Challenges and Advances on Big Data in Neuroimaging

An explosion of image acquisition tools, analytical software capabilities, postprocessing algorithms, and data-sharing infrastructures are all components of the current era of “Big Data” in neuroimaging. This flood of neuroimaging information coincides with an exponential parallel growth in data derived from the increasing utilization of electronic health record systems, and the mounting efforts at the genetic profiling of central nervous system diseases. One would expect revolutionary solutions for a multitude of health problems considering this burst of technical progress. Yet, transforming this flood of new information into clinically meaningful knowledge remains challenging. Our high-speed academic chase to develop new neuroimaging tools and add to the already “Big” pool of data needs to be matched by a clear vision for how to apply this information to better patient care, both at the individual and public health levels. In this talk, we will highlight “Big” opportunities in need of fruitful and innovative “Big data” applications. Population Level Analyses of Clinical Brain Imaging Ciprian Crainiceanu, PhD Johns Hopkins University Abstract: The talk will focus on structural brain imaging and will introduce two distinct scientific areas. The first part of the talk will be dedicated to the association of stroke location and stroke severity using computed tomography (CT) brain imaging in a large clinical trial. The second part will focus on structural MRI (sMRI) and its application to longitudinal lesion segmentation, tracking, and quantification in Multiple Sclerosis patients. Emphasis will be on the scientific problems and datasets. I will discuss the complexity associated with population analyses of clinical brain imaging and differences from analyses of populations of healthy brains. General information about Dr. Crainiceanu's work can be found at www.smart-stats.org. The talk will focus on structural brain imaging and will introduce two distinct scientific areas. The first part of the talk will be dedicated to the association of stroke location and stroke severity using computed tomography (CT) brain imaging in a large clinical trial. The second part will focus on structural MRI (sMRI) and its application to longitudinal lesion segmentation, tracking, and quantification in Multiple Sclerosis patients. Emphasis will be on the scientific problems and datasets. I will discuss the complexity associated with population analyses of clinical brain imaging and differences from analyses of populations of healthy brains. General information about Dr. Crainiceanu's work can be found at www.smart-stats.org. Exploratory Analysis of High Dimensional Brain Signals Hernando Ombao, Ph.D. Department of Statistics UC Irvine Abstract: We present statistical exploratory approaches to modeling dependence between components of high-dimensional brain signals. This work is motivated by a We present statistical exploratory approaches to modeling dependence between components of high-dimensional brain signals. This work is motivated by a