Voice biomarkers as indicators of cognitive changes in middle and later adulthood

Objective biomarkers of cognitive performance in older adults with mild cognitive impairment: acoustic features of affective prosody

OBJECTIVES

To explore negative affective prosody production patterns in older adults with mild cognitive impairment (MCI) and identify acoustic features as objective biomarkers for cognitive performances.

METHODS

We recruited 42 older adults with MCI and 29 healthy controls (HC). The cognitive performances were assessed with the auditory verbal learning test-HuaShan version, the Trail Making Test A, the Digit Span-Forward, and the Digit Span-Backward. Acoustic features were extracted with the PRAAT software.

RESULTS

There were significant differences between older adults with MCI and HC on anger sentences (t=-2.14, p=0.036, Cohen's d=0.52) and disgust sentences (Z=1.97, p=0.049, |r|=0.23). Duration in fearful sentences (β=-0.511, p=0.002) and degree of voice breaks in angry sentences (β=0.423, p=0.009) were identified as objective biomarkers for predicting verbal memory.

CONCLUSIONS

Older adults with MCI have deficits in the production pattern of angry and disgusted prosody. Clinicians could assess verbal memory decline by temporal features in negative emotional prosody.

Alzheimer disease seen through the lens of sex and gender

Alzheimer disease (AD) is a life-limiting neurodegenerative disorder that disproportionately affects women. Indeed, sex and gender are emerging as crucial modifiers of diagnostic and therapeutic pathways in AD. This Review provides an overview of the interactions of sex and gender with important developments in AD and offers insights into priorities for future research to facilitate the development and implementation of personalized approaches in the shifting paradigm of AD care. In particular, this Review focuses on the influence of sex and gender on important advances in the treatment and diagnosis of AD, including disease-modifying therapies, fluid-based biomarkers, cognitive assessment tools and multidomain lifestyle interventional studies.

Precision neurocognition: An emerging diagnostic paradigm leveraging digital cognitive assessment technology

Strategies that may modify Alzheimer's disease (AD) and other dementia disorders are being developed. To maximize the benefits of these strategies, it is critical that indicators suggesting neurocognitive decline are identified as early as possible. 'Precision neurocognition' is a heuristic that seeks to develop methodologies capable of identifying subtle behavior(s) that may flag emerging AD and other dementia related syndromes. Recent research suggests that digital neuropsychological assessment technology may be the platform that can realize the goals of precision neurocognition, i.e., the early detection of neurocognitive difficulties that are prognostic for mild cognitive impairment (MCI) and dementia. Past research associating 100% correct or statistically within normal limits responding using neuropsychological tests with time-based parameters obtained while participants undergo assessment is reviewed. Recent research with community dwelling and memory clinic participants examined test scores obtained using commonly available neuropsychological tests. This research extracted a number of discrete latency measures that clearly dissociate between groups, despite final test scores that are either 100% correct or statistically within normal limits. In sum, past research using digitally administered neuropsychological tests suggests that the goals of precision neurocognition as related to the early identification of neurodegenerative illness may be realized via an analysis of time derived, process-based behavior using digital assessment technology. Latency or time-based parameters as described in recent research could form the basis of a range of neurocognitive biomarkers for identifying people at risk for developing AD, other dementing disorders, and MCI.

Obfuscation via pitch-shifting for balancing privacy and diagnostic utility in voice-based cognitive assessment

INTRODUCTION

Digital voice analysis is an emerging tool for differentiating cognitive states, but it poses privacy risks as automated systems may inadvertently identify speakers.

METHODS

We developed a computational framework to evaluate the trade-off between voice obfuscation and cognitive assessment accuracy, using pitch-shifting as a representative method. This framework was applied to voice recordings from the Framingham Heart Study (FHS, n = 128) and the DementiaBank Delaware (DBD, n = 85) corpus, both featuring responses to neuropsychological tests. Speaker obfuscation was measured via equal error rate (EER), and diagnostic utility was assessed through machine learning models distinguishing cognitive states: normal cognition (NC), mild cognitive impairment (MCI), and dementia (DE).

