Index of content:
Volume 37, Issue 11, November 2010
37(2010); http://dx.doi.org/10.1118/1.3481357View Description Hide Description
- MEDICAL PHYSICS LETTERS
37(2010); http://dx.doi.org/10.1118/1.3491675View Description Hide DescriptionPurpose:
To develop a novel aperture-based algorithm for volumetric modulated arc therapy (VMAT) treatment plan optimization with high quality and high efficiency.Methods:
The VMAT optimization problem is formulated as a large-scale convex programming problem solved by a column generation approach. The authors consider a cost function consisting two terms, the first enforcing a desired dose distribution and the second guaranteeing a smooth dose rate variation between successive gantry angles. A gantry rotation is discretized into 180 beam angles and for each beam angle, only one MLC aperture is allowed. The apertures are generated one by one in a sequential way. At each iteration of the column generation method, a deliverable MLC aperture is generated for one of the unoccupied beam angles by solving a subproblem with the consideration of MLC mechanic constraints. A subsequent master problem is then solved to determine the dose rate at all currently generated apertures by minimizing the cost function. When all 180 beam angles are occupied, the optimization completes, yielding a set of deliverable apertures and associated dose rates that produce a high quality plan.Results:
The algorithm was preliminarily tested on five prostate and five head-and-neck clinical cases, each with one full gantry rotation without any couch/collimator rotations. High quality VMAT plans have been generated for all ten cases with extremely high efficiency. It takes only 5–8 min on CPU (MATLAB code on an Intel Xeon 2.27 GHz CPU) and 18–31 s on GPU (CUDA code on an NVIDIA Tesla C1060 GPU card) to generate such plans.Conclusions:
The authors have developed an aperture-based VMAT optimization algorithm which can generate clinically deliverable high quality treatment plans at very high efficiency.
- RADIATION THERAPY PHYSICS
Statistical analysis of dose heterogeneity in circulating blood: Implications for sequential methods of total body irradiation37(2010); http://dx.doi.org/10.1118/1.3495816View Description Hide DescriptionPurpose:
Improvements in delivery techniques for total body irradiation (TBI) using Tomotherapy® and intensity modulated radiation therapy have been proven feasible. Despite the promise of improved dose conformality, the application of these “sequential” techniques has been hampered by concerns over dose heterogeneity to circulating blood. The present study was conducted to provide quantitative evidence regarding the potential clinical impact of this heterogeneity.Methods:
Blood perfusion was modeled analytically as possessing linear, sinusoidal motion in the craniocaudal dimension. The average perfusion period for human circulation was estimated to be approximately 78 s. Sequential treatment delivery was modeled as a Gaussian-shaped dose cloud with a 10 cm length that traversed a 183 cm patient length at a uniform speed. Total dose to circulating blood voxels was calculated via numerical integration and normalized to 2 Gy per fraction. Dose statistics and equivalent uniform dose (EUD) were calculated for relevant treatment times, radiobiological parameters, blood perfusion rates, and fractionation schemes. The model was then refined to account for random dispersion superimposed onto the underlying periodic blood flow. Finally, a fully stochastic model was developed using binomial and trinomial probability distributions. These models allowed for the analysis of nonlinear sequential treatment modalities and treatment designs that incorporate deliberate organ sparing.Results:
The dose received by individual blood voxels exhibited asymmetric behavior that depended on the coherence among the blood velocity, circulation phase, and the spatiotemporal characteristics of the irradiation beam. Heterogeneity increased with the perfusion period and decreased with the treatment time. Notwithstanding, heterogeneity was less than for perfusion periods less than 150 s. The EUD was compromised for radiosensitive cells, long perfusion periods, and short treatment times. However, the EUD was unaffected (within 10%) for perfusion periods of less than 150 s or treatment times of 20 min or greater. Treatment over six fractions improved the EUD per fraction such that all parametric combinations resulted in unaffected EUD. The stochastic models confirmed these results.Conclusions:
Dose heterogeneity in circulating blood cells is clinically acceptable for typical treatment times, perfusion rates, and cell types. Development of conformal, sequential TBI treatment techniques should not be withheld based on concerns over circulating blood dose heterogeneity.
