Volume 41, Issue 8, August 2014
Index of content:
41(2014); http://dx.doi.org/10.1118/1.4883875View Description Hide Description
- RADIATION THERAPY PHYSICS
A critical evaluation of worst case optimization methods for robust intensity-modulated proton therapy planning41(2014); http://dx.doi.org/10.1118/1.4883837View Description Hide DescriptionPurpose:
To critically evaluate and compare three worst case optimization methods that have been previously employed to generate intensity-modulated proton therapy treatment plans that are robust against systematic errors. The goal of the evaluation is to identify circumstances when the methods behave differently and to describe the mechanism behind the differences when they occur.Methods:
The worst case methods optimize plans to perform as well as possible under the worst case scenario that can physically occur (composite worst case), the combination of the worst case scenarios for each objective constituent considered independently (objectivewise worst case), and the combination of the worst case scenarios for each voxel considered independently (voxelwise worst case). These three methods were assessed with respect to treatment planning for prostate under systematic setup uncertainty. An equivalence with probabilistic optimization was used to identify the scenarios that determine the outcome of the optimization.Results:
If the conflict between target coverage and normal tissue sparing is small and no dose-volume histogram (DVH) constraints are present, then all three methods yield robust plans. Otherwise, they all have their shortcomings: Composite worst case led to unnecessarily low plan quality in boundary scenarios that were less difficult than the worst case ones. Objectivewise worst case generally led to nonrobust plans. Voxelwise worst case led to overly conservative plans with respect to DVH constraints, which resulted in excessive dose to normal tissue, and less sharp dose fall-off than the other two methods.Conclusions:
The three worst case methods have clearly different behaviors. These behaviors can be understood from which scenarios that are active in the optimization. No particular method is superior to the others under all circumstances: composite worst case is suitable if the conflicts are not very severe or there are DVH constraints whereas voxelwise worst case is advantageous if there are severe conflicts but no DVH constraints. The advantages of composite and voxelwise worst case outweigh those of objectivewise worst case.
41(2014); http://dx.doi.org/10.1118/1.4884019View Description Hide DescriptionPurpose:
Accurate calculation of radiobiological parameters is crucial to predicting radiation treatment response. Modeling differences may have a significant impact on derived parameters. In this study, the authors have integrated two existing models with kinetic differential equations to formulate a new tumor regression model for estimation of radiobiological parameters for individual patients.Methods:
A system of differential equations that characterizes the birth-and-death process of tumor cells in radiation treatment was analytically solved. The solution of this system was used to construct an iterative model (Z-model). The model consists of three parameters: tumor doubling time Td, half-life of dead cells Tr, and cell survival fraction SFD under dose D. The Jacobian determinant of this model was proposed as a constraint to optimize the three parameters for six head and neck cancer patients. The derived parameters were compared with those generated from the two existing models: Chvetsov's model (C-model) and Lim's model (L-model). The C-model and L-model were optimized with the parameter Td fixed.Results:
With the Jacobian-constrained Z-model, the mean of the optimized cell survival fractions is 0.43 ± 0.08, and the half-life of dead cells averaged over the six patients is 17.5 ± 3.2 days. The parameters Tr and SFD optimized with the Z-model differ by 1.2% and 20.3% from those optimized with the Td-fixed C-model, and by 32.1% and 112.3% from those optimized with the Td-fixed L-model, respectively.Conclusions:
The Z-model was analytically constructed from the differential equations of cell populations that describe changes in the number of different tumor cells during the course of radiation treatment. The Jacobian constraints were proposed to optimize the three radiobiological parameters. The generated model and its optimization method may help develop high-quality treatment regimens for individual patients.
