Cone-beam CT Projection Images Research Articles

Improving dose calculation accuracy in preclinical radiation experiments using multi-energy element resolved cone-beam CT

Objective. Cone-beam CT (CBCT) in modern pre-clinical small-animal radiation research platforms provides volumetric images for image guidance and experiment planning purposes. In this work, we implemented multi-energy element-resolved (MEER) CBCT using three scans with different kVps on a SmART platform (Precision x-ray Inc.) to determine images of relative electron density (rED) and elemental composition (EC) that are needed for Monte Carlo-based radiation dose calculation. Approach. We performed comprehensive calibration tasks to achieve sufficient accuracy for this quantitative imaging purpose. For geometry calibration, we scanned a ball bearing phantom and used an analytical method together with an optimization approach to derive gantry angle specific geometry parameters. Intensity calibration and correction included the corrections for detector lag, glare, and beam hardening. The corrected CBCT projection images acquired at 30, 40, and 60 kVp in multiple scans were used to reconstruct CBCT images using the Feldkamp–Davis–Kress reconstruction algorithm. After that, an optimization problem was solved to determine images of rED and EC. We demonstrated the effectiveness of our CBCT calibration steps by showing improvements in image quality and successful material decomposition in cases with a small animal CT calibration phantom and a plastinated mouse phantom. Main results. It was found that artifacts induced by geometry inaccuracy, detector lag, glare, and beam hardening were visually reduced. CT number mean errors were reduced from 19% to 5%. In the CT calibration phantom case, median errors in H, O, and Ca fractions for all the inserts were below 1%, 2%, and 4% respectively, and median error in rED was less than 5%. Compared to the standard approach deriving material type and rED via CT number conversion, our approach improved Monte Carlo simulation-based dose calculation accuracy in bone regions. Mean dose error was reduced from 47.5% to 10.9%. Significance. The MEER-CBCT implemented on an existing CBCT system of a small animal irradiation platform achieved accurate material decomposition and significantly improved Monte Carlo dose calculation accuracy.

Automatic diaphragm segmentation for real-time lung tumor tracking on cone-beam CT projections: a convolutional neural network approach

Purpose: To automatically segment the diaphragm on individual lung cone-beam CT projection images, to enable real-time tracking of lung tumors using kilovoltage imaging. Methods: The deep neural network Mask R-CNN was trained on 3500 raw cone-beam CT projection images from 10 lung cancer patients, with the diaphragm manually segmented on each image used as a ground truth label. Ground-truth breathing traces were extracted from each patient for both diaphragm hemispheres, and apex positions were compared against the predicted output of the neural network. Ten-fold cross-validation was used to evaluate the segmentation accuracy. Results: The mean diaphragm apex prediction error was 4.4 mm. The mean percentage of projection images for which a successful prediction could me made was 87.3%. Prediction accuracy at some lateral gantry angles was worse due to overlap between diaphragm hemispheres, and the increased amount of fatty tissue. Conclusions: The neural network was able to track the diaphragm apex position successfully. This allows accurate assessment of the breathing phase, which can be used to estimate the position of the lung tumor in real time.

Effect of anatomical noise on the detectability of cone beam CT images with different slice direction, slice thickness, and volume glandular fraction.

We investigate the effect of anatomical noise on the detectability of cone beam CT (CBCT) images with different slice directions, slice thicknesses, and volume glandular fractions (VGFs). Anatomical noise is generated using a power law spectrum of breast anatomy, and spherical objects with diameters from 1mm to 11mm are used as breast masses. CBCT projection images are simulated and reconstructed using the FDK algorithm. A channelized Hotelling observer (CHO) with Laguerre-Gauss (LG) channels is used to evaluate detectability for the signal-known-exactly (SKE) binary detection task. Detectability is calculated for various slice thicknesses in the transverse and longitudinal planes for 15%, 30% and 60% VGFs. The optimal slice thicknesses that maximize the detectability of the objects are determined. The results show that the β value increases as the slice thickness increases, but that thicker slices yield higher detectability in the transverse and longitudinal planes, except for the case of a 1mm diameter spherical object. It is also shown that the longitudinal plane with a 0.1mm slice thickness provides higher detectability than the transverse plane, despite its higher β value. With optimal slice thicknesses, the longitudinal plane exhibits better detectability for all VGFs and spherical objects.

Reconstruction of implanted marker trajectories from cone-beam CT projection images using interdimensional correlation modeling.

