Search

Background: Circadian function is altered with aging, leading to changes in the ∼24‐h rhythms of behavior and physiology. Based on circadian rest‐activity rhythms (CRAR), we previously constructed a biological age biomarker to characterize the aging profile of circadian function—circadian age (CircAge)—in older adults (>65 years) and showed its association with risk of Alzheimer's dementia. Here, we constructed CircAge in middle‐to‐older‐aged adults and tested whether it predicts all‐cause dementia. Method: Participants from the UK Biobank with valid actigraphy data were studied (n = ∼97,000; 42‐78 years old). We developed a machine‐learning‐based regression model with 23 CRAR features derived from actigraphy (using cosinor, nonparametric, uniform‐phase‐empirical‐mode‐decomposition, and detrended‐fluctuation analyses). The fitting target was chronological age (ChrAge), and the prediction output was defined as CircAge. Model parameters were optimized using cross validation on 47,234 participants not performing shift work and without mental/behavioral or nervous system diseases. The optimized model was applied to all dementia‐free participants with valid CRAR (n = 80,805) to obtain their CircAge. Result: CircAge was 1.03±0.01 years older per one‐year older in ChrAge that explained ∼25% of the variance of CircAge (p<0.0001). Male sex, European ethnicity, or having mental/behavioral/nervous system diseases were associated with a greater CircAge (all p<0.0001). We compared CircAge of current shift workers (n = 6,389) with that of matched non‐shift workers to examine whether CircAge captures CRAR perturbations and found that CircAge of current shift workers was 1.1±0.2 years greater (p<0.0001) than non‐shift workers. We then associated the difference between CircAge and ChrAge (DCC, = CircAge‐ChrAge) to incident all‐cause dementia using a Cox proportional hazards model adjusting for ChrAge, sex, ethnic groups, mental/behavioral diseases, and nervous system diseases. Results showed that greater DCC predicted higher adjusted risk of all‐cause dementia (n = 451): each decade higher DCC corresponding to a 1.10‐fold higher risk (95% CI: 1.02‐1.18; p = 0.009). Conclusion: CRAR patterns can be translated into a biological aging biomarker for circadian function (CircAge), and greater CircAge than ChrAge is independently associated with risk of dementia in middle‐to‐older‐aged adults. Future studies are warranted to examine the use of CircAge (and/or in combination with other risk markers) for identifying at‐risk individuals in a clinical setting. [ABSTRACT FROM AUTHOR]

Background: Actigraphy derived measures including amplitude, regularity, and variability of daily rhythm, have been shown to predict incident Alzheimer's dementia. Here we developed an integrated actigraphy biomarker (IAB) for Alzheimer' risk and investigated whether the IAB predicted the conversion from mild cognitive impairment (MCI) to Alzheimer' dementia. Method: We studied 1195 participants (age 80.8±7.2yrs [SD]) from the Rush Memory and Aging Project who had finished baseline actigraphy assessment (∼10 days), were free from Alzheimer's dementia at actigraphy baseline and had been followed for up to 15 years with annual cognitive assessment and clinical diagnoses. Ten sleep/circadian related features were derived from baseline actigraphy recordings and were fed into a random forest survival model for prediction of time and incident Alzheimer's dementia. IAB was derived from the model as the relative risk. Cox proportional hazards and logistic regression models were used to evaluate the performance of IAB in predicting incident Alzheimer's dementia (all 1195 participants) or MCI (858 without MCI at baseline) and predicting the conversion from MCI to Alzheimer's dementia. Demographic variables including age, sex, and education were controlled in all Cox and logistic regression models. Result: Total 287 participants developed Alzheimer's dementia during the follow‐up. The derived IAB was 0.6 SD larger in the participants developed Alzheimer's dementia as compared with the controls. Larger IAB was associated with increased risk of AD with a hazard ratio (HR) = 1.63 (95% CI = 1.46‐1.81, P<0.0001) for 1‐SD increase in IAB. IAB did not predict incident MCI (P = 0.8). Within participants with MCI (337 at baseline and 308 developed), larger IAB was associated with a higher risk for AD, i.e., HR = 1.34 for 1 SD increase (95% CI = 1.18‐1.51, P<0.0001). The logistic model using the cutoff of 3 years for the MCI‐Alzheimer's dementia conversion gives the odd ratio = 1.56 for 1 SD increase of IAB, and results in AUC = 0.68, with a sensitivity = 0.68 and specificity = 0.61. Conclusion: Derived actigraphy biomarker was predictive of Alzheimer's risk at preclinical stages, and the conversion from MCI. Actigraphy provides useful information for early prediction and detection of AD thought its performance needs to be improved. [ABSTRACT FROM AUTHOR]

