Optical design and the development of high accuracy automated tick classification using computer vision

使用计算机视觉进行光学设计和高精度自动蜱分类的开发

• 批准号: 10325667
• 负责人: Autumn Goodwin
• 金额: $ 29.57万
• 依托单位: VECTECH, LLC
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-06-01 至 2022-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10325667
• 关键词: Adult; Agreement; Algorithm Design; Algorithms; Anatomy; Area; Artificial Intelligence; Car Phone; Cellular Phone; Classification; Collaborations; Computer Vision Systems; Culicidae; Data; Data Set; Databases; Detection; Development; Diagnostic; Diagnostic Imaging; Disease; Disease Surveillance; Disease Vectors; Future; Goals; Grain; Human; Image; Incidence; Insecta; Larva; Learning; Leg; Life; Lighting; Methods; Modeling; Molecular; Monitor; Morphology; Nymph; Optics; Phase; Process; Psychological Transfer; Reporting; Research; Resolution; Resources; Scientist; Spatial Distribution; Specimen; Standardization; Surveillance Methods; System; Telephone; Testing; Tick-Borne Diseases; Ticks; Time; Training; Vaccines; Validation; Visual; Visualization; Work; base; citizen science; convolutional neural network; design; detection method; disorder prevention; field study; flexibility; high resolution imaging; high risk; human disease; imaging platform; imaging system; improved; insight; intelligent algorithm; interest; mobile application; novel; sample collection; sex; tick bite; tool; validation studies; vector; vector control; vector tick; Adult; Agreement; Algorithm Design; Algorithms; Anatomy; Area; Artificial Intelligence; Car Phone; Cellular Phone; Classification; Collaborations; Computer Vision Systems; Culicidae; Data; Data Set; Databases; Detection; Development; Diagnostic; Diagnostic Imaging; Disease; Disease Surveillance; Disease Vectors; Future; Goals; Grain; Human; Image; Incidence; Insecta; Larva; Learning; Leg; Life; Lighting; Methods; Modeling; Molecular; Monitor; Morphology; Nymph; Optics; Phase; Process; Psychological Transfer; Reporting; Research; Resolution; Resources; Scientist; Spatial Distribution; Specimen; Standardization; Surveillance Methods; System; Telephone; Testing; Tick-Borne Diseases; Ticks; Time; Training; Vaccines; Validation; Visual; Visualization; Work; base; citizen science; convolutional neural network; design; detection method; disorder prevention; field study; flexibility; high resolution imaging; high risk; human disease; imaging platform; imaging system; improved; insight; intelligent algorithm; interest; mobile application; novel; sample collection; sex; tick bite; tool; validation studies; vector; vector control; vector tick

Abstract. The incidence of US tick-borne diseases has more than doubled in the last two decades. Due to lack of effective vaccines for tick-borne diseases, prevention of tick bites remains the primary focus of disease mitigation. Tick vector surveillance—monitoring an area to understand tick species composition, abundance, and spatial distribution—is key to providing the public with accurate and up-to-date information when they are in areas of high risk, and enabling precision vector control when necessary. Despite the importance of vector surveillance, current practices are highly resource intensive and require significant labor and time to collect and identify vector specimens. Acarologist or field taxonomist expertise is a limited resource required for tick identification, creating a significant capability barrier for national tick surveillance practice. While mobile applications to facilitate passive surveillance and reporting of human-tick encounters have grown in popularity, variable image quality, limited engagement, and scientist misidentification of rare, invasive, or morphologically similar tick species hinder the scalability of this approach. No automated solutions exist to build tick identification capacity. We seek to develop the first imaging and automated identification system capable of instantaneously and accurately identifying the top nine tick vectors in the US. This proposal will first characterize the optical requirements necessary to image diagnostic morphological features associated with adult ticks and develop a standardized imaging platform for tick identification. This will enable the development of a high-quality tick image dataset in partnership with the Walter Reed Biosystems Unit (WRBU) which will be used to train high-accuracy computer vision models for tick species and sex identification. Ultimately the approaches developed here will enable new tick identification tools for both the lab and citizen scientists; allowing vector surveillance managers to leverage image recognition in a practical system that will increase capacity and capability for biosurveillance, and equipping citizen scientists with improved tools to identify tick species during a human-tick encounter.

抽象的。在过去的两个 几十年。由于缺乏tick传播疾病的有效疫苗，预防滴答叮咬 仍然是缓解疾病的主要重点。刻度矢量监视 - 将一个区域监视到 了解壁虱物种组成，抽象和空间分布 - 是提供的关键 公众在高风险领域中具有准确和最新的信息，并且 必要时启用精度向量控制。尽管矢量很重要 监视，当前做法是高度资源密集的，需要大量劳动力， 是时候收集和识别矢量标本了。 Acarogist或现场分类学家专业知识是 刻度识别所需的资源有限，为 国家tick监视实践。而移动应用以促进被动监视 人类挑战相遇的报道越来越受欢迎，可变图像质量，有限 参与度，科学家错误识别稀有，侵入性或形态上相似的壁虱 物种阻碍了这种方法的可扩展性。没有自动解决方案来构建tick 识别能力。我们寻求开发第一个成像和自动识别系统 能够即时，准确地识别美国的前9个tick矢量。这 提案将首先描述图像诊断所需的光学要求 与成人壁虱相关的形态特征并开发标准化成像平台 用于刻度识别。这将使在中的高质量tick图像数据集中开发 与Walter Reed Biosystems（WRBU）的合作伙伴关系将用于培训 高准确的计算机视觉模型，用于tick规格和性别识别。最终 这里开发的方法将为实验室和公民提供新的tick识别工具 科学家；允许向量监视经理在实用中利用图像识别 系统将提高生物监视的能力和能力，并为公民装备 具有改进工具的科学家在人杀手中识别壁虱物种。