Context-Aware, Reference-Free Local Motion Metric for CBCT Deformable Motion Compensation

Purpose: Cone-beam CT is a common approach for 3D guidance in interventional radiology (IR), but its long image acquisition time poses a limit to image quality due to complex deformable motion of soft-tissue structures. Multi-region autofocus optimization showed successful compensation of deformable motion with gains in image quality. However, the use of sub-optimal conventional autofocus metrics, accompanied by the high-dimensionality and non-convexity of the optimization problem result in challenged convergence. This work presents a learning-based approach to the design of autofocus metrics tailored to the specific problem of motion compensation in soft-tissue CBCT. Methods: A deep convolutional neural network (CNN) was used to automatically extract image features quantifying the local motion amplitude and principal direction in volumetric regions of interest (128 x 128x 128 voxels) of a motion contaminated volume. The estimated motion amplitude is then used as the core component of the cost function of the deformable autofocus method, complemented by a regularization term favoring similar motion for regions close in space. The network consists of a siamese arrangement of three branches acting on the three central orthogonal planes of the volumetric ROI. The network was trained with simulated data generated from publicly available CT datasets, including deformable motion fields from which volumetric ROIs with locally rigid motion were extracted. The performance of motion amplitude estimation and of the final CNN-based deformable autofocus were assessed on synthetic CBCT data generated similarly to the training dataset and featuring deformable motion fields with 1 to 5 components with random direction and random amplitude ranging from 0 mm to 50 mm. Results: Predicted local motion amplitude showed good agreement with the true values, showing a liner relationship (R2 = 0.96, slope 0.95), and slight underestimation of the motion amplitude. Absolute errors in total motion amplitude and individual components remained < 2 mm throughout the explored range. Relative errors were higher for low amplitude motion, pointing to need of a larger training cohort, more focused on low motion amplitude scenarios. Motion compensation with the learning-based metric showed successful compensation of motion artifacts in single ROI environments, with 40% reduction in median RMSE using the motion static image as reference. Deformable motion compensation resulted in better visualization of soft-tissue structures, and overall sharper image details, with slight residual motion artifacts for regions combining moderate motion amplitude with complex anatomical structures. Conclusion: Deformable motion compensation with automatically learned autofocus metrics was proved feasible, opening the way to the design of metrics with potential for more reliable performance and easier optimization than conventional autofocus metrics.

To evaluate the influence of cone beam CT (CBCT) volume orientation on alveolar bone measurements for dental implant planning using CBCT in patients with different facial profiles. 74 CBCT volumes were selected from a database and classified according to the facial profile of the patient. Height and width measurements of the alveolar bone were carried out with the volume of the mandible in two different orientations: occlusal plane and mandibular base parallel to the horizontal plane. The data were subjected to the mixed model methodology for repeated measures, through the PROC MIXED procedure. Multiple comparisons were performed by Tukey Kramer test (α = 0.05). Alveolar bone width was significantly greater when the CBCT volume was oriented with the mandibular base parallel to the horizontal plane, for all facial profiles (p ≤ 0.05). Alveolar bone height was significantly higher (p ≤ 0.05) for dolichofacial individuals when compared to that of mesofacial and brachyfacial individuals, who did not differ significantly between each other (p > 0.05), regardless of the CBCT volume orientations used in this study. CBCT-based alveolar bone width is increased when the image volume is oriented with the mandibular base parallel to the horizontal plane and dolichofacial individuals present greater alveolar bone height.

The orientation of the head during the acquisition of cone beam computed tomography (CBCT) is crucial for accurate cephalometric measurements. However, involuntary head movements during the scan can result in misaligned images. This study presents a method to correct the natural head position (NHP) in CBCT skull images after acquisition. Initially, slices were reorganized, and volumes were oriented in RAI coordinates. Next, the Yaw angle in the axial plane was estimated using the fronto-zygomatic sutures (FZS). The Roll angle in the coronal plane was estimated using the centroids of the ocular cavities. Finally, the Pitch angle in the sagittal plane was measured (by the Sella Turcica (S) and Nasion (N) as reference points) and adjusted to reference values of 3.8° for males and 4.1° for females. These calculated angles are used to adjust the CBCT volumes, aligning the images with the NHP.•The method corrects misalignment in CBCT skull images due to involuntary head movements, ensuring accurate cephalometric measurements.•Reference points like fronto-zygomatic sutures, ocular cavity centroids, and the Sella-Nasion line were used to adjust head position in all anatomical planes.•The approach aligns CBCT volumes to the NHP after capture, reducing errors in cephalometric analysis and enhancing diagnostic accuracy.

