which uses this format.

Finally, the values in the ELISA results were changed from text values “0” and

“POSITIVE” to integer values zero (0) and one (1) respectively, to represent the boolean

values true and false.

Database creation and import

The next step in the processing was to import the data into the MySQL database. The

original application defined database structures to hold AVID information. It was designed

without the benefit of sample data and as a result the table structures would be revised to

reflect the actual data that had been received. A new database was created in the MySQL

RDBMS.


The new table structures were developed along the same guiding principles as those of the

original table, taking into account the significance of spatial data. This resulted in six tables

being defined. The tables are listed below with a short description and their structure:

1. Organisms – this defines all the different organisms from which samples are

collected. The primary key is “id” which is set to auto-increment with every new

row added to the table. The “name” field describes the type of organism and the

“comment” stores any additional descriptive information.

Table 3.1: Organisms table structure

Field Data Type Null Default Value Primary Key

id Integer (11) No None Yes

name Varchar (30) No None

comment Varchar (25) Yes Null

2. Samples – this defines the sample types prepared for biological analyses. The

primary key is “id” which is set to auto-increment with every new row added to the

table. The “sample_type” field is an integer identifier for the samples and the

“description” field holds the text description of this identifier.

Table 3.2: Samples table structure



sample_type Integer (3) No None

description Varchar (60) No None

3. Projects – this defines the various different projects for which samples are

collected and analysed. The primary key is “id” which is set to auto-increment with

every new row added to the table. The “Project” field stores an integer identifier

for the project that is conducting the sampling. The “Description” field stores the

text identifier for the project.


Table 3.3: Projects table structure



Project Integer (3) No None

Description Varchar (20) No None

4. ELISA – this table defines the pathogens that are targeted through the serology

analyses and holds the information on which samples tested positive for which

pathogens. The primary key in this table is “SampleID” which holds the unique

alphanumeric identifier that is assigned to each sample. The “id” field is set to

auto-increment with every new row added to the table. The “Rift_Valley_Fever”,

“Anaplasma_marginale”, “Typanosoma_congolense” and “Theileria_parva” store

integer values that are interpreted as boolean for the presence and absence of the

respective pathogens.

Table 3.4: ELISA table structure


id Integer (11) No None

SampleID Varchar (11) No None Yes

Rift_Valley_Fever Tinyint (4) No None

Anaplasma_marginale Tinyint (4) No None

Typanosoma_congolense Tinyint (4) No None

Theileria_parva Tinyint (4) No None

5. Storage – this table holds all the physical storage information for each sample and

the project it belongs to. The primary key in this table is “SampleID” which holds

the unique alphanumeric identifier that is assigned to each sample. The “id” field is

set to auto-increment with every new row added to the table. The “sample_type”

field is an integer identifier for the samples. The “Organism” field stores an integer

value that corresponds to the “id” field in the Organism table and is used as a key

to identify the sampled organism. The “Animal_ID” field holds a unique identifier

for animals that are sampled. The “Box_ID” field identifies the box in which a


sample is stored. The “Box_details” field identifies the location within the box that

the sample is stored. The “Tray_ID” holds the name of the box in which the

sample is stored. The “Project” field stores the integer identifier for the project that

is conducting the sampling.

Table 3.5: Storage table structure




sample_type Integer (3) No None

Organism Integer (2) No None

Animal_ID Varchar (7) No None

Box_ID Integer (5) No None

Box_details Varchar (4) No None

Tray_ID Varchar (10) No None

Project Integer (2) No None

6. Spatial – this table defines and holds all the spatial data associated with the

biological samples. The primary key in this table is “SampleID” which holds the

unique alphanumeric identifier that is assigned to each sample. The “id” field is set

to auto-increment with every new row added to the table. The “Visit_ID” stores the

name of the place where the sample was collected or an alphanumeric identifier for

the same. The “Visit_Date” field stores the date and time that the sample was

collected. The “Long” and “Lat” fields store the longitude and latitude coordinates

taken from the GPS device for the location where the sample was collected.

Table 3.6: Spatial table structure




Visit_ID Varchar (12) No None

Visit_Date datetime No None

Long Float (10,6) No None

Lat Float (10,6) No None


Importing was then completed by creating new worksheets in the filtered excel spreadsheet

containing only the fields that corresponded to each data table and then exporting the

contents of each into text files in comma separated value (csv) format which were then

imported into the corresponding MySQL database tables using the phpMyAdmin web

interface.

Overlay conversion

The overlays acquired for the GIS component are created in the ArcGIS software and thus

are in a proprietary file format that is used in that software. The file format is known as the

shapefile spatial data format. According to the Environmental Systems Research Institute

(1998), a shapefile consists of three files: a main file (extension .shp ), an index file

(extension .shx) and a dBASE table (extension .dbf) [61]. The main file contain the shapes

defined by vertices, the index file contains the offsets for each shape in relation to the

beginning of the main file and the dBASE file contains the feature attributes for each shape

in the order they appear.

This format is cannot be used by the Google Maps API and as such conversion needs to be

performed to Keyhole Markup Language in order for the overlays to be compatible with

our application. The conversion of the overlays from the shapefile format to KML format

is performed by a third party software called Shape2KML. This program allows for the

conversion of shapefiles to KML format as well as the assignment of colour schemes.

Shapefiles do not support colour directly with that feature being handled by the software

which is reading the file assisted by any colour information that may be contained in the

dBASE file as an attribute.

The five overlays were processed using the Shape2KML utility and the resulting KML files

were used in the application. However, two of the shapefiles, population density and

wildlife density were not used. This is because the conversion process did not yield files

with adequate colour differentiation to make the visual information useful. Therefore, only

three of the datasets yielded satisfactory overlays that could be used with the GIS

component.


3.1.3 Data Presentation

To present the data and run the application, we need a suitable environment. Since the

application is a web application we require a web server. We also require a database and

scripting engine as per our three-tier architectural model. Having already selected our tool

set previously we need to set up our environment.

To do this we use the WampServer program. WAMP is an acronym that stands for

Windows (Operating System), Apache (web server), MySQL (RDBMS) and PHP

(scripting engine). It provides a complete installation of the web server, scripting engine

and RDBMS on a Microsoft Windows computing environment along with associated

management tools. This type of environment is derived from a LAMP environment,

(Linux, Apache, MySQL, PHP), which had proven to be very popular among developers in

the open source community. There also exists a MAMP environment, (MacOS, Apache,

MySQL, PHP), for Apple Macintosh hardware computing environments that are running

MacOS. This allows our application to be set up on any one of the three major software

environments that exist currently. Appendix C outlines the installation process for the

WampServer 2.0i which is the version that is used in this project. The Wampserver

software installs all three components and includes a web interface for management of the

MySQL RDBMS, phpMyAdmin. This tool allows us to manage our databases from within

a web browser, simplifying the task of creating tables and views. Once installation is

complete, we are able to create a number of database tables and views that are crucial to

the presentation of our data. The database tables have been covered in the data processing

section and so we move on to database views.

A database view is a subset of of data created through a stored query and is accessible as a

virtual table in a database. Database views are useful in simplifying queries and collating

data from different tables within a database into one virtual table for querying. This has the

advantage of allowing flexible subsets to be created without affecting the underlying tables

holding the data. These views can be used in the creation of more views allowing for the

compilation of a virtual table containing fields from several “real” tables. To illustrate this

we shall look at three views that form the basis of our main data presentation.


The first view, named trial , is created with the following SQL query.

select `Spatial`.`SampleID` AS `SampleID`,`Spatial`.`VisitID` AS

`VisitID`,`Spatial`.`VisitDate` AS `VisitDate`,`Spatial`.`Long` AS

`Long`,`Spatial`.`Lat` AS `Lat`,`Storage`.Òrganism AS Òrganism` from (`Spatial`

left join `Storage` on((`Spatial`.`SampleID` = `Storage`.`SampleID`)))

This query creates a result set with fields from the “Spatial” table and the “Storage” table,

resulting in a virtual table with the following structure.

