ABSTRACT
PURPOSE: Organ-at-risk segmentation is essential in adaptive radiotherapy (ART). Learning-based automatic segmentation can reduce committed labor and accelerate the ART process. In this study, an auto-segmentation model was developed by employing individual patient datasets and a deep-learning-based augmentation method for tailoring radiation therapy according to the changes in the target and organ of interest in patients with prostate cancer. METHODS: Two computed tomography (CT) datasets with well-defined labels, including contoured prostate, bladder, and rectum, were obtained from 18 patients. The labels of the CT images captured during radiation therapy (CT2nd) were predicted using CT images scanned before radiation therapy (CT1st). From the deformable vector fields (DVFs) created by using the VoxelMorph method, 10 DVFs were extracted when each of the modified CT and CT2nd images were deformed and registered to the fixed CT1st image. Augmented images were acquired by utilizing 110 extracted DVFs and spatially transforming the CT1st images and labels. An nnU-net autosegmentation network was trained by using the augmented images, and the CT2nd label was predicted. A patient-specific model was created for 18 patients, and the performances of the individual models were evaluated. The results were evaluated by employing the Dice similarity coefficient (DSC), average Hausdorff distance, and mean surface distance. The accuracy of the proposed model was compared with those of models trained with large datasets. RESULTS: Patient-specific models were developed successfully. For the proposed method, the DSC values of the actual and predicted labels for the bladder, prostate, and rectum were 0.94 ± 0.03, 0.84 ± 0.07, and 0.83 ± 0.04, respectively. CONCLUSION: We demonstrated the feasibility of automatic segmentation by employing individual patient datasets and image augmentation techniques. The proposed method has potential for clinical application in automatic prostate segmentation for ART.
Subject(s)
Deep Learning , Prostatic Neoplasms , Radiotherapy Planning, Computer-Assisted , Tomography, X-Ray Computed , Humans , Prostatic Neoplasms/radiotherapy , Prostatic Neoplasms/diagnostic imaging , Male , Tomography, X-Ray Computed/methods , Radiotherapy Planning, Computer-Assisted/methods , Image Processing, Computer-Assisted/methods , Organs at Risk , Urinary Bladder/diagnostic imaging , Prostate/diagnostic imaging , Prostate/pathologyABSTRACT
BACKGROUND: Previous study proposed a method to measure linear energy transfer (LET) at specific points using the quenching magnitude of thin film solar cells. This study was conducted to propose a more advanced method for measuring the LET distribution. PURPOSE: This study focuses on evaluating the feasibility of estimating the proton LET distribution in proton therapy. The feasibility of measuring the proton LET and dose distribution simultaneously using a single-channel configuration comprising two solar cells with distinct quenching constants is investigated with the objective of paving the way for enhanced proton therapy dosimetry. METHODS: Two solar cells with different quenching constants were used to estimate the proton LET distribution. Detector characteristics (e.g., dose linearity and dose-rate dependency) of the solar cells were evaluated to assess their suitability for dosimetry applications. First, using a reference beam condition, the quenching constants of the two solar cells were determined according to the modified Birks equation. The signal ratios of the two solar cells were then evaluated according to proton LET in relation to the estimated quenching constants. The proton LET distributions of six test beams were obtained by measuring the signal ratios of the two solar cells at each depth, and the ratios were evaluated by comparing them with those calculated by Monte Carlo simulation. RESULTS: The detector characterization of the two solar cells including dose linearity and dose-rate dependence affirmed their suitability for use in dosimetry applications. The maximum difference between the LET measured using the two solar cells and that calculated by Monte Carlo simulation was 2.34 keV/µm. In the case of the dose distribution measured using the method proposed in this study, the maximum difference between range measured using the proposed method and that measured using a multilayered ionization chamber was 0.7 mm. The expected accuracy of simultaneous LET and dose distribution measurement using the method proposed in this study were estimated to be 3.82%. The signal ratios of the two solar cells, which are related to quenching constants, demonstrated the feasibility of measuring LET and dose distribution simultaneously. CONCLUSION: The feasibility of measuring proton LET and dose distribution simultaneously using two solar cells with different quenching constants was demonstrated. Although the method proposed in this study was evaluated using a single channel by varying the measuring depth, the results suggest that the proton LET and dose distribution can be simultaneously measured if the detector is configured in a multichannel form. We believe that the results presented in this study provide the envisioned transition to a multichannel configuration, with the promise of substantially advancing proton therapy's accuracy and efficacy in cancer treatment.
