Disease-driven domain generalization for neuroimaging-based assessment of Alzheimer's disease.

Development of deep learning models to evaluate structural brain changes caused by cognitive impairment in MRI scans holds significant translational value. The efficacy of these models often encounters challenges due to variabilities arising from different data generation protocols, imaging equipment, radiological artifacts, and shifts in demographic distributions. Domain generalization (DG) techniques show promise in addressing these challenges by enabling the model to learn from one or more source domains and apply this knowledge to new, unseen target domains. Here we present a framework that utilizes model interpretability to enhance the generalizability of classification models across various cohorts. We used MRI scans and clinical diagnoses from four independent cohorts: Alzheimer's Disease Neuroimaging Initiative (ADNI, n = 1821), the Framingham Heart Study (FHS, n = 304), the Australian Imaging Biomarkers & Lifestyle Study of Ageing (AIBL, n = 661), and the National Alzheimer's Coordinating Center (NACC, n = 4647). With this data, we trained a deep neural network to focus on areas of the brain identified as relevant to the disease for model training. Our approach involved training a classifier to differentiate between structural neurodegeneration in individuals with normal cognition (NC), mild cognitive impairment (MCI), and dementia due to Alzheimer's disease (AD). This was achieved by aligning class-wise attention with a unified visual saliency prior, which was computed offline for each class using all the training data. Our method not only competes with state-of-the-art approaches but also shows improved correlation with postmortem histology. This alignment with the gold standard evidence is a significant step towards validating the effectiveness of DG frameworks, paving the way for their broader application in the field.

DREAMER: a computational framework to evaluate readiness of datasets for machine learning.

BACKGROUND: Machine learning (ML) has emerged as the predominant computational paradigm for analyzing large-scale datasets across diverse domains. The assessment of dataset quality stands as a pivotal precursor to the successful deployment of ML models. In this study, we introduce DREAMER (Data REAdiness for MachinE learning Research), an algorithmic framework leveraging supervised and unsupervised machine learning techniques to autonomously evaluate the suitability of tabular datasets for ML model development. DREAMER is openly accessible as a tool on GitHub and Docker, facilitating its adoption and further refinement within the research community.. RESULTS: The proposed model in this study was applied to three distinct tabular datasets, resulting in notable enhancements in their quality with respect to readiness for ML tasks, as assessed through established data quality metrics. Our findings demonstrate the efficacy of the framework in substantially augmenting the original dataset quality, achieved through the elimination of extraneous features and rows. This refinement yielded improved accuracy across both supervised and unsupervised learning methodologies. CONCLUSION: Our software presents an automated framework for data readiness, aimed at enhancing the integrity of raw datasets to facilitate robust utilization within ML pipelines. Through our proposed framework, we streamline the original dataset, resulting in enhanced accuracy and efficiency within the associated ML algorithms.

Prediction of Alzheimer's disease progression within 6 years using speech: A novel approach leveraging language models.

INTRODUCTION: Identification of individuals with mild cognitive impairment (MCI) who are at risk of developing Alzheimer's disease (AD) is crucial for early intervention and selection of clinical trials. METHODS: We applied natural language processing techniques along with machine learning methods to develop a method for automated prediction of progression to AD within 6 years using speech. The study design was evaluated on the neuropsychological test interviews of n = 166 participants from the Framingham Heart Study, comprising 90 progressive MCI and 76 stable MCI cases. RESULTS: Our best models, which used features generated from speech data, as well as age, sex, and education level, achieved an accuracy of 78.5% and a sensitivity of 81.1% to predict MCI-to-AD progression within 6 years. DISCUSSION: The proposed method offers a fully automated procedure, providing an opportunity to develop an inexpensive, broadly accessible, and easy-to-administer screening tool for MCI-to-AD progression prediction, facilitating development of remote assessment. HIGHLIGHTS: Voice recordings from neuropsychological exams coupled with basic demographics can lead to strong predictive models of progression to dementia from mild cognitive impairment. The study leveraged AI methods for speech recognition and processed the resulting text using language models. The developed AI-powered pipeline can lead to fully automated assessment that could enable remote and cost-effective screening and prognosis for Alzehimer's disease.