Comparison of two methods for minimizing the effect of delayed charge on the dose delivered with a synchrotron based discrete spot scanning proton beam41(2014); http://dx.doi.org/10.1118/1.4885961View Description Hide DescriptionPurpose:
Delayed charge is a small amount of charge that is delivered to the patient after the planned irradiation is halted, which may degrade the quality of the treatment by delivering unwarranted dose to the patient. This study compares two methods for minimizing the effect of delayed charge on the dose delivered with a synchrotron based discrete spot scanning proton beam.Methods:
The delivery of several treatment plans was simulated by applying a normally distributed value of delayed charge, with a mean of 0.001(SD 0.00025) MU, to each spot. Two correction methods were used to account for the delayed charge. Method one (CM1), which is in active clinical use, accounts for the delayed charge by adjusting the MU of the current spot based on the cumulative MU. Method two (CM2) in addition reduces the planned MU by a predicted value. Every fraction of a treatment was simulated using each method and then recomputed in the treatment planning system. The dose difference between the original plan and the sum of the simulated fractions was evaluated. Both methods were tested in a water phantom with a single beam and simple target geometry. Two separate phantom tests were performed. In one test the dose per fraction was varied from 0.5 to 2 Gy using 25 fractions per plan. In the other test the number fractions were varied from 1 to 25, using 2 Gy per fraction. Three patient plans were used to determine the effect of delayed charge on the delivered dose under realistic clinical conditions. The order of spot delivery using CM1 was investigated by randomly selecting the starting spot for each layer, and by alternating per layer the starting spot from first to last. Only discrete spot scanning was considered in this study.Results:
Using the phantom setup and varying the dose per fraction, the maximum dose difference for each plan of 25 fractions was 0.37–0.39 Gy and 0.03–0.05 Gy for CM1 and CM2, respectively. While varying the total number of fractions, the maximum dose difference increased at a rate of 0.015 Gy and 0.0018 Gy per fraction for CM1 and CM2, respectively. For CM1, the largest dose difference was found at the location of the first spot in each energy layer, whereas for CM2 the difference in dose was small and showed no dependence on location. For CM1, all of the fields in the patient plans had an area where their excess dose overlapped. No such correlation was found when using CM2. Randomly selecting the starting spot reduces the maximum dose difference from 0.708 to 0.15 Gy. Alternating between first and last spot reduces the maximum dose difference from 0.708 to 0.37 Gy. In the patient plans the excess dose scaled linearly at 0.014 Gy per field per fraction for CM1 and standard delivery order.Conclusions:
The predictive model CM2 is superior to a cumulative irradiation model CM1 for minimizing the effects of delayed charge, particularly when considering maximal dose discrepancies and the potential for unplanned hot-spots. This study shows that the dose discrepancy potentially scales at 0.014 Gy per field per fraction for CM1.
Noninvasive referencing of intraocular tumors for external beam radiation therapy using optical coherence tomography: A proof of concept41(2014); http://dx.doi.org/10.1118/1.4885975View Description Hide DescriptionPurpose:
External beam radiation therapy is currently considered the most common treatment modality for intraocular tumors. Localization of the tumor and efficient compensation of tumor misalignment with respect to the radiation beam are crucial. According to the state of the art procedure, localization of the target volume is indirectly performed by the invasive surgical implantation of radiopaque clips or is limited to positioning the head using stereoscopic radiographies. This work represents a proof-of-concept for direct and noninvasive tumor referencing based on anterior eye topography acquired using optical coherence tomography (OCT).Methods:
A prototype of a head-mounted device has been developed for automatic monitoring of tumor position and orientation in the isocentric reference frame for LINAC based treatment of intraocular tumors. Noninvasive tumor referencing is performed with six degrees of freedom based on anterior eye topography acquired using OCT and registration of a statistical eye model. The proposed prototype was tested based on enucleated pig eyes and registration accuracy was measured by comparison of the resulting transformation with tilt and torsion angles manually induced using a custom-made test bench.Results:
Validation based on 12 enucleated pig eyes revealed an overall average registration error of 0.26 ± 0.08° in 87 ± 0.7 ms for tilting and 0.52 ± 0.03° in 94 ± 1.4 ms for torsion. Furthermore, dependency of sampling density on mean registration error was quantitatively assessed.Conclusions:
The tumor referencing method presented in combination with the statistical eye model introduced in the past has the potential to enable noninvasive treatment and may improve quality, efficacy, and flexibility of external beam radiotherapy of intraocular tumors.
41(2014); http://dx.doi.org/10.1118/1.4886015View Description Hide DescriptionPurpose:
Microbeam radiation therapy (MRT) uses narrow planes of high dose radiation beams to treat cancerous tumors. This experimental therapy method based on synchrotron radiation has been shown to spare normal tissue at up to 1000 Gy of peak entrance dose while still being effective in tumor eradication and extending the lifetime of tumor-bearing small animal models. Motion during treatment can lead to significant movement of microbeam positions resulting in broader beam width and lower peak to valley dose ratio (PVDR), which reduces the effectiveness of MRT. Recently, the authors have demonstrated the feasibility of generating microbeam radiation for small animal treatment using a carbon nanotube (CNT) x-ray source array. The purpose of this study is to incorporate physiological gating to the CNT microbeam irradiator to minimize motion-induced microbeam blurring.Methods:
The CNT field emission x-ray source array with a narrow line focal track was operated at 160 kVp. The x-ray radiation was collimated to a single 280 μm wide microbeam at entrance. The microbeam beam pattern was recorded using EBT2 Gafchromic© films. For the feasibility study, a strip of EBT2 film was attached to an oscillating mechanical phantom mimicking mouse chest respiratory motion. The servo arm was put against a pressure sensor to monitor the motion. The film was irradiated with three microbeams under gated and nongated conditions and the full width at half maximums and PVDRs were compared. An in vivo study was also performed with adult male athymic mice. The liver was chosen as the target organ for proof of concept due to its large motion during respiration compared to other organs. The mouse was immobilized in a specialized mouse bed and anesthetized using isoflurane. A pressure sensor was attached to a mouse's chest to monitor its respiration. The output signal triggered the electron extraction voltage of the field emission source such that x-ray was generated only during a portion of the mouse respiratory cycle when there was minimum motion. Parallel planes of microbeams with 12.4 Gy/plane dose and 900 μm pitch were delivered. The microbeam profiles with and without gating were analyzed using γ-H2Ax immunofluorescence staining.Results:
The phantom study showed that the respiratory motion caused a 50% drop in PVDR from 11.5 when there is no motion to 5.4, whereas there was only a 5.5% decrease in PVDR for gated irradiation compared to the no motion case. In thein vivo study, the histology result showed gating increased PVDR by a factor of 2.4 compared to the nongated case, similar to the result from the phantom study. The full width at tenth maximum of the microbeam decreased by 40% in gating in vivo and close to 38% with phantom studies.Conclusions:
The CNT field emission x-ray source array can be synchronized to physiological signals for gated delivery of x-ray radiation to minimize motion-induced beam blurring. Gated MRT reduces valley dose between lines during long-time radiation of a moving object. The technique allows for more precise MRT treatments and makes the CNT MRT device practical for extended treatment.
