Index of content:
Volume 37, Issue 1, January 2010
Medical physicists should be allowed by States to image and treat, just like radiologic technologists37(2010); http://dx.doi.org/10.1118/1.3253993View Description Hide Description
- RADIATION THERAPY PHYSICS
Evaluation of the interplay effect when using RapidArc to treat targets moving in the craniocaudal or right-left direction37(2010); http://dx.doi.org/10.1118/1.3263614View Description Hide DescriptionPurpose:
We have investigated the dosimetric errors caused by the interplay between the motions of the LINAC and the tumor during the delivery of a volume modulated arc therapy treatment. This includes the development of an IMRT QA technique, applied here to evaluate RapidArc plans of varying complexity.Methods:
An IMRT QA technique was developed, which involves taking a movie of the delivered dose (0.2 s frames) using a 2D ion chamber array. Each frame of the movie is then moved according to a respiratory trace and the cumulative dose calculated. The advantage of this approach is that the impact of turning the beam on at different points in the respiratory trace, and of different types of motion, can be evaluated using data from a single irradiation. We evaluated this technique by comparing with the results when we actually moved the phantom during irradiation. RapidArc plans were created to treat a 62 cc spherical tumor in a lung phantom (16 plans) and a 454 cc irregular tumor in an actual patient (five plans). The complexity of each field was controlled by adjusting the MU (312–966 MU). Each plan was delivered to a phantom, and a movie of the delivered dose taken using a 2D ion chamber array. Patient motion was modeled by shifting each dose frame according to a respiratory trace, starting the motion at different phases. The expected dose distribution was calculated by blurring the static dose distribution with the target motion. The dose error due to the interplay effect was then calculated by comparing the delivered dose with the expected dose distribution. Peak-to-peak motion of 0.5, 1.0, and 2.0 cm in the craniocaudal and right-left directions, with target periods of 3 and 5 s, were evaluated for each plan (252 different target motion/plan combinations).Results:
The daily dose error due to the interplay effect was less than 10% for 98.4% of all pixels in the target for all plans investigated. The percentage of pixels for which the daily dose error could be larger than 5% increased with increasing plan complexity (field MU), but was less than 15% for all plans if the motion was 1 cm or less. For 2 cm motion, the dose error could be larger than 5% for 40% of pixels, but was less than 5% for more than 80% of pixels for, and was less than 10% for 99% of all pixels. The interplay effect was smaller for 3 s periods than for 5 s periods.Conclusions:
The interplay between the motions of the LINAC and the target can result in an error in the delivered dose. This effect increases with plan complexity, and with target magnitude and period. It may average out after many fractions.
37(2010); http://dx.doi.org/10.1118/1.3264617View Description Hide DescriptionPurpose:
Recently, there has been an increasing interest in operating conventional linear accelerators without a flattening filter. The aim of this study was to determine beam quality variations as a function of off-axis ray angle for unflattened beams. In addition, a comparison was made with the off-axis energy variation in flattened beams.Methods:
Two Elekta Precise linear accelerators were modified in order to enable radiation delivery with and without the flattening filter in the beam line. At the Medical University Vienna (Vienna, Austria), half value layer (HVL) measurements were performed for 6 and 10 MV with an in-house developed device that can be easily mounted on the gantry. At St. Luke’s Hospital (Dublin, Ireland), measurements were performed at 6 MV in narrow beam geometry with the gantry tilted around 270° with pinhole collimators, an attenuator, and the chamber positioned on the table. All attenuation measurements were performed with ionization chambers and a buildup cap (2 mm brass) or a PMMA mini phantom (diameter 3 cm, measurement depth 2.5 cm).Results:
For flattened 6 and 10 MV photon beams from the Elekta linac the relative varies by about 11% for an off-axis ray angle . These results agree within ±2% with a previously proposed generic off-axis energy correction. For unflattened beams, the variation was less than 5% in the whole range of off-axis ray angles up to 10°. The difference in relative HVL data was less than 1% for unflattened beams at 6 and 10 MV.Conclusions:
Off-axis energy variation is rather small in unflattened beams and less than half the one for flattened beams. Thus, ignoring the effect of off-axis energy variation for dose calculations in unflattened beams can be clinically justified.
