TY - JOUR AU - Soroski, Thomas AU - da Cunha Vasco, Thiago AU - Newton-Mason, Sally AU - Granby, Saffrin AU - Lewis, Caitlin AU - Harisinghani, Anuj AU - Rizzo, Matteo AU - Conati, Cristina AU - Murray, Gabriel AU - Carenini, Giuseppe AU - Field, Thalia S AU - Jang, hyyeju PY - 2022 da - 2022/9/21 TI -评估基于web的阿尔茨海默语音数据自动转录:转录比较和机器学习分析JO - JMIR衰老SP - e33460 VL - 5 IS - 3 KW -阿尔茨海默病KW -轻度认知障碍KW -语音KW -自然语言处理KW -语音识别软件KW -机器学习KW -神经退行性疾病KW -转录软件KW -记忆AB -背景:用于医学研究的语音数据可以无创地大量收集。语音分析在诊断神经退行性疾病方面显示出前景。为了有效地利用语音数据，转录很重要，因为词汇内容中包含有价值的信息。人工转录虽然高度准确，但限制了潜在的可扩展性和与基于语言的筛选相关的成本节约。目的:为了更好地了解使用自动转录对神经退行性疾病的分类，即阿尔茨海默病(AD)、轻度认知障碍(MCI)或主观记忆投诉(SMC)与健康对照，我们比较了自动生成的转录本与手动校正的转录本。方法:我们招募了来自记忆诊所的个体(“患者”)，诊断为轻中度AD (n=44, 30%)， MCI (n=20, 13%)， SMC (n= 8,5%)，以及生活在社区的健康对照(n=77, 52%)。参与者被要求描述一幅标准化的图片，阅读一段文字，并回忆一段愉快的生活经历。我们通过检查转录置信度评分、转录错误率和机器学习分类准确性，将使用谷歌语音转文本软件生成的转录本与手动验证的转录本进行了比较。对于分类任务，使用了逻辑回归、高斯朴素贝叶斯和随机森林。 Results: The transcription software showed higher confidence scores (P<.001) and lower error rates (P>.05) for speech from healthy controls compared with patients. Classification models using human-verified transcripts significantly (P<.001) outperformed automatically generated transcript models for both spontaneous speech tasks. This comparison showed no difference in the reading task. Manually adding pauses to transcripts had no impact on classification performance. However, manually correcting both spontaneous speech tasks led to significantly higher performances in the machine learning models. Conclusions: We found that automatically transcribed speech data could be used to distinguish patients with a diagnosis of AD, MCI, or SMC from controls. We recommend a human verification step to improve the performance of automatic transcripts, especially for spontaneous tasks. Moreover, human verification can focus on correcting errors and adding punctuation to transcripts. However, manual addition of pauses is not needed, which can simplify the human verification step to more efficiently process large volumes of speech data. SN - 2561-7605 UR - https://aging.www.mybigtv.com/2022/3/e33460 UR - https://doi.org/10.2196/33460 UR - http://www.ncbi.nlm.nih.gov/pubmed/36129754 DO - 10.2196/33460 ID - info:doi/10.2196/33460 ER -