Search

Purpose: In the framework of developing the integration of a MRI-Linac system, configurations of MRI-Linac units were simulated in order to improve the dose distribution in tangential breast radiotherapy using transverse and longitudinal magnetic field geometries of Lorentz force for both medial and lateral tangential fields. Methods: In this work, the geant4 Monte Carlo (MC) code was utilized to compare dose distributions in breast radiotherapy for Linac-MR systems in the transverse and longitudinal geometries within humanoid phantoms across a range of magnetic field strengths of 0.5 and 1.5 T. The dose increment due to scattering from the coils was investigated for both geometries as well. Computed tomography images of two patients were used for MC simulations. One patient had intact breast while the other was mastectomized. In the simulations, planning and methods of chest wall irradiation were similar to the actual clinical planning. Results: In a longitudinal geometry, the magnetic field is shown to restrict the lateral spread of secondary electrons to the lung, heart, and contralateral organs, which reduced the mean dose of the ipsilateral lung and heart by means of 17.2% and 6% at 1.5 T, respectively. The transverse configuration exhibits a significant increase in tissue interface effects, which increased dose buildup in the entrance regions of the lateral and medial tangent beams to the planning target volume (PTV) and improved dose homogeneity within the PTV. The improved relative average homogeneity index for two patients to the PTV at magnetic field strength of 1.5 T with respect to no magnetic field case evaluated was 11.79% and 34.45% in the LRBP and TRBP geometries, respectively. In both geometries, the simulations show significant mean dose reductions in the contralateral breast and chest wall skin, respectively, by a mean of 16.6% and 24.9% at 0.5 T and 17.2% and 28.1% at 1.5 T in the transverse geometry, and 10.56% and 14.6% at 0.5 T and 11.3% and 16.3% at 1.5 T in the longitudinal geometry. Considering the scattered photons which reflected from the coils, the average relative dose of each voxel is slightly increased by 0.53% and 0.32% in the LRBP and TRBP geometries, respectively. Conclusions: Orienting the B0 magnetic field parallel to the photon beam axis, LRBP geometry, tends to restrict the radial spread of secondary electrons which resulted in dose reduction to the lung. Dosimetry issues observed in both Linac-MR geometries, such as changes to the lateral dose distribution, significantly exhibited dose reduction in the contralateral organs on a representative breast plan. Further, the results show sharper edge dose volume histogram curves at 1.5 T for both geometries, especially in the LRBP configuration.

Purpose: The aim of this study was to evaluate and analytically compare different calculation algorithms applied in our country radiotherapy centers base on the methodology developed by IAEA for treatment planning systems (TPS) commissioning (IAEA TEC‐DOC 1583). Material & Methods: Thorax anthropomorphic phantom (002LFC CIRS inc.), was used to measure 7 tests that simulate the whole chain of external beam TPS. The dose were measured with ion chambers and the deviation between measured and TPS calculated dose was reported. This methodology, which employs the same phantom and the same setup test cases, was tested in 4 different hospitals which were using 5 different algorithms/ inhomogeneity correction methods implemented in different TPS. The algorithms in this study were divided into two groups including correction based and model based algorithms. Results: A total of 84 clinical test case datasets for different energies and calculation algorithms were produced, which amounts of differences in inhomogeneity points with low density (lung) and high density (bone) was decreased meaningfully with advanced algorithms. The number of deviations outside agreement criteria was increased with the beam energy and decreased with advancement of the TPS calculation algorithm. Conclusion: Large deviations were seen in some correction based algorithms, so sophisticated algorithms, would be preferred in clinical practices, especially for calculation in inhomogeneous media. Use of model based algorithms with lateral transport calculation, is recommended. Some systematic errors which were revealed during this study, is showing necessity of performing periodic audits on TPS in radiotherapy centers.

PURPOSE Rectal toxicity is one of the most common side effects in prostate cancer radiotherapy. Prediction of such effect is of particularly interest to personalize the treatment. This study's purpose is to investigate the performance of predictive radiomic models developed based on clinical, dosimetric parameters and radiomic features extracted from rectal wall magnetic resonance image (MRI). METHODS This prospective study was conducted on 33 pathologically proven prostate cancer patients treated with intensity modulated radiation therapy (IMRT). Patient's clinical and dosimetric parameters were collected and rectal toxicity was assessed by common terminology criteria for adverse events (CTCAE). All patients underwent two MRI scans before and after IMRT course with same imaging protocol. A wide range of radiomic texture features were extracted from rectal wall MR images including T2 weighted (T2W) and apparent diffusion coefficient (ADC) scans (pre and post-IMRT) and robust features were found. Lasso regularization was used to create multivariable Elastic Net logistic regression model, based on the clinical, dosimetric and robust features for prediction and feature selection, simultaneously. All models were cross-validated using repeated 5-fold cross-validation and their performance were evaluated by area under the curve of the ROC curve (AUC), sensitivity and specificity. RESULTS Thirty-three patients were evaluated and eighteen patients were developed grade ≥ 1 rectal toxicity. From each ROI, 270 radiomic features were extracted and 37 radiomic features (14%) were found robust against delineations. Different clinical, dosimetric and imaging features were regularized and selected by Elastic Net. Thirteen predictive models including one clinical, two radiomics and eight integrated clinical-radiomic models were built. The predictive performance (AUC) for T2 and ADC-radiomic models were 0.636 and 0.788 respectively. We observed that the combination of T2 and ADC image features with clinical and dosimetric data have led to higher performance and T2-clinical-radiomic model had higher performance (AUC 0.933, Sensitivity 0.941 & Specificity 0.923) rather than ADC-clinical radiomics model (AUC 0.90, Sensitivity 0.941 & Specificity 0.846). Post, delta and total radiomic-clinical models also had good predictive performance. CONCLUSIONS Rectal wall MR radiomic features integrating with clinical and dose-volume parameters could improve the prediction performance of radiotherapy induced proctitis. The authors have demonstrated the performance of radiomic features extracted from pre IMRT T2W and ADC MR Images for prediction of IMRT induced rectal toxicity. These results may be used to tailor the therapy in terms of patient and treatment approach selection. Post-IMRT models also, may be used as hypothetical issues for predicting late rectal injuries.

