4D-CBCT Images Research Articles

SU‐E‐J‐183: Quantifying the Image Quality and Dose Reduction of Respiratory Triggered 4D Cone‐Beam Computed Tomography with Patient‐ Measured Breathing

Purpose:Respiratory triggered four dimensional cone‐beam computed tomography (RT 4D CBCT) is a novel technique that uses a patient's respiratory signal to drive the image acquisition with the goal of imaging dose reduction without degrading image quality. This work investigates image quality and dose using patient‐measured respiratory signals for RT 4D CBCT simulations instead of synthetic sinusoidal signals used in previous work.Methods:Studies were performed that simulate a 4D CBCT image acquisition using both the novel RT 4D CBCT technique and a conventional 4D CBCT technique from a database of oversampled Rando phantom CBCT projections. A database containing 111 free breathing lung cancer patient respiratory signal files was used to create 111 RT 4D CBCT and 111 conventional 4D CBCT image datasets from realistic simulations of a 4D RT CBCT system. Each of these image datasets were compared to a ground truth dataset from which a root mean square error (RMSE) metric was calculated to quantify the degradation of image quality. The number of projections used in each simulation is counted and was assumed as a surrogate for imaging dose.Results:Based on 111 breathing traces, when comparing RT 4D CBCT with conventional 4D CBCT the average image quality was reduced by 7.6%. However, the average imaging dose reduction was 53% based on needing fewer projections (617 on average) than conventional 4D CBCT (1320 projections).Conclusion:The simulation studies using a wide range of patient breathing traces have demonstrated that the RT 4D CBCT method can potentially offer a substantial saving of imaging dose of 53% on average compared to conventional 4D CBCT in simulation studies with a minimal impact on image quality.A patent application (PCT/US2012/048693) has been filed which is related to this work.

SU-E-J-246: A Deformation-Field Map Based Liver 4D CBCT Reconstruction Method Using Gold Nanoparticles as Constraints

Purpose: To investigate the feasibility of using nanoparticle markers to validate liver tumor motion together with a deformation field map-based four dimensional (4D) cone-beam computed tomography (CBCT) reconstruction method. Methods: A technique for lung 4D-CBCT reconstruction has been previously developed using a deformation field map (DFM)-based strategy. In this method, each phase of the 4D-CBCT is considered as a deformation of a prior CT volume. The DFM is solved by a motion modeling and free-form deformation (MM-FD) technique, using a data fidelity constraint and the deformation energy minimization. For liver imaging, there is low contrast of a liver tumor in on-board projections. A validation of liver tumor motion using implanted gold nanoparticles, along with the MM-FD deformation technique is implemented to reconstruct onboard 4D CBCT liver radiotherapy images. These nanoparticles were placed around the liver tumor to reflect the tumor positions in both CT simulation and on-board image acquisition. When reconstructing each phase of the 4D-CBCT, the migrations of the gold nanoparticles act as a constraint to regularize the deformation field, along with the data fidelity and the energy minimization constraints. In this study, multiple tumor diameters and positions were simulated within the liver for on-board 4D-CBCT imaging. The on-board 4D-CBCT reconstructed by the proposed method was compared with the “ground truth” image. Results: The preliminary data, which uses reconstruction for lung radiotherapy suggests that the advanced reconstruction algorithm including the gold nanoparticle constraint will Resultin volume percentage differences (VPD) between lesions in reconstructed images by MM-FD and “ground truth” on-board images of 11.5% (± 9.4%) and a center of mass shift of 1.3 mm (± 1.3 mm) for liver radiotherapy. Conclusion: The advanced MM-FD technique enforcing the additional constraints from gold nanoparticles, results in improved accuracy for reconstructing on-board 4D-CBCT of liver tumor. Varian medical systems research grant

WE-G-BRF-03: A Quasi-Cine CBCT Reconstruction Technique for Real-Time On- Board Target Tracking of Lung Cancer Treatment