Cone-beam CT (CBCT) is a widely used imaging modality for image-guided radiotherapy. Most vendors provide CBCT systems that are mounted on a linac gantry. Thus, CBCT can be used to estimate the actual 3-dimensional (3D) position of moving respiratory targets in the thoracic/abdominal region using 2D projection images. The authors have developed a method for estimating the 3D trajectory of respiratory-induced target motion from CBCT projection images using interdimensional correlation modeling. Because the superior-inferior (SI) motion of a target can be easily analyzed on projection images of a gantry-mounted CBCT system, the authors investigated the interdimensional correlation of the SI motion with left-right and anterior-posterior (AP) movements while the gantry is rotating. A simple linear model and a state-augmented model were implemented and applied to the interdimensional correlation analysis, and their performance was compared. The parameters of the interdimensional correlation models were determined by least-square estimation of the 2D error between the actual and estimated projected target position. The method was validated using 160 3D tumor trajectories from 46 thoracic/abdominal cancer patients obtained during CyberKnife treatment. The authors' simulations assumed two application scenarios: (1) retrospective estimation for the purpose of moving tumor setup used just after volumetric matching with CBCT; and (2) on-the-fly estimation for the purpose of real-time target position estimation during gating or tracking delivery, either for full-rotation volumetric-modulated arc therapy (VMAT) in 60 s or a stationary six-field intensity-modulated radiation therapy (IMRT) with a beam delivery time of 20 s. For the retrospective CBCT simulations, the mean 3D root-mean-square error (RMSE) for all 4893 trajectory segments was 0.41 mm (simple linear model) and 0.35 mm (state-augmented model). In the on-the-fly simulations, prior projections over more than 60° appear to be necessary for reliable estimations. The mean 3D RMSE during beam delivery after the simple linear model had established with a prior 90° projection data was 0.42 mm for VMAT and 0.45 mm for IMRT. The proposed method does not require any internal/external correlation or statistical modeling to estimate the target trajectory and can be used for both retrospective image-guided radiotherapy with CBCT projection images and real-time target position monitoring for respiratory gating or tracking.

Target-Matching Accuracy in Stereotactic Body Radiation Therapy of Lung Cancer: An Investigation Based on Four-Dimensional Digital Human Phantom

Aim: To investigate the effect of irregular respiratory on target-matching accuracy in lung stereotactic-body radiation therapy (SBRT). Methods: The four-dimensional extended cardiac torso (4D-XCAT) phantom was used to generate 4D computed tomography (4DCT) and cone-beam CT (CBCT) images. Images were generated for 1 regular and 10 irregular trajectories, and for 3 tumor size groups (1, 2, and 3 cm). Image registrations between CBCT and average intensity projection (AIP) images of 4DCT were performed based on target volume matching. Error of registration was determined as the difference between manual CBCT-to-AIP registration and known registration between the two. In addition, internal target volumes (ITVs) were contoured on and compared between AIP, maximum intensity projection (MIP), and CBCT. Results: Small inter-observer variations of registration were found: 0.2, 0.3, and 0.7 mm in the medial-lateral (ML), anterior-posterior (AP), and superior-inferior (SI) direction, respectively. Small errors of registration (median ≤ 0.5 mm) were found in all three directions for the regular respiratory profile. For the irregular profiles and on average of all tumor size groups, the median ± standard deviation (SD) errors of registration were 0.5 ± 0.3 mm, 0.4 ± 0.5 mm, and 1.9 ± 1.6 mm in the ML, AP, and SI direction, respectively. Significant differences were found between MIP and CBCT-based ITV volumes for the three tumor size groups (P = 0.011, 0.010, and 0.006 for 1, 2, and 3 cm tumor size group, respectively). Conclusion: Irregular breathing can induce error in CBCT-to-AIP registration in lung SBRT. This error increases as the breathing irregularity increases.

Digital tomosynthesis for verifying spine position during radiotherapy: a phantom study

Monitoring the stability of patient position is essential during high-precision radiotherapy such as spine stereotactic body radiotherapy (SBRT). We evaluated the combination of digital tomosynthesis (DTS) and triangulation for spine position detection, using non-clinical DTS software and an anthropomorphic pelvic phantom that includes a bone-like spine structure. Kilovoltage cone beam CT projection images over 2–16° gantry rotation were used to generate single slice DTS images. Each DTS slice was registered to a digitally reconstructed DTS derived from the planning CT scan to determine 2D shifts between actual phantom and treatment plan position. Two or more DTS registrations, central axes 4–22° apart, were triangulated to determine the 3D phantom position. Using sequentially generated DTS images, the phantom position can be updated every degree with a small latency of DTS and triangulation angle. The precision of position determination was investigated as function of DTS and triangulation angle. To mimic the scenario of spine SBRT, the effect on the standard deviation of megavoltage radiation delivery during kV image acquisition was tested. In addition, the ability of the system to detect different types of movement was investigated for a variety of small sudden and gradual movements during kV image acquisition.