ABSTRACT
BACKGROUND: Quality assurance (QA) is a prerequisite for safe and accurate pencil-beam proton therapy. Conventional measurement-based patient-specific QA (pQA) can only verify limited aspects of patient treatment and is labor-intensive. Thus, a better method is needed to ensure the integrity of the treatment plan. PURPOSE: Line scanning, which involves continuous and rapid delivery of pencil beams, is a state-of-the-art proton therapy technique. Machine performance in delivering scanning protons is dependent on the complexity of the beam modulations. Moreover, it contributes to patient treatment accuracy. A Monte Carlo (MC) simulation-based QA method that reflects the uncertainty related to the machine during scanning beam delivery was developed and verified for clinical applications to pQA. METHODS: Herein, a tool for particle simulation (TOPAS) for nozzle modeling was used, and the code was commissioned against the measurements. To acquire the beam delivery uncertainty for each plan, patient plans were delivered. Furthermore, log files recorded every 60 µs by the monitors downstream of the nozzle were exported from the treatment control system. The spot positions and monitor unit (MU) counts in the log files were converted to dipole magnet strengths and number of particles, respectively, and entered into the TOPAS. For the 68 clinical cases, MC simulations were performed in a solid water phantom, and two-dimensional (2D) absolute dose distributions at 20-mm depth were measured using an ionization chamber array (Octavius 1500, PTW, Freiburg, Germany). Consequently, the MC-simulated 2D dose distributions were compared with the measured data, and the dose distributions in the pre-treatment QA plan created with RayStation (RaySearch Laboratories, Stockholm, Sweden). Absolute dose comparisons were made using gamma analysis with 3%/3 mm and 2%/2 mm criteria for 47 clinical cases without considering daily machine output variation in the MC simulation and 21 cases with daily output variation, respectively. All cases were analyzed with 90% or 95% of passing rate thresholds. RESULTS: For 47 clinical cases not considering daily output variations, the absolute gamma passing rates compared with the pre-treatment QA plan were 99.71% and 96.97%, and the standard deviations (SD) were 0.70% and 3.78% with the 3%/3 mm or 2%/2 mm criteria, respectively. Compared with the measurements, the passing rate of 2%/2 mm gamma criterion was 96.76% with 3.99% of SD. For the 21 clinical cases compared with pre-treatment QA plan data and measurements considering daily output variations, the 2%/2 mm absolute gamma analysis result was 98.52% with 1.43% of SD and 97.67% with 2.72% of SD, respectively. With a 95% passing rate threshold of 2%/2 mm criterion, the false-positive and false-negative were 21.8% and 8.3% for without and with considering output variation, respectively. With a 90% threshold, the false-positive and false-negative reduced to 11.4% and 0% for without and with considering output variation, respectively. CONCLUSIONS: A log-file-based MC simulation method for patient QA of line-scanning proton therapy was successfully developed. The proposed method exhibited clinically acceptable accuracy, thereby exhibiting a potential to replace the measurement-based dosimetry QA method with a 90% gamma passing rate threshold when applying the 2%/2 mm criterion.
Subject(s)
Proton Therapy , Protons , Humans , Proton Therapy/methods , Monte Carlo Method , Radiotherapy Planning, Computer-Assisted/methods , Radiotherapy DosageABSTRACT
(1) In this study, we developed a deep learning (DL) model that can be used to predict late bladder toxicity. (2) We collected data obtained from 281 uterine cervical cancer patients who underwent definitive radiation therapy. The DL model was trained using 16 features, including patient, tumor, treatment, and dose parameters, and its performance was compared with that of a multivariable logistic regression model using the following metrics: accuracy, prediction, recall, F1-score, and area under the receiver operating characteristic curve (AUROC). In addition, permutation feature importance was calculated to interpret the DL model for each feature, and the lightweight DL model was designed to focus on the top five important features. (3) The DL model outperformed the multivariable logistic regression model on our dataset. It achieved an F1-score of 0.76 and an AUROC of 0.81, while the corresponding values for the multivariable logistic regression were 0.14 and 0.43, respectively. The DL model identified the doses for the most exposed 2 cc volume of the bladder (BD2cc) as the most important feature, followed by BD5cc and the ICRU bladder point. In the case of the lightweight DL model, the F-score and AUROC were 0.90 and 0.91, respectively. (4) The DL models exhibited superior performance in predicting late bladder toxicity compared with the statistical method. Through the interpretation of the model, it further emphasized its potential for improving patient outcomes and minimizing treatment-related complications with a high level of reliability.