37(2010); http://dx.doi.org/10.1118/1.3497152View Description Hide DescriptionPurpose:
Both spatial and biological information are necessary in order to perform true optimization of a treatment plan and for predicting clinical outcome. The goal of this work is to develop an enhanced treatment plan evaluation tool which incorporates biological parameters and retains spatialdoseinformation.Methods:
A software system is developed which provides biological plan evaluation with a novel combination of features. It incorporates hyper-radiosensitivity using the induced-repair model and applies the new concept of dose convolution filter (DCF) to simulate dose wash-out effects due to cell migration, bystander effect, and/or tissue motion during treatment. Further, the concept of spatial DVH (sDVH) is introduced to evaluate and potentially optimize the spatialdose distribution in the target volume. Finally, generalized equivalent uniform dose is derived from both the physical dose distribution (gEUD) and the distribution of equivalent dose in 2 Gy fractions and the software provides three separate models for calculation of tumor control probability (TCP), normal tissue complication probability (NTCP), and probability of uncomplicated tumor control . TCP, NTCP, and are provided as a function of prescribed dose and multivariable TCP, NTCP, and plots are provided to illustrate the dependence on individual parameters used to calculate these quantities. Ten plans from two clinical treatment sites are selected to test the three calculation models provided by this software.Results:
By retaining both spatial and biological information about the dose distribution, the software is able to distinguish features of radiotherapy treatment plans not discernible using commercial systems. Plans that have similar DVHs may have different spatial and biological characteristics and the application of novel tools such as sDVH and DCF within the software may substantially change the apparent plan quality or predicted plan metrics such as TCP and NTCP. For the cases examined, both the calculation method and the application of DCF can change the ranking order of competing plans. The voxel-by-voxel TCP model makes it feasible to incorporate spatial variations of clonogen densities, radiosensitivities , and fractionation sensitivities as those data become available.Conclusions:
The new software incorporates both spatial and biological information into the treatment planning process. The application of multiple methods for the incorporation of biological and spatialinformation has demonstrated that the order of application of biological models can change the order of plan ranking. Thus, the results of plan evaluation and optimization are dependent not only on the models used but also on the order in which they are applied. This software can help the planner choose more biologically optimal treatment plans and potentially predict treatment outcome more accurately.
37(2010); http://dx.doi.org/10.1118/1.3490083View Description Hide DescriptionPurpose:
Dose calculation is a key component in radiation treatment planning systems. Its performance and accuracy are crucial to the quality of treatment plans as emerging advanced radiation therapy technologies are exerting ever tighter constraints on dose calculation. A common practice is to choose either a deterministic method such as the convolution/superposition (CS) method for speed or a Monte Carlo(MC) method for accuracy. The goal of this work is to boost the performance of a hybrid Monte Carlo convolution/superposition (MCCS) method by devising a graphics processing unit (GPU) implementation so as to make the method practical for day-to-day usage.Methods:
Although the MCCS algorithm combines the merits of MC fluence generation and CS fluence transport, it is still not fast enough to be used as a day-to-day planning tool. To alleviate the speed issue of MC algorithms, the authors adopted MCCS as their target method and implemented a GPU-based version. In order to fully utilize the GPU computing power, the MCCS algorithm is modified to match the GPU hardware architecture. The performance of the authors’ GPU-based implementation on an Nvidia GTX260 card is compared to a multithreaded software implementation on a quad-core system.Results:
A speedup in the range of 6.7–11.4× is observed for the clinical cases used. The less than 2% statistical fluctuation also indicates that the accuracy of the authors’ GPU-based implementation is in good agreement with the results from the quad-core CPU implementation.Conclusions:
This work shows that GPU is a feasible and cost-efficient solution compared to other alternatives such as using cluster machines or field-programmable gate arrays for satisfying the increasing demands on computation speed and accuracy of dose calculation. But there are also inherent limitations of using GPU for accelerating MC-type applications, which are also analyzed in detail in this article.
37(2010); http://dx.doi.org/10.1118/1.3495539View Description Hide DescriptionPurpose:
Both the AccuBoost® D-shaped and round applicators have been dosimetrically characterized and clinically used to treat patients with breast cancer. While the round applicators provide conformal dose coverage, under certain clinical circumstances the breast skindose may be higher than preferred. The purpose of this study was to modify the round applicators to minimize skindose while not substantially affecting dose uniformity within the target volume and reducing the treatment time.Methods:
In order to irradiate the intended volume while sparing critical structures such as the skin, the current round applicator design has been augmented through the addition of an internal truncated cone (i.e., frustum) shield. Monte Carlo methods and clinical constraints were used to design the optimal cone applicator. With the cone applicator now defined as the entire assembly including the surrounding tungsten-alloy shell holding the HDR source catheter, the applicator height was reduced to diminish the treatment time while minimizing skindose.Monte Carlo simulation results were validated using both radiochromic film and ionization chamber measurements based on established techniques.Results:
The optimal cone applicators diminished the maximum skindose by 15%–32% (based on the applicator diameter and breast separation) with the tumordose reduced by less than 3% for a constant exposure time. Furthermore, reduction in applicator height diminished the treatment time by up to 30%. Radiochromic film and ionization chamberdosimetric results in phantom agreed with Monte Carlo simulation results typically within 3%. Larger differences were outside the treatment volume in low dose regions or associated with differences between the measurement and Monte Carlo simulation environments.Conclusions:
A new radiotherapytreatment device was developed and dosimetrically characterized. This set of applicators significantly reduces the skindose and treatment time while retaining uniform target dose.
