Andrew G. Jeung

Four‐dimensional cone beam CT with adaptive gantry rotation and adaptive data sampling

We have developed a new four-dimensional cone beam CT (4D-CBCT) on a Varian image-guided radiation therapy system, which has radiation therapy treatment and cone beam CT imaging capabilities. We adapted the speed of gantry rotation time of the CBCT to the average breath cycle of the patient to maintain the same level of image quality and adjusted the data sampling frequency to keep a similar level of radiation exposure to the patient. Our design utilized the real-time positioning and monitoring system to record the respiratory signal of the patient during the acquisition of the CBCT data. We used the full-fan bowtie filter during data acquisition, acquired the projection data over 200 deg of gantry rotation, and reconstructed the images with a half-scan cone beam reconstruction. The scan time for a 200-deg gantry rotation per patient ranged from 3.3 to 6.6 min for the average breath cycle of 3-6 s. The radiation dose of the 4D-CBCT was about 1-2 times the radiation dose of the 4D-CT on a multislice CT scanner. We evaluated the 4D-CBCT in scanning, data processing and image quality with phantom studies. We demonstrated the clinical applicability of the 4D-CBCT and compared the 4D-CBCT and the 4D-CT scans in four patient studies. The contrast-to-noise ratio of the 4D-CT was 2.8-3.5 times of the contrast-to-noise ratio of the 4D-CBCT in the four patient studies.

Evaluation of IsoCal geometric calibration system for Varian linacs equipped with on-board imager and electronic portal imaging device imaging systems

The purpose of this study is to evaluate the accuracy and reproducibility of the IsoCal geometric calibration system for kilovoltage (kV) and megavoltage (MV) imagers on Varian C‐series linear accelerators (linacs). IsoCal calibration starts by imaging a phantom and collimator plate using MV images with different collimator angles, as well as MV and kV images at different gantry angles. The software then identifies objects on the collimator plate and in the phantom to determine the location of the treatment isocenter and its relation to the MV and kV imager centers. It calculates offsets between the positions of the imaging panels and the treatment isocenter as a function of gantry angle and writes a correction file that can be applied to MV and kV systems to correct for those offsets in the position of the panels. We performed IsoCal calibration three times on each of five Varian C‐series linacs, each time with an independent setup. We then compared the IsoCal calibrations with a simplified Winston‐Lutz (WL)‐based system and with a Varian cubic phantom (VC)‐based system. The maximum IsoCal corrections ranged from 0.7 mm to 1.5 mm for MV and 0.9 mm to 1.8 mm for kV imagers across the five linacs. The variations in the three calibrations for each linac were less than 0.2 mm. Without IsoCal correction, the WL results showed discrepancies between the treatment isocenter and the imager center of 0.9 mm to 1.6 mm (for the MV imager) and 0.5 mm to 1.1 mm (for the kV imager); with IsoCal corrections applied, the differences were reduced to 0.2 mm to 0.6 mm (MV) and 0.3 mm to 0.6 mm (kV) across the five linacs. The VC system was not as precise as the WL system, but showed similar results, with discrepancies of less than 1.0 mm when the IsoCal corrections were applied. We conclude that IsoCal is an accurate and consistent method for calibration and periodic quality assurance of MV and kV imaging systems. PACS numbers: 87.55.Qr, 87.56.Fc

