Dataset Information

Development of digital voice biomarkers and associations with cognition, cerebrospinal biomarkers, and neural representation in early Alzheimer's disease.

ABSTRACT:

Introduction

Advances in natural language processing (NLP), speech recognition, and machine learning (ML) allow the exploration of linguistic and acoustic changes previously difficult to measure. We developed processes for deriving lexical-semantic and acoustic measures as Alzheimer's disease (AD) digital voice biomarkers.

Methods

We collected connected speech, neuropsychological, neuroimaging, and cerebrospinal fluid (CSF) AD biomarker data from 92 cognitively unimpaired (40 Aβ+) and 114 impaired (63 Aβ+) participants. Acoustic and lexical-semantic features were derived from audio recordings using ML approaches.

Results

Lexical-semantic (area under the curve [AUC] = 0.80) and acoustic (AUC = 0.77) scores demonstrated higher diagnostic performance for detecting MCI compared to Boston Naming Test (AUC = 0.66). Only lexical-semantic scores detected amyloid-β status (p = 0.0003). Acoustic scores associated with hippocampal volume (p = 0.017) while lexical-semantic scores associated with CSF amyloid-β (p = 0.007). Both measures were significantly associated with 2-year disease progression.

Discussion

These preliminary findings suggest that derived digital biomarkers may identify cognitive impairment in preclinical and prodromal AD, and may predict disease progression.

Highlights

This study derived lexical-semantic and acoustics features as Alzheimer's disease (AD) digital biomarkers.These features were derived from audio recordings using machine learning approaches.Voice biomarkers detected cognitive impairment and amyloid-β status in early stages of AD.Voice biomarkers may predict Alzheimer's disease progression.These markers significantly mapped to functional connectivity in AD-susceptible brain regions.

SUBMITTER: Hajjar I

PROVIDER: S-EPMC9899764 | biostudies-literature | 2023 Jan-Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Development of digital voice biomarkers and associations with cognition, cerebrospinal biomarkers, and neural representation in early Alzheimer's disease.

Hajjar Ihab I Okafor Maureen M Choi Jinho D JD Moore Elliot E Abrol Anees A Calhoun Vince D VD Goldstein Felicia C FC

Alzheimer's & dementia (Amsterdam, Netherlands) 20230101 1

<h4>Introduction</h4>Advances in natural language processing (NLP), speech recognition, and machine learning (ML) allow the exploration of linguistic and acoustic changes previously difficult to measure. We developed processes for deriving lexical-semantic and acoustic measures as Alzheimer's disease (AD) digital voice biomarkers.<h4>Methods</h4>We collected connected speech, neuropsychological, neuroimaging, and cerebrospinal fluid (CSF) AD biomarker data from 92 cognitively unimpaired (40 Aβ+) ...[more]

PMID: 36777093

Similar Datasets

Project description:BackgroundAfrican Americans have been reported to have a higher prevalence of Alzheimer's disease (AD) than Caucasians, but etiology-specific AD biomarkers have not been systematically analyzed in older African Americans. Coexisting cerebrovascular disease may also contribute to this increased prevalence. We hypothesized that cerebrospinal fluid (CSF) biomarkers of amyloid, neurodegeneration, and endothelial dysfunction would differ between older African Americans and Caucasians with normal cognition and cognitive impairment associated with AD.MethodsWe prospectively recruited 135 older Americans to undergo detailed clinical, neuropsychological, genetic, magnetic resonance imaging (MRI), and CSF analysis from 2013 to 2015 at Emory University (Atlanta, GA, USA). We compared levels of CSF markers for β-amyloid (Aβ42, Aβ40), total and phosphorylated tau (t-tau and p-tau181, respectively), endothelial dysfunction (soluble vascular cell adhesion molecule 1, soluble intercellular adhesion molecule 1), α-synuclein, and neurodegeneration (neurofilament light chain [NfL]), as well as MRI markers, for hippocampal atrophy and cerebrovascular disease (white matter hyperintensity [WMH] volume).ResultsSixty-five older African Americans (average age, 69.1 years) and 70 older Caucasians (average age, 70.8 years) were included. After adjusting for demographic variables, AD risk alleles, and cognitive function, older African Americans had lower CSF levels of p-tau181 (difference of 7.4 pg/ml; 95% CI, 3.7-11.2 pg/ml; p < 0.001), t-tau (difference of 23.6 pg/ml; 95% CI, 9.5-37.7; p = 0.001), and Aβ40 (difference of 1.35 ng/ml; 95% CI, 0.29-2.42 ng/ml; p = 0.013) despite similar levels of Aβ42, NfL, WMH volume, and hippocampal volume. Cognitively impaired African Americans also had lower CSF t-tau/Aβ42 (difference of 0.255 per 1-SD change in composite cognition; 95% CI, 0.100-0.409; p = 0.001) and p-tau181/Aβ42 (difference of 0.076 per 1-SD change in composite cognition; 95% CI, 0.031-0.122; p = 0.001). These could not be explained by measured biomarkers of non-AD processes, but African Americans may be more susceptible than Caucasians to the cognitive effects of WMH.ConclusionsDespite comparable levels of CSF Aβ42 and Aβ42/Aβ40, cognitive impairment in African Americans is associated with smaller changes in CSF tau markers but greater impact from similar WMH burden than Caucasians. Race-associated differences in CSF tau markers and ratios may lead to underdiagnosis of AD in African Americans.Trial registrationClinicalTrials.gov, NCT02089555 . Retrospectively registered on 14 March 2014.