Purpose: The OCTAVIUS detector729 is a 2D array of 729 air vented cubic plane parallel ion chambers used for pretreatment verification and QA. In this study we investigated dosimetric characteristics of this system for clinical photon beam dosimetry. Methods: Detector performance evaluation included determination of the location of the effective point of measurement (EPM), sensitivity, linearity, and reproducibility of detector response, as well as output factor, dose rate, and source to surface distance (SSD) dependence. Finally, assessment of wedge modulated fields was carried out. All the evaluations were performed five times for low and high photon energies. For reference measurements, a 0.6 cc ionization chamber was used. Data analysis and comparison of the OCTAVIUS detector with reference ion chamber data was performed using the VeriSoft patient plan verification software. Results: The reproducibility and stability of the measurements are excellent, the detector showed same signal with a maximum deviation of less than 0.5% in short and long term. Results of sensitivity test showed same signal with a maximum deviation of approximately 0.1%. As the detector 729 response is linear with dose and dose rate, it can be used for the measurement at regions of high dose gradient effectively. The detector agrees with the ionization chamber measurement to within 1% for SSD range of 75 to 125 cm. Also, its measured wedge modulated profiles matched very well with ion chamber dose profiles acquired in a water tank. Conclusions: As the response of the detector 729 is linear with dose and dose rate, it can be used for the measurements in the areas of dose gradients effectively. Based on the measurements and comparisons performed, this system is a reliable and accurate dosimeter for QA and pretreatment plan verification in radiotherapy.

Purpose: To investigate the improvement in dose distribution in tangential breast radiotherapy using a reversible transverse magnetic field that maintains the same direction of Lorentz force between two fields. The investigation has a potential application in future Linac-MR units. Methods: Computed tomography images of four patients and magnetic fields of 0.25–1.5 Tesla (T) were used for Monte Carlo simulation. Two patients had intact breast while the other two had mastectomy. Simulations of planning and chest wall irradiation were similar to the actual clinical process. The direction of superior-inferior magnetic field for the medial treatment beam was reversed for the lateral beam. Results: For the ipsilateral lung and heart mean doses were reduced by a mean (range) of 45.8% (27.6%–58.6%) and 26.0% (20.2%–38.9%), respectively, depending on various treatment plan setups. The mean V{sub 20} for ipsilateral lung was reduced by 55.0% (43.6%–77.3%). In addition acceptable results were shown after simulation of 0.25 T magnetic field demonstrated in dose-volume reductions of the heart, ipsilateral lung, and noninvolved skin. Conclusions: Applying a reversible magnetic field during breast radiotherapy, not only reduces the dose to the lung and heart but also produces a sharp drop dose volume histogram for planning target volume, because ofmore » bending of the path of secondary charged particles toward the chest wall by the Lorentz force. The simulations have shown that use of the magnetic field at 1.5 T is not feasible for clinical applications due to the increase of ipsilateral chest wall skin dose in comparison to the conventional planning while 0.25 T is suitable for all patients due to dose reduction to the chest wall skin.« less

Purpose: GZP6 machine is used for high dose rate (HDR) brachytherapy treatment. In this project, treatment planning of GYN brachytherapy of this machine has been evaluated using a newly fabricated phantom along Gafchromic films and TG‐43 calculation. Methods: A new QA phantom has been designed and fabricated from 90 slabs of Perspex to accommodate ovoids assembly of GYN brachytherapy treatment. Two large oviods with 3 cm diameter was used for this assembly. EBT Gafchoromic films were exactly sandwiched between slabs and completely fitted to shape of ovoids for verification of GZP6 treatment planning. Plan designed to deliver 2.5 Gy dose to surface of oviods. The measured doses values at several points of interest were evaluated. The measured data were compared with the calculated values by the GZP6 treatment planning software. With seeing large difference between film dosimetry and planning, Air kerma strength (Sk) was obtained from Monte Carlo simulation and considered for TG‐43 calculation to reduce error. Results: The Result of these investigations have indicated differences of up to ± 23.4 % between the treatment planning and measured film dosimetry data at different points in GYN implant with GZP6 60Co HDR system. Between air kerma strength used in planning and Sk from simulation, 26% difference was calculated. TG‐43 calculation with simulated Sk can reduces difference between measurements to around 9%. Conclusion: The new phantom could be utilized for the QA procedure of the GZP6 60Co HDR system or the Ir‐192 HDR system to confirmation the accuracy of treatment planning in GYN implants and to find source of systematic errors.