37(2010); http://dx.doi.org/10.1118/1.3266769View Description Hide DescriptionPurpose:
The aim of this work is to develop effective calibration methods for a novel four-dimensional (4D) diode array for pretreatment verification of intensity-modulated radiation therapy(IMRT) and rotational therapy.Methods:
A novel 4D diode array (ArcCHECK, Sun Nuclear, Melbourne, FL) was developed to meet the needs of appropriate and efficient quality assurance for IMRT and especially rotational radiotherapy. The diode array presents a consistent detectorimage in beam’s eye view at arbitrary gantry angles due to isotropic arrangement of diodes in a three-dimensional (3D) cylindrical phantom. The simultaneous update of all diodes on the detector array (fourth dimension) makes it capable of time-resolved beam delivery analysis with any rotational delivery techniques. The calibration procedure consisted of delivering and measuring a series of calibration beams with 5.8° angular spacing surrounding the cylindrical diode array. Correction factors for diode intrinsic sensitivity and directional response dependence were derived from these measurements. A real-time algorithm to derive gantry angles based on the detector signal was developed to interpolate and apply the corresponding angular correction factors.Results:
The calibration was validated with ion chamber scanned beam profiles in a 3D water tank. Excellent agreement was observed between diode array measurement and treatment planning system calculation. The accuracy of the gantry angle derivation algorithm was within 1° which caused a less than 0.2% dosimetric uncertainty.Conclusions:
With the proposed calibration method and the automatic gantry angle derivation algorithm, the 4D diode array achieved isotropic detector response and is suitable for both IMRT and rotational therapy pretreatment verification.
37(2010); http://dx.doi.org/10.1118/1.3266821View Description Hide DescriptionPurpose:
A Monte Carlo(MC) based QA process to validate the dynamic beam delivery accuracy for Varian RapidArc™ (Varian Medical Systems, Palo Alto, CA) using Linac delivery log files (DynaLog) is presented. Using DynaLog file analysis and MC simulations, the goal of this article is to (a) confirm that adequate sampling is used in the RapidArc optimization algorithm (177 static gantry angles) and (b) to assess the physical machine performance [gantry angle and monitor unit (MU) delivery accuracy].Methods:
Ten clinically acceptable RapidArc treatment plans were generated for various tumor sites and delivered to a water-equivalent cylindrical phantom on the treatment unit. Three Monte Carlo simulations were performed to calculate dose to the CT phantom image set: (a) One using a series of static gantry angles defined by 177 control points with treatment planning system (TPS) MLC control files (planning files), (b) one using continuous gantry rotation with TPS generated MLC control files, and (c) one using continuous gantry rotation with actual Linac delivery log files. Monte Carlo simulated dose distributions are compared to both ionization chamber point measurements and with RapidArc TPS calculated doses. The 3D dose distributions were compared using a 3D gamma-factor analysis, employing a distance-to-agreement criterion.Results:
The dose difference between MC simulations, TPS, and ionization chamber point measurements was less than 2.1%. For all plans, the MC calculated 3D dose distributions agreed well with the TPS calculated doses (gamma-factor values were less than 1 for more than 95% of the points considered). Machine performance QA was supplemented with an extensive DynaLog file analysis. A DynaLog file analysis showed that leaf position errors were less than for 94% of the time and there were no leaf errors greater than . The mean standard deviation in MU and gantry angle were and 0.355°, respectively, for the ten cases analyzed.Conclusions:
The accuracy and flexibility of the Monte Carlo based RapidArc QA system were demonstrated. Good machine performance and accurate dose distribution delivery of RapidArc plans were observed. The sampling used in the TPS optimization algorithm was found to be adequate.
