A Transfer Learning Method for Detecting Alzheimer's Disease Based on Speech and Natural Language Processing.
A Transfer Learning Method for Detecting Alzheimer's Disease Based on Speech and Natural Language Processing.
复制标题
DOI：
10.3389/fpubh.2022.772592
复制
发表时间：
2022
影响因子：
5.2
通讯作者：
Chen, Yan
中科院分区：
文献类型：
--
作者：
Liu, Ning;Luo, Kexue;Yuan, Zhenming;Chen, Yan
阿尔茨海默病（AD）是一种神经退行性疾病，很难用方便可靠的方法检测出来。阿尔茨海默病患者的语言变化是其认知状态的重要信号，可能有助于早期诊断。在这项研究中，我们开发了一个基于语音和自然语言处理（NLP）技术的迁移学习模型，用于AD的早期诊断。大型数据集的缺乏限制了没有特征工程的复杂神经网络模型的使用，而迁移学习可以有效地解决这一问题。迁移学习模型首先在大型文本数据集上进行预训练，得到预训练好的语言模型，然后在此模型的基础上，在小型训练集上进行AD分类模型。具体而言，采用蒸馏双向编码器表示（蒸馏bert）嵌入，结合逻辑回归分类器，将AD与正常对照区分开来。模型实验通过2020年的自发语音数据集对阿尔茨海默氏痴呆症的识别进行了评估，其中包括78名健康对照（HC）和78名AD患者。该模型的准确率为0.88，几乎与挑战赛的冠军分数相当，比挑战赛组织者设定的75%的基线有了相当大的提高。因此，本研究中的迁移学习方法提高了AD预测，不仅减少了对特征工程的需求，而且解决了缺乏足够大的数据集的问题。
Alzheimer's disease (AD) is a neurodegenerative disease that is difficult to be detected using convenient and reliable methods. The language change in patients with AD is an important signal of their cognitive status, which potentially helps in early diagnosis. In this study, we developed a transfer learning model based on speech and natural language processing (NLP) technology for the early diagnosis of AD. The lack of large datasets limits the use of complex neural network models without feature engineering, while transfer learning can effectively solve this problem. The transfer learning model is firstly pre-trained on large text datasets to get the pre-trained language model, and then, based on such a model, an AD classification model is performed on small training sets. Concretely, a distilled bidirectional encoder representation (distilBert) embedding, combined with a logistic regression classifier, is used to distinguish AD from normal controls. The model experiment was evaluated on Alzheimer's dementia recognition through spontaneous speech datasets in 2020, including the balanced 78 healthy controls (HC) and 78 patients with AD. The accuracy of the proposed model is 0.88, which is almost equivalent to the champion score in the challenge and a considerable improvement over the baseline of 75% established by organizers of the challenge. As a result, the transfer learning method in this study improves AD prediction, which does not only reduces the need for feature engineering but also addresses the lack of sufficiently large datasets.
登录
查看更多内容
影响因子：
2.6
作者：
Jonell, Patrik;Moell, Birger;Beskow, Jonas
通讯作者：
Beskow, Jonas
影响因子：
4.8
作者：
De Looze C;Dehsarvi A;Crosby L;Vourdanou A;Coen RF;Lawlor BA;Reilly RB
通讯作者：
Reilly RB
影响因子：
2.6
作者：
Laguarta, Jordi;Subirana, Brian
通讯作者：
Subirana, Brian
中国 60 岁及以上成年人痴呆和轻度认知障碍的患病率、危险因素及治疗：一项横断面研究
影响因子：
50
作者：
Jia, Longfei;Du, Yifeng;Jia, Jianping
通讯作者：
Jia, Jianping
影响因子：
3.7
作者：
Mahendran N;Vincent PMDR;Srinivasan K;Chang CY
通讯作者：
Chang CY