Newton001 A Software Infrastructure for Promoting Efficient Entomological Monitoring of Dengue Fever

Lead Research Organisation: Newcastle University
Department Name: Sch of Computer Science

Abstract

The activities primarily aim at reducing problems of poverty by improving the health of Brazilian communities. The development of an advanced and dependable data analytics infrastructure will enable the effective and efficient monitoring of Dengue fever outbreaks in Brazil. Dengue is an endemic problem in many Brazilian areas where public health assistance is inefficient. The Brazilian public health system cannot meet the demands of these areas due to the resources scarcity, which calls for developing trustworthy mechanisms to support efficiently decisions about the areas that have priority on resource allocation. This software infrastructure can be reused in the future in other developing countries suffering from Dengue outbreaks. The monitoring infrastructure will provide its data analytics capabilities through a unique software architecture that can be used through several mobile platforms and social networks. This infrastructure can only be developed through collaborative research with the leading UK researchers.

Technical Summary

The first planned activity of this project is to understand the requirements needed for the identification and control of dengue outbreaks. To identify dengue outbreaks, it is important to include the population as an entomological agent as a means to reveal environments favourable to mosquitoes breeding. The goal is to retrieve and control different sources of data to characterize regions that seem to be sources of dengue outbreaks.

Another goal is to gather additional data that must be generated and passed on to agencies and systems of the national health system, such as the National System for Surveillance and Control of Diseases. This first activity will produce a document describing requirements needed for the identification of dengue outbreaks.

The third planned activity is to provide a software platform where the population can act as an etymological agent, assisting in the surveillance and detection of dengue outbreaks. The population will use this platform to provide data from their communities to feed the data model of the public health agencies.

The fourth activity is to actually implement the software product line designed in the third activity. The goal is to provide software products that the population can use as a platform to support entomological surveillance of dengue.

The fifth activity is to deploy the products of the implemented SPL in an experimental environment. The goal is to analyse their performance and their acceptance, allowing refining and restructuring the SPL design according to the obtained results.

The sixth activity is to analyse the public data submitted by the population through the proposed platform in the experimental environments. The goal is to refine the models and algorithms proposed in the second activity.

The seventh and final activity is to design, deploy and monitor a pilot project during 6 months.

Planned Impact

N/A
 
Description Improvement of the Advanced MSc module on Large Data Analytics at Newcastle University
Geographic Reach National 
Policy Influence Type Influenced training of practitioners or researchers
Impact The substantial improvement of the MSc module consisted in integrating the knowledge gained in the project about how the users of social media react to the occurrences of dengue and zika, as well as about the advanced methods developed for filtering and classifying these type of big data.
 
Description Newton Fund Grant Institutional Links
Amount £99,926 (GBP)
Funding ID 273018638 
Organisation British Council 
Sector Charity/Non Profit
Country United Kingdom
Start 04/2017 
End 03/2019
 
Description Cooperation focusing on improvement of Twitter user ranking 
Organisation Fluminense Federal University
PI Contribution Newcastle team coordinates and leads this joint work on improving user ranking, focusing on learning a classifier for users, based on the collection of tweets sent by each user.
Collaborator Contribution FFU researchers brought into this join work a deep knowledge of natural language analytics.
Impact Only initial but promising results by now. The joint work with UFF continues.
Start Year 2017
 
Description Cooperation on preparation of a new EPSRC proposal to build on the work developed in this project 
Organisation Lancaster University
Country United Kingdom 
Sector Academic/University 
PI Contribution We have developed a novel approach to structuring complex fault tolerance in large systems: Holistic Fault Tolerance. It serves as a foundation of the new EPSRC proposal that we are developing now with the partner.
Collaborator Contribution The partner has developed new ideas of applying the holistic fault tolerance for the cloud computing.
Impact Too early to say
Start Year 2017
 
