Cone-beam CT Geometry Research Articles

SU-E-J-09: Performance Optimization of Thick, Segmented Scintillators for Radiotherapy Imaging.

Thick segmented scintillators, incorporating a 2-dimensional matrix of optically-isolated scintillator elements, have shown considerable potential for improving the performance of megavoltage active matrix, flat-panel imagers (MV AMFPIs). While over a factor of 20 improvement in DQE at zero spatial-frequency has been demonstrated for prototypes incorporating CsI(Tl) and BGO scintillators, less-than-optimal element-to-element alignment (misalignment) as well as mis-registration to the underlying AMFPI array pixels can result in spatial resolution loss, reducing DQE improvement at higher spatial frequencies. In this presentation, a method to restore spatial resolution and DQE, based on the use of a high resolution AMFPI array along with special binning techniques, is investigated. The effect of misalignment and mis-registration of segmented scintillators on imaging performance was investigated theoretically and empirically through determination of the modulation transfer function (MTF) and DQE, as well as through realization of reconstructed images of a phantom in a cone-beam CT geometry. The empirical investigation, which was conducted using a 6 MV photon beam, employed a prototype BGO segmented scintillator consisting of 120×60 elements separated by 50 μm-thick septal walls and an element-to-element pitch of 1016 μm. The scintillator was coupled to a higher resolution 127-μm-pitch AMFPI array. Misalignment and mis-registration result in significant degradation of spatial resolution, leading to DQE reduction at non-zero spatial frequencies. While mis-registration for a well-aligned scintillator can be overcome through 8×8 binning of the array pixels to match the scintillator elements, any misalignment will affect such binning and lead to spatial resolution loss. However, the use of 'selective' binning, consisting of the selection of those pixels corresponding to the interior locations of each element, improves resolution while preserving DQE. The use of high-resolution AMFPI arrays combined with selective binning allows prototype AMFPIs incorporating thick, segmented scintillators to achieve imager performance limited only by scintillator performance. Work Supported by NIH grant R01-CA051397.

Dose-Guided Radiotherapy: Potential Benefit of Online Dose Recalculation for Stereotactic Lung Irradiation in Patients With Non-Small-Cell Lung Cancer

To determine whether dose-guided radiotherapy (i.e., online recalculation and evaluation of the actual dose distribution) can improve decision making for lung cancer patients treated with stereotactic body radiotherapy. For this study 108 cone-beam computed tomography (CBCT) scans of 10 non-small-cell lung cancer patients treated with stereotactic body radiotherapy were analyzed retrospectively. The treatment plans were recalculated on the CBCT scans. The V(100%) of the internal target volume (ITV) and D(max) of the organs at risk (OARs) were analyzed. Results from the recalculated data were compared with dose estimates for target and OARs by superposition of the originally planned dose distribution on CBCT geometry (i.e., the original dose distribution was assumed to be spatially invariant). Before position correction was applied the V(100%) of the ITV was 100% in 65% of the cases when an ITV-PTV margin of 5 mm was used and 52% of the cases when a margin of 3 mm was used. After position correction, the difference of D(max) in the OARs with respect to the treatment plan was within 5% in the majority of the cases. When the dose was not recalculated but estimated assuming an invariant dose distribution, clinically relevant errors occurred in both the ITV and the OARs. Dose-guided radiotherapy can be used to determine the actual dose in OARs when the target has moved with respect to the OARs. When the workflow is optimized for speed, it can be used to prevent unnecessary position corrections. Estimating the dose by assuming an invariant dose instead of recalculation of the dose gives clinically relevant errors.

Estimation of CT cone-beam geometry using a novel method insensitive to phantom fabrication inaccuracy: Implications for isocenter localization accuracy

