Image processing

Dr. Trouard is an Associate Professor of Biomedical Engineering and Medical Imaging and a member of the Evelyn F. McKnight Brain Institute. His research involves the development and application of novel magnetic resonance imaging (MRI) techniques to understand human health and effectively treat disease. Dr. Trouard’s multidisciplinary work spans a range from basic studies of cell culture systems, to studies of preclinical animal models of disease, to clinical imaging in humans. Many aspects of this work are directed towards understanding and treatment of neurological disorders including Alzheimer's Disease, Niemann Pick Type C disease, Stroke and Cancer.

My research interests are in the field of theoretical image science with emphasis on medical imaging. I currently study task-based measures of image quality in which one must define the task the images are to be used for and the observer who will be performing this task, to properly measure and optimize the quality of images and imaging systems. We take the stance that imaging systems should be designed to best enable the observer to detect tumors and not base design decisions on resolution, contrast, etc. Topics in my research group include accurate system modeling, statistical modeling, observer performance metrics, signal-detection theory and general image science.

Biomedical imaging, in general, and various modalities of tomography are now an important part of medical practice and biomedical research. I develop mathematics of biomedical imaging. All modalities of tomography imaging rely heavily on mathematical algorithms for forming an image. My work involves developing the theory and the algorithm enabling this technology. By developing these techniques further, I contribute to improving health and life in the 21st century. Keywords: Electromagnetic and acoustic scattering; wave propagation; photonic crystals; spectral properties of high contrast band-gap materials and operators on graphs; computerized tomography.

Kobus Barnard, PhD, is an associate professor in the recently formed University of Arizona School of Information: Science, Technology, and Arts (SISTA), created to foster computational approaches across disciplines in both research and education. He also has University of Arizona appointments with Computer Science, ECE, Statistics, Cognitive Sciences, and BIO5. He leads the Interdisciplinary Visual Intelligence Lab (IVILAB) currently housed in SISTA. Research in the IVILAB revolves around building top-down statistical models that link theory and semantics to data. Such models support going from data to knowledge using Bayesian inference. Much of this work is in the context of inferring semantics and geometric form from image and video. For example, in collaboration with multiple researchers, the IVILAB has applied this approach to problems in computer vision (e.g., tracking people in 3D from video, understanding 3D scenes from images, and learning models of object structure) and biological image understanding (e.g., tracking pollen tubes growing in vitro, inferring the morphology of neurons grown in culture, extracting 3D structure of filamentous fungi from the genus Alternaria from brightfield microscopy image stacks, and extracting 3D structure of Arabidopsis plants). An additional IVILAB research project, Semantically Linked Instructional Content (SLIC) is on improving access to educational video through searching and browsing.Dr. Barnard holds an NSF CAREER grant, and has received support from three additional NSF grants, the DARPA Mind’s eye program, ONR, the Arizona Biomedical Research Commission (ABRC), and a BIO5 seed grant. He was supported by NSERC (Canada) during graduate and post-graduate studies (NSERC A, B and PDF). His work on computational color constancy was awarded the Governor General’s gold medal for the best dissertation across disciplines at SFU. He has published over 80 papers, including one awarded best paper on cognitive computer vision in 2002.