kV Projections Research Articles

SU‐G‐JeP4‐10: Measurement of Prostate Motion Trajectories During Prostate SBRT Boost Treatments with a Rectafix

Purpose:To determine prostate motion during SBRT boost treatments with a Rectafix rectal sparing device in place using kV imaging during treatment.Methods:Patients each had three gold fiducial markers inserted into the prostate and received two VMAT boost fractions of 9.5–10 Gy under the PROMETHEUS clinical trial protocol with a Rectafix rectal retractor in place. Two‐dimensional kilovoltage images of fiducial markers were acquired continuously during delivery. Three patients were treated on a Varian Clinac iX linear accelerator (6X, 600 MU/min), where kV images were acquired at 5 Hz during treatment. Seven patients were treated on a Varian Truebeam linear accelerator (10XFFF, 2400 MU/min) where kV images were acquired every 3 seconds. Images were processed off‐line using the Kilovoltage Intrafraction Monitoring (KIM) software after treatment. KIM determines prostate position in three dimensions from 2D kV projections using a probability density model and a pre‐treatment kV arc. The 3D displacement of the prostate was quantified as a function of time throughout each fraction.Results:From all fractions analyzed, it was found that the prostate had moved less than 1 mm in any direction from its initial position 84.6% of the time. The prostate was between 1 and 2 mm from its initial position 14.2% of the time, between 2 and 3 mm of its initial position 0.8% of the time and was greater than 3 mm from its initial position only 0.4% of the time.Conclusion:The amount of prostate motion observed during prostate SBRT boost treatments with a Rectafix device in place was minimal and lower than that observed in non‐Rectafix studies. The Rectafix device reduces rectal dose as well as immobilizing the prostate. Kimberley Legge is the recipient of an Australian Postgraduate Award.

A software tool of digital tomosynthesis application for patient positioning in radiotherapy.

Digital Tomosynthesis (DTS) is an image modality in reconstructing tomographic images from two‐dimensional kV projections covering a narrow scan angles. Comparing with conventional cone‐beam CT (CBCT), it requires less time and radiation dose in data acquisition. It is feasible to apply this technique in patient positioning in radiotherapy. To facilitate its clinical application, a software tool was developed and the reconstruction processes were accelerated by graphic processing unit (GPU). Two reconstruction and two registration processes are required for DTS application which is different from conventional CBCT application which requires one image reconstruction process and one image registration process. The reconstruction stage consists of productions of two types of DTS. One type of DTS is reconstructed from cone‐beam (CB) projections covering a narrow scan angle and is named onboard DTS (ODTS), which represents the real patient position in treatment room. Another type of DTS is reconstructed from digitally reconstructed radiography (DRR) and is named reference DTS (RDTS), which represents the ideal patient position in treatment room. Prior to the reconstruction of RDTS, The DRRs are reconstructed from planning CT using the same acquisition setting of CB projections. The registration stage consists of two matching processes between ODTS and RDTS. The target shift in lateral and longitudinal axes are obtained from the matching between ODTS and RDTS in coronal view, while the target shift in longitudinal and vertical axes are obtained from the matching between ODTS and RDTS in sagittal view. In this software, both DRR and DTS reconstruction algorithms were implemented on GPU environments for acceleration purpose. The comprehensive evaluation of this software tool was performed including geometric accuracy, image quality, registration accuracy, and reconstruction efficiency. The average correlation coefficient between DRR/DTS generated by GPU‐based algorithm and CPU‐based algorithm is 0.99. Based on the measurements of cube phantom on DTS, the geometric errors are within 0.5 mm in three axes. For both cube phantom and pelvic phantom, the registration errors are within 0.5 mm in three axes. Compared with reconstruction performance of CPU‐based algorithms, the performances of DRR and DTS reconstructions are improved by a factor of 15 to 20. A GPU‐based software tool was developed for DTS application for patient positioning of radiotherapy. The geometric and registration accuracy met the clinical requirement in patient setup of radiotherapy. The high performance of DRR and DTS reconstruction algorithms was achieved by the GPU‐based computation environments. It is a useful software tool for researcher and clinician in evaluating DTS application in patient positioning of radiotherapy.PACS number(s): 87.57.nf

4D cone beam CT-based dose assessment for SBRT lung cancer treatment

The purpose of this research is to develop a 4DCBCT-based dose assessment method for calculating actual delivered dose for patients with significant respiratory motion or anatomical changes during the course of SBRT. To address the limitation of 4DCT-based dose assessment, we propose to calculate the delivered dose using time-varying (‘fluoroscopic’) 3D patient images generated from a 4DCBCT-based motion model. The method includes four steps: (1) before each treatment, 4DCBCT data is acquired with the patient in treatment position, based on which a patient-specific motion model is created using a principal components analysis algorithm. (2) During treatment, 2D time-varying kV projection images are continuously acquired, from which time-varying ‘fluoroscopic’ 3D images of the patient are reconstructed using the motion model. (3) Lateral truncation artifacts are corrected using planning 4DCT images. (4) The 3D dose distribution is computed for each timepoint in the set of 3D fluoroscopic images, from which the total effective 3D delivered dose is calculated by accumulating deformed dose distributions. This approach is validated using six modified XCAT phantoms with lung tumors and different respiratory motions derived from patient data. The estimated doses are compared to that calculated using ground-truth XCAT phantoms. For each XCAT phantom, the calculated delivered tumor dose values generally follow the same trend as that of the ground truth and at most timepoints the difference is less than 5%. For the overall delivered dose, the normalized error of calculated 3D dose distribution is generally less than 3% and the tumor D95 error is less than 1.5%. XCAT phantom studies indicate the potential of the proposed method to accurately estimate 3D tumor dose distributions for SBRT lung treatment based on 4DCBCT imaging and motion modeling. Further research is necessary to investigate its performance for clinical patient data.