37(2010); http://dx.doi.org/10.1118/1.3267040View Description Hide DescriptionPurpose:
Recently, several authors have shown contrast improvements in megavoltage portal imaging and cone-beam computed tomography using low atomic number targets. This work compliments previous studies by investigating the effects of varying different beam production parameters including target atomic number, target thickness, and incident electron energy on spatial resolution.Methods:
Target materials of beryllium, aluminum, and tungsten were investigated over a range of thicknesses between 10% and 100% of the continuous slowing down approximation range of electrons. Incident electron kinetic energies of 4.5 and were used, in conjunction with custom targets installed above the carousel of a modern radiotherapy linear accelerator.Monte Carlo simulations of the accelerator were constructed and compared to the experimental results.Results:
The results showed that thinner targets, as well higher incident electron energies, generally produce more favorable modulation transfer function(MTF) curves. Due to an MTF dependence of the detector system on the photon energy, the experimental results showed that low- targets produced superior MTF curves. Simulations showed 14.5% and 21.5% increases in for the 7.0 and targets (Al; 60% ), respectively, when moved from the carousel to the location of the clinical target. values for the custom targets were compared to the clinical beam and were found to be between 10.4% lower and 15.5% higher .Conclusions:
Integration of low- external targets into the treatment head of a medical linear was achieved with only minor modifications. It was shown that reasonably high resolution images on par or better than those acquired with the clinical beam can be achieved using external low- targets.
37(2010); http://dx.doi.org/10.1118/1.3271338View Description Hide DescriptionPurpose:
The low effective atomic number, reusability, and other computed radiography-related advantages make europium doped potassium chloride a promising dosimetry material. The purpose of this study is to model point dosimeters with a Monte Carlo(MC) method and, using this model, to investigate the dose responses of two-dimensional (2D) storage phosphor films (SPFs).Methods:
point dosimeters were irradiated using a beam at four depths for each of five square field sizes. The dose measured by was compared to that measured by an ionization chamber to obtain the magnitude of energy dependent dose measurement artifact. The measurements were simulated using DOSXYZnrc with phase space files generated by BEAMnrcMP. Simulations were also performed for films with thicknesses ranging from . The work function of the prototype material was determined by comparing the sensitivity of a thick film to a commercial -based SPF with a known work function. The work function was then used to estimate the sensitivity of a thick film.Results:
The simulated dose responses of prototype point dosimeters agree well with measurement data acquired by irradiating the dosimeters in the beam with varying field size and depth. Furthermore, simulations with films demonstrate that an ultrathin film with thickness of the order of would have nearly water-equivalent dose response. The simulation results can be understood using classic cavity theories. Finally, preliminary experiments and theoretical calculations show that ultrathin film could provide excellent signal in a dose-to-water irradiation.Conclusions:
In conclusion, the authors demonstrate that-based dosimeters can be accurately modeled by a MC method and that 2D films of the order of thick would have minimal energy dependence. The data support the future research and development of a storage phosphor-based system for quantitative, high-resolution multidimensional radiation therapydosimetry.
Commissioning of the discrete spot scanning proton beam delivery system at the University of Texas M.D. Anderson Cancer Center, Proton Therapy Center, Houston37(2010); http://dx.doi.org/10.1118/1.3259742View Description Hide DescriptionPurpose:
To describe a summary of the clinical commissioning of the discrete spot scanning proton beam at the Proton Therapy Center, Houston (PTC-H).Methods:
Discrete spot scanning system is composed of a delivery system (Hitachi ProBeat), an electronic medical record (Mosaiq V 1.5), and a treatment planning system (TPS) (Eclipse V 8.1). Discrete proton pencil beams (spots) are used to deposit dose spot by spot and layer by layer for the proton distal ranges spanning from and over a maximum scan area at the isocenter of . An arbitrarily chosen reference calibration condition has been selected to define the monitor units (MUs). Using radiochromic film and ion chambers, the authors have measured spot positions, the spot sizes in air, depth dose curves, and profiles for proton beams with various energies in water, and studied the linearity of the dose monitors. In addition to dosimetric measurements and TPS modeling, significant efforts were spent in testing information flow and recovery of the delivery system from treatment interruptions.Results:
The main dose monitors have been adjusted such that a specific amount of charge is collected in the monitor chamber corresponding to a single MU, following the IAEA TRS 398 protocol under a specific reference condition. The dose monitor calibration method is based on the absolute dose per MU, which is equivalent to the absolute dose per particle, the approach used by other scanning beam institutions. The full width at half maximum for the spot size in air varies from approximately for for . The measured versus requested 90% depth dose in water agrees to within over ranges of . The beam delivery interlocks perform as expected, guarantying the safe and accurate delivery of the planned dose.Conclusions:
The dosimetric parameters of the discrete spot scanning proton beam have been measured as part of the clinical commissioning program, and the machine is found to function in a safe manner, making it suitable for patient treatment.