37(2010); http://dx.doi.org/10.1118/1.3501313View Description Hide DescriptionPurpose:
A project to construct a new treatment facility, as an extension of the existing HIMAC facility, has been initiated for the further development of carbon-ion therapy at NIRS. This new treatment facility is equipped with a 3D irradiation system with pencil-beam scanning. The challenge of this project is to realize treatment of a moving target by scanning irradiation. To achieve fast rescanning within an acceptable irradiation time, the authors developed a fast scanning system.Methods:
In order to verify the validity of the design and to demonstrate the performance of the fast scanning prior to use in the new treatment facility, a new scanning-irradiation system was developed and installed into the existing HIMAC physics-experiment course. The authors made strong efforts to develop (1) the fast scanning magnet and its power supply, (2) the high-speed control system, and (3) the beam monitoring. The performance of the system including 3D dose conformation was tested by using the carbon beam from the HIMAC accelerator.Results:
The performance of the fast scanning system was verified by beam tests. Precision of the scanned beam position was less than ±0.5 mm. By cooperating with the planning software, the authors verified the homogeneity of the delivered field within ±3% for the 3D delivery. This system took only 20 s to deliver the physical dose of 1 Gy to a spherical target having a diameter of 60 mm with eight rescans. In this test, the average of the spot-staying time was considerably reduced to, while the minimum staying time was .Conclusions:
As a result of this study, the authors verified that the new scanning delivery system can produce an accurate 3D dose distribution for the target volume in combination with the planning software.
37(2010); http://dx.doi.org/10.1118/1.3495969View Description Hide DescriptionPurpose:
The depth dose of a monoenergetic broad parallel protonbeam has been modeled in a number of ways, but evidently not yet for oblique incidence. The purpose of this investigation is to find an accurate analytic formula for this case, which can then be used to model the depth dose of a broad beam with an initial Gaussian angular distribution.Methods:
The Bortfeld model of depth dose in a broad normally incident protonbeam has been extended to the case of oblique incidence. This extension uses an empirically determined Gaussian parameter which (roughly) characterizes the off-axis dose of a proton pencil beam. As with Bortfeld’s work, the modeling is done in terms of parabolic cylinder functions. To obtain the depth dose for an initial angular distribution, the result is integrated over the angle of incidence, weighted by a Gaussian probability function. The predictions of the theory have been compared to MCNPXMonte Carlo calculations for four phantom materials (water, bone, aluminum, and copper) and for initial proton energies of 50, 100, 150, 200, and 250 MeV.Results:
Comparisons of the depth dose predicted by this theory with Monte Carlo calculations have established that with very good accuracy, can be taken to be independent both of the depth and of the angle of incidence. As a function of initial proton range or of initial proton energy, has been found to obey a power law to very high accuracy. Good fits to Monte Carlo calculations have also been found for an initial Gaussian angular distribution.Conclusions:
This investigation is the first step in the accurate modeling of a proton pencil beam with initial Gaussian angular distribution. It provides the longitudinal factor, with its Bragg peak buildup and sharp distal falloff. A transverse factor must still be incorporated into this theory and this will give the lateral penumbra of a collimated protonbeam. Also, it will be necessary to model the dose of product particles from nuclear interactions of the protonbeam. With the accurate modeling of a pencil beam, it will be possible to accurately take into account the effect of localized tissue inhomogeneities.
37(2010); http://dx.doi.org/10.1118/1.3495538View Description Hide DescriptionPurpose:
Pulmonary nodules present unique problems during radiation treatment due to nodule position uncertainty that is caused by respiration. The radiation field has to be enlarged to account for nodule motion during treatment. The purpose of this work is to provide a method of locating a pulmonary nodule in a megavolt portal image that can be used to reduce the internal target volume (ITV) during radiation therapy. A reduction in the ITV would result in a decrease in radiation toxicity to healthy tissue.Methods:
Eight patients with nonsmall cell lung cancer were used in this study. CT scans that include the pulmonary nodule were captured with a GE Healthcare LightSpeed RT 16 scanner. Megavolt portal images were acquired with a Varian Trilogy unit equipped with an AS1000 electronic portal imaging device. The nodule localization method uses grayscale morphological filtering and level-set segmentation with a prior. The treatment-time portion of the algorithm is implemented on a graphical processing unit.Results:
The method was retrospectively tested on eight cases that include a total of 151 megavolt portal image frames. The method reduced the nodule position uncertainty by an average of 40% for seven out of the eight cases. The treatment phase portion of the method has a subsecond execution time that makes it suitable for near-real-time nodule localization.Conclusions:
A method was developed to localize a pulmonary nodule in a megavolt portal image. The method uses the characteristics of the nodule in a prior CT scan to enhance the nodule in the portal image and to identify the nodule region by level-set segmentation. In a retrospective study, the method reduced the nodule position uncertainty by an average of 40% for seven out of the eight cases studied.
