ZHW: A Numerical CODEC for Big Data Scientific Computation
Explore the source record for details and available documents.
Engineering topics
Publications and source records attributed to Wu, Z.
Explore the source record for details and available documents.
Purpose: To quantitatively evaluate the strategic innovation and accuracy variation of deformation registration algorithm for parotid glands using the similarity Dice coefficient during the course of radiation therapy (RT) for nasopharyngeal cancer (NPC). Methods: Daily MVCT data for 10 patients with pathologically proven nasopharyngeal cancers were analyzed. The data were acquired using tomotherapy (TomoTherapy, Accuray) at the PLA General Hospital. The prescription dose to the primary target was 70Gy in 33 fractions. Two kinds of contours for parotid glands on daily MVCTs were obtained by populating these contours from planning CTs to the daily CTs via rigid-body registration with or without the rotation shifts using the in-house tools and the Adaptive plan software (Adaptive Plan, TomoTherapy), and were edited manually if necessary. The diffeomorphic Demons algorithm developed in the in-house tool was used to propagate the parotid structures from the daily CTs to planning CTs. The differences of the mapped parotid contours in two methods were evaluated using Dice similarity index (DSI). Two-tailed t-test analysis was carried out to compare the DSI changes during the course of RT. Results: For 10 patient plans, the accuracy of deformation image registration (DIR) with the rotation shift was obviously better than those without the rotation shift. The Dice scores of the ipsi- and contra-lateral parotids for with and without the rotation shifts were found to be correlated with each other [0.904±0.031 vs 0.919±0.030 (p<0.001); 0.900±0.031 vs 0.910±0.032 (p<0.001)]. The Dice scores for the parotids have shown the reduction with the changes of parotid volumes during RT. The DSI values between the first and last fraction were 0.932±0.020 vs 0.899±0.030 in 10 patient plans. Conclusion: DIR was successfully improved using the strategic innovation for ART. And the decrease of DIR accuracy has also been found during the delivery of fractionated radiotherapy. This work was supported in part by the grant from Chinese Natural Science Foundation (Grant No. 11105225).
Purpose: Electronic portal imaging device (EPID) can be used to acquire a two-dimensional exit dose distribution during treatment delivery, thus allowing the in-vivo verification of the dose delivery through a comparison of measured portal images to predicted portal dose images (PDI). The aim of this study was to present a novel method to easily and accurately predict PDI, and to establish an EPID-based in-vivo dose verification method during IMRT treatments. Methods: We developed a model to determine the predicted portal dose at the same plane of the EPID detector location. The Varian EPID (aS1000) positions at 150cm source-to-detector-distance (SDD), and can be used to acquire in-vivo exit dose using Portal Dosimetry (PD) function. Our model was generated to make an equivalent water thickness represent the buildup plate of EPID. The exit dose at extend SDD plane with patient CT data in the beam can be calculated as the predicted PDI in the treatment planning system (TPS). After that, the PDI was converted to the fluence at SDD of 150cm using the inverse square law coded in MATLAB. Five head-and-neck and prostate IMRT patient plans contain 32 fields were investigated to evaluate the feasibility of this new method. The measured EPID image was compared with PDI using the gamma analysis. Results: The average results for cumulative dose comparison were 81.9% and 91.6% for 3%, 3mm and 4%, 4mm gamma criteria, respectively. Results indicate that the patient transit dosimetry predicted algorithm compares well with EPID measured PD doses for test situations. Conclusion: Our new method can be used as an easy and feasible tool for online EPID-based in-vivo dose delivery verification for IMRT treatments. It can be implemented for fast detecting those obvious treatment delivery errors for individual field and patient quality assurance.