Maximizing utility of neuropsychological measures in sex-specific predictive models of incident Alzheimer's disease in the Framingham Heart Study.

INTRODUCTION: Sex differences in neuropsychological (NP) test performance might have important implications for the diagnosis of Alzheimer's disease (AD). This study investigates sex differences in neuropsychological performance among individuals without dementia at baseline. METHODS: Neuropsychological assessment data, both standard test scores and process coded responses, from Framingham Heart Study participants were analyzed for sex differences using regression model and Cox proportional hazards model. Optimal NP profiles were identified by machine learning methods for men and women. RESULTS: Sex differences were observed in both summary scores and composite process scores of NP tests in terms of adjusted means and their associations with AD incidence. The optimal NP profiles for men and women have 10 and 8 measures, respectively, and achieve 0.76 mean area under the curve for AD prediction. DISCUSSION: These results suggest that NP tests can be leveraged for developing more sensitive, sex-specific indices for the diagnosis of AD.

Speech patterns during memory recall relates to early tau burden across adulthood.

INTRODUCTION: Early cognitive decline may manifest in subtle differences in speech. METHODS: We examined 238 cognitively unimpaired adults from the Framingham Heart Study (32-75 years) who completed amyloid and tau PET imaging. Speech patterns during delayed recall of a story memory task were quantified via five speech markers, and their associations with global amyloid status and regional tau signal were examined. RESULTS: Total utterance time, number of between-utterance pauses, speech rate, and percentage of unique words significantly correlated with delayed recall score although the shared variance was low (2%-15%). Delayed recall score was not significantly different between ß-amyoid-positive (Aß+) and -negative (Aß-) groups and was not associated with regional tau signal. However, longer and more between-utterance pauses, and slower speech rate were associated with increased tau signal across medial temporal and early neocortical regions. DISCUSSION: Subtle speech changes during memory recall may reflect cognitive impairment associated with early Alzheimer's disease pathology. HIGHLIGHTS: Speech during delayed memory recall relates to tau PET signal across adulthood. Delayed memory recall score was not associated with tau PET signal. Speech shows greater sensitivity to detecting subtle cognitive changes associated with early tau accumulation. Our cohort spans adulthood, while most PET imaging studies focus on older adults.

Flortaucipir tau PET findings from former professional and college American football players in the DIAGNOSE CTE research project.

INTRODUCTION: Tau is a key pathology in chronic traumatic encephalopathy (CTE). Here, we report our findings in tau positron emission tomography (PET) measurements from the DIAGNOSE CTE Research Project. METHOD: We compare flortaucipir PET measures from 104 former professional players (PRO), 58 former college football players (COL), and 56 same-age men without exposure to repetitive head impacts (RHI) or traumatic brain injury (unexposed [UE]); characterize their associations with RHI exposure; and compare players who did or did not meet diagnostic criteria for traumatic encephalopathy syndrome (TES). RESULTS: Significantly elevated flortaucipir uptake was observed in former football players (PRO+COL) in prespecified regions (p < 0.05). Association between regional flortaucipir uptake and estimated cumulative head impact exposure was only observed in the superior frontal region in former players over 60 years old. Flortaucipir PET was not able to differentiate TES groups. DISCUSSION: Additional studies are needed to further understand tau pathology in CTE and other individuals with a history of RHI.

Optimal blood tau species for the detection of Alzheimer's disease neuropathology: an immunoprecipitation mass spectrometry and autopsy study.