A neural network based 3D/3D image registration quality evaluator for the head-and-neck patient setup in the absence of a ground truth37(2010); http://dx.doi.org/10.1118/1.3502756View Description Hide DescriptionPurpose:
To develop a neural network based registration quality evaluator (RQE) that can identify unsuccessful 3D/3D image registrations for the head-and-neck patient setup in radiotherapy.Methods:
A two-layer feed-forward neural network was used as a RQE to classify 3D/3D rigid registration solutions as successful or unsuccessful based on the features of the similarity surface near the point-of-solution. The supervised training and test data sets were generated by rigidly registering daily cone-beam CTs to the treatment planning fan-beam CTs of six patients with head-and-neck tumors. Two different similarity metrics (mutual information and mean-squared intensity difference) and two different types of image content (entire image versus bony landmarks) were used. The best solution for each registration pair was selected from 50 optimizing attempts that differed only by the initial transformation parameters. The distance from each individual solution to the best solution in the normalized parametrical space was compared to a user-defined error threshold to determine whether that solution was successful or not. The supervised training was then used to train the RQE. The performance of the RQE was evaluated using the test data set that consisted of registration results that were not used in training.Results:
The RQE constructed using the mutual information had very good performance when tested using the test data sets, yielding the sensitivity, the specificity, the positive predictive value, and the negative predictive value in the ranges of 0.960–1.000, 0.993–1.000, 0.983–1.000, and 0.909–1.000, respectively. Adding a RQE into a conventional 3D/3D image registration system incurs only about 10%–20% increase of the overall processing time.Conclusions:
The authors’ patient study has demonstrated very good performance of the proposed RQE when used with the mutual information in identifying unsuccessful 3D/3D registrations for daily patient setup. The classifier had very good generality and required only to be trained once for each implementation. When the RQE is incorporated with an automated 3D/3D image registration system, it can improve the robustness of the system.
37(2010); http://dx.doi.org/10.1118/1.3503849View Description Hide DescriptionPurpose:
When comparing binary test results from two diagnostic systems, superiority in both “sensitivity” and “specificity” also implies differences in all conventional summary indices and locally in the underlying receiver operating characteristics (ROC) curves. However, when one of the two binary tests has higher sensitivity and lower specificity (or vice versa), comparisons of their performance levels are nontrivial and the use of different summary indices may lead to contradictory conclusions. A frequently used approach that is free of subjectivity associated with summary indices is based on the comparison of the underlying ROC curves that requires the collection of rating data using multicategory scales, whether natural or experimentally imposed. However, data for reliable estimation of ROC curves are frequently unavailable. The purpose of this article is to develop an approach of using “diagnostic likelihood ratios,” namely, likelihood ratios of “positive” or “negative” responses, to make simple inferences regarding the underlying ROC curves and associated areas in the absence of reliable rating data or regarding the relative binary characteristics, when these are of primary interest.Methods:
For inferences related to underlying curves, the authors exploit the assumption of concavity of the true underlying ROC curve to describe conditions under which these curves have to be different and under which the curves have different areas. For scenarios when the binary characteristics are of primary interest, the authors use characteristics of “chance performance” to demonstrate that the derived conditions provide strong evidence of superiority of one binary test as compared to another. By relating these derived conditions to hypotheses about the true likelihood ratios of two binary diagnostic tests being compared, the authors enable a straightforward statistical procedure for the corresponding inferences.Results:
The authors derived simple algebraic and graphical methods for describing the conditions for superiority of one of two diagnostic tests with respect to their binary characteristics, the underlying ROC curves, or the areas under the curves. The graphical regions are useful for identifying potential differences between two systems, which then have to be tested statistically. The simple statistical tests can be performed with well known methods for comparison of diagnostic likelihood ratios. The developed approach offers a solution for some of the more difficult to analyze scenarios, where diagnostic tests do not demonstrate concordant differences in terms of both sensitivity and specificity. In addition, the resulting inferences do not contradict the conclusions that can be obtained using conventional and reasonably defined summary indices.Conclusions:
When binary diagnostic tests are of primary interest, the proposed approach offers an objective and powerful method for comparing two binary diagnostic tests. The significant advantage of this method is that it enables objective analyses when one test has higher sensitivity but lower specificity, while ensuring agreement with study conclusions based on other reasonable and widely acceptable summary indices. For truly multicategory diagnostic tests, the proposed method can help in concluding inferiority of one of the diagnostic tests based on binary data, thereby potentially saving the need for conducting a more expensive multicategory ROC study.