Project description:BackgroundIn Parkinson's disease (PD), levels of cerebrospinal fluid (CSF) biomarkers and progression of non-motor symptoms are associated, but the specifics are not yet clear.ObjectiveThe aim of this study was to investigate the associations of non-motor symptoms with CSF biomarkers in PD.Materials and methodsWe assessed 487 individuals from the Parkinson's Progression Markers Initiative (PPMI), consisting of 155 healthy controls (HCs) and 332 individuals with PD. Patients with PD were grouped according to non-motor symptoms and compared CSF α-synuclein (α-syn), amyloid-beta 1-42 (Aβ1-42), and total tau (t-tau) levels. Multiple linear regressions were used in baseline analysis and linear mixed-effects models in longitudinal analysis. Analyses of mediating effects between cognition and CSF biomarkers were also performed.ResultsAt baseline, PD patients with cognitive impairment (PDCI) exhibited significantly lower CSF α-syn (β = -0.1244; P = 0.0469), Aβ (β = -0.1302; P = 0.0447), and t-tau (β = -0.1260; P = 0.0131) levels than PD patients without cognitive impairment (PDCU). Moreover, a faster decline of α-syn (β = -0.2152; P = 0.0374) and Aβ (β = -0.3114; P = 0.0023) and a faster rise of t-tau (β = -0.1534; P = 0.0274) have been found in longitudinal analysis. The Aβ positive group showed an earlier decline in cognitive performance (β = -0.5341; P = 0.0180) compared with the negative Aβ group in both analyses. In addition, we found that PD patients with probable rapid eye movement sleep behavior disorder (pRBD) showed decreased CSF α-syn (β = -0.1343; P = 0.0033) levels. Finally, mediation analysis demonstrated that olfactory function partially mediated the relationship between cognition and CSF biomarkers levels.ConclusionOur study shows that CSF biomarkers are associated with cognition at baseline and longitudinally. Cognitive impairment is more severe in patients with a heavier Aβ burden. CSF α-syn decreased in PD patients with pRBD. This study suggests that early recognition of the increased risk of non-motor symptoms is important for disease surveillance and may be associated with the pathological progression of CSF markers.

Project description:ObjectiveWe analyzed the longitudinal profile of Alzheimer's disease (AD) cerebrospinal fluid (CSF) biomarkers in early Parkinson's disease (PD) compared with healthy controls (HCs) and tested baseline CSF biomarkers for prediction of clinical decline in PD.MethodsAmyloid-β 1 to 42 (Aβ42 ), total tau (t-tau) and phosphorylated tau (p-tau) at the threonine 181 position were measured using the high-precision Roche Elecsys electrochemiluminescence immunoassay in all available CSF samples from longitudinally studied patients with PD (n = 416) and HCs (n = 192) followed for up to 3 years in the Parkinson's Progression Markers Initiative (PPMI). Longitudinal CSF and clinical data were analyzed with linear-mixed effects models.ResultsWe found patients with PD had lower CSF t-tau (median = 157.7 pg/mL; range = 80.9-467.0); p-tau (median = 13.4 pg/mL; range = 8.0-40.1), and Aβ42 (median = 846.2 pg/mL; range = 238.8-3,707.0) than HCs at baseline (CSF t-tau median = 173.5 pg/mL; range = 82.0-580.8; p-tau median = 15.4 pg/mL; range = 8.1-73.6; and Aβ42 median = 926.5 pg/mL; range = 239.1-3,297.0; p < 0.05-0.001) and a moderate-to-strong correlation among these biomarkers in both patients with PD and HCs (Rho = 0.50-0.97; p < 0.001). Of the patients with PD, 31.5% had pathologically low levels of CSF Aβ42 at baseline and these patients with PD had lower p-tau levels (median = 10.8 pg/mL; range = 8.0-32.8) compared with 27.7% of HCs with pathologically low CSF Aβ42 (CSF p-tau median = 12.8 pg/mL; range 8.2-73.6; p < 0.03). In longitudinal CSF analysis, we found patients with PD had greater decline in CSF Aβ42 (mean difference = -41.83 pg/mL; p = 0.03) and CSF p-tau (mean difference = -0.38 pg/mL; p = 0.03) at year 3 compared with HCs. Baseline CSF Aβ42 values predicted small but measurable decline on cognitive, autonomic, and motor function in early PD.InterpretationOur data suggest baseline CSF AD biomarkers may have prognostic value in early PD and that the dynamic change of these markers, although modest over a 3-year period, suggest biomarker profiles in PD may deviate from healthy aging. ANN NEUROL 2020;88:574-587.

Dataset Information

Development of digital voice biomarkers and associations with cognition, cerebrospinal biomarkers, and neural representation in early Alzheimer's disease.

Introduction

Methods

Results

Discussion

Highlights

Publications

Development of digital voice biomarkers and associations with cognition, cerebrospinal biomarkers, and neural representation in early Alzheimer's disease.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets