Intensity-modulated Radiation Therapy Treatment Plans Research Articles

Variable Dose-Constraints Method for Enhancing Intensity-Modulated Radiation Therapy Treatment Planning

The conventional approach to intensity-modulated radiation therapy treatment planning involves two distinct strategies: optimizing an evaluation function while accounting for dose constraints, and solving feasibility problems using feasibility-seeking projection methods that incorporate inequality constraints. This paper introduces a novel iterative scheme within the framework of continuous dynamical systems, wherein constraint conditions dynamically evolve to enhance the optimization process. The validity of dynamically varying dose constraints is theoretically established through the foundation of continuous-time dynamical systems theory. In particular, we formalize a system of differential equations, with both beam coefficients and dose constraints modeled as state variables. The asymptotic stability of the system’s equilibrium is rigorously proven, ensuring convergence to a solution. In practical terms, we leverage a discretized iteration formula derived from the continuous-time system to achieve rapid computational speed. The mathematical structure of the proposed approach, which directly incorporates dose-volume constraints into the objective function, facilitates significant computational efficiency and solution refinement. The proposed method has an inherent dynamics that approaches more desirable solutions within the set of solutions when the solution to the optimization problem is not an isolated point. This property guarantees the identification of optimal solutions that respect the prescribed dose-volume constraints while enhancing accuracy when such constraints are feasible. By treating dose constraints as variables and concurrently solving the optimization problem with beam coefficients, we can achieve more accurate results when compared with using fixed values for prescribed dose conditions.

Deep learning architecture with shunted transformer and 3D deformable convolution for voxel-level dose prediction of head and neck tumors.

Intensity-modulated radiation therapy (IMRT) has been widely used in treating head and neck tumors. However, due to the complex anatomical structures in the head and neck region, it is challenging for the plan optimizer to rapidly generate clinically acceptable IMRT treatment plans. A novel deep learning multi-scale Transformer (MST) model was developed in the current study aiming to accelerate the IMRT planning for head and neck tumors while generating more precise prediction of the voxel-level dose distribution. The proposed end-to-end MST model employs the shunted Transformer to capture multi-scale features and learn a global dependency, and utilizes 3D deformable convolution bottleneck blocks to extract shape-aware feature and compensate the loss of spatial information in the patch merging layers. Moreover, data augmentation and self-knowledge distillation are used to further improve the prediction performance of the model. The MST model was trained and evaluated on the OpenKBP Challenge dataset. Its prediction accuracy was compared with three previous dose prediction models: C3D, TrDosePred, and TSNet. The predicted dose distributions of our proposed MST model in the tumor region are closest to the original clinical dose distribution. The MST model achieves the dose score of 2.23 Gy and the DVH score of 1.34 Gy on the test dataset, outperforming the other three models by 8%-17%. For clinical-related DVH dosimetric metrics, the prediction accuracy in terms of mean absolute error (MAE) is 2.04% for , 1.54% for , 1.87% for , 1.87% for , 1.89% for , respectively, superior to the other three models. The quantitative results demonstrated that the proposed MST model achieved more accurate voxel-level dose prediction than the previous models for head and neck tumors. The MST model has a great potential to be applied to other disease sites to further improve the quality and efficiency of radiotherapy planning.

Dose-volume metric-based prediction of radiotherapy-induced lymphocyte loss in patients with non-small-cell lung cancer treated with modern radiotherapy techniques

Background and PurposeRadiation-induced lymphopenia (RIL) is a common side effect of radiotherapy (RT) that may negatively impact survival. We aimed to identify RIL predictors in patients with non-small-cell lung cancer (NSCLC) treated intensity modulated radiotherapy (IMRT) and volumetric modulated arc therapy (VMAT). Materials and MethodsWe retrospectively analysed data of 306 patients who underwent radical RT for NSCLC. Absolute lymphocyte count (ALC) loss was evaluated for each patient by fitting an exponential decay curve to data from first 45 days since treatment start, and percentage ALC loss relative to baseline was calculated based on area under the decay curve and baseline ALC. We compared IMRT and VMAT treatment plans and used linear regression to predict ALC loss. ResultsALC decreased during RT in the whole patient group, while neutrophil counts remained stable and decreased only in those treated with concurrent chemoradiotherapy (CRT). Percentage ALC loss ranged between 11 and 78 % and was more strongly than lymphocyte nadir correlated with dose-volume metrics for relevant normal structures. We found evidence for the association of high radiation dose to the lungs, heart and body with percentage ALC loss, with lung volume exposed to 20–30 Gy being most important predictors in patients treated with IMRT. A multivariable model based on CRT use, baseline ALC and first principal component (PC1) of the dose-volume predictors showed good predictive performance (bias-corrected R2 of 0.40). ConclusionPercentage lymphocyte loss is a robust measure of RIL that is predicted by baseline ALC, CRT use and dose-volume parameters to the lungs, heart and body.

Automatic IMRT treatment planning through fluence prediction and plan fine-tuning for nasopharyngeal carcinoma

BackgroundAt present, the implementation of intensity-modulated radiation therapy (IMRT) treatment planning for geometrically complex nasopharyngeal carcinoma (NPC) through manual trial-and-error fashion presents challenges to the improvement of planning efficiency and the obtaining of high-consistency plan quality. This paper aims to propose an automatic IMRT plan generation method through fluence prediction and further plan fine-tuning for patients with NPC and evaluates the planning efficiency and plan quality.MethodsA total of 38 patients with NPC treated with nine-beam IMRT were enrolled in this study and automatically re-planned with the proposed method. A trained deep learning model was employed to generate static field fluence maps for each patient with 3D computed tomography images and structure contours as input. Automatic IMRT treatment planning was achieved by using its generated dose with slight tightening for further plan fine-tuning. Lastly, the plan quality was compared between automatic plans and clinical plans.ResultsThe average time for automatic plan generation was less than 4 min, including fluence maps prediction with a python script and automated plan tuning with a C# script. Compared with clinical plans, automatic plans showed better conformity and homogeneity for planning target volumes (PTVs) except for the conformity of PTV-1. Meanwhile, the dosimetric metrics for most organs at risk (OARs) were ameliorated in the automatic plan, especially Dmax of the brainstem and spinal cord, and Dmean of the left and right parotid glands significantly decreased (P < 0.05).ConclusionWe have successfully implemented an automatic IMRT plan generation method for patients with NPC. This method shows high planning efficiency and comparable or superior plan quality than clinical plans. The qualitative results before and after the plan fine-tuning indicates that further optimization using dose objectives generated by predicted fluence maps is crucial to obtain high-quality automatic plans.

Dosimetric comparison of 3D-Conformal and IMRT techniques used in radiotherapy of gastric cancer: A retrospective study

This study aimed to compare three-dimensional conformal radiotherapy (3D-CRT) and intensity-modulated radiotherapy (IMRT) planning techniques commonly used in gastric cancer radiotherapy with dose volume histograms. Computed tomography (CT) images of 20 gastric cancer patients were retrospectively analyzed. 3D-CRT, 5F- IMRT and 7F- IMRT treatment plans were created for each patient. The 3 plans were compared on dose volume histogram (DVH). 3D-CRT, 5F-IMRT and 7F-IMRT plans achieved a prescribed dose of 45 Gy for 95% of PTV volume. D95 values were 45.06±0.47 Gy for 3DCRT, 46.39±0.38 Gy for 5F-IMRT and 45.20±0.11 Gy for 7F-IMRT. In 3D-CRT, 5F-IMRT and 7F-IMRT techniques, the 13 Gy receiving volumes of the right kidney were found to be 35.08+9.59, 36.25+7.97 and 37.03+9.03 respectively. Moreover, the volume of the right kidney receiving a 20 Gy and 30 Gy dose received less dose with 5F-IMRT than with 7F-IMRT. Since each patient's critical organs are at different distances from the target and each technique has its own advantages in terms of critical organs, we suggest that the useful technique should be decided by clinical consensus.

