While aging is typically associated with cognitive decline, some individuals are able to diverge from the characteristic downward slope and maintain very high levels of cognitive performance. Prior studies have found that cortical thickness in the cingulate cortex, a region involved in information processing, memory, and attention, distinguish those with exceptional cognitive abilities when compared to their cognitively more typical elderly peers. Others major areas outside of the cingulate, such as the prefrontal cortex and insula, are also key in successful aging well into late age, suggesting that structural properties across a wide range of areas may better explain differences in cognitive abilities. Here, we aim to assess the role of regional cortical thickness, both in the cingulate and the whole brain, in modeling Top Cognitive Performance (TCP), measured by performance in the top 50th percentile of memory and executive function. Using data from National Alzheimer’s Coordinating Center and The 90 + Study, we examined healthy subjects aged 70–100 years old. We found that, while thickness in cingulate regions can model TCP status with some degree of accuracy, a whole-brain, network-level approach out-performed the localist, cingulate models. These findings suggests a need for more network-style approaches and furthers our understanding of neurobiological factors contributing to preserved cognition.

In this study, the characteristics of the air quality in current years were comprehensively analyzed according to the data between 2016 and 2020. The difference of six pollutants concentrations within 5 years in Beijing, Tianjin, and Hebei was compared at the same time. The control effectiveness in three areas was evaluated. Control measures would benefit the air quality improvement, which had the potential to give advice for the area that are less effective. Furthermore, comparing the change of concentration of six major pollutants within 5 years, the effectiveness of control measures on each concentration is evaluated. Under this circumstance, we considered different factors that contribute to the result, and found their implication on future control. From this study, it was concluded that (1) The control measure between 2016 and 2020 in Beijing-Tianjin-Hebei was effective for CO2, CO, SO2, PM2.5, and PM10. (2) The control of ozone and nitrogen dioxide is less effective, which requires efforts to reduce vehicle emissions and VOC emissions. (3) The effectiveness among the three places is equally effective, which indicated the positive tendency should be sustained.

BackgroundEarly detection and diagnosis are critical for effective treatment of Alzheimer’s disease (AD). Statistical machine learning holds promise for disease prediction using large AD repositories, such as the National Alzheimer’s Coordinating Center (NACC) dataset. Past work has extensively explored binary classifications of AD diagnosis. Here, we extended this to address two major, remaining challenges: data imputation and multiclass classification (e.g., AD vs Mild Cognitive Impairment, MCI vs. Healthy Cognition, HC).MethodFor imputation, we tested single‐value (baseline), k‐NN, and Bayesian imputation on a subset of complete training data with artificially removed values (completely at random). Per modality of interest, a systematic search was conducted for the best input feature set to impute the missing values. The mean absolute deviation and mismatch rate between the imputed and true values were used to evaluate the methods on a held‐out test set. For classification, we then implemented both “flat” multiclass and hierarchical classification, an extension of multiclass classification that organizes classes into a tree, to predict clinical diagnosis using multimodal inputs. Several pipelines with combinations of imputation methods, classifiers, and input modalities were tested on different hierarchical classification strategies, e.g., HC vs [AD vs MCI], compared with the flat, multiclass baseline (AD vs MCI vs HC).ResultCompared to the baseline methods for both imputation and classification, the proposed alternatives performed better, as measured by the predetermined imputation and classification metrics evaluated on the test set. For imputation, behavioral and neuropsychiatric features were more easily and accurately imputed while ApoE genotype was more difficult to impute. For classification, using a combination of classifiers and input features in the hierarchy outperformed the baseline method and support the notion that select biomarkers are better able to characterize AD subpopulations within the hierarchy.ConclusionEstablishing a systematic imputation method tailored to AD biomarker modalities fills a critical gap in achieving more robust disease prediction. Building a hierarchical classifier parallels working through a differential diagnosis for AD, which offers greater translational value than binary classification of AD. In combination, these results form an important step toward developing reliable, holistic classifiers of AD disease status.