Implication of spot position error on plan quality and patient safety in pencil-beam-scanning proton therapya)41(2014); http://dx.doi.org/10.1118/1.4885956View Description Hide DescriptionPurpose:
To quantitatively and systematically assess dosimetric effects induced by spot positioning error as a function of spot spacing (SS) on intensity-modulated proton therapy (IMPT) plan quality and to facilitate evaluation of safety tolerance limits on spot position.Methods:
Spot position errors (PE) ranging from 1 to 2 mm were simulated. Simple plans were created on a water phantom, and IMPT plans were calculated on two pediatric patients with a brain tumor of 28 and 3 cc, respectively, using a commercial planning system. For the phantom, a uniform dose was delivered to targets located at different depths from 10 to 20 cm with various field sizes from 22 to 152 cm2. Two nominal spot sizes, 4.0 and 6.6 mm of 1 σ in water at isocenter, were used for treatment planning. The SS ranged from 0.5 σ to 1.5 σ, which is 2–6 mm for the small spot size and 3.3–9.9 mm for the large spot size. Various perturbation scenarios of a single spot error and systematic and random multiple spot errors were studied. To quantify the dosimetric effects, percent dose error (PDE) depth profiles and the value of percent dose error at the maximum dose difference (PDE [ΔDmax]) were used for evaluation.Results:
A pair of hot and cold spots was created per spot shift. PDE[ΔDmax] is found to be a complex function of PE, SS, spot size, depth, and global spot distribution that can be well defined in simple models. For volumetric targets, the PDE [ΔDmax] is not noticeably affected by the change of field size or target volume within the studied ranges. In general, reducing SS decreased the dose error. For the facility studied, given a single spot error with a PE of 1.2 mm and for both spot sizes, a SS of 1σ resulted in a 2% maximum dose error; a SS larger than 1.25 σ substantially increased the dose error and its sensitivity to PE. A similar trend was observed in multiple spot errors (both systematic and random errors). Systematic PE can lead to noticeable hot spots along the field edges, which may be near critical structures. However, random PE showed minimal dose error.Conclusions:
Dose error dependence for PE was quantitatively and systematically characterized and an analytic tool was built to simulate systematic and random errors for patient-specific IMPT. This information facilitates the determination of facility specific spot position error thresholds.
41(2014); http://dx.doi.org/10.1118/1.4887778View Description Hide DescriptionPurpose:
Modeling dose to a moving target in lung is a very difficult task. Current approaches to planning lung stereotactic body radiotherapy (SBRT) generally calculate dose on either free breathing or average computed tomography (CT) scans, which do not always accurately predict dose to parts of the target volume not occupied by tumor on the planning scan. In this work, the authors look at using density overrides of the target volumes to more accurately predict dose for lung SBRT using the analytic anisotropic algorithm (AAA).Methods:
Volumetric modulated arc therapy plans were created on free breathing scans (FBP), time average scans (AVGP), free breathing scans with the internal target volume overridden to tumor density (ITVP), free breathing scans with the planning target volume overridden to tumor density (PTVP), and free breathing scan using a hybrid scheme with the internal target volume set to tumor density and the planning target volume minus the internal target volume set to a density intermediate between lung and tumor (HP) for the case of a 4D motion phantom and five patient cases. Radiochromic film measurements were made for the phantom plans, with gamma analysis used to compare the planned to delivered dose. The patient plans were recalculated on each of the phases of a 4DCT to evaluate tumor coverage and conformity index (CI). A modified modulation complexity score (MCSv) and average open area per control point (AA) metrics were used to evaluate multileaf collimator (MLC) modulation for each of the plans.Results:
The HP plans showed significantly higher gamma passing rates (p < 0.05) than the FBP, AVGP, and ITVP for criteria of 2 mm/2% and 1 mm/1%. No significant correlation was observed between gamma values and AA or MCSv. The tumor volume was covered by the prescription dose on all phases of the 4DCT for all patient plans. The PTVP and HP yielded lower mean CI than the other plans for all five patients, with three of the cases showing statistically significant differences (p < 0.05). No meaningful correlation was observed between the mean CI and AA or MCSv.Conclusions:
These measurements suggest that the HP planning method may provide more accurate dose modeling and decreased normal lung irradiation for lung SBRT compared to the commonly used FBP and AVG planning methods when used with the AAA. The HP method does not appear to have a strong relationship with MLC modulation.
