Vocabulary Size in Speech May Be an Early Indicator of Cognitive Impairment

Beyond electronic health record data: leveraging natural language processing and machine learning to uncover cognitive insights from patient-nurse verbal communications

BACKGROUND

Mild cognitive impairment and early-stage dementia significantly impact healthcare utilization and costs, yet more than half of affected patients remain underdiagnosed. This study leverages audio-recorded patient-nurse verbal communication in home healthcare settings to develop an artificial intelligence-based screening tool for early detection of cognitive decline.

OBJECTIVE

To develop a speech processing algorithm using routine patient-nurse verbal communication and evaluate its performance when combined with electronic health record (EHR) data in detecting early signs of cognitive decline.

METHOD

We analyzed 125 audio-recorded patient-nurse verbal communication for 47 patients from a major home healthcare agency in New York City. Out of 47 patients, 19 experienced symptoms associated with the onset of cognitive decline. A natural language processing algorithm was developed to extract domain-specific linguistic and interaction features from these recordings. The algorithm's performance was compared against EHR-based screening methods. Both standalone and combined data approaches were assessed using F1-score and area under the curve (AUC) metrics.

RESULTS

The initial model using only patient-nurse verbal communication achieved an F1-score of 85 and an AUC of 86.47. The model based on EHR data achieved an F1-score of 75.56 and an AUC of 79. Combining patient-nurse verbal communication with EHR data yielded the highest performance, with an F1-score of 88.89 and an AUC of 90.23. Key linguistic indicators of cognitive decline included reduced linguistic diversity, grammatical challenges, repetition, and altered speech patterns. Incorporating audio data significantly enhanced the risk prediction models for hospitalization and emergency department visits.

DISCUSSION

Routine verbal communication between patients and nurses contains critical linguistic and interactional indicators for identifying cognitive impairment. Integrating audio-recorded patient-nurse communication with EHR data provides a more comprehensive and accurate method for early detection of cognitive decline, potentially improving patient outcomes through timely interventions. This combined approach could revolutionize cognitive impairment screening in home healthcare settings.

Fractional amplitude of low-frequency fluctuations during music-evoked autobiographical memories in neurotypical older adults

Introduction

Researchers have shown that music-evoked autobiographical memories (MEAMs) can stimulate long-term memory mechanisms while requiring little retrieval effort and may therefore be used in promising non-pharmacological interventions to mitigate memory deficits. Despite an increasing number of studies on MEAMs, few researchers have explored how MEAMs are bound in the brain.

Methods

In the current study activation indexed by fractional amplitude of low frequency fluctuations (fALFF) during familiar and unfamiliar MEAM retrieval was compared in a sample of 24 healthy older adults. Additionally, we aimed to investigate the impact of age-related gray matter volume (GMV) reduction in key regions associated with MEAM-related activation. In addition to a T1 structural scan, neuroimaging data were collected while participants listened to familiar music (MEAM retrieval) versus unfamiliar music.

Results

When listening to familiar compared to unfamiliar music, greater fALFF activation patterns were observed in the right parahippocampal gyrus, controlling for age and GMV. The current findings for the familiar (MEAM) condition have implications for cognitive aging as persons experiencing age-related memory decline are particularly susceptible to volumetric reduction in the parahippocampal cortex. Post-hoc analyses to explore correlations between brain activity and the content of MEAMs were performed using the text analysis program Linguistic Inquiry and Word Count.

Discussion

Our findings suggest that MEAM-related activation of the parahippocampal cortex is evident in normative older adults. However, it is yet to be determined whether such brain states are attainable in older adult populations diagnosed with mild cognitive impairment and/or prodromal Alzheimer's disease.

Plasma amyloid-β oligomerization tendency as a potential predictor for conversion from mild cognitive impairment to Alzheimer's dementia: Findings from the GMCII cohort

INTRODUCTION

This study aimed to explore the association between plasma amyloid-β oligomerization tendency (OAβ) and cognitive performance in Alzheimer's disease (AD) and determine its predictive value for outcomes of mild cognitive impairment (MCI).

METHODS

Plasma from 727 subjects (286 AD, 260 MCI, and 181 controls) in a case registry was analyzed using the multimer detection system (MDS) to measure plasma OAβ.

RESULTS

Elevated plasma OAβ was strongly correlated with multidomain cognitive performance in patients with MCI and AD. Patients with MCI with high baseline plasma OAβ demonstrated a higher risk of progressing to dementia (hazard ratio = 1.083, 95% confidence interval [CI] 1.032-1.137). Baseline plasma OAβ effectively predicted MCI-dementia conversion (area under the curve [AUC] = 0.824, 95% CI 0.752-0.897).

DISCUSSION

The real-world findings underscore the clinical relevance of plasma OAβ as a potential predictor for the conversion from mild cognitive impairment (MCI) to dementia.

Highlights

We recruit study participants of Alzheimer's dementia (AD), mild cognitive impairment (MCI), and cognitively normal controls in a case registry.We use the multimer detection system (MDS) to measure plasma amyloid-β oligomerization tendency (OAβ).We observe that elevated plasma OAβ strongly correlates with multidomain cognitive performance in patients with MCI and AD.MCI individuals with high baseline plasma OAβ demonstrate a higher risk of progressing to dementia.The real-world findings underscore the clinical relevance of plasma Oaβ as a potential predictor for the conversion from MCI to dementia.

