Improving the Assessment of Mild Cognitive Impairment in Advanced Age With a Novel Multi-Feature Automated Speech and Language Analysis of Verbal Fluency

Abstract

Introduction: Clinically relevant information can go uncaptured in the conventional scoring of a verbal fluency test. We hypothesize that characterizing the temporal aspects of the response through a set of time related measures will be useful in distinguishing those with MCI from cognitively intact controls. Methods: Audio recordings of an animal fluency test administered to 70 demographically matched older adults (mean age 90.4 years), 28 with mild cognitive impairment (MCI) and 42 cognitively intact (CI) were professionally transcribed and fed into an automatic speech recognition (ASR) system to estimate the start time of each recalled word in the response. Next, we semantically cluster participant generated animal names and through a novel set of time-based measures, we characterize the semantic search strategy of subjects in retrieving words from animal name clusters. This set of time-based features along with standard count-based features (e.g., number of correctly retrieved animal names) were then used in a machine learning algorithm trained for distinguishing those with MCI from CI controls. Results: The combination of both count-based and time-based features, automatically derived from the test response, achieved 77% on AUC-ROC of the support vector machine (SVM) classifier, outperforming the model trained only on the raw test score (AUC, 65%), and well above the chance model (AUC, 50%). Conclusion: This approach supports the value of introducing time-based measures to the assessment of verbal fluency in the context of this generative task differentiating subjects with MCI from those with intact cognition.

Consistency and applicability of different brief screen instrument of cognitive function in elderly population

“…Many neuropsychological instruments developed may be used for quick screening of cognitive functions. Verbal fluency test mostly evaluates attention and executive function [ 43 , 48 ]. Performance on the SDMT is mostly underpinned by attention, perceptual speed, motor speed, and visual scanning [ 49 ].…”

Mental health and aging in the 2020s.

“…Digital assistants, such as Apple’s Siri, Amazon’s Alexa, Microsoft’s Cortana, and Google Assistant, have the potential to aid memory, enhance medication adherence, and offer some form of social interaction (Pradhan et al, 2019), as do increasingly sophisticated robots (Pu et al, 2019) and automated chatbots. Sensors placed throughout the home and other monitoring systems may provide unobtrusive, objective assessment data (Chen et al, 2020) and enable older adults with cognitive impairments to stay safely at home (Robinson et al, 2020). Of course, the widening availability of consumer-oriented products means clinicians need to keep up-to-date on the science, complicated privacy policies, and steps for implementation.…”

Assessing Deception in Questionnaire Surveys With Eye-Tracking

“…The most promising classification accuracy came from a linear SVM, and the performance of all classifiers is shown in Supplementary Table S1 . SVM classifier is a type of supervised machine learning approach that attempts to distinguish between two classes of data points separated by a hyperplane in a high dimensional space ( Cortes and Vapnik, 1995 ; Chen et al, 2020 ). SVM is widely used to deal with classification problems in machine learning ( Byvatov et al, 2003 ; Peltier et al, 2009 ), many studies of lie detection ( Mottelson et al, 2018 ; Mazza et al, 2020 ; Mathur and Matarić, 2020 ) or eye movements ( Huang et al, 2015 ; Dalrymple et al, 2019 ; Steil et al, 2019 ; Kang et al, 2020 ) have used SVM for classification.…”