Table 3.7: trial view structure

Field Data Type

SampleID Varchar (12)

Visit_ID Varchar (12)

Visit_Date datetime

Long Float (10,6)

Lat Float (10,6)

Organism Integer (2)

The query uses a left join on the fields Spatial.SampleID and Storage.SampleID to achieve

this.

This gives us the type of the organism as identified by the number key appended to the

fields found in the Spatial table. In order to make this information easier to understand, we

use this view as a source to create another, called trial1. We use the following SQL query.

select `trial`.`SampleID` AS `SampleID`,`trial`.`VisitID` AS

`VisitID`,`trial`.`VisitDate` AS `VisitDate`,`trial `.`Long` AS `Long`,`trial`.`Lat` AS

`Lat`,Òrganisms`.`name` AS `name` from (`trial` left join Òrganisms`

on((`trial`.Òrganism` = Òrganisms`.ìd`)))


This query creates a result set with fields from the “trial” view and the “Organism” table,


Table 3.8: trial1 view structure

Field Data Type



Visit_Date datetime

Long Float (10,6)

Lat Float (10,6)

name Varchar (30)

The query uses a left join on the fields trial.Organism and Organisms.id to achieve this.

This gives us the name of the organism as identified by the “name” field in the Organisms

table, which effectively replaces the “Organism” field that was found in the “trial1” view.

Using this second view, we create a third which adds fields from the ELISA table using the

following SQL query.

select `trial1`.`SampleID` AS `SampleID`,`trial1`.VisitID` AS

`VisitID`,`trial1`.`VisitDate` AS `VisitDate`,`tria l1`.`Long` AS `Long`,`trial1`.`Lat` AS

`Lat`,`trial1`.`name` AS `name`,ÈLISA`.`Rift Valley Fever` AS `Rift Valley

Fever`,ÈLISA`.Ànaplasma marginale` AS Ànaplasma

marginale`,ÈLISA`.`Trypansoma congolense` AS `Trypansoma

congolense`,ÈLISA`.`Theileria parva` AS `Theileria parva` from (`trial1` left join

ÈLISA` on((`trial1`.`SampleID` = ÈLISA`.`SampleID`)))

This query creates a result set with fields from the “trial1” view and the “Organism” table,



Table 3.9: avidresults1 view structure

Field Data Type



Visit_Date datetime

Long Float (10,6)

Lat Float (10,6)

name Varchar (30)

Rift_Valley_Fever Tinyint (4)

Anaplasma_marginale Tinyint (4)

Typanosoma_congolense Tinyint (4)

Theileria_parva Tinyint (4)

The query uses a left join on the fields trial1.SampleID and ELISA.SampleID to achieve

this.

This gives us the ELISA results for each sample in our new view, named “avidresults1”.

This view forms the table that we shall query from our web pages.

While the view “avidresults1” provides the spatial and serology result data, we create

another view to provide additional details on the sample. We follow the same pattern of

building views, starting with the following query.

select `Storage`.`SampleID` AS `SampleID`,`Storage`.Ànimal_ID` AS

Ànimal_ID`,`Storage`.`Box_ID` AS `Box_ID`,`Storage.`Box_Details` AS

`Box_Details`,`Storage`.`Tray_ID` AS `Tray_ID`,`Storage`.`Project` AS

`Project`,`Samples`.`description` AS `description` from (`Storage` left join `Samples`

on((`Storage`.`sample_type` = `Samples`.`sample_type`)))

This query creates a result set with fields from the “Storage” table and the “Samples” table,



Table 3.10: storage1 view structure

Field Data Type


Animal_ID Varchar (7)

Box_ID Integer (5)

Box_details Varchar (4)

Tray_ID Varchar (10)

Project Integer (2)

description Varchar (60)

The query uses a left join on the fields Storage.sample_type and Samples.sample_type to

give us the description of the sample type appended to a subset of the fields found in the

Storage table.

To this view, named “storage1”, we add the name of the project that the sample belongs to

by creating a new view with the following SQL query.

select `storage1`.`SampleID` AS `SampleID`,`storage1`.Ànimal_ID` AS

Ànimal_ID`,`storage1`.`Box_ID` AS `Box_ID`,`storage1`.`Box_Details` AS

`Box_Details`,`storage1`.`Tray_ID` AS `Tray_ID`,`storage1`.`description` AS `Sample

Type`,`Projects`.`Description` AS Òrigin` from (`storage1` left join `Projects`

on((`storage1`.`Project` = `Projects`.`Project`)))

This query creates a result set with the fields from the “storage1” view and the “Samples”

table, resulting in a virtual table with the following structure.

Table 3.11: storage2 view structure

Field Data Type


Animal_ID Varchar (7)

Box_ID Integer (5)

Box_details Varchar (4)

Tray_ID Varchar (10)

Sample Type Varchar (60)

Origin Varchar (20)


The query uses a left join on the fields storage1.Project and Projects.Project to give us the

project name as the field “Origin”, effectively replacing the “Project” field in the

“storage1” view.

These two views, “avidresults1” and “storage2”, are the two tables that our PHP scripts

shall be querying.

The queries that the PHP scripts pass to the MySQL RDBMS are user driven. This means

that they are created from user input via a web form. This form is contains several input

fields as can be seen in figure 3.1 below.

The user can enter a sample id in the search term field as well as dates to search within.

The list of pathogens that AVID monitors is included as a drop down list. The PHP code

that converts the user input into a formatted SQL query for the database is shown below.


Figure 3.1: Avid Search form

<?php

// Has the user provided the parameter?

if (empty($searchValue))

{

// No, the user hasn't provided a parameter or initial page load

// Do nothing

} // end of if empty($searchValue) body

else

{

// Start a query ...

$query = "Select SampleID,

`VisitID`,

`VisitDate`,

`Lat`,

`Long`,

`name` AS 'type'

from àvidresults1`";

// ... then, if the user has specified a search term,

// add the searchValue as a WHERE clause ...

if ($searchValue != "All")

{$query .= " WHERE SampleID Like '$searchValue%'";}

else {$query .= " WHERE 1";}

// ... then, if the user has specified a "From" date term,

// add the timestampFrom field as a WHERE clause ...

if ($fromValue != "")

$query .= " AND VisitDate >= '$fromValue'";

// ... then, if the user has specified a "To" date term,

// add the timestampFrom field as a WHERE clause ...

if ($toValue != "")

$query .= " AND VisitDate <= '$toValue'";

// ... then, if the user has selected a pathogen,

// add the selection value as a WHERE clause ...

if ($disease != "0")

$query .= " AND `$disease` IS TRUE";

// ... and then complete the query.

$query .= " ORDER BY VisitDate LIMIT 0, 100";


The results are limited in the query to 100 for the sake of map performance. This shall be

elaborated on later. The comments in the code provide a step by step description of the

assembling of the query from the user parameters filled in the form before the query is

completed. Once the query is complete, the following code passes it on the the MySQL

RDBMS.

// run the query and show the results

displayAVIDList($connection, $query, $searchValue);

The function displayAVIDList() takes three arguments; a database connection, an SQL

query as a string and finally the search parameter entered by the user in the form. The

function submits the query to the database and prints any results in a table together with a

button for mapping them. If no results were found, the user is informed accordingly.