Forecasting longitudinal changes in oropharyngeal tumor morphology throughout the course of head and neck radiation therapy41(2014); http://dx.doi.org/10.1118/1.4887815View Description Hide DescriptionPurpose:
To create models that forecast longitudinal trends in changing tumor morphology and to evaluate and compare their predictive potential throughout the course of radiation therapy.Methods:
Two morphology feature vectors were used to describe 35 gross tumor volumes (GTVs) throughout the course of intensity-modulated radiation therapy for oropharyngeal tumors. The feature vectors comprised the coordinates of the GTV centroids and a description of GTV shape using either interlandmark distances or a spherical harmonic decomposition of these distances. The change in the morphology feature vector observed at 33 time points throughout the course of treatment was described using static, linear, and mean models. Models were adjusted at 0, 1, 2, 3, or 5 different time points (adjustment points) to improve prediction accuracy. The potential of these models to forecast GTV morphology was evaluated using leave-one-out cross-validation, and the accuracy of the models was compared using Wilcoxon signed-rank tests.Results:
Adding a single adjustment point to the static model without any adjustment points decreased the median error in forecasting the position of GTV surface landmarks by the largest amount (1.2 mm). Additional adjustment points further decreased the forecast error by about 0.4 mm each. Selection of the linear model decreased the forecast error for both the distance-based and spherical harmonic morphology descriptors (0.2 mm), while the mean model decreased the forecast error for the distance-based descriptor only (0.2 mm). The magnitude and statistical significance of these improvements decreased with each additional adjustment point, and the effect from model selection was not as large as that from adding the initial points.Conclusions:
The authors present models that anticipate longitudinal changes in tumor morphology using various models and model adjustment schemes. The accuracy of these models depended on their form, and the utility of these models includes the characterization of patient-specific response with implications for treatment management and research study design.
Independent calculation of dose distributions for helical tomotherapy using a conventional treatment planning system41(2014); http://dx.doi.org/10.1118/1.4887779View Description Hide DescriptionPurpose:
The dosimetric verification of treatment plans in helical tomotherapy usually is carried out via verification measurements. In this study, a method for independent dose calculation of tomotherapy treatment plans is presented, that uses a conventional treatment planning system with a pencil kernel dose calculation algorithm for generation of verification dose distributions based on patient CT data.Methods:
A pencil beam algorithm that directly uses measured beam data was configured for dose calculation for a tomotherapy machine. Tomotherapy treatment plans were converted into a format readable by an in-house treatment planning system by assigning each projection to one static treatment field and shifting the calculation isocenter for each field in order to account for the couch movement. The modulation of the fluence for each projection is read out of the delivery sinogram, and with the kernel-based dose calculation, this information can directly be used for dose calculation without the need for decomposition of the sinogram. The sinogram values are only corrected for leaf output and leaf latency. Using the converted treatment plans, dose was recalculated with the independent treatment planning system. Multiple treatment plans ranging from simple static fields to real patient treatment plans were calculated using the new approach and either compared to actual measurements or the 3D dose distribution calculated by the tomotherapy treatment planning system. In addition, dose–volume histograms were calculated for the patient plans.Results:
Except for minor deviations at the maximum field size, the pencil beam dose calculation for static beams agreed with measurements in a water tank within 2%/2 mm. A mean deviation to point dose measurements in the cheese phantom of 0.89% ± 0.81% was found for unmodulated helical plans. A mean voxel-based deviation of −0.67% ± 1.11% for all voxels in the respective high dose region (dose values >80%), and a mean local voxel-based deviation of −2.41% ± 0.75% for all voxels with dose values >20% were found for 11 modulated plans in the cheese phantom. Averaged over nine patient plans, the deviations amounted to −0.14% ± 1.97% (voxels >80%) and −0.95% ± 2.27% (>20%, local deviations). For a lung case, mean voxel-based deviations of more than 4% were found, while for all other patient plans, all mean voxel-based deviations were within ±2.4%.Conclusions:
The presented method is suitable for independent dose calculation for helical tomotherapy within the known limitations of the pencil beam algorithm. It can serve as verification of the primary dose calculation and thereby reduce the need for time-consuming measurements. By using the patient anatomy and generating full 3D dose data, and combined with measurements of additional machine parameters, it can substantially contribute to overall patient safety.