Purpose: Complex, involuntary, non-periodic, deformable motion presents a confounding factor to cone-beam CT (CBCT) image quality due to long (>10 s) scan times. We report and demonstrate an image-based deformable motion compensation method for CBCT, including phantom, cadaver, and animal studies as precursors to clinical studies. Methods: The method corrects deformable motion in CBCT scan data by solving for a motion vector field (MVF) that optimizes a sharpness criterion in the 3D image (viz., gradient entropy). MVFs are estimated by interpolating M locally rigid motion trajectories across N temporal nodes and are incorporated in a modified 3D filtered backprojection approach. The method was evaluated in a cervical spine phantom under flexion, and a cadaver undergoing variable magnitude of complex motion while imaged on a mobile C-arm (Cios Spin 3D, Siemens Healthineers, Forchheim, Germany). Further assessment was performed on a preclinical animal study using a clinical fixed-room C-arm (Artis Zee, Siemens Healthineers, Forchheim, Germany). Results: In phantom studies, the algorithm resolved visibility of cervical vertebrae under situations of strong flexion, reducing the root-mean-square error by 60% when compared to a motion-free reference. Reduced motion artifacts (blurring, streaks, and loss of soft-tissue edges) were evident in abdominal CBCT of a cadaver imaged during small, medium, and large motion-induced deformation. The animal study demonstrated reduction of streaks from complex motion of bowel gas during the scan. Conclusion: Overall, the studies demonstrate the robustness of the algorithm to a broad range of motion amplitudes, frequencies, data sources (i.e., mobile or fixed-room C-arms) and other confounding factors in real (not simulated) experimental data (e.g., truncation and scatter). These preclinical studies successfully demonstrate reduction of motion artifacts in CBCT and support translation of the method to clinical studies in interventional body radiology.

Objective. Dynamic cone-beam computed tomography (CBCT) can capture high-spatial-resolution, time-varying images for motion monitoring, patient setup, and adaptive planning of radiotherapy. However, dynamic CBCT reconstruction is an extremely ill-posed spatiotemporal inverse problem, as each CBCT volume in the dynamic sequence is only captured by one or a few x-ray projections, due to the slow gantry rotation speed and the fast anatomical motion (e.g. breathing). Approach. We developed a machine learning-based technique, prior-model-free spatiotemporal implicit neural representation (PMF-STINR), to reconstruct dynamic CBCTs from sequentially acquired x-ray projections. PMF-STINR employs a joint image reconstruction and registration approach to address the under-sampling challenge, enabling dynamic CBCT reconstruction from singular x-ray projections. Specifically, PMF-STINR uses spatial implicit neural representations to reconstruct a reference CBCT volume, and it applies temporal INR to represent the intra-scan dynamic motion of the reference CBCT to yield dynamic CBCTs. PMF-STINR couples the temporal INR with a learning-based B-spline motion model to capture time-varying deformable motion during the reconstruction. Compared with the previous methods, the spatial INR, the temporal INR, and the B-spline model of PMF-STINR are all learned on the fly during reconstruction in a one-shot fashion, without using any patient-specific prior knowledge or motion sorting/binning. Main results. PMF-STINR was evaluated via digital phantom simulations, physical phantom measurements, and a multi-institutional patient dataset featuring various imaging protocols (half-fan/full-fan, full sampling/sparse sampling, different energy and mAs settings, etc). The results showed that the one-shot learning-based PMF-STINR can accurately and robustly reconstruct dynamic CBCTs and capture highly irregular motion with high temporal (∼ 0.1 s) resolution and sub-millimeter accuracy. Significance. PMF-STINR can reconstruct dynamic CBCTs and solve the intra-scan motion from conventional 3D CBCT scans without using any prior anatomical/motion model or motion sorting/binning. It can be a promising tool for motion management by offering richer motion information than traditional 4D-CBCTs.

