Deformable image registration for cone-beam CT guided transoral robotic base-of-tongue surgery

Abstract

Transoral robotic surgery (TORS) offers a minimally invasive approach to resection of base of tongue tumors. However, precise localization of the surgical target and adjacent critical structures can be challenged by the highly deformed intraoperative setup. We propose a deformable registration method using intraoperative cone-beam CT (CBCT) to accurately align preoperative CT or MR images with the intraoperative scene. The registration method combines a Gaussian mixture (GM) model followed by a variation of the Demons algorithm. First, following segmentation of the volume of interest (i.e., volume of the tongue extending to the hyoid), a GM model is applied to surface point clouds for rigid initialization (GM rigid) followed by nonrigid deformation (GM nonrigid). Second, the registration is refined using the Demons algorithm applied to distance map transforms of the (GM-registered) preoperative image and intraoperative CBCT. Performance was evaluated in repeat cadaver studies (25 image pairs) in terms of target registration error (TRE), entropy correlation coefficient (ECC), and normalized pointwise mutual information (NPMI). Retraction of the tongue in the TORS operative setup induced gross deformation >30 mm. The mean TRE following the GM rigid, GM nonrigid, and Demons steps was 4.6, 2.1, and 1.7 mm, respectively. The respective ECC was 0.57, 0.70, and 0.73 and NPMI was 0.46, 0.57, and 0.60. Registration accuracy was best across the superior aspect of the tongue and in proximity to the hyoid (by virtue of GM registration of surface points on these structures). The Demons step refined registration primarily in deeper portions of the tongue further from the surface and hyoid bone. Since the method does not use image intensities directly, it is suitable to multi-modality registration of preoperative CT or MR with intraoperative CBCT. Extending the 3D image registration to the fusion of image and planning data in stereo-endoscopic video is anticipated to support safer, high-precision base of tongue robotic surgery.

Deformable image registration with local rigidity constraints for cone-beam CT-guided spine surgery

“…In this work, Demons optimization was performed in a morphological image pyramid in which I 0 and I 1 were repeatedly downsampled to coarser resolution by a factor of 2 (e.g., downsampling factors of 8, 4, 2, and 1). The MSL was fixed at 0.5 voxels based on previous work and additional analysis (not shown for brevity) (Reaungamornrat et al, 2013). The convergence criterion was based on diminishing change in MSD as in previous work (Reaungamornrat et al, 2013).…”

“…The MSL was fixed at 0.5 voxels based on previous work and additional analysis (not shown for brevity) (Reaungamornrat et al, 2013). The convergence criterion was based on diminishing change in MSD as in previous work (Reaungamornrat et al, 2013). The compositive form in equation (2) was favored over the diffeomorphic form, since exponentiation of the motion field was found to accumulate numerical errors in the deformation estimate (Arsigny et al, 2006, Vercauteren et al, 2009).…”

Transoral Robotic Surgery: Image Guidance and Augmented Reality

“…Using CBCT as the primary intra-operative imaging modality, Liu et al [7,8] and Reaungamornrat et al [9] have described the development of an image guidance system through a sequence of experimental studies, beginning with 3D-printed anatomy combined with ex vivo tissue samples, progressing through in vivo animal (porcine) trials, and most recently culminating in human cadaveric investigation. While TORS applied to the resection of oropharyngeal cancers can significantly reduce morbidity and improve postoperative function in comparison to open surgery, the approach can present substantial challenges on account of the typically deep submucosal extent of the target lesion, and the surgeon's inability to palpate anatomical structures directly.…”

Deformable registration for image-guided spine surgery: preserving rigid body vertebral morphology in free-form transformations

“…Variation of the smoothing filters yields different deformation models with varying properties. 10,11 For example, smoothing only the update field results in a fluid deformation model, smoothing only the displacement field results in an elastic deformation model, and smoothing both fields results in a viscoelastic model. In this work, Demons optimization was performed in a multi-resolution pyramid in which I0 and I1 were downsampled by factors of 4, 2, and 1.…”

“…The MSL was fixed at 0.5 voxels based on previous work and additional analysis (not shown for brevity). 9,11 The compositive form in Eq. (1) was favored over the diffeomorphic form, since exponentiation of the motion field was found to accumulated numerical errors in the deformation estimate.…”