Verification of patient-specific dose distributions in proton therapy using a commercial two-dimensional ion chamber array37(2010); http://dx.doi.org/10.1118/1.3505011View Description Hide DescriptionPurpose:
The purpose of this study was to determine whether a two-dimensional (2D) ion chamber array detector quickly and accurately measures patient-specific dose distributions in treatment with passively scattered and spot scanning proton beams.Methods:
The 2D ion chamber array detector MatriXX was used to measure the dose distributions in plastic water phantom from passively scattered and spot scanning proton beam fields planned for patient treatment. Planar dose distributions were measured using MatriXX, and the distributions were compared to those calculated using a treatment-planning system. The dose distributions generated by the treatment-planning system and a film dosimetry system were similarly compared.Results:
For passively scatteredproton beams, the gamma index for the dose-distribution comparison for treatment fields for three patients with prostate cancer and for one patient with lung cancer was less than 1.0 for 99% and 100% of pixels for a 3% dose tolerance and 3 mm distance-to-dose agreement, respectively. For spot scanning beams, the mean (± standard deviation) percentages of pixels with gamma indices meeting the passing criteria were and for MatriXX and film dosimetry, respectively, for 20 fields used to treat patients with prostate cancer.Conclusions:
Unlike film dosimetry, MatriXX provides not only 2D dose-distribution information but also absolute dosimetry in fractions of minutes with acceptable accuracy. The results of this study indicate that MatriXX can be used to verify patient-field specific dose distributions in proton therapy.
Investigation of an implantable dosimeter for single-point water equivalent path length verification in proton therapy37(2010); http://dx.doi.org/10.1118/1.3504609View Description Hide DescriptionPurpose:
In vivo range verification in proton therapy is highly desirable. A recent study suggested that it was feasible to use point dosemeasurement for in vivo beam range verification in proton therapy, provided that the spread-out Bragg peak dose distribution is delivered in a different and rather unconventional manner. In this work, the authors investigate the possibility of using a commercial implantable dosimeter with wireless reading for this particular application.Methods:
The traditional proton treatment technique delivers all the Bragg peaks required for a SOBP field in a single sequence, producing a constant dose plateau across the target volume. As a result, a point dosemeasurement anywhere in the target volume will produce the same value, thus providing no information regarding the water equivalent path length to the point of measurement. However, the same constant dose distribution can be achieved by splitting the field into a complementary pair of subfields, producing two oppositely “sloped” depth-dose distributions, respectively. The ratio between the two distributions can be a sensitive function of depth and measuring this ratio at a point inside the target volume can provide the water equivalent path length to the dosimeter location. Two types of field splits were used in the experiment, one achieved by the technique of beam current modulation and the other by manipulating the location and width of the beam pulse relative to the range modulator track. Eight MOSFET-based implantable dosimeters at four different depths in a water tank were used to measure the dose ratios for these field pairs. A method was developed to correct the effect of the well-known LET dependence of the MOSFETdetectors on the depth-dose distributions using the columnar recombination model. The LET-corrected dose ratios were used to derive the water equivalent path lengths to the dosimeter locations to be compared to physical measurements.Results:
The implantable dosimetersmeasured the dose ratios with a reasonable relative uncertainty of 1%–3% at all depths, except when the ratio itself becomes very small. In total, 55% of the individual measurements reproduced the water equivalent path lengths to the dosimeters within 1 mm. For three dosimeters, the difference was consistently less than 1 mm. Half of the standard deviations over the repeated measurements were equal or less than 1 mm.Conclusions:
With a single fitting parameter, the LET-correction method worked remarkably well for the MOSFETdetectors. The overall results were very encouraging for a potential method ofin vivo beam range verification with millimeter accuracy. This is sufficient accuracy to expand range of clinical applications in which the authors could use the distal fall off of the proton depth dose for tight margins.