Markerless tumor tracking using short kilovoltage imaging arcs for lung image-guided radiotherapy

The ability to monitor tumor motion without implanted markers is clinically advantageous for lung image-guided radiotherapy (IGRT). Existing markerless tracking methods often suffer from overlapping structures and low visibility of tumors on kV projection images. We introduce the short arc tumor tracking (SATT) method to overcome these issues. The proposed method utilizes multiple kV projection images selected from a nine-degree imaging arc to improve tumor localization, and respiratory-correlated 4D cone-beam CT (CBCT) prior knowledge to minimize the effects of overlapping anatomies. The 3D tumor position is solved as an optimization problem with prior knowledge incorporated via regularization. We retrospectively validated SATT on 11 clinical scans from four patients with central tumors. These patients represent challenging scenarios for markerless tumor tracking due to the inferior adjacent contrast. The 3D trajectories of implanted fiducial markers were used as the ground truth for tracking accuracy evaluation. In all cases, the tumors were successfully tracked at all gantry angles. Compared to standard pre-treatment CBCT guidance alone, trajectory errors were significantly smaller with tracking in all cases, and the improvements were the most prominent in the superior-inferior direction. The mean 3D tracking error ranged from 2.2–9.9 mm, which was 0.4–2.6 mm smaller compared to pre-treatment CBCT. In conclusion, we were able to directly track tumors with inferior visibility on kV projection images using SATT. Tumor localization accuracies are significantly better with tracking compared to the current standard of care of lung IGRT. Future work involves the prospective evaluation and clinical implementation of SATT.

Intrafractional 3D localization using kilovoltage digital tomosynthesis for sliding-window intensity modulated radiation therapy

To implement novel imaging sequences integrated into intensity modulated radiation therapy (IMRT) and determine 3D positions for intrafractional patient motion monitoring and management.In one method, we converted a static gantry IMRT beam into a series of arcs in which dose index and multileaf collimator positions for all control points were unchanged, but gantry angles were modified to oscillate ±3° around the original angle. Kilovoltage (kV) projections were acquired continuously throughout delivery and reconstructed to provide a series of 6° arc digital tomosynthesis (DTS) images which served to evaluate the in-plane positions of embedded-fiducials/vertebral-body. To obtain out-of-plane positions via triangulation, a 20° gantry rotation with beam hold-off was inserted during delivery to produce a pair of 6° DTS images separated by 14°. In a second method, the gantry remained stationary, but both kV source and detector moved over a 15° longitudinal arc using pitch and translational adjustment of the robotic arms. Evaluation of localization accuracy in an anthropomorphic Rando phantom during simulated intrafractional motion used programmed couch translations from customized scripts. Purpose-built software was used to reconstruct DTS images, register them to reference template images and calculate 3D fiducial positions.No significant dose difference (<0.5%) was found between the original and converted IMRT beams. For a typical hypofractionated spine treatment, 200 single DTS (6° arc) and 10 paired DTS (20° arc) images were acquired for each IMRT beam, providing in-plane and out-of-plane monitoring every 1.6 and 34.5 s, respectively. Mean ± standard deviation error in predicted position was −0.3 ± 0.2 mm, −0.1 ± 0.1 mm in-plane, and 0.2 ± 0.4 mm out-of-plane with rotational gantry, 0.8 ± 0.1 mm, −0.7 ± 0.3 mm in-plane and 1.1 ± 0.1 mm out-of-plane with translational source/detector.Acquiring 3D fiducial positions from kV-DTS during fixed gantry IMRT is technically feasible, and is capable of providing reliable guidance for intrafractional patient motion management.

Generation of hybrid sinograms for the recovery of kV-CT images with metal artifacts for helical tomotherapy.

