RESUMO
BACKGROUND: Spatiotemporal dengue forecasting using machine learning (ML) can contribute to the development of prevention and control strategies for impending dengue outbreaks. However, training data for dengue incidence may be inflated with frequent zero values because of the rarity of cases, which lowers the prediction accuracy. This study aimed to understand the influence of spatiotemporal resolutions of data on the accuracy of dengue incidence prediction using ML models, to understand how the influence of spatiotemporal resolution differs between quantitative and qualitative predictions of dengue incidence, and to improve the accuracy of dengue incidence prediction with zero-inflated data. METHODOLOGY: We predicted dengue incidence at six spatiotemporal resolutions and compared their prediction accuracy. Six ML algorithms were compared: generalized additive models, random forests, conditional inference forest, artificial neural networks, support vector machines and regression, and extreme gradient boosting. Data from 2009 to 2012 were used for training, and data from 2013 were used for model validation with quantitative and qualitative dengue variables. To address the inaccuracy in the quantitative prediction of dengue incidence due to zero-inflated data at fine spatiotemporal scales, we developed a hybrid approach in which the second-stage quantitative prediction is performed only when/where the first-stage qualitative model predicts the occurrence of dengue cases. PRINCIPAL FINDINGS: At higher resolutions, the dengue incidence data were zero-inflated, which was insufficient for quantitative pattern extraction of relationships between dengue incidence and environmental variables by ML. Qualitative models, used as binary variables, eased the effect of data distribution. Our novel hybrid approach of combining qualitative and quantitative predictions demonstrated high potential for predicting zero-inflated or rare phenomena, such as dengue. SIGNIFICANCE: Our research contributes valuable insights to the field of spatiotemporal dengue prediction and provides an alternative solution to enhance prediction accuracy in zero-inflated data where hurdle or zero-inflated models cannot be applied.
RESUMO
Background: Insecticide-treated mosquito bed nets and indoor residual spraying are widely used for malaria vector control. However, their effectiveness can be affected by household members' habits, requiring alternative approaches toward malaria vector control. Objective: To assess the effectiveness of modified houses in preventing mosquito entry; to assess the impact of house modifications on indoor air conditions and evaluate the acceptability of modified houses in the community where the study was conducted. Methods: Five traditional and five modified houses were constructed in Nampula district, Mozambique and underwent a 90-day overnight indoor mosquito collection using Centers for Disease Control and nitride ultraviolet light traps during the rainy season. Mosquitoes were identified morphologically. Indoor temperature, relative humidity, carbon dioxide levels and wind speed were also collected. The Student's t-test was used to compare the means of the number of mosquitos and environmental factors between both house types. A binomial form of the Generalized Linear Model identified the factors associated with the community volunteer's preference for house type. Results: Modified houses reduced the number of Anopheles by an average of 14.97 mosquitos (95% CI, 11.38-18.56, p < 0.000) and non-Anopheles by 16.66 mosquitoes (95% CI, 8.23-25.09, p < 0.000). Although fewer mosquitoes were trapped in modified houses compared to traditional ones, the modifications were more effective against Anopheles (94% reduction) than for non-Anopheles (71% reduction). The average temperature increased at 0.25°C in modified houses but was not statistically significant (95% CI, -0.62 to 0.12, p = 0.181). Community volunteers preferred modified houses due to reduced mosquito buzzing. The efficacy of modified houses including its acceptability by community, highlight its potential to lower malaria risk. Effective integration of modified houses into the vector control strategy will require raising awareness among communities about malaria risks associated with house structure and training them to modify their houses.