Purpose: To develop a quasi-cine CBCT reconstruction technique that uses extremely-small angle (∼3°) projections to generate real-time high-quality lung CBCT images. Method: 4D-CBCT is obtained at the beginning and used as prior images. This study uses extremely-small angle (∼3°) on-board projections acquired at a single respiratory phase to reconstruct the CBCT image at this phase. An adaptive constrained free-form deformation (ACFD) method is developed to deform the prior 4D-CBCT volume at the same phase to reconstruct the new CBCT. Quasi-cine CBCT images are obtained by continuously reconstructing CBCT images at subsequent phases every 3° angle (∼0.5s). Note that the prior 4D-CBCT images are dynamically updated using the latest CBCT images. The 4D digital extended-cardiac-torso (XCAT) phantom was used to evaluate the efficacy of ACFD. A lung patient was simulated with a tumor baseline shift of 2mm along superior-inferior (SI) direction after every respiratory cycle for 5 cycles. Limited-angle projections were simulated for each cycle. The 4D-CBCT reconstructed by these projections were compared with the ground-truth generated in XCAT.Volume-percentage-difference (VPD) and center-of-mass-shift (COMS) were calculated between the reconstructed and the ground-truth tumors to evaluate their geometric differences.The ACFD was also compared to a principal-component-analysis based motion-modeling (MM) method. Results: Using orthogonal-view 3° projections, the VPD/COMS values for tumor baseline shifts of 2mm, 4mm, 6mm, 8mm, 10mm were 11.0%/0.3mm, 25.3%/2.7mm, 22.4%/2.9mm, 49.5%/5.4mm, 77.2%/8.1mm for the MM method, and 2.9%/0.7mm, 3.9%/0.8mm, 6.2%/1mm, 7.9%/1.2mm, 10.1%/1.1mm for the ACFD method. Using orthogonal-view 0° projections (1 projection only), the ACFD method yielded VPD/COMS results of 5.0%/0.9mm, 10.5%/1.2mm, 15.1%/1.4mm, 20.9%/1.6mm and 24.8%/1.6mm. Using single-view instead of orthogonal-view projections yielded less accurate results for ACFD. Conclusion: The ACFD method accurately reconstructs snapshot CBCT images using orthogonal-view 3° projections. It has a great potential to provide real-time quasi-cine CBCT images for verification in lung radiation therapy. The research is supported by grant from Varian Medical Systems.

Image quality in thoracic 4D cone‐beam CT: A sensitivity analysis of respiratory signal, binning method, reconstruction algorithm, and projection angular spacing

Respiratory signal, binning method, and reconstruction algorithm are three major controllable factors affecting image quality in thoracic 4D cone-beam CT (4D-CBCT), which is widely used in image guided radiotherapy (IGRT). Previous studies have investigated each of these factors individually, but no integrated sensitivity analysis has been performed. In addition, projection angular spacing is also a key factor in reconstruction, but how it affects image quality is not obvious. An investigation of the impacts of these four factors on image quality can help determine the most effective strategy in improving 4D-CBCT for IGRT. Fourteen 4D-CBCT patient projection datasets with various respiratory motion features were reconstructed with the following controllable factors: (i) respiratory signal (real-time position management, projection image intensity analysis, or fiducial marker tracking), (ii) binning method (phase, displacement, or equal-projection-density displacement binning), and (iii) reconstruction algorithm [Feldkamp-Davis-Kress (FDK), McKinnon-Bates (MKB), or adaptive-steepest-descent projection-onto-convex-sets (ASD-POCS)]. The image quality was quantified using signal-to-noise ratio (SNR), contrast-to-noise ratio, and edge-response width in order to assess noise/streaking and blur. The SNR values were also analyzed with respect to the maximum, mean, and root-mean-squared-error (RMSE) projection angular spacing to investigate how projection angular spacing affects image quality. The choice of respiratory signals was found to have no significant impact on image quality. Displacement-based binning was found to be less prone to motion artifacts compared to phase binning in more than half of the cases, but was shown to suffer from large interbin image quality variation and large projection angular gaps. Both MKB and ASD-POCS resulted in noticeably improved image quality almost 100% of the time relative to FDK. In addition, SNR values were found to increase with decreasing RMSE values of projection angular gaps with strong correlations (r ≈ -0.7) regardless of the reconstruction algorithm used. Based on the authors' results, displacement-based binning methods, better reconstruction algorithms, and the acquisition of even projection angular views are the most important factors to consider for improving thoracic 4D-CBCT image quality. In view of the practical issues with displacement-based binning and the fact that projection angular spacing is not currently directly controllable, development of better reconstruction algorithms represents the most effective strategy for improving image quality in thoracic 4D-CBCT for IGRT applications at the current stage.

Respiratory motion guided four dimensional cone beam computed tomography: encompassing irregular breathing