A preliminary investigation on core lexicon analysis in dementia of the Alzheimer's type

BACKGROUND

Core lexicon (CL) analysis is a time efficient and possibly reliable measure that captures discourse production abilities. For people with aphasia, CL scores have demonstrated correlations with aphasia severity, as well as other discourse and linguistic measures. It was also found to be clinician-friendly and clinically sensitive enough to capture longitudinal changes in aphasia. To our knowledge, CL has never been investigated in individuals with neurologically progressive disease.

AIMS

As a preliminary investigation, we sought to investigate (1) whether CL scores correlate with dementia severity, (2) whether CL scores correlate with measures of discourse quality, and (3) whether CL scores correlate with other measures of lexical/semantic access.

METHODS & PROCEDURES

Twelve participants with a cognitive impairment associated with dementia of the Alzheimer's type (DAT) completed several measures of language and cognitive ability, as well as provide a language sample from the wordless picture book, Picnic.

RESULTS & CONCLUSION

Results are informative, as they provide insight into characteristics of CL and provide support for potential use of CL in individuals with neurologically progressive disease. The results indicated that CL scores do correlate with dementia severity and several measures of language ability, indicating they may provide a useful measure of language abilities in DAT, but more research is needed.

WHAT THIS PAPER ADDS

What is already known on the subject Core lexicon (CL) analysis is an assessment measure of discourse ability, most closely related to informativeness or productivity, used in aphasiology that is easier to use and less time consuming than previous measures of informativeness, such as correct information units or type-token ratio (TTR). For people with aphasia, CL analysis correlates with aphasia severity, measures of informativeness, as well as other measures of discourse quality. It has also been shown to be faster and more reliable between scorers than other informativeness measures. What this study adds Core lexicon analysis is a new simple and online method for assessing the informativeness of a discourse sample without the need to record or transcribe the language sample. CL is receiving a lot of attention in aphasia, correlating with everything from aphasia severity to measures of productivity and lexical access, as well as measures of informativeness. Unfortunately, no one has investigated CL analysis in dementia. The study demonstrates the first evidence that CL analysis may be a useful measure for determining dementia severity and language quality in people with dementia. What are the clinical implications of this work? Core lexicon analysis may provide clinicians and researchers with an easy method for assessing the discourse of people with a cognitive impairment associated with dementia of the Alzheimer's type. This will improve initial assessment, as well as improve ongoing language assessment that may provide clues into their functional ability to communicate effectively.

Written discourse in diagnosis for acquired neurogenic communication disorders: current evidence and future directions

Purpose

We aimed to perform the first review of research focusing on written discourse performance in people with acquired neurogenic communication disorders. In studies from 2000 onward, we specifically sought to determine: (1) the differences between patient populations and control groups, (2) the differences between different patient populations, (3) longitudinal differences between patient populations, and (4) modality differences between spoken and written discourse performance.

Methods

We completed a thorough search on MEDLINE, Embase, Cochrane, APAPsycinfo, Web of Science, and Scopus databases. We identified studies that focus on written discourse performance in people with aphasia, primary progressive aphasia, mild cognitive impairment, and Alzheimer's disease.

Results

Nineteen studies were identified from the review of literature, some of which addressed more than one of our review questions. Fifteen studies included a comparison between clinical populations and controls. Six studies compared different characteristics of patient populations. Three studies reported changes over time in progressive disorders. Six studies targeted different modalities of discourse.

Conclusion

Differences in linguistic features by patient populations are not yet clear due to the limited number of studies and different measures and tasks used across the studies. Nevertheless, there is substantial evidence of numerous linguistic features in acquired neurogenic communication disorders that depart from those of healthy controls. Compared to the controls, people with aphasia tend to produce fewer words, and syntactically simpler utterances compared to the controls. People with Alzheimer's disease produce less information content, and this feature increases over time, as reported in longitudinal studies. Our review imparts additional information that written and spoken discourse provide unique insights into the cognitive and linguistic deficits experienced by people with aphasia, Alzheimer's disease, mild cognitive impairment and primary progressive aphasia and provide targets for treatment to improve written communication in these groups.

Natural language processing techniques for studying language in pathological ageing: A scoping review

BACKGROUND

In the past few years there has been a growing interest in the employment of verbal productions as digital biomarkers, namely objective, quantifiable behavioural data that can be collected and measured by means of digital devices, allowing for a low-cost pathology detection, classification and monitoring. Numerous research papers have been published on the automatic detection of subtle verbal alteration, starting from written texts, raw speech recordings and transcripts, and such linguistic analysis has been singled out as a cost-effective method for diagnosing dementia and other medical conditions common among elderly patients (e.g., cognitive dysfunctions associated with metabolic disorders, dysarthria).