The overall goal of this study is to restore kilovoltage computed tomography (kV-CT) images which are disfigured by patients' metal prostheses. By generating a hybrid sinogram that is a combination of kV and megavoltage (MV) projection data, the authors suggest a novel metal artifact-reduction (MAR) method that retains the image quality to match that of kV-CT and simultaneously restores the information of metal prostheses lost due to photon starvation. CT projection data contain information about attenuation coefficients and the total length of the attenuation. By normalizing raw kV projections with their own total lengths of attenuation, mean attenuation projections were obtained. In the same manner, mean density projections of MV-CT were obtained by the normalization of MV projections resulting from the forward projection of density-calibrated MV-CT images with the geometric parameters of the kV-CT device. To generate the hybrid sinogram, metal-affected signals of the kV sinogram were identified and replaced by the corresponding signals of the MV sinogram following a density calibration step with kV data. Filtered backprojection was implemented to reconstruct the hybrid CT image. To validate the authors' approach, they simulated four different scenarios for three heads and one pelvis using metallic rod inserts within a cylindrical phantom. Five inserts describing human body elements were also included in the phantom. The authors compared the image qualities among the kV, MV, and hybrid CT images by measuring the contrast-to-noise ratio (CNR), the signal-to-noise ratio (SNR), the densities of all inserts, and the spatial resolution. In addition, the MAR performance was compared among three existing MAR methods and the authors' hybrid method. Finally, for clinical trials, the authors produced hybrid images of three patients having dental metal prostheses to compare their MAR performances with those of the kV, MV, and three existing MAR methods. The authors compared the image quality and MAR performance of the hybrid method with those of other imaging modalities and the three MAR methods, respectively. The total measured mean of the CNR (SNR) values for the nonmetal inserts was determined to be 14.3 (35.3), 15.3 (37.8), and 25.5 (64.3) for the kV, MV, and hybrid images, respectively, and the spatial resolutions of the hybrid images were similar to those of the kV images. The measured densities of the metal and nonmetal inserts in the hybrid images were in good agreement with their true densities, except in cases of extremely low densities, such as air and lung. Using the hybrid method, major streak artifacts were suitably removed and no secondary artifacts were introduced in the resultant image. In clinical trials, the authors verified that kV and MV projections were successfully combined and turned into the resultant hybrid image with high image contrast, accurate metal information, and few metal artifacts. The hybrid method also outperformed the three existing MAR methods with regard to metal information restoration and secondary artifact prevention. The authors have shown that the hybrid method can restore the overall image quality of kV-CT disfigured by severe metal artifacts and restore the information of metal prostheses lost due to photon starvation. The hybrid images may allow for the improved delineation of structures of interest and accurate dose calculations for radiation treatment planning for patients with metal prostheses.

WE-AB-303-01: FEATURED PRESENTATION: A Dual-Detector Phase-Matched Digital Tomosynthesis (DTS) Imaging Scheme Using Aggregated KV and MV Projections for Intra-Treatment Lung Tumor Tracking

Purpose: To develop a dual-detector phase-matched DTS technique for continuous and fast intra-treatment lung tumor localization. Methods: Tumor localization accuracy of limited-angle DTS imaging is affected by low inter-slice resolution. The dual-detector DTS technique aims to overcome this limitation through combining orthogonally acquired beam's eye view MV projections and kV projections for intra-treatment DTS reconstruction and localization. To aggregate the kV and MV projections for reconstruction, the MV projections were linearly converted to synthesize corresponding kV projections. To further address the lung motion induced localization errors, this technique uses respiratory phase-matching to match the motion information between on-board DTS and reference DTS to offset the adverse effects of motion blurriness in tumor localization.A study was performed using the CIRS008A lung phantom to simulate different on-board target variation scenarios for localization. The intra-treatment kV and MV acquisition was achieved through the Varian TrueBeam Developer Mode. Four methods were compared for their localization accuracy: 1. the proposed dual-detector phase-matched DTS technique; 2. the single-detector phase-matched DTS technique; 3. the dual-detector 3D-DTS technique without phase-matching; and 4. the single-detector 3D-DTS technique without phase-matching. Results: For scan angles of 2.5°, 5°, 10°, 20° and 30°, the dual-detector phase-matched DTS technique localized the tumor with average(±standard deviations) errors of 0.4±0.3 mm, 0.5±0.3 mm, 0.6±0.2 mm, 0.9±0.4 mm and 1.0±0.3 mm, respectively. The corresponding values of single-detector phase-matched DTS technique were 4.0±2.5 mm, 2.7±1.1 mm, 1.7±1.2 mm, 2.2±0.9 mm and 1.5±0.8 mm, respectively. The values of dual-detector 3D-DTS technique were 6.2±1.7 mm, 6.3±1.2 mm, 5.3±1.3 mm, 2.0±2.2 mm and 1.5±0.5 mm, respectively. And the values of single-detector 3D-DTS technique were 9.7±8.9 mm, 9.8±8.8 mm, 10.0±9.7 mm, 3.9±2.7 mm and 2.2±1.3 mm, respectively. Conclusion: The dual-detector phase-matched DTS technique substantially improves the tumor localization accuracy, which can be applied to real-time intra-treatment lung tumor localization. The research was funded by the National Institutes of Health Grant No. R01-CA184173 and a grant from Varian Medical Systems.

