Atlases and statistical modeling of vascular networks from medical images

点击翻译按钮获取中文摘要

Statistical atlases are powerful tools for modeling inter-individual variability within a population. An atlas built from a healthy population allows one to define the normality of the anatomical structures of interest and to efficiently detect the abnormalities in pathological cases. The overall objective of this research program is to develop novel computational tools for constructing statistical atlases of vascular networks. These tools will be applied in ophthalmology and in cardiology to objectively define vascular abnormalities and better understand their exact role in ocular and cardiovascular diseases. Furthermore, given the systemic nature of vascular diseases, the eye represents a unique location in the body to visualize blood vessels non-invasively and can be used as a natural window to the heart. Thus, the relationship between microvascular abnormalities in the retina and cardiovascular abnormalities will be explored. The short-term objectives are: OBJ 1) build and validate a statistical atlas of retinal vessel networks from fundus images in healthy populations; OBJ 2) build and validate a statistical atlas of coronary artery networks from multi-view angiographic sequences; OBJ 3) demonstrate the usefulness of these vascular atlases in understanding the correlation between topological / morphological changes in retinal vessels and functional changes in coronary arteries in a population affected by cardiovascular disease.The long-term vision is to contribute to a paradigm shift in healthcare by promoting blindness prevention and cardiovascular diseases risk stratification for lifestyle intervention. The outcome of this DG will be the first step toward that goal. The long-term objective will be to use the proposed atlases in the non-invasive assessment of an individual's retinal vascular changes over time. Precise and accurate analysis methods will be required to automatically detect subtle but significant changes in retinal microvascular networks related to manifestations of ocular or cardiovascular diseases in the pre-symptomatic stages.We believe that our program will lead to major advances in computational modeling of vascular networks. This research program will address significant technical challenges to develop statistical atlases of the retinal vessel network and the coronary arteries. These models will provide powerful tools that will allow an early detection of subtle changes in the topology, morphology and functionality of vascular networks. Ocular diseases are the most common cause of blindness in the working-age population, while cardiovascular disease continues to be the leading cause of death in North America. Thus, the proposed research will lead to significant cost savings to the Canadian healthcare system and to the economy and society by preventing serious, even life-threatening, health conditions.