RESULTS

With the top 20 acoustic features, our framework achieved classification accuracies of 62.2% (EER: 0.3335) on the FHS dataset for NC, MCI, and DE differentiation, and 63.7% (EER: 0.1796) on the DBD dataset for NC and MCI differentiation, using obfuscated speech files.

DISCUSSION

Our results demonstrate the feasibility of privacy-preserving voice markers, offering a scalable solution for voice-based cognitive assessments.

HIGHLIGHTS

We developed a computational framework using pitch-shifting and acoustic transformations to balance speaker privacy and diagnostic utility in voice-based cognitive assessments. We evaluated the framework on two independent datasets, Framingham Heart Study (FHS, n = 128) and DementiaBank Delaware (DBD, n = 85) corpus, assessing the trade-off between privacy (measured by equal error rate [EER]) and classification accuracy. Our framework achieved classification accuracies of 62.2% (EER: 0.3335) for distinguishing normal cognition (NC), mild cognitive impairment (MCI), and dementia in the FHS dataset and 63.7% (EER: 0.1796) for NC and MCI differentiation in the DBD dataset, using obfuscated speech files. Our framework demonstrates that pitch-shifting levels can preserve diagnostic utility while protecting speaker identity, offering a scalable and privacy-preserving solution.

Obfuscation via pitch-shifting for balancing privacy and diagnostic utility in voice-based cognitive assessment

Introduction

Digital voice analysis is gaining traction as a tool to differentiate cognitively normal from impaired individuals. However, voice data poses privacy risks due to the potential identification of speakers by automated systems.

Methods

We developed a framework that uses weighted linear interpolation of privacy and utility metrics to balance speaker obfuscation and cognitive integrity in cognitive assessments. This framework applies pitch-shifting for speaker obfuscation while preserving cognitive speech features. We tested it on digital voice recordings from the Framingham Heart Study (N=128) and Dementia Bank Delaware corpus (N=85), both containing responses to neuropsychological tests.

Results

The tool effectively obfuscated speaker identity while maintaining cognitive feature integrity, achieving an accuracy of 0.6465 in classifying individuals with normal cognition, mild cognitive impairment, and dementia in the FHS cohort.

Discussion

Our approach enables the development of digital markers for dementia assessment while protecting sensitive personal information, offering a scalable solution for privacy-preserving voice-based diagnostics.

Identifying and Estimating Frailty Phenotypes by Vocal Biomarkers: Cross-Sectional Study

BACKGROUND

Researchers have developed a variety of indices to assess frailty. Recent research indicates that the human voice reflects frailty status. Frailty phenotypes are seldom discussed in the literature on the aging voice.

OBJECTIVE

This study aims to examine potential phenotypes of frail older adults and determine their correlation with vocal biomarkers.

METHODS

Participants aged ≥60 years who visited the geriatric outpatient clinic of a teaching hospital in central Taiwan between 2020 and 2021 were recruited. We identified 4 frailty phenotypes: energy-based frailty, sarcopenia-based frailty, hybrid-based frailty-energy, and hybrid-based frailty-sarcopenia. Participants were asked to pronounce a sustained vowel "/a/" for approximately 1 second. The speech signals were digitized and analyzed. Four voice parameters-the average number of zero crossings (A1), variations in local peaks and valleys (A2), variations in first and second formant frequencies (A3), and spectral energy ratio (A4)-were used for analyzing changes in voice. Logistic regression was used to elucidate the prediction model.

RESULTS

Among 277 older adults, an increase in A1 values was associated with a lower likelihood of energy-based frailty (odds ratio [OR] 0.81, 95% CI 0.68-0.96), whereas an increase in A2 values resulted in a higher likelihood of sarcopenia-based frailty (OR 1.34, 95% CI 1.18-1.52). Respondents with larger A3 and A4 values had a higher likelihood of hybrid-based frailty-sarcopenia (OR 1.03, 95% CI 1.002-1.06) and hybrid-based frailty-energy (OR 1.43, 95% CI 1.02-2.01), respectively.

CONCLUSIONS

Vocal biomarkers might be potentially useful in estimating frailty phenotypes. Clinicians can use 2 crucial acoustic parameters, namely A1 and A2, to diagnose a frailty phenotype that is associated with insufficient energy or reduced muscle function. The assessment of A3 and A4 involves a complex frailty phenotype.