TH‐EF‐BRB‐06: A Method to Synthesize Hybrid KV/MV Projections for Metal Artifact Corrections in CBCT

Purpose:Simultaneous utilization of kV and MV projections is a promising approach to correct metal artifacts in CBCT. However, due to inherent differences in attenuation, beam hardening, and scattering properties of kV and MV beams, it is challenging to employ kV and MV projections in the same CBCT set. To address these limitations, a novel method was developed to generate hybrid kV/MV projections, and to correct metal artifacts in CBCT images.Methods:First, a calibration dataset‐based method was developed to correct beam hardening and scatter in kV and MV projections, and to convert MV projections to kV‐equivalent attenuation maps. In the next stage, hybrid kV/MV projections were synthesized by weighted summation of corrected kV and MV projections; a user‐controlled, generalized logistic curve was utilized as the weighting function, and its shape determined the relative weight of kV and MV projections with respect to relative attenuation differences on a pixel‐by‐pixel basis.Results:Phantoms with integrated metal hardware were imaged using 125 kVp and 6 MV beams from a linac with on‐board KV imaging. CBCT images were reconstructed from KV‐only projections and from hybrid kV/MV projections. In kV‐only CBCT sets, metal hardware lead to severe artifacts and HU variations. HU values were underestimated by — 400±117 in a Catphan phantom with femoral implants. On the other hand, CBCT reconstructed from hybrid kV/MV projections exhibited significantly less metal artifacts, and the underestimation in HU values was reduced to — 100±73. Similar results were observed in all imaging experiments.Conclusion:Our method was successful in reducing metal artifacts, and improving CT number accuracy in CBCT images. Consistency of corrections appeared to be robust under diverse imaging conditions. The user‐controlled weighting function allowed fine tuning the balance between the level of desired metal artifact suppression and the preservation of soft tissue contrast in CBCT images.

WE-D-303-03: 3D Delivered Dose Assessment Using a 4DCT-Based Motion Model

Purpose: To develop a clinically feasible method of calculating actual delivered dose for patients with significant respiratory motion during the course of SBRT. Methods: This approach can be specified in three steps. (1) At planning stage, a patient-specific motion model is created from planning 4DCT using a principal components analysis (PCA) algorithm. (2) During treatment, 2D time-varying projection images (either kV or MV projections) are acquired, from which time-varying ‘fluoroscopic’ 3D images of the patient are reconstructed using the motion model. (3) A 3D dose distribution is computed for each timepoint in the set of 3D fluoroscopic images, from which the total effective 3D delivered dose is calculated by accumulating dose distributions onto a reference image. This approach was validated using two modified XCAT phantoms with lung tumors and different respiratory motions. The estimated doses were compared to the dose that would be calculated for 4DCT-based planning and to the actual delivered dose that was calculated using “ground truth” XCAT phantoms. The approach was also tested using one set of patient data. Results: For the XCAT phantom with a regular breathing pattern, the errors in D95 are 0.11% and 0.83% respectively for kV and MV reconstructions compared to the ground truth, which is comparable to 4DCT (0.093%). For the XCAT phantom with an irregular breathing pattern, the errors are 0.81% and 1.75% for kV and MV reconstructions, both better than that of 4DCT (4.01%). For real patient, the dose estimation is clinically reasonable and demonstrates differences between 4DCT and MV reconstruction-based estimation. Conclusions: Using kV or MV projections, the proposed approach is able to assess delivered doses for all respiratory phases during treatment. Compared to the 4DCT dose, the dose estimation using reconstructed 3D fluoroscopic images is as good for regular respiratory pattern and better for irregular respiratory pattern.

WE-AB-303-03: A Scheme for Real-Time, Marker-Less and Low-Dose Tumor Tracking Using Scattering Imaging: Monte Carlo Simulation Study

Purpose: In-treatment tumor localization is critical for the management of tumor motion in lung cancer radiotherapy. Conventional tumor-tracking methods using a kV or MV x-ray projection has limited contrast. To facilitate real-time, marker-less and low-dose in-treatment image tumor tracking, we propose a novel scheme using Compton scatter imaging. This study reports Monte Carlo (MC) simulations on this scheme for the purpose of proof-of-principle. Methods: A slit x-ray beam along the patient superior-inferior (SI) direction is directed to the patient, intersecting the patient lung at a 2D plane containing majority part of the tumor motion trajectory. X-ray photons are scattered due to Compton effect from this plane, which are spatially collimated by, e.g., a pinhole, on one side of the plane and then captured by a detector behind it. The captured image, after correcting for x-ray attenuation and scatter angle variation, reflects the electron density, which allows visualization of the instantaneous anatomy on this plane. We performed MC studies on a phantom and a patient case for the initial test of this proposed method. Results: In the phantom case, the contrast-resolution calculated using tumor/lung as foreground/background for kV fluoroscopy, cone-beam CT, and scattering image were 0.0625, 0.6993, and 0.5290, respectively. In the patient case, tumor motion can be clearly observed in the scatter images. Compared to fluoroscopy, scattering imaging also significantly reduced imaging dose because of its narrower beam design. Conclusion: MC simulation studies demonstrated the potential of the proposed scheme in terms of capturing the instantaneous anatomy of a patient on a 2D plane. Clear visualization of the tumor will probably facilitate ‘marker-less’ and ‘real-time’ tumor tracking with low imaging dose. NIH (1R01CA154747-01, 1R21CA178787-01A1 and 1R21EB017978-01A1)