AIMS

To provide a critical appraisal and synthesis of evidence concerning the application of natural language processing (NLP) techniques for clinical purposes in the geriatric population. In particular, we discuss the state of the art on studying language in healthy and pathological ageing, focusing on the latest research efforts to build non-intrusive language-based tools for the early identification of cognitive frailty due to dementia. We also discuss some challenges and open problems raised by this approach.

METHODS & PROCEDURES

We performed a scoping review to examine emerging evidence about this novel domain. Potentially relevant studies published up to November 2021 were identified from the databases of MEDLINE, Cochrane and Web of Science. We also browsed the proceedings of leading international conferences (e.g., ACL, COLING, Interspeech, LREC) from 2017 to 2021, and checked the reference lists of relevant studies and reviews.

MAIN CONTRIBUTION

The paper provides an introductory, but complete, overview of the application of NLP techniques for studying language disruption due to dementia. We also suggest that this technique can be fruitfully applied to other medical conditions (e.g., cognitive dysfunctions associated with dysarthria, cerebrovascular disease and mood disorders).

CONCLUSIONS & IMPLICATIONS

Despite several critical points need to be addressed by the scientific community, a growing body of empirical evidence shows that NLP techniques can represent a promising tool for studying language changes in pathological aging, with a high potential to lead a significant shift in clinical practice.

WHAT THIS PAPER ADDS

What is already known on this subject Speech and languages abilities change due to non-pathological neurocognitive ageing and neurodegenerative processes. These subtle verbal modifications can be measured through NLP techniques and used as biomarkers for screening/diagnostic purposes in the geriatric population (i.e., digital linguistic biomarkers-DLBs). What this paper adds to existing knowledge The review shows that DLBs can represent a promising clinical tool, with a high potential to spark a major shift to dementia assessment in the elderly. Some challenges and open problems are also discussed. What are the potential or actual clinical implications of this work? This methodological review represents a starting point for clinicians approaching the DLB research field for studying language in healthy and pathological ageing. It summarizes the state of the art and future research directions of this novel approach.

AI-based assessments of speech and language impairments in dementia

Recent advancements in the artificial intelligence (AI) domain have revolutionized the early detection of cognitive impairments associated with dementia. This has motivated clinicians to use AI-powered dementia detection systems, particularly systems developed based on individuals' and patients' speech and language, for a quick and accurate identification of patients with dementia. This paper reviews articles about developing assessment tools using machine learning and deep learning algorithms trained by vocal and textual datasets.

ADscreen: A speech processing-based screening system for automatic identification of patients with Alzheimer's disease and related dementia

Alzheimer's disease and related dementias (ADRD) present a looming public health crisis, affecting roughly 5 million people and 11 % of older adults in the United States. Despite nationwide efforts for timely diagnosis of patients with ADRD, >50 % of them are not diagnosed and unaware of their disease. To address this challenge, we developed ADscreen, an innovative speech-processing based ADRD screening algorithm for the protective identification of patients with ADRD. ADscreen consists of five major components: (i) noise reduction for reducing background noises from the audio-recorded patient speech, (ii) modeling the patient's ability in phonetic motor planning using acoustic parameters of the patient's voice, (iii) modeling the patient's ability in semantic and syntactic levels of language organization using linguistic parameters of the patient speech, (iv) extracting vocal and semantic psycholinguistic cues from the patient speech, and (v) building and evaluating the screening algorithm. To identify important speech parameters (features) associated with ADRD, we used the Joint Mutual Information Maximization (JMIM), an effective feature selection method for high dimensional, small sample size datasets. Modeling the relationship between speech parameters and the outcome variable (presence/absence of ADRD) was conducted using three different machine learning (ML) architectures with the capability of joining informative acoustic and linguistic with contextual word embedding vectors obtained from the DistilBERT (Bidirectional Encoder Representations from Transformers). We evaluated the performance of the ADscreen on an audio-recorded patients' speech (verbal description) for the Cookie-Theft picture description task, which is publicly available in the dementia databank. The joint fusion of acoustic and linguistic parameters with contextual word embedding vectors of DistilBERT achieved F1-score = 84.64 (standard deviation [std] = ±3.58) and AUC-ROC = 92.53 (std = ±3.34) for training dataset, and F1-score = 89.55 and AUC-ROC = 93.89 for the test dataset. In summary, ADscreen has a strong potential to be integrated with clinical workflow to address the need for an ADRD screening tool so that patients with cognitive impairment can receive appropriate and timely care.

Language patterns in Japanese patients with Alzheimer disease: A machine learning approach

AIM

The authors applied natural language processing and machine learning to explore the disease-related language patterns that warrant objective measures for assessing language ability in Japanese patients with Alzheimer disease (AD), while most previous studies have used large publicly available data sets in Euro-American languages.

METHODS

The authors obtained 276 speech samples from 42 patients with AD and 52 healthy controls, aged 50 years or older. A natural language processing library for Python was used, spaCy, with an add-on library, GiNZA, which is a Japanese parser based on Universal Dependencies designed to facilitate multilingual parser development. The authors used eXtreme Gradient Boosting for our classification algorithm. Each unit of part-of-speech and dependency was tagged and counted to create features such as tag-frequency and tag-to-tag transition-frequency. Each feature's importance was computed during the 100-fold repeated random subsampling validation and averaged.

