EN FR
EN FR


Section: New Results

Biomechanics-based graph matching for augmented CT-CBCT

Augmenting intraoperative cone beam computed tomography (CBCT) images with preoperative computed tomography (CT) data in the context of image-guided liver therapy is proposed (see Fig. 5). The expected benefit is an improved visualization of tumor(s), vascular system and other internal structures of interest. An automatic elastic registration based on matching of vascular trees extracted from both the preoperative and intraoperative images is presented. Although methods dedicated to non-rigid graph matching exist, they are not efficient when large intraoperative deformations of tissues occur, as is the case during the liver surgery. First, an improved graph matching algorithm using Gaussian process is introduced by imposing additional constraints during the matching when the number of hypotheses is large; this extended version does not require a manual initialization of matching. Second, a fast biomechanical model is employed to make the method capable of handling large deformations. The proposed automatic intraoperative augmentation is evaluated on both synthetic and real data. It is demonstrated that the algorithm is capable of handling large deformations, thus being more robust and reliable than previous approaches. Moreover, the time required to perform the elastic registration is compatible with the intraoperative navigation scenario. The input data and result of the biomechanics-based graph matching method, which can handle large deformations and augment intraoperative CBCT, is shown in Fig. 5 .

Figure 5. In the top left the preoperative CT image with complete portal and hepatic vessels clearly visible. In the bottom left the intraoperative deform, noisy CBCT image with partial only partial portal vessels visible. In the right the augmented CBCT view with the complete vessels added form preoperative image.
IMG/CBCTaugmented.png
  • Authors: Jaime Garcia Guevara and Igor Peterlik and Marie-Odile Berger and Stephane Cotin

  • Type: Journal publication IJCARS 2018