Plasma-to-autopsy studies are essential for validation of blood biomarkers and understanding their relation to Alzheimer's disease (AD) pathology. Few such studies have been done on phosphorylated tau (p-tau) and those that exist have made limited or no comparison of the different p-tau variants. This study is the first to use immunoprecipitation mass spectrometry (IP-MS) to compare the accuracy of eight different plasma tau species in predicting autopsy-confirmed AD. The sample included 123 participants (AD = 69, non-AD = 54) from the Boston University Alzheimer's disease Research Center who had an available ante-mortem plasma sample and donated their brain. Plasma samples proximate to death were analyzed by targeted IP-MS for six different tryptic phosphorylated (p-tau-181, 199, 202, 205, 217, 231), and two non-phosphorylated tau (195-205, 212-221) peptides. NIA-Reagan Institute criteria were used for the neuropathological diagnosis of AD. Binary logistic regressions tested the association between each plasma peptide and autopsy-confirmed AD status. Area under the receiver operating curve (AUC) statistics were generated using predicted probabilities from the logistic regression models. Odds Ratio (OR) was used to study associations between the different plasma tau species and CERAD and Braak classifications. All tau species were increased in AD compared to non-AD, but p-tau217, p-tau205 and p-tau231 showed the highest fold-changes. Plasma p-tau217 (AUC = 89.8), p-tau231 (AUC = 83.4), and p-tau205 (AUC = 81.3) all had excellent accuracy in discriminating AD from non-AD brain donors, even among those with CDR < 1). Furthermore, p-tau217, p-tau205 and p-tau231 showed the highest ORs with both CERAD (ORp-tau217 = 15.29, ORp-tau205 = 5.05 and ORp-tau231 = 3.86) and Braak staging (ORp-tau217 = 14.29, ORp-tau205 = 5.27 and ORp-tau231 = 4.02) but presented increased levels at different amyloid and tau stages determined by neuropathological examination. Our findings support plasma p-tau217 as the most promising p-tau species for detecting AD brain pathology. Plasma p-tau231 and p-tau205 may additionally function as markers for different stages of the disease.

Dissociating Statistically Determined Normal Cognitive Abilities and Mild Cognitive Impairment Subtypes with DCTclock.

OBJECTIVE: To determine whether the DCTclock can detect differences across groups of patients seen in the memory clinic for suspected dementia. METHOD: Patients (n = 123) were classified into the following groups: cognitively normal (CN), subtle cognitive impairment (SbCI), amnestic cognitive impairment (aMCI), and mixed/dysexecutive cognitive impairment (mx/dysMCI). Nine outcome variables included a combined command/copy total score and four command and four copy indices measuring drawing efficiency, simple/complex motor operations, information processing speed, and spatial reasoning. RESULTS: Total combined command/copy score distinguished between groups in all comparisons with medium to large effects. The mx/dysMCI group had the lowest total combined command/copy scores out of all groups. The mx/dysMCI group scored lower than the CN group on all command indices (p < .050, all analyses); and lower than the SbCI group on drawing efficiency (p = .011). The aMCI group scored lower than the CN group on spatial reasoning (p = .019). Smaller effect sizes were obtained for the four copy indices. CONCLUSIONS: These results suggest that DCTclock command/copy parameters can dissociate CN, SbCI, and MCI subtypes. The larger effect sizes for command clock indices suggest these metrics are sensitive in detecting early cognitive decline. Additional research with a larger sample is warranted.

Ante-mortem plasma phosphorylated tau (181) predicts Alzheimer's disease neuropathology and regional tau at autopsy.