RESULTS

The model resulted in an accuracy of 0.84 (SD = 0.06), and an area under the curve of 0.90 (SD = 0.03). Among the features that were important for such predictions, seven of the top 10 features were related to part-of-speech, while the remaining three were related to dependency. A box plot analysis demonstrated that the appearance rates of content words-related features were lower among the patients, whereas those with stagnation-related features were higher.

CONCLUSION

The current study demonstrated a promising level of accuracy for predicting AD and found the language patterns corresponding to the type of lexical-semantic decline known as 'empty speech', which is regarded as a characteristic of AD.

Analysis of spontaneous speech in Parkinson's disease by natural language processing

INTRODUCTION

Patients with Parkinson's disease (PD) encounter a variety of speech-related problems, including dysarthria and language disorders. To elucidate the pathophysiological mechanisms for linguistic alteration in PD, we compared the utterance of patients and that of healthy controls (HC) using automated morphological analysis tools.

METHODS

We enrolled 53 PD patients with normal cognitive function and 53 HC, and assessed their spontaneous speech using natural language processing. Machine learning algorithms were used to identify the characteristics of spontaneous conversation in each group. Thirty-seven features focused on part-of-speech and syntactic complexity were used in this analysis. A support-vector machine (SVM) model was trained with ten-fold cross-validation.

RESULTS

PD patients were found to speak less morphemes on one sentence than the HC group. Compared to HC, the speech of PD patients had a higher rate of verbs, case particles (dispersion), and verb utterances, and a lower rate of common noun utterances, proper noun utterances, and filler utterances. Using these conversational changes, the respective discrimination rates for PD or HC were more than 80%.

CONCLUSIONS

Our results demonstrate the potential of natural language processing for linguistic analysis and diagnosis of PD.

Identifying an Appropriate Picture Stimulus for a Bangla Picture Description Task

PURPOSE

The absence of culture- and language-specific aphasia assessment in Bangla underscores a critical problem in communication sciences and disorders. Aphasia occurs in ~41% of Bangla-speaking stroke survivors. In the past 40 years, stroke incidence has doubled in low- and middle-income countries, such as Bangladesh and India, where there are ~250 million native Bangla speakers. This study aims to initiate the first step toward identifying an appropriate picture stimulus for the Bangla picture description task (PDT) intended for inclusion in a Bangla aphasia assessment. Researchers have reported the importance of cultural relevance and three visuographic variables of a picture (high-context, color, and photograph vs. black-and-white line drawing) for faster comprehension and comprehensive language production in people with aphasia and neurologically healthy adults.

METHOD

Ninety-six neurologically healthy native Bangla speakers of three age groups (young 19-30, middle age 40-55, and older 65-89 years) were recruited to compare spontaneous language production for four selected culturally related and nonrelated picture stimuli with and without the three visuographic variables. Five linguistic variables were used to analyze the language samples.

RESULTS

The results demonstrated a significant (a) picture type effect for moving-average type-token ratio, words per minute (WPM), and mean length of utterance in morphemes (MLU) and (b) age group effect for WPM, MLU, and complexity index.

CONCLUSIONS

This study suggests that a culturally related high-context color photograph is the optimal choice for the Bangla PDT. This study also indicates reduced fluency, grammatical complexity, and syntactic complexity in healthy Bangla-speaking adults aged 65 years and above.

SUPPLEMENTAL MATERIAL

https://doi.org/10.23641/asha.22233664.

Structural invariants and semantic fingerprints in the "ego network" of words

Well-established cognitive models coming from anthropology have shown that, due to the cognitive constraints that limit our "bandwidth" for social interactions, humans organize their social relations according to a regular structure. In this work, we postulate that similar regularities can be found in other cognitive processes, such as those involving language production. In order to investigate this claim, we analyse a dataset containing tweets of a heterogeneous group of Twitter users (regular users and professional writers). Leveraging a methodology similar to the one used to uncover the well-established social cognitive constraints, we find regularities at both the structural and semantic levels. In the former, we find that a concentric layered structure (which we call ego network of words, in analogy to the ego network of social relationships) very well captures how individuals organise the words they use. The size of the layers in this structure regularly grows (approximately 2-3 times with respect to the previous one) when moving outwards, and the two penultimate external layers consistently account for approximately 60% and 30% of the used words, irrespective of the number of layers of the user. For the semantic analysis, each ring of each ego network is described by a semantic profile, which captures the topics associated with the words in the ring. We find that ring #1 has a special role in the model. It is semantically the most dissimilar and the most diverse among the rings. We also show that the topics that are important in the innermost ring also have the characteristic of being predominant in each of the other rings, as well as in the entire ego network. In this respect, ring #1 can be seen as the semantic fingerprint of the ego network of words.