The code for the function is displayed below,

// Show all sample records in a <table>

function displayAVIDList($connection, $query, $searchValue)

{

// Run the query on the DBMS

if (!($result = @ mysql_query ($query, $connection)))

mysql_error( );

// Find out how many rows are available

$rowsFound = @ mysql_num_rows($result);

// If the query has results ...

if ($rowsFound > 0)

{

// Define $color=1

$color="1";

// ... print out a header

echo "<p>Results for $searchValue<br>";


// Map link

echo '<p>

<form name="goMap" action="mapthis.php" method="POST">

<input type="hidden" name="filters" value="';

echo rawurlencode($query);

echo '"><input type="submit" value="Map this result set">

</form>

</p>';

// and start a <table>.

echo '<table width="400" border="0" align="left" cellpadding="2" cellspacing="1">';

//Column titles

echo '<tr><td>Sample ID</td><td>Location</td><td>Organism</td></tr>';

while($rows=mysql_fetch_array($result)){

// If $color==1 table row color = #FFC600

if($color==1){

echo "<tr bgcolor='#bfd9ff'>

<td>".$rows['SampleID']."</td><td>".$rows['VisitID']."</td><td>".$rows['type']."</td>

</tr>";

// Set $color==2, for switching to other color

$color="2";

}

// When $color not equal 1, use this table row color

else {

echo "<tr bgcolor='#e8e8e8'>

<td>".$rows['SampleID']."</td><td>".$rows['VisitID']."</td><td>".$rows['type']."</td>

</tr>";

// Set $color back to 1

$color="1";

}

} // end while loop body

// Finish the <table>

echo '</table>';


} // end if $rowsFound body

else

{

// Ask user to check selection

echo "<p>No records were found. Please review your search parameters.</p>";

}

// Report how many rows were found

echo "$rowsFound records found matching your

criteria.</p>";

} // end of function

The results are displayed in a table as shown in figure 3.2, together with the button used to

implement the mapping option.

The form that is created for the mapping facility holds the SQL query in a hidden field.

This is necessary for the mapping page to retrieve the coordinates of the samples in the

result set to display on the map. The form passes the query to the mapping page which in

turn passes it to another PHP script that uses it to recreate the dataset and retrieve the

coordinates. These are then written to an XML file which serves as a temporary data store

for the Google Maps script that renders them on the map. The page is based on a sample

script provided in the Google Maps API documentation for performing exactly this type of

map manipulation.


Figure 3.2 : AVID search results

The mapping page relies on a PHP script to convert the output of the database into the

XML store. This PHP script is invoked by a Google Maps API function, downloadUrl(url,

callback), which is a function for XML and data parsing. The code snippet that does this is

listed below.

downloadUrl("avid_genxml.php"+"?option="+ postVars, function(data) {

var xml = data.responseXML;

var markers = xml.documentElement.getElementsByTagName("marker");

for (var i = 0; i < markers.length; i++) {

var name = markers[i].getAttribute("name");

var address = markers[i].getAttribute("address");

var type = markers[i].getAttribute("type");

var point = new google.maps.LatLng(

parseFloat(markers[i].getAttribute("lat")),

parseFloat(markers[i].getAttribute("lng")));

var html = '<b><a href="details.php?sample=' + name + '">' + name +'</a></b> <br/>' + address +

"<br/>" + type;

var icon = customIcons[type] || {};

var marker = new google.maps.Marker({

map: map,

position: point,

icon: icon.icon,

shadow: icon.shadow

});

bindInfoWindow(marker, map, infoWindow, html);

pointArray.push(marker);

}

});

}

The function invokes the script “avid_genxml.php” and passes the query to it in the HTTP

GET parameter “option”. The Javascript variable “postVars” contains the query string that

it url encoded, having been retrieved from the HTTP POST variable that was passed from

the form in the previous page.


The PHP script “avid_genxml.php” carries out the following functions when invoked:

• Opens a connection to the database server and selects the appropriate database.

• Retrieves the query string from the HTTP GET variable passed to it.

• Executes the query

• Outputs each row result into an XML document as elements.

• Closes the database connection.

The code snippet for the PHP script performing the above functions is below:

// Opens a connection to a MySQL server

$connection=mysql_connect ($localhost, $username, $password);

if (!$connection) {

die('Not connected : ' . mysql_error());

}

// Set the active MySQL database

$db_selected = mysql_select_db($database, $connection);

if (!$db_selected) {

die ('Can\'t use db : ' . mysql_error());

}

// Use get value for query

$query= rawurldecode($_GET['option']);

// Select the matching rows according to the query

$result = mysql_query($query);

if (!$result) {

die('Invalid query: ' . mysql_error());

}

header("Content-type: text/xml");

// Start XML file, echo parent node

echo '<markers>';

// Iterate through the rows, printing XML nodes for each

while ($row = @mysql_fetch_assoc($result)){

// ADD TO XML DOCUMENT NODE

echo '<marker ';


echo 'name="' . parseToXML($row['SampleID']) . '" ';

echo 'address="' . parseToXML($row['VisitID']) . '" ';

echo 'lat="' . $row['Lat'] . '" ';

echo 'lng="' . $row['Long'] . '" ';

echo 'type="' . $row['type'] . '" ';

echo '/>';

}

// End XML file

echo '</markers>';

// Close the DBMS connection

mysql_close($connection);

The downloadUrl() function will execute the callback argument, function(data), after the

PHP script has concluded and make use of the the XML document created to create the

points on the map and place markers at the corresponding locations. Each point represents

a sample from the result set and is marked by a custom icon depending on organism type.

These icons are defined in the page and correspond to the Organism name retrieved from

the database. Each type of organism will have a different coloured icon to represent it on

the map. Clicking on an icon brings up supplementary information about that sample,

written into the information window component of the marker. The information window

content was modified to include a link to a further script that retrieves the storage details of

the sample. Additionally, the downloadUrl() function was also modified so that it stores

the points into a global array for use in the animation function.


The results of the mapping page on a set of results is shown in the screen capture below.

This is the achievement of our first objective: visual representation of sample collection

points within the geographical research area. The first thing we notice is that a selection of

our points lie in the Indian Ocean. This is explained under “Challenges” in the next

chapter.

Having arrived at this point, we proceed to implement the code to satisfy our second

objective: overlay of additional datasets of information that are relating to the geographical

research area. There are three datasets available as overlays. These are implemented by

three buttons on the mapping page interface, along with a button to remove them from the

map window. Furthermore, custom overlays are supported by use of a feature that takes the

URL of a valid KML file entered in text box and overlays it on the map. This allows any

valid KML overlay to be added to the map, provided it is publicly accessible on the

internet.


Figure 3.3: Points on map

Figure 3.4. shows the interface that performs these functions.

Each button is mapped to a Javascript function to perform the overlay. The code snippets

that achieve this are listed below.

Button Mapping

<div id="overlays">

<input onclick="showRainOverlay();" type=button value="Rainfall distribution"/>

<input onclick="showTsetseOverlay();" type=button value="Tseste distribution"/>

<input onclick="showCattleOverlay();" type=button value="Cattle distribution"/>

<input onclick="clearOverlays();" type=button value="Clear Overlays"/>

<br/>

<form action="#" name="overlay">

<input name="url" type="text">

<input onclick="showCustomOverlay(this.form);" type="button" value="Custom overlay">

</form>

</div>

Overlay case 1: a fixed overlay

// Shows rain overlay

function showRainOverlay() {

var rainfallLayer = new

google.maps.KmlLayer('http://sites.google.com/site/eaoverlays/home/kenya_rainfall_distribution.kml');

rainfallLayer.setMap(map);

kmlArray.push(rainfallLayer);

}

Overlay case 2: Custom overlay

// Shows Custom overlay

function showCustomOverlay(form) {

var layer = form.url.value;

var customLayer = new google.maps.KmlLayer(layer);

customLayer.setMap(map);

kmlArray.push(customLayer); }


Figure 3.4: Overlay choices

Custom overlays must be in the form http://www.foo.com/bar.kml . The domain name must

be fully qualified due to the way in which the Google Maps API processes overlays.