Blood-based biomarkers such as tau phosphorylated at threonine 181 (phosphorylated-tau181) represent an accessible, cost-effective and scalable approach for the in vivo detection of Alzheimer's disease pathophysiology. Plasma-pathological correlation studies are needed to validate plasma phosphorylated-tau181 as an accurate and reliable biomarker of Alzheimer's disease neuropathological changes. This plasma-to-autopsy correlation study included participants from the Boston University Alzheimer's Disease Research Center who had a plasma sample analysed for phosphorylated-tau181 between 2008 and 2018 and donated their brain for neuropathological examination. Plasma phosphorelated-tau181 was measured with single molecule array technology. Of 103 participants, 62 (60.2%) had autopsy-confirmed Alzheimer's disease. Average time between blood draw and death was 5.6 years (standard deviation = 3.1 years). Multivariable analyses showed higher plasma phosphorylated-tau181 concentrations were associated with increased odds for having autopsy-confirmed Alzheimer's disease [AUC = 0.82, OR = 1.07, 95% CI = 1.03-1.11, P < 0.01; phosphorylated-tau standardized (z-transformed): OR = 2.98, 95% CI = 1.50-5.93, P < 0.01]. Higher plasma phosphorylated-tau181 levels were associated with increased odds for having a higher Braak stage (OR = 1.06, 95% CI = 1.02-1.09, P < 0.01) and more severe phosphorylated-tau across six cortical and subcortical brain regions (ORs = 1.03-1.06, P < 0.05). The association between plasma phosphorylated-tau181 and Alzheimer's disease was strongest in those who were demented at time of blood draw (OR = 1.25, 95%CI = 1.02-1.53), but an effect existed among the non-demented (OR = 1.05, 95% CI = 1.01-1.10). There was higher discrimination accuracy for Alzheimer's disease when blood draw occurred in years closer to death; however, higher plasma phosphorylated-tau181 levels were associated with Alzheimer's disease even when blood draw occurred >5 years from death. Ante-mortem plasma phosphorylated-tau181 concentrations were associated with Alzheimer's disease neuropathology and accurately differentiated brain donors with and without autopsy-confirmed Alzheimer's disease. These findings support plasma phosphorylated-tau181 as a scalable biomarker for the detection of Alzheimer's disease.

BMI decline patterns and relation to dementia risk across four decades of follow-up in the Framingham Study.

BACKGROUND: Obesity has been associated with increased risk of dementia with several studies reporting a reverse causality, with weight loss preceding the onset of dementia. METHODS: Two thousand forty-five non-demented Framingham Offspring participants, aged 30 to 50 years, were included to determine effect of body mass index (BMI) decline patterns from mid- to late life over a 39-year follow-up. Group-based trajectory models were used to create BMI trajectories. RESULTS: Decreasing BMI trends were associated with higher risk of developing dementia in late life. Decliners with first early mid-life increasing and then later mid-life declining patterns of BMI were at greater increased risk of dementia compared to non-decliners (hazard ratio 3.84, 95% confidence interval 1.39-10.60). CONCLUSION: While patterns of decline in BMI were associated with dementia, a subgroup with a pattern of initial increasing BMI followed by declining BMI, both occurring within mid-life, appeared to be central to declining BMI-dementia association. Further validations are needed to provide robust conclusions.

Long-term blood pressure patterns in midlife and dementia in later life: Findings from the Framingham Heart Study.

INTRODUCTION: Long-term blood pressure (BP) measures, such as visit-to-visit BP variability (BPV) and cumulative BP, are strong indicators of cardiovascular risks. This study modeled up to 20 years of BP patterns representative of midlife by using BPV and cumulative BP, then examined their associations with development of dementia in later life. METHODS: For 3201 individuals from the Framingham Heart Study, multivariate logistic regression analyses were performed to examine the association between long-term BP patterns during midlife and the development of dementia (ages ≥ 65). RESULTS: After adjusting for covariates, every quartile increase in midlife cumulative BP was associated with a sequential increase in the risk of developing dementia (e.g., highest quartile of cumulative systolic blood pressure had approximately 2.5-fold increased risk of all-cause dementia). BPV was not significantly associated with dementia. DISCUSSION: Findings suggest that cumulative BP over the course of midlife predicts risk of dementia in later life. HIGHLIGHTS Long-term blood pressure (BP) patterns are strong indicators of vascular risks. Cumulative BP and BP variability (BPV) were used to reflect BP patterns across midlife. High cumulative BP in midlife is associated with increased dementia risk. Visit-to-visit BPV was not associated with the onset of dementia.