Predicting Scores on the Mini-Mental State Examination (MMSE) from Spontaneous Speech

The aim of this study was to examine the relationship between language components, such as lexical density, length, and content in terms of “Time, Space and Action”, with MMSE scores. For this reason, a group of 33 older participants, without a diagnosis of dementia, was examined, providing information regarding recent and future events. Participants with higher MMSE scores showed higher lexical density, speech length, as well as number of tokens related to Time, Place and Action in their speech. However, these differences only reach the statistical level for lexical density when participants were divided into two groups (MCI and healthy controls). Word frequency was lower for participants with MCI but this difference was not statistically significant. Lastly, lexical density was positively correlated with MMSE scores and predicted MMSE scores. These results could be of interest at the applied level in the screening of MCI.

Health Professionals’ Experience Using an Azure Voice-Bot to Examine Cognitive Impairment (WAY2AGE)

Virtual Assistants (VA) are a new groundbreaking tool for screening cognitive impairment by healthcare professionals. By providing the volume of data needed in healthcare guidance, better treatment monitoring and optimization of costs are expected. One of the first steps in the development of these items is the experience of the healthcare professionals in their use. The general goal of the current project, WAY2AGE, is to examine healthcare professionals’ experience in using an Azure voice-bot for screening cognitive impairment. In this way, back-end services, such as the ChatBot, Speech Service and databases, are provided by the cloud platform Azure (Paas) for a pilot study. Most of the underlying scripts are implemented in Python, Net, JavaScript and open software. A sample of 30 healthcare workers volunteered to participate by answering a list of question in a survey set-up, following the example provided in the previous literature. Based on the current results, WAY2AGE was evaluated very positively in several categories. The main challenge of WAY2AGE is the articulation problems of some older people, which can lead to errors in the transcription of audio to text that will be addressed in the second phase. Following an analysis of the perception of a group of thirty health professionals on its usability, potential limitations and opportunities for future research are discussed.

Using Digital Speech Assessments to Detect Early Signs of Cognitive Impairment

Detecting early signs of cognitive decline is crucial for early detection and treatment of Alzheimer's Disease. Most of the current screening tools for Alzheimer's Disease represent a significant burden, requiring invasive procedures, or intensive and costly clinical testing. Recent findings have highlighted changes to speech and language patterns that occur in Alzheimer's Disease, and may be detectable prior to diagnosis. Automated tools to assess speech have been developed that can be used on a smartphone or tablet, from one's home, in under 10 min. In this study, we present the results of a study of older adults who completed a digital speech assessment task over a 6-month period. Participants were grouped according to those who scored above (N = 18) or below (N = 18) the recommended threshold for detecting cognitive impairment on the Montreal Cognitive Assessment (MoCA) and those with diagnoses of mild cognitive impairment (MCI) or early Alzheimer's Disease (AD) (N = 14). Older adults who scored above the MoCA threshold had better performance on speech composites reflecting language coherence, information richness, syntactic complexity, and word finding abilities. Those with MCI and AD showed more rapid decline in the coherence of language from baseline to 6-month follow-up, suggesting that this score may be useful both for detecting cognitive decline and monitoring change over time. This study demonstrates that automated speech assessments have potential as sensitive tools to detect early signs of cognitive impairment and monitor progression over time.

Tablet-Based Automatic Assessment for Early Detection of Alzheimer's Disease Using Speech Responses to Daily Life Questions

Health-monitoring technologies for automatically detecting the early signs of Alzheimer's disease (AD) have become increasingly important. Speech responses to neuropsychological tasks have been used for quantifying changes resulting from AD and differentiating AD and mild cognitive impairment (MCI) from cognitively normal (CN). However, whether and how other types of speech tasks with less burden on older adults could be used for detecting early signs of AD remains unexplored. In this study, we developed a tablet-based application and compared speech responses to daily life questions with those to neuropsychological tasks in terms of differentiating MCI from CN. We found that in daily life questions, around 80% of speech features showing significant differences between CN and MCI overlapped those showing significant differences in both our study and other studies using neuropsychological tasks, but the number of significantly different features as well as their effect sizes from life questions decreased compared with those from neuropsychological tasks. On the other hand, the results of classification models for detecting MCI by using the speech features showed that daily life questions could achieve high accuracy, i.e., 86.4%, comparable to neuropsychological tasks by using eight questions against all five neuropsychological tasks. Our results indicate that, while daily life questions may elicit weaker but statistically discernable differences in speech responses resulting from MCI than neuropsychological tasks, combining them could be useful for detecting MCI with comparable performance to using neuropsychological tasks, which could help develop health-monitoring technologies for early detection of AD in a less burdensome manner.

Psycholinguistic Screening for Cognitive Decline in Cancer Survivors: A Feasibility Study

OBJECTIVES

To test the feasibility of using psycholinguistic speech analysis as a proxy for cognitive function in men undergoing treatment for prostate cancer.

SAMPLE & SETTING

Audio-recorded speech samples were collected from 13 men enrolled in a parent study at the University of Kansas Cancer Center in Kansas City.

METHODS & VARIABLES

Audio-recorded speech samples, collected from clinical interviews and in response to a prompt question during the parent study at two time points, were evaluated to determine feasibility relationships between neurocognitive and psycholinguistic measures.

RESULTS

Correlations between neurocognitive and psycholinguistic measures were identified for prompted speech, but the strength of relationships varied between time points. No relationships were identified in clinical interview speech samples.

