A machine learning approach for potential Super-Agers identification using neuronal functional connectivity networks.

INTRODUCTION: Aging is often associated with cognitive decline. Understanding neural factors that distinguish adults in midlife with superior cognitive abilities (Positive-Agers) may offer insight into how the aging brain achieves resilience. The goals of this study are to (1) introduce an optimal labeling mechanism to distinguish between Positive-Agers and Cognitive Decliners, and (2) identify Positive-Agers using neuronal functional connectivity networks data and demographics. METHODS: In this study, principal component analysis initially created latent cognitive trajectories groups. A hybrid algorithm of machine learning and optimization was then designed to predict latent groups using neuronal functional connectivity networks derived from resting state functional magnetic resonance imaging. Specifically, the Optimal Labeling with Bayesian Optimization (OLBO) algorithm used an unsupervised approach, iterating a logistic regression function with Bayesian posterior updating. This study encompassed 6369 adults from the UK Biobank cohort. RESULTS: OLBO outperformed baseline models, achieving an area under the curve of 88% when distinguishing between Positive-Agers and cognitive decliners. DISCUSSION: OLBO may be a novel algorithm that distinguishes cognitive trajectories with a high degree of accuracy in cognitively unimpaired adults. Highlights: Design an algorithm to distinguish between a Positive-Ager and a Cognitive-Decliner.Introduce a mathematical definition for cognitive classes based on cognitive tests.Accurate Positive-Ager identification using rsfMRI and demographic data (AUC = 0.88).Posterior default mode network has the highest impact on Positive-Aging odds ratio.

Bioenergetic and vascular predictors of potential super-ager and cognitive decline trajectories-a UK Biobank Random Forest classification study.

Aging has often been characterized by progressive cognitive decline in memory and especially executive function. Yet some adults, aged 80 years or older, are "super-agers" that exhibit cognitive performance like younger adults. It is unknown if there are adults in mid-life with similar superior cognitive performance ("positive-aging") versus cognitive decline over time and if there are blood biomarkers that can distinguish between these groups. Among 1303 participants in UK Biobank, latent growth curve models classified participants into different cognitive groups based on longitudinal fluid intelligence (FI) scores over 7-9 years. Random Forest (RF) classification was then used to predict cognitive trajectory types using longitudinal predictors including demographic, vascular, bioenergetic, and immune factors. Feature ranking importance and performance metrics of the model were reported. Despite model complexity, we achieved a precision of 77% when determining who would be in the "positive-aging" group (n = 563) vs. cognitive decline group (n = 380). Among the top fifteen features, an equal number were related to either vascular health or cellular bioenergetics but not demographics like age, sex, or socioeconomic status. Sensitivity analyses showed worse model results when combining a cognitive maintainer group (n = 360) with the positive-aging or cognitive decline group. Our results suggest that optimal cognitive aging may not be related to age per se but biological factors that may be amenable to lifestyle or pharmacological changes.

Using machine learning to predict COVID-19 infection and severity risk among 4510 aged adults: a UK Biobank cohort study.

Many risk factors have emerged for novel 2019 coronavirus disease (COVID-19). It is relatively unknown how these factors collectively predict COVID-19 infection risk, as well as risk for a severe infection (i.e., hospitalization). Among aged adults (69.3 ± 8.6 years) in UK Biobank, COVID-19 data was downloaded for 4510 participants with 7539 test cases. We downloaded baseline data from 10 to 14 years ago, including demographics, biochemistry, body mass, and other factors, as well as antibody titers for 20 common to rare infectious diseases in a subset of 80 participants with 124 test cases. Permutation-based linear discriminant analysis was used to predict COVID-19 risk and hospitalization risk. Probability and threshold metrics included receiver operating characteristic curves to derive area under the curve (AUC), specificity, sensitivity, and quadratic mean. Model predictions using the full cohort were marginal. The "best-fit" model for predicting COVID-19 risk was found in the subset of participants with antibody titers, which achieved excellent discrimination (AUC 0.969, 95% CI 0.934-1.000). Factors included age, immune markers, lipids, and serology titers to common pathogens like human cytomegalovirus. The hospitalization "best-fit" model was more modest (AUC 0.803, 95% CI 0.663-0.943) and included only serology titers, again in the subset group. Accurate risk profiles can be created using standard self-report and biomedical data collected in public health and medical settings. It is also worthwhile to further investigate if prior host immunity predicts current host immunity to COVID-19.