WE-AB-303-08: Direct Lung Tumor Tracking Using Short Imaging Arcs

Purpose: Most current tumor tracking technologies rely on implanted markers, which suffer from potential toxicity of marker placement and mis-targeting due to marker migration. Several markerless tracking methods have been proposed: these are either indirect methods or have difficulties tracking lung tumors in most clinical cases due to overlapping anatomies in 2D projection images. We propose a direct lung tumor tracking algorithm robust to overlapping anatomies using short imaging arcs. Methods: The proposed algorithm tracks the tumor based on kV projections acquired within the latest six-degree imaging arc. To account for respiratory motion, an external motion surrogate is used to select projections of the same phase within the latest arc. For each arc, the pre-treatment 4D cone-beam CT (CBCT) with tumor contours are used to estimate and remove the contribution to the integral attenuation from surrounding anatomies. The position of the tumor model extracted from 4D CBCT of the same phase is then optimized to match the processed projections using the conjugate gradient method. The algorithm was retrospectively validated on two kV scans of a lung cancer patient with implanted fiducial markers. This patient was selected as the tumor is attached to the mediastinum, representing a challenging case for markerless tracking methods. The tracking results were converted to expected marker positions and compared with marker trajectories obtained via direct marker segmentation (ground truth). Results: The root-mean-squared-errors of tracking were 0.8 mm and 0.9 mm in the superior-inferior direction for the two scans. Tracking error was found to be below 2 and 3 mm for 90% and 98% of the time, respectively. Conclusions: A direct lung tumor tracking algorithm robust to overlapping anatomies was proposed and validated on two scans of a lung cancer patient. Sub-millimeter tracking accuracy was observed, indicating the potential of this algorithm for real-time guidance applications.

TU‐EF‐204‐06: Waveform Measurements On a Fast‐KV Switching CT System

Purpose:To evaluate the timing and shape of the kilovoltage (kV) waveform for a fast‐kV switching dual‐energy CT system using a non‐invasive method.Methods:A kV divider with mobile filter pack was connected to a digital oscilloscope and scanned using a fast‐kV switching dual‐energy CT system in service mode. The tube was stationary, so exposure time was equivalent to rotation time. In this system, the generator switches quickly between two nominal voltages of 80kV and 140kV. The predicted waveform period was calculated given 1968 projections per exposure/rotation. Two Gemstone Spectral Imaging (GSI) service settings, GSI‐2 and GSI‐3, with exposure times of 0.8sec and 0.5sec, respectively, were analyzed. The data were binned into high (&gt;110kV) and low (&lt;110kV) projection regions. Waveform periods were calculated using the intersections of the waveform with 110kV. The waveform was analyzed for period accuracy, minimum and maximum kV, and average kV of the high and low kV projection regions. The effective spectral separation was the difference between the average kVs of the low and high projection regions.Results:The shape of the waveform was neither square nor sinusoidal. The minimum and maximum kVs were 81.6kV and 139.6kV for GSI‐2 (0.8sec) and 84.8kV and 131.6kV for GSI‐3 (0.5sec), which were within 6% of the nominal kVs. GSI‐2 and GSI‐3 had average waveform periods of 0.8134msec and 0.5064msec, respectively, within 0.5% of the calculated periods. The average kV for the high and low projections yielded effective spectral separations of 38.3kV (89.2kV to 127.5kV) for GSI‐2 (0.8sec) and 27.2kV (93.7kV to 121.4kV) for GSI‐3 (0.5sec).Conclusion:A method for analyzing the kV waveform of a fast‐kV switching x‐ray tube was demonstrated. The pseudo‐sinusoidal shape of the kV waveform lowers spectral separation compared to that of an ideal square wave, which may impact the quality of dual‐energy CT material decomposition.This research was conducted at the MD Anderson Center for Advanced Biomedical Imaging in‐part with equipment support from General Electric Healthcare

SU‐D‐207‐05: Real‐Time Intrafractional Motion Tracking During VMAT Delivery Using a Conventional Elekta CBCT System

Purpose:Real‐time kV projection streaming capability has become recently available for Elekta XVI version 5.0. This study aims to investigate the feasibility and accuracy of real‐time fiducial marker tracking during CBCT acquisition with or without simultaneous VMAT delivery using a conventional Elekta linear accelerator.Methods:A client computer was connected to an on‐board kV imaging system computer, and receives and processes projection images immediately after image acquisition. In‐house marker tracking software based on FFT normalized cross‐correlation was developed and installed in the client computer. Three gold fiducial markers with 3 mm length were implanted in a pelvis‐shaped phantom with 36 cm width. The phantom was placed on a programmable motion platform oscillating in anterior‐posterior and superior‐inferior directions simultaneously. The marker motion was tracked in real‐time for (1) a kV‐only CBCT scan with treatment beam off and (2) a kV CBCT scan during a 6‐MV VMAT delivery. The exposure parameters per projection were 120 kVp and 1.6 mAs. Tracking accuracy was assessed by comparing superior‐inferior positions between the programmed and tracked trajectories.Results:The projection images were successfully transferred to the client computer at a frequency of about 5 Hz. In the kV‐only scan, highly accurate marker tracking was achieved over the entire range of cone‐beam projection angles (detection rate / tracking error were 100.0% / 0.6±0.5 mm). In the kV‐VMAT scan, MV‐scatter degraded image quality, particularly for lateral projections passing through the thickest part of the phantom (kV source angle ranging 70°‐110° and 250°‐290°), resulting in a reduced detection rate (90.5%). If the lateral projections are excluded, tracking performance was comparable to the kV‐only case (detection rate / tracking error were 100.0% / 0.8±0.5 mm).Conclusion:Our phantom study demonstrated a promising Result for real‐time motion tracking using a conventional Elekta linear accelerator. MV‐scatter suppression is needed to improve tracking accuracy during MV delivery.This research is funded by Motion Management Research Grant from Elekta.