IMPLICATIONS FOR NURSING

Feasibility was demonstrated for recording, transcribing, and analyzing speech from clinical interviews, and results suggest relationships between neurocognitive and psycholinguistic measures in prompted speech. If validated, psycholinguistic assessments may be used to assess cognitive function in cancer survivors. Advances in natural language processing may provide opportunities for automated speech analyses for cancer treatment-related cognitive decline.

Using Speech Data From Interactions With a Voice Assistant to Predict the Risk of Future Accidents for Older Drivers: Prospective Cohort Study

Background

With the rapid growth of the older adult population worldwide, car accidents involving this population group have become an increasingly serious problem. Cognitive impairment, which is assessed using neuropsychological tests, has been reported as a risk factor for being involved in car accidents; however, it remains unclear whether this risk can be predicted using daily behavior data.

Objective

The objective of this study was to investigate whether speech data that can be collected in everyday life can be used to predict the risk of an older driver being involved in a car accident.

Methods

At baseline, we collected (1) speech data during interactions with a voice assistant and (2) cognitive assessment data—neuropsychological tests (Mini-Mental State Examination, revised Wechsler immediate and delayed logical memory, Frontal Assessment Battery, trail making test-parts A and B, and Clock Drawing Test), Geriatric Depression Scale, magnetic resonance imaging, and demographics (age, sex, education)—from older adults. Approximately one-and-a-half years later, we followed up to collect information about their driving experiences (with respect to car accidents) using a questionnaire. We investigated the association between speech data and future accident risk using statistical analysis and machine learning models.

Results

We found that older drivers (n=60) with accident or near-accident experiences had statistically discernible differences in speech features that suggest cognitive impairment such as reduced speech rate (P=.048) and increased response time (P=.040). Moreover, the model that used speech features could predict future accident or near-accident experiences with 81.7% accuracy, which was 6.7% higher than that using cognitive assessment data, and could achieve up to 88.3% accuracy when the model used both types of data.

Conclusions

Our study provides the first empirical results that suggest analysis of speech data recorded during interactions with voice assistants could help predict future accident risk for older drivers by capturing subtle impairments in cognitive function.

Scaling laws in natural conversations among elderly people

Language is a result of brain function; thus, impairment in cognitive function can result in language disorders. Understanding the aging of brain functions in terms of language processing is crucial for modern aging societies. Previous studies have shown that language characteristics, such as verbal fluency, are associated with cognitive functions. However, the scaling laws in language in elderly people remain poorly understood. In the current study, we recorded large-scale data of one million words from group conversations among healthy elderly people and analyzed the relationship between spoken language and cognitive functions in terms of scaling laws, namely, Zipf's law and Heaps' law. We found that word patterns followed these scaling laws irrespective of cognitive function, and that the variations in Heaps' exponents were associated with cognitive function. Moreover, variations in Heaps' exponents were associated with the ratio of new words taken from the other participants' speech. These results indicate that the exponents of scaling laws in language are related to cognitive processes.

"What should I do when I get home?" treatment plan discussion at discharge between specialist physicians and older in-patients: mixed method study

Background

During discharge from hospital, older patients and physicians discuss the plan for managing patients’ health at home. If not followed at home, it can result in poor medication management, readmissions, or other adverse events. Comorbidities, polypharmacy and cognitive impairment may create challenges for older patients. We assessed discharge conversations between older in-patients and physicians for treatment plan activities and medication information, with emphasis on the role of cognitive function in the ongoing conversation.

Methods

We collected 11 videos of discharge consultations, medication lists, and self-reported demographic information from hospitalised patients ≥65 years at the Geriatric department in a general hospital. Mini Mental State Examination score < 25 was classified as low cognitive function. We used microanalysis of face-to-face dialogue to identify and characterise sequences of interaction focused on and distinguishing the treatment plan activities discussed. In addition to descriptive statistics, we used a paired-sample t-test and Mann-Whitney U test for non-parametric data.

Results

Patients’ median age was 85 (range: 71–90);7 were females and 4 males. Median of 17 (range: 7 to 23) treatment plan activities were discussed. The proportions of the activities, grouped from a patient perspective, were: 0.40 my medications, 0.21 something the hospital will do for me, 0.18 someone I visit away from home, 0.12 daily routine and 0.09 someone coming to my home. Patients spoke less (mean 190.9 words, SD 133.9) during treatment plan activities compared to other topics (mean 759 words, SD 480.4), (p = .001). Patients used on average 9.2 (SD 3.1) medications; during the conversations, an average of 4.5 (SD 3.3) were discussed, and side effects discussed on average 1.2 (SD 2.1) times. During treatment plan discussions, patients with lower cognitive function were less responsive and spoke less (mean 116.5 words, SD 40.9), compared to patients with normal cognition (mean 233.4 words, SD 152.4), (p = .089).

Conclusion

Physicians and geriatric patients discuss many activities during discharge conversations, mostly focusing on medication use without stating side effects. Cognitive function might play a role in how older patients respond. These results may be useful for an intervention to improve communication between physicians and older hospitalised patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12913-020-05860-9.