SU‐FF‐I‐04: A Fast Variable‐Intensity Ring Suppression Algorithm

Purpose: Gain drifts and nonlinearities in amorphous silicon flat‐panel x‐ray detectors can produce ring artifacts in reconstructed cone‐beam computed tomography(CBCT)images. We have found that the magnitude of these artifacts can exceed 50 HU in clinical situations, and that the intensity of a given ring may not be uniform throughout an image. In some cases (e.g. half‐fan pelvic scans), discrete arcs may be produced. The goal of this study was to develop a post‐processing algorithm to efficiently suppress such variable‐intensity rings in axial slices. Method and Materials: Our approach builds upon the work of Sijbers and Postnov who showed that constant‐intensity rings can be estimated via radial median filtering of the input image after its transformation to polar coordinates. To characterize variable‐intensity rings and arcs, we developed a 2‐D estimation technique that uses a combination of row‐based (radial) and column‐based (angular) filters operating in the polar domain. The 2‐D estimates were transformed back to Cartesian space for subtraction from the original image. The new algorithm was implemented in C++ and tested on clinical and phantom CBCTimages acquired using a Varian 4030CB detector.Results: Correction times (3.2GHz Intel Pentium4 processor), including coordinate transformations, averaged 55 msec/slice for 512×512 matrix sizes. Rings and arcs were reduced in intensity by more than an order of magnitude to levels well below the background noise intensity. By subtracting ring estimates in Cartesian space, the polar matrix size could be reduced without sacrificing spatial resolution in the final image. This permitted for a 4× reduction in execution time compared to the original Sijbers‐Postnov approach where subtraction occurs in polar space. Conclusion: The Sijbers‐Postnov algorithm ring suppression algorithm was modified to provide improved image quality and fast execution times suitable for clinical implementation. Conflict of Interest: Funding provided by Varian Medical Systems.

Detection and localization of radiotherapy targets by template matching

Radio opaque fiducials are implanted in tumors for the purpose of tracking the target motion using X-ray projections during radiation therapy dose delivery. In this paper we describe and evaluate a novel method based on template matching for detection and localization of arbitrary shaped fiducials. Segmentation methods are not adequate for these fiducials because their appearance in online X-ray projections can vary greatly as a function of imaging angle. The algorithm is based on using the planning CT image to generate templates that correspond to the imaging angles of the online images. We demonstrate successful tracking of complex shape fiducials in clinical images of lung and abdomen. We also validate the algorithm by comparing the results with a segmentation approach for one case in which the fiducials could be tracked by both methods. We also show how by adaptive thresholding of the match scores, we can control the false detection rate.

Optimal parameters for clinical implementation of breast cancer patient setup using Varian DTS software

Digital tomosynthesis (DTS) was evaluated as an alternative to cone‐beam computed tomography (CBCT) for patient setup. DTS is preferable when there are constraints with setup time, gantry‐couch clearance, and imaging dose using CBCT. This study characterizes DTS data acquisition and registration parameters for the setup of breast cancer patients using nonclinical Varian DTS software. DTS images were reconstructed from CBCT projections acquired on phantoms and patients with surgical clips in the target volume. A shift‐and‐add algorithm was used for DTS volume reconstructions, while automated cross‐correlation matches were performed within Varian DTS software. Triangulation on two short DTS arcs separated by various angular spread was done to improve 3D registration accuracy. Software performance was evaluated on two phantoms and ten breast cancer patients using the registration result as an accuracy measure; investigated parameters included arc lengths, arc orientations, angular separation between two arcs, reconstruction slice spacing, and number of arcs. The shifts determined from DTS‐to‐CT registration were compared to the shifts based on CBCT‐to‐CT registration. The difference between these shifts was used to evaluate the software accuracy. After findings were quantified, optimal parameters for the clinical use of DTS technique were determined. It was determined that at least two arcs were necessary for accurate 3D registration for patient setup. Registration accuracy of 2 mm was achieved when the reconstruction arc length was > 5° for clips with HU ≥ 1000°; larger arc length (≥ 8°) was required for very low HU clips. An optimal arc separation was found to be ≥ 20° and optimal arc length was 10°. Registration accuracy did not depend on DTS slice spacing. DTS image reconstruction took 10–30 seconds and registration took less than 20 seconds. The performance of Varian DTS software was found suitable for the accurate setup of breast cancer patients. Optimal data acquisition and registration parameters were determined. PACS numbers: 87.57.‐s, 87.57.nf, 87.57.nj