37(2010); http://dx.doi.org/10.1118/1.3495537View Description Hide DescriptionPurpose:
To use EGSnrc Monte Carlo simulations to directly calculate beam quality conversion factors,, for 32 cylindrical ionization chambers over a range of beam qualities and to quantify the effect of systematic uncertainties on Monte Carlo calculations of . These factors are required to use the TG-51 or TRS-398 clinical dosimetry protocols for calibrating external radiotherapy beams.Methods:
Ionization chambers are modeled either from blueprints or manufacturers’ user’s manuals. The dose-to-air in the chamber is calculated using the EGSnrc user-code egs_chamber using 11 different tabulated clinical photon spectra for the incident beams. The dose to a small volume of water is also calculated in the absence of the chamber at the midpoint of the chamber on its central axis. Using a simple equation, is calculated from these quantities under the assumption that is constant with energy and compared to TG-51 protocol and measured values.Results:
Polynomial fits to the Monte Carlo calculated factors as a function of beam quality expressed as and are given for each ionization chamber. Differences are explained between Monte Carlo calculated values and values from the TG-51 protocol or calculated using the computer program used for TG-51 calculations. Systematic uncertainties in calculated values are analyzed and amount to a maximum of one standard deviation uncertainty of 0.99% if one assumes that photon cross-section uncertainties are uncorrelated and 0.63% if they are assumed correlated. The largest components of the uncertainty are the constancy of and the uncertainty in the cross-section for photons in water.Conclusions:
It is now possible to calculate directly using Monte Carlo simulations.Monte Carlo calculations for most ionization chambers give results which are comparable to TG-51 values. Discrepancies can be explained using individual Monte Carlo calculations of various correction factors which are more accurate than previously used values. For small ionization chambers with central electrodes composed of high-Z materials, the effect of the central electrode is much larger than that for the aluminumelectrodes in Farmer chambers.
The effect of an inconsistent breathing amplitude on the relationship between an external marker and internal lung deformation in a porcine model37(2010); http://dx.doi.org/10.1118/1.3496325View Description Hide DescriptionPurpose:
Investigate the relationship between the motion of the Varian Real-time Position Management (RPM) device and the internal motion of a pig during induced inconsistencies in the amplitude of breathing.Methods:
Twelve studies were performed on four ventilated female Landrace cross pigs using a GE Healthcare, Discovery CT 750 HD scanner. In each study, a 4.0 cm section (64 slices) of the pig’s lungs was repeatedly scanned 20 times using cine mode, each time lasting more than one breathing cycle. During these cine scans, a Varian RPM device was used to collect respiratory amplitudes and the ventilator air return tube was periodically crimped to induce inconsistent breathing amplitudes. Each breathing cycle and its associated cine scan were categorized as either consistent or inconsistent, based on thresholds of the minimum expiration and maximum inspiration amplitudes. From the group of consistent amplitude cine scans in a study, a reference scan was chosen. The effect of inconsistent breathing amplitudes on the relationship between the motion of the RPM marker and the motion within three regions of interest (in each lung and the chest wall) was investigated with two methods: (1) A 4D-CT sorting algorithm based on RPM amplitude was used to sort volumes into 4D-CT phase bins. Within each phase bin, the nonlinear deformation of volumes collected during consistent and inconsistent breathing amplitude was calculated with respect to the reference volume. The magnitude of the deformations (in mm) were compared to determine if inconsistent breathing amplitude caused greater deformations. (2) Nonlinear deformations between each CT volume from a cine scan and the maximum expiration volume of the reference scan were calculated. Regression analyses between the nonlinear deformations within three regions of interest (in each lung and the chest wall) and the RPM amplitudes were performed to test the effect of inconsistent breathing amplitudes on the linearity of the relationship between the 3D motion of internal anatomy and the 1D motion of the RPM external marker.Results:
(1) Inconsistent versus consistent breathing amplitudes caused a significant increase in deformation relative to the reference scan within the left lung ( versus , ). (2) One-to-one correspondences between motions of internal anatomies and motion of the RPM external marker did not exist. The regression lines between the two types of motions did not yield an identity relationship (unity slope and zero intercept). Inconsistent breathing produced significantly different regression lines than consistent breathing in ten of the 12 studies within a left lung region of interest.Conclusions:
The results of these two studies indicate that inconsistency in the amplitude of breathing disrupted the correspondence between the motion of the external marker and internal anatomies. As a consequence, radiation therapy of tumors embedded in lungtissue may be prone to significant errors if inconsistent breathing amplitudes occur during treatment.