WE-AB-303-10: The Use of On-Board KV Imaging During Respiratory-Gated VMAT Delivery to Determine the Correlation and Phase Shift Between External Marker Motion and Internal Tumour Motion

Purpose: To determine the correlation between internal tumor motion and external surrogate motion and any potential phase shifts during respiratory-gated VMAT delivery using on-board kV imaging. Methods: A respiratory motion phantom was used to trace the motion of an internal target, embedded with a 3 cm moving target, and an external surrogate. On-board kV projections were acquired at 11 frames per second during VMAT delivery on a TrueBeam v.1.6 (Varian Medical Systems, Palo Alto, CA) operating in developer mode. Full motion encompassed treatment delivery using sinusoidal motion with 4 second period and 2 cm peak-to-peak amplitude, and real-patient breathing (RPB) motion was programmed to the phantom. Respiratory-gated treatment deliveries (30–60% gating window) on five additional RPB waveforms were programmed to investigate correlation within the gating window. To investigate the capability of during treatment kV imaging to detect phase shifts during gated delivery, a second phantom with independent motion was used. Eight controlled phase shifts in intervals of 0.4 seconds were added to the sinusoidal motion. For each phase shift, the Pearson linear correlation coefficient statistical test was performed for each set of respiratory traces. The phase shift was calculated by shifting the external trace, in time, until maximum correlation was obtained. Results: A strong internal and external correlation was obtained for both the free-breathing (sinusoidal R2=0.993 and RPB R2=0.990) and respiratory-gated cases (range from R2=0.986–0.996). There was no significant difference between the programmed shift and that acquired using on-board kV imaging (p=0.899, R-squared = 0.997). Conclusion: During treatment kV imaging has the capability to verify intrafractional anatomical position. It can be accurately used as a tool to quantify internal and external correlation, and determine phase shifts within the gating window, and ultimately, the accuracy of respiratory-gating treatment delivery.

WE-D-303-04: 4DCBCT-Based Dose Assessment for SBRT Lung Cancer Treatment

Purpose: To develop a 4DCBCT-based dose assessment method for calculating actual delivered dose for patients with significant respiratory motion during the course of SBRT or anatomical changes between treatment days. Methods: To address the limitation of 4DCT-based dose assessment, we propose to calculate the delivered dose using time varying (‘fluoroscopic’) 3D patient images generated from a 4DCBCT-based motion model. The method includes four steps: (1) before each treatment, 4DCBCT data is acquired with the patient in treatment position, based on which a patient-specific motion model is created using a principal components analysis (PCA) algorithm. (2) During treatment, 2D time-varying kV projection images are continuously acquired, from which time-varying ‘fluoroscopic’ 3D images of the patient are reconstructed using the motion model. (3) Lateral truncation artifacts are corrected using planning 4DCT images. (4) the 3D dose distribution is computed for each timepoint in the set of 3D fluoroscopic images, from which the total effective 3D delivered dose is calculated by accumulating deformed dose distributions. This approach is validated using six modified XCAT phantoms with lung tumors and different respiratory motions derived from patient data. The estimated doses are compared to that calculated using ground-truth XCAT phantoms. Results: For each XCAT phantom, the delivered tumor dose values generally follow the same trend as that of the ground truth and at most timepoints the difference is less than 5%. For the overall delivered dose, the normalized error of calculated 3D dose distribution is generally less than 3% and the tumor D95 error is less than 1.5%. Conclusions: XCAT phantom studies indicate the potential of the proposed method to accurately estimate 3D tumor dose distributions for SBRT lung treatment based on 4DCBCT imaging and motion modeling. Further research is necessary to investigate its performance for clinical patients data.

3D delivered dose assessment using a 4DCT-based motion model.