Four dimensional cone beam computed tomography (4DCBCT) images suffer from angular under sampling and bunching of projections due to a lack of feedback between the respiratory signal and the acquisition system. To address this problem, respiratory motion guided 4DCBCT (RMG-4DCBCT) regulates the gantry velocity and projection time interval, in response to the patient’s respiratory signal, with the aim of acquiring evenly spaced projections in a number of phase or displacement bins during the respiratory cycle. Our previous study of RMG-4DCBCT was limited to sinusoidal breathing traces. Here we expand on that work to provide a practical algorithm for the case of real patient breathing data. We give a complete description of RMG-4DCBCT including full details on how to implement the algorithms to determine when to move the gantry and when to acquire projections in response to the patient’s respiratory signal. We simulate a realistic working RMG-4DCBCT system using 112 breathing traces from 24 lung cancer patients. Acquisition used phase-based binning and parameter settings typically used on commercial 4DCBCT systems (4 min acquisition time, 1200 projections across 10 respiratory bins), with the acceleration and velocity constraints of current generation linear accelerators. We quantified streaking artefacts and image noise for conventional and RMG-4DCBCT methods by reconstructing projection data selected from an oversampled set of Catphan phantom projections. RMG-4DCBCT allows us to optimally trade-off image quality, acquisition time and image dose. For example, for the same image quality and acquisition time as conventional 4DCBCT approximately half the imaging dose is needed. Alternatively, for the same imaging dose, the image quality as measured by the signal to noise ratio, is improved by 63% on average. C-arm cone beam computed tomography systems, with an acceleration up to 200°/s2, a velocity up to 100°/s and the acquisition of 80 projections per second, allow the image acquisition time to be reduced to below 60 s. We have made considerable progress towards realizing a system to reduce projection clustering in conventional 4DCBCT imaging and hence reduce the imaging dose to the patient.

4D CBCT image guidance in SBRT for lung tumors

Introduction and objective: To analyze 4D image guided accuracy and dose volume histograms in SBRT for lung tumors.

An Integrated Simulation System Based on Digital Human Phantom for 4D Radiation Therapy of Lung Cancer

Purpose: To develop and test an integrated simulation system based on the digital Extended Cardio Torso (XCAT) phantom for 4-dimensional (4D) radiation therapy of lung cancer. Methods: A computer program was developed to facilitate the characterization and implementation of the XCAT phantom for 4D radiation therapy applications. To verify that patient-specific motion trajectories are reproducible with the XCAT phantom, motion trajectories of the diaphragm and chest were extracted from previously acquired MRI scans of five subjects and were imported into the XCAT phantom. The input versus the measured trajectories was compared. Simulation methods of 4D-CT and 4D-cone-beam CT (CBCT) based on the XCAT phantom were developed and tested for regular and irregular respiratory patterns. Simulation of 4D dose delivery was illustrated in a simulated lung stereotactic-body radiation therapy (SBRT) case based on the XCAT phantom. Dosimetric comparison was performed between the planned dose and simulated delivered dose. Result: The overall mean (±standard deviation) difference in motion amplitude between the input and measured trajectories was 1.19 (±0.79) mm for the XCAT phantoms with voxel size of 2 mm. 4D-CT and 4D-CBCT images simulated based on the XCAT phantom were validated using regular respiratory patterns and tested for irregular respiratory patterns. Comparison between simulated 4D dose delivery and planned dose for the lung SBRT case showed comparable results in all dosimetric matrices: the relative differences were 0.3%, 4.0%, 0%, and 2.8%, respectively, for max cord dose, max esophagus dose, mean heart dose, and V20Gy of the lungs. 97.5% of planning target volume (PTV) received prescription dose in the simulated 4D delivery, as compared to 95% of PTV received prescription dose in the plan. Conclusion: We developed an integrated simulation system based on the XCAT digital phantom and illustrated its utility in 4D radiation therapy of lung cancer. This simulation system is potentially a useful tool for quality control and development of imaging and treatment techniques for 4D radiation therapy of lung cancer.

Motion‐map constrained image reconstruction (MCIR): Application to four‐dimensional cone‐beam computed tomography

Utilization of respiratory correlated four-dimensional cone-beam computed tomography (4DCBCT) has enabled verification of internal target motion and volume immediately prior to treatment. However, with current standard CBCT scan, 4DCBCT poses challenge for reconstruction due to the fact that multiple phase binning leads to insufficient number of projection data to reconstruct and thus cause streaking artifacts. The purpose of this study is to develop a novel 4DCBCT reconstruction algorithm framework called motion-map constrained image reconstruction (MCIR), that allows reconstruction of high quality and high phase resolution 4DCBCT images with no more than the imaging dose as well as projections used in a standard free breathing 3DCBCT (FB-3DCBCT) scan. The unknown 4DCBCT volume at each phase was mathematically modeled as a combination of FB-3DCBCT and phase-specific update vector which has an associated motion-map matrix. The motion-map matrix, which is the key innovation of the MCIR algorithm, was defined as the matrix that distinguishes voxels that are moving from stationary ones. This 4DCBCT model was then reconstructed with compressed sensing (CS) reconstruction framework such that the voxels with high motion would be aggressively updated by the phase-wise sorted projections and the voxels with less motion would be minimally updated to preserve the FB-3DCBCT. To evaluate the performance of our proposed MCIR algorithm, we evaluated both numerical phantoms and a lung cancer patient. The results were then compared with the (1) clinical FB-3DCBCT reconstructed using the FDK, (2) 4DCBCT reconstructed using the FDK, and (3) 4DCBCT reconstructed using the well-known prior image constrained compressed sensing (PICCS). Examination of the MCIR algorithm showed that high phase-resolved 4DCBCT with sets of up to 20 phases using a typical FB-3DCBCT scan could be reconstructed without compromising the image quality. Moreover, in comparison with other published algorithms, the image quality of the MCIR algorithm is shown to be excellent. This work demonstrates the potential for providing high-quality 4DCBCT during on-line image-guided radiation therapy (IGRT), without increasing the imaging dose. The results showed that (at least) 20 phase images could be reconstructed using the same projections data, used to reconstruct a single FB-3DCBCT, without streak artifacts that are caused by insufficient projections.

