Web platform using digital image processing and geographic information system tools: a Brazilian case study on dengue

Abstract

BackgroundDengue fever is endemic in Asia, the Americas, the East of the Mediterranean and the Western Pacific. According to the World Health Organization, it is one of the diseases of greatest impact on health, affecting millions of people each year worldwide. A fast detection of increases in populations of the transmitting vector, the Aedes aegypti mosquito, is essential to avoid dengue outbreaks. Unfortunately, in several countries, such as Brazil, the current methods for detecting populations changes and disseminating this information are too slow to allow efficient allocation of resources to fight outbreaks. To reduce the delay in providing the information regarding A. aegypti population changes, we propose, develop, and evaluate a system for counting the eggs found in special traps and to provide the collected data using a web structure with geographical location resources.MethodsOne of the most useful tools for the detection and surveillance of arthropods is the ovitrap, a special trap built to collect the mosquito eggs. This allows for an egg counting process, which is still usually performed manually, in countries such as Brazil. We implement and evaluate a novel system for automatically counting the eggs found in the ovitraps’ cardboards. The system we propose is based on digital image processing (DIP) techniques, as well as a Web based Semi-Automatic Counting System (SCSA-WEB). All data collected are geographically referenced in a geographic information system (GIS) and made available on a Web platform. The work was developed in Gama’s administrative region, in Brasília/Brazil, with the aid of the Environmental Surveillance Directory (DIVAL-Gama) and Brasília’s Board of Health (SSDF), in partnership with the University of Brasília (UnB). The system was built based on a field survey carried out during three months and provided by health professionals. These professionals provided 84 cardboards from 84 ovitraps, sized 15 × 5 cm. In developing the system, we conducted the following steps:Obtain images from the eggs on an ovitrap’s cardboards, with a microscope.Apply a proposed image-processing-based semi-automatic counting system. The system we developed uses the Java programming language and the Java Server Faces technology. This is a framework suite for web applications development. This approach will allow a simple migration to any Operating System platform and future applications on mobile devices.Collect and store all data into a Database (DB) and then georeference them in a GIS. The Database Management System used to develop the DB is based on PostgreSQL. The GIS will assist in the visualization and spatial analysis of digital maps, allowing the location of Dengue outbreaks in the region of study. This will also facilitate the planning, analysis, and evaluation of temporal and spatial epidemiology, as required by the Brazilian Health Care Control Center.Deploy the SCSA-WEB, DB and GIS on a single Web platform.ResultsThe statistical results obtained by DIP were satisfactory when compared with the SCS...

Influence of shading and pedestrian traffic on the preference ofAedes (Stegomyia) aegypti(Diptera: Culicidae) for oviposition microenvironments

“…Its capacity for transmission is due to its rapid adaptation to different environmental contexts. This plasticity is essential for initiating epidemics of major demographic magnitudes, especially in tropical regions (Costa et al , Brasil et al ). In this context, deforestation and urbanization stimulate the synanthropy of Ae.…”

Rapid Surveillance for Vector Presence (RSVP): Development of a novel system for detecting Aedes aegypti and Aedes albopictus

“…Egg counting software has the potential to remove the need for manual counting and improve the accuracy and consistency of subjective rapid visual estimates [68]. We demonstrated that systems developed to count eggs harvested under laboratory conditions [69, 70] may not be suitable for rapid analyses of field collections due to greater exposure of ovistrips to contaminants.…”

Uma revisão das técnicas computacionais para contagem de ovos de Aedes aegypti em imagens de ovitrampas

“…O processo quando realizado desta forma é bastante cansativo e sujeito a falhas, uma vez que demanda muito esforço visual contínuo do responsável. Além disso, como citado por Brasil et al (2015), pode haver ainda variações nas leituras, quando feitas por diferentes profissionais ou quando há sobreposição de ovos.…”

“…Em geral, uma ovitrampa tem cerca de 12 cm de profundidade com uma abertura circular com 5 cm de raio. Os pesquisadores sugerem ainda a adoção de cerca de 200 ml de líquido por armadilha (Brasil et al, 2015).…”

“…Elas são indicadas para regiões com mão de obra disponível, uma vez que deverão ser vistoriadas e substituídas, impreterivelmente no período que antecede à eclosão dos ovos, sob pena de contribuir para sua proliferação. Porém, quando bem monitoradas, segundo Brasil et al (2015), elas viabilizam identificar com antecedência possíveis infestações.…”