Overlays are first retrieved by the Google Maps Server for pre-processing before the data

is sent back to the calling page for display. As a result, the overlays must be accessible

publicly and referenced through fully qualified domain names or public i.p. addresses. The

overlays used in the application have been placed on a public website to facilitate this. The

firewall configuration of the University network does not allow the Google Maps service

to access the files in the manner required and so this is a workaround to remedy the

situation.

Overlays can also be used in tandem. This is useful to compare two or more related or

contrasting datasets. There are however, practical limitations in what a user may be

visually able to distinguish when too much information is added to a map. The following

screen captures show a single overlay and the combination of two overlays on the map.


Figure 3.5: Single overlay

This is the achievement of our second objective: overlay of additional datasets of

information that are relating to the geographical research area. The ability to overlay

allows for visual comparisons and the ability to zoom to high levels means that visual

analyses can be carried out to a detailed level. We move on to implement our third and

final objective: the representation of the AVID dataset within the bounds of a time period.

Time and animation KML elements are not supported in the Google Maps API at this time,

only in the Google earth plug-in as was discussed earlier in this document. Therefore, we

use standard Javascript and other Google Maps API functions to implement an animation

function. The animation function makes use of a global array that holds all the markers on

the map. The SQL query that generates the dataset from which the points are plotted has an

ORDER BY clause on the “VisitDate” field that causes the results and consequently the

markers to be ordered according to when the samples were collected. The animation is

effected by the “Show sequence” button on the mapping page. This button has an onclick

event associated with it that calls the function executePan(). This function first clears all

the markers on the map by iterating through the array and removing them from the map.

This is done by passing a null value to the marker method setMap().


Figure 3.6: Combined overlays

The function then calls another function, moveTo(), with a Javascript method

panIntervalId(). This method allows the calling of a function repeatedly at a set interval, in

this case 2 seconds. The moveTo() function restores a market to the map each time it is

called and updates a pointer value. Each time the function is called, the pointer will be

checked against the array length before proceeding to make sure the end has not been

reached. If not, a marker will be restored from the array until the end is reached, after

which it calls the Javascript function clearIntervalId() that stops the repetition.

The code snippet containing the two functions is below.

// function iterates through the different points in the array

function moveTo() {

if (last < pointArray.length)

{

pointArray[last].setMap(map);

last++;

}

else // We have reached the end

{

//Reset value of pointer and stop

last = 0;

clearIntervalId(panIntervalId);

}

}

// Panning function

function executePan() {

var pointer = 0;

// Clear map first

while (pointer < pointArray.length){

pointArray[pointer].setMap(null);

pointer++;

}

// Set zoom and call pan function

map.setZoom(7);

panIntervalId = setInterval("moveTo()",2000);


}


The effect is to animate the map by restoring the markers in a time sequence according to

when they the samples were collected. Therefore if a search for a pathogen is carried out,

the result set, when mapped in time sequence, will show the detection of the disease in

samples over a period of time. The more accurate the source data, the more accurate the

animation will be. The following screen captures show an animation being executed.

This forms the implementation of the third objective. Putting it all together we are able

present a typical use case.


Figure 3.7a: Clear map Figure 3.7b: Animation begins

Figure 3.7c: Animation progressing

3.2 Typical use case

Before we begin, the application must be installed. Installation instructions are contained in

Appendix D. The application provides user access control via a login page requiring a

username and password. The default username and password pair consists of “John” and

“1234”.

The user access mechanism was maintained from the previous application. Correct user

credentials allow us access to the search form.

Figure 3.9 shows a populated search form with dates selected using the calendar control

and a pathogen selected from the drop down list. Clicking the “Go” button submits the

form. The script executes and returns the results. This is shown in figure 3.10 below.


Figure 3.8: Login screen

Figure 3.9: Populated search form

Mapping the result set produces a map with the 43 markers. Each marker can be clicked

and the information window that pops up displays the sample id, the location and the

organism associated with the sample. The sample id is hyperlinked to a page that displays

the storage information for that sample. Figures 4.11a and 4.11b show the map, with the

information window clicked, and the details page respectively.

This concludes our typical use case and our implementation.


Figure 3.10: Results

Figure 3.11a: Marker clicked

Figure 3.11b: Sample storage details

Chapter 4 Conclusion

4.1 Review

This project set out to establish a GIS component that would make use of the spatial data

collected in the AVID project samples. Three objectives were identified as follows:

1. Visual representation of sample collection points within the geographical research


2. Overlay of additional datasets of information that are relating to the geographical

research area as may be made available, such as weather conditions or livestock

population densities.

3. Representation of the AVID dataset within the bounds of a time period. In order to

represent events occurring over a timespan of selected weeks or months it should

be possible represent the incidences in an animated form showing the occurrence

of disease over the map with time.

These three objectives were achieved with the following levels of success. Objective 1 was

achieved fully with the sample collection locations plotted on the map. Custom icons were

used to differentiate between organisms and additional data relating to the samples

displayed in the information window of the markers.

Objective 2 was achieved fully with the ability to overlay both fixed (hard coded) and

custom (dynamic) overlays implemented. Three overlays were prepared and coded into the

application.

Objective 3 was partly achieved. The animation of the occurrence of disease was

implemented but without indication of any time scale. This was due to the limitation

inherent in the Google Maps API.


4.2 Challenges

Several challenges were encountered during the design and execution of the project.

1. Non-uniform identifiers for samples. Discrepancies in the sample identifiers for

samples from human sources and all other samples resulted in the data relating to

the those samples being excluded from the dataset during data preparation.

2. Degraded spatial data. Non-disclosure and confidentiality concerns resulted in a

degradation of spatial data by the use of random offsets by AVID before the data

was provided. This caused the locations plotted on the map to appear in illogical

areas such as several miles off the Kenyan coast in the Indian Ocean.

3. Shapefile conversion. The files used to generate the overlays for the application

were in the Shapefile format. The conversion process was not ideal as it was

performed by a third party tool. Colour information as well as shape definition in

some instances suffered.

4. Availability of up to date overlay datasets. Up to date demographic datasets as well

as those relating to wildlife proved difficult to acquire within the time allocated to

the project. Most of the datasets were over a decade old and may not be an accurate

reflection of the current state of the various environments.

5. The Google Maps API was unable to process large numbers of markers. This

restriction placed a limitation on the size of the result set that could be mapped at

any given time.

6. The time data available only represents the sample collection data and time and not

the date of infection. This means that there is always a slight error inherent the

detection of disease occurrence as the time data is not corrected for the phase of the

infection detected.


These challenges represent real world constraints that would need to be overcome to refine

the application before integration into a wider information management system. The

measures required to overcome them require more time to explore than was available for

his project and that in itself forms the overriding challenge encountered.

4.3 Recommendations

Based on the challenges encountered, the following recommendations are put forward:

1. Sample identifiers for all samples should be harmonized into a single system so

that data from all the samples can be included in the querying.

2. Spatial data will need to be provided without degradation in order to plot markers

that lie within the bounds of the land masses where sampling is taking place.

3. Shapefile conversion should be performed from within the application that

generated them, This will result in high quality KML files that accurately reflect the

original files.

4. Where available, up to date GIS data should always be used for the overlay

datasets. This will give accurate comparisons for trend analyses. In particular,

environmental factors that affect disease prevalence should be matched up with the

time bounds that are being queried to give an accurate picture of any correlation.

5. The Google Maps API support team should be contacted to provide a means to ease

this restriction and allow for the mapping of large result sets.

These recommendations are a minimal set, designed to improve the quality of application

and are by no means exhaustive. Rather, the application would benefit greatly from

integration with a wider information system in order for maximum utility to be realised.

Thus a wider context for recommendations would be available in which this application

would be viewed as a module.