MO‐F‐WAB‐10: Evaluation of Scan Time Required for Motion Tracking in Four‐Dimensional Cone Beam CT Using Patient Data Sets

Purpose: Applying four‐dimensional cone beam CT in a radiation treatment setting requires a tradeoff between image quality and scan time. The aim of this study is to assess the minimum scan time required to determine the motion of tumors immediately prior to radiation treatment. Methods: Long CBCT scans of approximately 10 minutes were taken from 11 patients with lung lesions. Shorter scan times (down to about 1 minute) for these patients were simulated by removing projections. Four dimensional images were created from each full and simulated scan time using a ROI based FDK reconstruction algorithm with 8 breathing phases. Tumor motion for each phase relative to the first phase was determined through deformable image registration in a region of interest. The ability to track motion was assessed by comparing the results of the image registration of the simulated scans to the results for the full data scan. Results: For each comparison between the full data set and a reduced data set, a root‐mean‐square error was calculated comparing the two motions on a phase by phase basis, with the summation being over the phases. As anticipated, the RMS error as well as the spread in the RMS error increased with decreasing scan time. Good tumor tracking was observed down to 2 minute scan times (RMS error = 1.9 ± 0.6 mm), while 1 minute scan times demonstrated a large spread in the error (RMS error = 2.4 ± 1.8 mm). Conclusion: This study suggests scan times down to two minutes are sufficient for the extraction of tumor motion trajectories from 4D CBCT images. Additionally, this study sets up the ability to compare other reconstruction methods to the traditional FDK method.

SU-D-144-02: Four-Dimensional Cone Beam CT for a Small Animal Image Guided Radiation Therapy System Without Use of An External Respiratory Monitoring System

Purpose: To develop four‐dimensional cone beam CT for an image guided radiation system for small animals without the use of external respiration monitoring systems. Methods: Nine mice were scanned using the cone beam CT capabilities of an X‐Rad 225Cx system. The mice were anesthetized with 1.5 to 3% isoflurane and scanned (60 kVp and 4 mA) at a sampling rate of 15 fps and gantry rotation rate of 0.5 rpm to yield a total of approximately 1800 975×975 projection images. The projections were used to obtain a respiratory signal using a modified Amsterdam Shroud method. Inspirations appeared approximately as vertical lines in the Amsterdam Shroud image due to the small ratio of inspiration to expiration times in mice anesthetized with isoflurane. The locations of these lines were extracted using vertical edge detection, dilation and projection onto one dimension. Four‐dimensional cone beam CT images (voxel size 0.1 × 0.1 × 0.3 mm) were then obtained using these peak inspirations to sort the projections into 6 phases. The peak inspirations of two mice were manually extracted from the projections to test the peak extraction method. Results: 4D CBCT images were successfully created for each mouse with minimal streak artifacts. Some blurring in the initial phase was seen due to the sharpness of the inspiration peak. The peak finding algorithm determined the peaks correctly to within an average of 0.7 and 0.4 projections for the first and second mouse, respectively. 82% and 89% of the projections were sorted into the correct phase bin for the two mice. Conclusion: The four‐dimensional CBCT method developed here can be used to analyze lung motion in mice and assist in preclinical targeted radiation therapy studies.

WE-G-134-06: Image Quality in Thoracic 4D Cone-Beam CT: A Sensitivity Analysis of Respiratory Signal Source, Binning Method, and Reconstruction Algorithm