Predicting Mini-Mental Status Examination Scores through Paralinguistic Acoustic Features of Spontaneous Speech

Speech analysis could provide an indicator of cognitive health and help develop clinical tools for automatically detecting and monitoring cognitive health progression. The Mini Mental Status Examination (MMSE) is the most widely used screening tool for cognitive health. But the manual operation of MMSE restricts its screening within primary care facilities. An automatic screening tool has the potential to remedy this situation. This study aims to assess the association between acoustic features of spontaneous speech and assess whether acoustic features can be used to automatically predict MMSE score. We assessed the effectiveness of paralinguistic feature set for MMSE score prediction on a balanced sample of DementiaBank's Pitt spontaneous speech dataset, with patients matched by gender and age. Linear regression analysis shows that fusion of acoustic features, age, sex and years of education provides better results (mean absolute error, MAE = 4.97, and R² = 0.261) than acoustic features alone (MAE = 5.66 and R² =0.125) and age, gender and education level alone (MAE of 5.36 and R² =0.17). This suggests that the acoustic features of spontaneous speech are an important part of an automatic screening tool for cognitive impairment detection.Clinical relevance- We hereby present a method for automatic screening of cognitive health. It is based on acoustic information of speech, a ubiquitous source of data, therefore being cost-efficient, non-invasive and with little infrastructure required.

Narrative video scene description task discriminates between levels of cognitive impairment in Alzheimer's disease

The process of interpreting and acting upon the visual environment requires both intact cognitive and visual systems. The narrative description (ND) task, initially developed to detect changes in ecologically relevant visual function in people with impaired vision, is an objective measure of the ability to perceive, understand, and describe a visual scene in a movie clip.

OBJECTIVE

Because the ND task draws heavily on semantic and working memory ability in addition to basic visual perception, we aimed to assess the discriminative performance of this task across levels of cognitive impairment.

METHOD

We recruited 56 participants with cognitive status ranging from normal cognition to mild dementia (median age 82, range 66 to 99 years) to watch 20 30-s video clips and describe the visual content without time constraints. These verbal responses were transcribed and processed to generate ND shared word scores using a "wisdom of the crowd," natural-language processing approach. We compared ND scores across diagnostic groups, and used linear mixed models to examine decrements in task performance.

RESULTS

There was a stepwise decline of ND scores with increasing levels of cognitive impairment. Additional analyses showed that ND performance was highly related to performance on the Montreal Cognitive Assessment (MoCA) and domain-specific neuropsychological tests for semantic fluency and set shifting. Other models demonstrated differences in ND performance related video content between cognitively normal and impaired participants.

CONCLUSION

The ND test was able to detect decrements in task performance between levels of cognitive impairment and was related to other global neuropsychological measures. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

A Systematic Review and Implementation of IoT-Based Pervasive Sensor-Enabled Tracking System for Dementia Patients

In today's world, 46.8 million people suffer from brain related diseases. Dementia is most prevalent of all. In general scenario, a dementia patient lacks proper guidance in searching out the way to return back at his/her home. Thus, increasing the risk of getting damaged at individual-health level. Therefore, it is important to track their movement in more sophisticated manner as possible. With emergence of wearables, GPS sensors and Internet of Things (IoT), such devices have become available in public domain. Smartphone apps support caregiver to locate the dementia patients in real-time. RF, GSM, 3G, Wi-Fi and 4G technology fill the communication gap between patient and caregiver to bring them closer. In this paper, we incorporated 7 most popular wearables for investigation to seek appropriateness for dementia tracking in recent times in systematic manners. We performed an in-depth review of these wearables as per the cost, technology wise and application wise characteristics. A case novel study i.e. IoT-based Force Sensor Resistance enabled System-FSRIoT, has been proposed and implemented to validate the effectiveness of IoT in the domain of smarter dementia patient tracking in wearable form factor. The results show promising aspect of a whole new notion to leverage efficient assistive physio-medical healthcare to the dementia patients and the affected family members to reduce life risks and achieve a better social life.

Using Tablet-Based Assessment to Characterize Speech for Individuals with Dementia and Mild Cognitive Impairment: Preliminary Results

Early detection of dementia as well as improvement in diagnosis coverage has been increasingly important. Previous studies involved extracting speech features during neuropsychological assessments by humans, such as medical pro- fessionals, and succeeded in detecting patients with dementia and mild cognitive impairment (MCI). Enabling such assessment in an automated fashion by using computer devices would extend the range of application. In this study, we developed a tablet-based application for neuropsychological assessments and collected speech data from 44 Japanese native speakers including healthy controls (HCs) and those with MCI and dementia. We first extracted acoustic and phonetic features and showed that several features exhibited significant difference between HC vs. MCI and HC vs. dementia. We then constructed classification models by using these features and demonstrated that these models could differentiate MCI and dementia from HC with up to 82.4 and 92.6% accuracy, respectively.