On the suitability of ultrathin detectors for absorbed dose assessment in the presence of high-density heterogeneities41(2014); http://dx.doi.org/10.1118/1.4886760View Description Hide DescriptionPurpose:
The aim of this study was to evaluate the suitability of several detectors for the determination of absorbed dose in bone.Methods:
Three types of ultrathin LiF-based thermoluminescent dosimeters (TLDs)—two LiF:Mg,Cu,P-based (MCP-Ns and TLD-2000F) and a7Li-enriched LiF:Mg,Ti-based (MTS-7s)—as well as EBT2 Gafchromic films were used to measure percentage depth-dose distributions (PDDs) in a water-equivalent phantom with a bone-equivalent heterogeneity for 6 and 18 MV and a set of field sizes ranging from to . MCP-Ns, TLD-2000F, MTS-7s, and EBT2 have active layers of 50, 20, 50, and 30 μm, respectively. Monte Carlo (MC) dose calculations (PENELOPE code) were used as the reference and helped to understand the experimental results and to evaluate the potential perturbation of the fluence in bone caused by the presence of the detectors. The energy dependence and linearity of the TLDs’ response was evaluated.Results:
TLDs exhibited flat energy responses (within 2.5%) and linearity with dose (within 1.1%) within the range of interest for the selected beams. The results revealed that all considered detectors perturb the electron fluence with respect to the energy inside the bone-equivalent material. MCP-Ns and MTS-7s underestimated the absorbed dose in bone by 4%–5%. EBT2 exhibited comparable accuracy to MTS-7s and MCP-Ns. TLD-2000F was able to determine the dose within 2% accuracy. No dependence on the beam energy or field size was observed. The MC calculations showed that a thick detector can provide reliable dose estimations in bone regardless of whether it is made of LiF, water or EBT’s active layer material.Conclusions:
TLD-2000F was found to be suitable for providing reliable absorbed dose measurements in the presence of bone for high-energy x-ray beams.
A dosimetric comparison of proton and photon therapy in unresectable cancers of the head of pancreas41(2014); http://dx.doi.org/10.1118/1.4887797View Description Hide DescriptionPurpose:
Uncontrolled local growth is the cause of death in ∼30% of patients with unresectable pancreatic cancers. The addition of standard-dose radiotherapy to gemcitabine has been shown to confer a modest survival benefit in this population. Radiation dose escalation with three-dimensional planning is not feasible, but high-dose intensity-modulated radiation therapy (IMRT) has been shown to improve local control. Still, dose-escalation remains limited by gastrointestinal toxicity. In this study, the authors investigate the potential use of double scattering (DS) and pencil beam scanning (PBS) proton therapy in limiting dose to critical organs at risk.Methods:
The authors compared DS, PBS, and IMRT plans in 13 patients with unresectable cancer of the pancreatic head, paying particular attention to duodenum, small intestine, stomach, liver, kidney, and cord constraints in addition to target volume coverage. All plans were calculated to 5500 cGy in 25 fractions with equivalent constraints and normalized to prescription dose. All statistics were by two-tailed paired t-test.Results:
Both DS and PBS decreased stomach, duodenum, and small bowel dose in low-dose regions compared to IMRT (p < 0.01). However, protons yielded increased doses in the mid to high dose regions (e.g., 23.6–53.8 and 34.9–52.4 Gy for duodenum using DS and PBS, respectively; p < 0.05). Protons also increased generalized equivalent uniform dose to duodenum and stomach, however these differences were small (<5% and 10%, respectively; p < 0.01). Doses to other organs-at-risk were within institutional constraints and placed no obvious limitations on treatment planning.Conclusions:
Proton therapy does not appear to reduce OAR volumes receiving high dose. Protons are able to reduce the treated volume receiving low-intermediate doses, however the clinical significance of this remains to be determined in future investigations.
41(2014); http://dx.doi.org/10.1118/1.4886757View Description Hide Description
In this study, the authors introduce and demonstrate quality control procedures for evaluating the geometric and dosimetric fidelity of dynamic treatment delivery techniques involving treatment couch motion synchronous with gantry and multileaf collimator (MLC). Tests were designed to evaluate positional accuracy, velocity constancy and accuracy for dynamic couch motion under a realistic weight load. A test evaluating the geometric accuracy of the system in delivering treatments over complex dynamic trajectories was also devised. Custom XML scripts that control the Varian TrueBeam™ STx (Serial #3) axes in Developer Mode were written to implement the delivery sequences for the tests. Delivered dose patterns were captured with radiographic film or the electronic portal imaging device. The couch translational accuracy in dynamic treatment mode was 0.01 cm. Rotational accuracy was within 0.3°, with 0.04 cm displacement of the rotational axis. Dose intensity profiles capturing the velocity constancy and accuracy for translations and rotation exhibited standard deviation and maximum deviations below 3%. For complex delivery involving MLC and couch motions, the overall translational accuracy for reproducing programmed patterns was within 0.06 cm. The authors conclude that in Developer Mode, TrueBeam™ is capable of delivering dynamic treatment delivery techniques involving couch motion with good geometric and dosimetric fidelity.
