The purpose of the proposed work is to develop a method for non-rigid registration of magnetic resonance (MR) images of the human lung. The motivation is the quantification of lung motion from the registration of successive images in serial MR acquisitions during normal respiration. MR quantification of pulmonary motion enables in vivo assessment of parenchymal mechanics within the lung in order to assist disease diagnosis or treatment monitoring. In particular, the long-term aim is to determine those mechanical characteristics which predict the severity of disease and the extent of recovery in the afflicted lung.
Specific aims for this proposal are as follows 1. To develop a method for non-rigid registration of MR images of the lung. There are two issues of concern when adapting existing registration algorithms to work with pulmonary MR images: first, we require a measure of image similarity to drive the registration process, and, second, we need to consider the deformation model governing the displacement of lung parenchyma during respiration. We will investigate both of these issues in this project 2. To characterize the registration algorithm over synthetic data. We will use synthetic motion data, for which perfect ground truth data is readily available, to systematically evaluate the performance of the proposed registration method under different (a) noise levels, (b) spatial resolutions and (c) degrees of parenchymal deformation. 3. To validate the registration algorithm over serial MR data. We will validate the proposed method on a set of serial MR studies, in which successive images of the lung acquired at different phases of the respiratory cycle will be registered and the results then evaluated by comparison against expert defined anatomical landmark correspondences between the images.
| Kiryu, Shigeru; Sundaram, Tessa; Kubo, Shigeto et al. (2008) MRI assessment of lung parenchymal motion in normal mice and transgenic mice with sickle cell disease. J Magn Reson Imaging 27:49-56 |
| Avants, Brian B; Schoenemann, P Thomas; Gee, James C (2006) Lagrangian frame diffeomorphic image registration: Morphometric comparison of human and chimpanzee cortex. Med Image Anal 10:397-412 |
| Sundaram, Tessa A; Gee, James C (2005) Towards a model of lung biomechanics: pulmonary kinematics via registration of serial lung images. Med Image Anal 9:524-37 |
| Gee, James; Sundaram, Tessa; Hasegawa, Ichiro et al. (2003) Characterization of regional pulmonary mechanics from serial magnetic resonance imaging data. Acad Radiol 10:1147-52 |