4.4 Future work

In view of the limited time available for this project and the evolving nature of the user

requirements, the following are suggestions for future work.

1. The Google Earth plug-in should be explored as it matures to provide better

animation features.

2. An import function for the AVID data should be designed and export templates

perhaps supplied so that database extracts from the AVID database can be used

directly, saving data preparation time. Alternatively, database views can be

generated on the AVID database to allow the application to work directly off it.

3. Report generation features should be explored to produce portable document format

maps as well as statistics on the AVID data.

4. User authentication and management remains rudimentary and there is no facility

to manage user access easily. This would be an area to develop a more complete

user management facility.


References

1. MedicineNet.Com “Definition of Emerging infectious disease”.http://www.medterms.com/script/main/art.asp?articlekey=22801 last accessedFriday 13th August 2010.

2. MedicineNet.Com, “Definition of Emerging infectious disease”.http://www.medterms.com/script/main/art.asp?articlekey=22801 last accessedFriday 13th August 2010.

3. Centre for Disease Control Special Pathogens Branch, Rift Valley Fever Questionsand Answers, http://www.cdc.gov/ncidod/dvrd/spb/mnpages/dispages/rvf.htm lastaccessed Friday 13th August 2010.

4. Leggat P., Goldsmid J., 2001. “Zoonosis”, Dictionary of tropical medicine forHealth professionals, Australia: ACTM.

5. World Health Organization, “Zoonoses and public health”, Veterinary public health(VPH), http://www.who.int/zoonoses/vph/en/ last accessed Friday 13th August2010.

6. World Health Organization, “Emerging Zoonoses”, Zoonoses and veterinary publichealth, http://www.who.int/zoonoses/emerging_zoonoses/en/index.html lastaccessed Friday 13th August 2010.

7. World Health Organization, Avian Influenza,http://www.who.int/topics/avian_influenza/en/ last accessed Friday 13th August2010.

8. World Health Organization, Cumulative Number of Confirmed Human Cases ofAvian Influenza A/(H5N1) Reported to WHO,http://www.who.int/csr/disease/avian_influenza/country/cases_table_2010_08_12/en/index.html last accessed Friday 13th August 2010.

9. Mcleod A. et al, 2005. “Severe localised effects of 2003 – 4 outbreaks”, Economicand social impacts of Avian Influenza, FAO.

10. The World Bank, 2006. Table 1.7 Possible economic impacts of flu pandemic,Global Development Finance 2006, Page 37, USA: World Bank.

11. The World Bank, 2006. Table 1.8 A breakdown of economic impacts of a potentialhuman-to-human pandemic , Global Development Finance 2006, Page 38, USA:World Bank.

12. World Health Organization, 2010, “What is the pandemic (H1N1) 2009 virus?” ,Pandemic (H1N1) 2009: frequently asked questions,http://www.who.int/csr/disease/swineflu/frequently_asked_questions/about_disease


/en/index.html last accessed Friday 13th August 2010.

13. World Health Organization, 2010, “Laboratory-confirmed cases of pandemic(H1N1) 2009 as officially reported to WHO by States Parties to the IHR (2005) asof 1 August 2010” , Pandemic (H1N1) – Update 112,http://www.who.int/csr/don/2010_08_06/en/index.html last accessed Friday 13th

August 2010.

14. AVID, 2009. “Objectives”, An Integrated Response System for EmergingInfectious Diseases in East Africa .


16. World Health Organization, “Overview”, Rift Valley Fever,http://www.who.int/mediacentre/factsheets/fs207/en/index.html last accessedFriday 13th August 2010.


18. Food and Agriculture Market Information System, FAMIS, 2010. “Kenya: PoorLivestock Disease Monitoring Confines Herders to Poverty”,http://famis.comesa.int/com/option.com_news/task.viewarticle/sid.360/Itemid.129/pillar.sps/lang.fr/sectionid./ last accessed Friday 13th August 2010.

19. Global Early Warning and Response System for Major Animal Diseases includingZoonoses, 2010, “Objectives of GLEWS”, http://www.glews.net/index.php?option=com_content&view=article&id=50&Itemid=54 last accessed Friday 13th

August 2010.

20. Global Early Warning and Response System for Major Animal Diseases includingZoonoses, 2010, “Project Background”, http://www.glews.net/index.php?option=com_content&view=article&id=49&Itemid=34 last accessed Friday 13th

August 2010.

21. Leggat P., Goldsmid J., 2001. “Arbovirus”, Dictionary of tropical medicine forHealth professionals, Australia: ACTM.

22. Leggat P., Goldsmid J., 2001. “Arthropod”, Dictionary of tropical medicine forHealth professionals, Australia: ACTM.

23. Leggat P., Goldsmid J., 2001. “Vector”, Dictionary of tropical medicine for Healthprofessionals, Australia: ACTM.

24. World Health Organization Scientific Group, 1967. “Criteria for definition of avirus as an Arbovirus”, W.H.O. Technical Report Series no. 369, Page 9. Geneva.


25. World Health Organization, Rift Valley Fever: Transmission in Humans,http://www.who.int/mediacentre/factsheets/fs207/en/ last accessed Friday 13th

August 2010

26. AVID, 2010. AVID sample data sets, Organisms worksheet.


28. Harper T. K., 2004. “Outbreaks and History of Infection”, TKH Virology Notes:Rift Valley Fever http://www.tarakharper.com/v_rift.htm#out last accessed Friday13th August 2010.

29. World Health Organization Global Alert and Response,http://www.who.int/csr/don/archive/disease/rift_valley_fever/en/ last accessedFriday 13th August 2010


31. "serology". Oxford Dictionaries. April 2010. Oxford Dictionaries. April 2010.Oxford University Press. http://oxforddictionaries.com/definition/serology(accessed August 17, 2010).

32. Enzyme-Linked Immunosorbent Assay (ELISA), Microbial Life, Second Edition.James T. Stanley, Robert P. Gunsalus, Stephen Lory, and Jerome J. Perry.Sunderland, MA: Sinauer Associates, (2007).

33. Linthicum, K., et al. , 1999. Climate and Satellite Indicators to Forecast Rift ValleyFever Epidemics in Kenya, Science 16 Vol. 285, page 397.

34. Low S. P., 2005. “The Impacts of ENSO in Africa”, Climate Change and Africa,Page 221, Cambridge: Cambridge University Press.

35. Linthicum, K., et al. , 1999. Climate and Satellite Indicators to Forecast Rift ValleyFever Epidemics in Kenya , Science 16 Vol. 285, page 398.

36. Ford TE, Colwell RR, Rose JB, Morse SS, Rogers DJ, Yates TL., 2009. “Usingsatellite images of environmental changes to predict infectious disease outbreaks”,Emerging Infectious Diseases, Volume 15, No. 9, Page 1342.

37. Manishika, Jain., 2009. GIS and remote sensing techniques. New Delhi: HimanshuPublications.

38. Wikipedia contributors, "Cartography," Wikipedia, The Free Encyclopaedia,http://en.wikipedia.org/w/index.php?title=Cartography&oldid=360485185(accessed May 12, 2010).


39. Bixby, R. O., 1996. “The Importance of Undergraduate. Graduate, Technical, andApplied GIS and Remote Sensing Education,” Remote sensing and GIS for sitecharacterization: applications and standards, pp 89 – 92. West Conshohocken, PA:American Society for Testing and Materials.

40. Baumann, Paul R., 2001, History of Remote Sensing, Aerial Photography, Dept. ofGeography, State University of New York, College at Oneonta, Oneonta, NewYork,http://employees.oneonta.edu/baumanpr/geosat2/RSHistory/HistoryRSPart1.htm(accessed May 12, 2010)

41. Easa, S., Chan, Y. & American Society of Civil Engineers, 2000. Urban planningand development applications of GIS, pp 33 – 34. USA: American Society of CivilEngineers.