Atheromas can be detected incidentally in routine dental cone beam computed tomography (CBCT) images. This study aims to assess prevalence and risk factors associated with these vascular lesions. The maxillofacial CBCTs of 458 subjects were evaluated and divided into 4 groups based on the presence of calcified atheroma: subjects with no calcified atheroma, subjects with intracranial calcified atheroma (ICA), subjects with extracranial calcified atheroma (ECA), and subjects exhibiting combined lesions. Age, sex, medical conditions, family history, and size were documented. Analysis of variance followed by a multiple comparison test was used for data satisfying parametric test assumptions. Chi-squared tests were used to assess categorical data. The Spearman Rho test was used to assess the correlation between the incidence of calcified atheroma and subjects' medical condition. Of the 458 CBCTs evaluated, 29.90% presented with calcified atheroma. Calcified atheroma prevalence was significantly higher in older patients versus younger patients (p=0.004) and in males compared to females (p=0.004). Males were more likely to have the combination of ICA and ECA, whereas females were more likely to have ICA alone (p≤0.040). Patients with calcified atheroma were significantly more likely to have a history of hyperlipidemia (p=0.001), hypertension (p=0.001), and myocardial infarction/coronary artery diseases (p=0.001). Overall, patients exhibiting both intracranial and extracranial lesions were more likely to have cardiovascular risk factors (p=0.001). Incidentally detected calcified atheromas in CBCTs are common. Subjects with combined atheroma lesions are at higher risk for cardiovascular disease. The diagnosis of incidental calcified atheromas in CBCT's warrants early referral to medical specialists, especially if there is no medical history of existing cardiovascular disease.

Cone Beam Computed Tomography (CBCT) has gained a broad acceptance in dentomaxillofacial imaging. Computed Tomography (CT) is another imaging modality for diagnosis and preoperative assessments of the head and neck region. Considering the increased radiation exposure and high cost of CT, this study sought to subjectively assess the image quality of CBCT and Multi Slice CT (MSCT). A dry human mandible was scanned by five CBCT systems (New Tom 3G, Scanora, CRANEX 3D, Promax and Galileos) and one MSCT system. Three independent oral and maxillofacial radiologists reviewed the CBCT and MSCT scans for the quality of 10 landmarks namely mental foramen, trabecular bone, Periodontal Ligament (PDL), dentin, incisive canal, mandibular canal, dental pulp, enamel, lamina dura and cortical bone using a five-point scale. Significant differences were found between MSCT and CBCT and among the five CBCT systems (p<0.05) in visualization of different anatomical structures. A fine structure such as the incisive canal was significantly less visible and more variable among the systems in comparison with other anatomical landmarks such as the mental foramen, mandibular canal, cortical bone, dental pulp, enamel and dentin (p<0.05). The Cranex 3D and Promax systems were superior to MSCT and all other CBCT systems in visualizing anatomical structures. The CBCT image quality was superior to that of MSCT even though some variability existed among different CBCT systems in visualizing fine structures. Considering the low radiation dose and high resolution, CBCT may be beneficial for dentomaxillofacial imaging.

Feasibility of MR-Generated Synthetic CT Images for Planning Stereotactic Radiosurgery of Brain Metastases

Using a single cone-beam computed tomography scan to obtain full occlusal details, with the mandible in centric relation and maximum intercuspation.

PurposeThe aim of this study was to improve the image quality of cone-beam computed tomography (CBCT) mounted on the gantry of a linear accelerator used in radiation therapy based on the image information provided by planning multi-detector CT (MDCT).MethodsMDCT-based shading correction for CBCT and virtual monochromatic CT (VMCT) synthesized using the dual-energy method were performed. In VMCT, the high-energy data were obtained from CBCT, while the low-energy data were obtained from MDCT. An electron density phantom was used to investigate the efficacy of shading correction and VMCT on improving the target detectability, Hounsfield unit (HU) accuracy and variation, which were quantified by calculating the contrast-to-noise ratio (CNR), the percent difference (%Diff) and the standard deviation of the CT numbers for tissue equivalent background material, respectively. Treatment plan studies for a chest phantom were conducted to investigate the effects of image quality improvement on dose planning.ResultsFor the electron density phantom, the mean value of CNR was 17.84, 26.78 and 34.31 in CBCT, shading-corrected CBCT and VMCT, respectively. The mean value of %Diff was 152.67%, 11.93% and 7.66% in CBCT, shading-corrected CBCT and VMCT, respectively. The standard deviation within a uniform background of CBCT, shading-corrected CBCT and VMCT was 85, 23 and 15 HU, respectively. With regards to the chest phantom, the monitor unit (MU) difference between the treatment plan calculated using MDCT and those based on CBCT, shading corrected CBCT and VMCT was 6.32%, 1.05% and 0.94%, respectively.ConclusionsEnhancement of image quality in on-board CBCT can contribute to daily patient setup and adaptive dose delivery, thus enabling higher confidence in patient treatment accuracy in radiation therapy. Based on our results, VMCT has the highest image quality, followed by the shading corrected CBCT and the original CBCT. The research results presented in this study should be able to provide a route to reach a high level of image quality for CBCT imaging in radiation oncology.

