Meeting Banner
Abstract #0482

Quantitative Analysis of Cardiac Motion Effects on in vivo Diffusion Tensor Parameters

Hongjiang Wei1, Magalie Viallon1, 2, Benedicte M.A. Delattre1, Lihui Wang1, Vinay M. Pai3, Han Wen3, Pierre Croisille1, 4, Yuemin Zhu1

1CREATIS,, CNRS (UMR 5220), INSERM (U1044),INSA Lyon,University of Lyon,, Lyon, France; 2Department of Radiology,University Hospitals of Geneva,, Geneva, Switzerland; 3Imaging Physics Lab, BBC/NHLBI/NIH,, Bethesda, Maryland 20892, United States; 4Jean-Monnet University,, Saint-Etienne, France

Cardiac motion is a crucial problem in in vivo diffusion tensor imaging (DTI) of the human heart. Its effects of on diffusion tensor parameters of the human heart have not been well established. Recently, an efficient method was proposed that acquires cardiac diffusion weighted (DW) images at different time points of the cardiac cycle and removes motion-induced signal loss using PCA filtering and temporal MIP techniques (PCATMIP). Meanwhile, polarized light imaging (PLI) provides us the ground-truth of the heart fiber architecture, and DENSE technique offers us higher spatial resolution 3D displacement fields of the human heart. These different imaging possibilities have led us to investigate a multimodal approach to quantitatively analyze the effects of cardiac motion on diffusion tensor parameters.

Abstract PDF

Keywords

acceleration accuracy achieved acquisition allow allowing analyzing appears applied applying architecture assessing background behavior blocks border boulder cardiac cardiovascular clinical combined compliance component computing conditions consecutive constructing contain corrected crucial cycle datasets delay dense derived describe described despite diffusion diffusivity dimension displacement echoes effectively efficient elevation empirical encoded encoding entire established ethical experimental fiber fields filtering flowchart fractional function generation gradient gray great ground guidelines healthy heart histological hospitals human importance improved in vivo induced intensity invest investigating legal light longitudinal loss mainly manual maps matrix measured measuring methyl model molecules motion myocardial myocardium narrower noise normalized original overestimation partial patterns physics polarized position possibilities post presently principal problem processed processing projection properties proposed quality quantitative quantitatively radiology realistic reduced regular relation removes represent represented resolution respectively sections segmentation series significantly simulated simulation solutions spatial stimulated strength structure subject subsequent suggests temporal tensor throughout tissues trans trigger truth variation volunteers water white yielding