42. El-Rabbany, A., 2002. “Types of GPS Receivers”, Introduction to GPS: the GlobalPositioning System, pp 16 – 19, Norwood, MA: Artech House, Inc.

43. Kolb, L., 2008, Toys to Tools: Connecting Student Cell Phones to Education, page81, USA: International Society for Technology in Education.

44. Easa, S., Chan, Y. & American Society of Civil Engineers, 2000. “Section II: GISApplications”,Urban planning and development applications of GIS, USA:American Society of Civil Engineers.

45. Netzband, Maik; Stefanov, William L.; Redman, Charles (Eds.), 2007. “Chapter 1 -Remote Sensing as a Tool for Urban Planning and Sustainability ” Applied RemoteSensing for Urban Planning, Governance and Sustainability, pp 1 – 18, Berlin:Springer.

46. Gilbert, E.W., 1958. Pioneer maps and health and disease in England. TheGeographical Journal Vol. 124, No. 2. Blackwell Publishing on behalf of The RoyalGeographical Society (with the Institute of British Geographers)

47. Keithley, C., Renton, F., Echavarria, F. & Cantarero, R., 1994. “Mapping the spatialpatterns of health risks among minority groups in Lincoln, Nebraska”, GIS/LIS(1994), pp 477 - 480. USA: American Congress on Surveying and Mapping,American Society for Photogrammetry and Remote Sensing, AM/FM International,the Association of American Geographers, and the Urban and Regional InformationSystems Association

48. Keithley, C., Renton, F., Echavarria, F. & Cantarero, R., 1994. “Mapping thespatial patterns of health risks among minority groups in Lincoln, Nebraska”,GIS/LIS (1994), page 482. USA: American Congress on Surveying and Mapping,American Society for Photogrammetry and Remote Sensing, AM/FM International,the Association of American Geographers, and the Urban and Regional InformationSystems Association


49. Keithley, C., Renton, F., Echavarria, F. & Cantarero, R., 1994. “Mapping the spatialpatterns of health risks among minority groups in Lincoln, Nebraska”, GIS/LIS(1994), page 484. USA: American Congress on Surveying and Mapping, AmericanSociety for Photogrammetry and Remote Sensing, AM/FM International, theAssociation of American Geographers, and the Urban and Regional InformationSystems Association

50. Gamma, E. et al, 1995. “Design Patterns in Smalltalk MVC”, Design patterns:elements of resusable object-oriented software. Page 4. USA: Addison-Wesley.

51. Williams H. & Lane D., 2006. “Chapter 1. Database Applications and the Web”,Web Database Applications with PHP, and MySQL, Second Edition, page 4. USA:O'Reilly Media.

52. Infrastructure for Spatial Information in Europe, website, “About INSPIRE”,http://inspire.jrc.ec.europa.eu/index.cfm/pageid/48 , (accessed May 12, 2010).

53. Infrastructure for Spatial Information in Europe, website, “INSPIRE Roadmap”,http://inspire.jrc.ec.europa.eu/index.cfm/pageid/44, (accessed May 12, 2010).

54. Open Source Geospatial Foundation, “OpenLayers: Home”, http://openlayers.org/,(accessed May 12, 2010).

55. Daratech Inc., 2009. GIS/Geospatial Markets & Opportunities report, August 2009,Cambridge, MA: Daratech Incorporated.

56. Google Inc., “Corporate Information: Google milestones”,http://www.google.com/corporate/history.html, (accessed May 12, 2010).

57. Lerdorf R., Tatroe K. & MacIntyre P., 2006. “Chapter 1. Introduction to PHP”,Programming PHP, Second Edition, page 2. USA: O'Reilly Media.

58. Netcraft.com website, “Market Share for Top Servers Across All DomainsAugust 1995 - August 2010”, August 2010 Web Server Survey,http://news.netcraft.com (last accessed August 23, 2010).

59. Udell S., 2009. “Chapter 1. Introduction to the Geoweb”, Beginning Google Mapsmashups with mapplets, KML and GeoRSS: from novice to Professional, page 6.USA: Apress.

60. Google Incorporated, website, “KML – What is KML?”, KML Documentation,http://code.google.com/apis/kml/documentation/whatiskml.html, (last accessedAugust 23, 2010).

61. Environmental Systems Research Institute, 1998, “ESRI Shapefile TechnicalDescription, An ESRI White Paper” , USA , ESRI.


Appendices

Appendix A: H1N1 Worldwide Incidences


Appendix B: Global Risk Map for Rift Valley Fever


Appendix C: WampServer 2.0i Installation Guide

(Source: BulletProof Templates)

Once you have downloaded the latest version of WAMP 2.0, follows these steps to install it

to your local PC:

1. Locate the downloaded set up file and double-click on it. You will be faced with an

alert window warning you not to upgrade from WAMP5 1.x. Click on Yes to

continue

2. The Welcome setup window will load. Click on Next to proceed


3. On the License Agreement screen select the radio button for I accept the

agreement then click on Next

4. The Select Destination Location screen will load. Change the default location if

you desire then click on Next

5. Now the Select Additional Tasks screen is loaded. Select the checkboxes for any


icons you want installing then click on Next

6. You will be faced with the Ready to Install screen. Review the settings and use the

Back button to go back and change any of the settings. If the settings are correct,

click on Install to install WAMP 2.0


7. If you have Mozilla Firefox installed on your PC, you may be faced with the

following window that prompts whether you want FireFox to be your default

browser, so select the appropriate choice

8. The PHP mail parameters screen will load. Review the default settings and

change accordingly then click on Next. The default values can generally be used

when installing WAMP 2.0 to a local PC

9. The final screen to load is the installation completed screen. Click on Finish to

close the window and start WAMP


10.WAMP 2.0 is started. The WampServer icon is loaded onto the system tray


Appendix D: Installing the Application

The application is provided in the form of a compressed file named “Avid.zip”. In order to

install the application, the necessary files need to be extracted to an appropriate directory

within the web server root.

The following steps should be followed:

1. Make sure the WampServer software is running by clicking on the WampServer

icon in the system tray.

2. Copy the compressed file to the hard disk of the target computer where the

WampServer software is installed.

3. Double-click on the file uncompress the files and select the target directory as

“c:\wamp\www”.


4. Open the “c:\wamp\www” directory using Windows Explorer and locate the folder

“avid”. Open it and locate the file “map.sql”. This contains the information that

needs to be imported into the MySQL database. Move the file onto the desktop

from the “c:\wamp\www” directory.

5. Access the phpMyAdmin interface by opening a web browser window and

accessing the location “http://localhost”. This will bring up the WampServer local

home page. Click on the phpMyAdmin link.

6. The default username does not include a password for the MySQL server. This can

be changed by using the MySQL console under the MySQL option in the

WampServer icon. The console accepts text commands and the following should be

executed to change the password; set password=password(“<password>”); ,

where <password> is replaced with the new password.

7. The file, “phpsqlajax_dbinfo.php” should be edited to contain the new password in

the $password variable before running the application. Failure to do this will

prevent the application from running correctly.

8. Create a new database within MySQL using the create database screen and name it

“map”.


9. Use the MySQL import command to populate the database using the file “map.sql”

that was copied to the desktop.

10. The application can now be run by clicking on the link in the WampServer

homepage or directly by using the URL “http://localhost/avid/”.


Appendix E: Source Codes

The source code for the main application pages is contained here.