Evaluation of gross target volumes in CECT vs MRI in head and neck cancer and its implication on concordance indices and dose-volume parameters of IMRT treatment plan

ABSTRACT Background: Although radiotherapy treatment planning (RTP) for head and neck cancers (HNCs) is based on contrast enhanced computed tomography (CECT), soft tissue contrasts are better evident on magnetic resonance imaging (MRI). We therefore evaluated dose-volume histogram (DVH) parameters along with concordance index (ConI), conformity index (CI), and homogeneity index (HI) of planning target volume (PTV) of GTV delineated on CECT vs MRI in HNCs enrolled for intensity modulated radiotherapy (IMRT). Methodology: Forty consecutive HNCs were enrolled in this study. All underwent CECT and MRI simulations with immobilization devices. GTVp and GTVn were delineated independently on CECT and MRI images. Corresponding MRI volumes were then copied to CECT. IMRT plans were generated on the CECT incorporating PTV margins. DVH parameters of PTVpn for both CECT and MRI were compared. In addition, mean (±SD) percentage difference of GTVp, GTVn, GTVpn, ConI, CI, and HI were evaluated using paired t-test. Results: The GTVp (P = 0.005), GTVn (P = 0.009), and GTVpn (P &lt; 0.001) delineated on MRI were found to be significantly larger than GTV delineated on CECT. In eight patients, GTV outlined on CECT were larger than MRI. Significant mean differences in CECT vs MRI of CI (P &lt; 0.001), HI (P &lt; 0.001), ConI (P &lt; 0.001), and DVH parameters (D2, D95, D98, V95, and V100 all P &lt; 0.001; D50: P = 0.009) were noted. Conclusion: The GTVs and corresponding PTVs were significantly larger on MRI compared to CECT. This resulted in significant differences in DVH parameters, CI, ConI, and HI. This could be improved by co-registered MRI-CECT volumes during routine IMRT treatment planning for HNCs.

A dosimetric evaluation of intensity modulated radiotherapy and three-dimensional conformal radiotherapy for prostate cancer in Ghana.

External beam radiotherapy incorporates treatment techniques such as three-dimensional conformal radiotherapy (3DCRT), intensity-modulated radiotherapy (IMRT), image-guided radiotherapy and volumetric modulated arc therapy to deliver high-energy radiation to cancer. The use of IMRT for cancer treatment is also associated with significant costs for patients in low-middle-income countries. The purpose of this study was to compare the dosimetric properties of 3DCRT and IMRT treatment plans for the external beam irradiation of patients with prostate cancer (Pca) to ascertain the superiority of IMRT in terms of dose homogeneity, conformity and dose limitation to organs at risk (OAR) in a resource-limited setting. One hundred and sixty treatment plans for 80 patients were created using 3DCRT and IMRT on the Eclipse treatment planning system (version 13.6). Data were collected and assessed from the dose-volume histogram of each plan. The conformity and homogeneity index (HI) for each of the plans were calculated. The doses to the OAR were also recorded and evaluated. The mean HIs for the IMRT and 3DCRT treatment techniques were 0.04 ± 0.02 (range: 0.01-0.011) and 0.09 ± 0.02 (range: 0.04-0.016), respectively. The mean conformity index (CI) for IMRT and 3DCRT techniques were 1.257 ± 0.112 (range: 0.99-1.58) and 1.302 ± 0.196 (range: 1.10-2.26). IMRT had a better significant mean HI and CI compared to 3DCRT. Generally, for this study, IMRT had better organ sparing compared to 3DCRT. The mean doses for the OARs ranged from 4.3-74.6 Gy for IMRT and 3.1-75.9 Gy for the 3DCRT technique. Overall, this study demonstrates that IMRT may offer an enhanced therapeutic profile, potentially reducing toxicity to the patient and ensuring more precise dose delivery to the target volume compared to 3DCRT in PCa external beam irradiation.

The multiple-set split feasibility problem and its application in intensity-modulated radiation therapy

In this paper, we apply a parallel iterative method for solving the split common fixed point problem in Hilbert spaces, without prior knowledge of the transfer operator norm, to the multiple-set split feasibility problem, which plays an important role in medical image reconstruction and signal processing. By formulating a model of the intensity-modulated radiation therapy treatment in a mathematical framework based on the multiple-set split feasibility problem, we also offer an effective approach for solving the intensity-modulated radiation therapy treatment planning.

A deep learning-based dose prediction method for evaluation of radiotherapy treatment planning

[Purpose]To address the issue of low accuracy in dose distribution prediction in radiotherapy, we propose a deep learning-based model for predicting three-dimensional dose distribution in tumor radiation therapy. The model utilizes quantitative evaluation methods to assess the treatment plans. [Methods]We selected a dataset of 130 cervical cancer patients, including CT images and target region files. A deep learning U-Net model based on convolutional neural networks and residual blocks was employed to automatically extract multi-scale and multi-level feature maps of CT images, target regions, and anatomical structures of critical organs for intensity-modulated radiation therapy (IMRT) treatment plans and perform three-dimensional dose distribution prediction. Quantitative analysis methods, including error measures such as maximum dose (Dmax), mean dose (Dmean), V20, and D95, were used. [Results]For cervical cancer cases, DVH (dose-volume histogram) graphs were generated based on the evaluation results, directly reflecting the differences between the actual and predicted doses. The actual errors met the basic requirements, and a quantitative evaluation approach was used to optimize the dosimetric parameters. The specific quantification results are: DSC: 86.52 ± 9.31, 95% HD: 3.74 ± 1.49, JD: 0.112 ± 0.026, MSD: 0.067 ± 0.031. [Conclusion]Through training the deep learning model, we have successfully captured the complex nonlinear relationship between IMRT plan feature map parameters and three-dimensional dose distribution. In practical clinical applications, this trained model can accurately predict personalized three-dimensional dose distribution for new patients and effectively assess treatment plans in a quantitative manner. The source code is available at: https://github.com/xiebw9509/Radiotherapy_dose_prediction.

Treatment planning comparison for head and neck cancer between photon, proton, and combined proton–photon therapy – From a fixed beam line to an arc

Background and purposeThis study investigates whether combined proton–photon therapy (CPPT) improves treatment plan quality compared to single-modality intensity-modulated radiation therapy (IMRT) or intensity-modulated proton therapy (IMPT) for head and neck cancer (HNC) patients. Different proton beam arrangements for CPPT and IMPT are compared, which could be of specific interest concerning potential future upright-positioned treatments. Furthermore, it is evaluated if CPPT benefits remain under inter-fractional anatomical changes for HNC treatments. Material and methodsFive HNC patients with a planning CT and multiple (4–7) repeated CTs were studied. CPPT with simultaneously optimized photon and proton fluence, single-modality IMPT, and IMRT treatment plans were optimized on the planning CT and then recalculated and reoptimized on each repeated CT. For CPPT and IMPT, plans with different degrees of freedom for the proton beams were optimized. Fixed horizontal proton beam line (FHB), gantry-like, and arc-like plans were compared. ResultsThe target coverage for CPPT without adaptation is insufficient (average V95%=88.4 %), while adapted plans can recover the initial treatment plan quality for target (average V95%=95.5 %) and organs-at-risk. CPPT with increased proton beam flexibility increases plan quality and reduces normal tissue complication probability of Xerostomia and Dysphagia. On average, Xerostomia NTCP reductions compared to IMRT are −2.7 %/-3.4 %/-5.0 % for CPPT FHB/CPPT Gantry/CPPT Arc. The differences for IMPT FHB/IMPT Gantry/IMPT Arc are + 0.8 %/-0.9 %/-4.3 %. ConclusionCPPT for HNC needs adaptive treatments. Increasing proton beam flexibility in CPPT, either by using a gantry or an upright-positioned patient, improves treatment plan quality. However, the photon component is substantially reduced, therefore, the balance between improved plan quality and costs must be further determined.

Generalisation of radiotherapy dose calculation for Monte Carlo algorithm combined with 3D Swin-Unet: a multi-institutional IMRT evaluation

Objective. Accurate dose calculations are essential prerequisites for precise radiotherapy. The integration of deep learning into dosimetry could consider computational accuracy and efficiency and has potential applicability to clinical dose calculation. The generalisation of a deep learning dose calculation method (hereinafter referred to as TERMA-Monte Carlo network, T-MC net) was evaluated in clinical practice using intensity-modulated radiotherapy (IMRT) plans for various human body regions and multiple institutions, with the Monte Carlo (MC) algorithm serving as a benchmark. Approach. Sixty IMRT plans were selected from four institutions for testing the head and neck, chest and abdomen, and pelvis regions. Using the MC results as the benchmark, the T-MC net calculation results were used to perform three-dimensional dose distribution and dose-volume histogram (DVH) comparisons of the entire body, planning target volume (PTV) and organs at risk (OARs), respectively, and calculate the mean ±95% confidence interval of gamma pass rate (GPR), percentage of agreement (PA) and dose difference ratio (DDR) of dose indices D95, D50, and D5. Main results. For the entire body, the GPRs of 3%/3 mm, 2%/2 mm, 2%/1 mm, and the PA were 99.62 ± 0.32%, 98.50 ± 1.09%, 95.60 ± 2.90% and 97.80 ± 1.12%, respectively. For the PTV, the GPRs of 3%/3 mm, 2%/2 mm, 2%/1 mm and the PA were 98.90 ± 1.00%, 95.78 ± 2.83%, 92.23 ± 4.74% and 98.93 ± 0.62%, respectively. The absolute value of average DDR was less than 1.4%. Significance. We proposed a general dose calculation framework based on deep learning, using the MC algorithm as a benchmark, performing a generalisation test for IMRT treatment plans across multiple institutions. The framework provides high computational speed while maintaining the accuracy of MC and may become an effective dose algorithm engine in treatment planning, adaptive radiotherapy, and dose verification.

Superiorization of projection algorithms for linearly constrained inverse radiotherapy treatment planning.

We apply the superiorization methodology to the constrained intensity-modulated radiation therapy (IMRT) treatment planning problem. Superiorization combines a feasibility-seeking projection algorithm with objective function reduction: The underlying projection algorithm is perturbed with gradient descent steps to steer the algorithm towards a solution with a lower objective function value compared to one obtained solely through feasibility-seeking. Within the open-source inverse planning toolkit matRad, we implement a prototypical algorithmic framework for superiorization using the well-established Agmon, Motzkin, and Schoenberg (AMS) feasibility-seeking projection algorithm and common nonlinear dose optimization objective functions. Based on this prototype, we apply superiorization to intensity-modulated radiation therapy treatment planning and compare it with (i) bare feasibility-seeking (i.e., without any objective function) and (ii) nonlinear constrained optimization using first-order derivatives. For these comparisons, we use the TG119 water phantom, the head-and-neck and the prostate patient of the CORT dataset. Bare feasibility-seeking with AMS confirms previous studies, showing it can find solutions that are nearly equivalent to those found by the established piece-wise least-squares optimization approach. The superiorization prototype solved the linearly constrained planning problem with similar dosimetric performance to that of a general-purpose nonlinear constrained optimizer while showing smooth convergence in both constraint proximity and objective function reduction. Superiorization is a useful alternative to constrained optimization in radiotherapy inverse treatment planning. Future extensions with other approaches to feasibility-seeking, e.g., with dose-volume constraints and more sophisticated perturbations, may unlock its full potential for high performant inverse treatment planning.

Novel hybrid treatment planning approach for irradiation a pediatric craniospinal axis

This study presents a new treatment planning approach merging 3D-CRT and VMAT fields into a hybrid treatment plan (HybTP), in order to achieve an optimum dose coverage of the planning target volume (PTV) and protection of OAR. Craniospinal axis irradiation (CSI) treated with 3D conformal radiotherapy (3D-CRT) is associated with high doses to the heart and eye lenses but provides better sparing of lungs and kidneys compared to volumetric modulated arc therapy (VMAT). VMAT treatment spares eye lenses and the heart, but lungs and kidneys are not as effective as 3D-CRT. Thus, a combination of both techniques (HybTP) may be optimal in sparing all these organs at risk (OAR). The results of HybTP are compared with helical tomotherapy (HT), intensity modulated radio therapy (IMRT), VMAT, and 3D-CRT plans. Hybrid, HT, VMAT, IMRT, and 3D-CRT treatment plans for a male child (age 6 years) with medulloblastoma were created and compared. A total dose of 35.2 Gy (PTV) with a dose per fraction of 1.6 Gy was prescribed. The following dose acceptance criteria were defined:(1)Mean PTV dose should be 100% of the prescribed dose, while the maximum dose should not exceed 107%.(2)At least 98% of the PTV should receive 95% of the prescribed dose.(3)The cribriform plate should be covered by the 95% isodose line.(4)The acceptance criteria for the OARs were: lenses Dmax<10 Gy, lungs Dmean<7 Gy, kidneys Dmean<15 Gy, heart Dmean<26 Gy, and heart V25<10%.The plans were compared regarding dose homogeneity index (HI) and conformity index (CI), PTV coverage, (particularly at cribriform plate) and doses at OARs. Best conformity was achieved with HT (CI = 0.98) followed by VMAT (CI = 0.96), IMRT (CI = 0.91), HybTP (CI = 0.86), and 3D-CRT (CI = 0.83). The homogeneity index varied marginally. For both HT and IMRT the HI was 0.07, and for 3D-CRT, VMAT and HybTP the HI was between 0.13 and 0.15. The cribriform plate was sufficiently covered by HybTP, VMAT, and 3D-CRT. The dose acceptance criteria for OARs were met by HT and HybTP. VMAT did not meet the criteria for lung (Dmean = right 10.4 Gy/left 10.2 Gy), 3D-CRT did not meet the criteria for eye lenses (Dmax = right 32.3 Gy/left 33.1), and heart (V25≈44%) and IMRT did not meet the criteria for lung (Dmean = right 11.1 Gy/left 11.2 Gy) and eye lenses (Dmax = right 12.2 Gy/left 13.1). HybTP meets all defined acceptance criteria and has proved to be a reasonable alternative for CSI. With HybTP that combines VMAT at the brain and heart with 3D-CRT posterior spinal fields (to spare lungs and kidneys), both appropriate coverage of the PTV and sparing of OAR can be achieved.