The purpose of this work is to develop a clinically feasible method of calculating actual delivered dose distributions for patients who have significant respiratory motion during the course of stereotactic body radiation therapy (SBRT). A novel approach was proposed to calculate the actual delivered dose distribution for SBRT lung treatment. This approach can be specified in three steps. (1) At the treatment planning stage, a patient-specific motion model is created from planning 4DCT data. This model assumes that the displacement vector field (DVF) of any respiratory motion deformation can be described as a linear combination of some basis DVFs. (2) During the treatment procedure, 2D time-varying projection images (either kV or MV projections) are acquired, from which time-varying "fluoroscopic" 3D images of the patient are reconstructed using the motion model. The DVF of each timepoint in the time-varying reconstruction is an optimized linear combination of basis DVFs such that the 2D projection of the 3D volume at this timepoint matches the projection image. (3) 3D dose distribution is computed for each timepoint in the set of 3D reconstructed fluoroscopic images, from which the total effective 3D delivered dose is calculated by accumulating deformed dose distributions. This approach was first validated using two modified digital extended cardio-torso (XCAT) phantoms with lung tumors and different respiratory motions. The estimated doses were compared to the dose that would be calculated for routine 4DCT-based planning and to the actual delivered dose that was calculated using "ground truth" XCAT phantoms at all timepoints. The approach was also tested using one set of patient data, which demonstrated the application of our method in a clinical scenario. For the first XCAT phantom that has a mostly regular breathing pattern, the errors in 95% volume dose (D95) are 0.11% and 0.83%, respectively for 3D fluoroscopic images reconstructed from kV and MV projections compared to the ground truth, which is clinically comparable to 4DCT (0.093%). For the second XCAT phantom that has an irregular breathing pattern, the errors are 0.81% and 1.75% for kV and MV reconstructions, both of which are better than that of 4DCT (4.01%). In the case of real patient, although it is impossible to obtain the actual delivered dose, the dose estimation is clinically reasonable and demonstrates differences between 4DCT and MV reconstruction-based dose estimates. With the availability of kV or MV projection images, the proposed approach is able to assess delivered doses for all respiratory phases during treatment. Compared to the planning dose based on 4DCT, the dose estimation using reconstructed 3D fluoroscopic images was as good as 4DCT for regular respiratory pattern and was a better dose estimation for the irregular respiratory pattern.

An accurate algorithm to match imperfectly matched images for lung tumor detection without markers.

In order to locate lung tumors on kV projection images without internal markers, digitally reconstructed radiographs (DRRs) are created and compared with projection images. However, lung tumors always move due to respiration and their locations change on projection images while they are static on DRRs. In addition, global image intensity discrepancies exist between DRRs and projections due to their different image orientations, scattering, and noises. This adversely affects comparison accuracy. A simple but efficient comparison algorithm is reported to match imperfectly matched projection images and DRRs. The kV projection images were matched with different DRRs in two steps. Preprocessing was performed in advance to generate two sets of DRRs. The tumors were removed from the planning 3D CT for a single phase of planning 4D CT images using planning contours of tumors. DRRs of background and DRRs of tumors were generated separately for every projection angle. The first step was to match projection images with DRRs of background signals. This method divided global images into a matrix of small tiles and similarities were evaluated by calculating normalized cross‐correlation (NCC) between corresponding tiles on projections and DRRs. The tile configuration (tile locations) was automatically optimized to keep the tumor within a single projection tile that had a bad matching with the corresponding DRR tile. A pixel‐based linear transformation was determined by linear interpolations of tile transformation results obtained during tile matching. The background DRRs were transformed to the projection image level and subtracted from it. The resulting subtracted image now contained only the tumor. The second step was to register DRRs of tumors to the subtracted image to locate the tumor. This method was successfully applied to kV fluoro images (about 1000 images) acquired on a Vero (BrainLAB) for dynamic tumor tracking on phantom studies. Radiation opaque markers were implanted and used as ground truth for tumor positions. Although other organs and bony structures introduced strong signals superimposed on tumors at some angles, this method accurately located tumors on every projection over 12 gantry angles. The maximum error was less than 2.2 mm, while the total average error was less than 0.9 mm. This algorithm was capable of detecting tumors without markers, despite strong background signals.PACS numbers: 87.57.cj, 87.57.cp87.57.nj, 87.57.np, 87.57.Q‐, 87.59.bf, 87.63.lm

3D fluoroscopic image estimation using patient-specific 4DCBCT-based motion models

3D fluoroscopic images represent volumetric patient anatomy during treatment with high spatial and temporal resolution. 3D fluoroscopic images estimated using motion models built using 4DCT images, taken days or weeks prior to treatment, do not reliably represent patient anatomy during treatment. In this study we developed and performed initial evaluation of techniques to develop patient-specific motion models from 4D cone-beam CT (4DCBCT) images, taken immediately before treatment, and used these models to estimate 3D fluoroscopic images based on 2D kV projections captured during treatment. We evaluate the accuracy of 3D fluoroscopic images by comparison to ground truth digital and physical phantom images. The performance of 4DCBCT-based and 4DCT-based motion models are compared in simulated clinical situations representing tumor baseline shift or initial patient positioning errors. The results of this study demonstrate the ability for 4DCBCT imaging to generate motion models that can account for changes that cannot be accounted for with 4DCT-based motion models. When simulating tumor baseline shift and patient positioning errors of up to 5 mm, the average tumor localization error and the 95th percentile error in six datasets were 1.20 and 2.2 mm, respectively, for 4DCBCT-based motion models. 4DCT-based motion models applied to the same six datasets resulted in average tumor localization error and the 95th percentile error of 4.18 and 5.4 mm, respectively. Analysis of voxel-wise intensity differences was also conducted for all experiments. In summary, this study demonstrates the feasibility of 4DCBCT-based 3D fluoroscopic image generation in digital and physical phantoms and shows the potential advantage of 4DCBCT-based 3D fluoroscopic image estimation when there are changes in anatomy between the time of 4DCT imaging and the time of treatment delivery.

