METHODS: We analyzed speech recordings from 109 patients (52F, 57M; Age = 72.63± 8.61) who were diagnosed with various neurodegenerative diseases, including Alzheimer\'s disease, Frontotemporal Dementia, and Vascular Cognitive Impairment, in a cognitive neurology memory clinic. Speech recordings of an open-ended picture description task were processed using the Winterlight speech analysis platform which generates >500 acoustic and linguistic features. We investigated the linear relationship between the speech features and clinical assessments including the Mini Mental State Examination (MMSE), Western Aphasia Battery (WAB), and Mattis Dementia Rating Scale while controlling for age, sex and years of education. Speech features that were significantly associated with clinical measures were then included in group comparisons with healthy older adults (N = 74, ∼39F; Age = 61.31±7.29).
RESULTS: Speech features including lexical and syntactic features were significantly correlated with clinical assessments in patients, across diagnoses. Lower MMSE scores were associated with the use of more familiar nouns (β = -1.60, p<.001). Similarly, increased impairment as assessed by the WAB was correlated with the use of higher frequency nouns (β = -0.01, p<.001). Patients used significantly more nouns (z = 6.25, p<.001) and shorter words (z = 8.33, p<.001) than the healthy older adults. Their speech duration was also significantly shorter (z = 7.98, p<.001) and they paused more (z = 5.19, p<.001).
CONCLUSIONS: Speech changes representing decreased speech, with simpler vocabularies and syntax, were detectable in patients with different neurodegenerative diseases and correlated with clinical outcomes. These same speech patterns differed in patients with neurodegenerative disease compared to healthy older adults. Speech has the potential to be a sensitive measure for detecting cognitive impairments across various neurodegenerative diseases.
方法:我们分析了109例患者的语音记录(52F,57M;年龄=72.63±8.61),被诊断患有各种神经退行性疾病,包括老年痴呆症,额颞叶痴呆,和血管性认知障碍,在认知神经学记忆诊所。使用Winterlight语音分析平台处理开放式图片描述任务的语音记录,该平台生成>500个声学和语言特征。我们调查了言语特征与临床评估之间的线性关系,包括迷你精神状态检查(MMSE),西部失语症电池(WAB),和马蒂斯痴呆症评定量表,同时控制年龄,性别和教育年限。然后将与临床指标显着相关的言语特征与健康老年人进行组比较(N=74,〜39F;年龄=61.31±7.29)。
结果:包括词汇和句法特征在内的语音特征与患者的临床评估显着相关,跨诊断。较低的MMSE评分与使用更熟悉的名词相关(β=-1.60,p<.001)。同样,WAB评估的损害增加与使用较高频率名词相关(β=-0.01,p<.001)。与健康老年人相比,患者使用的名词(z=6.25,p<.001)和较短的单词(z=8.33,p<.001)明显更多。他们的语音持续时间也明显较短(z=7.98,p<.001),并且他们暂停更多(z=5.19,p<.001)。
结论:语音变化代表语音减少,使用更简单的词汇表和语法,在患有不同神经退行性疾病的患者中均可检测到,并与临床结局相关。与健康的老年人相比,神经退行性疾病患者的这些相同的语音模式有所不同。言语有可能成为检测各种神经退行性疾病的认知障碍的敏感措施。