Events Calendar

Anthropomorphic Digital Reference Objects: Extensible Tools for Evaluating Quantitative Imaging Algorithms

Speaker: Dr. Ryan Bosca , University of Wisconsin

Talk Title: Anthropomorphic Digital Reference Objects: Extensible Tools for Evaluating Quantitative Imaging Algorithms

Series: Medical Imaging Seminar Series

Abstract: Assessing and mitigating the various sources of bias and variance associated with MR image quantification algorithms (e.g., pharmacokinetic modeling of dynamic contrast enhanced MRI) is essential to the use of such algorithms in clinical research and practice. Grid based digital reference objects (DRO) have been used traditionally to assess such algorithms. More recently, a number of publicly available, normal patient derived, digital anthropomorphic tissue models have been developed. By assigning physical parameters to these tissue models in conjunction with a physiological model, for example, the general kinetic model (GKM), a new DRO can be generated. Furthermore, by incorporating a disease state (e.g., a tumor), the DRO can be made to more realistically represent standard operating conditions for quantitative imaging algorithms. Specifically, such DROs provide a means of assessing algorithm performance across a model parameter space and facilitating investigation of any spatially dependent biases and variances. This talk will provide an overview of the methodology for generating such DROs in addition to illustrating some potential example applications.

Biography: Dr. Bosca is a research associate in Medical Physics at the University of Wisconsin in Madison. He received his master degree in physics from the University of North Texas and his PhD in medical physics from the Graduate School of Biomedical Sciences at The University of Texas Health Science Center in Houston. His research interests include quantitative MR imaging, particularly pharmacokinetic modeling, the development of realistic digital phantoms to aid in assessing and mitigating sources of bias and variance in quantitative algorithms, and the application of multi parametric quantitative MR imaging (e.g., combing quantitative imaging biomarkers from dynamic contrast enhanced, diffusion tensor, dynamic susceptibility contrast MR techniques) for assessing treatment response. In addition, he is keenly interested in the development and validation of freely available, open source, software tools for use in quantitative imaging studies.

Host: Professor Krishna Nayak

Location: Hughes Aircraft Electrical Engineering Center (EEB) - 132

Audiences: Everyone Is Invited

Contact: Talyia White