Markerless tracking of small lung tumors for stereotactic radiotherapy.

(1) To validate retrospective markerless tracking software for small lung tumors by comparing tracked motion in 4-dimensional planning computed tomography (4DCT) derived kV projection images and known tumor motion in the same 4DCT. (2) To evaluate variability of tumor motion using kV projection images from cone-beam computed tomography (CBCT) scans acquired on different days. Nonclinical tumor tracking software (TTS) used a normalized cross correlation algorithm to track the tumor on enhanced kV projection images (e.g., from a CBCT scan). The reference dataset consisted of digitally reconstructed radiographs (DRRs) from one phase of a planning 4DCT. TTS matches two in-plane coordinates and obtains the out-of-plane coordinate by triangulating with match results from other projections. (1) To validate TTS, tracking results were compared with known 4DCT tumor motion for two patients (A and B). Projection images (1 image/1°) were digitally reconstructed for each 4DCT phase. From these, kV projection series were composed simulating full breathing cycles every 20° of gantry rotation [breathing period = 20°/(6°/s) = 3.33 s]. Reference templates were 360 "tumor enhanced" DRRs from the 4DCT expiration phase. TTS-derived tumor motion was compared to known tumor motion on 4DCT. (2) For five patients, TTS-assessed motion during clinical CBCT acquisition was compared with motion on the planning 4DCT, and the motion component in the Y (cranio-caudal)-direction was compared with the motion of an external marker box (RPM, real-time position management). (1) Validation results: TTS for case A (tumor 6.2 cm(3), 32 mm axial diameter) over 360° showed mean motion X (medial-lateral) = 3.4, Y = 11.5, and Z (ventral-dorsal) = 4.9 mm (1 SD < 1.0 mm). Corresponding 4DCT motion was X = 3.1, Y = 11.3, and Z = 5.1 mm. Correlation coefficients between TTS tumor motion and displacement of the tumor's center of mass (CoM) on 4DCT were 0.64, 0.96, and 0.82 (X, Y, and Z, respectively). For case B (4.1 cm(3), 20 mm diameter), due to temporarily decreased tumor visibility preventing TTS from resolving the tumor, robust tracking data were only available between angles 300°-40° and 120°-220°. Mean motion according to TTS was X = 2.0, Y = 7.7, and Z = 8.2 mm (1 SD < 0.9 mm). Tumor motion on 4DCT was X = 1.8, Y = 7.6, and Z = 9.5 mm and correlation coefficients between TTS motion and CoM displacement were 0.59, 0.95, and 0.93 (X, Y, and Z, respectively). (2) CLINICAL RESULTS: TTS revealed a mean intrafraction variation in tumor motion in Y-direction of >2.0 mm (1 SD) in four of five patients. In addition, clinical tumor motion amplitude differed from that seen on planning 4DCT. Internal and external structures that create abrupt density change (e.g., table-top edge, interface between lung/mediastinum and lung/heart) were observed to prevent 360° tracking of the tumor. Correlation coefficients between TTS motion in the Y-direction and the RPM signal (22 observations) ranged from 0.78 to 0.96. In 2D, 241 TTS matches at end-inspiration and end-expiration were visually validated: mean difference was 0.8 mm (SD = 0.7) for both. TTS can track small lung tumors if these are visible in kV projections. A 4DCT dataset can be used to validate kV tracking of moving targets. TTS and 4DCT displacement agreed to within 2 mm. TTS and RPM motion were closely associated but tumor motion during CBCT can vary from the planning 4DCT.

Analysis on the Positional Accuracy of the Non-orthogonal Two-pair kV Imaging Systems for Real-time Tumor Tracking Using XCAT

In this study, we aim to design the architecture of the kV imaging system for tumor tracking in the dual-head gantry system and analyze its accuracy by simulations. We established mathematical formulas and algorithms to track the tumor position with the two-pair kV imaging systems when they are in the non-orthogonal positions. The algorithms have been designed in the homogeneous coordinate framework and the position of the source and the detector coordinates are used to estimate the tumor position. 4D XCAT (4D extended cardiac-torso) software was used in the simulation to identify the influence of the angle between the two-pair kV imaging systems and the resolution of the detectors to the accuracy in the position estimation. A metal marker fiducial has been inserted in a numerical human phantom of XCAT and the kV projections were acquired at various angles and resolutions using CT projection software of the XCAT. As a result, a positional accuracy of less than about 1mm was achieved when the resolution of the detector is higher than 1.5 mm/pixel and the angle between the kV imaging systems is approximately between 90 o and 50 o . When the resolution is lower than 1.5 mm/pixel, the positional errors were higher than 1mm and the error fluctuation by the angles was greater. The resolution of the detector was critical in the positional accuracy for the tumor tracking and determines the range for the acceptable angle range between the kV imaging systems. Also, we found that the positional accuracy analysis method using XCAT developed in this study is highly useful and will be a invaluable tool for further refined design of the kV imaging systems for tumor tracking systems.