ABSTRACT
PURPOSE: The amount of luminescent light detected in a scintillator is reduced with increased proton linear energy transfer (LET) despite receiving the same proton dose, through a phenomenon called quenching. This study evaluated the ability of a solar cell coated with scintillating powder (SC-SP) to measure therapeutic proton LET by measuring the quenching effect of the scintillating powder using a solar cell while simultaneously measuring the dose of the proton beam. METHODS: SC-SP was composed of a flexible thin film solar cell and scintillating powder. The LET and dose of the pristine Bragg peak in the 14 cm range were calculated using a validated Monte Carlo model of a double scattering proton beam nozzle. The SC-SP was evaluated by measuring the proton beam under the same conditions at specific depths using SC-SP and Markus chamber. Finally, the 10 and 20 cm range pristine Bragg peaks and 5 cm spread-out Bragg peak (SOBP) in the 14 cm range were measured using the SC-SP and the Markus chamber. LETs measured using the SC-SP were compared with those calculated using Monte Carlo simulations. RESULTS: The quenching factors of the SC-SP and solar cell alone, which were slopes of linear fit obtained from quenching correction factors according to LET, were 0.027 and 0.070 µm/keV (R2 : 0.974 and 0.975). For pristine Bragg peaks in the 10 and 20 cm ranges, the maximum differences between LETs measured using the SC-SP and calculated using Monte Carlo simulations were 0.5 keV/µm (15.7%) and 1.2 keV/µm (12.0%), respectively. For a 5 cm SOBP proton beam, the LET measured using the SC-SP and calculated using Monte Carlo simulations differed by up to 1.9 keV/µm (18.7%). CONCLUSIONS: Comparisons of LETs for pristine Bragg peaks and SOBP between measured using the SC-SP and calculated using Monte Carlo simulations indicated that the solar cell-based system could simultaneously measure both LET and dose in real-time and is cost-effective.
Subject(s)
Proton Therapy , Protons , Powders , Linear Energy Transfer , Monte Carlo MethodABSTRACT
PURPOSE: A real-time solar cell based in vivo dosimetry system (SC-IVD) was developed using a flexible thin film solar cell and scintillating powder. The present study evaluated the clinical feasibility of the SC-IVD in electron beam therapy. METHODS: A thin film solar cell was coated with 100 mg of scintillating powder using an optical adhesive to enhance the sensitivity of the SC-IVD. Calibration factors were obtained by dividing the dose, measured at a reference depth for 6-20 MeV electron beam energy, by the signal obtained using the SC-IVD. Dosimetric characteristics of SC-IVDs containing variable quantities of scintillating powder (0-500 mg) were evaluated, including energy, dose rate, and beam angle dependencies, as well as dose linearity. To determine the extent to which the SC-IVD affected the dose to the medium, doses at R90 were compared depending on whether the SC-IVD was on the surface. Finally, the accuracy of surface doses measured using the SC-IVD was evaluated by comparison with surface doses measured using a Markus chamber. RESULTS: Charge measured using the SC-IVD increased linearly with dose and was within 1% of the average signal according to the dose rate. The signal generated by the SC-IVD increased as the beam angle increased. The presence of the SC-IVD on the surface of a phantom resulted in a 0.5%-2.2% reduction in dose at R90 for 6-20 MeV electron beams compared with the bare phantom. Surface doses measured using the SC-IVD system and Markus chamber differed by less than 5%. CONCLUSIONS: The dosimetric characteristics of the SC-IVD were evaluated in this study. The results showed that it accurately measured the surface dose without a significant difference of dose in the medium when compared with the Markus chamber. The flexibility of the SC-IVD allows it to be attached to a patient's skin, enabling real-time and cost-effective measurement.