Mechanical instabilities that occur during gantry rotation of on-board cone-beam computed tomography (CBCT) imaging systems limit the efficacy of image-guided radiotherapy. Various methods for calibrating the CBCT geometry and correcting errors have been proposed, including some that utilize dedicated fiducial phantoms. The purpose of this work was to investigate the role of phantom fabrication imprecision on the accuracy of a particular CT cone-beam geometry estimate and to test a new method to mitigate errors in beam geometry arising from imperfectly fabricated phantoms. The authors implemented a fiducial phantom-based beam geometry estimation following the one described by Cho et al. [Med Phys 32(4), 968-983 (2005)]. The algorithm utilizes as input projection images of the phantom at various gantry angles and provides a full nine parameter beam geometry characterization of the source and detector position and detector orientation versus gantry angle. A method was developed for recalculating the beam geometry in a coordinate system with origin at the source trajectory center and aligned with the axis of gantry rotation, thus making the beam geometry estimation independent of the placement of the phantom. A second CBCT scan with the phantom rotated 180 degrees about its long axis was averaged with the first scan to mitigate errors from phantom imprecision. Computer simulations were performed to assess the effect of 2D fiducial marker positional error on the projections due to image discretization, as well as 3D fiducial marker position error due to phantom fabrication imprecision. Experimental CBCT images of a fiducial phantom were obtained and the algorithm used to measure beam geometry for a Varian Trilogy with an on-board CBCT. Both simulations and experimental results reveal large sinusoidal oscillations in the calculated beam geometry parameters with gantry angle due to displacement of the phantom from CBCT isocenter and misalignment with the gantry axis, which are eliminated by recalculating the beam geometry in the source coordinate system. Simulations and experiments also reveal an additional source of oscillations arising from fiducial marker position error due to phantom fabrication imprecision that are mitigated by averaging the results with those of a second CBCT scan with phantom rotated. With a typical fiducial marker position error of 0.020 mm (0.001 in.), source and detector position are found in simulations to be within 250 microm of the true values, and detector and gantry angles less than 0.2 degrees. Detector offsets are within 100 microm of the known value. Experimental results verify the efficacy of the second scan in mitigating beam geometry errors, as well as large apparent source/detector isocenter offsets arising from phantom fabrication imprecision. The authors have developed and validated a novel fiducial phantom-based CBCT beam geometry estimation algorithm that does not require precise positioning of the phantom at machine isocenter and is insensitive to positional imprecision of fiducial markers within the phantom due to fabrication errors. The method can accurately locate source and detector isocenters even when using an imprecise phantom, which is very important for measurement of isocenter coincidence of the therapy and on-board imaging systems.

The effects of compensator and imaging geometry on the distribution of x‐ray scatter in CBCT

X-ray scatter contributes significantly to image degradation in cone-beam CT (CBCT) reconstructed images in the form of CT number inaccuracy image artifacts and loss of contrast. The need for an understanding of the relationship between the scatter distribution and common imaging parameters (cone angle, air gap, filtration, object size) is an essential element to developing methods for efficiently correcting for the effects of scatter in CBCT. The first objective of this study is to validate the scatter distributions calculated using a CBCT Monte Carlo (MC) model against measured scatter estimates. The second objective is to use the CBCT MC model to investigate the effects of common imaging parameters and bowtie compensators on the resulting scatter distribution. This investigation employs the use of a CBCT MC model, developed using the EGSnrc code, to simulate the primary and scatter fluence arriving at the detector. The simulation is validated against projection images, scatter-to-open field ratio (SOR), and scatter-to-primary ratio (SPR) measurements taken using a bench-top CBCT system. The CBCT MC model is used to simulate the scatter distribution arriving at the detector for different cone angles {1.4 degrees, 2.8 degrees, 5.7 degrees, and 11.3 degrees}, source-to-axis distances (SADs) {50, 75 and 100 cm}, and axis-to-detector distances (ADDs) {9, 18, 30, 44, 56 cm} for both a 16.4 and 30.6 cm diameter water cylinder. The effects of different bowtie filters are also simulated using the CBCT MC model for the aforementioned cylinder sizes. Profiles of the simulated and measured projection images agree within 6%. The measurements of the SOR and SPR, taken using beam stop techniques, show good agreement with the simulated results. Limitations of current beam stop techniques in estimating scatter profiles of objects with varying thickness are also demonstrated. A functional relationship between scatter (SOR, SPR) and air gap, cone angle is reported. The bowtie filter was found to have the beneficial effect of decreasing the magnitude of scatter in the projection images (SPR decreased by 56%) as well as altering the spatial distribution of the scatter. The CBCT MC model accurately simulates scatter and primary fluences in the CBCT imaging components and geometry. The CBCT MC model provides a useful tool in investigating the effects of varying CBCT imaging parameters on the scatter distribution. Increasing the air gap, decreasing the cone angle, and the use of bowtie filtration were all found to be effective ways to minimize scatter in CBCT. The bowtie filter was particularly effective in both minimizing the magnitude and modifying the spatial distribution of the scattered photons. This observation directs further research in optimizing bowtie filter design in an effort to minimize scatter induced artifacts.