37(2010); http://dx.doi.org/10.1118/1.3490477View Description Hide DescriptionPurpose:
A learning-based approach integrating the use of pixel-level statistical modeling and spiculation detection is presented for the segmentation of mammographic masses with ill-defined margins and spiculations.Methods:
The algorithm involves a multiphase pixel-level classification, using a comprehensive group of features computed from regional intensity, shape, and textures, to generate a mass-conditional probability map (PM). Then, the mass candidate, along with the background clutters consisting of breast fibroglandular and other nonmass tissues, is extracted from the PM by integrating the prior knowledge of shape and location of masses. A multiscale steerable ridge detection algorithm is employed to detect spiculations. Finally, all the object-level findings, including mass candidate, detected spiculations, and clutters, along with the PM, are integrated by graph cuts to generate the final segmentation mask.Results:
The method was tested on 54 masses (51 malignant and 3 benign), all with ill-defined margins and irregular shape or spiculations. The ground truth delineations were provided by five experienced radiologists. Area overlapping ratio of 0.689 (±0.160) and 0.540 (±0.164) were obtained for segmenting entire mass and margin portion only, respectively. Williams index of area and contour based measurements indicated that the segmentation results of the algorithm agreed well with the radiologists’ delineation.Conclusions:
The proposed approach could closely delineate the mass body. Most importantly, it is capable of including mass margin and its spicule extensions which are considered as key features for breast lesion analyses.
37(2010); http://dx.doi.org/10.1118/1.3505452View Description Hide DescriptionPurpose:
The aim of this work was to characterize a multi-axis ion chamber array (IC PROFILER™; Sun Nuclear Corporation, Melbourne, FL USA) that has the potential to simplify the acquisition of LINAC beam data.Methods:
The IC PROFILER™ (orpanel) measurement response was characterized with respect to radiation beam properties, including dose,dose per pulse, pulse rate frequency (PRF), and energy. Panel properties were also studied, including detector-calibration stability, power-on time, backscatter dependence, and the panel’s agreement with water tank measurements [profiles, fractional depth dose (FDD), and output factors].Results:
The panel’s relative deviation was typically within (±) 1% of an independent (or nominal) response for all properties that were tested. Notable results were (a) a detectable relative field shape change of with linear accelerator PRF changes; (b) a large range in backscatter thickness had a minimal effect on the measured dose distribution (typically less than 1%); (c) the error spread in profile comparison between the panel and scanning water tank (Blue Phantom™, CC13™; IBA Schwarzenbruck, DE) was approximately (±) 0.75%.Conclusions:
The ability of the panel to accurately reproduce water tank profiles, FDDs, and output factors is an indication of its abilities as a dosimetry system. The benefits of using the panel versus a scanning water tank are less setup time and less error susceptibility. The same measurements (including device setup and breakdown) for both systems took 180 min with the water tank versus 30 min with the panel. The time-savings increase as the measurement load is increased.
- RADIATION IMAGING PHYSICS
Evaluation of an improved algorithm for producing realistic 3D breast software phantoms: Application for mammography37(2010); http://dx.doi.org/10.1118/1.3491812View Description Hide DescriptionPurpose:
This work presents an improved algorithm for the generation of 3D breast software phantoms and its evaluation for mammography.Methods:
The improved methodology has evolved from a previously presented 3D noncompressed breast modeling method used for the creation of breast models of different size, shape, and composition. The breast phantom is composed of breast surface, duct system and terminal ductal lobular units, Cooper’s ligaments, lymphatic and blood vessel systems, pectoral muscle, skin, 3D mammographic background texture, and breast abnormalities. The key improvement is the development of a new algorithm for 3D mammographic texture generation. Simulated images of the enhanced 3D breast model without lesions were produced by simulating mammographicimage acquisition and were evaluated subjectively and quantitatively. For evaluation purposes, a database with regions of interest taken from simulated and real mammograms was created. Four experienced radiologists participated in a visual subjective evaluation trial, as they judged the quality of the simulated mammograms, using the new algorithm compared to mammograms, obtained with the old modeling approach. In addition, extensive quantitative evaluation included power spectral analysis and calculation of fractal dimension, skewness, and kurtosis of simulated and real mammograms from the database.Results:
The results from the subjective evaluation strongly suggest that the new methodology for mammographic breast texture creates improved breast models compared to the old approach. Calculated parameters on simulated images such as exponent deducted from the power law spectral analysis and fractal dimension are similar to those calculated on real mammograms. The results for the kurtosis and skewness are also in good coincidence with those calculated from clinical images. Comparison with similar calculations published in the literature showed good agreement in the majority of cases.Conclusions:
The improved methodology generated breast models with increased realism compared to the older model as shown in evaluations of simulated images by experienced radiologists. It is anticipated that the realism will be further improved using an advanced image simulator so that simulated images may be used in feasibility studies in mammography.