Background: Dementia is a growing cause of disability and loss of independence in the elderly, yet remains largely under‐diagnosed. A biomarker for dementia that can identify individuals with or at risk for developing dementia may help close this diagnostic gap. Deviations from the normal aging trajectory, known as Brain Age Index (BAI), have shown potential to serve as biomarkers for cognitive impairment. We aimed to investigate the association between BAI and dementia and determine whether an overnight sleep EEG‐based BAI could serve as a useful biomarker for dementia. We hypothesized that patients with dementia‐related diseases have significantly higher brain age indices than patients without dementia. Method: Using a dataset of polysomnograms from 11,039 patients, we computed BAI using an EEG‐based brain age algorithm. Patients were categorized into four groups, including dementia, mild cognitive impairment (MCI), symptomatic, and non‐dementia, based on clinical diagnoses, Montreal Cognitive Assessment (MoCA), and/or Mini‐Mental State Exam (MMSE) scores. Result: We found an overall significant trend across dementia groups using Cuzick's test (p=0.0007). The BAI of patients with dementia (4.11 ± 10.02 yrs) were significantly higher than those of patients with no dementia (0.516 ±10.40 yrs), with p‐value of 0.002, and those of healthy subjects (‐0.67 ± 9.52 yrs), with p‐value of 0.0002. BAI was negatively correlated with MoCA (R = ‐0.1359, p = 0.006) and MMSE (R = ‐0.1201, p = 0.005). Features related to delta activity in non‐REM sleep stages 2 (N2) and 3 (N3) tend to correlate with non‐dementia, and features related to theta and delta waves in the Wake and N1 stages tend to correlate with dementia. Conclusion: Our results suggests a potential for an EEG‐based Brain Age Index to serve as a biomarker of dementia. This opens new possibilities for using BAI as an assessment tool for the presence of underlying neurodegenerative disease and monitoring of disease progression. [ABSTRACT FROM AUTHOR]

Background: In older adults, circadian rhythms show alterations in multiple dimensions with aging and during Alzheimer's dementia progression. We propose that circadian function may serve as a unique gauge of biological age (i.e., circadian age), and that individuals with greater circadian age than chronological age have increased risk of Alzheimer's dementia. Method: We constructed a regression model with chronological age as the predicting target. The model prediction was termed circadian age. A softplus activation function was used to ensure positivity. The model was optimized to minimize the mean‐squared prediction error while penalizing the covariance between chronological age and prediction error. Cosinor analysis, nonparametric analysis, uniform phase empirical mode decomposition, and detrended fluctuation analysis were performed on baseline actigraphy data of 882 participants (age: 80.1±7.3; female/male: 692/190) from the Rush Memory and Aging Project who were cognitively intact and were not taking sleep medications. These analyses resulted in 23 features for circadian daily activity patterns as model inputs. The optimal model was trained through cross‐validation for the penalization parameter and applied to follow‐up actigraphy data. We compared circadian age among participants at different cognitive stages, evaluated the longitudinal change of circadian age, and examined the association of circadian age with incident Alzheimer's dementia. Result: Chronological age explained 24.8% of the variance in circadian age. At analytical baseline, participants with Alzheimer's dementia had significantly greater circadian–chronological age deviation (CCD) than cognitively normal participants (2.98±1.43 vs. ‐0.13±0.42 years, p=0.04). For each 1‐year increase in chronological age, circadian age increased by 1.05±0.06 years (p<0.001), suggesting accelerated circadian aging. Among baseline cognitively normal participants, greater CCD was associated with an increased risk of Alzheimer's dementia independent from chronological age, sex, total daily activity, sleep fragmentation, vascular disease and risk factors, and depressive symptoms. For a 10‐year increase in CCD, the hazard ratio was 1.23 (95% CI: 1.06‐1.42). Conclusion: Circadian daily patterns can be integrated to predict chronological age. Older circadian age than chronological age is an independent risk factor for Alzheimer's dementia. Future work should refine the determination of health status for training dataset and examine whether circadian age links with endogenous circadian function. [ABSTRACT FROM AUTHOR]