Subject(s)
Electrons , In Vivo Dosimetry , Humans , Powders , Radiometry/methods , Film Dosimetry/methodsABSTRACT
This research addresses the problem of interobserver variability (IOV), in which different oncologists manually delineate varying primary gross tumor volume (pGTV) contours, adding risk to targeted radiation treatments. Thus, a method of IOV reduction is urgently needed. Hypothesizing that the radiation oncologist's IOV may shrink with the aid of IOV maps, we propose IOV prediction network (IOV-Net), a deep-learning model that uses the fuzzy membership function to produce high-quality maps based on computed tomography (CT) images. To test the prediction accuracy, a ground-truth pGTV IOV map was created using the manual contour delineations of radiation therapy structures provided by five expert oncologists. Then, we tasked IOV-Net with producing a map of its own. The mean squared error (prediction vs. ground truth) and its standard deviation were 0.0038 and 0.0005, respectively. To test the clinical feasibility of our method, CT images were divided into two groups, and oncologists from our institution created manual contours with and without IOV map guidance. The Dice similarity coefficient and Jaccard index increased by ~6 and 7%, respectively, and the Hausdorff distance decreased by 2.5 mm, indicating a statistically significant IOV reduction (p < 0.05). Hence, IOV-net and its resultant IOV maps have the potential to improve radiation therapy efficacy worldwide.
ABSTRACT
For accurate respiration gated radiation therapy, compensation for the beam latency of the beam control system is necessary. Therefore, we evaluate deep learning models for predicting patient respiration signals and investigate their clinical feasibility. Herein, long short-term memory (LSTM), bidirectional LSTM (Bi-LSTM), and the Transformer are evaluated. Among the 540 respiration signals, 60 signals are used as test data. Each of the remaining 480 signals was spilt into training and validation data in a 7:3 ratio. A total of 1000 ms of the signal sequence (Ts) is entered to the models, and the signal at 500 ms afterward (Pt) is predicted (standard training condition). The accuracy measures are: (1) root mean square error (RMSE) and Pearson correlation coefficient (CC), (2) accuracy dependency on Ts and Pt, (3) respiratory pattern dependency, and (4) error for 30% and 70% of the respiration gating for a 5 mm tumor motion for latencies of 300, 500, and 700 ms. Under standard conditions, the Transformer model exhibits the highest accuracy with an RMSE and CC of 0.1554 and 0.9768, respectively. An increase in Ts improves accuracy, whereas an increase in Pt decreases accuracy. An evaluation of the regularity of the respiratory signals reveals that the lowest predictive accuracy is achieved with irregular amplitude patterns. For 30% and 70% of the phases, the average error of the three models is <1.4 mm for a latency of 500 ms and >2.0 mm for a latency of 700 ms. The prediction accuracy of the Transformer is superior to LSTM and Bi-LSTM. Thus, the three models have clinically applicable accuracies for a latency <500 ms for 10 mm of regular tumor motion. The clinical acceptability of the deep learning models depends on the inherent latency and the strategy for reducing the irregularity of respiration.
Subject(s)
Deep Learning , Neoplasms , Humans , Motion , Respiration , Protein Sorting SignalsABSTRACT
PURPOSE: To evaluate the dosimetric characteristics and applications of a dosimetry system composed of a flexible amorphous silicon thin-film solar cell and scintillator screen (STFSC-SS) for therapeutic X-rays. METHODS: The real-time dosimetry system was composed of a flexible a-Si thin-film solar cell (0.2-mm thick), a scintillator screen to increase its efficiency, and an electrometer to measure the generated charge. The dosimetric characteristics of the developed system were evaluated including its energy dependence, dose linearity, and angular dependence. Calibration factors for the signal measured by the system and absorbed dose-to-water were obtained by setting reference conditions. The application and correction accuracy of the developed system were evaluated by comparing the absorbed dose-to-water measured using a patient treatment beam with that measured using the ion chamber. RESULTS: The responses of STFSC-SS to energy, field size, depth, and source-to-surface distance (SSD) were more dependent on measurement conditions than were the responses of the ion chamber, although the former dependence was due to the scintillator screen, not the solar cell. The signals of STFSC-SS were also dependent on dose rate, while the responses of solar cell alone and scintillator screen were not dependent on dose rate. The scintillator screen reduced the output of solar cell at 6 and 15 MV by 0.60 and 0.55%, respectively. The different absorbed dose-to-water measured using STFSC-SS for patient treatment beam differed by 0.4% compared to those measured using the ionization chamber. The uncertainties of the developed system for 6 and 15 MV photon beams were 1.8 and 1.7%, respectively, confirming the accuracy and applicability of this system. CONCLUSIONS: The thin-film solar cell-based detector developed in this study can accurately measure absorbed dose-to-water. The increased signal resulting from the use of the scintillator screen is advantageous for measuring low doses and stable signal output. In addition, this system is flexible, making it applicable to curved surfaces, such as a patient's body, and is cost-effective.