Object-constrained meshless deformable algorithm for high speed 3D nonrigid registration between CT and CBCT37(2010); http://dx.doi.org/10.1118/1.3271389View Description Hide DescriptionPurpose:
High-speed nonrigid registration between the planning CT and the treatmentCBCT data is critical for real time image guided radiotherapy (IGRT) to improve the dose distribution and to reduce the toxicity to adjacent organs. The authors propose a new fully automatic 3D registration framework that integrates object-based global and seed constraints with the grayscale-based “demons” algorithm.Methods:
Clinical objects were segmented on the planning CTimages and were utilized as meshless deformable models during the nonrigid registration process. The meshless models reinforced a global constraint in addition to the grayscale difference between CT and CBCT in order to maintain the shape and the volume of geometrically complex 3D objects during the registration. To expedite the registration process, the framework was stratified into hierarchies, and the authors used a frequency domain formulation to diffuse the displacement between the reference and the target in each hierarchy. Also during the registration of pelvis images, they replaced the air region inside the rectum with estimated pixel values from the surrounding rectal wall and introduced an additional seed constraint to robustly track and match the seeds implanted into the prostate. The proposed registration framework and algorithm were evaluated on 15 real prostate cancer patients. For each patient, prostate gland, seminal vesicle, bladder, and rectum were first segmented by a radiationoncologist on planning CTimages for radiotherapy planning purpose. The same radiationoncologist also manually delineated the tumor volumes and critical anatomical structures in the corresponding CBCTimages acquired at treatment. These delineated structures on the CBCT were only used as the ground truth for the quantitative validation, while structures on the planning CT were used both as the input to the registration method and the ground truth in validation. By registering the planning CT to the CBCT, a displacement map was generated. Segmented volumes in the CTimages deformed using the displacement field were compared against the manual segmentations in the CBCTimages to quantitatively measure the convergence of the shape and the volume. Other image features were also used to evaluate the overall performance of the registration.Results:
The algorithm was able to complete the segmentation and registration process within 1 min, and the superimposed clinical objects achieved a volumetric similarity measure of over 90% between the reference and the registered data. Validation results also showed that the proposed registration could accurately trace the deformation inside the target volume with average errors of less than 1 mm. The method had a solid performance in registering the simulated images with up to 20 Hounsfield unit white noise added. Also, the side by side comparison with the original demons algorithm demonstrated its improved registration performance over the local pixel-based registration approaches.Conclusions:
Given the strength and efficiency of the algorithm, the proposed method has significant clinical potential to accelerate and to improve the CBCT delineation and targets tracking in online IGRT applications.
37(2010); http://dx.doi.org/10.1118/1.3271582View Description Hide DescriptionPurpose:
Cone beam CT is increasingly being used for daily patient positioning verification during radiation therapy treatments. The daily use of CBCT could lead to accumulated patient doses higher than the older technique of weekly portal imaging. There have been several studies focusing on measurement or calculation of the patient dose from CBCT recently.Methods:
This study investigates the feasibility of configuring a kV x-ray source in a commercial treatment planning system to calculate the dose to patient resulting from an IGRT procedure. The method proposed in this article can be used to calculate dose from CBCTimaging procedure and include that in the patient treatment plans.Results:
The kilovoltage beam generated by the CBCTimager has been modeled using the planning system. The modeled profiles agree with the measured ones to within 5%. The modeled beam was used to calculate dose to phantom in the pelvic region and the calculations were compared to TLD measurements. The agreement between calculated and measured doses ranges from 0% to 19% in soft tissue with larger variations observed near and within the bone.Conclusions:
The modeling of the beam produces reasonable results and the dose calculation comparisons indicate the potential for computing kilovoltage CBCTdoses using a treatment planning system. Further improvements in the dose calculation algorithm are necessary, especially for dose calculations in and near the bone.