Midlife lipid and glucose levels are associated with Alzheimer's disease.

INTRODUCTION: It is unknown whether vascular and metabolic diseases assessed in early adulthood are associated with Alzheimer's disease (AD) later in life. METHODS: Association of AD with lipid fractions, glucose, blood pressure, body mass index (BMI), and smoking obtained prospectively from 4932 Framingham Heart Study (FHS) participants across nine quadrennial examinations was evaluated using Cox proportional hazard and Kaplan-Meier models. Age-, sex-, and education-adjusted models were tested for each factor measured at each exam and within three adult age groups (early = 35-50, middle = 51-60, and late = 61-70). RESULTS: A 15 mg/dL increase in high density lipoprotein (HDL) cholesterol was associated with decreased AD risk during early (15.4%, P = 0.041) and middle (17.9%, P = 0.014) adulthood. A 15 mg/dL increase in glucose measured during middle adulthood was associated with 14.5% increased AD risk (P = 0.00029). These findings remained significant after adjusting for treatment. DISCUSSION: Our findings suggest that careful management of cholesterol and glucose beginning in early adulthood can lower AD risk.

Blood levels of MCP-1 modulate the genetic risks of Alzheimer's disease mediated by HLA-DRB1 and APOE for Alzheimer's disease.

INTRODUCTION: C-Reactive protein (CRP) and monocyte chemoattractant protein-1 (MCP-1) are both implicated in the peripheral proinflammatory cascade and blood-brain barrier (BBB) disruption. Since the blood CRP level increases Alzheimer's disease (AD) risk depending on the apolipoprotein E (APOE) genotype, we hypothesized that the blood MCP-1 level exerts different effects on the AD risk depending on the genotypes. METHODS: Using multiple regression analyses, data from the Framingham Heart Study (n = 2884) and Alzheimer's Disease Neuroimaging Initiative study (n = 231) were analyzed. RESULTS: An elevated blood MCP-1 level was associated with AD risk in major histocompatibility complex, Class II, DR beta 1 (HLA-DRB1) rs9271192-AC/CC (hazard ratio [HR] = 3.07, 95% confidence interval [CI] = 1.50-6.28, p = 0.002) and in APOE ε4 carriers (HR = 3.22, 95% CI = 1.59-6.53, p = 0.001). In contrast, among HLA-DRB1 rs9271192-AA and APOE ε4 noncarriers, blood MCP-1 levels were not associated with these phenotypes. DISCUSSION: Since HLA-DRB1 and APOE are expressed in the BBB, blood MCP-1 released in the peripheral inflammatory cascade may function as a mediator of the effects of HLA-DRB1 rs9271192-AC/CC and APOE ε4 genotypes on AD pathogenesis in the brain via the BBB pathways.

Examination of potentially modifiable dementia risk factors across the adult life course: The Framingham Heart Study.

INTRODUCTION: We examined for associations between potentially modifiable risk factors across the adult life course and incident dementia. METHODS: Participants from the Framingham Heart Study were included (n = 4015). Potential modifiable risk factors included education, alcohol intake, smoking, body mass index (BMI), physical activity, social network, diabetes, and hypertension. Cox models were used to examine associations between each factor and incident dementia, stratified by early adult life (33-44 years), midlife (45-65 years), and late life (66-80 years). RESULTS: Increased dementia risk was associated with diabetes (hazard ratio [HR] = 1.62; 95% confidence interval [CI] = 1.07-2.46) and physical inactivity (HR = 1.57; 95% CI = 1.12-2.20) in midlife, and with obesity (HR = 1.76; 95% CI = 1.08-2.87) in late life. Having multiple potential modifiable risk factors in midlife and late life was associated with greater risk. DISCUSSION: Potentially modifiable risk factors individually have limited impact on dementia risk in this population across the adult life course, although in combination they may have a synergistic effect. HIGHLIGHTS: Diabetes and physical inactivity in midlife is associated with increased dementia risk. Obesity in late life is associated with increased dementia risk. Having more potentially modifiable risk factors in midlife and late life is associated with greater dementia risk.