Purpose: Respiratory signal source, binning method, and reconstruction algorithm are three major factors affecting thoracic 4D cone-beam CT (4D-CBCT) image quality. Various studies have investigated each of them individually, but no sensitivity analysis to these factors has been performed. This study quantitatively compares the impacts of the three factors on thoracic 4D-CBCT image quality. Methods: A 4D-CBCT patient projection dataset acquired using audio-visual biofeedback with known degree of baseline shift in respiratory motion was reconstructed with the following factors: (i) respiratory signal source (via real-time position management or projection image based method), (ii) binning method (phase, displacement, or equal-projection-density (EPD) displacement binning), and (iii) reconstruction algorithm (Feldkamp-Davis-Kress (FDK), McKinnon-Bates (MKB), or adaptive-steepest-descent projection-onto-convex-sets (ASD-POCS)). The image quality was quantified using signal-to-noise ratio (SNR), contrast ratio (CR), and edge-response width (ERW) to assess noise, contrast, and sharpness, respectively. Results: Respiratory signal sources had no significant effect on any of the three metrics. Displacement binning was found to produce slightly more inconsistent SNR values owing to unevenly distributed angular views. The binning method otherwise had minimal effects on image quality. For the reconstruction algorithms, MKB resulted in a slightly higher SNR than FDK, while ASD-POCS improved SNR by a factor of three. Conversely, degradation in CR was observed with MKB and ASD-POCS. FDK and MKB produced very similar ERW, while slightly higher values were observed with ASD-POCS, indicating a small degree of blurriness. Conclusion: Respiratory signal sources and binning methods had a negligible impact on image quality, and did not introduce apparent motion artifacts. The MKB and ASD-POCS reconstruction algorithms reduced the noise and streaking but produced poorer image contrast. Of the reconstruction factors tested, reconstruction algorithms play the most important role in improving 4D-CBCT image quality and therefore represent the highest priority for further development. This project is supported by an NHMRC Australia Fellowship, NHMRC project grant 1034060, US NCI P01CA116602, an International Postgraduate Research Scholarships, and an Australian Postgraduate Award.

WE-G-134-07: Respiratory Motion Guided Four Dimensional Cone Beam Computed Tomography: Image Quality Analysis

Purpose: The aim of Respiratory Motion Guided-4DCBCT (RMG-4DCBCT) is to improve image quality and reduce radiation dose in 4DCBCT imaging. In current generation 4DCBCT, the gantry rotation speed and imaging frequency are constant and independent of the patient's breathing which leads to projection clustering. We have developed a software system which regulates the gantry velocity and projection time interval, in response to the patient's real time respiratory signal, so that a full set of near-evenly spaced projections can be acquired for each respiratory phase. We present our results analysing the quality of images produced with RMG-4DCBCT using the Catphan phantom. Methods: The RMG-4DCBCT software: (1) optimizes the gantry trajectory and projection pulse rate schedule based on the patients predicted breathing trace, (2) adjusts the gantry velocity and projection pulse rate according to the patients real time breathing signal and (3) sends commands to the gantry and kilovoltage imager. Acquisition of 60, RMG-4DCBCT(60), and 120, RMG-4DCBCT(120), half fan projections in each of 10 respiratory phases was simulated in real-time using 111 breathing traces from 24 lung cancer patients. A dataset containing 608 half fan projections of the Catphan phantom was sampled to reconstruct 4DCBCT images using both conventional 4DCBCT and RMG-4DCBCT. From the reconstructed images a streak ratio (SR) was calculated to quantify image quality. Results: The average SR was reduced from 5.16 with conventional 4DCBCT to 1.5 with RMG-4DCBCT(120) and 2.9 with RMG-4DCBCT(60). The average time required to acquire the dataset was 240 seconds with conventional 4DCBCT, 599 seconds for RMG-4DCBCT(120) and 260 seconds for RMG-4DCBCT(60). If we force RMG-4DCBCT to acquire 120 projections in exactly 240 seconds the SR is 4.2. Conclusion: There are less streak artifacts with RMG-4DCBCT than conventional 4DCBCT. The imaging time for RMG-4DCBCT depends on the patients breathing rate and number of projections acquired. This project is supported by an NHMRC Australia Fellowship and NHMRC project grant 1034060.

SU‐C‐141‐04: Robustness of 4DCT and 4DCBCT Object Volume Measurement with a Motion‐Capable Lung Phantom That Mimics Realistic Patient Tumor Trajectories