37(2010); http://dx.doi.org/10.1118/1.3271390View Description Hide DescriptionPurpose:
In proton therapy, as in other forms of radiation therapy, scattered and secondary particles produce undesired dose outside the target volume that may increase the risk of radiation-induced secondary cancer and interact with electronic devices in the treatment room. The authors implement a Monte Carlo model of this dose deposited outside passively scattered fields and compare it to measurements, determine the out-of-field equivalent dose, and estimate the change in the dose if the same target volumes were treated with an active beam scanning technique.Methods:
Measurements are done with a thimble ionization chamber and the Wellhofer MatriXX detector inside a Lucite phantom with field configurations based on the treatment of prostate cancer and medulloblastoma. The authors use aGEANT4Monte Carlo simulation, demonstrated to agree well with measurements inside the primary field, to simulate fields delivered in the measurements. The partial contributions to the dose are separated in the simulation by particle type and origin.Results:
The agreement between experiment and simulation in the out-of-field absorbed dose is within 30% at 10–20 cm from the field edge and 90% of the data agrees within 2 standard deviations. In passive scattering, the neutron contribution to the total dose dominates in the region downstream of the Bragg peak (65%–80% due to internally produced neutrons) and inside the phantom at distances more than 10–15 cm from the field edge. The equivalent doses using 10 for the neutron weighting factor at the entrance to the phantom and at 20 cm from the field edge are 2.2 and 2.6 mSv/Gy for the prostate cancer and cranial medulloblastoma fields, respectively. The equivalent dose at 15–20 cm from the field edge decreases with depth in passive scattering and increases with depth in active scanning. Therefore, active scanning has smaller out-of-field equivalent dose by factors of 30–45 in the entrance region and this factor decreases with depth.Conclusions:
The dose deposited immediately downstream of the primary field, in these cases, is dominated by internally produced neutrons; therefore, scattered and scanned fields may have similar risk of second cancer in this region. The authors confirm that there is a reduction in the out-of-field dose in active scanning but the effect decreases with depth.GEANT4 is suitable for simulating the dose deposited outside the primary field. The agreement with measurements is comparable to or better than the agreement reported for other implementations of Monte Carlo models. Depending on the position, the absorbed dose outside the primary field is dominated by contributions from primary protons that may or may not have scattered in the brass collimating devices. This is noteworthy as the quality factor of the low LET protons is well known and the relative dose risk in this region can thus be assessed accurately.
37(2010); http://dx.doi.org/10.1118/1.3264177View Description Hide DescriptionPurpose:
First, to show that accurate formulas for scattering power must take into account the competition between the Gaussian core and the single scattering tail of the angular distribution, which affects the rate of change in the Gaussian width and leads to the single scattering correction (SSC). Second, to show that the SSC requires that be nonlocal: Besides material properties and energy at the point of interest, it must depend in some fashion on how much multiple scattering has already taken place. Third, after reviewing five previous formulas (three local and two nonlocal), to derive an improved “differential Molière” formula . Last, to investigate, by studying some practical cases, when an accurate formula for is actually needed.Methods:
We first take the numerical derivative of the Molière/Fano/Hanson in order to find the true SSC. We simplify the formula for (ICRU Report 35) for protons, introducing a new material dependent property, the “scattering length” , analogous to radiation length . We then use as a basis for by including a nonlocal correction factor which, by virtue of the Øverås approximation, parametrizes the single scattering correction.Results:
The improved scattering power is where , (MeV) is the initial product of proton momentum and speed, is the same at the point of interest, and . is easily computed and generalizes readily to mixed slabs because is not material dependent.Conclusions:
Whether an accurate formula for is required depends very much on the problem at hand. For beam spreading in water, five of the six formulas for give almost identical results, suggesting that patient dose calculations are insensitive to . That is not true, however, of beam spreading in Pb. At the opposite extreme, the projected rms beam width at the end of a Pb/Lexan/air stack, analogous to the upstream modulator in a passive beam spreading system, is sensitive to . In this case a simple experiment would discriminate between all but two of the six formulas discussed. Scattering power applies as much to Monte Carlo as to deterministic transport calculations. Using in any of its forms will avoid step size dependence. Using the best available could be important in general purpose Monte Carlo codes, which are expected to give the correct answer to many different problems.