Subject(s)
Radiometry , Silicon , Humans , Radiography , Radiometry/methods , Water , X-RaysABSTRACT
To automatically identify optimal beam angles for proton therapy configured with the double-scattering delivery technique, a beam angle optimization method based on a convolutional neural network (BAODS-Net) is proposed. Fifty liver plans were used for training in BAODS-Net. To generate a sequence of input data, 25 rays on the eye view of the beam were determined per angle. Each ray collects nine features, including the normalized Hounsfield unit and the position information of eight structures per 2° of gantry angle. The outputs are a set of beam angle ranking scores (S beam) ranging from 0° to 359°, with a step size of 1°. Based on these input and output designs, BAODS-Net consists of eight convolution layers and four fully connected layers. To evaluate the plan qualities of deep-learning, equi-spaced, and clinical plans, we compared the performances of three types of loss functions and performed K-fold cross-validation (K = 5). For statistical analysis, the volumes V27Gy and V30Gy as well as the mean, minimum, and maximum doses were calculated for organs-at-risk by using a paired-samples t-test. As a result, smooth-L1 loss showed the best optimization performance. At the end of the training procedure, the mean squared errors between the reference and predicted S beam were 0.031, 0.011, and 0.004 for L1, L2, and smooth-L1 loss, respectively. In terms of the plan quality, statistically, PlanBAO has no significant difference from PlanClinic (P >.05). In our test, a deep-learning based beam angle optimization method for proton double-scattering treatments was developed and verified. Using Eclipse API and BAODS-Net, a plan with clinically acceptable quality was created within 5 min.
ABSTRACT
BACKGROUND: Patient-specific dose prediction improves the efficiency and quality of radiation treatment planning and reduces the time required to find the optimal plan. In this study, a patient-specific dose prediction model was developed for a left-sided breast clinical case using deep learning, and its performance was compared with that of conventional knowledge-based planning using RapidPlan™. METHODS: Patient-specific dose prediction was performed using a contour image of the planning target volume (PTV) and organs at risk (OARs) with a U-net-based modified dose prediction neural network. A database of 50 volumetric modulated arc therapy (VMAT) plans for left-sided breast cancer patients was utilized to produce training and validation datasets. The dose prediction deep neural network (DpNet) feature weights of the previously learned convolution layers were applied to the test on a cohort of 10 test sets. With the same patient data set, dose prediction was performed for the 10 test sets after training in RapidPlan. The 3D dose distribution, absolute dose difference error, dose-volume histogram, 2D gamma index, and iso-dose dice similarity coefficient were used for quantitative evaluation of the dose prediction. RESULTS: The mean absolute error (MAE) and one standard deviation (SD) between the clinical and deep learning dose prediction models were 0.02 ± 0.04%, 0.01 ± 0.83%, 0.16 ± 0.82%, 0.52 ± 0.97, - 0.88 ± 1.83%, - 1.16 ± 2.58%, and - 0.97 ± 1.73% for D95%, Dmean in the PTV, and the OARs of the body, left breast, heart, left lung, and right lung, respectively, and those measured between the clinical and RapidPlan dose prediction models were 0.02 ± 0.14%, 0.87 ± 0.63%, - 0.29 ± 0.98%, 1.30 ± 0.86%, - 0.32 ± 1.10%, 0.12 ± 2.13%, and - 1.74 ± 1.79, respectively. CONCLUSIONS: In this study, a deep learning method for dose prediction was developed and was demonstrated to accurately predict patient-specific doses for left-sided breast cancer. Using the deep learning framework, the efficiency and accuracy of the dose prediction were compared to those of RapidPlan. The doses predicted by deep learning were superior to the results of the RapidPlan-generated VMAT plan.