WE‐C‐T‐617‐10: Geometry Calibration of An On‐Board KV Imaging System

Purpose: We have developed a method of geometrically calibrating the MV and kV radiographic isocenters of a Varian linear accelerator equipped with an on‐board kV imaging system. Method and Materials: The calibration uses a cylindrical phantom containing 16 BBs placed in a spiral pattern and ∼400–600 radiographs of the phantom acquired using both kV and MV imaging systems. Since the phantom has a known geometry, the 7 degrees of freedom (DOF) needed to fit the predicted projections of the BBs to the measured projections can be calculated. During MV imaging a partial transmission plate is placed in the accelerator wedge slot to project a reference shape onto the EPIDimages of the phantom. The reference shape accounts for motion of the EPID and generates the radiographic isocenter for the MV x‐ray beam alone. Geometry correction parameters can be calculated for the kV imaging system that map the center of radiographicimages or CBCT projections to the radiographic isocenter of the MV beam. The method has been tested with linear accelerator and lab bench imaging systems. Results: BB identification accuracy is ∼0.3 pixels and is limited by noise in the images. While the method is insensitive to changes in SID (0.3 pixels ∼= 2.5 cm SID change), the method can measure known in‐plane imager displacements to an accuracy of better than 0.1 mm. Imager displacements versus gantry angle are small − 0.75×0.50 mm. CBCT reconstruction using parameters from this method results in spatial frequencies as high as 18 lp/cm and adjustments in parameters of ±0.1 mm reduce the reconstructed spatial resolution.Conclusion: Our method to match MV and kV radiographic isocenters for on‐board imaging systems shows that accelerator flex is quite modest but that the calibration method can improve the spatial resolution of the CBCT reconstructions.

WE‐D‐204B‐08: Tracking 3D Trajectory of Internal Markers Using Radiographic Sequential Stereo Imaging: Estimation of Breathing Motion

Purpose: To show that Sequential Stereo Algorithm can track internal markers that move with respiration using a single imager and to determine the accuracy and required number of images (dose) for typical breathing trajectories. Method and Materials: The Sequential Stereo Algorithm, developed at Varian Medical Systems, Palo Alto, CA (patent pending) uses images acquired at different times and from different directions to track point targets moving in 3D. It performs best for trajectories that are approximately repeating in space, but not necessarily periodic; no prior knowledge or additional surrogate signal is needed. One application is to track radio opaque markers moving by respiration in the lung or liver, using kV images acquired on a gantry‐based onboard imager while the gantry is rotating. We evaluated the algorithm using three different studies: 1) The Tracking concept was proven with projections from clinical CBCT scans (used retrospectively) exhibiting respiration motion. 2) Tracking accuracy was measured in a phantom experiment by kV imaging of a marker that moved on a known trajectory using a computer controlled 3‐axis stage. 3) The sensitivity of accuracy to imaging frequency was determined by simulation study involving a virtual moving marker. Actual patient trajectories that included irregular breathing were driving both the stage and the simulated virtual point. Results: Using 3.5fps imaging over 30 Sec and 180° gantry rotation the 3D tracking error at measurement points was 0.65mm RMS (1.83mm max) for the phantom imaging study, and 0.75mm RMS (1.64mm max) for the simulation. For an irregular breathing interval the error was 1.01mm RMS (3.64mm max). Simulation showed that error remains consistently under 1mm for imaging frequency greater than 3fps. Conclusion: Sequential Stereo can accurately track internal markers that move with respiration in 3D, and may potentially be useful in real‐time adaptive radiotherapy techniques.

SU‐E‐J‐57: Clinical Protocol for DTS‐Based APBI Setup: Optimal Data Acquisition, Reconstruction and Registration Parameters Using Varian DTS Software

Purpose: Digital tomosynthesis (DTS) was evaluated as an alternative to CBCT for minimizing possibility of collisions and reducing imaging dose. While feasibility of DTS was demonstrated for APBI patient setup, clinical implementation has not been optimized for this technique. This work characterizes data acquisition/registration parameters and establishes clinical protocol for accurate setup using: Varian OBI system for data acquisition; and non‐clinical Varian DTS software for DTS reconstruction/registration for APBI patient setup. Methods: Backprojection‐and‐deblurring algorithms were used for DTS volume reconstructions. DTS volume registrations were done manually and automatically (cross‐correlation). Subsequent triangulation on two short DTS arcs was done to improve registration accuracy. Software performance was evaluated on a breast phantom and nine breast cancer patients, under an IRB‐approved protocol. Parameters investigated include arc lengths, arc orientations, number of arcs, reconstruction slice spacing and other limiting factors relevant to clinical practice. Shifts determined from the registration of DTS volumes were compared to the shifts based on registration between planning CT and CBCT. The difference between these shifts was used to evaluate the software performance and accuracy. The findings were quantified and optimal parameters for clinical use of DTS technique were determined. Results: At least two arcs were necessary for accurate setup evaluation. Registration accuracy of 2 mm was achieved when reconstruction arc length was > 5 deg for clips with HU>1000; larger arc length (> 8 deg) was required for low HU clips. Optimal arc separation was found to be > 20 deg. Optimal arc length was determined to be 8–10 deg. No dependence on DTS slice spacing was found. Time required for DTS reconstruction was 10s– 45 s and it was less than 20s for registration. Conclusions: Optimal data acquisition/registration parameters were determined for DTS imaging utilized for APBI patient setup, and performance of the software was objectively quantified. This study is supported by grant from Varian Medical Systems Inc. and Kayes grant

