Events

Time Matters: Spatiotemporal Modeling and Analysis for Image Time Series

Lecture / Panel
 
For NYU Community

Speaker:  Guido Gerig, Scientific Computing and Imaging Institute (SCI) School of Computing, University of Utah

Abstract:

Clinical assessment routinely uses terms such as development, growth trajectory, aging, degeneration, disease progress, recovery or prediction. This terminology inherently carries the aspect of dynamic processes, suggesting that snapshots in time and cross-sectional analysis are not sufficient. This leads to longitudinal imaging studies where subjects are imaged repeatedly over time, or where dynamic imaging is applied to study functional changes. With volumetric 3D imaging in clinical applications, we therefore get 4D datasets which require suitable 4D image processing and analysis methodologies. Methods include image registration, establishing spatial correspondence across structures that may change shape and appearance, and segmentation and modeling tools that make use of the notion of repeated data.  

This talk discusses the development of new image analysis methodologies that carry the notion of linear and nonlinear regression, now applied to complex, high-dimensional data such as images and image-derived shapes and structures. This leads to the modeling of spatiotemporal time trajectories of structures relevant to specific applications in medicine, biology or even video-sequence analysis. Where repeated data for individuals is available, we will demonstrate

that statistical concepts of longitudinal data analysis such as nonlinear mixed-effect modeling (NLME) can be extended to structures and shapes modeled from dynamic image data. This research is driven by clinical studies such as the analysis of early brain growth in healthy subjects at risk for mental illness, monitoring of efficacy of therapeutic intervention in traumatic brain injury, and understanding neurodegeneration in normal aging and Alzheimer's disease. However, methodologies are generic and will find applications in a wide range of areas where we would like to extract spatiotemporal models from dynamic image series.

Bio: 

Guido Gerig received his Ph.D. in 1987 from the Swiss Federal Institute of Technology, ETH Zurich, Switzerland. In 1998, he was appointed Taylor Grandy Professor at the University of North Carolina at Chapel Hill with joint appointments in the departments of Computer Science and Psychiatry. Guido Gerig joined the SCI Institute at University of Utah in 2007, where he has a faculty position at the School of Computing, with adjunct positions at the departments of Bioengineering and Psychiatry. He serves as associate director of the SCI Institute and the director of the Utah Center for Neuroimage Analysis. He was appointed as a fellow of the American Institute for Medical and Biological Engineering (AIMBE) in 2010 and is a fellow of the MICCAI  society. 

Guido Gerig supports a number of clinical neuroimaging projects with methodology for image processing, registration, atlas building, segmentation, shape analysis, and statistical analysis. Current key research topics are segmentation of volumetric image data, spatiotemporal shape modelling from longitudinal image data, building of population atlases of volumetric images and embedded structures, and new methodologies for statistical analysis of diffusion tensor imaging (DTI). Driving clinical problems are studies of early brain development in subject at risk for mental illness, longitudinal infant studies of autism, assessment of pathology and changes due to therapeutic intervention in traumatic brain injury (TBI), and analysis of anatomical shape changes during pre-clinical Huntington's disease. Tools and methods developed through driving clinical applications are open source (ITK) and made available to the public via the NITRC resource.