Rigid Registration

3D-3D MR - CT Example

(Click on image for animation)

2D-3D MR - Example From the work of Lilla Zollei

Before and After Registration of 3D CT to a Pair of X-Ray Images

The SPL has used several technologies for the rigid registration of medical data. During neurosurgical procedures performed in the conventional operating room, rigid registration has been used to relate the position of the patient to pre-operatively obtained images and 3D models. This registration has been carried out using several methods. Originally it was done manually (using the term "video registration"). Subsequently,an automated procedure was devised that made use of skin surface measurements obtained from a laser scanner. This same automated method has also been used in conjunction with skin surface measurements obtained from a hand-held wand.

Mutual Information (MI) is another technology that is commonly used at the SPL for the rigid registration of 3D medical image data. This technique is based on the intensities in the data, and is suitable for registering data of differing modalities. It is used in the SPL for the pre-operative fusion of conventional MRI, MR angiography, CT , fMRI and SPECT data. This method has been applied in neurosurgical procedures carried out in the intervantional MRI. In this application, a fast volume scan is performed after the patient is clamped, and this scan is registered to a pre-operative MRI in order tobring the previously prepared suite of imagery and models into conjunction with the current positioning of the patient. The first example above shows an mpeg move of a run of the maximization of mutual information registration method on CT and MRI data.

The MI approach has also recently been applied to the the rigid registration of volumetric data with pairs of projection x-ray images (see second example above).

People involved in these research projects:

Sandy Wells

Eric Grimson

David Gering

Ron Kikinis

Lilla Zollei

Related papers:

L . Zöllei, E. Grimson, A. Norbash, W. Wells: "2D-3D Rigid Registration of X-Ray Fluoroscopy and CT Images Using Mutual Information and Sparsely Sampled Histogram Estimators", IEEE CVPR, 2001.

D. Gering, A. Nabavi, R. Kikinis, N. Hata, L. Odonnell, W. Eric L. Grimson, F. Jolesz, P. Black, W. Wells III. An Integrated Visualization System for Surgical Planning and Guidance Using Image Fusion and an Open MR. Journal of Magnetic Resonance Imaging, Vol 13, pp. 967-975, June, 2001.

Hata et al. Multimodality Deformable Registration of Pre- and Intraoperative Images for MRI-Guided Brain SurgeryProceedings of First International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI'98),pp. 1067-1074, ISBN 3-540-65136-5, Lecture Notes in Computer Science, Eds. W.M. Wells, A. Colchester, and S. Delp, SpringerVerlag, October, 1998. Boston.

Wells et al. Multi-Modal Volume Registration by Maximization of Mutual Information.Medical Image Analysis, 1(1):35-51 (1996).

Grimson et al. An Automatic Registration Method for Frameless Stereotaxy, Image Guided Surgery and Enhanced Reality Visualization. IEEE TMI, April 1996.

Wells et al. Multi-Modal Volume Registration by Maximization of Mutual Information.Proc. Second International Symposium on Medical Robotics and Computer Assisted Surgery, 1995. p55-62. Wiley.