Voice biomarkers in middle and later adulthood as predictors of cognitive changes

Background

Prosody voice measures, especially jitter and shimmer, have been associated with cognitive impairment and hold potential as early indicators of risk for cognitive decline. Prior research suggests that voice measures assessed concurrently with longitudinal cognitive outcomes are associated with 10-year cognitive declines in middle-age and older adults from Midlife in the U.S. (MIDUS) study.

Results

Using a subsample from the same study, we expanded previous research to examine voice measures that were (1) collected 8 years before cognitive outcomes, (2) derived from narrative speech in logical memory tests instead of word list recall tests, and (3) independent of the cognitive outcomes. Multilevel analyses controlled for covariates of age, sex, education, neurological conditions, depressive symptoms, and chronic conditions. The results indicated that higher jitter and lower shimmer predicted greater 10-year declines in episodic memory and working memory.

Conclusion

These findings extend previous research by highlighting prosody voice measures assessed 8 years earlier as predictors of subsequent cognitive declines over a decade.

Digital assessment of cognition in neurodegenerative disease: a data driven approach leveraging artificial intelligence

Introduction

A rapid and reliable neuropsychological protocol is essential for the efficient assessment of neurocognitive constructs related to emergent neurodegenerative diseases. We developed an AI-assisted, digitally administered/scored neuropsychological protocol that can be remotely administered in ~10 min. This protocol assesses the requisite neurocognitive constructs associated with emergent neurodegenerative illnesses.

Methods

The protocol was administered to 77 ambulatory care/memory clinic patients (56.40% women; 88.50% Caucasian). The protocol includes a 6-word version of the Philadelphia (repeatable) Verbal Learning Test [P(r)VLT], three trials of 5 digits backward from the Backwards Digit Span Test (BDST), and the "animal" fluency test. The protocol provides a comprehensive set of traditional "core" measures that are typically obtained through paper-and-pencil tests (i.e., serial list learning, immediate and delayed free recall, recognition hits, percent correct serial order backward digit span, and "animal" fluency output). Additionally, the protocol includes variables that quantify errors and detail the processes used in administering the tests. It also features two separate, norm-referenced summary scores specifically designed to measure executive control and memory.

Results

Using four core measures, we used cluster analysis to classify participants into four groups: cognitively unimpaired (CU; n = 23), amnestic mild cognitive impairment (MCI; n = 17), dysexecutive MCI (n = 23), and dementia (n = 14). Subsequent analyses of error and process variables operationally defined key features of amnesia (i.e., rapid forgetting, extra-list intrusions, profligate responding to recognition foils); key features underlying reduced executive abilities (i.e., BDST items and dysexecutive errors); and the strength of the semantic association between successive responses on the "animal" fluency test. Executive and memory index scores effectively distinguished between all four groups. There was over 90% agreement between how cluster analysis of digitally obtained measures classified patients compared to classification using a traditional comprehensive neuropsychological protocol. The correlations between digitally obtained outcome variables and analogous paper/pencil measures were robust.

Discussion

The digitally administered protocol demonstrated a capacity to identify patterns of impaired performance and classification similar to those observed with standard paper/pencil neuropsychological tests. The inclusion of both core measures and detailed error/process variables suggests that this protocol can detect subtle, nuanced signs of early emergent neurodegenerative illness efficiently and comprehensively.

Digital Neuropsychology beyond Computerized Cognitive Assessment: Applications of Novel Digital Technologies

Compared with other health disciplines, there is a stagnation in technological innovation in the field of clinical neuropsychology. Traditional paper-and-pencil tests have a number of shortcomings, such as low-frequency data collection and limitations in ecological validity. While computerized cognitive assessment may help overcome some of these issues, current computerized paradigms do not address the majority of these limitations. In this paper, we review recent literature on the applications of novel digital health approaches, including ecological momentary assessment, smartphone-based assessment and sensors, wearable devices, passive driving sensors, smart homes, voice biomarkers, and electronic health record mining, in neurological populations. We describe how each digital tool may be applied to neurologic care and overcome limitations of traditional neuropsychological assessment. Ethical considerations, limitations of current research, as well as our proposed future of neuropsychological practice are also discussed.