Description Evaluation of the project progress and feedback on the project work 
Organisation Fiocruz - Research Center Renê Rachou
Country Brazil 
Sector Public 
PI Contribution Our technical work on developing A Software Infrastructure for Promoting Efficient Entomological Monitoring of Dengue Fever.
Collaborator Contribution Regular feedback on the project work and regular updates on the current state of work in FioCruz and the current situation with the dengue and zika in Brazil.
Impact This partnership allowed us to improve our platform as well as to find the right dissemination and exploitation channels for the platform to be developed in the project.
Start Year 2015
 
Description Feedback on the design of the VazaDengue platform 
Organisation Fluminense Federal University
PI Contribution The project team has been improving the platform using the feedback from UFF.
Collaborator Contribution The person we met had a wide expertise in developing Dengue-related apps - she was involved in a number of projects on the topic. She provided a very useful feedback on the design of our platform (both the portal and the apps).
Impact The main outcome of this collaboration has been in developing an improved version of the portal and the app.
Start Year 2016
 
Description Feedback on the design of the project platform (both the portal and the apps) 
Organisation Federal University of Rio de Janeiro
Country Brazil 
Sector Academic/University 
PI Contribution The feedback allowed us to further improve the infrastructure we are developing in the project
Collaborator Contribution The professor cooperating with us is an expert on telemedicine and the coordinator of a network on excellence on telemedicine in Brazil. He is also an expert on studying various diseases in poorly developed areas in Brazil. He provided a very useful feedback on the design of our platform (both the portal and the apps)
Impact An improved version of the infrastructure that takes into account the needs of various stakeholders.
Start Year 2016
 
Description Improvement of the classification of the dengue tweets 
Organisation Fluminense Federal University
PI Contribution We have developed the main approach for harvesting and filtering tweets. And an initial classification algorithm.
Collaborator Contribution Our classification algorithm was improved by the partner.
Impact We produced a high-quality joint paper: Paolo Missier, Callum McClean, Jonathan Carlton, Diego Cedrim, Leonardo da Silva Sousa, Alessandro F. Garcia, Alexandre Plastino, Alexander Romanovsky: Recruiting from the Network: Discovering Twitter Users Who Can Help Combat Zika Epidemics. ICWE 2017 - Web Engineering - 17th International Conference, ICWE 2017, Rome, Italy, June 5-8, 2017, Proceedings, pp.437-445 This work contributed to the development of our Portal that presents the results of the analysis to the public and health professionals
Start Year 2016
 
Description Leveraging Gamification and Social Networks for Improving Prevention and Control of Zika 
Organisation Federal University of Alagoas
Country Brazil 
Sector Academic/University 
PI Contribution Development of the advanced techniques for tweet processing, filtering, classification and for user ranking.
Collaborator Contribution Leveraging Gamification to improve the engagement of the health organization, health professionals and the public into combating zika by using social networks.
Impact A successful joint proposal to the BC Institutional Links grant scheme.
Start Year 2016
 
Description Providing tweet datasets for athen initial study 
Organisation Federal University of Minas Gerais
Country Brazil 
Sector Academic/University 
PI Contribution Our team worked with these datasets to improve the classifiers we develop in the porject
Collaborator Contribution The researchers from UFMG provided us with tweet datasets so that we could perform an initial study of the first version of our tweet classifier, when we were starting our project.
Impact These data sets allowed us to improve our method and understand better the way Twitter delivers data on request.
Start Year 2015
 
Description The VazaDengue system for early identification of potential Dengue outbreaks 
Organisation Pontifical Catholic University of Rio de Janeiro
Country Brazil 
Sector Academic/University 
PI Contribution We contribute into the design of the overall architecture of the system. We develop the classification algorithms and an innovative supervised algorithm for continually mining the relevant content.
Collaborator Contribution The development, the integration and the deployment of the system and the app. The subsystems that collect tweets and publish the outcomes of the analysis.
Impact The VazaDengue system: - mines the relevant content about Dengue posted in Portuguese on social networks, - classifies it in one of the following categories: suspected cases of the disease, mosquito focus, news, and jokes, - publishes results in dynamic maps to allows its users to report mosquito focus and suspicious disease. Such published content can be useful to support the work of different categories of health professionals.
Start Year 2015
 