Recommendations to address key recruitment challenges of Alzheimer's disease clinical trials.

Clinical trials for Alzheimer's disease (AD) are slower to enroll study participants, take longer to complete, and are more expensive than trials in most other therapeutic areas. The recruitment and retention of a large number of qualified, diverse volunteers to participate in clinical research studies remain among the key barriers to the successful completion of AD clinical trials. An advisory panel of experts from academia, patient-advocacy organizations, philanthropy, non-profit, government, and industry convened in 2020 to assess the critical challenges facing recruitment in Alzheimer's clinical trials and develop a set of recommendations to overcome them. This paper briefly reviews existing challenges in AD clinical research and discusses the feasibility and implications of the panel's recommendations for actionable and inclusive solutions to accelerate the development of novel therapies for AD.

White matter hyperintensities in former American football players.

INTRODUCTION: The presentation, risk factors, and etiologies of white matter hyperintensities (WMH) in people exposed to repetitive head impacts are unknown. We examined the burden and distribution of WMH, and their association with years of play, age of first exposure, and clinical function in former American football players. METHODS: A total of 149 former football players and 53 asymptomatic unexposed participants (all men, 45-74 years) completed fluid-attenuated inversion recovery magnetic resonance imaging, neuropsychological testing, and self-report neuropsychiatric measures. Lesion Segmentation Toolbox estimated WMH. Analyses were performed in the total sample and stratified by age 60. RESULTS: In older but not younger participants, former football players had greater total, frontal, temporal, and parietal log-WMH compared to asymptomatic unexposed men. In older but not younger former football players, greater log-WMH was associated with younger age of first exposure to football and worse executive function. DISCUSSION: In older former football players, WMH may have unique presentations, risk factors, and etiologies. HIGHLIGHTS: Older but not younger former football players had greater total, frontal, temporal, and parietal lobe white matter hyperintensities (WMH) compared to same-age asymptomatic unexposed men. Younger age of first exposure to football was associated with greater WMH in older but not younger former American football players. In former football players, greater WMH was associated with worse executive function and verbal memory.

Integrative brain transcriptome analysis links complement component 4 and HSPA2 to the APOE ε2 protective effect in Alzheimer disease.

Mechanisms underlying the protective effect of apolipoprotein E (APOE) ε2 against Alzheimer disease (AD) are not well understood. We analyzed gene expression data derived from autopsied brains donated by 982 individuals including 135 APOE ɛ2/ɛ3 carriers. Complement pathway genes C4A and C4B were among the most significantly differentially expressed genes between ɛ2/ɛ3 AD cases and controls. We also identified an APOE ε2/ε3 AD-specific co-expression network enriched for astrocytes, oligodendrocytes and oligodendrocyte progenitor cells containing the genes C4A, C4B, and HSPA2. These genes were significantly associated with the ratio of phosphorylated tau at position 231 to total Tau but not with amyloid-ß 42 level, suggesting this APOE ɛ2 related co-expression network may primarily be involved with tau pathology. HSPA2 expression was oligodendrocyte-specific and significantly associated with C4B protein. Our findings provide the first evidence of a crucial role of the complement pathway in the protective effect of APOE ε2 for AD.

Associations Between the Digital Clock Drawing Test and Brain Volume: Large Community-Based Prospective Cohort (Framingham Heart Study).