Purpose: Accurate volume estimation for time‐resolved fan‐beam CT (4DCT) and cone‐beam CT (4DCBCT) is critical to planning and verification of many motion‐sensitive radiotherapy applications, including stereotactic lung radiotherapy. We used a 4D torso phantom to acquire realistic respiratory images with 4DCT and 4DCBCT, and compared object size estimation of the two modalities. Methods: A torso phantom consisting of a water‐filled sphere surrounded by foam pellets was driven by a motion device capable of reproducing patient respiratory traces. Ten typical respiratory traces were imaged along with static and sinusoidal cases with clinically‐relevant 4DCT and 4DCBCT techniques. 4DCT images were reconstructed in ten phases at 512×512×104 voxels and 0.977×0.977×2.50 mm3/voxel (GE Advantage). 4DCBCT images were reconstructed in ten phases at 128×205×205 voxels and 2.00×2.00×2.00 mm3/voxel (Elekta XVI). Objects were contoured in MIM 5.6.3 using threshold segmentation and volumes compared between image phases. Results: Reference object size was 7.03 mL and 7.01 mL for the static CT and CBCT acquisitions, respectively. When averaged over all phases, 4DCT volume was 6.78+/−0.72 mL; however, variation between phases was large and significant in Kruskal‐Wallis rank‐sum tests (2.68‐‐10.96 mL, p<0.001). 4DCBCT showed reduced accuracy of object volume estimation relative to the static case and 4DCT in paired sign‐rank tests (μ =9.08+/−0.07 mL, p<0.001); however, 4DCBCT greatly improved consistency of estimation, as no significant differences were detected between phases (8.47‐‐9.10 mL, p=0.4). Conclusion: Variation in object volume between respiratory phases was dramatically inferior for 4DCT versus 4DCBCT, possibly due to fan‐beam CT interplay effects or superiority of diaphragm sorting versus a linear surrogate. Consequently, radiotherapy planning or gating on individual phases of 4DCT datasets has potential to cause significant error. 4DCBCT did overestimate object size relative to 4DCT due to partial volume effects. As such, higher‐resolution 4DCBCT presets should be investigated for patient verification imaging.

TU-C-141-04: Evaluation of Clinical Acceptability of DIR Mapped Contours for Adaptive Radiotherapy with 4D Cone-Beam CT

Purpose: To evaluate clinical acceptability of contours generated using deformable image registration of 4D fan beam (4DFBCT) to 4D cone beam (4DCBCT) images for image-guided adaptive radiotherapy (IGART). Methods: Sixteen locally advanced non-small cell lung cancer (NSCLC) patients underwent one planning 4DFBCT and weekly 4DCBCT scans. 4DFBCT to 4DCBCT registrations were performed using two intensity-driven deformable image registration (DIR) algorithms: 1) small deformation, inverse consistent linear elastic (SICLE) algorithm and 2) Insight Toolkit diffeomorphic demons (DEMONS). Multiple physicians delineated the gross tumor volume in all images (MANUAL). Manual contours were mapped from 4DFBCT to 4DCBCT for two scans in each patient, and were compared with manual contours using Dice similarity coefficient (DSC), average symmetric distance (ASD), false positive (FP), false negative (FN) and percentage volume difference (%VD). A physician was asked to rate DIR generated and manual contours as A) clinically acceptable (CA) B) clinically acceptable after minor modifications (CAMM) and C) clinically unacceptable (CU) using planning CT contour as reference. Reviewing physician was unaware that some of the contours were manual. The results were analyzed with respect to similarity measures, volume and regression characteristics of the tumor. Results: No contours were rated as CU by the physician. The number of contours rated as CA and CAMM was respectively 14 (44%) and 18 (56%) for SICLE, 3 (9%) and 29 (91%) for DEMONS, and 26 (81%) and 6 (19%) for MANUAL. No correlation was observed between physician ratings and DSC, ASD, FP, FN, %VD, tumor shrinkage, day of CBCT scan, and volume of the GTV. Conclusion: DIR generated contours were considered clinically acceptable for IGART, but may require some manual adjustment prior to use. Volume and surface similarity measures did not correlate with physician judgment of clinical acceptability. Supported by NIH Grant P01 CA 116602; This work was supported by a grant (P01CA116602) from the National Cancer Institute. Gary E. Christensen holds a papent related to the SICLE registration algorithm; E. Weiss and J. Williamson has grants from Varian medical systems and Philips Radiation Oncology Systems.

TU‐G‐141‐04: Phase‐Selected 4d CBCT Acquisition Based On An Internal‐External Motion Correlation Model

Purpose: X‐ray projections acquired with linac‐based scanners for 4d CBCT reconstruction are traditionally binned into breathing phases ‘after‐the‐fact’, i.e. after the image acquisition. This approach leads to an uneven spreading of projections over breathing phases, which compromises the image quality. We have therefore developed a novel approach based on actively triggered projections employing the forward‐predicted position of the tumor at the time of image acquisition. Methods: The prediction is based on a high‐frequency external motion sensor and uses linear regression to compensate for the effective latency of the external motion sensor and the x‐ray imaging chain (160 ms). An internal‐to‐external motion model is established from the first 40 images, by correlating the external chest‐wall displacement (AP) with the internal motion (SI) derived from the images (using an ‘Amsterdam shroud’). We used a lung phantom moving on a sinusoidal trajectory (15 mm SI amplitude) with a 3.5/5.0 s period. The AP motion (5 mm) was phase‐shifted by 13°. The gantry rotation times were adapted depending on the breathing period during prediction training (first 30 s). Results: The first 40–50 projections were acquired at 6.25 Hz to cover 1–2 breathing cycles. During post‐processing, these images were binned and used alongside the actively triggered frames. For the 4d CBCT mode, we were able to evenly spread the projections over all breathing phases. Motion‐induced artifacts were effectively mitigated and the tumor‐to‐lung contrast became more evenly distributed over the respiratory phases. We also demonstrated a quasi‐4d mode consisting of actively triggered peak‐exhale/peak‐inhale images only. By increasing the image frequency for the quasi‐4d mode, it was possible to reduce the gantry rotation to 2.5–3.5 min without compromising the number of projections. Conclusion: Our new approach to 4d CBCT imaging allows for active phase selection as opposed to the traditional retrospective phase binning.