Purpose: This study will use Monte Carlo to simulate the Cyberknife system, and intend to develop the third-party tool to evaluate the dose verification of specific patient plans in TPS. Methods: By simulating the treatment head using the BEAMnrc and DOSXYZnrc software, the comparison between the calculated and measured data will be done to determine the beam parameters. The dose distribution calculated in the Raytracing, Monte Carlo algorithms of TPS (Multiplan Ver4.0.2) and in-house Monte Carlo simulation method for 30 patient plans, which included 10 head, lung and liver cases in each, were analyzed. The γ analysis with the combined 3mm/3% criteria would be introduced to quantitatively evaluate the difference of the accuracy between three algorithms. Results: More than 90% of the global error points were less than 2% for the comparison of the PDD and OAR curves after determining the mean energy and FWHM.The relative ideal Monte Carlo beam model had been established. Based on the quantitative evaluation of dose accuracy for three algorithms, the results of γ analysis shows that the passing rates (84.88±9.67% for head,98.83±1.05% for liver,98.26±1.87% for lung) of PTV in 30 plans between Monte Carlo simulation and TPS Monte Carlo algorithms were good. And the passing rates (95.93±3.12%,99.84±0.33% in each) of PTV in head and liver plans between Monte Carlo simulation and TPS Ray-tracing algorithms were also good. But the difference of DVHs in lung plans between Monte Carlo simulation and Ray-tracing algorithms was obvious, and the passing rate (51.263±38.964%) of γ criteria was not good. It is feasible that Monte Carlo simulation was used for verifying the dose distribution of patient plans. Conclusion: Monte Carlo simulation algorithm developed in the CyberKnife system of this study can be used as a reference tool for the third-party tool, which plays an important role in dose verification of patient plans. This work was supported in part by the grant from Chinese Natural Science Foundation (Grant No. 11275105). Thanks for the support from Accuray Corp.
Purpose: The aim of this study is to investigate the feasibility of using RapidPlan for breast cancer radiotherapy and to evaluate its performance for planners with different planning experiences. Methods: A training database was collected with 80 expert plan datasets from patients previously received left breast conserving surgery and IMRT-simultaneously integrated boost radiotherapy. The models were created on the RapidPlan. Five patients from the training database and 5 external patients were used for internal and external validation, respectively. Three planners with different planning experiences (beginner, junior, senior) designed manual and RapidPlan based plans for additional ten patients. The plan qualities were compared with manual and RapidPlan based ones. Results: For the internal and external validations, there were no significant dose differences on target coverage for plans from RapidPlan and manual. Also, no difference was found in the mean doses to contralateral breast and lung. The RapidPlan improved the heart (V5, V10, V20, V30, and mead dose) and ipsilateral lung (V5, V10, V20, V30, and mean dose) sparing for the beginner and junior planners. Compare to the plans from senior planner, 6 out of 16 clinically checked parameters were improved in RapidPlan, and the left parameters were similar. Conclusion: It is feasible to generate clinical acceptable plans using RapidPlan for breast cancer radiotherapy. The RapidPlan helps to systematically improve the quality of IMRT plans against the benchmark of clinically accepted plans. The RapidPlan shows promise for homogenizing plan quality by transferring planning expertise from more experienced to less experienced planners.
Purpose Daily setup for definitive prostatic radiotherapy is challenged by suboptimal visibility of the prostate boundary and daily variation of rectum shape and position. For patients with improved bowel preparation, we conducted a dosimetric comparison between prostate implanted marker (IM)-based daily setup and anterior rectal wall (ARW)-based setup, with the hypothesis that the former leads to adequate target coverage with better rectal sparing. Methods Five IMRT/VMAT prostate cases with implanted markers were selected for analysis. Daily CBCT showed improvement of the rectal volume compared to planning CT. For each patient, the prostate and rectum were contoured on three CBCT images (fraction 5/15/25) with subsequent physician review. The CBCTs were then registered to a planning CT using IM-based registration. The deviation of ARW positions from planning CT to CBCT were analyzed at various sup-inf levels (−1.8 cm to 1.8 cm from level of prostate center). To estimate the potential dosimetric impact from ARW-based setup, the treatment plans were recalculated using A-P shifts ranging from −1mm to +6mm. Clinically important rectum DVH values including Dmax, D3cc and Dmean were computed. Results For the studied patients, we observed on average 32% rectum volume reduction from planning CT to CBCT. As a Results, the ARW on average shifts posteriorly by −1mm to +5mm, depending on the sup-inf level of observation, with larger shifts observed at more superior levels. Recalculation shows that when ARW shifts 1mm posteriorly, ARW-based CBCT setup leads to a 1.0%, 4.2%, and 3.2% increase in rectum Dmax, D3cc, and Dmean, respectively, compared to IM-based setup. The dosimetric deviations increase to 4.7%, 25.8% and 24.7% when ARW shifts 6mm posteriorly. No significant prostate-only dose difference was observed. Conclusion For patients with improved bowel preparation, IM-based CBCT setup leads to accurate prostate coverage along with significantly lower rectal dose, compared to ARW-based setup.