PD-0044: Respiratory-induced motion of pancreatic cancer: time trends during the course of radiotherapy

Purpose/Objective: Irradiation of pancreatic tumors requires large margins to account for respiratory-induced intrafractional tumor motion. These margins are derived from a single pre-treatment 4DCT scan. This respiratory-induced intrafractional motion can be visualized using implanted fiducial markers. The purpose of the current study is to quantify the respiratoryinduced motion of pancreatic cancer, both on the 4DCT scan as well as on daily cone beam CT (CBCT) scans during a 5 week course of radiation to reveal possible trends during the course of treatment. Materials and Methods: Eight patients (4 male / 4 female) with borderline resectable pancreatic cancer were treated with 25 x 2Gy plus weekly gemcitabine 300 mg/m. All patients received 2 to 4 fiducial markers by endoscopic ultrasound-guided implantation. The amplitude of respiratory-induced motion was measured on the pretreatment 4DCT scan and also on all available daily CBCT scans. Phase reconstructions of the CBCT scans were made by an automated selection of the 2D projection images corresponding to the peak inhale and exhale phases. These reconstructions were matched to the (reference) planning CT. The amplitudes (differences between the 2 extreme positions) were analyzed by matching on the markers (in three directions) and on the diaphragm (superior-inferior (SI) direction only). Rotations and deformations of the tissue were not analyzed. The amplitudes were linearly fit as a function of time (i.e. treatment day). The fits were tested for significance using a Student’s t-test and statistical significance was assumed at p≤0.05. Results: The amplitude of intrafractional motion of the markers was greatest in the SI direction (mean 5.9mm, SD: 2.7mm on CBCT). Differences up to 5.1mm (range: 0.3 5.1 mm) were found between the amplitude on the 4DCT scan and the mean amplitude of all CBCT scans of a single patient. The analysis of daily respiratory-induced motion of the markers revealed significant trends in the amplitude as a function of time (table and figure). The diaphragm did not always show the same trend as the markers. For the markers neither the amplitudes nor the trends were significant in the other 2 directions (data not shown). Conclusions: The 4DCT used for treatment planning is often not representative of the respiratory-induced motion amplitude during treatment. The CBCT projection images showed significant time trends in the amplitude of respiratory-induced motion during a five week course of radiation. These time trends in the amplitudes indicate the usefulness of adaptive treatment procedures. PD-0045 Time-resolved dose reconstruction of VMAT delivery to moving targets with and without dynamic MLC tracking T. Ravkilde, P.J. Keall, C. Grau, M. Hoyer, P.R. Poulsen Aarhus University Hospital, Department of Oncology, Aarhus, Denmark University of Sydney, Sydney Medical School, Sydney, Australia

SU-F-BRCD-04: Efficient Scatter Distribution Estimation and Correction in CBCT Using Concurrent Monte Carlo Fitting.

X-ray scatter is a known source of image artifacts, contrast reduction, and CT number inaccuracy in cone-beam CT (CBCT). We present and demonstrate the performance of a novel scatter correction method based on an algorithm that simultaneously combines multiple Monte Carlo (MC) CBCT scatter simulations through the use of a fitting function. The scatter estimation system consists of concurrently run MC CBCT scatter projection simulations that are a sub-sample of the projection angles used in the reconstruction projection set, P, to be corrected. The photons generated by each MC simulation are simultaneously aggregated in an algorithm which computes the scatter detector response, S(i,j,k), for each down-sampled projection location (i,j) and angle (k). S(i,j,k) is fit to a function, FS, and if FS is determined to have a specified goodness of fit value the simulations are terminated. FS is subtracted from P which is subsequently used to create a scatter corrected reconstruction. The scatter correction method was applied to simulated phantoms using a frequency limited sum of sines and cosines as the fitting function. Image quality in the corrected reconstruction was evaluated using metrics looking at contrast, noise, and artifact reduction. Fitting the scatter distribution to a limited sum of sine and cosine functions, using a low-pass filtered Fast Fourier transform, provides a computationally efficient and accurate fit. Scatter distribution estimates for a 360 image projection set were computed in under one minute. The scatter correction algorithm increased the contrast-to-noise ratio by 46%, reduced the shading artifact by 87%, and decreased the skin line artifact by 79% in reconstructions of a simulated pelvis phantom. The algorithm provides an efficient method for estimating and removing the scatter distribution in from CBCT projection images. The results on simulated data show a significant increase in image quality.