ESTIMATION OF COMPUTED TOMOGRAPHY DOSE INDEX IN CONE BEAM COMPUTED TOMOGRAPHY: MOSFET MEASUREMENTS AND MONTE CARLO SIMULATIONS

To address the lack of accurate dose estimation method in cone beam computed tomography (CBCT), we performed point dose metal oxide semiconductor field-effect transistor (MOSFET) measurements and Monte Carlo (MC) simulations. A Varian On-Board Imager (OBI) was employed to measure point doses in the polymethyl methacrylate (PMMA) CT phantoms with MOSFETs for standard and low dose modes. A MC model of the OBI x-ray tube was developed using BEAMnrc/EGSnrc MC system and validated by the half value layer, x-ray spectrum and lateral and depth dose profiles. We compared the weighted computed tomography dose index (CTDIw) between MOSFET measurements and MC simulations. The CTDIw was found to be 8.39 cGy for the head scan and 4.58 cGy for the body scan from the MOSFET measurements in standard dose mode, and 1.89 cGy for the head and 1.11 cGy for the body in low dose mode, respectively. The CTDIw from MC compared well to the MOSFET measurements within 5% differences. In conclusion, a MC model for Varian CBCT has been established and this approach may be easily extended from the CBCT geometry to multi-detector CT geometry.

Poster — Wed Eve—40: Comparison of CBCT CT Numbers to CT‐Simulator CT Numbers using Mutual Information and Joint Probability Distributions

The Varian On‐Board Imaging (OBI) system is an imaging tool used by the London Regional Cancer Program for image‐guided radiation therapy. The system consists of a kilovoltage source (120 kVp), emitting a conical X‐ray beam, and a flat panel amorphous silicon detector. Both source and detector are mounted on the linear accelerator gantry, rotating 360° about the patient creating 3‐dimensional volumetric computed tomography images. Unfortunately, cone‐beam computed tomography (CBCT) images are susceptible to many forms of artifacts resulting primarily from: increased amounts of scatter due to CBCT geometry, spectral X‐ray attenuation, and intra‐scanning organ motion. In order to assess the extent of CBCT image artifacts, we propose a new technique that compares CT Numbers in corresponding tissue regions obtained by the CBCT and Phillips Brilliance CT‐simulator systems. Our approach registers CT images from two imaging modalities using a mutual information (MI) algorithm. Once aligned, quantitative CT Number analysis is performed on the CT images using a 2‐dimensional histogram (or joint probability distribution) of pixel counts for geographically‐matched pairs of CT Numbers. Phantom studies have shown similarity between CBCT and CT‐simulation CT Numbers when the imaged medium is head‐like in diameter (20 cm) and homogeneous (Catphan 504 phantom). Greater discrepancies were observed (400 HU) when CT images of larger phantoms (30 cm diameter, Gammex RMI Tomo phantom) and increased heterogeneity were compared. Presently, quantitative CBCT CT Number inaccuracy precludes the use of CBCT images for whole body treatment planning, and adaptive dose computations.

SU‐FF‐I‐18: Organ Doses and Projection Angles in Cone Beam CT

Purpose: To calculate how organ doses varies with projection angle for two types of cone beam CT geometry. Method: We used the PCXMC software package to compute organ doses to patients. We modeled a typical x‐ray spectrum using 120 kV x‐ray tube voltage and 6 mm Al filtration. Two cone beam irradiation geometries were modeled: (a) a standard CT gantry; and (b) a C‐arm. Both systems made use of a detector with dimensions of 30 cm × 40 cm. For the CT gantry system, the source to isocenter distance was 57 cm, and the SID was 93 cm. For the C‐arm system, the source to isocenter distance was 78.5 cm, and the SID was 120 cm. Organ dose to the five most radiosensitive organs (breast, colon, lung, red bone marrow, and stomach) were obtained for each cone beam arrangement at 15 degree intervals as the x‐ray tube rotates through 360 degrees around the patient. Results: On average, patient doses using the C‐arm were ∼60% of those obtained using the CT gantry. Ratios of maximum to minimum doses as a function of x‐ray tube angle ranged from ∼2.3:1 for the lung to ∼31:1 for the stomach. For the breast, colon, and stomach the highest doses were obtained for AP projections, whereas for the lung and red bone marrow, PA projections resulted in the highest organ doses. In all cases, dose minima were obtained when the x‐ray beam entered the patient in a lateral orientation. Conclusions: The high maximum to minimum dose ratios for radiosensitive organs such as the stomach and breast indicate that modulation of the x‐ray tube current with projection angle could result in substantial dose reductions in cone beam CT.