Using voice biomarkers for frailty classification

Clinicians use the patient's voice intuitively to evaluate general health and frailty. Voice is an emerging health indicator but has been scarcely studied in the context of frailty. This study explored voice parameters as possible predictors of frailty in older adults. Fifty-three participants over 70 years old were recruited from rehabilitation wards at a tertiary medical center. Participants' frailty was assessed using Rockwood frailty index and they were classified as most-frail (n = 33, 68%) or less-frail (n = 20, 32%). Participants were recorded counting from 1 to 10 and backwards using a smartphone recording application. The following voice biomarkers were derived: peak and average volume, peak/average volume ratio, pauses' total length, and pause length standard deviation. The most-frail group had a higher peak volume/average volume ratio (p = 0.03) and greater variance in lengths of pauses between speech segments (p = 0.002). These parameters indicate greater speech irregularity in the most-frail, compared to the less-frail. The most-frail group also had a longer total duration of pauses (p = 0.02). No statistically significant difference was found in peak and average volume (p = 0.75 and 0.39). Most-frail participants' speech had different characteristics, compared to participants in the less-frail group. This is a first step to developing an AI-based frailty assessment tool that can assist in identifying our most vulnerable patients.

Early ultrasonic vocalization deficits and related thyroarytenoid muscle pathology in the transgenic TgF344-AD rat model of Alzheimer's disease

Background

Alzheimer's disease (AD) is a progressive neurologic disease and the most common cause of dementia. Classic pathology in AD is characterized by inflammation, abnormal presence of tau protein, and aggregation of β-amyloid that disrupt normal neuronal function and lead to cell death. Deficits in communication also occur during disease progression and significantly reduce health, well-being, and quality of life. Because clinical diagnosis occurs in the mid-stage of the disease, characterizing the prodrome and early stages in humans is currently challenging. To overcome these challenges, we use the validated TgF344-AD (F344-Tg(Prp-APP, Prp-PS1)19/Rrrc) transgenic rat model that manifests cognitive, behavioral, and neuropathological dysfunction akin to AD in humans.

Objectives

The overarching goal of our work is to test the central hypothesis that pathology and related behavioral deficits such as communication dysfunction in part manifest in the peripheral nervous system and corresponding target tissues already in the early stages. The primary aims of this study are to test the hypotheses that: (1) changes in ultrasonic vocalizations (USV) occur in the prodromal stage at 6 months of age and worsen at 9 months of age, (2) inflammation as well as AD-related pathology can be found in the thyroarytenoid muscle (TA) at 12 months of age (experimental endpoint tissue harvest), and to (3) demonstrate that the TgF344-AD rat model is an appropriate model for preclinical investigations of early AD-related vocal deficits.

Methods

USVs were collected from male TgF344-AD (N = 19) and wildtype (WT) Fischer-344 rats (N = 19) at 6 months (N = 38; WT: n = 19; TgF344-AD: n = 19) and 9 months of age (N = 18; WT: n = 10; TgF344-AD: n = 8) and acoustically analyzed for duration, mean power, principal frequency, low frequency, high frequency, peak frequency, and call type. RT-qPCR was used to assay peripheral inflammation and AD-related pathology via gene expressions in the TA muscle of male TgF344-AD rats (n = 6) and WT rats (n = 6) at 12 months of age.

Results

This study revealed a significant reduction in mean power of ultrasonic calls from 6 to 9 months of age and increased peak frequency levels over time in TgF344-AD rats compared to WT controls. Additionally, significant downregulation of AD-related genes Uqcrc2, Bace2, Serpina3n, and Igf2, as well as downregulation of pro-inflammatory gene Myd88 was found in the TA muscle of TgF344-AD rats at 12 months of age.

Discussion

Our findings demonstrate early and progressive vocal deficits in the TgF344-AD rat model. We further provide evidence of dysregulation of AD-pathology-related genes as well as inflammatory genes in the TA muscles of TgF344-AD rats in the early stage of the disease, confirming this rat model for early-stage investigations of voice deficits and related pathology.