Background: Timely diagnosis of dementia is important to patients and their caregivers for advanced planning, yet dementia is under‐diagnosed by healthcare professionals and under‐coded in claims data. Sensitive and specific tools to detect cognitive concerns in diverse clinical settings could prompt referral for cognitive evaluation and specialist care. Method: We developed a deep learning natural language processing (NLP) method to detect cognitive concerns in unstructured clinician notes from electronic health records (EHR). We leveraged a gold‐standard set of ∼1000 patients sampled randomly from three strata: patients with diagnosis codes, patients with specialist visits but no code, or patients with neither. The physician performed a detailed chart review and adjudication of cognitive status, noting "cognitive concern" (i.e., any evidence of cognitive difficulties) and rating patients on a 5‐point scale: normal, normal vs. MCI, MCI, MCI vs. dementia, and dementia. We used 10% of the labeled data as a test set and the remaining 90% for training and validation of the model to classify patients with any cognitive concerns (normal vs. other). We also built a web‐based chart review annotation tool that facilitates labeling and enables an active learning loop to scale up labeling to thousands of charts. Result: In a random sample from the gold‐standard dataset, 30 out of 80 patients with cognitive concerns had no diagnosis code or medication related to dementia; we hypothesized that our deep learning tool could leverage clinical text to improve detection of cognitive concerns. Indeed, a model with codes and medications had an area under the receiver operating characteristic (AUROC) curve of 0.79, sensitivity of 0.59, and specificity of 1.00 for the binary classification task. The deep learning model improved the AUROC to 0.90, increased sensitivity to 0.79, and maintained specificity of 0.98. Notes from primary care, specialties such as neurology and psychiatry, and social workers had the highest likelihood of containing information. Conclusion: The deep learning model was successful in detecting cases without a dementia‐related diagnosis code or medication. Automatic processing of electronic medical records with a deep learning tool can be used for early detection of cognitive concern to optimize patient care and predict hospital readmission. [ABSTRACT FROM AUTHOR]

Background: Dementia is a growing cause of disability and loss of independence in the elderly, yet remains largely under‐diagnosed. Early detection and classification of dementia may help close this diagnostic gap and improve management of disease progression. EEG sleep patterns have been identified as a potential biomarker to detect Alzheimer's disease and other neurodegenerative diseases. Method: From a dataset of 9834 polysomnograms, sleep architecture and microstructure features such as frequency band powers, EEG coherence, and spindle density were extracted. Patients were labeled as belonging to dementia, mild cognitive impairment (MCI), or cognitively normal (CN) groups based on clinical diagnosis, Montreal Cognitive Assessment (MoCA), Mini‐Mental State Exam (MMSE) scores, Clinical Dementia Rating (CDR) and medications. We trained logistic regression, random forest, and XGBoost models to classify patients into Dementia, MCI, and CN groups. Result: Nested cross validation results show an AUC of 0.81 (F1 = 0.76) for binary classification of dementia vs CN groups and a mean AUC of 0.75 (F1 =0.57) for multiclass classification of dementia vs. MCI vs CN groups. REM latency, spindle activity, duration, frequency, slow wave oscillation, delta/alpha band power in wake, N1 theta/alpha band power in N1 were among the top weighted features. Conclusion: Our dementia classification algorithms show promise for incorporating dementia screening techniques into routine sleep EEG and providing diagnostic, monitoring, and prognostication capabilities. [ABSTRACT FROM AUTHOR]