Purpose: To assess the potential of dynamic infrared imaging to evaluate early skin reactions during radiation therapy in cancer patients. Methods: Thermal images were captured by our home-built system consisting of two flash lamps and an infrared (IR) camera. The surface temperature of the skin was first raised by ∼ 6 °C from ∼1 ms short flashes; the camera then captured a series of IR images for 10 seconds. For each image series, a basal temperature was recorded for 0.5 seconds before flash was triggered. The temperature gradients (ε) were calculated between a reference point (immediately after the flash) and at a time point of 2sec, 4sec and 9sec after that. A 1.0 cm region of interest (ROI) on the skin was drawn; the mean and standard deviations of the ROIs were calculated. The standard ε values for normal human skins were evaluated by imaging 3 healthy subjects with different skin colors. All of them were imaged on 3 separate days for consistency checks. Results: The temperature gradient, which is the temperature recovery rate, depends on the thermal properties of underlying tissue, i.e. thermal conductivity. The average ε for three volunteers averaged over 3 measurements were 0.64±0.1, 0.72±0.2 and 0.80±0.3 at 2sec, 4sec and 9sec respectively. The standard deviations were within 1.5%–3.2%. One of the volunteers had a prior small skin burn on the left wrist and the ε values for the burned site were around 9% (at 4sec) and 13% (at 9sec) lower than that from the nearby normal skin. Conclusion: The temperature gradients from the healthy subjects were reproducible within 1.5%–3.2 % and that from a burned skin showed a significant difference (9%–13%) from the normal skin. We have an IRB approved protocol to image head and neck patients scheduled for radiation therapy.
Purpose: Intensity-modulated proton therapy (IMPT) is increasingly used in proton therapy. For IMPT optimization, Monte Carlo (MC) is desired for spots dose calculations because of its high accuracy, especially in cases with a high level of heterogeneity. It is also preferred in biological optimization problems due to the capability of computing quantities related to biological effects. However, MC simulation is typically too slow to be used for this purpose. Although GPU-based MC engines have become available, the achieved efficiency is still not ideal. The purpose of this work is to develop a new optimization scheme to include GPU-based MC into IMPT. Methods: A conventional approach using MC in IMPT simply calls the MC dose engine repeatedly for each spot dose calculations. However, this is not the optimal approach, because of the unnecessary computations on some spots that turned out to have very small weights after solving the optimization problem. GPU-memory writing conflict occurring at a small beam size also reduces computational efficiency. To solve these problems, we developed a new framework that iteratively performs MC dose calculations and plan optimizations. At each dose calculation step, the particles were sampled from different spots altogether with Metropolis algorithm, such that the particle number is proportional to the latest optimized spot intensity. Simultaneously transporting particles from multiple spots also mitigated the memory writing conflict problem. Results: We have validated the proposed MC-based optimization schemes in one prostate case. The total computation time of our method was ∼5–6 min on one NVIDIA GPU card, including both spot dose calculation and plan optimization, whereas a conventional method naively using the same GPU-based MC engine were ∼3 times slower. Conclusion: A fast GPU-based MC dose calculation method along with a novel optimization workflow is developed. The high efficiency makes it attractive for clinical usages.
Purpose: To investigate the performance of a new penalized-likelihood PET image reconstruction algorithm using the 1{sub 1}-norm total-variation (TV) sum of the 1st through 4th-order gradients as the penalty. Simulated and brain patient data sets were analyzed. Methods: This work represents an extension of the preconditioned alternating projection algorithm (PAPA) for emission-computed tomography. In this new generalized algorithm (GPAPA), the penalty term is expanded to allow multiple components, in this case the sum of the 1st to 4th order gradients, to reduce artificial piece-wise constant regions (“staircase” artifacts typical for TV) seen in PAPA images penalized with only the 1st order gradient. Simulated data were used to test for “staircase” artifacts and to optimize the penalty hyper-parameter in the root-mean-squared error (RMSE) sense. Patient FDG brain scans were acquired on a GE D690 PET/CT (370 MBq at 1-hour post-injection for 10 minutes) in time-of-flight mode and in all cases were reconstructed using resolution recovery projectors. GPAPA images were compared PAPA and RMSE-optimally filtered OSEM (fully converged) in simulations and to clinical OSEM reconstructions (3 iterations, 32 subsets) with 2.6 mm XYGaussian and standard 3-point axial smoothing post-filters. Results: The results from the simulated data show a significant reduction in the 'staircase' artifact for GPAPA compared to PAPA and lower RMSE (up to 35%) compared to optimally filtered OSEM. A simple power-law relationship between the RMSE-optimal hyper-parameters and the noise equivalent counts (NEC) per voxel is revealed. Qualitatively, the patient images appear much sharper and with less noise than standard clinical images. The convergence rate is similar to OSEM. Conclusions: GPAPA reconstructions using the 1{sub 1}-norm total-variation sum of the 1st through 4th-order gradients as the penalty show great promise for the improvement of image quality over that currently achieved with clinical OSEM reconstructions.
