Developing an App-Based Voice Clinical Decision Support Tool to Augment the Sensitivity of the Bedside Swallow Evaluation in Older Adults

项目摘要/摘要 Rameau博士的Beeson应用程序的总体目标是(1)开发一个易于使用、高度 准确的床边临床决策支持工具，通过分析确定老年人的吞咽功能障碍 他们录音的声音、咳嗽和讲话的特征，以及(2)成为新生领域的领导者 老年性耳鼻喉科，重点是老年人的言语和吞咽问题的管理 阿尔茨海默病和相关痴呆(AD/ADRD)。 应聘者：安娜伊·S·拉莫博士，威尔·康奈尔医学院耳鼻咽喉科医生 世卫组织在老年耳鼻喉科方面表现出极大的研究能力和临床兴趣，特别是在 对患有吞咽困难的老年人的护理。拉莫博士的长期职业目标是成为一名独立的 老年吞咽困难的研究人员和学术领袖。她已经为此奠定了重要的基础。 目标，发展密切的导师关系，获得广泛的研究、临床、管理和领导能力 技能，以及在多个机构和基金会资助方面担任私人投资的经验。 导师：为了实现她的目标，Rameau博士聘请了一支杰出的、跨学科的团队 导师，他们是非常成功的，由NIH资助的研究人员，以及在他们的 各自的字段。她的指导团队包括WCMC事业部联席主任马克·拉克斯博士 老年医学，WCM耳鼻喉科头颈外科主任Michael Stewart博士，以及 WCM老龄化和行为研究中心主任Sara Chaaja博士说。 研究：与康奈尔鸟类学实验室的生物声学家合作，我们的研究都发表在 犬类切除模型和一项人类试点研究表明，传统的声音和声音声学指标 在有和没有吸入风险的患者之间，咳嗽的声音有重要的不同。这些特点 可作为吞咽困难的生物标志物，以及对患者数字记录的声音和 咳嗽有可能有效地区分正常和异常吞咽。包括多个 特征，如那些专门评估口腔脆弱的特征，并结合结果可能会增加 吞咽功能障碍识别的准确性，特别是在更容易患AD/ADRD的老年人中 为沉默的渴望干杯。将非线性方法和机器学习添加到常规声学分析可以 还可以提高精确度。Rameau博士建议使用这些技术来开发临床决策支持 该工具结合了对患者记录的声音和咳嗽的多个特征的分析，以确定 他们的抱负风险。为了增加在床边使用这种工具的可能性，她将对其进行修改以进行录音 在智能手机上作为移动应用程序的一部分。在完成这项研究后，她计划提交一份R01 为了前瞻性地检查临床决策支持工具在多站点随机对照试验中的性能， 不同的老年人队列。

PROJECT SUMMARY/ABSTRACT The over-arching goals of Dr, Rameau's Beeson application are (1) to develop an easy-to-use, highly accurate bedside clinical decision support tool to identify swallowing dysfunction in older adults using analysis of features of their recorded voice, cough and speech, and (2) to become a leader in the nascent field of geriatric otolaryngology with a focus on management of speech and swallowing issues in older adults with Alzheimer's disease and related dementias (AD/ADRD). Candidate: The applicant, Dr. Anaïs Rameau, is an otolaryngologist at Weill Cornell Medical College (WCMC) who has demonstrated significant research ability and clinical interest in geriatric otolaryngology, especially in the care of older adults with dysphagia. Dr. Rameau's long-term career goal is to become an independent researcher and academic leader in geriatric dysphagia. She already laid significant foundations towards this goal, developing close mentorship relationships, gaining broad research, clinical, management and leadership skills, and experience as PI on multiple institutional and foundation grants. Mentors: To achieve her objectives, Dr. Rameau has engaged an exceptional, transdisciplinary team of mentors, who are highly successful, NIH-funded investigators and well-recognized leaders and experts in their respective fields. Her mentorship team includes Dr. Mark Lachs, Co-Director of the WCMC Division of Geriatrics, Dr. Michael Stewart, Chair of the WCM Department of Otolaryngology – Head & Neck Surgery, and Dr. Sara Czaja, Director of the Center on Aging and Behavioral Research at WCM. Research: In collaboration with bio-acousticians at the Cornell Lab of Ornithology, our research, both in an excised canine model and a human pilot study, has suggested that conventional acoustic metrics of voice and cough sound differ in important ways between patients with and without aspiration risk. These characteristics may serve as biomarkers of dysphagia, and acoustic analysis of the patient's digitally recorded voice and cough has the potential to effectively differentiate between normal and abnormal swallowing. Including multiple characteristics, such as those specifically evaluating oral frailty, and combining the results may increase accuracy of swallowing dysfunction identification, especially in older adults with AD/ADRD who are more prone to silent aspiration. Adding non-linear methods and machine learning to conventional acoustic analysis may also improve accuracy. Dr. Rameau proposes to use these techniques to develop a clinical decision support tool that incorporates analysis of multiple characteristics of a patient's recorded voice and cough to determine their aspiration risk. To increase the potential for use of this tool at the bedside, she will modify it for recordings on a smartphone as part of a mobile application. Upon completion of this research, she plans to submit an R01 to prospectively examine the performance of the clinical decision support tool in a multi-site RCT in large, diverse cohort of older adults.