语音重复任务在轻度认知功能障碍检测中的应用
Application of speech repetition task in the detection of mild cognitive impairment
殷潇潇 1王思文 1王贺 1高琳琳 2任智 1王钦文3
作者信息
- 1. 宁波大学教师教育学院(宁波 315211)
- 2. 宁波大学信息科学与工程学院
- 3. 宁波大学医学院
- 折叠
摘要
轻度认知功能障碍(mild cognitive impairment,MCI)通常被视为痴呆的前驱阶段,其主要特征为认知功能轻度下降.研究表明,MCI患者中语言变化可能先于其他认知功能症状,这为早期识别和干预提供了机会.MCI患者语言特点包括语速、发音和语调等异常.五个单词测验、数字延迟匹配测试和句子重复测试等语音重复任务,是评估MCI患者语言特点的有效方法,这些任务要求患者重复特定内容,分析重复准确性,从而评估其语言功能.机器学习和深度学习技术的应用,能自动提取语音重复任务数据中的MCI相关特征,提高诊断准确性.这些技术的结合应用有助于早期发现MCI,为及时干预提供依据.
Abstract
Mild cognitive impairment(MCI)is often regarded as a prodromal stage of dementia,primarily characterized by mild decline in cognitive function.Due to the mild nature of its symptoms,many MCI cases miss the opportunity for intervention.Research indicates that linguistic changes in MCI may precede other cognitive symptoms,providing an opportunity for early identification and intervention.These linguistic changes include abnormalities in speech rate,pronunciation,and intonation.Speech repetition tasks,such as the five-word test,delayed digit matching,and sentence repetition tests,are effective methods for assessing the speech characteristics of MCI patients.These tasks involve asking patients to repeat specific content,and analyzing the accuracy of their repetitions to assess their speech function.The application of machine learning and deep learning techniques enables automatic extraction of MCI-related features from speech data,improving diagnostic accuracy.The combined application of these techniques contributes to the early detection of MCI,providing a basis for timely intervention.
关键词
认知功能障碍/任务重复/阿尔茨海默病/语言/语音识别/机器学习/深度学习Key words
Mild cognitive impairment/Task repetition/Alzheimer disease/Language/Speech repetition/Ma-chine learning/Deep learning引用本文复制引用
基金项目
国家自然科学基金(32171035)
浙江省"尖兵""领雁"研发攻关计划(2024C03101)
宁波市"科技创新"重大专项(2025)(2019B10034)
出版年
2024
NETL
NSTL
万方数据