Subject(s)
Breast Neoplasms/radiotherapy , Deep Learning , Radiotherapy Planning, Computer-Assisted/methods , Radiotherapy, Intensity-Modulated/methods , Female , Humans , Middle Aged , Organs at Risk , Radiotherapy DosageABSTRACT
PURPOSE: We developed a compact and lightweight time-resolved mirrorless scintillation detector (TRMLSD) employing image processing techniques and a convolutional neural network (CNN) for high-resolution two-dimensional (2D) dosimetry. METHODS: The TRMLSD comprises a camera and an inorganic scintillator plate without a mirror. The camera was installed at a certain angle from the horizontal plane to collect scintillation from the scintillator plate. The geometric distortion due to the absence of a mirror and camera lens was corrected using a projective transform. Variations in brightness due to the distance between the image sensor and each point on the scintillator plate and the inhomogeneity of the material constituting the scintillator were corrected using a 20.0 × 20.0 cm2 radiation field. Hot pixels were removed using a frame-based noise-reduction technique. Finally, a CNN-based 2D dose distribution deconvolution model was applied to compensate for the dose error in the penumbra region and a lack of backscatter. The linearity, reproducibility, dose rate dependency, and dose profile were tested for a 6 MV X-ray beam to verify dosimeter characteristics. Gamma analysis was performed for two simple and 10 clinical intensity-modulated radiation therapy (IMRT) plans. RESULTS: The dose linearity with brightness ranging from 0.0 cGy to 200.0 cGy was 0.9998 (R-squared value), and the root-mean-square error value was 1.010. For five consecutive measurements, the reproducibility was within 3% error, and the dose rate dependency was within 1%. The depth dose distribution and lateral dose profile coincided with the ionization chamber data with a 1% mean error. In 2D dosimetry for IMRT plans, the mean gamma passing rates with a 3%/3 mm gamma criterion for the two simple and ten clinical IMRT plans were 96.77% and 95.75%, respectively. CONCLUSION: The verified accuracy and time-resolved characteristics of the dosimeter may be useful for the quality assurance of machines and patient-specific quality assurance for clinical step-and-shoot IMRT plans.
Subject(s)
Image Processing, Computer-Assisted/methods , Radiometry/instrumentation , Radiometry/methods , Radiotherapy, Intensity-Modulated/methods , Scintillation Counting/instrumentation , Scintillation Counting/methods , Gamma Cameras , Humans , Neural Networks, Computer , Radiotherapy Dosage , Reproducibility of Results , X-RaysABSTRACT
This study aimed to investigate the performance of a deep learning-based survival-prediction model, which predicts the overall survival (OS) time of glioblastoma patients who have received surgery followed by concurrent chemoradiotherapy (CCRT). The medical records of glioblastoma patients who had received surgery and CCRT between January 2011 and December 2017 were retrospectively reviewed. Based on our inclusion criteria, 118 patients were selected and semi-randomly allocated to training and test datasets (3:1 ratio, respectively). A convolutional neural network-based deep learning model was trained with magnetic resonance imaging (MRI) data and clinical profiles to predict OS. The MRI was reconstructed by using four pulse sequences (22 slices) and nine images were selected based on the longest slice of glioblastoma by a physician for each pulse sequence. The clinical profiles consist of personal, genetic, and treatment factors. The concordance index (C-index) and integrated area under the curve (iAUC) of the time-dependent area-under-the-curve curves of each model were calculated to evaluate the performance of the survival-prediction models. The model that incorporated clinical and radiomic features showed a higher C-index (0.768 (95% confidence interval (CI): 0.759, 0.776)) and iAUC (0.790 (95% CI: 0.783, 0.797)) than the model using clinical features alone (C-index = 0.693 (95% CI: 0.685, 0.701); iAUC = 0.723 (95% CI: 0.716, 0.731)) and the model using radiomic features alone (C-index = 0.590 (95% CI: 0.579, 0.600); iAUC = 0.614 (95% CI: 0.607, 0.621)). These improvements to the C-indexes and iAUCs were validated using the 1000-times bootstrapping method; all were statistically significant (p < 0.001). This study suggests the synergistic benefits of using both clinical and radiomic parameters. Furthermore, it indicates the potential of multi-parametric deep learning models for the survival prediction of glioblastoma patients.
