Target Registration Error Reduction Research Articles

Scale-adaptive deep network for deformable image registration.

Multiresolution hierarchical strategy is typically used in conventional optimization-based image registration to capture varying magnitudes of deformations while avoiding undesirable local minima. A rough concept of the scale is captured in deep networks by the reception field of kernels, and it has been realized to be both desirable and challenging to capture convolutions of different scales simultaneously in registration networks. In this study, we propose a registration network that is conscious of and self-adaptive to deformation of various scales to improve registration performance. Dilated inception modules (DIMs) are proposed to incorporate receptive fields of different sizes in a computationally efficient way. Scale adaptive modules (SAMs) are proposed to guide and adjust shallow features using convolutional kernels with spatially adaptive dilation rate learned from deep features. DIMs and SAMs are integrated into a registration network which takes a U-net structure. The network is trained in an unsupervised setting and completes registration with a single evaluation run. Experiment with two-dimensional (2D) cardiac MRIs showed that the adaptive dilation rate in SAM corresponded well to the deformation scale. Evaluated with left ventricle segmentation, our method achieved a Dice coefficient of (0.93±0.02), significantly better than SimpleElastix and networks without DIM or SAM. The average surface distance was less than 2mm, comparable to SimpleElastix without statistical significance. Experiment with synthetic data demonstrated the effectiveness of DIMs and SAMs, which led to a significant reduction in target registration error (TRE) based on dense deformation field. The three-dimensional (3D) version of the network achieved a 2.52mm mean TRE on anatomical landmarks in DIR-Lab thoracic 4DCTs, lower than SimpleElastix and networks without DIM or SAM with statistical significance. The average registration times were 0.002s for 2D images with size 256×256 and 0.42s for 3D images with size 256×256×96. The introduction and integration of DIMs and SAMs addressed the heterogeneous scale problem in an efficient and self-adaptive way. The proposed method provides an alternative to the inefficient multiresolution registration setups.

Target registration error reduction for percutaneous abdominal intervention.

A real-time methodology that finds spatio-temporal correspondence between the positions of the target point in the pre-treatment 3DCT image and during the procedure was proposed. It based on minimizing the target registration error in III tier registration circuits. Particle Swarm Optimization and Differential Evaluation were used to find optimal values of Elastic Body Spline parameters in the generation of abdominal deformation field. Different transformation classes have been tested: rigid, affine, Thin Plate Spline, Elastic Body Spline. The lowest TRE was obtained for the swarm optimization algorithm - differential evolution for the rigid and affine version: 3.47 and 3.73 mm, respectively.

Intraoperative image guidance in transoral robotic surgery: A pilot study.

Intraoperative image guidance during transoral robotic surgery (TORS) is hampered by imaging-friendly instrumentation and intraoperative positioning. The purpose of this study was to develop and validate an accurate image-guidance system for TORS. A custom radiolucent mouth retractor was fabricated from biocompatible material (Med-610; Stratasys, Minneapolis, MN). Teflon beads were placed in the oropharynx and carotid arteries of 3 cadavers. CT scans were obtained in the preoperative and intraoperative positions. Displacement of targets between preoperative and intraoperative scans was measured. Surgical navigation was based on the open-source Image-Guided Surgery Toolkit. Target registration error (TRE) was determined by measuring the distance between the tracker and bead registered to preoperative versus intraoperative scans. The inferior oropharyngeal targets demonstrated the greatest displacement between positions. A significant reduction in TRE was observed when registering the tracker to the intraoperative compared to the preoperative scan. This study describes an accurate intraoperative image-guidance system for TORS.

Intraoperative Image Guidance in Transoral Robotic Surgery – A Pilot Study

Background Intraoperative image guidance during transoral robotic surgery (TORS) is hampered by imaging-friendly instrumentation and intraoperative positioning. The purpose of this study was to develop and validate an accurate image-guidance system for TORS. Methods A custom radiolucent mouth retractor was fabricated from biocompatible material (Med-610; Stratasys, Minneapolis, MN). Teflon beads were placed in the oropharynx and carotid arteries of 3 cadavers. CT scans were obtained in the preoperative and intraoperative positions. Displacement of targets between preoperative and intraoperative scans was measured. Surgical navigation was based on the open-source Image-Guided Surgery Toolkit. Target registration error (TRE) was determined by measuring the distance between the tracker and bead registered to preoperative versus intraoperative scans. Results The inferior oropharyngeal targets demonstrated the greatest displacement between positions. A significant reduction in TRE was observed when registering the tracker to the intraoperative compared to the preoperative scan. Conclusion This study describes an accurate intraoperative image-guidance system for TORS.

Biomechanically Constrained Surface Registration: Application to MR-TRUS Fusion for Prostate Interventions.

In surface-based registration for image-guided interventions, the presence of missing data can be a significant issue. This often arises with real-time imaging modalities such as ultrasound, where poor contrast can make tissue boundaries difficult to distinguish from surrounding tissue. Missing data poses two challenges: ambiguity in establishing correspondences; and extrapolation of the deformation field to those missing regions. To address these, we present a novel non-rigid registration method. For establishing correspondences, we use a probabilistic framework based on a Gaussian mixture model (GMM) that treats one surface as a potentially partial observation. To extrapolate and constrain the deformation field, we incorporate biomechanical prior knowledge in the form of a finite element model (FEM). We validate the algorithm, referred to as GMM-FEM, in the context of prostate interventions. Our method leads to a significant reduction in target registration error (TRE) compared to similar state-of-the-art registration algorithms in the case of missing data up to 30%, with a mean TRE of 2.6 mm. The method also performs well when full segmentations are available, leading to TREs that are comparable to or better than other surface-based techniques. We also analyze robustness of our approach, showing that GMM-FEM is a practical and reliable solution for surface-based registration.

Minimization of target registration error for vertebra in image-guided spine surgery

The accuracy of pedicle screw placement during image-guided spine surgery (IGSS) can be characterized by estimating the target registration error (TRE). The major factors that influence TRE were identified, minimized, and verified with in vitro experiments. Computed-tomography-compatible markers are placed over anatomical landmarks of lumbar vertebral segments in locations that are feasible and routinely used in surgical procedures. TRE was determined directly for markers placed on the pedicles of vertebra segments. First, optimum selections of landmarks are proposed for different landmarks according to the minimum achievable TRE values in different configurations. These anatomical landmarks are feasible and accessible to overcome constraints that may be imposed during surgical procedures. Second, the effect of fiducial weighting on corresponding points to overcome anisotropic localization error based on maximum likelihood approach is evaluated. Third, an experimental model for fiducial localization error (FLE) is derived to obtain the weights. At the end, an error zone was obtained for each marker to indicate the possible acceptable deviation from the marker's exact location in practice. This study was performed in vitro on a spine phantom. Optimal landmark selection led to a 30% reduction in TRE. In addition, optimum weighting of the fiducials in an FLE model that incorporates anisotropic localization error in the registration algorithm led to a 28% reduction in the TRE. Landmark configuration, transformation parameters, and fiducial localization error are factors that significantly affect the total TRE. These factors should be optimized to minimize the TRE. Both the optimum configuration of landmarks and the anisotropic weighing of fiducials have significant impact on the registration accuracy for IGSS.