Marie Curie Action

Radio Frequency Ablation

Radio frequency ablation was the first clinical application to be targeted by the ARIS*ER consortium. ARIS*ER has invested a lot of effort into researching relevant technological aspects to improve upon the current state of RFA procedures.

At the mid-way stage of the ARIS*ER project, a demo was developed which integrated all of the research results which potentially can improve RFA procedures. This demo was developed such that user evalutation can be carried out evaluating the user-requirements which were coined by the user centred design group and evaluating visulization tools developed in cooperation between the visualization group, human factors group and clinical personnel. In addition the demo will act as a testbed to test new real-time image processing tools and visualization tools. The demo will thereby act as a bridge for discussions between engineers and surgeons to further improve the RFA procedures.

Video of RFA Demo

Figure 1: Movie of RFA Demo

Research

The research focused on four areas.

User centered design and human factors

New methods for finding and organizing user-requirements were developed. These methods were used in clinical focus groups to pinpoint all the relevant user-requirements for RFA-procedures. In cooperation with both doctors and engineers, a new innovative design for a system which supports RFA was developed.

Image registration

Clinical personnel performing RFA procedures are dependent on having intra-operative real-time image guidance tools to accurately hit the tumor with the RF-probe. Hence, ultrasound is commonly the intra-operative real-time image modality used. To be able to incorporate pre-operative surface models such as vessel structures, tumor and liver with the intra-operative ultrasound, new registration algorithms were developed. In particular, a method for registration of intra-operative US with MR was developed.

Image segmentation

Hitting the pathological structure with the RF-probe guided by an intra-operative image guidance tools based on pre-operative segmentations of the structure requires very accurate image segmentation algorithms. Accurate and robust image segmentation algorithms were developed for segmentation of liver, liver tumors and liver vessels.

Image Visualization and Navigation

An integral part of any image navigation tools are how the information available in the system is presented to the end-user. In this research area we have developed a number of visualization tools to improve the spatial understanding of where the structures are in relation to the RF-probe as well as tools which warn the surgeon when he is about to hit critical structures such as vessels.

RFA System

The final system consists of three modules which supports the pre-, intra- and post-operative parts of the surgery.

Pre-Operative Module

This module supports the pre-operative exploration of pre-operative CT/MR volumetric images in addition to 3-D views of segmented structures in relation 2-D slices of the volumetric images.

 

Figure 2a: The left image shows the post-operative module where the pre-operative image is visualized.

Figure 2b: The module supports a seemless transition between pre-operative and post-operative images through the use of a transparency slider. The right image shows the post-operative image.

Intra-Operative Module

Surgeons have to make many tough decisions during surgical procedures. Any intra-operative guidance system should take this into account and provide the surgeon with easy accessible information which increases his ability to make quick and accurate decisions. The intra-operative module of the RFA system does exactly this through providing the surgeon with a number of image modalities all fused together.

   

Figure 3: These images shows the hepatic (red) and portal (blue) vessels, tumor (green) and liver (brown) overlayed on the phantom.

   

Figure 4: The system allows fusion of the intra-operative Ultrasound image with segmented structures from pre-operative images. (In this case the vessel structure).

   

Figure 5: For instance when planning the insertion path of the RF probe it is useful to just see the relation of the tumor (green) to the vessel structures.

   

Figure 6: Ultrasound contains a low signal to noise ratio. Hence, in some cases it is useful to rather see the pre-operative CT overlayed on with the relevant anatomical structures which facilitates easier orientation and planning of insertion path.

Post-Operative Module

After surgery it is important to validate the results. Hence, the post-operative module supports fusion of pre-operative and post-operative CT/MR images to assess the effect of the ablation.

 

Figure 7a: This left image shows post-operative module where the pre-operative image is visualized.

Figure 7b: The module supports a seemless transition between the pre-operative and post-operative image through the use of a transparency slider. The right image shows the post-operative image.

Contacts