Elastic registration of 3D ultrasound images

DRAMMS: Deformable registration via attribute matching and mutual-saliency weighting

A general-purpose deformable registration algorithm referred to as "DRAMMS" is presented in this paper. DRAMMS bridges the gap between the traditional voxel-wise methods and landmark/feature-based methods with primarily two contributions. First, DRAMMS renders each voxel relatively distinctively identifiable by a rich set of attributes, therefore largely reducing matching ambiguities. In particular, a set of multi-scale and multi-orientation Gabor attributes are extracted and the optimal components are selected, so that they form a highly distinctive morphological signature reflecting the anatomical and geometric context around each voxel. Moreover, the way in which the optimal Gabor attributes are constructed is independent of the underlying image modalities or contents, which renders DRAMMS generally applicable to diverse registration tasks. A second contribution of DRAMMS is that it modulates the registration by assigning higher weights to those voxels having higher ability to establish unique (hence reliable) correspondences across images, therefore reducing the negative impact of those regions that are less capable of finding correspondences (such as outlier regions). A continuously-valued weighting function named "mutual-saliency" is developed to reflect the matching uniqueness between a pair of voxels implied by the tentative transformation. As a result, voxels do not contribute equally as in most voxel-wise methods, nor in isolation as in landmark/feature-based methods. Instead, they contribute according to the continuously-valued mutual-saliency map, which dynamically evolves during the registration process. Experiments in simulated images, inter-subject images, single-/multi-modality images, from brain, heart, and prostate have demonstrated the general applicability and the accuracy of DRAMMS.

Real-time MRI-TRUS fusion for guidance of targeted prostate biopsies

Targeted prostate biopsy is challenging because no currently established imaging modality is both accurate for prostate cancer diagnosis and cost-effective for real-time procedure guidance. A system that fuses real-time transrectal ultrasound images with previously acquired endorectal coil MRI images for prostate biopsy guidance is presented here. The system uses electromagnetic tracking and intraoperative image registration to superimpose the MRI data on the ultrasound image. Prostate motion is tracked and compensated for without the need for fiducial markers. The accuracy of the system in phantom studies was shown to be 2.4 +/- 1.2 mm. The fusion system has been used in more than 20 patients to guide biopsies with almost no modification of the conventional protocol. Retrospective clinical evaluation suggests that clinically acceptable spatial accuracy can be achieved.

Towards 3D ultrasound image based soft tissue tracking: a transrectal ultrasound prostate image alignment system

The emergence of real-time 3D ultrasound (US) makes it possible to consider image-based tracking of subcutaneous soft tissue targets for computer guided diagnosis and therapy. We propose a 3D transrectal US based tracking system for precise prostate biopsy sample localisation. The aim is to improve sample distribution, to enable targeting of unsampled regions for repeated biopsies, and to make post-interventional quality controls possible. Since the patient is not immobilized, since the prostate is mobile and due to the fact that probe movements are only constrained by the rectum during biopsy acquisition, the tracking system must be able to estimate rigid transformations that are beyond the capture range of common image similarity measures. We propose a fast and robust multi-resolution attribute-vector registration approach that combines global and local optimization methods to solve this problem. Global optimization is performed on a probe movement model that reduces the dimensionality of the search space and thus renders optimization efficient. The method was tested on 237 prostate volumes acquired from 14 different patients for 3D to 3D and 3D to orthogonal 2D slices registration. The 3D-3D version of the algorithm converged correctly in 96.7% of all cases in 6.5s with an accuracy of 1.41mm (r.m.s.) and 3.84mm (max). The 3D to slices method yielded a success rate of 88.9% in 2.3s with an accuracy of 1.37mm (r.m.s.) and 4.3mm (max).

Fast deformable registration of 3D-ultrasound data using a variational approach

We present an intensity based deformable registration algorithm for 3D ultrasound data. The proposed method uses a variational approach and combines the characteristics of a multilevel algorithm and the properties of ultrasound data in order to provide a fast and accurate deformable registration method. In contrast to previously proposed approaches, we use no feature points and no interpolation technique, but compute a dense displacement field directly. We demonstrate that this approach, although it includes solving large PDE systems, reduces the computation time if implemented using efficient numerical techniques. The performance of the algorithm is tested on multiple 3D US images of the liver. Validation is performed by simulations, similarity comparisons between original and deformed images, visual inspection of the displacement fields and visual assessment of the deformed images by physicians.

Towards real-time registration of 4D ultrasound images

In this paper, we demonstrate a method for fast registration of sequences of 3D liver images, which could be used for the future real-time applications. In our method, every image is elastically registered to a so called fixed ultrasound image exploiting the information from previous registration. A few feature points are automatically selected, and tracked inside the images, while the deformation of other points are extrapolated with respect to the tracked points employing a fast free-form approach. The main intended application of the proposed method is real-time tracking of tumors for radiosurgery. The algorithm is evaluated on both naturally and artificially deformed images. Experimental results show that for around 85 percent accuracy, the process of tracking is completed very close to real time.