ABSTRACT
PURPOSE: The objective of this work is to determine mechanical, radiation, and imaging isocentres in three-dimensional (3D) coordinates and verifying coincidence of isocentres of passively scattered proton beam using a visual tracking system (VTS) and an in-house developed phantom named the Eagle. METHODS: The Eagle phantom consists of two modules: The first, named Eagle-head, is used for determining 3D mechanical isocentre of gantry rotation. The second, named Eagle-body, is used for determining 3D radiation and imaging isocentres. The Eagle-body has four slots wherein radiochromic films were inserted for measuring the 3D radiation isocentre and a metal bead was embedded in the centre of one cube to determine the imaging isocentre; this was determined by analysing cone-beam computed tomography images of the cube. Infrared reflective markers that can be tracked by VTS were attached to the Eagle at predetermined locations. The tracked data were converted into 3D treatment room coordinates. The developed method was compared with other methods to assess accuracy. RESULTS: The isocentres were determined in mm with respect to the laser isocentre. The mechanical, radiation, and imaging isocentres were (-0.289, 0.189, 0.096), (-0.436, -0.217, 0.009), and (0.134, 0.142, 0.103), respectively. When compared with other methods, the difference in coordinates was (-0.033, -0.107, 0.014) and (0.003, 0.067, 0.039) for radiation and imaging isocentres, respectively. CONCLUSION: The developed system was found to be useful in providing fast and accurate measurements of the three isocentres in the 3D treatment room coordinate system.
Subject(s)
Cone-Beam Computed Tomography/instrumentation , Proton Therapy/methods , Protons , Quality Assurance, Health Care/statistics & numerical data , Algorithms , Equipment Design , Humans , Motion , Phantoms, Imaging , Radioactive TracersABSTRACT
PURPOSE: The purpose of this study was to investigate the feasibility of two-dimensional (2D) dose distribution deconvolution using convolutional neural networks (CNNs) instead of an analytical approach for an in-house scintillation detector that has a detector-interface artifact in the penumbra region. METHODS: Datasets of 2D dose distributions were acquired from a medical linear accelerator of Novalis Tx. The datasets comprise two different sizes of square radiation fields and 13 clinical intensity-modulated radiation treatment (IMRT) plans. These datasets were divided into two datasets (training and test) to train and validate the developed network, called PenumbraNet, which is a shallow linear CNN. The PenumbraNet was trained to transform the measured dose distribution [M(x, y)] to calculated distribution [D(x, y)] by the treatment planning system. After training of the PenumbraNet was completed, the performance was evaluated using test data, which were 10 × 10 cm2 open field and ten clinical IMRT cases. The corrected dose distribution [C(x, y)] was evaluated against D(x, y) with 2%/2 mm and 3%/3 mm criteria of the gamma index for each field. The M(x, y) and deconvolved dose distribution with the analytically obtained kernel using Wiener filtering [A(x, y)] were also evaluated for comparison. In addition, we compared the performance of the shallow depth of linear PenumbraNet with that of nonlinear PenumbraNet and a deep nonlinear PenumbraNet within the same training epoch. RESULTS: The mean gamma passing rates were 84.77% and 95.81% with 3%/3 mm gamma criteria for A(x, y) and C(x, y) of the PenumbraNet, respectively. The mean gamma pass rates of nonlinear PenumbraNet and the deep depth of nonlinear PenumbraNet were 96.62%, 93.42% with 3%/3 mm gamma criteria, respectively. CONCLUSIONS: We demonstrated the feasibility of the PenumbraNets for 2D dose distribution deconvolution. The nonlinear PenumbraNet which has the best performance improved the gamma passing rate by 11.85% from the M(x, y) at 3%/3 mm gamma criteria.
