Integrated approach to 3D warping and registration from lung images

Computerized volumetric warping and registration of 3D lung images can provide objective, accurate, and reproducible measures to the understanding of human lung structure and function. It is also invaluable to the assessment of the presence of diseases and their response to therapy. However, due to the complexity of breathing motion, little work has been carried out in this research area. In this paper, we propose an integrated approach to implement volumetric lung warping and registration from 3D CT images obtained at different stages of breathing. Both feature points and lung surfaces at consecutive frames are incorporated as a priori knowledge for 3D warping to derive an initial sparse comprehensive displacement field. This comprehensive displacement field is then interpolated over the entire volume in an iterative fashion governed by a model derived from continuum mechanics and 3D optical flow. The iteration is based on an objective function defined by a weighted sum of continuity equation, brightness constraint of 3D optical flow and motion-discontinuity-preserving smoothness constraint. Therefore, the 3D warping is accomplished by minimizing such objective function. This integrated scheme is less sensitive to the distribution of feature points and is resilient to the errors introduced in the process of feature point matching. Preliminary results are visualized by overlaying the displacement field with the original images. Effectiveness of the algorithm is also evaluated according to several checking measures. We believe the proposed approach will open up new areas of research in lung image analysis that can make use of the results from lung volumes warping.