Impact of COVID-19 on the Onset and Progression of Alzheimer's Disease and Related Dementias: A Roadmap for Future Research.

COVID-19 causes lasting neurological symptoms in some survivors. Like other infections, COVID-19 may increase risk of cognitive impairment. This perspective highlights four knowledge gaps about COVID-19 that need to be filled to avoid this possible health issue. The first is the need to identify the COVID-19 symptoms, genetic polymorphisms and treatment decisions associated with risk of cognitive impairment. The second is the absence of model systems in which to test hypotheses relating infection to cognition. The third is the need for consortia for studying both existing and new longitudinal cohorts in which to monitor long term consequences of COVID-19 infection. A final knowledge gap discussed is the impact of the isolation and lack of social services brought about by quarantine/lockdowns on people living with dementia and their caregivers. Research into these areas may lead to interventions that reduce the overall risk of cognitive decline for COVID-19 survivors.

Inflammation, negative affect, and amyloid burden in Alzheimer's disease: Insights from the kynurenine pathway.

BACKGROUND: Depressive symptoms in Alzheimer's disease (AD) predict worse cognitive and functional outcomes. Both AD and major depression inflammatory processes are characterized by shunted tryptophan metabolism away from serotonin (5-HT) and toward the neuroinflammatory kynurenine (Kyn) pathway. The present study assessed associations between Kyn and behavioral, neuroanatomical, neuropathological, and physiological outcomes common to both AD and negative affect across the AD continuum. METHODS: In 58 cognitively normal, 396 mild cognitive impairment, and 112 AD participants from the Alzheimer's Disease Neuroimaging Initiative-1 (ADNI1) cohort, serum markers of 5-HT, tryptophan, and Kyn were measured and their relationships investigated with immunologic markers, affect and functional outcomes, CSF markers of beta-amyloid (Aß) and tau, and regional gray matter. RESULTS: A higher Kyn/Tryptophan ratio was linked to many inflammatory markers, as well as lower functional independence and memory scores. A higher Kyn/5-HT ratio showed similar associations, but also strong relationships with negative affect and neuropsychiatric disturbance, executive dysfunction, and global cognitive decline. Further, gray matter atrophy was seen in hippocampus, anterior cingulate, and prefrontal cortices, as well as greater amyloid and total tau deposition. Finally, using moderated-mediation, several pro-inflammatory factors partially mediated Kyn/5-HT and negative affect scores in participants with subclinical Aß (i.e., Aß-), whereas such associations were fully mediated by Complement 3 in Aß+ participants. CONCLUSION: These findings suggest that inflammatory signaling cascades may occur during AD, which is associated with increased Kyn metabolism that influences the pathogenesis of negative affect. Aß and the complement system may be critical contributing factors in this process.

Using machine learning to predict COVID-19 infection and severity risk among 4,510 aged adults: a UK Biobank cohort study.

BACKGROUND: Many risk factors have emerged for novel 2019 coronavirus disease (COVID-19). It is relatively unknown how these factors collectively predict COVID-19 infection risk, as well as risk for a severe infection (i.e., hospitalization). METHODS: Among aged adults (69.3 ± 8.6 years) in UK Biobank, COVID-19 data was downloaded for 4,510 participants with 7,539 test cases. We downloaded baseline data from 10-14 years ago, including demographics, biochemistry, body mass, and other factors, as well as antibody titers for 20 common to rare infectious diseases. Permutation-based linear discriminant analysis was used to predict COVID-19 risk and hospitalization risk. Probability and threshold metrics included receiver operating characteristic curves to derive area under the curve (AUC), specificity, sensitivity, and quadratic mean. RESULTS: The "best-fit" model for predicting COVID-19 risk achieved excellent discrimination (AUC=0.969, 95% CI=0.934-1.000). Factors included age, immune markers, lipids, and serology titers to common pathogens like human cytomegalovirus. The hospitalization "best-fit" model was more modest (AUC=0.803, 95% CI=0.663-0.943) and included only serology titers. CONCLUSIONS: Accurate risk profiles can be created using standard self-report and biomedical data collected in public health and medical settings. It is also worthwhile to further investigate if prior host immunity predicts current host immunity to COVID-19.