3D in vivo dose verification of entire hypo-fractionated IMRT treatments using an EPID and cone-beam CT

As radiotherapy becomes more complicated, dose and geometry verification become more necessary. The aim of this study was to use back-projected EPID-based 3D in vivo dosimetry and cone-beam CT (CBCT) to obtain a complete account of the entire treatment for a select patient group. Nine hypo-fractionated rectum IMRT patient plans were investigated. The absolute dose was reconstructed at multiple planes using patient contours and EPID images acquired for all fields during treatment. The meso-rectal fat (m-R) was re-delineated on daily CBCT scans, acquired prior to each fraction. The total accumulated dose was determined by mapping the m-R surface of each fraction to the planned m-R surface. Average planned and measured isocentre dose ratios were 0.98 (±0.01SD). 3D gamma analysis (3% maximum dose and 3 mm) revealed mean γ, 〈 γ mean〉 = 0.35 (±0.03 SD), maximum 1% of γ points, 〈 γ max1%〉 = 1.02 (±0.14SD) and the percentage of points with γ ⩽ 1, 〈 P γ⩽1 〉 = 99% (range [96%, 100%]), averaged over all patients. CBCT m-R volumes varied by up to 20% of planned volumes, but remained in the high dose region. Over-dosage of up to 4.5% in one fraction was measured in the presence of gas pockets. By combining EPID dosimetry with CBCT geometry information, the total dose can be verified in 3D in vivo and compared with the planned dose distribution. This method can provide a safety net for advanced treatments involving dose escalation, as well as a full account of the delivered dose to specific volumes, allowing adaptation of the treatment from the original plan if necessary.

WE-D-L100E-01: Cone Beam CT Workshop

The development of practical large field-of-view flat panel x-ray detectors has lead to an explosion in cone-beam computed tomography (CT) applications. In this workshop, a number of different clinical applications for cone-beam CT will be described. Methodologies for addressing issues such as x-ray scatter, detector lag, and limited field-of-view or angular coverage will also be discussed. C-arm fluoroscopic systems have been used for image guidance in the operating room for many years. With the advent of flat panel-based C-arm systems, the availability of portable, operating room-based computed tomography systems exists. The three dimensional characteristics of CT will likely improve surgical outcomes involving surgical device implantation. The development of a portable C-arm-based CT system, for use in the operating theater as well as in other clinical settings, will be discussed. Angiographic suites are designed to optimize vascular imaging geometry. Flat panel-based angiography suites have undergone modifications to allow CT imaging for some specific vascular imaging applications. These systems are also capable of performance in orthopedic applications. The use of such systems in a number of clinical applications will be described, and corrections methods for x-ray scatter and detector lag will also be presented. The use of image guidance in radiation therapy applications has skyrocketed, and in many clinical radiation therapy systems, system-mounted cone-beam CT systems are available for use for guiding RT delivery. The use of cone-beam CT for radiation therapy applications will be described, and some of the challenges and limitations of this technology will be discussed. Overall, a sampling of cone-beam CT applications will be presented in this workshop. The overall flexibility of cone-beam CT systems to accommodate the wide range of applications described is a testimony to the long term clinical potential of cone-beam CT systems. This workshop will also describe some of the compromises and limitations which are required when cone-beam CT geometry is utilized.

Evaluation of x‐ray scatter properties in a dedicated cone‐beam breast CT scanner

The magnitude of scatter contamination on a first-generation prototype breast computed tomography (CT) scanner was evaluated using the scatter-to-primary ratio (SPR) metric. The SPR was measured and characterized over a wide range of parameters relevant to breast CT imaging, including x-ray beam energy, breast diameter, breast composition, isocenter-to-detector distance, collimated slot thickness, and grid ratio. The results demonstrated that in the absence of scatter reduction techniques, the SPR levels for the average breast (e.g., 14 cm diameter 50/50 composition cylindrical phantom) are quite high (approximately 0.5 at the center of the phantom for 80 kVp in true cone-beam CT geometry), and increases as the diameter of the phantom is increased (to approximately 1.0 at the center of a 18 cm diameter 50/50 phantom). The x-ray beam energy and the phantom compositions had only minimal impact on the measured SPR. When an ideal bowtie filter was used, the SPRs at the central axis of the 14 and 18 cm cylindrical phantoms were reduced while the SPRs at the edge of the phantoms were increased. Lastly, collimation in the vertical direction had a significant impact on the SPRs at the central axis of the phantoms. These high SPR levels might lead to cupping artifacts and increased noise in the reconstructed CT images, and this suggests that efficient scatter rejection and/or correction techniques may be required to improve the quality and accuracy of cone beam CT images.