Principles for an Implementation of a Complete CT Reconstruction Tool Chain for Arbitrary Sized Data Sets and Its GPU Optimization

This article describes the implementation of an efficient and fast in-house computed tomography (CT) reconstruction framework. The implementation principles of this cone-beam CT reconstruction tool chain are described here. The article mainly covers the core part of CT reconstruction, the filtered backprojection and its speed up on GPU hardware. Methods and implementations of tools for artifact reduction such as ring artifacts, beam hardening, algorithms for the center of rotation determination and tilted rotation axis correction are presented. The framework allows the reconstruction of CT images of arbitrary data size. Strategies on data splitting and GPU kernel optimization techniques applied for the backprojection process are illustrated by a few examples.

Workflow and performance of intraoperative CT, cone-beam CT, and robotic cone-beam CT for spinal navigation in 503 consecutive patients

OBJECTIVE A direct comparison of intraoperative CT (iCT), cone-beam CT (CBCT), and robotic cone-beam CT (rCBCT) has been necessary to identify the ideal imaging solution for each individual user’s need. Herein, the authors sought to analyze workflow, handling, and performance of iCT, CBCT, and rCBCT imaging for navigated pedicle screw instrumentation across the entire spine performed within the same surgical environment by the same group of surgeons. METHODS Between 2014 and 2018, 503 consecutive patients received 2673 navigated pedicle screws using iCT (n = 1219), CBCT (n = 646), or rCBCT (n = 808) imaging during the first 24 months after the acquisition of each modality. Clinical and demographic data, workflow, handling, and screw assessment and accuracy were analyzed. RESULTS Intraoperative CT showed image quality and workflow advantages for cervicothoracic cases, obese patients, and long-segment instrumentation, whereas CBCT and rCBCT offered independent handling, around-the-clock availability, and the option of performing 2D fluoroscopy. All modalities permitted reliable intraoperative screw assessment. Navigated screw revision was possible with each modality and yielded final accuracy rates > 92% in all groups (iCT 96.2% vs CBCT 92.3%, p < 0.001) without a difference in the accuracy of cervical pedicle screw placement or the rate of secondary screw revision surgeries. CONCLUSIONS Continuous training and an individual setup of iCT, CBCT, and rCBCT has been shown to permit safe and precise navigated posterior instrumentation across the entire spine with reliable screw assessment and the option of immediate revision. The perceived higher image quality and larger scan area of iCT should be weighed against the around-the-clock availability of CBCT and rCBCT technology with the option of single-handed robotic image acquisition.

Evaluation of Anatomoradiological Findings on Trigeminal Neuralgia Patients Using Computed Tomography and Cone-Beam Computed Tomography

The study aimed to establish and evaluate anatomoradiological landmarks in trigeminal neuralgia patients using computed tomography (CT) and cone-beam CT. CT images of 40 trigeminal neuralgia (TN) and 40 healthy individuals were retrospectively analyzed and enrolled in the study. The width and length of the foramen rotundum (FR), foramen ovale (FO), foramen supraorbitale, and infraorbitale were measured. The distances between these foramen, between these foramen to the median plane, and between the superior orbital fissure, FO, and FR to clinoid processes were also measured bilaterally. Variations were evaluated according to groups. Significant differences were found for width and length of the foramen ovale, length of the foramen supraorbitale, and infraorbitale between TN and control subjects (p < 0.05). On both sides, FO gets narrower and the length of the infraorbital and supraorbital foramen shortens in the TN group. In most of the control patients, the plane which passes through the infraorbital and supraorbital foramen intersects with impression trigeminale; 70% on the right-side, and 67% in the left-side TN groups. This plane does not intersect with impression trigeminale and deviates in certain degrees. The determination of specific landmarks allows customization to individual patient anatomy and may help the surgeon achieve a more selective effect with a variety of percutaneous procedures in trigeminal neuralgia patients.

Identification of Carotid Artery Microstructure and Plaque Rupture Using C-Arm Cone-Beam CT: A Case Report

C-arm cone-beam computed tomography (CBCT) offers a high imaging resolution with a wide range of contrast to visualize vessels, soft tissue, and bone. We report the usefulness of CBCT in observing neovascularization, microcalcification, and plaque rupture. A 56-year-old man presented with vertigo and complain of an unsteady gait for 5 months. Catheter angiography demonstrated right severe carotid stenosis with irregular filling defect, which on high-resolution MRI showed vessel wall enhancement. The CBCT showed high density structures and linear contrast enhancement from the vascular lumen to the plaque, related to microstructure and plaque rupture. Carotid endarterectomy was performed, and histopathology confirmed that the high-density areas represented neovascularization and microcalcification, with linear enhancement representing plaque rupture. This is the first report showing that microcalcifications and plaque rupture can be identified by CBCT. Thus, CBCT can be used as a promising supplement to current imaging modalities to evaluate plaque components more accurately.

Low-Dose Cone-Beam CT Iterative Reconstruction via Total Variation and Gradient Total Variation

The compressed sensing (CS) technique has been utilized to reconstruct Cone-beam computed tomography (CBCT) images via limited projection from under-sampled measurements. However, the condition of limited projection is an ill-posed problem. Since the CBCT image itself doesn’t have sparse features, the total variation (TV) transform has been widely adopted in CBCT reconstruction. This method, which penalizes the weight of each voxel at a constant rate regardless of different spatial gradient, may not recover qualified CBCT images from ill-posed projection data. This work presents a new strategy to deal with the deficits stated above by utilizing non-uniform weighting penalization in CBCT reconstruction. The proposed new strategy combines TV and gradient total variation (GTV) for reconstruction in a hybrid weighting penalization way, where the total variation is penalized by the gradient total variation in advance. The proposed penalty not only retains the benefits of TV, including artifact and noise suppression, but also maintains the structures in regions with gradual gradient intensity transition more effectively. This study tested the proposed method by under-sampled projections of 2 objects and 2 experiments (2 digital phantom). We assessed its performance against the OS-SART method, FDK method, conventional TV method and TV+GTV method in the tissue contrast, reconstruction accuracy, and imaging resolution by comparing the root mean squared error (RMSE), the correlation coefficient (CC), the structural similarity (SSIM), and profiles intensity of the reconstructed images. The proposed method produced the reconstructed image with the lowest RMSEs and the highest CCs and SSIMs for each experiment.