We present a new image-based technique to rigidly register intra-operative 3D ultrasound (US) with pre-operative Magnetic Resonance (MR) images. Automatic registration is achieved by maximization of a similarity measure which generalizes the correlation ratio (CR), and whose novelty is to incorporate multivariate information from the MR data (intensity and gradient). In addition, the similarity measure is built upon a robust intensity-based distance measure, which makes it possible to handle a variety of US artifacts. A cross-validation study has been carried out using a number of phantom and clinical data. This indicates that the method is quite robust and that the worst registration errors are of the order of the MR image resolution. Keywords: image registration, ultrasound, magnetic resonance, correlation ratio, robust estimation. 1

Automatic processing of 3D ultrasound (US) is of great interest for the development of innovative and low-cost computer-assisted surgery tools. In this paper, we present a new image-based technique to rigidly register intra-operative 3D US with pre-operative Magnetic Resonance (MR) data. Automatic registration is achieved by maximization of a similarity measure that generalizes the correlation ratio (CR). This novel similarity measure has been designed to better take into account the nature of US images. A preliminary cross-validation study has been carried out using a number of phantom and clinical data. This indicates that the worst registration errors are of the order of the MR resolution.

During a neurosurgical intervention, the brain tissues shift and warp. In order to keep an accurate positioning of the surgical instruments, one has to estimate this deformation from intra-operative images. We present in this article a feasibility study of a tracking tool based on intra-operative 3D ultrasound (US) image sequences. The automatic processing of this kind of images is of great interest for the development of innovative and low-cost image guided surgery tools. The difficulty relies both in the complex nature of the ultrasound image, and in the amount of data to be treated as fast as possible.

During a neuro-surgical intervention, the brain tissues shift and warp. In order to keep an accurate positioning of the surgical instruments, one has to estimate this deformation from intra-operative images. 3D ultrasound (US) imaging is an innovative and low-cost modality which appears to be suited for such computer-assisted surgery tools. In this paper, we present a new image-based technique to register intra-operative 3D US with pre-operative Magnetic Resonance (MR) data. A first automatic rigid registration is achieved by the maximisation of a similarity measure that generalises the correlation ratio. Then, brain deformations are tracked in the 3D US time-sequence using a “demon’s” like algorithm. Experiments show that a registration accuracy of the MR voxel size is achieved for the rigid part, and a qualitative accuracy of a few millimetres could be obtained for the complete tracking system. 1.

Abstract. 3D registration of ultrasound images is an important and fast-growing research area with various medical applications, such as image-guided radiotherapy and surgery. However, this registration process is extremely challenging due to the deformation of soft tissue and the existence of speckles in these images. This paper presents a novel intra-modality elastic registration technique for 3D ultrasound images. It uses the general concept of attribute vectors to find the corresponding voxels in the fixed and moving images. The method does not require any pre-segmentation and does not employ any numerical optimization procedure. Therefore, the computational requirements are very low and it has the potential to be used for real-time applications. The technique is implemented and tested for 3D ultrasound images of liver, captured by a 3D ultrasound transducer. The results show that the method is sufficiently accurate and robust and is not easily trapped with local minima. 1

Freehand three-dimensional ultrasound is a technique for acquiring ultrasonic data of a 3D volume by recording the trajectory of the ultrasound probe using a position sensor. In planning and registration, a freehand ultrasound systems is used to track a two-dimensional probe. Probe calibration is necessary to find the rigid body transformation from the coordinate system of the B-scan to that of the mobile part of the position sensor. Numerous techniques for this have been developed over the past decade. In this review, we give a comprehensive description of existing calibration techniques and classify them according to the mathematical principles on which they are based. We give a thorough analysis of these approaches based on their accuracy, ease of use, reliability, and speed of calibration. To ensure consistency, these comparisons are done by the authors based on experimental results and not on figures quoted in previous papers. 1

Objective: Applications of three-dimensional ultrasound (3D US) are emerging throughout the field of medicine. In this study, tracked, free-hand 3D phantom US images were mapped to computed tomograms (CT) as a development for image-guided surgery (IGS) of the spine. In the operating room, the registration of tracked 3D US images to other imaging modalities, such as CT, could allow the surgeon to identify more precisely the surgical target area prior to the incision. An independent quantitative measure of registration accuracy using a fiducial marker system was provided. Methods: Three-dimensional free-hand US images of a phantom spine were created by tracking the transducer with an optical sensing system. Two sets of images were acquired from three lumbar vertebrae using 4.5- and 7.5-MHz transducers. These images were then segmented for the extraction of the posterior vertebral surface. Next, a surface-based registration of US to the corresponding segmented CT images was performed. Registration errors were computed as the distance between a set of target points transformed using the experimental transformation and the same set of target points transformed using fiducial markers as a gold standard. Results: Results indicated that alignment of these image sets is feasible using only part of the

This paper considers registration of 3D ultrasound volumes. One way to acquire 3D data is to use a mechanically-swept 3D probe. However, the usefulness of these probes is restricted by their limited field of view. While this problem can be overcome by attaching a position sensor to the probe, an external position sensor can be an inconvenience in a clinical setting and does not align the volumes correctly when there is tissue displacement or deformation. The objective of this paper is to replace the 6 degree of freedom (DOF) sensor with a combination of 3 DOF image registration and an integrated intertial sensor for measuring orientation. We examine a range of optimisation algorithms and similarity measures for registration and compare them in in vitro and in vivo experiments. We register based on multiple reslice images rather than a whole voxel array. In this paper, we use a large number of reslices for improved reliability at the expense of computational speed. We have found that the Levenberg-Marquardt method is very fast but is not guaranteed to give the correct solution all the time. We conclude that normalised mutual information used in the Nelder-Mead simplex algorithm is potentially suitable for the registration task with an average execution time of around five minutes, in the majority of cases, with two restarts in a C++ implementation on a 3.0 GHz Intel Core 2 Duo CPU machine. 1