Search for: All records

Prodromal detection of Alzheimer’s Disease(AD) is a substantial challenge in the research community. Among the tools used in AD diagnosis, cognitive exams are standard in most procedures. However, the barrage of cognitive examinations is both time and resource consuming. With the use of Machine Learning, Feature Elimination (FE) can be combined with classification algorithms to determine which cognitive exams are best suited for diagnosis. Using the results of FE, it can be determined if subsections of different composite scores can be combined to create a new enhanced and exhaustive exam. This paper implements a Recursive Feature Elimination with Cross Validation (RFECV) machine learning algorithm to determine which cognitive exams perform best for AD classification tasks. Out of 119 features, an average of 16 features were selected as optimal. These optimal features average 75% Accuracy, 70% Precision, and 75% Recall and an F1 Weighted score of 71% in classification. 

Abstract Depression has been associated with greater risk of Alzheimer’s disease (AD), and existing research has identified structural differences in brain regions in depressed subjects compared to healthy samples, but results have been heterogeneous. We sought to determine the effect of depression on regional brain volumes by cognitive and APOE e4 status. Secondary analysis of the National Alzheimer’s Coordinating Center (NACC) Uniform Data Set was conducted using complete MRI data from 1,371 participants (mean age: 70.5; SD: 11.7). Multiple linear regression was used to estimate the adjusted effect of depression (via the Neuropsychiatric Inventory Questionnaire) on regional brain volumes through measurement of 30 structural MRIs. Depression in the prior two years was associated with lower total brain, cerebrum,, and gray matter volumes and greater total brain white matter hyperintensities (p<.05). Greater volumes were also observed in all ventricular volume measures. Lower mean volumes were observed in six additional frontal lobe and parietal lobe cortical regions. Alternately, depression antecedent to the past 2 years correlated only with occipital lobe gray matter volumes (right, left, total). Our findings suggest that depression in the prior two years is associated with atrophy across multiple brain regions and related ventricular enlargement, even after controlling for intracranial volume and demographic covariates. The duration of depression influences results, however, as depression prior to 2 years before assessment was correlated with significantly fewer and different regional brain volume changes. 

Abstract Sleep disruption has been associated with increased beta-amyloid deposition and greater risk for later development of Alzheimer’s disease. Studies indicate that sleep disturbance correlates with regional brain volumes, but data are limited. We sought to determine the effect of sleep disturbance on regional brain volumes by cognitive and apolipoprotein e (APOE) e4 status. We conducted a secondary analysis of the National Alzheimer’s Coordinating Center (NACC) Uniform Data Set using complete structural imaging data from 1,371 participants (mean age: 70.5; SD: 11.7). Multiple linear regression was used to estimate the adjusted effect of sleep disturbance (via Neuropsychiatric Inventory Questionnaire) on regional brain volumes through measurement of 30 structural MRI biomarkers. Sleep disruption was associated with greater volumes in the right and left lateral ventricles and greater volume of total white matter hyperintensities (p<.05). Lower mean volumes in total brain, total gray matter, and total cerebrum grey matter volumes, and in 12 hippocampal, frontal, parietal, and temporal lobe volumes were observed among participants who reported sleep disturbance. Males, Hispanic participants, and those with less education were more likely to report sleep disruption. Cognitive status moderated the relationship between sleep disturbance and lateral ventricular volumes, while APOE e4 moderated the effect between sleep disturbance and parietal lobe volumes. These findings suggest that disrupted sleep is associated with atrophy across multiple brain regions and ventricular hydrocephalus ex vacuo, after controlling for intracranial volume and demographic covariates. The influence of cognition and APOE e4 status indicates that this relationship is affected by co-occurring physiological processes. 

Abstract Anxiety has been associated with greater risk of Alzheimer’s disease (AD) and existing research has identified structural differences in regional brain tissue in anxious compared to healthy samples, but results have been variable and somewhat inconsistent. We sought to determine the effect of anxiety on regional brain volumes by cognitive and apolipoprotein e (APOE) e4 status using data from a large, national dataset. A secondary analysis of the National Alzheimer’s Coordinating Center Uniform (NACC) Data Set was conducted using complete MRI data from 1,371 participants (mean age: 70.5; SD: 11.7). Multiple linear regression was used to estimate the adjusted effect of anxiety (via the Neuropsychiatric Inventory Questionnaire) on regional brain volumes through measurement of 30 structural MRI biomarkers. Anxiety was associated with lower total brain and total cortical gray matter volumes and increased lateral ventricular volume (p<.05). Lower mean volumes were also observed in all hippocampal, frontal lobe, parietal lobe, temporal lobe, and right occipital lobe volumes among participants who reported anxiety. Conversely, greater ventricular volumes were also correlated with anxiety. Findings suggest that anxiety is associated with significant atrophy in multiple brain regions and ventricular enlargement, even after controlling for intracranial volume and demographic covariates. Anxiety-related changes to brain morphology may contribute to greater AD risk.