An initial study on the estimation of time-varying volumetric treatment images and 3D tumor localization from single MV cine EPID images41(2014); http://dx.doi.org/10.1118/1.4889779View Description Hide DescriptionPurpose:
In this work the authors develop and investigate the feasibility of a method to estimate time-varying volumetric images from individual MV cine electronic portal image device (EPID) images.Methods:
The authors adopt a two-step approach to time-varying volumetric image estimation from a single cine EPID image. In the first step, a patient-specific motion model is constructed from 4DCT. In the second step, parameters in the motion model are tuned according to the information in the EPID image. The patient-specific motion model is based on a compact representation of lung motion represented in displacement vector fields (DVFs). DVFs are calculated through deformable image registration (DIR) of a reference 4DCT phase image (typically peak-exhale) to a set of 4DCT images corresponding to different phases of a breathing cycle. The salient characteristics in the DVFs are captured in a compact representation through principal component analysis (PCA). PCA decouples the spatial and temporal components of the DVFs. Spatial information is represented in eigenvectors and the temporal information is represented by eigen-coefficients. To generate a new volumetric image, the eigen-coefficients are updated via cost function optimization based on digitally reconstructed radiographs and projection images. The updated eigen-coefficients are then multiplied with the eigenvectors to obtain updated DVFs that, in turn, give the volumetric image corresponding to the cine EPID image.Results:
The algorithm was tested on (1) Eight digital eXtended CArdiac-Torso phantom datasets based on different irregular patient breathing patterns and (2) patient cine EPID images acquired during SBRT treatments. The root-mean-squared tumor localization error is (0.73 ± 0.63 mm) for the XCAT data and (0.90 ± 0.65 mm) for the patient data.Conclusions:
The authors introduced a novel method of estimating volumetric time-varying images from single cine EPID images and a PCA-based lung motion model. This is the first method to estimate volumetric time-varying images from single MV cine EPID images, and has the potential to provide volumetric information with no additional imaging dose to the patient.
Utilization of optical tracking to validate a software-driven isocentric approach to robotic couch movements for proton radiotherapy41(2014); http://dx.doi.org/10.1118/1.4890588View Description Hide DescriptionPurpose:
An optical tracking and positioning system (OTPS) was developed to validate the software-driven isocentric (SDI) approach to control the six-degrees-of-freedom movement of a robotic couch.Methods:
The SDI approach to movements rotating around a predefined isocenter, referred to as a GeoIso, instead of a mechanical pivot point was developed by the robot automation industry. With robotic couch-sag corrections for weight load in a traditional SDI approach, movements could be accurately executed for a GeoIso located within a 500 mm cubic volume on the couch for treatments. The accuracy of SDI movement was investigated using the OTPS. The GeoIso was assumed to align with the proton beam isocenter (RadIso) for gantry at the reference angle. However, the misalignment between GeoIso and RadIso was quantitatively investigated by measuring the displacements at various couch angles for a target placed at the RadIso at an initial couch angle. When circular target displacements occur on a plane, a relative isocenter shift (RIS) correction could be applied in the SDI movement to minimize target displacements. Target displacements at a fixed gantry angle without and with RIS correction were measured for 12 robotic couches. Target displacements for various gantry angles were performed on three couches in gantry rooms to study the gantry-induced RadIso shift. The RIS correction can also be applied for the RadIso shift. A new SDI approach incorporating the RIS correction with the couch sag is described in this study. In parallel, the accuracy of SDI translation movements for various weight loads of patients on the couch was investigated during positioning of patients for proton prostate treatments.Results:
For a fixed gantry angle, measured target displacements without RIS correction for couch rotations in the horizontal plane varied from 4 to 20 mm. However, measured displacements perpendicular to couch rotation plane were about 2 mm for all couches. Extracted misalignments of GeoIso and RadIso in the horizontal plane were about 10 mm for one couch and within 3 mm for the rest of couches. After applying the RIS correction, the residual target displacements for couch rotations were within 0.5 mm to RadIso for all couches. For various gantry angles, measured target location for each angle was within 0.5 mm to its excepted location by the preset RadIso shift. Measured target displacements for ±30° of couch rotations were within 0.5 mm for gantry angles at 0° and 180°. Overall, nearly 85% of couch movements were within 0.5 mm in the horizontal plane and 0.7 mm vector distance from required displacements.Conclusions:
The authors present an optical tracking methodology to quantify for software-driven isocentric movements of robotic couches. By applying proper RIS correction for misaligned GeoIso and RadIso for each couch, and the RadIso shifts for a moving gantry, residual target displacements for isocentric couch movements around the actual RadIso can be reduced to submillimeter tolerance.
