Studying the Laryngeal Mechanisms Underlying Dysphonia in Connected Speech

研究连语语音中发声困难的喉部机制

• 批准号: 10378024
• 负责人: Maryam Naghibolhosseini
• 金额: $ 13.78万
• 依托单位: MICHIGAN STATE UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-04-01 至 2024-03-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10378024
• 关键词: Acoustics; Address; Age; Area; Auditory; Behavior; Biomechanics; Categories; Characteristics; Clinical; Communication; Coupling; Data; Data Set; Development; Diagnosis; Dysphonia; Endoscopes; Evaluation; Functional disorder; Goals; Gold; Health; Human; Image; Knowledge; Larynx; Lead; Machine Learning; Measurement; Measures; Methodology; Methods; Mining; National Institute on Deafness and Other Communication Disorders; Operative Surgical Procedures; Outcome; Outcomes Research; Paralysed; Patients; Persons; Physics; Physiological; Prevention; Production; Protocols documentation; Research; Series; Severities; Source; Spastic Dysphonias; Speech; Speed; Statistical Data Interpretation; Statistical Models; Strategic Planning; System; Techniques; Testing; Therapeutic; Time; Tremor; Visual; Voice; Voice Disorders; Voice Disturbances; Voice Quality; base; clinical application; clinical development; clinical practice; clinically relevant; cohort; flexibility; image processing; imaging approach; improved; innovation; kinematics; machine learning model; sex; temporal measurement; time use; tool; treatment strategy; vibration; vocal cord; vocalization

Project Summary/Abstract This proposal aims to employ the recent advancement of coupling fiberoptic endoscopes with high-speed videoendoscopy (HSV) systems to obtain HSV recordings during connected speech. The goal is to study vocal mechanisms underlying dysphonia in patients with neurogenic voice disorders. The long-term goal of this line of research is to create clinically applicable quantitative methods for functional measurement of vocal fold vibration in connected speech using innovative laryngeal imaging, an approach that could advance clinical voice assessment and treatment practice. In Aim 1, HSV-based measures of vocal fold kinematics will be developed and the influence of these measures on voice audio-perceptual qualities in the patients will be determined. Image processing techniques will be developed to extract such measures from the HSV data in connected speech. The extracted measures will be given as inputs to the statistical models to determine the source of the differences between the normal controls and the patients for different speech phonetic contexts and words. This aim provides an unbiased HSV-based method to predict voice quality. Developing such HSV-based methodology for functional laryngeal examination in connected speech can enhance clinical voice assessment. In addition, better understanding the influence of phonetic context would lead to optimizing the protocols for functional voice assessment through laryngeal imaging in connected speech. In Aim 2, machine learning approaches will be employed to discover hidden physics and unknown laryngeal mechanisms of voice production in the dysphonic patients. The findings of this project will help make necessary adjustments in biomechanical or physiological characteristics of vocal folds to enhance voice quality in patients with neurogenic voice disorders. Therefore, the outcome of this research will aid clinicians in properly selecting, and developing new treatment strategies (therapeutic, medicinal, or surgical), which are based on the gained knowledge of laryngeal mechanisms of dysphonia. The proposed research is in harmony with multiple priority areas of the NIDCD, described in the 2017-2021 Strategic Plan. Both aims support Priority 3 (improve methods of diagnosis, treatment, and prevention) through developing objective HSV-based measures and predicting the voice quality. Comparing laryngeal mechanisms in normal and disordered voices addresses Priority 1 (deepen our understanding of the normal function of the systems of human communication). Both aims propose to study laryngeal mechanisms in patients with neurogenic and functional voice disorders, which addresses Priority 2 (increase our knowledge about conditions that alter or diminish communication and health).

项目概要/摘要 该提案旨在利用高速耦合光纤内窥镜的最新进展 视频内窥镜 (HSV) 系统可在连接语音期间获取 HSV 记录。目标是学习声乐 神经源性声音障碍患者发声困难的机制。这条线的长远目标 研究旨在创建临床适用的定量方法来测量声带振动的功能 使用创新的喉部成像进行互联语音，这是一种可以推进临床语音的方法 评估和治疗实践。在目标 1 中，将开发基于 HSV 的声带运动学测量方法 并将确定这些措施对患者语音音频感知质量的影响。图像 将开发处理技术来从互联语音的 HSV 数据中提取此类测量值。这 提取的测量值将作为统计模型的输入，以确定差异的来源 正常对照者和患者之间针对不同语音语境和单词的差异。这一目标提供了 一种基于 HSV 的无偏预测语音质量的方法。开发这种基于 HSV 的功能方法 连接语音的喉部检查可以增强临床语音评估。另外，更好的是 了解语音上下文的影响将有助于优化功能语音协议 通过连接语音中的喉部成像进行评估。在目标 2 中，机器学习方法将是 用于发现语音障碍中隐藏的物理和未知的发声机制 患者。该项目的研究结果将有助于对生物力学或生理学进行必要的调整 声带的特征，以提高神经源性声音障碍患者的声音质量。因此， 这项研究的结果将帮助临床医生正确选择和制定新的治疗策略 （治疗、药物或手术），其基于对喉部机制的了解 发声困难。拟议的研究与 NIDCD 的多个优先领域相一致，如 2017-2021 年战略计划。这两个目标都支持优先事项 3（改进诊断、治疗和治疗方法） 通过制定基于 HSV 的客观措施并预测语音质量来预防）。比较 正常和紊乱声音中的喉部机制解决了优先事项 1（加深我们对声音的理解） 人类交流系统的正常功能）。这两个目标都旨在研究喉部机制 患有神经源性和功能性声音障碍的患者，这解决了优先事项 2（增加我们的知识 关于改变或削弱沟通和健康的情况）。