Multimodal Image Registration for Pre- and Intra-Operative Imaging
using Ultrasound (MulRegUS)

Motivation

This project is motivated by the need to improve the safety and effectiveness of interventional and minimally invasive procedures guided by different imaging modalities. This can be accomplished by developing methods that are able to be aware of the  anatomical context and accurately produce a combined visualization of several image sources in a real time.  Ultrasonography (US) is a real time image acquisition modality capable of providing 2D and 3D imaging with high frame rate. In addition, it is also a convenient tool because it is inexpensive and  non-ionizing [1]. However, US images often  contain more image artifacts than most other modalities, and also often suffers from the existence of shadows, noise and speckle [2,3]. On the other hand, there are imaging modalities are able to provide better anatomical definition, such as CT (Computed Tomography) imaging provides higher details in bony structures, while MR (Magnetic Resonance) imaging provides higher details in soft tissues [4].

Combining several imaging modalities could fuse the advantages of each modality to create more comprehensive information about the anatomy of interest. This is made possible by image registration, which is the process of transforming different sets of data into one coordinate system [5]. In this case, combining US and CT/MR or fluoroscopic images could result in images with comprehensive information and in real time mode. It could lead to a more detailed information thus better understanding of the medical situation, and hence could lead to a better diagnostic and intra-operative guidance, and finally to a better treatment results.

The result of this project has many potential applications. In Image Guided Surgery (IGS), it could be implemented as a part of the IGS platform, for example Resection Map [6]. This navigation system resulting from the ARISER-project is now revitalized in the current Hepa-Navi project, which got funding for a postdoc from Helse Sør-Øst in 2014 (Hepa-Navi: Planning and Navigation System for Liver Resection). Integration of US to the IGS platform Hepa-Navi could provide valuable real-time updated images of the organ to the rigid map intra-operatively. The most important user of the results from this work implemented into the Hepa-Navi surgical navigation platform will be surgeons, especially laparoscopic surgeons, performing liver resection and possibly pancreas surgery. 

US registration can also be implemented in surgical robotics systems.  As an example, in I-SUR (Intelligent Surgical Robotics), US is used as continuous guidance of the robot to perform a needle insertion procedure, see www.isur.eu , in which IVS is one of partners.

3D transesophageal echocardiography (TEE) mixed with fluoroscopy can be beneficial for Improved visualization in Mitral Clip placement and Transcatheter Aortic valve implantation (TAVI). Recently two new catheter based interventions is presented clinically. The proposed project aims at studying how pre-operative 3D echocardiography can be used and improved for intra-procedural guidance of two interventions:- Mitral Clip placement and Transcatheter Aortic valve implantation (TAVI). The repair and replacement of valves through catheterized procedures has just recently become a reality. Currently, catheter guidance and positioning is primarily supported by X-ray fluoroscopy with support from 2D ultrasound, either an external cardiac probe or TEE. The aim of this sub-project is to provide a better and more intuitive visualization of the combination of this new 3D TEE probe with fluoroscopic image by means of image fusion techniques.

The proposed project overall is aimed at developing methods to register/align 2D or 3D Ultrasound images to other pre-operative imaging modalities such as CT/MRI images or fluoroscopic projection images. Its expected contribution is towards the development of new methodology for Ultrasound (US) registration with improved accuracy and speed. The result of this will contribute to a major challenge in soft tissue surgery and catheter navigation. 

Aims

The project aims to develop registration techniques between intra-operative images of US and pre-operative images of MR/CT or a pre-operative model such as a 3D model.

The main specific targets to be achieved in this project are:

• to develop a image registration method that shows improved accuracy and speed compared to existing comparable techniques
• to extend the proposed method into a non-rigid registration method
• to apply the developed method for real-time update of model in surgical soft-tissue navigation or for mixed visualization in catheter navigation

Secondary objectives:

• to study existing techniques of US probe calibration and registration
• to develop and validate a US probe calibration method
• to develop and validate a rigid registration method of US and another pre-operative image or its 3D model

References

1. C. Leung, L.H-Zaad, P. Foroughi, and P. Abolmaesumi. ”A Real-Time Intrasubject Elastic Registration Algorithm for Dynamic 2-D Ultrasound Images.” Ultrasound in Medicine and Biology 35(7), pp. 1159-1176, 2009.
2 . G.P. Penney, et. al. , ”Registration of freehand 3D ultrasound and magnetic resonance liver images." Medical Image Analysis 8, pp. 81-91, 2004.
3. R.D. Bucholz, et. al. ”The Correction of Stereotactic Inaccuracy Caused by Brain Shift Using an Intraoperative Ultrasound Device.” Proceedings of CVR Med-MRCAS, pp. 459-466, 1997.
4. -." CT Scan vs MRI - Difference and Comparison." [Online] Dec 4, 2011. [Cited: May 13, 2011.] <http://www.diffen.com/difference/CT_Scan_vs_MRI>.
5. -."Image Registration". Wikipedia. [Online] Nov 21, 2011. [Cited: May 13, 2014.] <http://en.wikipedia.org/wiki/Image_registration>.
6. Lamata P,  et. al. " Use of the Resection Map system as guidance during hepatectomy." Surg Endosc. 24(9), pp. 2327-37, Sep 2010.