Extending WAAVES+: An animal and environment-agnostic, automated USV scoring platform for high-throughput social, behavioral, and neuropharmacological studies

项目摘要 超声波发声（USV）被认为反映了情绪处理、脑神经化学和脑功能。 功能-动物模型研究中的关键观察结果。收集和处理USV数据是时间密集型的， 手工制作，而且很贵。最重要的是，它限制了研究人员充分有效地利用 实验设计，是其他研究人员进入的障碍。啮齿动物自动制表 发声数据将降低成本，并使研究人员能够在数小时内充分探索机制， 多时段研究。解决这一限制将提高研究的效率，并可能加快 心理健康的新进展。 我们的总体目标是满足整个社区对上级自动USV评分的需求-减轻 USV分析的手动负担，同时促进和增强基于啮齿动物和小鼠的情感 加工研究。我们的具体目标是（a）开发一种与环境无关的自动化USV 评分工具，“开箱即用”，无需大量定制或环境特定的培训集;（B） 自动化呼叫分类并扩展声学特性的参数化，为研究人员提供 用于调查重要和行为相关结果的额外数据。 这项工作将消除与基于USV的研究相关的成本和时间障碍，同时 扩大相关数据的数量。我们保守估计，如果 不是1 M。这项工作将形成一个产品的基础，将提高研究吞吐量，减少 治疗开发时间和成本，并允许研究人员设计和实施更复杂的 实验设计加速解决精神健康障碍的进展。

Project Summary Ultrasonic vocalizations (USVs) are known to reflect emotional processing, brain neurochemistry and brain function - key observations in animal model studies. Collecting and processing USV data is time-intensive, manual, and costly. Most importantly, it limits researchers’ ability to employ fully effective, and nuanced experimental designs, and is a barrier to entry for other researchers. Automated tabulation of rodent vocalization data would reduce cost and enable researchers to fully explore mechanisms through multi-hour, multi-session studies. Addressing this limitation will enhance the efficiency of research and could accelerate new mental health advancements. Our overall goal is to serve a community-wide need for superior, automated USV scoring - alleviating the manual burden of USV analyses while facilitating and enhancing rodent- and mouse-based emotional processing research. Our specific objectives are to (a) develop an environment-agnostic, automated USV scoring tool that works “out-of-the-box” without extensive tailoring or environment-specific training sets; (b) automate call classification and expand the parameterization of acoustic properties to provide researchers with additional data for investigating significant and behaviorally-relevant outcomes. This work will eliminate the cost and time barrier associated with USV-based research while simultaneously expanding the quantity of relevant data. We conservatively estimate an annual cost savings in the $100Ks, if not $1Ms. This work will form the foundation of a product that will enhance research throughput, reduce treatment development time and cost, and allow researchers to design and implement more complex experimental designs accelerating advancements addressing mental health disorders.