Evaluation of deep learning for detecting intraosseous jaw lesions in cone beam computed tomography volumes

Objectives: To evaluate the influence of sharpening enhancement filters in the detection of root fractures using cone-beam computed tomography (CBCT) at different levels of milliamperage (mA) in teeth with and without endodontic filling. Materials and Methods:Eighty-four CBCT volumes acquired at three mA levels (4, 6.3 and 10) of 28 teeth inserted in the dental socket of dry human skull were selected from a previous study.The teeth were divided into four groups according to the presence and absence of root fracture and endodontic filling.Five radiologists evaluated all CBCT volumes for the presence of root fracture with and without the application of the "Sharpen 1x" and "Sharpen 2x" image filters of the OnDemand 3D software.Area under the ROC curve (AUC), sensitivity, specificity, and inter-and intra-observer concordance were calculated and compared (α=0.05).Results: Enhancement filters did not lead to significant differences in the values of AUC, sensitivity, and specificity at the three mA levels tested (p>0.05),regardless of the presence of endodontic filling (p>0.05).However, the significant reduction of AUC observed in CBCT volumes at 4 mA without filter (p<0.05)ceased to exist after the application of filters (p>0.05).Sensitivity and specificity ranged from low and moderate. Conclusions:The use of sharpening enhancement filters can be recommended in CBCT volumes at 4 mA for root fracture detection for leading to the same performance as those at 6.3 and 10 mA.The presence of endodontic filling material did not influence the action of filters in the diagnosis of root fracture.Clinical relevance: Digital enhancement filters can contribute to the diagnosis of root fracture in CBCT volumes acquired with reduced radiation dose.

Cone beam computed tomography (CBCT) plays an important role in many medical fields nowadays. Unfortunately, the potential of this imaging modality is hampered by lower image quality compared to the conventional CT, and producing accurate reconstructions remains challenging. A lot of recent research has been directed towards reconstruction methods relying on deep learning, which have shown great promise for various imaging modalities. However, practical application of deep learning to CBCT reconstruction is complicated by several issues, such as exceedingly high memory costs of deep learning methods when working with fully 3D data. Additionally, deep learning methods proposed in the literature are often trained and evaluated only on data from a specific region of interest, thus raising concerns about possible lack of generalization to other regions. In this work, we aim to address these limitations and propose LIRE: a learned invertible primal-dual iterative scheme for CBCTreconstruction. LIRE is a learned invertible primal-dual iterative scheme for CBCT reconstruction, wherein we employ a U-Net architecture in each primal block and a residual convolutional neural network (CNN) architecture in each dual block. Memory requirements of the network are substantially reduced while preserving its expressive power through a combination of invertible residual primal-dual blocks and patch-wise computations inside each of the blocks during both forward and backward pass. These techniques enable us to train on data with isotropic 2 mm voxel spacing, clinically-relevant projection count and detector panel resolution on current hardware with 24 GBvideo random access memory (VRAM). Two LIRE models for small and for large field-of-view (FoV)setting were trained and validated on a set of 260 + 22 thorax CT scans and tested using a set of 142 thorax CT scans plus an out-of-distribution dataset of 79 head and neck CT scans. For both settings, our method surpasses the classical methods and the deep learning baselines on both test sets. On the thorax CT set, our method achieves peak signal-to-noise ratio (PSNR) of 33.84 ± 2.28 for the small FoV setting and 35.14 ± 2.69 for the large FoV setting; U-Net baseline achieves PSNR of 33.08 ± 1.75 and 34.29 ± 2.71 respectively. On the head and neck CT set, our method achieves PSNR of 39.35 ± 1.75 for the small FoV setting and 41.21 ± 1.41 for the large FoV setting; U-Net baseline achieves PSNR of 33.08 ± 1.75 and 34.29 ± 2.71 respectively. Additionally, we demonstrate that LIRE can be finetuned to reconstruct high-resolution CBCT data with the same geometry but 1 mm voxel spacing and higher detector panel resolution, where it outperforms the U-Net baseline aswell. Learned invertible primal-dual schemes with additional memory optimizations can be trained to reconstruct CBCT volumes directly from the projection data with clinically-relevant geometry and resolution. Such methods can offer better reconstruction quality and generalization compared to classical deep learning baselines.