TH‐C‐103‐08: BEST IN PHYSICS (IMAGING) — A Hybrid 4D Cone Beam CT Reconstruction Algorithm for Highly Under‐Sampled Projections From the 1‐Minute Cone Beam Scan

Purpose: To facilitate fast and accurate 4D image guidance for lung and upper abdomen radiotherapy, we propose a novel 4D cone beam CT (4D‐CBCT) reconstruction algorithm that offers high‐quality images from highly under‐sampled projections under the conventional 1‐min CBCT scan. Methods: As opposed to conventional reconstruction approaches via image intensity domain, our method integrates patient‐specific CT information and reconstructs 4D‐CBCT image via motion vector domain. It solves an optimization problem, where a deformation vector field is determined, such that, when being used to deform the patient prior CT image, the projections match the measurements. We develop a forward‐backward splitting (FBS) method to solve this problem. Specifically, an intermediate variable is introduced to split the original problem into two iteratively preformed sub‐problems, i.e., reconstruction of the intermediate variable based on projections and the currently reconstructed 4DCBCT image, and deformable registration between the prior image and the intermediate variable. This hybrid algorithm achieves a solution with both right geometrical information defined by measured CBCT projections and correct intensity values from the prior CT, yielding high quality 4D‐CBCT images. Both experiments on a moving ball phantom and patient studies have been performed to validate the proposed method. Results: Even with very limited number of projections from a 1‐min CBCT scan, artifacts‐free 4D‐CBCT images with correct HU values can be obtained by the proposed algorithm in both phantom and patient studies. To validate the anatomical geometry accuracy, the ground‐truth ball center location along the superior‐inferior direction is compared to that measured in the reconstructed 4D‐CBCT. Satisfactory agreements (0.364mm average and 0.504mm maximum error) are observed. Conclusion: A hybrid 4D‐CBCT reconstruction algorithm is developed for highly under‐sampled projections from the conventional 1‐min CBCT scan. It is capable of reconstructing high‐quality images free of under‐sampled streaking with accurate HU values. This work is supported in part by NIH (1R01CA154747‐01), Varian Medical Systems through a Master Research Agreement, the Early Career Award from Thrasher Research Fund, and the University of California Lab Fees Research Program.

TH‐C‐103‐01: Motion‐Map Constrained Image Reconstruction for 4DCBCT Reconstruction in IGRT

Purpose: We developed a novel low‐dose 4DCBCT reconstruction algorithm, named Motion‐Map Constrained Image Reconstruction (MCIR), that allows reconstruction of high quality and high phase resolution 4DCBCT images with no more than the imaging dose as well as the X‐ray projections used in a standard free‐breathing 3DCBCT (FB‐3DCBCT) scan. Methods: The unknown 4DCBCT volume at each phase was mathematically modeled as combination of FB‐3DCBCT and motion‐map matrix with phase‐specific update vector. The motion‐map matrix, which is the key innovation of MCIR algorithm, was defined as the matrix that distinguishes voxels between ones that are moving and stationary. This 4DCBCT model was then reconstructed with compressed sensing (CS) reconstruction framework such that the voxels with high motion would be aggressively updated by the input phase information and the voxels with less motion would be minimally updated to preserve the FB‐3DCBCT. To evaluate the performance of our proposed MCIR algorithm, we have tested on both numerical phantoms and lung cancer patient CBCT scans. The results were then compared with the 1) clinical FB‐3DCBCT reconstructed using FDK, 2) 4DCBCT reconstructed using FDK, and 3) 4DCBCT reconstructed using the best‐known algorithm to date: PICCS. Results: Examination of the MCIR algorithm showed that high phase resolution 4DCBCT with sets up to 20 phases using typical patient FB‐3DCBCT scan could be reconstructed without compromising the image quality. Moreover, both in numerical as well as in real patient data sets, the MCIR algorithm outperformed the other algorithms. This is achieved without using any a priori data, such as the planning CT. Conclusion: This work demonstrates the potential for providing high‐quality 4DCBCT information during on‐line IGRT and ART environment, without sacrificing imaging dose. Proposed algorithm is also entirely compatible to GPU programming, which could sufficiently speed up the process to achieve within clinically usable time‐frame (<30 sec).