Automatic dose prediction using deep learning and plan optimization with finite-element control for intensity modulated radiation therapy.

Automatic solutions for generating radiotherapy treatment plans using deep learning (DL) have been investigated by mimicking the voxel's dose. However, plan optimization using voxel-dose features has not been extensively studied. This study aims to investigate the efficiency of a direct optimization strategy with finite elements (FEs) after DL dose prediction for automatic intensity-modulated radiation therapy (IMRT) treatment planning. A double-UNet DL model was adapted for 220 cervical cancer patients (200 for training and 20 for testing), who underwent IMRT between 2016 and 2020 at our clinic. The model inputs were computed tomography (CT) slices, organs at risk (OARs), and planning target volumes (PTVs), and the outputs were dose distributions of uniformly generated high-dose region-controlled plans. The FEs were discretized into equal intervals of the dose prediction value within the [OARs avoid PTV(O-P)] and [body avoids OARs & PTV(B-OP)] regions in the test cohort and used to define the objectives for IMRT plan optimization. The plans were optimized using a two-step process. In the beginning, the plans of two extra cases with and without low-dose region control were compared to pursue robust and optimal dose adjustment degree pattern of FEs. In the first step, the mean dose of O-P FEs were constrained to differing degrees according to the pattern. The further the FEs from the PTV, the tighter the constraints. In the second step, the mean dose of O-P FEs from first step were constrained again but weakly and the dose of the B-OP FEs from dose prediction and PTV were tightly regulated. The dosimetric parameters of the OARs and PTV were evaluated and compared using an interstep approach. In another 10 cases, the plans optimized via the aforementioned steps (method 1) were compared with those directly generated by the double-UNet dose prediction model trained by low and high region-controlled plans (method 2). The mean differences in dose metrics between the UNet-predicted dose and the clinical plans were: 0.47Gy for bladder D50% ; 0.62Gy for rectum D50% ; 0% for small intestine V30Gy ; 1% for small intestine V40Gy ; 4% for left femoral head V30Gy ; and 6% for right femoral head V30Gy . The reductions in mean dose (p<0.001) after FE-based optimization were: 4.0, 1.9, 2.8, 5.9, and 5.7Gy for the bladder, rectum, small intestine, left femoral head, and right femoral head, respectively, with flat PTV homogeneity and conformity. Method 1 plans produced lower mean doses than those of method 2 for the bladder (0.7Gy), rectum (1.0Gy), and small intestine (0.6Gy), while maintaining PTV homogeneity and conformity. FE-based direct optimization produced lower OAR doses and adequate PTV doses after DL prediction. This solution offers rapid and automatic plan optimization without manual adjustment, particularly in low-dose regions.

Novel in-house knowledge-based automated planning system for lung cancer treated with intensity-modulated radiotherapy.