Title The VazaDengue system for early identification of potential Dengue outbreaks 
Description The system consists of a web portal and a mobile app. The portal collects the twits, categorized them and ranks the Twitter users to identify the social sensors. The users of the portal can see on the map the uptodate information about the relevant twits, their location, hitmaps, etc. The portal is connected to the apps that collect the information provided by the users about dengue focuses and pass it it the portal, and allow the users to see on the map the uptodate information about the relevant twits, their location, hitmaps, etc. The portal and the apps are used to engage the social sensors identified. We developed a web portal and a mobile app for Android and iOS. The app is distributed as open source and open distribution with the GPL licence. 
Type Of Technology Webtool/Application 
Year Produced 2016 
Impact Nothing to report now 
URL http://www.vazadengue.inf.puc-rio.br/
 
Title The VazaZika application for combating zika 
Description The VazaZika application is available online at http://www.vazazika.com.br It has been developed in cooperation with health professionals  from the Municipal Health Secretariat of Maceió An initial version of the platform is available and is in experimental use by a group of PG students in 4 partner universities in Brazil and by health professionals. This is a new, very much re-worked version of the application for the mobile phones. It supports a number of gamification strategies for involving users. 
Type Of Technology Webtool/Application 
Year Produced 2018 
Impact Too early to say 
URL http://www.vazazika.com.br
 
Description A working meeting with the health professionals in Maceio, Alagoas. 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact We organised a major project meeting in Maceió (UFAL) on April 10-12, 2018. We had 15 attendees from the project partners (UFAL, UFPE, PUC Rio, Newcastle U.) and three health professionals from Municipal Health Secretariat of Maceió (the Department of Control of Vectors and Diseases Transmitted by Poisonous Animals, Municipal Health Secretariat of Maceió):
-Carmem Samico: Coordinator of Control of Vectors and Diseases Transmitted by Poisonous Animals, Municipal Health Secretariat of Maceió;
-Charles Nunes: Veterinarian of the Municipal Health Secretariat of Maceió
-Gilson Vieira: Coordinator of Aedes aegypti Control Program

During this three day meeting we
-discussed with the experts the project and the approaches we are developing
-evaluated various strategies for disseminating and exploiting the project results
-received feedback on the usability of our platform/application
-discussed the ways to approach educational authorities in Alagoes to involve students
-clarified the ways the authorities in the State combat zika and conduct the campaigns
Year(s) Of Engagement Activity 2018
 
Description Engagement with the Brazilian Community of Health Agents 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact In June 2017 a description of the VazaDengue platform was published as a news in the Facebook group: Brazilian Community of Health Agents, that includes the Epidemic Control Agents and Community Health Agents.
Year(s) Of Engagement Activity 2017
 
Description Meeting with the Head of the Health Department of Rio de Janeiro, Felipe Peixoto 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Professional Practitioners
Results and Impact The results of the ongoing work on the Newton project were presented. The followup discussions allowed us to answer various questions, to show demos and to discuss possible exploitation channels. This meeting was followed by two telephone calls from the Health Department of Rio de Janeiro to discuss concrete exploitation options.
Year(s) Of Engagement Activity 2016
 
Description Two talks given to Brazilian Academy of Science (ABC - Academia Brasileira de Ciência) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Other audiences
Results and Impact Due to the interdisciplinary nature of our project (computer science, health/medical research, social engagement/social media), these were important meeting to raise the awareness of the Brazilian Academy of Science in our work on the project and to engage into discussions with the experts in different subjects.

In these meetings, there were two health department employees that were very enthusiastic about collaborating with us and integrating our system with the Health Department of Rio de Janeiro
Year(s) Of Engagement Activity 2016