Harnessing the Power of Voice: A Deep Neural Network Model for Alzheimer's Disease Detection

Background and Purpose

Voice, reflecting cerebral functions, holds potential for analyzing and understanding brain function, especially in the context of cognitive impairment (CI) and Alzheimer's disease (AD). This study used voice data to distinguish between normal cognition and CI or Alzheimer's disease dementia (ADD).

Methods

This study enrolled 3 groups of subjects: 1) 52 subjects with subjective cognitive decline; 2) 110 subjects with mild CI; and 3) 59 subjects with ADD. Voice features were extracted using Mel-frequency cepstral coefficients and Chroma.

Results

A deep neural network (DNN) model showed promising performance, with an accuracy of roughly 81% in 10 trials in predicting ADD, which increased to an average value of about 82.0%±1.6% when evaluated against unseen test dataset.

Conclusions

Although results did not demonstrate the level of accuracy necessary for a definitive clinical tool, they provided a compelling proof-of-concept for the potential use of voice data in cognitive status assessment. DNN algorithms using voice offer a promising approach to early detection of AD. They could improve the accuracy and accessibility of diagnosis, ultimately leading to better outcomes for patients.

Voice analysis results in individuals with Alzheimer's disease: How do age and cognitive status affect voice parameters?

BACKGROUND

Reports of acoustic changes in the voice in individuals with Alzheimer's disease (AD) and the relationship of acoustic changes with age and cognitive status are still limited.

OBJECTIVE

This study aims to determine the changes in voice analysis results in AD, as well as the effects of age and cognitive status on voice parameters.

METHODS

The study included 47 (AD: 30; healthy: 17) women with a mean age of 76.13 years. The acoustic voice parameters mean fundamental frequency (F0), relative average perturbation (RAP), jitter percent (Jitt), shimmer percent (Shim), and noise-to-harmonic ratio were detected. The mini-mental state examination (MMSE) was utilized.

RESULTS

F0, Shim, Jitt, and RAP values were found to be statistically significantly higher in individuals with AD compared to healthy individuals. There was a significant negative correlation between MMSE and F0, Jitt, RAP and Shim, and the MMSE score had a significant negative effect on F0, Jitt, and RAP (p < .05).

CONCLUSION

Cognitive status was discovered to significantly impact the voice, with fundamental frequency and frequency and amplitude perturbations increasing as cognitive level decreases. In order to contribute to the therapy process for voice disorders, cognitive functions can be focused on in addition to voice therapy.

Fusion of Low-Level Descriptors of Digital Voice Recordings for Dementia Assessment

Digital voice recordings can offer affordable, accessible ways to evaluate behavior and function. We assessed how combining different low-level voice descriptors can evaluate cognitive status. Using voice recordings from neuropsychological exams at the Framingham Heart Study, we developed a machine learning framework fusing spectral, prosodic, and sound quality measures early in the training cycle. The model's area under the receiver operating characteristic curve was 0.832 (±0.034) in differentiating persons with dementia from those who had normal cognition. This offers a data-driven framework for analyzing minimally processed voice recordings for cognitive assessment, highlighting the value of digital technologies in disease detection and intervention.

Broadened assessments, health education and cognitive aids in the remote memory clinic

The prevalence of dementia is increasing and poses a health challenge for individuals and society. Despite the desire to know their risks and the importance of initiating early therapeutic options, large parts of the population do not get access to memory clinic-based assessments. Remote memory clinics facilitate low-level access to cognitive assessments by eschewing the need for face-to-face meetings. At the same time, patients with detected impairment or increased risk can receive non-pharmacological treatment remotely. Sensor technology can evaluate the efficiency of this remote treatment and identify cognitive decline. With remote and (partly) automatized technology the process of cognitive decline can be monitored but more importantly also modified by guiding early interventions and a dementia preventative lifestyle. We highlight how sensor technology aids the expansion of assessments beyond cognition and to other domains, e.g., depression. We also illustrate applications for aiding remote treatment and describe how remote tools can facilitate health education which is the cornerstone for long-lasting lifestyle changes. Tools such as transcranial electric stimulation or sleep-based interventions have currently mostly been used in a face-to-face context but have the potential of remote deployment-a step already taken with memory training apps. Many of the presented methods are readily scalable and of low costs and there is a range of target populations, from the worried well to late-stage dementia.