Search

Repeated measures designs are widely used in practice to increase power, reduce sample size, and increase efficiency in data collection. Correlation between repeated measurements is one of the first research questions that needs to be addressed in a repeated-measure study. In addition to an estimate for correlation, confidence interval should be computed and reported for statistical inference. The asymptotic interval based on the delta method is traditionally calculated due to its simplicity. However, this interval is often criticized for its unsatisfactory performance with regards to coverage and interval width. Bootstrap could be utilized to reduce the interval width, and the widely used bootstrap intervals include the percentile interval, the bias-corrected interval, and the bias-corrected with acceleration interval. Wilcox (Comput Stat Data Anal 22:89–98,1996) suggested a modified percentile interval with the interval levels adjusted by sample size to have the coverage probability close to the nominal level. For a study with repeated measures, more parameters in addition to sample size would affect the coverage probability. For these reasons, we propose modifying the percentiles in the percentile interval to guarantee the coverage probability based on simulation studies. We analyze the correlation between imaging volumes and memory scores from the Alzheimer's Disease Neuroimaging Initiative (ADNI) study to illustrate the application of the considered intervals. The proposed interval is exact with the coverage probability guaranteed, and is recommended for use in practice. [ABSTRACT FROM AUTHOR]

Amyloid- β (A β ) is the target in many clinical trials for Alzheimer's disease (AD). Preclinical AD patients are heterogeneous with regards to different backgrounds and diagnosis. Accurately predicting A β status of participants by using machine learning (ML) models based on easily accessible data, could improve the effectiveness of AD clinical trials. We will develop optimal ML models for each subpopulation stratified by sex and disease stages using sub scores from screening neurological tests. Data from the AD Neuroimaging Initiative (ADNI) were used to build the ML models, for three groups: individuals with significant memory concern, early mild cognitive impairment (MCI), and late MCI. Data were further separated into 6 groups by disease stage (3 levels) and sex (2 categories). The outcome was defined as the A β status confirmed by the PET imaging, and the features include demographic data, newly identified risk factors, screening tests, and the domain scores from screening tests. Monte Carlo simulation studies were used together with k-fold cross-validation technique to compute model performance metric. We also develop a new feature selection method based on the stochastic ordering to avoiding searching all possible combinations of features. Accuracy of the identified optimal model for SMC male was over 90% by using domain scores, and accuracy for LMCI female was above 86%. Domain scores can improve the ML model prediction as compared to the total scores. Accurate ML prediction models can identify the proper population for AD clinical trials. [ABSTRACT FROM AUTHOR]