This project undertakes the development and application of statistical methodology to neuroimaging. In particular, while brain imaging is a fundamental tool in neuroscience, the statistical treatment of the quantification of such images has lagged behind imaging technology. Numerous statistical problems are just beginning to be addressed in the analysis of neuroimages. These include: design of experiments to limit the search volume in image analysis by either a priori knowledge or a previous scan; the analysis of voxel (volume element) subtraction of images to investigate brain volumes of change in positron emission tomography (PET) or magnetic resonance imaging (MRI) scans of the same individuals under different tasks or drugs; multiple comparison issues to exploit the spatial correlation and to control the experiment-wise Type 1 error of any inference concerning a brain volume of apparent activity (where there may be 16,000 voxels per slice); techniques to analyze time-course data that is fundamental in functional MRI experiments; and the planning of experiments to ensure adequate power. Further, the resolution of these problems is all the more crucial as the imaging technology continues to improve dramatically. Research has been conducted concerning receiver operating characteristic (ROC) methodology that has direct application to the evaluation of different imaging modalities. Papers have been submitted, are in review or were published in FY94 on the following topics: computational and statistical tools for paired digital analysis, variability and covariability in magnetic resonance functional neuro-imaging, and induced ischemia in the motor areas of the brains of normal volunteers as assessed by PET scans (MNB); new computationally-intensive methodologies to compare ROC plots that are useful in evaluation of imaging modalities, an investigation of quantitative MRI and clinical staging parameters for autosomal dominant cerebral ataxias (CNB); a functional MRI study on cortical activation during mental calculation (MNB); statistical methodology for analysis of functional MRI data; a statistical analysis based on estimation of the intrinsic temporal and spatial autocorrelations in the spectral domain for functional MR images in a study to assess the changes in the motor areas of the brains of normal volunteers that are activated by a finger movement task versus by its ideation only (MNB); and the variability of metabolites among normal volunteers using MR spectroscopy (NB).