41(2014); http://dx.doi.org/10.1118/1.4890592View Description Hide DescriptionPurpose:
To compare the dosimetric performance of Acuros XB (AXB), anisotropic analytical algorithm (AAA), and x-ray voxel Monte Carlo (XVMC) in heterogeneous phantoms and lung stereotactic body radiotherapy (SBRT) plans.Methods:
Water- and lung-equivalent phantoms were combined to evaluate the percentage depth dose and dose profile. The radiation treatment machine Novalis (BrainLab AG, Feldkirchen, Germany) with an x-ray beam energy of 6 MV was used to calculate the doses in the composite phantom at a source-to-surface distance of 100 cm with a gantry angle of 0°. Subsequently, the clinical lung SBRT plans for the 26 consecutive patients were transferred from the iPlan (ver. 4.1; BrainLab AG) to the Eclipse treatment planning systems (ver. 11.0.3; Varian Medical Systems, Palo Alto, CA). The doses were then recalculated with AXB and AAA while maintaining the XVMC-calculated monitor units and beam arrangement. Then the dose-volumetric data obtained using the three different radiation dose calculation algorithms were compared.Results:
The results from AXB and XVMC agreed with measurements within ±3.0% for the lung-equivalent phantom with a 6 × 6 cm2 field size, whereas AAA values were higher than measurements in the heterogeneous zone and near the boundary, with the greatest difference being 4.1%. AXB and XVMC agreed well with measurements in terms of the profile shape at the boundary of the heterogeneous zone. For the lung SBRT plans, AXB yielded lower values than XVMC in terms of the maximum doses of ITV and PTV; however, the differences were within ±3.0%. In addition to the dose-volumetric data, the dose distribution analysis showed that AXB yielded dose distribution calculations that were closer to those with XVMC than did AAA. Means ± standard deviation of the computation time was 221.6 ± 53.1 s (range, 124–358 s), 66.1 ± 16.0 s (range, 42–94 s), and 6.7 ± 1.1 s (range, 5–9 s) for XVMC, AXB, and AAA, respectively.Conclusions:
In the phantom evaluations, AXB and XVMC agreed better with measurements than did AAA. Calculations differed in the density-changing zones (substance boundaries) between AXB/XVMC and AAA. In the lung SBRT cases, a comparative analysis of dose-volumetric data and dose distributions with XVMC demonstrated that the AXB provided better agreement with XVMC than AAA. The computation time of AXB was faster than that of XVMC; therefore, AXB has better balance in terms of the dosimetric performance and computation speed for clinical use than XVMC.
Dose-to-water conversion for the backscatter-shielded EPID: A frame-based method to correct for EPID energy response to MLC transmitted radiation41(2014); http://dx.doi.org/10.1118/1.4890677View Description Hide DescriptionPurpose:
To develop a frame-by-frame correction for the energy response of amorphous silicon electronic portal imaging devices (a-Si EPIDs) to radiation that has transmitted through the multileaf collimator (MLC) and to integrate this correction into the backscatter shielded EPID (BSS-EPID) dose-to-water conversion model.Methods:
Individual EPID frames were acquired using a Varian frame grabber and iTools acquisition software then processed using in-house software developed inMATLAB. For each EPID image frame, the region below the MLC leaves was identified and all pixels in this region were multiplied by a factor of 1.3 to correct for the under-response of the imager to MLC transmitted radiation. The corrected frames were then summed to form a corrected integrated EPID image. This correction was implemented as an initial step in the BSS-EPID dose-to-water conversion model which was then used to compute dose planes in a water phantom for 35 IMRT fields. The calculated dose planes, with and without the proposed MLC transmission correction, were compared to measurements in solid water using a two-dimensional diode array.Results:
It was observed that the integration of the MLC transmission correction into the BSS-EPID dose model improved agreement between modeled and measured dose planes. In particular, the MLC correction produced higher pass rates for almost all Head and Neck fields tested, yielding an average pass rate of 99.8% for 2%/2 mm criteria. A two-sample independentt-test and fisher F-test were used to show that the MLC transmission correction resulted in a statistically significant reduction in the mean and the standard deviation of the gamma values, respectively, to give a more accurate and consistent dose-to-water conversion.Conclusions:
The frame-by-frame MLC transmission response correction was shown to improve the accuracy and reduce the variability of the BSS-EPID dose-to-water conversion model. The correction may be applied as a preprocessing step in any pretreatment portal dosimetry calculation and has been shown to be beneficial for highly modulated IMRT fields.