BACKGROUND: The digital Clock Drawing Test (dCDT) has been recently used as a more objective tool to assess cognition. However, the association between digitally obtained clock drawing features and structural neuroimaging measures has not been assessed in large population-based studies. OBJECTIVE: We aimed to investigate the association between dCDT features and brain volume. METHODS: This study included participants from the Framingham Heart Study who had both a dCDT and magnetic resonance imaging (MRI) scan, and were free of dementia or stroke. Linear regression models were used to assess the association between 18 dCDT composite scores (derived from 105 dCDT raw features) and brain MRI measures, including total cerebral brain volume (TCBV), cerebral white matter volume, cerebral gray matter volume, hippocampal volume, and white matter hyperintensity (WMH) volume. Classification models were also built from clinical risk factors, dCDT composite scores, and MRI measures to distinguish people with mild cognitive impairment (MCI) from those whose cognition was intact. RESULTS: A total of 1656 participants were included in this study (mean age 61 years, SD 13 years; 50.9% women), with 23 participants diagnosed with MCI. All dCDT composite scores were associated with TCBV after adjusting for multiple testing (P value <.05/18). Eleven dCDT composite scores were associated with cerebral white matter volume, but only 1 dCDT composite score was associated with cerebral gray matter volume. None of the dCDT composite scores was associated with hippocampal volume or WMH volume. The classification model for differentiating MCI and normal cognition participants, which incorporated age, sex, education, MRI measures, and dCDT composite scores, showed an area under the curve of 0.897. CONCLUSIONS: dCDT composite scores were significantly associated with multiple brain MRI measures in a large community-based cohort. The dCDT has the potential to be used as a cognitive assessment tool in the clinical diagnosis of MCI.

Association Between Acoustic Features and Neuropsychological Test Performance in the Framingham Heart Study: Observational Study.

BACKGROUND: Human voice has increasingly been recognized as an effective indicator for the detection of cognitive disorders. However, the association of acoustic features with specific cognitive functions and mild cognitive impairment (MCI) has yet to be evaluated in a large community-based population. OBJECTIVE: This study aimed to investigate the association between acoustic features and neuropsychological (NP) tests across multiple cognitive domains and evaluate the added predictive power of acoustic composite scores for the classification of MCI. METHODS: This study included participants without dementia from the Framingham Heart Study, a large community-based cohort with longitudinal surveillance for incident dementia. For each participant, 65 low-level acoustic descriptors were derived from voice recordings of NP test administration. The associations between individual acoustic descriptors and 18 NP tests were assessed with linear mixed-effect models adjusted for age, sex, and education. Acoustic composite scores were then built by combining acoustic features significantly associated with NP tests. The added prediction power of acoustic composite scores for prevalent and incident MCI was also evaluated. RESULTS: The study included 7874 voice recordings from 4950 participants (age: mean 62, SD 14 years; 4336/7874, 55.07% women), of whom 453 were diagnosed with MCI. In all, 8 NP tests were associated with more than 15 acoustic features after adjusting for multiple testing. Additionally, 4 of the acoustic composite scores were significantly associated with prevalent MCI and 7 were associated with incident MCI. The acoustic composite scores can increase the area under the curve of the baseline model for MCI prediction from 0.712 to 0.755. CONCLUSIONS: Multiple acoustic features are significantly associated with NP test performance and MCI, which can potentially be used as digital biomarkers for early cognitive impairment monitoring.

Automated detection of mild cognitive impairment and dementia from voice recordings: A natural language processing approach.

INTRODUCTION: Automated computational assessment of neuropsychological tests would enable widespread, cost-effective screening for dementia. METHODS: A novel natural language processing approach is developed and validated to identify different stages of dementia based on automated transcription of digital voice recordings of subjects' neuropsychological tests conducted by the Framingham Heart Study (n = 1084). Transcribed sentences from the test were encoded into quantitative data and several models were trained and tested using these data and the participants' demographic characteristics. RESULTS: Average area under the curve (AUC) on the held-out test data reached 92.6%, 88.0%, and 74.4% for differentiating Normal cognition from Dementia, Normal or Mild Cognitive Impairment (MCI) from Dementia, and Normal from MCI, respectively. DISCUSSION: The proposed approach offers a fully automated identification of MCI and dementia based on a recorded neuropsychological test, providing an opportunity to develop a remote screening tool that could be adapted easily to any language.