E1: avid_test1.php

<?php include 'login_check.php'; ?> <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd"> <html> <head> <title>Listing the AVID database</title> <link rel="stylesheet" href="avid.css" type="text/css" />

<script language="JavaScript" src="ts_picker.js">

//Script by Denis Gritcyuk: [email protected] //Submitted to JavaScript Kit (http://javascriptkit.com) //Visit http://javascriptkit.com for this script

</script> </head>

<body bgcolor="white"> <?php include("header.html");?> <?php include("navbar.html");?> <?php // error_reporting(E_ALL); // ini_set('display_errors', true);

include 'phpsqlajax_dbinfo.php';

// Opens a connection to a MySQL server $connection=mysql_connect ($localhost, $username, $password);

// Connect to the MySQL DBMS if (!($connection = @ mysql_connect($localhost, $username, $password))) die("Could not connect");

// Set the active MySQL database $db_selected = mysql_select_db($database, $connection); if (!$db_selected) { die ('Can\'t use db : ' . mysql_error());


}

// Show all sample records in a <table> function displayAVIDList($connection, $query, $searchValue) { // Run the query on the DBMS if (!($result = @ mysql_query ($query, $connection))) mysql_error( );

// Find out how many rows are available $rowsFound = @ mysql_num_rows($result);

// If the query has results ... if ($rowsFound > 0) {

// Define $color=1 $color="1";

// ... print out a header echo "<p>Results for $searchValue<br>";

// Map link echo '<p>

<form name="goMap" action="mapthis.php" method="POST"> <input type="hidden" name="filters" value="';

echo rawurlencode($query); echo '"><input type="submit" value="Map this result set">

</form> </p>';

// and start a <table>.

echo '<table width="400" border="0" align="left" cellpadding="2"cellspacing="1">';

//Column titles echo '<tr><td>Sample ID</td><td>Location</td><td>Organism</td></tr>'; while($rows=mysql_fetch_array($result)){

// If $color==1 table row color = #FFC600 if($color==1){ echo "<tr bgcolor='#bfd9ff'> <td>".$rows['SampleID']."</td><td>".$rows['VisitID']."</td><td>".

$rows['type']."</td> </tr>"; // Set $color==2, for switching to other color


$color="2"; }

// When $color not equal 1, use this table row color else { echo "<tr bgcolor='#e8e8e8'> <td>".$rows['SampleID']."</td><td>".$rows['VisitID']."</td><td>".

$rows['type']."</td> </tr>"; // Set $color back to 1 $color="1"; }

} // end while loop body

// Finish the <table> echo '</table>';

} // end if $rowsFound body else { // Ask user to check selection

echo "<p>No records were found. Please review your search parameters.</p>"; }

// Report how many rows were found echo "$rowsFound records found matching your criteria.</p>"; } // end of function

$scriptName = "avid_test_1.php"; //Assign search value if (!(@$_POST['Value'] == null)||!(@$_POST['Value'] == "")) $searchValue =

@$_POST['Value'];

// Assign time bounds if (!(@$_POST['timestampFrom'] == null)||!(@$_POST['timestampFrom'] == ""))

$fromValue = @$_POST['timestampFrom']; if (!(@$_POST['timestampTo'] == null)||!(@$_POST['timestampTo'] == ""))

$toValue = @$_POST['timestampTo'];

// Assign options for drop down $sql="SELECT id, name FROM pathogens"; $result=mysql_query($sql);

$options="";

while ($row=mysql_fetch_array($result)) {

$id=$row["name"];


$name=$row["name"]; $options.="<OPTION VALUE=\"$id\">".$name; }

//Assign disease value if (!(@$_POST['pathogens'] == null)||!(@$_POST['pathogens'] == "")) $disease =

@$_POST['pathogens'];

// Create and Display Search form ?> <h2>AVID Sample Collection Results</h2>

<p><form name="Samples" action="<?php $scriptName;?>" method="POST"> <table cellspacing="2" border="1"><tr><TH COLSPAN=2>AVID DatabaseSample Search</TH> </tr>

<tr><td>Enter a search term :<br>(type All to display complete dataset)</td> <td><input type="text" name="Value" value="All" size="10"><br> </td></tr>

<tr><TH COLSPAN=2>Select dates using the calendar icons:</TH> </tr> <tr><td>From: </td><td><input type="Text" name="timestampFrom" value=""

readonly="true"> <a href="javascript:show_calendar('document.Samples.timestampFrom',document.Samples.timestampFrom.value);"><img src="cal.gif" width="16" height="16"border="0" alt="Click Here to Pick up the timestamp"></a></td></tr>

<tr><td>To: </td><td><input type="Text" name="timestampTo" value=""readonly="true"> <a href="javascript:show_calendar('document.Samples.timestampTo',document.Samples.timestampTo.value);"><img src="cal.gif" width="16" height="16"border="0" alt="Click Here to Pick up the timestamp"></a></td></tr>

<tr><td>Please select a pathogen from the list below: <br> <SELECT NAME=pathogens> <OPTION VALUE=0>Choose <?=$options?> </SELECT> </td><td>

<input type="submit" value="Go"></td></tr></table> </form><br></p> <?php // Has the user provided the parameter?

if (empty($searchValue)) { // No, the user hasn't provided a parameter or initial page load

// Do nothing

} // end of if empty($searchValue) body

else


{ // Start a query ... $query = "Select SampleID,

`VisitID`, `VisitDate`, `Lat`, `Long`, `name` AS 'type'

from àvidresults1`";

// ... then, if the user has specified a search term, // add the searchValue as a WHERE clause ... if ($searchValue != "All") {$query .= " WHERE SampleID Like '$searchValue%'";}

else {$query .= " WHERE 1";}

// ... then, if the user has specified a "From" date term, // add the timestampFrom field as a WHERE clause ... if ($fromValue != "") $query .= " AND VisitDate >= '$fromValue'";

// ... then, if the user has specified a "To" date term, // add the timestampFrom field as a WHERE clause ... if ($toValue != "") $query .= " AND VisitDate <= '$toValue'";

// ... then, if the user has selected a pathogen, // add the selection value as a WHERE clause ... if ($disease != "0") $query .= " AND `$disease` IS TRUE";

// ... and then complete the query. $query .= " ORDER BY VisitDate LIMIT 0, 100";

// run the query and show the results displayAVIDList($connection, $query, $searchValue);

// Close the DBMS connection mysql_close($connection); } // end of else if empty($searchValue) body ?> </body> </html>


E2: mapthis.php

<?php include 'login_check.php'; ?> <html xmlns="http://www.w3.org/1999/xhtml"> <head> <meta http-equiv="content-type" content="text/html; charset=utf-8"/> <link rel="stylesheet" href="avid.css" type="text/css" /> <title>AVID GIS concept version 1.1</title> <script src="http://maps.google.com/maps/api/js?sensor=false" type="text/javascript"></script> <script type="text/javascript"> //<![CDATA[

//Global map variable var map;

//Array that holds overlays var kmlArray = [];

// Array for panning var pointArray = [];

// Panning Control var panIntervalId = 0; var last = 0;

var customIcons = { Sheep: { icon: 'images/symbol_blue_s.png', shadow: 'http://labs.google.com/ridefinder/images/mm_20_shadow.png' }, Goat: { icon: 'images/symbol_red_s.png', shadow: 'http://labs.google.com/ridefinder/images/mm_20_shadow.png' } };

function load() { map = new google.maps.Map(document.getElementById("map"), { center: new google.maps.LatLng(-1.5, 38.12), zoom: 6, mapTypeId: 'terrain' }); var infoWindow = new google.maps.InfoWindow;

// Variable for dynamic querying var postVars = "<?php echo $_POST['filters'];?>;"