37(2010); http://dx.doi.org/10.1118/1.3271104View Description Hide DescriptionPurpose:
Dose distributions are calculated for various models of and seeds in the standardized plaques of the Collaborative Ocular Melanoma Study (COMS). The sensitivity to seed model of dose distributions and dose distributions relative to TG-43 are investigated.Methods:
Monte Carlo simulations are carried out with the EGSnrc user-code BrachyDose. Brachytherapy seeds and eye plaques are fully modeled. Simulations of one seed in the central slot of a 20 mm Modulay (gold alloy) plaque backing with and without the Silastic (silicone polymer) insert and of a 16 mm fully loaded Modulay/Silastic plaque are performed. Dose distributions are compared to those calculated under TG-43 assumptions, i.e., ignoring the effects of the plaque backing and insert and interseed attenuation. Three-dimensional dose distributions for different and seed models are compared via depth-dose curves, isodose contours, and tabulation of doses at points of interest in the eye. Results are compared to those of our recent BrachyDose study for COMS plaques containing model 6711 or 200 seeds [R. M. Thomson et al., Med. Phys.35, 5530–5543 (2008)].Results:
Along the central axis of a plaque containing one seed, variations of less than 1% are seen in the effect of the Modulay backing alone for different seed models; for the Modulay/Silastic combination, variations are 2%. For a 16 mm plaque fully loaded with seeds, dose decreases relative to TG-43 doses are 11%–12% (19%–20%) and 14%–15% (20%) at distances of 0.5 and 1 cm from the inner sclera along the plaque’s central axis, respectively. For the same prescription dose,doses at points of interest vary by up to 8% with seed model. Doses to critical normal structures are lower for all seed models than for with the possible exception of the sclera adjacent to the plaque; scleral doses vary with seed model and are not always higher for than for .Conclusions:
Dose decreases relative to doses calculated under TG-43 assumptions vary slightly with seed model (for each radionuclide). Dose distributions are sensitive to seed model; however, variations are generally no larger than the magnitudes of other systematic uncertainties in eye plaque therapy.
Dosimetric verification of IMAT delivery with a conventional EPID system and a commercial portal dose image prediction tool37(2010); http://dx.doi.org/10.1118/1.3271107View Description Hide DescriptionPurpose:
The electronic portal imaging device(EPID) is a system for checking the patient setup; as a result of its integration with the linear accelerator and software customized for dosimetry, it is increasingly used for verification of the delivery of fixed-field intensity-modulated radiation therapy(IMRT). In order to extend such an approach to intensity-modulated arc therapy (IMAT), the combined use of an EPID system and a portal doseimage prediction (PDIP) tool has been investigated.Methods:
The dosimetric behavior of an EPID system, mechanically reinforced to maintain its positional stability during the accelerator gantry rotation, has been studied to assess its ability to measure portal dose distributions for IMAT treatment beams. In addition, the PDIP tool of a commercial treatment planning system, commonly used for static IMRTdosimetry, has been validated for simulating the PDIs of IMAT treatment fields. The method has been applied to the delivery verification of 23 treatment fields that were measured in their dual mode of IMRT and IMAT modalities.Results:
The EPID system has proved to be appropriate for measuring the PDIs of IMAT fields; additionally the PDIP tool was able to simulate these accurately. The results are quite similar to those obtained for static IMRT treatment verification, although it was necessary to investigate the dependence of the EPID signal and of the accelerator monitor chamber response on variable dose rate.Conclusions:
Our initial tests indicate that the EPID system, together with the PDIP tool, is a suitable device for the verification of IMAT plan delivery; however, additional tests are necessary to confirm these results.
