SCH: INT: Surveillance and Control of Mosquito-Borne Diseases through Automated Species Identification and Spatiotemporal Modeling

蚊媒疾病的传播对国家和世界的健康和福利构成了紧迫的威胁。其中许多疾病(西尼罗河病、登革热、疟疾、寨卡病毒)已经成为地方病，据估计，每年全世界有270万人死于疫情。佛罗里达州是美国国内蚊媒疾病的震中，2018年爆发了毁灭性的寨卡病毒疫情，2019年和2020年出现了当地传播的登革热病例。已知的大多数蚊媒疾病是通过三个常见的蚊属传播的，即伊蚊、按蚊和库蚊。由于这些疾病中的许多都没有疫苗或治疗方法，实时监测在部署对策方面至关重要，例如更有针对性的杀虫剂治疗和公共宣传运动，以消除滋生生境和减少疾病暴发。该奖项支持开发一个通过智能手机图像大规模自动识别蚊子属和物种的平台的研究。该平台将使公民能够上传智能手机图像，从而为全球蚊子种群的实时数据做出贡献。该项目将研究深度学习技术，从智能手机图像中自动分类蚊子物种。识别蚊子是一个具有挑战性的问题，因为对于没有受过训练的人来说，物种差异并不明显。识别技术将基于对蚊子不同解剖特征的分割。该项目将产生经过验证的算法，用于大规模的物种自动分类。这些算法将嵌入一个平台，用于通过众包方式输入带有地理标记的蚊子和死鸟图像。这些数据将被用来检测入侵蚊子的引入，生成蚊子分布图，并生成实时风险图，以便及早发现疾病暴发。这些鉴定方法有望用于其他昆虫物种的分类，以及进一步研究蚊子生态学和进化生物学，以改善公众健康。这一奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The spread of mosquito-borne diseases poses an urgent threat to the Nation's and the world's health and welfare. Many of these diseases (West Nile disease, dengue fever, malaria, Zika) have become endemic, and outbreaks have been estimated to result annually in 2.7 million deaths worldwide. The state of Florida is a domestic epicenter for mosquito-borne diseases, with a devastating Zika outbreak in 2018 and locally transmitted cases of dengue fever in 2019 and 2020. The majority of known mosquito-borne diseases are transmitted by three common mosquito genera, namely Aedes, Anopheles, and Culex. Because there are no vaccines or cures available for many of these diseases, real-time surveillance is critical in deploying countermeasures, such as more targeted insecticide treatment and public information campaigns, to eliminate breeding habitats and mitigate disease outbreaks. This award supports research to develop a platform for large-scale automated identification of mosquito genera and species via smartphone images. The platform will enable citizens to upload smartphone images to contribute to real-time data data on mosquito populations worldwide.The project will investigate deep learning techniques for automated classification of mosquito species from smartphone images. Mosquito identification is a challenging problem, as species differences are not obvious to the untrained eye. Identification techniques will be based on segmentation of different anatomical features of mosquitoes. The project will result in validated algorithms for automated classification of species at scale. The algorithms will be embedded in a platform for crowd-sourced input of geographically-tagged images of mosquitoes and dead birds. These data will be leveraged to detect introductions of invasive mosquitoes, generate mosquito distribution maps, and produce real-time risk maps to enable early detection of disease outbreaks. The identification methods are expected to be useful for the classification of other insect species and to further investigations in mosquito ecology and evolutionary biology with the goal of improving public health.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.