Automatic tracking of implanted fiducial markers in cone beam CT projection images

This paper describes a novel method for simultaneous intrafraction tracking of multiple fiducial markers. Although the proposed method is generic and can be adopted for a number of applications including fluoroscopy based patient position monitoring and gated radiotherapy, the tracking results presented in this paper are specific to tracking fiducial markers in a sequence of cone beam CT projection images. The proposed method is accurate and robust thanks to utilizing the mean shift and random sampling principles, respectively. The performance of the proposed method was evaluated with qualitative and quantitative methods, using data from two pancreatic and one prostate cancer patients and a moving phantom. The ground truth, for quantitative evaluation, was calculated based on manual tracking preformed by three observers. The average dispersion of marker position error calculated from the tracking results for pancreas data (six markers tracked over 640 frames, 3840 marker identifications) was 0.25 mm (at iscoenter), compared with an average dispersion for the manual ground truth estimated at 0.22 mm. For prostate data (three markers tracked over 366 frames, 1098 marker identifications), the average error was 0.34 mm. The estimated tracking error in the pancreas data was < 1 mm (2 pixels) in 97.6% of cases where nearby image clutter was detected and in 100.0% of cases with no nearby image clutter. The proposed method has accuracy comparable to that of manual tracking and, in combination with the proposed batch postprocessing, superior robustness. Marker tracking in cone beam CT (CBCT) projections is useful for a variety of purposes, such as providing data for assessment of intrafraction motion, target tracking during rotational treatment delivery, motion correction of CBCT, and phase sorting for 4D CBCT.

SU‐E‐I‐116: Spatial Frequency Spectrum of the X‐Ray Scatter Distribution in CBCT

Purpose: X‐ray scatter causes significant diminishing effects to image quality in cone‐beam CT (CBCT) such as: image artifacts, loss in contrast‐ to‐noise ratio, and CT number inaccuracy. The purpose of this study is to examine the spatial frequency content of the scatter distribution in CBCT projection images using the Fourier transform. This work forms the basis for accelerated Monte Carlo scatter estimation. Methods and Materials: A previously validated CBCT MC system is used to generate the resulting scatter distribution for various imaging sites (head and neck, abdomen, and pelvis) and projection angles using a voxelized anthropomorphic phantoms. The spatial frequency is examined by taking the fast Fourier transform (FFT) of the simulated scatter distributions. The effects of using a bowtie filter on the spatial frequencies of the scatter distribution is also examined. Results: The spatial frequency of the scatter distribution, in the horizontal and vertical direction, is largely contained in the lower frequencies of the Fourier transform (&lt;0.1 1/cm). The angular component of the spatial frequency of scatter distribution shows two strong periodic signals with a period of pi and pi/2. The use of the bowtie filter has an effect of decreasing the higher frequency content of the horizontal component of the spatial frequency. Conclusions: The computational time of estimating the scatter distribution using MC can be reduced by simulating fewer photons and using a low‐pass filter. Periodic signals in the projection angle component of the scatter distribution indicate that a significant reduction in number of projection angles simulated to accurately estimate the scatter distribution is also possible.

TH-E-220-10: Cine Cone Beam CT for Image Guided Lung Cancer Radiotherapy

Purpose: Cone beam CT(CBCT) has been widely utilized for localizing daily tumor position and patient positioning in lungcancerradiotherapy. However, due to the respiratory motion during the CBCT scan, the tumorimage is blurred and may introduce inaccuracy in treatment delivery. To solve this problem, and to obtain accurate daily lungtumor motion information, in this work we develop a cine CBCT technology, which can reconstruct a series of 3D CBCTimages from a regular CBCT scan. Methods: Provided a set of volumetric images of a patient at N breathing phases obtained by 4DCT as training data, a deformable image registration is first carried out between a reference phase and the other N‐1 phases. As a result, N‐1 deformation vector fields (DVFs) are generated. These DVFs are represented by a few eigenvectors and coefficients obtained from principal component analysis (PCA). New DVFs relative to reference images as a function of space and time can be generated by varying those PCA coefficients, resulting in new volumetric images. The volumetric reconstruction from a CBCT projection is realized by optimizing the PCA coefficients such that the computed projection matches the measured CBCT projection. Results: The cine‐CBCT method was tested on real patient data. 200 CBCT projection images acquired during a regular CBCT scan are used to reconstruct 200 corresponding cine CBCTimages. Each reconstructed cine CBCTimage precisely represent the patient anatomy and tumor geometry at a specific time point during the regular CBCT scan and thus can be utilized to improve the treatment accuracy. Conclusions: Cine‐CBCT, a method to reconstruct volumetric images based on a single CBCT projection, is introduced in this study. Experimental results based on real patient data also demonstrate that volumetric images can be precisely generated. Acknowledgement: The authors would like to acknowledge the support from Varian Medical Systems, Inc.