To evaluate how additional information from Cone Beam CT (CBCT) impacts on periapical assessment and treatment planning based on clinical examination and periapical radiographs (PR) in cases followed up five to eleven years after surgical endodontic retreatment (SER). Patients receiving SER during 2004-2010 were reinvited for follow-up examination including clinical examination, PR, and CBCT. In total, 108 patients (119 teeth) were reinvited, 74 patients (83 teeth) accepted to participate. Three observers initially assessed PR according to the four-scaled, increasing disease severity criteria by Rud etal. (International Journal of Oral Surgery, 1, 1972 and 195) and Molven etal. (International Journal of Oral and Maxillofacial Surgery, 16, and 432): 'Radiographic assessment A'. By including clinical information 'Treatment plan A' was made as follows: 1) no treatment, 2) further observation, 3) SER reoperation (SER-R), or 4) extraction. Hereafter, the CBCT volume was assessed and the information incorporated for 'Radiographic assessment B' followed by 'Treatment plan B'. Agreement between radiographic assessments and between treatment plans was recorded and assessed statistically by Stuart-Maxwell test for marginal homogeneity. Nine teeth had been extracted; thus, the final analysis included 74 teeth (66 patients). The radiographic assessment was changed as a result of the CBCT evaluation in 38 cases (51.4%), of which 35 (47.3%) were to a higher Rud & Molven score, P<0.001. The treatment plan was changed for 18 teeth (24.3%). For 14 teeth (18.9%), the change was from no treatment or further observation to a more invasive treatment plan (SER-R or extraction), P=0.005. The use of CBCT for long-term follow-up after SER led to more cases diagnosed with persisting or recurrent apical periodontitis and hence often to the recommendation of a more invasive treatment modality.

The cone beam computed tomography (CBCT) guided small animal radiation research platform (SARRP) has been developed for focal tumor irradiation, allowing laboratory researchers to test basic biological hypotheses that can modify radiotherapy outcomes in ways that were not feasible previously. CBCT provides excellent bone to soft tissue contrast, but is incapable of differentiating tumors from surrounding soft tissue. Bioluminescence tomography (BLT), in contrast, allows direct visualization of even subpalpable tumors and quantitative evaluation of tumor response. Integration of BLT with CBCT offers complementary image information, with CBCT delineating anatomic structures and BLT differentiating luminescent tumors. This study is to develop a systematic method to calibrate an integrated CBCT and BLT imaging system which can be adopted onboard the SARRP to guide focal tumor irradiation. The integrated imaging system consists of CBCT, diffuse optical tomography (DOT), and BLT. The anatomy acquired from CBCT and optical properties acquired from DOT serve as a priori information for the subsequent BLT reconstruction. Phantoms were designed and procedures were developed to calibrate the CBCT, DOT/BLT, and the entire integrated system. Geometrical calibration was performed to calibrate the CBCT system. Flat field correction was performed to correct the nonuniform response of the optical imaging system. Absolute emittance calibration was performed to convert the camera readout to the emittance at the phantom or animal surface, which enabled the direct reconstruction of the bioluminescence source strength. Phantom and mouse imaging were performed to validate the calibration. All calibration procedures were successfully performed. Both CBCT of a thin wire and a euthanized mouse revealed no spatial artifact, validating the accuracy of the CBCT calibration. The absolute emittance calibration was validated with a 650 nm laser source, resulting in a 3.0% difference between simulated and measured signal. The calibration of the entire system was confirmed through the CBCT and BLT reconstruction of a bioluminescence source placed inside a tissue-simulating optical phantom. Using a spatial region constraint, the source position was reconstructed with less than 1 mm error and the source strength reconstructed with less than 24% error. A practical and systematic method has been developed to calibrate an integrated x-ray and optical tomography imaging system, including the respective CBCT and optical tomography system calibration and the geometrical calibration of the entire system. The method can be modified and adopted to calibrate CBCT and optical tomography systems that are operated independently or hybrid x-ray and optical tomography imaging systems.