Facilitating fine-grained intra-urban dengue forecasting by integrating urban environments measured from street-view images

Abstract

Background: Dengue fever (DF) is a mosquito-borne infectious disease that has threatened tropical and subtropical regions in recent decades. An early and targeted warning of a dengue epidemic is important for vector control. Current studies have primarily determined weather conditions to be the main factor for dengue forecasting, thereby neglecting that environmental suitability for mosquito breeding is also an important factor, especially in fine-grained intra-urban settings. Considering that street-view images are promising for depicting physical environments, this study proposes a framework for facilitating fine-grained intra-urban dengue forecasting by integrating the urban environments measured from street-view images.

Methods: The dengue epidemic that occurred in 167 townships of Guangzhou City, China, between 2015 and 2019 was taken as a study case. First, feature vectors of street-view images acquired inside each township were extracted by a pre-trained convolutional neural network, and then aggregated as an environmental feature vector of the township. Thus, townships with similar physical settings would exhibit similar environmental features. Second, the environmental feature vector is combined with commonly used features (e.g., temperature, rainfall, and past case count) as inputs to machine-learning models for weekly dengue forecasting.

Results: The performance of machine-learning forecasting models (i.e., MLP and SVM) integrated with and without environmental features were compared. This indicates that models integrating environmental features can identify high-risk urban units across the city more precisely than those using common features alone. In addition, the top 30% of high-risk townships predicted by our proposed methods can capture approximately 50-60% of dengue cases across the city.

Conclusions: Incorporating local environments measured from street view images is effective in facilitating fine-grained intra-urban dengue forecasting, which is beneficial for conducting spatially precise dengue prevention and control.

Keywords: Dengue forecasting; Fine-grained; Intra-urban; Street-view image; Urban environment.

Fig. 1

Study area. All 167 townships in Guangzhou City, China

Fig. 2

Spatial distribution of dengue cases in Guangzhou City, China between January 2015 and September 2019

Fig. 3

Weekly dengue case count of Guangzhou City from January 2015 to September 2019

Fig. 4

Meteorological data of Guangzhou City. A Weekly mean temperature and B weekly cumulative rainfall of a township from January 2015 to September 2019. C Weekly mean temperature and D weekly cumulative rainfall of all townships within Guangzhou City during the week of September 12–18, 2016

Fig. 5

Population data (2017) of Guangzhou City. A 100-m gridded population counts provided by the WorldPop. B Township-based population aggregated from the 100 m-grid population counts

Fig. 6

Street-view images acquired at a location with heading degrees of 0, 90, 180, and 270

Fig. 7

Framework of the proposed approach

Fig. 8

Extracting the environmental feature vector of an urban unit from street-view images using the pre-trained PSPNet model

Fig. 9

Cosine similarity matrix of four street-view images measured by their feature vectors extracted from the pre-trained PSPNet model

Fig. 10

Time series of weekly local case count before and after data-smoothing

Fig. 11

Temporal distribution of predicted case counts of three townships

Fig. 12

Predicted (smoothed) case counts of all townships during two different weeks. The smoothed case counts were predicted by the 1-week ahead MLP-based model using both common and environmental features

Fig. 13

Forecasting performance evaluated by Pearson correlation coefficient. a MLP-based forecasting. b SVM-based forecasting

Fig. 14

Forecasting performance of MLP-based models evaluated by the hit rate metric

Fig. 15

Forecasting performance of SVM-based models evaluated by the hit rate metric