Using cone-beam CT projection images to estimate the average and complete trajectory of a fiducial marker moving with respiration

Stereotactic body radiotherapy of lung cancer often makes use of a static cone-beam CT (CBCT) image to localize a tumor that moves during the respiratory cycle. In this work, we developed an algorithm to estimate the average and complete trajectory of an implanted fiducial marker from the raw CBCT projection data. After labeling the CBCT projection images based on the breathing phase of the fiducial marker, the average trajectory was determined by backprojecting the fiducial position from images of similar phase. To approximate the complete trajectory, a 3D fiducial position is estimated from its position in each CBCT project image as the point on the source-image ray closest to the average position at the same phase. The algorithm was tested with computer simulations as well as phantom experiments using a gold seed implanted in a programmable phantom capable of variable motion. Simulation testing was done on 120 realistic breathing patterns, half of which contained hysteresis. The average trajectory was reconstructed with an average root mean square (rms) error of less than 0.1 mm in all three directions, and a maximum error of 0.5 mm. The complete trajectory reconstruction had a mean rms error of less than 0.2 mm, with a maximum error of 4.07 mm. The phantom study was conducted using five different respiratory patterns with the amplitudes of 1.3 and 2.6 cm programmed into the motion phantom. These complete trajectories were reconstructed with an average rms error of 0.4 mm. There is motion information present in the raw CBCT dataset that can be exploited with the use of an implanted fiducial marker to sub-millimeter accuracy. This algorithm could ultimately supply the internal motion of a lung tumor at the treatment unit from the same dataset currently used for patient setup.

Reconstruction of a cone‐beam CT image via forward iterative projection matching

To demonstrate the feasibility of reconstructing a cone-beam CT (CBCT) image by deformably altering a prior fan-beam CT (FBCT) image such that it matches the anatomy portrayed in the CBCT projection data set. A prior FBCT image of the patient is assumed to be available as a source image. A CBCT projection data set is obtained and used as a target image set. A parametrized deformation model is applied to the source FBCT image, digitally reconstructed radiographs (DRRs) that emulate the CBCT projection image geometry are calculated and compared to the target CBCT projection data, and the deformation model parameters are adjusted iteratively until the DRRs optimally match the CBCT projection data set. The resulting deformed FBCT image is hypothesized to be an accurate representation of the patient's anatomy imaged by the CBCT system. The process is demonstrated via numerical simulation. A known deformation is applied to a prior FBCT image and used to create a synthetic set of CBCT target projections. The iterative projection matching process is then applied to reconstruct the deformation represented in the synthetic target projections; the reconstructed deformation is then compared to the known deformation. The sensitivity of the process to the number of projections and the DRR/CBCT projection mismatch is explored by systematically adding noise to and perturbing the contrast of the target projections relative to the iterated source DRRs and by reducing the number of projections. When there is no noise or contrast mismatch in the CBCT projection images, a set of 64 projections allows the known deformed CT image to be reconstructed to within a nRMS error of 1% and the known deformation to within a nRMS error of 7%. A CT image nRMS error of less than 4% is maintained at noise levels up to 3% of the mean projection intensity, at which the deformation error is 13%. At 1% noise level, the number of projections can be reduced to 8 while maintaining CT image and deformation errors of less than 4% and 13%, respectively. The method is sensitive to contrast mismatch between the simulated projections and the target projections when the soft-tissue contrast in the projections is low. By using prior knowledge available in a FBCT image, the authors show that a CBCT image can be iteratively reconstructed from a comparatively small number of projection images, thus saving acquisition time and reducing imaging dose. This will enable more frequent daily imaging during radiation therapy. Because the process preserves the CT numbers of the FBCT image, the resulting 3D image intensities will be more accurate than a CBCT image reconstructed via conventional backprojection methods. Reconstruction errors are insensitive to noise at levels beyond what would typically be found in CBCT projection data, but are sensitive to contrast mismatch errors between the CBCT projection data and the DRRs.

