TY -的T1 Dysprosody标记PPA (P3.194) JF -神经学乔-神经学六世- 90 - 1首页5补充SP - P3.194盟Naomi Nevler盟沙龙灰AU -大卫欧文盟-马克Liberman盟-穆雷格罗斯曼Y1 - 2018/04/10 UR - //www.ez-admanager.com/content/90/15_Supplement/P3.194.abstract N2 -目的:量化目标标记的Dysprosody原发性进行性失语(PPA)和与他们特定clinico-anatomic特性。背景:大多数研究在韵律依靠主观的评估,经常关注情感因素。我们开发了一个自动化的技术演讲分析,演讲活动检测器(SAD),检查语音的韵律特征在PPA患者。我们假设不同的声波标记dysprosody PPA的变体。设计/方法:我们分析了半结构化的、数字化的语音样本49失语症的患者(软羊革= 14日svPPA = 18日lvPPA = 17;年龄50 - 85年,43%的男性)和31个匹配的健康对照组(年龄54 - 89年,36%的男性)。我们实现了一个自动化、语音分析协议,并提取声学测量基频(f0)和演讲和暂停时间。然后我们计算f0范围和停顿率每分钟和比较这些之间的组织和检查与临床试验的关系,灰质萎缩,CSF水平的磷酸化τ。结果:我们发现f0范围缩小在光面(平均3.96±1.14半音来)患者与健康对照组相比(6.06±1.95半音来;svPPA患者p = 0.003)和(5.85±1.96半音来;p = 0.023)。意味着在软羊革停顿率显著增加(平均58.7±21.4每分钟停顿)。 ROC curve analysis for a combined classifier (pause-rate/f0-range) gave an AUC of 93% for detection of naPPA versus healthy controls and 84% for distinguishing naPPA from svPPA. Regression analysis associated f0 range in non-fluent patients with cortical atrophy in their left premotor and middle temporal cortices. In a subset of patients, CSF level of phosphorylated tau was associated with abnormal acoustic features of speech.Conclusions: This study validates the implementation of SAD in the clinical evaluation of patients with variants of PPA, and suggests that this tool may provide effective and inexpenssive means for screening and monitoring disease in PPA.Disclosure: Dr. Nevler has nothing to disclose. Dr. Ash has nothing to disclose. Dr Irwin has nothing to disclose. Dr. Liberman has nothing to disclose. Dr. Grossman has received personal compensation for consulting, serving on a scientific advisory board, speaking, or other activities with GE Whitney. Dr. Grossman has received personal compensation in an editorial capacity for Neurology. ER -