Subject(s)
Neural Networks, Computer , Radiation Dosage , Radiotherapy Planning, Computer-Assisted/methods , Feasibility Studies , Humans , Radiometry , Radiotherapy Dosage , Radiotherapy, Intensity-ModulatedABSTRACT
A robotic couch capable of six degrees of freedom (6-DoF) of motion was introduced for state-of-the-art radiation therapy. Patient treatment requires precise quality assurance (QA) of 6-DoF. Unfortunately, conventional methods do not provide the requisite accuracy and precision. Therefore, we developed a high-precision automated QA system using a visual tracking system (VTS). The VTS comprises four motion-sensing cameras, a cube with infrared reflective markers. To acquire data in treatment room coordinates, a transformation matrix from VTS coordinates to treatment room coordinates was determined. The mean error and standard deviation of linear and rotational motions, as well as couch sagging were analyzed from continuously acquired images in the moving couch. The accuracy of VTS was 0.024â¯mm deviation for the sinusoidal motion, and the accuracy of the transformation matrix was 0.02â¯mm. In a cross-comparison, the difference between Laser Tracker (FARO) measurements was 0.14⯱â¯0.12â¯mm for translation and 0.032⯱â¯0.026° on average for yaw rotation. The new system provides QA of yaw, pitch and roll motion as well as sagging of the couch and sub-millimeter/degree accuracy together with precision.
Subject(s)
Motion , Quality Assurance, Health Care , Radiotherapy/instrumentation , Robotics/instrumentation , Humans , Patient Positioning/instrumentationABSTRACT
Gold nanoparticles (GNPs) injected in a body for dose enhancement in radiation therapy are known to form clusters. We investigated the dependence of dose enhancement on the GNP morphology using Monte-Carlo simulations and compared the model predictions with experimental data. The cluster morphology was approximated as a body-centred cubic (BCC) structure by placing GNPs at the 8 corners and the centre of a cube with an edge length of 0.22-1.03 µm in a 4 × 4 × 4 µm3 water-filled phantom. We computed the dose enhancement ratio (DER) for 50 and 260 kVp photons as a function of the distance from the cube centre for 12 different cube sizes. A 10 nm-wide concentric shell shaped detector was placed up to 100 nm away from a GNP at the cube centre. For model validation, simulations based on BCC and nanoparticle random distribution (NRD) models were performed using parameters that corresponded to the experimental conditions, which measured increases in the relative biological effect due to GNPs. We employed the linear quadratic model to compute cell surviving fraction (SF) and sensitizer enhancement ratio (SER). The DER is inversely proportional to the distance to the GNPs. The largest DERs were 1.97 and 1.80 for 50 kVp and 260 kVp photons, respectively. The SF predicted by the BCC model agreed with the experimental value within 10%, up to a 5 Gy dose, while the NRD model showed a deviation larger than 10%. The SERs were 1.21 ± 0.13, 1.16 ± 0.11, and 1.08 ± 0.11 according to the experiment, BCC, and NRD models, respectively. We most accurately predicted the GNP radiosensitization effect using the BCC approximation and suggest that the BCC model is effective for use in nanoparticle dosimetry.
Subject(s)
Gold/chemistry , Metal Nanoparticles/chemistry , Monte Carlo Method , Phantoms, Imaging , Photons/therapeutic use , Radiation-Sensitizing Agents , Humans , Radiometry , Radiotherapy Dosage , WaterABSTRACT
Target motion-induced uncertainty in particle therapy is more complicated than that in X-ray therapy, requiring more accurate motion management. Therefore, a hybrid motion-tracking system that can track internal tumor motion and as well as an external surrogate of tumor motion was developed. Recently, many correlation tests between internal and external markers in X-ray therapy have been developed; however, the accuracy of such internal/external marker tracking systems, especially in particle therapy, has not yet been sufficiently tested. In this article, the process of installing an in-house hybrid internal/external motion-tracking system is described and the accuracy level of tracking system was acquired. Our results demonstrated that the developed in-house external/internal combined tracking system has submillimeter accuracy, and can be clinically used as a particle therapy system as well as a simulation system for moving tumor treatment.