Assuntos
Anopheles , Habitação , Malária , Controle de Mosquitos , Moçambique , Malária/prevenção & controle , Animais , Controle de Mosquitos/métodos , Humanos , Mosquitos Vetores , FemininoRESUMO
BACKGROUND: Dengue is an endemic vector-borne disease influenced by environmental factors such as landscape and climate. Previous studies separately assessed the effects of landscape and climate factors on mosquito occurrence and dengue incidence. However, both factors concurrently coexist in time and space and can interact, affecting mosquito development and dengue disease transmission. For example, eggs laid in a suitable environment can hatch after being submerged in rain water. It has been difficult for conventional statistical modeling approaches to demonstrate these combined influences due to mathematical constraints. OBJECTIVES: To investigate the combined influences of landscape and climate factors on mosquito occurrence and dengue incidence. METHODS: Entomological, epidemiological, and landscape data from the rainy season (July-December) were obtained from respective government agencies in Metropolitan Manila, Philippines, from 2012 to 2014. Temperature, precipitation and vegetation data were obtained through remote sensing. A random forest algorithm was used to select the landscape and climate variables. Afterward, using the identified key variables, a model-based (MOB) recursive partitioning was implemented to test the combined influences of landscape and climate factors on ovitrap index (vector mosquito occurrence) and dengue incidence. RESULTS: The MOB recursive partitioning for ovitrap index indicated a high sensitivity of vector mosquito occurrence on environmental conditions generated by a combination of high residential density areas with low precipitation. Moreover, the MOB recursive partitioning indicated high sensitivity of dengue incidence to the effects of precipitation in areas with high proportions of residential density and commercial areas. CONCLUSIONS: Dengue dynamics are not solely influenced by individual effects of either climate or landscape, but rather by their synergistic or combined effects. The presented findings have the potential to target vector surveillance in areas identified as suitable for mosquito occurrence under specific climatic conditions and may be relevant as part of urban planning strategies to control dengue.
Assuntos
Culicidae , Dengue , Animais , Dengue/epidemiologia , Aprendizado de Máquina , Mosquitos Vetores , FilipinasRESUMO
BACKGROUND: For the last 10 years, mobile phones have provided the global health community with innovative and cost-effective strategies to address the challenges in the prevention and management of dengue fever. OBJECTIVE: The aim is to introduce and describe the design and development process of Mozzify, an integrated mobile health (mHealth) app that features real-time dengue fever case reporting and mapping system, health communication (real-time worldwide news and chat forum/timeline, within-app educational videos, links to local and international health agency websites, interactive signs and symptoms checker, and a hospital directions system), and behavior modification (reminders alert program on the preventive practices against dengue fever). We also aim to assess Mozzify in terms of engagement and information-sharing abilities, functionality, aesthetics, subjective quality, and perceived impact. METHODS: The main goals of the Mozzify app were to increase awareness, improve knowledge, and change attitudes about dengue fever, health care-seeking behavior, and intention-to-change behavior on preventive practices for dengue fever among users. It was assessed using the Mobile Application Rating Scale (MARS) among 50 purposively sampled individuals: public health experts (n=5), environment and health-related researchers (n=23), and nonclinical (end users) participants (n=22). RESULTS: High acceptability and excellent satisfaction ratings (mean scores ≥4.0 out of 5) based on the MARS subscales indicate that the app has excellent user design, functionality, usability, engagement, and information among public health experts, environment and health-related researchers, and end users. The app's subjective quality (recommending the app to other people and the app's overall star rating), and specific quality (increase awareness, improve knowledge, and change attitudes about dengue fever; health care-seeking behavior; and intention-to-change behavior on preventive practices for dengue fever) also obtained excellent satisfaction ratings from the participants. Some issues and suggestions were raised during the focus group and individual discussions regarding the availability of the app for Android devices, language options limitations, provision of predictive surveillance, and inclusion of other mosquito-borne diseases. CONCLUSIONS: Mozzify may be a promising integrated strategic health intervention system for dengue fever case reporting and mapping; increase awareness, improve knowledge, and change attitude about dengue fever; and disseminating and sharing information on dengue fever among the general population and health experts. It also can be an effective aid in the successful translation of knowledge on preventive measures against dengue fever to practice.