Large scale detailed mapping of dengue vector breeding sites using street view images

Abstract

Targeted environmental and ecosystem management remain crucial in control of dengue. However, providing detailed environmental information on a large scale to effectively target dengue control efforts remains a challenge. An important piece of such information is the extent of the presence of potential dengue vector breeding sites, which consist primarily of open containers such as ceramic jars, buckets, old tires, and flowerpots. In this paper we present the design and implementation of a pipeline to detect outdoor open containers which constitute potential dengue vector breeding sites from geotagged images and to create highly detailed container density maps at unprecedented scale. We implement the approach using Google Street View images which have the advantage of broad coverage and of often being two to three years old which allows correlation analyses of container counts against historical data from manual surveys. Containers comprising eight of the most common breeding sites are detected in the images using convolutional neural network transfer learning. Over a test set of images the object recognition algorithm has an accuracy of 0.91 in terms of F-score. Container density counts are generated and displayed on a decision support dashboard. Analyses of the approach are carried out over three provinces in Thailand. The container counts obtained agree well with container counts from available manual surveys. Multi-variate linear regression relating densities of the eight container types to larval survey data shows good prediction of larval index values with an R-squared of 0.674. To delineate conditions under which the container density counts are indicative of larval counts, a number of factors affecting correlation with larval survey data are analyzed. We conclude that creation of container density maps from geotagged images is a promising approach to providing detailed risk maps at large scale.

Added-value of mosquito vector breeding sites from street view images in the risk mapping of dengue incidence in Thailand

“…The frequency of occurrence and the suitability of containers as breeding sites vary, with ceramic containers generally more suitable than plastic containers. Haddawy et al [ 4 ] detected outdoor open containers which constitute potential dengue vector breeding sites from geotagged Google street view (GSV) images using convolutional neural networks. In this study, we make use of their dataset and provide here a brief description of their approach and their data.…”

“…As noted by Campbell et al [ 37 ], seasonal cycles of dengue disease are observed in every province in Thailand and the public health officials are familiar with the seasonal projections of disease. Besides, we wish to relate the analyses in this study to the results from the previous work [ 4 ] which examined the seasonal correlation between container densities and the Breteau index and found a strong correlation during the dengue season. The distribution of Aedes vectors was found to be influenced seasonally by breeding outdoors rather than indoors in a study in Thailand [ 16 ].…”

“…We, therefore, sought to investigate this empirically. Using the container density values from Haddawy et al [4] and four years of dengue incidence data, we employ a risk (https://www.volkswagenstiftung.de/en/ foundation) through a Lichtenberg professorship for JS and conducted within the Bremen Spatial Cognition Center (BSCC), Germany. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.…”

“…Haddawy et al [ 4 ] presented a novel approach to detect outdoor open containers that constitute potential dengue vector breeding sites in geotagged images and demonstrated the approach on Google street view (GSV) images. Eight of the most common containers are detected in the images using a convolutional neural network.…”

“…We, therefore, sought to investigate this empirically. Using the container density values from Haddawy et al [ 4 ] and four years of dengue incidence data, we employ a risk prediction approach to determine the added value of container densities obtained from GSV images in predictive models for three provinces in Thailand. We develop Linear Mixed Effects Models (LMER) at the sub-district level, and along with the container density, population data, and metrological covariates are included in the models.…”