No abstract

Introduction Mexican Americans remain severely underrepresented in Alzheimer's disease (AD) research. The Health & Aging Brain among Latino Elders (HABLE) study was created to fill important gaps in the existing literature. Methods Community‐dwelling Mexican Americans and non‐Hispanic White adults and elders (age 50 and above) were recruited. All participants underwent comprehensive assessments including an interview, functional exam, clinical labs, informant interview, neuropsychological testing, and 3T magnetic resonance imaging (MRI) of the brain. Amyloid and tau positron emission tomography (PET) scans were added at visit 2. Blood samples were stored in the Biorepository. Results Data was examined from n = 1705 participants. Significant group differences were found in medical, demographic, and sociocultural factors. Cerebral amyloid and neurodegeneration imaging markers were significantly different between Mexican Americans and non‐Hispanic Whites. Discussion The current data provide strong support for continued investigations that examine the risk factors for and biomarkers of AD among diverse populations.

No abstract

The benefits of locally adaptive statistical methods for fMRI research have been shown in recent years, as these methods are more proficient in detecting brain activations in a noisy environment. One such method is local canonical correlation analysis (CCA), which investigates a group of neighboring voxels instead of looking at the single voxel time course. The value of a suitable test statistic is used as a measure of activation. It is customary to assign the value to the center voxel for convenience. The method without constraints is prone to artifacts, especially in a region of localized strong activation. To compensate for these deficiencies, the impact of different spatial constraints in CCA on sensitivity and specificity are investigated. The ability of constrained CCA (cCCA) to detect activation patterns in an episodic memory task has been studied. This research shows how any arbitrary contrast of interest can be analyzed by cCCA and how accurate P-values optimized for the contrast of interest can be computed using nonparametric methods. Results indicate an increase of up to 20% in detecting activation patterns for some of the advanced cCCA methods, as measured by ROC curves derived from simulated and real fMRI data.

The receiver operating characteristic (ROC) method is a useful and popular tool for testing the efficiency of various diagnostic tests applicable to functional MRI (fMRI) data. Typically, the diagnostic tests are applied on simulated and pseudo-human fMRI data, and the area under the ROC curve is used as a measure of the efficiency of the diagnostic test. The effectiveness of such a method depends on how well the simulated data approximate the real data. For multivariate statistical methods, however, this technique is usually inadequate, as the spatial dependence among voxels is ignored for simulated data. In this work a modified ROC method using real fMRI data with a broader scope is proposed. This method can be applied to most fMRI postprocessing techniques, including multivariate analyses such as canonical correlation analysis (CCA). Also, the relationship of the modified ROC method with the conventional ROC method is discussed in detail. The goal of all postprocessing algorithms in functional MRI (fMRI) is to detect signal from a noisy background. It is important to assess the performance of these different methods in terms of accuracy (few false positives (FPs)) and efficiency (few missed positives). The most popular and useful tools for such assessments are methods based on the receiver operating characteristic (ROC) (1). One of the first studies of ROC methods in fMRI was made by Constable et al. (2). These methods require knowledge about the distributions of true signal and noise. If these are already known, ROC curves can be obtained from purely theoretical considerations, and there is no need for data. However, the actual distributions are unknown in most cases (if not always) and it is necessary to obtain data for an estimate of the ROC curve. In conventional ROC methods, the dataset consists of true signal as well as noise, and it must be determined which one is the true signal and which one is not. This information is not normally revealed by real data; therefore, simulated data are required. For any specific postprocessing method, one typically uses simulated or pseudo-human fMRI datasets and plots the ROC curve after the postprocessing algorithm is implemented (3). The area under the ROC curve serves as a measure of the power of a particular postprocessing algorithm.Despite its advantages, the conventional ROC method in its present form is limited in scope. This is most evident in multivariate statistical analysis. The conventional ROC method looks at the time-course of a single voxel and assumes spatial independence among neighboring voxels. This simple assumption enables us to simulate fMRI data in order to implement the ROC methods. However, in real fMRI data the assumption of spatial independence is not valid, and it is necessary to use multivariate statistical methods to take into account the spatial dependence between neighboring voxels. Since we normally have very little information about the nature of spatial dependence, it is difficult to simulate fMRI data that incorporate the spatial de...