- RADIATION IMAGING PHYSICS
CT colonography: Advanced computer-aided detection scheme utilizing MTANNs for detection of “missed” polyps in a multicenter clinical trial37(2010); http://dx.doi.org/10.1118/1.3263615View Description Hide DescriptionPurpose:
The purpose of this study was to develop an advanced computer-aided detection(CAD) scheme utilizing massive-training artificial neural networks (MTANNs) to allow detection of “difficult” polyps in CT colonography (CTC) and to evaluate its performance on false-negative (FN) CTC cases that radiologists “missed” in a multicenter clinical trial.Methods:
The authors developed an advanced CAD scheme consisting of an initial polyp-detection scheme for identification of polyp candidates and a mixture of expert MTANNs for substantial reduction in false positives (FPs) while maintaining sensitivity. The initial polyp-detection scheme consisted of (1) colon segmentation based on anatomy-based extraction and colon-based analysis and (2) detection of polyp candidates based on a morphologic analysis on the segmented colon. The mixture of expert MTANNs consisted of (1) supervised enhancement of polyps and suppression of various types of nonpolyps, (2) a scoring scheme for converting output voxels into a score for each polyp candidate, and (3) combining scores from multiple MTANNs by the use of a mixing artificial neural network. For testing the advanced CAD scheme, they created a database containing 24 FN cases with 23 polyps (range of; average of ) and a mass , which were “missed” by radiologists in CTC in the original trial in which 15 institutions participated.Results:
The initial polyp-detection scheme detected 63% of the missed polyps with 21.0 FPs per patient. The MTANNs removed 76% of the FPs with loss of one true positive; thus, the performance of the advanced CAD scheme was improved to a sensitivity of 58% with 8.6 FPs per patient, whereas a conventional CAD scheme yielded a sensitivity of 25% at the same FP rate (the difference was statistically significant).Conclusions:
With the advanced MTANN CAD scheme, 58% of the polyps missed by radiologists in the original trial were detected and with a reasonable number of FPs. The results suggest that the use of an advanced MTANN CAD scheme may potentially enhance the detection of “difficult” polyps.
37(2010); http://dx.doi.org/10.1118/1.3263618View Description Hide DescriptionPurpose:
Reverse helical cone-beam computed tomography(CBCT) is a scanning configuration for potential applications in image-guided radiation therapy in which an accurate anatomic image of the patient is needed for image-guidance procedures. The authors previously developed an algorithm for image reconstruction from nontruncated data of an object that is completely within the reverse helix. The purpose of this work is to develop an image reconstruction approach for reverse helical CBCT of a long object that extends out of the reverse helix and therefore constitutes data truncation.Methods:
The proposed approach comprises of two reconstruction steps. In the first step, a chord-based backprojection-filtration (BPF) algorithm reconstructs a volumetric image of an object from the original cone-beam data. Because there exists a chordless region in the middle of the reverse helix, the image obtained in the first step contains an unreconstructed central-gap region. In the second step, the gap region is reconstructed by use of a Pack–Noo-formula-based filteredbackprojection (FBP) algorithm from the modified cone-beam data obtained by subtracting from the original cone-beam data the reprojection of the imagereconstructed in the first step.Results:
The authors have performed numerical studies to validate the proposed approach in image reconstruction from reverse helical cone-beam data. The results confirm that the proposed approach can reconstruct accurate images of a long object without suffering from data-truncation artifacts or cone-angle artifacts.Conclusions:
They developed and validated a BPF-FBP tandem algorithm to reconstructimages of a long object from reverse helical cone-beam data. The chord-based BPF algorithm was utilized for converting the long-object problem into a short-object problem. The proposed approach is applicable to other scanning configurations such as reduced circular sinusoidal trajectories.
Method for evaluating bow tie filter angle-dependent attenuation in CT: Theory and simulation results37(2010); http://dx.doi.org/10.1118/1.3264616View Description Hide DescriptionPurpose:
Dosimetry in computed tomography(CT) is increasingly based on Monte Carlo studies that define the dose in the patient (in mGy) as a function of air kerma (free in air) at isocenter (mGy). The accuracy of Monte Carlo studies depends in part on the accuracy of the characterization of the bow tie filter for a given CTscanner model. A simple method for characterizing the bow tie filter attenuation profile in CTscanners would therefore be very useful. The theory behind such a method is proposed.Methods:
A measurement protocol is discussed mathematically and demonstrated using computer simulation. The proposed method requires the placement of a radiation monitor at the periphery of the CT field, and the time domain signal (kerma rate versus time) is measured with good temporal resolution ( or better) and with all other objects (e.g., patient couch) retracted from the field of view. Knowledge of the source to isocenter distance (or alternately, the isocenter to probe distance) is required. The stationary detector records the kerma rate versus time signal as the gantry rotates through several revolutions. From this temporal data, signal processing techniques are used to extract in-phase peaks, as well as out-of-phase kerma rate levels. From these data, the distance from isocenter to the probe can be determined (or, alternatively, the source to isocenter distance), and the angle-dependent bow tie filter attenuation can be computed. By measuring the angle-dependent bow tie filter attenuation at several kVp settings, the bow tie composition versus fan angle can be computed using basis decomposition techniques.Results:
The simulations illustrated that with 2% added noise in the kerma rate versus time signal, the attenuation properties of a hypothetical two component (aluminum and polymethyl methacrylate) bow tie filter could be determined. Although the computed basis material thicknesses were not exactly equal to the actual thicknesses, their combined attenuation factors matched that of the actual filter across kVp’s to within an average of 0.057%.Conclusions:
It is concluded that the proposed method may provide a simple noninvasive approach to characterizing the performance of bow tie filters in CT systems; however, experimental validation is necessary.