Purpose: To ascertain the feasibility of acquiring real time images of small lung tumors from scattered photons while undergoing radiation treatment.There are several methodologies currently used to track tumor location such as MV-cine acquisition and kV fluoroscopy. However, MVcine offers no information parallel to the beam axis while kV fluoroscopy offers little potential for soft tissue discernability while also increasing the patient dose. This study investigates the feasibility of observing an actual simulated tumor while exploring techniques that may improve image quality. Methods: A prototype imager consisting of a gamma camera pinhole collimator and a computed radiography (CR) plate were used in conjunction with a Varian TrueBeam linac. One study consisted of a 2.5 cm diameter solid water cylinder representing a solid tumor imbedded within a lung equivalent material slab. The cylinder with the lung slab was sandwiched between 1 cm lung equivalent slabs and these were sandwiched between 2 slabs of solid water. The top water slab was 1 cm thick. The other imaging study consisted of three different density plugs, 0.46, 1.09, and 1.82 g/cm3 placed on the accelerator couch. The gantry was orientated 70° relative to the CR plate. The slabs and plugs were irradiated with 2000 MU and 500 MU respectively using the 6FFF mode. Results: The solid water plug was visually discernible in the slab phantom. The ratio of the signal coming from the higher density plugs (placed on the treatment couch) to that between the plugs increased from 1.02 to about 3.0 after subtracting the background image acquired with no plugs present. Conclusion: Preliminary results indicate that a lung tumor could be visualized with scattered radiation during treatment. Improvements in discerning an object can be enhanced by filtering out the head leakage and background scattered radiation not emanating from the imaged object.
Purpose: To evaluate the dosimetric implications resulting from organ segmentation performed by different clinical experts Methods: Twelve patients received SBRT treatment to thoracic region within the past year were selected for this study. Three physicians contoured a set of organs following RTOG guideline. DVHs of all contours were generated from the approved plans used for treatment, and were compared to those produced during planning. Most OARs were evaluated on their max dose, some, such as heart and chest wall, were also evaluated on metrics such as max dose to 4cc of volume, or 30Gy volume dose. Results: In general, there is a greater dosimetric difference between the RTOG contour sets and clinical contour sets than among the three RTOG contour sets themselves for each patient. For example, there was no difference in esophagus max dose between the RTOG contour sets for ten patients. However, they showed an average of 2.3% higher max dose than the clinical contour set, with a standard deviation of 6.6%. The proximal bronchial tree (PBT) showed a similar behavior. The average difference of PBT max dose for seven patients is 0% between the three RTOG contour sets, with standard deviation of 1%. They showed an average of 16.1% higher max dose than the clinical contour set, with a standard deviation of 126%. Conclusion: This study shows that using RTOG contouring standards improves segmentation consistency between different physicians; most of the contours examined showed less than 1% dose difference. When RTOG contour sets were compared to the clinical contour set, the differences are much more significant. Thus it is important to standardize contouring guidelines in radiation therapy treatment planning. This will reduce uncertainties in clinical outcome analysis and research studies.
Abstract not provided
Laser-driven acceleration in dielectric structures can provide gradients on the order of GeV/m. The small transverse dimension and tiny feature sizes introduce challenges in design, fabrication, and simulation studies of these structures. In this paper we present the results of beam dynamic simulation and short range longitudinal wakefield simulation of the double grating structure. We show the linear trend of acceleration in a dielectric accelerator design and calculate the maximum achievable gradient equal to 0.47E{sub 0} where E0 is maximum electric field of the laser excitation. On the other hand, using wakefield simulations, we show that the loss factor of the structure with 400nm gap size will be 0.12GV/m for a 10fC, 100as electron bunch which is an order of magnitude less than expected gradient near damage threshold of the device.