// Change this depending on the name of your PHP file downloadUrl("avid_genxml.php"+"?option="+ postVars, function(data) { var xml = data.responseXML; var markers = xml.documentElement.getElementsByTagName("marker"); for (var i = 0; i < markers.length; i++) { var name = markers[i].getAttribute("name"); var address = markers[i].getAttribute("address"); var type = markers[i].getAttribute("type"); var point = new google.maps.LatLng( parseFloat(markers[i].getAttribute("lat")), parseFloat(markers[i].getAttribute("lng"))); var html = '<b><a href="details.php?sample=' + name + '">' + name +'</a></b><br>' + address + "<br>" + type; var icon = customIcons[type] || {}; var marker = new google.maps.Marker({ map: map, position: point, icon: icon.icon,

shadow: icon.shadow }); bindInfoWindow(marker, map, infoWindow, html);

pointArray.push(marker); } }); }

function bindInfoWindow(marker, map, infoWindow, html) { google.maps.event.addListener(marker, 'click', function() { infoWindow.setContent(html); infoWindow.open(map, marker); }); }

function downloadUrl(url, callback) { var request = window.ActiveXObject ? new ActiveXObject('Microsoft.XMLHTTP') : new XMLHttpRequest;

request.onreadystatechange = function() { if (request.readyState == 4) { request.onreadystatechange = doNothing; callback(request, request.status); } };

request.open('GET', url, true); request.send(null); }


function doNothing() {}

// Shows rain overlay function showRainOverlay() { var rainfallLayer = newgoogle.maps.KmlLayer('http://sites.google.com/site/eaoverlays/home/kenya_rainfall_distribution.kml'); rainfallLayer.setMap(map);

kmlArray.push(rainfallLayer); }

// Shows tsetse overlay function showTsetseOverlay() { var tsetseLayer = newgoogle.maps.KmlLayer('http://sites.google.com/site/eaoverlays/home/kenya_tsetsedistribn.kml'); tsetseLayer.setMap(map);

kmlArray.push(tsetseLayer); }

// Shows cattle overlay function showCattleOverlay() { var cattleLayer = newgoogle.maps.KmlLayer('http://sites.google.com/site/eaoverlays/home/kenya_cattledensity.kml'); cattleLayer.setMap(map);

kmlArray.push(cattleLayer); }

// Shows Custom overlay function showCustomOverlay(form) {

var layer = form.url.value; var customLayer = new google.maps.KmlLayer(layer);

customLayer.setMap(map); kmlArray.push(customLayer);

}

// Removes the overlays from the map and clears the array function clearOverlays() { if (kmlArray) { for (i in kmlArray) { kmlArray[i].setMap(null); } } kmlArray.length = 0; }

// function iterates through the different points in the array


function moveTo() { if (last < pointArray.length) {

pointArray[last].setMap(map); last++;

} else // We have reached the end {

//Reset value of pointer and stop last = 0; clearIntervalId(panIntervalId);

} }

// Panning function function executePan() {

var pointer = 0;

// Clear map first while (pointer < pointArray.length){

pointArray[pointer].setMap(null); pointer++;

}

// Set zoom and call pan function map.setZoom(7); panIntervalId = setInterval("moveTo()",2000);

}

//]]> </script> </head>

<body onload="load()"> <?php include("header.html");?> <div id="overlays"> <input onclick="showRainOverlay();" type=button value="Rainfall distribution"/> <input onclick="showTsetseOverlay();" type=button value="Tseste distribution"/> <input onclick="showCattleOverlay();" type=button value="Cattle distribution"/> <input onclick="clearOverlays();" type=button value="Clear Overlays"/> <br/> <form action="#" name="overlay">

<input name="url" type="text"> <input onclick="showCustomOverlay(this.form);" type="button" value="Customoverlay"> </form>


</div>

<div id="animate"> <input onclick="executePan();" type=button value="Show sequence"/> </div>

<div id="map" style="width: 800px; height: 600px" border="1"></div> </body> </html>


E3: avid_genxml.php

<?php include 'login_check.php'; require("phpsqlajax_dbinfo.php");

function parseToXML($htmlStr) { $xmlStr=str_replace('<','<',$htmlStr); $xmlStr=str_replace('>','>',$xmlStr); $xmlStr=str_replace('"','"',$xmlStr); $xmlStr=str_replace("'",''',$xmlStr); $xmlStr=str_replace("&",'&',$xmlStr); return $xmlStr; }

// Opens a connection to a MySQL server $connection=mysql_connect ($localhost, $username, $password); if (!$connection) { die('Not connected : ' . mysql_error()); }

// Set the active MySQL database $db_selected = mysql_select_db($database, $connection); if (!$db_selected) { die ('Can\'t use db : ' . mysql_error()); }

// Use get value for query $query= rawurldecode($_GET['option']);

// Select the matching rows according to the query $result = mysql_query($query); if (!$result) { die('Invalid query: ' . mysql_error()); }

header("Content-type: text/xml");

// Start XML file, echo parent node echo '<markers>';

// Iterate through the rows, printing XML nodes for each while ($row = @mysql_fetch_assoc($result)){ // ADD TO XML DOCUMENT NODE echo '<marker '; echo 'name="' . parseToXML($row['SampleID']) . '" '; echo 'address="' . parseToXML($row['VisitID']) . '" '; echo 'lat="' . $row['Lat'] . '" '; echo 'lng="' . $row['Long'] . '" ';


echo 'type="' . $row['type'] . '" '; echo '/>'; }

// End XML file echo '</markers>'; // Close the DBMS connection mysql_close($connection); ?>E4: details.php

<?php include 'login_check.php'; ?> <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd"> <html> <head> <title>Sample Storage Details</title> <link rel="stylesheet" href="avid.css" type="text/css" /> </head> <?php include("header.html");?> <?php include("navbar.html");?> <body bgcolor="white"> <?php

include 'phpsqlajax_dbinfo.php';

if (!($_GET['sample'] == null)||!($_GET['sample'] == "")){

// Opens a connection to a MySQL server $connection=mysql_connect ($localhost, $username, $password);

// Connect to the MySQL DBMS if (!($connection = @ mysql_connect($localhost, $username, $password))) die("Could not connect");

// Set the active MySQL database $db_selected = mysql_select_db($database, $connection); if (!$db_selected) { die ('Can\'t use db : ' . mysql_error()); } // Start a query ... $searchterm = $_GET['sample'];

$query = "Select `SampleID`, Òrigin`, Ànimal_ID`, `Sample Type`, `Tray_ID`,`Box_ID`, `Box_Details` from `storage2` where SampleID = \"$searchterm\"";


// Run the query on the DBMS if (!($result = @ mysql_query ($query, $connection))) mysql_error( );

// Output a heading echo '<h1>Details</h1>'; // and start a <table>. echo '<table width="400" border="1" cellpadding="2" cellspacing="1">'; //Populate it while($rows=mysql_fetch_array($result)){ echo '<tr><TH COLSPAN=2>AVID Sample Storage details</TH> </tr>'; echo '<tr><td>Sample ID</td><td>'.$rows['SampleID'].'</td></tr>'; echo "<tr><td>Project</td><td>".$rows['Origin']."</td></tr>"; echo "<tr><td>Animal ID</td><td>".$rows['Animal_ID']."</td></tr>"; echo "<tr><td>Sample Type</td><td>".$rows['Sample Type']."</td></tr>"; echo "<tr><td>Tray ID</td><td>".$rows['Tray_ID']."</td></tr>"; echo "<tr><td>Box ID</td><td>".$rows['Box_ID']."</td></tr>"; echo "<tr><td>Location in box</td><td>".$rows['Box_Details']."</td></tr>"; }

// Finish the <table> echo '</table>';

// Close the DBMS connection mysql_close($connection);

echo '<p><a href="avid_test_1.php">Click here</a> to perform a new search.</p>'; } else {

// Do nothing echo "<h2>oops ... something went wrong <br> Please check your referring

page.</h2>"; }// end of else

?> </body> </html>


GIS SUPPORT FOR EMERGING INFECTIOUS DISEASES IN EAST ... · GIS support for emerging infectious diseases in East Africa 9. Emerging infectious diseases cause significant health and

Documents