37(2010); http://dx.doi.org/10.1118/1.3264619View Description Hide DescriptionPurpose:
Quantitativein vivoimaging of lung perfusion in rodents can provide critical information for preclinical studies. However, the combined challenges of high temporal and spatial resolution have made routine quantitative perfusion imaging difficult in small animals. The purpose of this work is to demonstrate 4D micro-CT for perfusion imaging in rodents at heartbeat temporal resolution and isotropic spatial resolution.Methods:
We have recently developed a dual tube/detector micro-CT scanner that is well suited to capture first pass kinetics of a bolus of contrast agent used to compute perfusion information. Our approach is based on the paradigm that similar time density curves can be reproduced in a number of consecutive, small volume injections of iodinated contrast agent at a series of different angles. This reproducibility is ensured by the high-level integration of the imaging components of our system with a microinjector, a mechanical ventilator, and monitoring applications. Sampling is controlled through a biological pulse sequence implemented inLABVIEW. Image reconstruction is based on a simultaneous algebraic reconstruction technique implemented on a graphic processor unit. The capabilities of 4D micro-CT imaging are demonstrated in studies on lung perfusion in rats.Results:
We report 4D micro-CT imaging in the rat lung with a heartbeat temporal resolution (approximately 150 ms) and isotropic 3D reconstruction with a voxel size of based on sampling using 16 injections of each. The total volume of contrast agent injected during the experiments (0.8 mL) was less than 10% of the total blood volume in a rat. This volume was not injected in a single bolus, but in multiple injections separated by at least 2 min interval to allow for clearance and adaptation. We assessed the reproducibility of the time density curves with multiple injections and found that these are very similar. The average time density curves for the first eight and last eight injections are slightly different, i.e., for the last eight injections, both the maximum of the average time density curves and its area under the curve are decreased by 3.8% and 7.2%, respectively, relative to the average time density curves based on the first eight injections. The radiation dose associated with our 4D micro-CT imaging is 0.16 Gy and is therefore in the range of a typical micro-CT dose.Conclusions:
4D micro-CT-based perfusion imaging demonstrated here has immediate application in a wide range of preclinical studies such as tumor perfusion, angiogenesis, and renal function. Although our imagingsystem is in many ways unique, we believe that our approach based on the multiple injection paradigm can be used with the newly developed flat-panel slip-ring-based micro-CT to increase their temporal resolution in dynamic perfusion studies.
37(2010); http://dx.doi.org/10.1118/1.3264662View Description Hide DescriptionPurpose:
A system is presented for automated estimation of progression of interstitial lung disease in serial thoracic CT scans.Methods:
The system compares corresponding 2D axial sections from baseline and follow-up scans and concludes whether this pair of sections represents regression, progression, or unchanged disease status. The correspondence between serial CT scans is achieved by intrapatient volumetric image registration. The system classification function is trained with two different feature sets. Features in the first set represent the intensity distribution of a difference image between the baseline and follow-up CT sections. Features in the second set represent dissimilarities computed between the baseline and follow-up images filtered with a bank of general purpose texture filters.Results:
In an experiment on 74 scan pairs, the system classification accuracies were 76.1% and 79.5% for the two feature sets, respectively, while the accuracies of two observer radiologist were 78.5% and 82%, respectively. The agreements of the system with the reference standard, measured by weighted kappa statistics, were 0.611 and 0.683 for the two feature sets, respectively.Conclusions:
The system employing the second feature set showed good agreement with the reference standard, and its accuracy approached that of two radiologists.