Robarts Research

Assessment of cardiac function and early detection of abnormality play an essential role in the diagnosis, treatment, and control of several heart diseases. Visual assessment of cardiac function may result in higher inter- and intra-observer variability. Alternatively, automating abnormality scoring and producing clinically relevant quantitative measurements are desirable to facilitate the disease diagnosis. Due to the similarity between the statistical information associated with normal and abnormal heart motions, the heart motion abnormality detection is acknowledged as a difficult problem, which has recently bestirred a significant research attention. Our study investigates several fundamental problems related to cardiac images acquired in regular clinical routine. First, this study investigates automating the analysis of global cardiac function, and proposes effective measures to process the information from the data. Then, it addresses the problem of uncertainties associated with cardiac motion and proposes solutions based on a more complex motion models. Finally, a solution to the regional myocardial abnormality analysis is developed in accordance with the clinical standard issued by the American Heart Association.

Neurodegenerative diseases afflict a significant portion of the world's aging population, with Alzheimer's disease (AD) alone affecting close to 500,000 Canadians.  Non-invasive magnetic resonance imaging has shown potential to detect structural changes at an earlier stage than neuropsychological or cognitive evaluations, however, the high degree of anatomical variability among our brains has proven to be a challenge in developing general computational methods. Neuroanatomical registration and segmentation are fundamental components of many structural and functional analysis techniques, thus advances here can lead to the development of biomarkers for early disease progression.  In this presentation I will present methods for improved accuracy and robustness using context-specific anatomical guidance in registration and intelligent atlas selection in multi-atlas segmentation, then demonstrate applications of these in shape analysis of the hippocampus in AD.