Department of Biostatistics and Epidemiology, College of Public Health, East Tennessee State University, Johnson City, TN, United States of America.
Clayton School of Information Technology, Monash University, Melbourne, Victoria, Australia.
PLoS One. 2018 Nov 7;13(11):e0205636. doi: 10.1371/journal.pone.0205636. eCollection 2018.
It has been quite a challenge to diagnose Mild Cognitive Impairment due to Alzheimer's disease (MCI) and Alzheimer-type dementia (AD-type dementia) using the currently available clinical diagnostic criteria and neuropsychological examinations. As such we propose an automated diagnostic technique using a variant of deep neural networks language models (DNNLM) on the verbal utterances of affected individuals. Motivated by the success of DNNLM on natural language tasks, we propose a combination of deep neural network and deep language models (D2NNLM) for classifying the disease. Results on the DementiaBank language transcript clinical dataset show that D2NNLM sufficiently learned several linguistic biomarkers in the form of higher order n-grams to distinguish the affected group from the healthy group with reasonable accuracy on very sparse clinical datasets.
使用目前可用的临床诊断标准和神经心理学检查来诊断阿尔茨海默病（AD）所致轻度认知障碍（MCI）和阿尔茨海默型痴呆（AD 型痴呆）是一项极具挑战性的任务。因此，我们提出了一种使用深度神经网络语言模型（DNNLM）变体对受影响个体的口头表达进行自动诊断的技术。受 DNNLM 在自然语言任务上取得成功的启发，我们提出了一种将深度神经网络和深度语言模型（D2NNLM）相结合的方法来对疾病进行分类。在 DementiaBank 语言转录临床数据集上的结果表明，D2NNLM 充分学习了几种语言生物标志物，这些生物标志物以高阶 n-gram 的形式存在，能够在非常稀疏的临床数据集中以合理的准确性将受影响组与健康组区分开来。
