Highly accurate fast lung CT registration

Abstract

Lung registration in thoracic CT scans has received much attention in the medical imaging community. Possible applications range from follow-up analysis, motion correction for radiation therapy, monitoring of air flow and pulmonary function to lung elasticity analysis. In a clinical environment, runtime is always a critical issue, ruling out quite a few excellent registration approaches. In this paper, a highly efficient variational lung registration method based on minimizing the normalized gradient fields distance measure with curvature regularization is presented. The method ensures diffeomorphic deformations by an additional volume regularization. Supplemental user knowledge, like a segmentation of the lungs, may be incorporated as well. The accuracy of our method was evaluated on 40 test cases from clinical routine. In the EMPIRE10 lung registration challenge, our scheme ranks third, with respect to various validation criteria, out of 28 algorithms with an average landmark distance of 0.72 mm. The average runtime is about 1:50 min on a standard PC, making it by far the fastest approach of the top-ranking algorithms. Additionally, the ten publicly available DIR-Lab inhale-exhale scan pairs were registered to subvoxel accuracy at computation times of only 20 seconds. Our method thus combines very attractive runtimes with state-of-the-art accuracy in a unique way

Automatic lung segmentation method for MRI-based lung perfusion studies of patients with chronic obstructive pulmonary disease

“…where F denotes the domain of F. NGF focuses on edges instead of intensities and has proven to be a reliable distance measure in inhale/exhale registrations of chest CT scans as shown by Rühaak et al [18]. As regularization term elastic regularization is chosen (see, e.g., Modersitzki [15]):…”

Evaluation of Scan-Line Optimization for 3D Medical Image Registration

“…It is described in [24] as a variational approach, which is based on second order regularization and a data term that penalizes deviations of gradient orientations. It employs affine linear transformation prior to nonlinear registration.…”

“…Many methods in this field follow a variational approach and often utilize prior knowledge, such as lung segmentation masks [1,17,16,24,22,26], an initial solution from an affine-linear pre-registration [24], a sparse set of landmark pairs for initialization [22], or they incorporate a diffeomorphic motion assumption into the energy model [1,26] and perform symmetric registration [1,13]. Results of a recent challenge and study presented in [20] established a comprehensive overview of general state-of-the-art methods.…”

A fast alignment method for breast MRI follow-up studies using automated breast segmentation and current-prior registration

“…Thirdly, the segmented breast volumes are registered with a non-rigid registration algorithm producing a final deformation field. The proposed registration scheme is inspired by the works of Rühaak et al, 6 which uses the discretize-then-optimize paradigm in a multilevel Gauss-Newton optimization framework. 7 A schematic overview of the entire workflow is given in Fig.…”