Subjective rating scale for discourse: Evidence from the efficacy of subjective rating scale in amnestic mild cognitive impairments

In clinical settings, the language ability of patients with neurologic communication disorders have been measured by quantitative parameters such as the total number of words in dialogue and picture description tasks. However, this quantitative analysis requires a long period of time in order to analyze the quantitative parameters, and results can differ according to discourse tasks. The purposes of this study are to explore whether SR-D may predict the quantitative measures of discourse tasks. Forty patients with amnestic MCI and 40 normal elderly participated in the study. We gathered responses to 10 items regarding SR-D and analyzed the quantitative measures of narrative discourse through 3 discourse tasks (i.e, picture description, dialogue, procedural discourse). We found significant differences in MLTW, CIU, and SR-D scores between the 2 groups. In particular, 4 items were significantly correlated with the performance of MLTW and CIU. Sensitivity and specificity of these 4 items were 100% and 75%, respectively. In terms of economic opportunity costs, objective measures cannot be evaluated to be practical, since it is used in research rather than clinical diagnosis in general. Therefore, evaluation of discourse using a few items proven in its sensitivity and specificity could allow a wide use of such measure in not only research but also in clinical diagnosis. These findings suggest that subjective measures of narrative discourse may be valid with objective language tests to predict individual discourse performance.

Connected speech and language in mild cognitive impairment and Alzheimer's disease: A review of picture description tasks

INTRODUCTION

The neuropsychological profile of people with mild cognitive impairment (MCI) and Alzheimer's disease (AD) dementia includes a history of decline in memory and other cognitive domains, including language. While language impairments have been well described in AD dementia, language features of MCI are less well understood. Connected speech and language analysis is the study of an individual's spoken discourse, usually elicited by a target stimulus, the results of which can facilitate understanding of how language deficits typical of MCI and AD dementia manifest in everyday communication. Among discourse genres, picture description is a constrained task that relies less on episodic memory and more on semantic knowledge and retrieval, within the cognitive demands of a communication context. Understanding the breadth of evidence across the continuum of cognitive decline will help to elucidate the areas of strength and need in terms of using this method as an evaluative tool for both cognitive changes and everyday functional communication.

METHOD

We performed an extensive literature search of peer-reviewed journal articles that focused on the use of picture description tasks for evaluating language in persons with MCI or AD dementia. We selected articles based on inclusion and exclusion criteria and described the measures assessed, the psychometric properties that were reported, the findings, and the limitations of the included studies.

RESULTS

36 studies were selected and reviewed. Across all 36 studies, there were 1, 127 patients with AD dementia and 274 with MCI or early cognitive decline. Multiple measures were examined, including those describing semantic content, syntactic complexity, speech fluency, vocal parameters, and pragmatic language. Discriminant validity widely reported and distinct differences in language were observable between adults with dementia and controls; fewer studies were able to distinguish language differences between typically aging adults and those with MCI.

DISCUSSION

Our review shows that picture description tasks are useful tools for detecting differences in a wide variety of language and communicative measures. Future research should expand knowledge about subtle changes to language in preclinical AD and Mild Cognitive Impairment (MCI) which may improve the utility of this method as a clinically meaningful screening tool.

Declines in Connected Language Are Associated with Very Early Mild Cognitive Impairment: Results from the Wisconsin Registry for Alzheimer's Prevention

Changes to everyday spoken language ("connected language") are evident in persons with AD dementia, yet little is known about when these changes are first detectable on the continuum of cognitive decline. The aim of this study was to determine if participants with very early, subclinical memory declines were also showing declines in connected language. We analyzed connected language samples obtained from a simple picture description task at two time points in 264 participants from the Wisconsin Registry for Alzheimer's Prevention (WRAP). In parallel, participants were classified as either "Cognitively Healthy" or "Early Mild Cognitive Impairment" based on longitudinal neuropsychological test performance. Linear mixed effects models were used to analyze language parameters that were extracted from the connected language samples using automated feature extraction. Participants with eMCI status declined faster in features of speech fluency and semantic content than those who were cognitively stable. Measures of lexical diversity and grammatical complexity were not associated with eMCI status in this group. These findings provide novel insights about the relationship between cognitive decline and everyday language, using a quick, inexpensive, and performance-based method.

Detecting Dementia Through Interactive Computer Avatars

This paper proposes a new approach to automatically detect dementia. Even though some works have detected dementia from speech and language attributes, most have applied detection using picture descriptions, narratives, and cognitive tasks. In this paper, we propose a new computer avatar with spoken dialog functionalities that produces spoken queries based on the mini-mental state examination, the Wechsler memory scale-revised, and other related neuropsychological questions. We recorded the interactive data of spoken dialogues from 29 participants (14 dementia and 15 healthy controls) and extracted various audiovisual features. We tried to predict dementia using audiovisual features and two machine learning algorithms (support vector machines and logistic regression). Here, we show that the support vector machines outperformed logistic regression, and by using the extracted features they classified the participants into two groups with 0.93 detection performance, as measured by the areas under the receiver operating characteristic curve. We also newly identified some contributing features, e.g., gap before speaking, the variations of fundamental frequency, voice quality, and the ratio of smiling. We concluded that our system has the potential to detect dementia through spoken dialog systems and that the system can assist health care workers. In addition, these findings could help medical personnel detect signs of dementia.