SU‐GG‐J‐81: Geometric Accuracy of Imaging Systems on Trilogy MX Using an Automated Geometric Test Tool

Purpose: To characterize the geometric accuracy of the imaging systems of a newly designed accelerator (Trilogy MX, Varian Medical Systems, Palo Alto, CA) using a fully automated, geometric calibration tool. Methods: A geometric calibration tool ‐ consisting of a phantom containing 16 tungstencarbide ball‐bearings and a MV collimator insert with a central pin ‐ has been used to characterize the geometric coincidence of MV and kV imaging equipment with the radiation isocenter of Trilogy MX. Three sets of data are acquired: MV images at multiple MV collimator rotations, and both MV and kV images at multiple gantry angles. The MV collimatorimages are used to identify the central axis of the MV beam, and the MV images at different gantry angles are then used to identify the MV radiation isocenter. Calculations then determine the corrections needed to move both the MV and kV imagers so that their central pixels exactly align with the projection of the radiation isocenter. The Trilogy MX employs a new control system so corrections can be downloaded to the control system and used to physically correct the imager positions while the gantry is being rotated. The positions of the imagers have been verified using the geometric calibration tool and a Winston‐Lutz test. Results: Coincidence between the MV isocenter ‐ measured using the geometric tool ‐ and the central imager pixels of both imagers was less than ±0.1mm. The coincidence between the central pixels of both imagers and the projection of the Winston‐Lutz BB was better than ±0.3mm. Conclusions: An automated tool for calibrating the geometry of MV and kV imaging equipment has been developed and tested. The tool gives similar results to conventional Winston‐Lutz measurements. The geometric accuracy of the imaging systems of Trilogy MX is better than ±0.3mm.

WE‐A‐134‐07: RapidTrack: A Versatile Tool for Template‐Based Target Tracking During Radiotherapy

PURPOSE To investigate the accuracy that can be obtained by using various template-based methods to track target motion from radiographic projections during radiation therapy dose delivery. METHODS We have developed a non-clinical software suite (RapidTrack) in two parts to accomplish template-based tracking. RapidTrack-Planning is an application used to generate templates from plan CT. These templates are oriented to correspond with the imaging angle of the online images to which they will be matched. Templates may be derived from the actual voxel values of the plan CT (using MIP or tomosynthesis) or may be generated from the plan contours without reference to the voxel values. The RapidTrack-Lung application performs the actual matching with online projections using a cross-correlation method. The match locations can be used to derive a 3D position of the target. Quality of the match is measured using a peak-to-sidelobe ratio (PSR) metric, which reduces the incidence of false detection. RESULTS The RapidTrack suite was shown to be quite versatile for tracking of radiotherapy targets. The generated templates are not dependent on any particular shape or structure of target; hence they are applicable to a variety of fiducial shapes as well as to fiducial-less scenarios. RapidTrack was used on intra-fraction images of seeds in pancreas; quality of the match was visually illustrated by overlaying the planning contour on the image at the match location. PSR of the match was consistently over 2.2 at all angles. Tracking of embolization coils in lung produced similar results. Breathing motion of a lung tumor was successfully tracked using a markerless algorithm based on synthetic contour template. CONCLUSION The template-based algorithm demonstrated shows promise for tracking radiotherapy targets during treatment, both in fiducial and fiducial-less scenarios. The PSR metric is an effective means of measuring the quality of tracking. Varian Medical Systems.