Verification of Planning Target Volume Settings in Volumetric Modulated Arc Therapy for Stereotactic Body Radiation Therapy by Using In-Treatment 4-Dimensional Cone Beam Computed Tomography

To evaluate setup error and tumor motion during beam delivery by using 4-dimensional cone beam computed tomography (4D CBCT) and to assess the adequacy of the planning target volume (PTV) margin for lung cancer patients undergoing volumetric modulated arc therapy for stereotactic body radiation therapy (VMAT-SBRT). Fifteen lung cancer patients treated by single-arc VMAT-SBRT were selected in this analysis. All patients were treated with an abdominal compressor. The gross tumor volumes were contoured on maximum inspiration and maximum expiration CT datasets from 4D CT respiratory sorting and merged into internal target volumes (ITVs). The PTV margin was isotropically taken as 5 mm. Registration was automatically performed using "pre-3D" CBCT. Treatment was performed with a D95 prescription of 50 Gy delivered in 4 fractions. The 4D tumor locations during beam delivery were determined using in-treatment 4D CBCT images acquired in each fraction. Then, the discrepancy between the actual tumor location and the ITV was evaluated in the lateral, vertical, and longitudinal directions. Overall, 55 4D CBCT sets during VMAT-SBRT were successfully obtained. The amplitude of tumor motion was less than 10 mm in all directions. The average displacements between ITV and actual tumor location during treatment were 0.41 ± 0.93 mm, 0.15 ± 0.58 mm, and 0.60 ± 0.99 mm for the craniocaudal, left-right, and anteroposterior directions, respectively. The discrepancy in each phase did not exceed 5 mm in any direction. With in-treatment 4D CBCT, we confirmed the required PTV margins when the registration for moving target was performed using pre-3D CBCT. In-treatment 4D CBCT is a direct method for quantitatively assessing the intrafractional location of a moving target.

Evaluation of 4-dimensional Computed Tomography to 4-dimensional Cone-Beam Computed Tomography Deformable Image Registration for Lung Cancer Adaptive Radiation Therapy

To evaluate 2 deformable image registration (DIR) algorithms for the purpose of contour mapping to support image-guided adaptive radiation therapy with 4-dimensional cone-beam CT (4DCBCT). One planning 4D fan-beam CT (4DFBCT) and 7 weekly 4DCBCT scans were acquired for 10 locally advanced non-small cell lung cancer patients. The gross tumor volume was delineated by a physician in all 4D images. End-of-inspiration phase planning 4DFBCT was registered to the corresponding phase in weekly 4DCBCT images for day-to-day registrations. For phase-to-phase registration, the end-of-inspiration phase from each 4D image was registered to the end-of-expiration phase. Two DIR algorithms-small deformation inverse consistent linear elastic (SICLE) and Insight Toolkit diffeomorphic demons (DEMONS)-were evaluated. Physician-delineated contours were compared with the warped contours by using the Dice similarity coefficient (DSC), average symmetric distance, and false-positive and false-negative indices. The DIR results are compared with rigid registration of tumor. For day-to-day registrations, the mean DSC was 0.75 ± 0.09 with SICLE, 0.70 ± 0.12 with DEMONS, 0.66 ± 0.12 with rigid-tumor registration, and 0.60 ± 0.14 with rigid-bone registration. Results were comparable to intraobserver variability calculated from phase-to-phase registrations as well as measured interobserver variation for 1 patient. SICLE and DEMONS, when compared with rigid-bone (4.1 mm) and rigid-tumor (3.6 mm) registration, respectively reduced the average symmetric distance to 2.6 and 3.3 mm. On average, SICLE and DEMONS increased the DSC to 0.80 and 0.79, respectively, compared with rigid-tumor (0.78) registrations for 4DCBCT phase-to-phase registrations. Deformable image registration achieved comparable accuracy to reported interobserver delineation variability and higher accuracy than rigid-tumor registration. Deformable image registration performance varied with the algorithm and the patient.

4-dimensional Cone Beam CT Ventilation Imaging to Facilitate Adaptive Functional Avoidance Lung Cancer Radiation Therapy

Four-dimensional-CT ventilation imaging has been proposed as an alternative to nuclear medicine (NM) ventilation imaging and may allow functional avoidance of the healthiest lung regions in lung cancer radiation therapy. However, the efficacy of functional avoidance will diminish if there are temporal changes in the ventilation pattern throughout a course of radiation therapy. To account for the temporal changes in ventilation patterns, we investigated 4D cone-beam CT (4D-CBCT) imaging to quantify ventilation.