- RADIATION IMAGING PHYSICS
Computerized detection of noncalcified plaques in coronary CT angiography: Evaluation of topological soft gradient prescreening method and luminal analysis41(2014); http://dx.doi.org/10.1118/1.4885958View Description Hide DescriptionPurpose:
The buildup of noncalcified plaques (NCPs) that are vulnerable to rupture in coronary arteries is a risk for myocardial infarction. Interpretation of coronary CT angiography (cCTA) to search for NCP is a challenging task for radiologists due to the low CT number of NCP, the large number of coronary arteries, and multiple phase CT acquisition. The authors conducted a preliminary study to develop machine learning method for automated detection of NCPs in cCTA.Methods:
With IRB approval, a data set of 83 ECG-gated contrast enhanced cCTA scans with 120 NCPs was collected retrospectively from patient files. A multiscale coronary artery response and rolling balloon region growing (MSCAR-RBG) method was applied to each cCTA volume to extract the coronary arterial trees. Each extracted vessel was reformatted to a straightened volume composed of cCTA slices perpendicular to the vessel centerline. A topological soft-gradient (TSG) detection method was developed to prescreen for NCP candidates by analyzing the 2D topological features of the radial gradient field surface along the vessel wall. The NCP candidates were then characterized by a luminal analysis that used 3D geometric features to quantify the shape information and gray-level features to evaluate the density of the NCP candidates. With machine learning techniques, useful features were identified and combined into an NCP score to differentiate true NCPs from false positives (FPs). To evaluate the effectiveness of the image analysis methods, the authors performed tenfold cross-validation with the available data set. Receiver operating characteristic (ROC) analysis was used to assess the classification performance of individual features and the NCP score. The overall detection performance was estimated by free response ROC (FROC) analysis.Results:
With our TSG prescreening method, a prescreening sensitivity of 92.5% (111/120) was achieved with a total of 1181 FPs (14.2 FPs/scan). On average, six features were selected during the tenfold cross-validation training. The average area under the ROC curve (AUC) value for training was 0.87 ± 0.01 and the AUC value for validation was 0.85 ± 0.01. Using the NCP score, FROC analysis of the validation set showed that the FP rates were reduced to 3.16, 1.90, and 1.39 FPs/scan at sensitivities of 90%, 80%, and 70%, respectively.Conclusions:
The topological soft-gradient prescreening method in combination with the luminal analysis for FP reduction was effective for detection of NCPs in cCTA, including NCPs causing positive or negative vessel remodeling. The accuracy of vessel segmentation, tracking, and centerline identification has a strong impact on NCP detection. Studies are underway to further improve these techniques and reduce the FPs of the CADe system.
Task-based detectability in CT image reconstruction by filtered backprojection and penalized likelihood estimation41(2014); http://dx.doi.org/10.1118/1.4883816View Description Hide DescriptionPurpose:
Nonstationarity is an important aspect of imaging performance in CT and cone-beam CT (CBCT), especially for systems employing iterative reconstruction. This work presents a theoretical framework for both filtered-backprojection (FBP) and penalized-likelihood (PL) reconstruction that includes explicit descriptions of nonstationary noise, spatial resolution, and task-based detectability index. Potential utility of the model was demonstrated in the optimal selection of regularization parameters in PL reconstruction.Methods:
Analytical models for local modulation transfer function (MTF) and noise-power spectrum (NPS) were investigated for both FBP and PL reconstruction, including explicit dependence on the object and spatial location. For FBP, a cascaded systems analysis framework was adapted to account for nonstationarity by separately calculating fluence and system gains for each ray passing through any given voxel. For PL, the point-spread function and covariance were derived using the implicit function theorem and first-order Taylor expansion according toFessler [“Mean and variance of implicitly defined biased estimators (such as penalized maximum likelihood): Applications to tomography,” IEEE Trans. Image Process.5(3), 493–506 (1996)]. Detectability index was calculated for a variety of simple tasks. The model for PL was used in selecting the regularization strength parameter to optimize task-based performance, with both a constant and a spatially varying regularization map.Results:
Theoretical models of FBP and PL were validated in 2D simulated fan-beam data and found to yield accurate predictions of local MTF and NPS as a function of the object and the spatial location. The NPS for both FBP and PL exhibit similar anisotropic nature depending on the pathlength (and therefore, the object and spatial location within the object) traversed by each ray, with the PL NPS experiencing greater smoothing along directions with higher noise. The MTF of FBP is isotropic and independent of location to a first order approximation, whereas the MTF of PL is anisotropic in a manner complementary to the NPS. Task-based detectability demonstrates dependence on the task, object, spatial location, and smoothing parameters. A spatially varying regularization “map” designed from locally optimal regularization can improve overall detectability beyond that achievable with the commonly used constant regularization parameter.Conclusions:
Analytical models for task-based FBP and PL reconstruction are predictive of nonstationary noise and resolution characteristics, providing a valuable framework for understanding and optimizing system performance in CT and CBCT.