SBIR-STTR Award

The incidence of US tick-borne diseases has more than doubled in the last twodecades. Due to lack of effective vaccines for tick-borne diseases, prevention of tick bitesremains the primary focus of disease mitigation. Tick vector surveillance-monitoring an area tounderstand tick species composition, abundance, and spatial distribution-is key to providingthe public with accurate and up-to-date information when they are in areas of high risk, andenabling precision vector control when necessary. Despite the importance of vectorsurveillance, current practices are highly resource intensive and require significant labor andtime to collect and identify vector specimens. Acarologist or field taxonomist expertise is alimited resource required for tick identification, creating a significant capability barrier fornational tick surveillance practice. While mobile applications to facilitate passive surveillanceand reporting of human-tick encounters have grown in popularity, variable image quality, limitedengagement, and scientist misidentification of rare, invasive, or morphologically similar tickspecies hinder the scalability of this approach. No automated solutions exist to build tickidentification capacity. We seek to develop the first imaging and automated identification systemcapable of instantaneously and accurately identifying the top nine tick vectors in the US. Thisproposal will first characterize the optical requirements necessary to image diagnosticmorphological features associated with adult ticks and develop a standardized imaging platformfor tick identification. This will enable the development of a high-quality tick image dataset inpartnership with the Walter Reed Biosystems Unit (WRBU) which will be used to trainhigh-accuracy computer vision models for tick species and sex identification. Ultimately theapproaches developed here will enable new tick identification tools for both the lab and citizenscientists; allowing vector surveillance managers to leverage image recognition in a practicalsystem that will increase capacity and capability for biosurveillance, and equipping citizenscientists with improved tools to identify tick species during a human-tick encounter. Project Narrative. Despite the importance of tick vector surveillance for disease prevention, current practices to collect and identify specimens are resource intensive, limiting the quality and quantity of the data informing control efforts. Here we propose the determination of optical requirements for visualization of diagnostic features of the top nine US tick vectors, and the development of high-accuracy computer vision algorithms for the identification of tick species and sex for use in a standardized optical configuration. The high-accuracy tick classification system developed through this proposal promises to expand capacity and capability for tick vector surveillance. Adult ; 21+ years old ; Adult Human ; adulthood ; Algorithms ; Anatomy ; Anatomic ; Anatomic Sites ; Anatomic structures ; Anatomical Sciences ; Artificial Intelligence ; AI system ; Computer Reasoning ; Machine Intelligence ; Classification ; Systematics ; Computer Vision Systems ; computer vision ; Diagnostic Imaging ; Disease ; Disorder ; Disease Vectors ; Future ; Goals ; Human ; Modern Man ; Incidence ; Insecta ; Insects ; Insects Invertebrates ; Larva ; Learning ; Lighting ; Illumination ; Methods ; Culicidae ; Mosquitoes ; Nymph ; Optics ; optical ; Research ; Resources ; Research Resources ; Spatial Distribution ; Standardization ; Telephone ; Phone ; Testing ; Ticks ; Ixodida ; Time ; Psychological Transfer ; learning transfer ; training transfer ; Vaccines ; Work ; Data Set ; Dataset ; Tick-Borne Diseases ; tick-borne illness ; tickborne disease ; tickborne illness ; base ; improved ; sample collection ; specimen collection ; Area ; Phase ; Training ; insight ; Visual ; Databases ; Data Bases ; data base ; Collaborations ; Morphology ; tool ; Diagnostic ; Research Specimen ; Specimen ; Life ; Scientist ; System ; interest ; field based data ; field learning ; field test ; field study ; novel ; disease prevention ; disorder prevention ; Agreement ; validation studies ; Reporting ; Modeling ; Cell Phone ; Cellular Telephone ; iPhone ; smart phone ; smartphone ; Cellular Phone ; Mobile Phones ; Car Phone ; Leg ; Data ; Detection ; Resolution ; Surveillance Methods ; Validation ; Monitor ; Molecular ; Process ; sex ; Development ; developmental ; Image ; imaging ; vector control ; vector ; design ; designing ; human disease ; high risk ; flexibility ; flexible ; Algorithm Design ; Algorithmic Design ; Algorithmic Engineering ; algorithm engineering ; algorithmic composition ; mobile application ; mobile app ; mobile device application ; imaging platform ; imaging system ; citizen science ; amateur science ; amateur scientists ; citizen scientists ; civic science ; crowd science ; crowd-sourced science ; scientific citizenship ; high resolution imaging ; Disease Surveillance ; Grain ; convolutional neural network ; ConvNet ; convolutional network ; convolutional neural nets ; vector tick ; Visualization ; tick bite ; intelligent algorithm ; smart algorithm ; detection method ; detection procedure ; detection technique ;

The incidence of US tick-borne diseases has more than doubled in the last two decades. Today,Lyme disease is the most common vector-borne disease in the United States, impacting overhalf-a-million Americans each year. Due to lack of effective vaccines for tick-borne diseases, preventionof tick bites and early tick bite treatment is the primary focus of disease mitigation. Tick vectorsurveillance-monitoring an area to understand tick species composition, abundance, and spatialdistribution-is key to providing the public with accurate and up-to-date information when they are inareas of high risk, and enabling precision vector control when necessary. Despite the importance of vectorsurveillance, current practices are highly resource intensive and require significant labor and time tocollect and identify vector specimens. Acarologist or field taxonomist expertise is a limited resourcerequired for tick identification, creating a significant capability barrier for national tick surveillancepractice. While mobile applications to facilitate passive surveillance and reporting of human-tickencounters have grown in popularity, variable image quality, limited engagement, and scientistmisidentification of rare, invasive, or morphologically similar tick species hinder the scalability of thisapproach. To date, no automated solutions exist to build tick identification capacity. We seek to advancePhase I work that successfully achieved an imaging and automated identification system capable ofinstantaneously and accurately identifying twelve adult tick species with 98% accuracy. This proposalwill first improve the Phase I optical design for scalability to accommodate imaging of additionalintra-specific tick species variability as nymphs, adult males, and unfed or engorged adult females. Inparallel, we develop methods to optimize quality of guided user imaging of ticks in a mobile appapproach for the general public. This will enable the development of a representative image database withpartners including TickSpotters, TickCheck, the Walter Reed Biosystems Unit (WRBU), and others. Theresulting database will be used to train, validate, test and deploy high-accuracy computer vision models intwo tick identification products for professional public health and the general public. Ultimately theapproaches developed here will enable vector management organizations to leverage image recognition ina practical system that will increase capacity and capability for biosurveillance, and equip the generalpublic with improved tools to identify ticks during a human-tick encounter.

Public Health Relevance Statement:
Project Narrative. Current tick identification methods are highly resource and labor intensive, requiring physical collection of specimens and subsequent identification of species, sex, life stage, and engorgement by an acarologist based on visual morphological inspection. Here we propose to advance a prototype deep learning system capable of instantaneously and accurately identifying twelve medically-relevant adult tick species with 98% accuracy for practical deployment in professional public health and the general public. The resulting products developed through this proposal will ultimately expand tick surveillance capability and capacity, and strengthen public health response to tick-borne diseases.

Project Terms:
<21+ years old>