SU-GG-J-20: A Trajectory Estimation Method of Moving Target From Cone Beam CT Projection Images

Purpose: The aim of this work was to develop and investigate a method for trajectory estimation of a moving target from cone beam CT projection images. It can be useful for assessment of respiratory tumor motion just before or during treatment, and for developing further respiratory compensation strategies. Method and Materials: Six lungtumor trajectories with range of motion larger than 10 mm, acquired using a CyberKnife system, were used to simulate CBCT acquisition. To reconstruct 3D positions from projected positions, the two nearest angular projection images with the same SI‐projection positions were used for triangulation. With the triangulated positions, a linear correlation was assumed for tumor motion. LR and AP positions were then derived from SI positions using the linear model. The estimated 3D tumor positions were compared with the known positions. To demonstrate the real simulation, CBCT projection images for a seed‐embedded moving phantom were acquired using a Varian OBI system. A 3D motion platform was used to drive the phantom with one of six trajectories. The marker position from each projection image was extracted using Varian RPM‐Fluoro application. Results: The mean estimation errors were within 0.3±1.8 mm while errors of up to 4 mm were observed. In general, higher correlation coefficients with SI‐motion provide better estimation of AP and LR positions. Applying the center offset of OBI imager (xOffset=−2 pixel, yOffset=0 pixel), the remaining uncertainty associated with gantry sag and marker extraction (∼1 pixel) was within ∼0.5 mm. Conclusion: The proposed method provides useful information for tumor motion characteristics such as the mean position, range, and principal direction of motion which can be useful for respiratory compensation treatments. Conflict of Interest: Research supported by Varian Medical Systems.

Measurement of inter and intra fraction organ motion in radiotherapy using cone beam CT projection images

A method is presented for extraction of intra and inter fraction motion of seeds/markers within the patient from cone beam CT (CBCT) projection images. The position of the marker is determined on each projection image and fitted to a function describing the projection of a fixed point onto the imaging panel at different gantry angles. The fitted parameters provide the mean marker position with respect to the isocentre. Differences between the theoretical function and the actual projected marker positions are used to estimate the range of intra fraction motion and the principal motion axis in the transverse plane. The method was validated using CBCT projection images of a static marker at known locations and of a marker moving with known amplitude. The mean difference between actual and measured motion range was less than 1 mm in all directions, although errors of up to 5 mm were observed when large amplitude motion was present in an orthogonal direction. In these cases it was possible to calculate the range of motion magnitudes consistent with the observed marker trajectory. The method was shown to be feasible using clinical CBCT projections of a pancreas cancer patient.

SU‐EE‐A1‐04: Verifying Internal Target Volume Using Cone‐Beam CT for Stereotactic Body Radiotherapy Treatment

Purpose: The internal target volume (ITV) could be determined using 4D CT simulation images and be verified in the treatment room using on‐board cone‐beam CT (CBCT) since the CBCT projection images are acquired over approximately 10 breathing cycles. This study used a 4D dynamic phantom to verify the accuracy of this technique and also to develop a procedure for using CBCT to clinically verify ITV in stereotactic body radiotherapy (SBRT) treatment. Method and Materials: A CIRS 4D dynamic phantom, with a target ball and precisely controlled motion, was imaged using a 4D CT scanner. A Varian RPM system was used for respiratory gating. Ten 3‐D image sets were generated corresponding to 10 breathing phases. The ITV was determined based on the phase images. To assess concordance, on‐board CBCT images of the target ball were compared with the 4D‐CT defined ITV. SBRT patient with tumor targets located in the thorax and upper abdomen were similarly scanned using phase gating 4D CT. The ITVs were compared between simulation CT and CBCT scans to identify localization error. Results: The dynamic phantom motion was 20mm along the inferior‐superior direction, 5mm along the anterior‐posterior direction, and 2mm along the left‐right direction with a cycle time of 4 seconds. The concordance of the CBCT and ITV matching was within 1 mm. For the lung SBRT patient, the target volume based on the CT images without respiratory gating was 0.7 cc. The ITV was 2 cc. The ITV matches well with the CBCT images. The localization errors between free‐breathing CT and CBCT were 2 mm to the right, 1 mm to the anterior, and 2 mm to the superior. Conclusion: CBCT provides an accurate assessment of the ITV for targets affected by respiratory motion.