Abstract Text<\/strong><\/div>\nAs a constituent component of CIFASD5, this project will focus on improving the identification of individuals with prenatal alcohol exposure (PAE), across the lifespan, with the primary goal of translation for clinical application. We target two main categories outlined in the Request for Application (RFA) 1. To identify FASD cases early and accurately, and 2. To expand and translate basic and mechanistic understandings of the effects of PAE. To address these topics, we aim to improve the accuracy and reliability of the clinical recognition of FASD, controlling for the influence of sex, co- exposures, and age. We seek to understand better how neurocognitive and facial morphology are correlative and covariant in FASD and exploit this relationship to improve screening accuracy. Previously, we have implemented methods and associated software for analyzing face shape and face – brain-cognitive relationships to detect the effects of PAE. In doing so, we have created an effective toolkit to automatically and objectively identify cardinal and subtle PAE-associated facial dysmorphism across the FASD spectrum. It is now a priority to refine and optimize our methods, utilizing low-cost 3D imaging devices to validate and distribute to the clinical community. To accomplish our goals, we propose to follow specific aims: 1. Explore the impact of factors secondary to alcohol exposure that may influence facial morphology; 2. Detect facial, cranial, and neural effects of prenatal alcohol exposure across the lifespan; 3. Develop, test, and validate automated methods for FASD screening, integrating neurocognitive assessment tools and distributing to the clinical community. Achieving aims 1 and 2 will dramatically refine FASD recognition by understanding the factors secondary to alcohol that may influence outcomes across the lifespan. Successful demonstration of aim 2 will achieve the earliest possible diagnosis enabling further research and feedback for intervention studies and also focus on recognizing traits in adults – a population significantly underreported in the literature. Ultimately, aim 3 encompasses the primary goal of the project, developing, testing, and validating multi-domain screening tools that assist with clinical identification. We will work closely with consortium partners who will recruit subjects and provide facial images and neurocognitive assessment data (U01: Chambers; U01: Coles & Weinberg; U01: Wozniak; U01: DiClemente; U01: Petrenko). Particularly close collaboration is required for the face-neurocognitive analysis, where we will integrate and combine facial and neurocognitive screening tools (U01: Mattson). We will rely on research resources for data management (R24: Wetherill) and to validate screening tools and gain clinical feedback (R24: Del Campo).<\/p>\n
Public Health Relevance Statement<\/strong><\/div>\nAutomated tools will be developed to better identify those with suspected prenatal alcohol exposure using a combination of 3D facial imaging and neurocognitive testing using low-cost imaging devices. We will utilize outputs from online cognitive assessment (FASD-Tree and Brain-Online) tools to improve sensitivity\/specificity for the identification of FASD. Machine learning approaches will be applied to achieve automated objective analysis of facial images and deployed on a cloud server for worldwide, multi-device accessibility. To ensure our tools are optimized, we will investigate the impact on facial dysmorphism of age, ethnicity, sexual dimorphism, and co-exposures. These tools will facilitate and support the recognition of individuals with FASD, who may need further support, and will include infants through to adults using a fully automated and objective approach, using low-cost solutions and providing an accessible toolkit for clinicians.<\/p>\n