37(2010); http://dx.doi.org/10.1118/1.3488899View Description Hide DescriptionPurpose:
The purpose of this work was to evaluate the visibility of simulated lesions in 2D digital mammography (DM) and breast tomosynthesis (BT) images of patients.Methods:
Images of the same women were acquired on both a DM system (Mammomat Novation, Siemens Healthcare, Erlangen, Germany) and a BT prototype system adapted from the same type of DM system. Using the geometrical properties of the two systems, simulated lesions were projected and added to each DM image as well as to each BT projection image prior to 3D reconstruction. The same beam quality and approximately the same total absorbed dose to the glandular tissue were used for each breast image acquisition on the two systems. A series of four-alternative forced choice human observer experiments was conducted for each of five simulated lesion diameters: 0.2, 1, 3, 8, and 25 mm. An additional experiment was conducted for the 0.2 mm lesion in BT only at twice the dose level. Threshold signal was defined as the lesion signal intensity required for a detectability index of 2.5. Four medical physicists participated in all experiments. One experiment, consisting of 60 cases, was conducted per test condition (i.e., lesion size and signal combination).Results:
For the smallest lesions (0.2 mm), the threshold signal for DM was 21% lower than for BT at equivalent dose levels, and was 26% lower than DM. For the lesions larger than 1 mm, the threshold signal increased linearly (in log space) with the lesion diameter for both DM and BT, with DM requiring around twice the signal as BT. The difference in the threshold signal between BT and DM at each lesion size was statistically significant, except for the 0.2 mm lesion between and DM.Conclusions:
The results of this study indicate that low-signal lesions larger than 1.0 mm may be more visible in BT compared to DM, whereas 0.2 mm lesions may be better visualized with DM compared to BT, when compared at equal dose.
Combining scatter reduction and correction to improve image quality in cone-beam computed tomography (CBCT)37(2010); http://dx.doi.org/10.1118/1.3497272View Description Hide DescriptionPurpose:
The authors propose a combined scatter reduction and correction method to improve image quality in cone-beam computed tomography(CBCT). Although using a beam-block approach similar to previous techniques to measure the scatter, this method differs in that the authors utilize partially blocked projection data obtained during scatter measurement for CBCTimage reconstruction. This study aims to evaluate the feasibility of the proposed approach.Methods:
A 1D grid, composed of lead septa, was placed between the radiation source and the imaging object for scatter measurement. Image data were collected from the grid interspace regions while the scatter distribution was measured in the blocked regions under the grid. Scatter correction was performed by subtracting the measured scatter from the imaging data. Image information in the penumbral regions of the grid was derived. Three imaging modes were developed to reconstruct full CBCTimages from partial projection data. The single-rotation half-fan mode uses interpolation to fill the missing data. The dual-rotation half-fan mode uses two rotations, with the grid offset by half a grid cycle, to acquire two complementary sets of projections, which are then merged to form complete projections for reconstruction. The single-rotation full-fan mode was designed for imaging a small object or a region of interest. Full-fan projection images were acquired over a 360° scan angle with the grid shifting a distance during the scan. An enlarged Catphan phantom was used to evaluate potential improvement in image quality with the proposed technique. An anthropomorphic pelvis phantom was used to validate the feasibility of reconstructing a complete set of CBCTimages from the partially blocked projections using three imaging modes. Rigid-body image registration was performed between the CBCTimages from the single-rotation half-fan mode and the simulation CT and the results were compared to that for the CBCTimages from dual-rotation mode and conventional CBCTimages.Results:
The proposed technique reduced the streak artifact index from 58% to 1% in comparison with the conventional CBCT. It also improved CT number linearity from 0.880 to 0.998 and the contrast-to-noise ratio(CNR) from 4.29 to 6.42. Complete sets of CBCTimages with overall improved image quality were achieved for all three image modes. The longitudinal resolution was slightly compromised for the single-rotation half-fan mode. High resolution was retained for the dual-rotation half-fan and single-rotation full-fan modes in the longitudinal direction. The registration error for the CBCTimages from the single-rotation half-fan mode was in the longitudinal direction and negligible in the other directions.Conclusions:
The proposed method provides combined scatter correction and direct scatter reduction. Scatter correction may eliminate scatter artifacts, while direct scatter reduction may improve the CNR to compensate the CNR degradation due to scatter correction. Complete sets of CBCTimages are reconstructed in all three imaging modes. The single-rotation mode can be used for rigid-body patient alignment despite degradation in longitudinal resolution. The dual-rotation mode may be used to improve CBCTimage quality for soft tissue delineation in adaptive radiation therapy.