The goal of this study was to propose aknowledge-based planning system which could automatically design plans for lung cancer patients treated with intensity-modulated radiotherapy (IMRT). From May 2018 to June 2020, 612 IMRT treatment plans of lung cancer patients were retrospectively selected to construct aplanning database. Knowledge-based planning (KBP) architecture named αDiar was proposed in this study. It consisted of two parts separated by afirewall. One was the in-hospital workstation, and the other was the search engine in the cloud. Based on our previous study, A‑Net in the in-hospital workstation was used to generate predicted virtual dose images. Asearch engine including athree-dimensional convolutional neural network (3D CNN) was constructed to derive the feature vectors of dose images. By comparing the similarity of the features between virtual dose images and the clinical dose images in the database, the most similar feature was found. The optimization parameters (OPs) of the treatment plan corresponding to the most similar feature were assigned to the new plan, and the design of anew treatment plan was automatically completed. After αDiar was developed, we performed two studies. The first retrospective study was conducted to validate whether this architecture was qualified for clinical practice and involved 96patients. The second comparative study was performed to investigate whether αDiar could assist dosimetrists in improving the quality of planning for the patients. Two dosimetrists were involved and designed plans for only one trial with and without αDiar; 26patients were involved in this study. The first study showed that about 54% (52/96) of the automatically generated plans would achieve the dosimetric constraints of the Radiation Therapy Oncology Group (RTOG) and about 93% (89/96) of the automatically generated plans would achieve the dosimetric constraints of the National Comprehensive Cancer Network (NCCN). The second study showed that the quality of treatment planning designed by junior dosimetrists was improved with the help of αDiar. Our results showed that αDiar was an effective tool to improve planning quality. Over half of the patients' plans could be designed automatically. For the remaining patients, although the automatically designed plans did not fully meet the clinical requirements, their quality was also better than that of manual plans.

A dosimetric comparative study following RTOG and ESTRO contouring guidelines for breast radiation therapy

PurposeTo compare the dosimetric parameters considering the Radiation Therapy Oncology Group (RTOG) and European Society for Radiotherapy and Oncology (ESTRO) guidelines for breast cancer radiotherapy. Two radiotherapy techniques, intensity-modulated radiation therapy (IMRT) and volumetric-modulated arc therapy (VMAT), were considered. Patients and methodsTwenty-eight patients with left-sided medially-located TanyN2M0 tumors were contoured based on RTOG and ESTRO guidelines. 9-field IMRT, 10-field IMRT, 11-field IMRT, and VMAT treatment plans were applied as radiotherapy (RT) techniques for both contouring sets. The dosimetric parameters of the RT plans were extracted and compared. ResultsComparing dose-volume histogram (DVH) parameters, equivalent uniform dose (EUD), and normal tissue complication probability (NTCP) of OARs across the contouring guidelines and considering each RT technique showed that the only significant differences were higher Dmax, Dmean, V30, and V45, EUD, and NTCP of the thyroid in all treatment modalities when the RTOG guideline had been adopted. Using the VMAT technique, PTV's EUD and the tumor control probability (TCP) were considerably higher when the ESTRO guideline was adopted. Moreover, the conformity index (CI) of VMAT plans was significantly higher when the ESTRO guideline was used. ConclusionUnless having higher doses to thyroid when the RTOG guideline was adopted, the doses to other organs-at-risk (OAR) were similar between the two considering guidelines. Moreover, except for higher EUD, TCP, and CI for VMAT when the ESTRO guideline was used, no other significant differences were obtained between dosimetric parameters of target volumes considering the RT techniques and contouring guidelines.

Simultaneous dose distribution and fluence prediction for nasopharyngeal carcinoma IMRT

BackgroundCurrent intensity-modulated radiation therapy (IMRT) treatment planning is still a manual and time/resource consuming task, knowledge-based planning methods with appropriate predictions have been shown to enhance the plan quality consistency and improve planning efficiency. This study aims to develop a novel prediction framework to simultaneously predict dose distribution and fluence for nasopharyngeal carcinoma treated with IMRT, the predicted dose information and fluence can be used as the dose objectives and initial solution for an automatic IMRT plan optimization scheme, respectively.MethodsWe proposed a shared encoder network to simultaneously generate dose distribution and fluence maps. The same inputs (three-dimensional contours and CT images) were used for both dose distribution and fluence prediction. The model was trained with datasets of 340 nasopharyngeal carcinoma patients (260 cases for training, 40 cases for validation, 40 cases for testing) treated with nine-beam IMRT. The predicted fluence was then imported back to treatment planning system to generate the final deliverable plan. Predicted fluence accuracy was quantitatively evaluated within projected planning target volumes in beams-eye-view with 5 mm margin. The comparison between predicted doses, predicted fluence generated doses and ground truth doses were also conducted inside patient body.ResultsThe proposed network successfully predicted similar dose distribution and fluence maps compared with ground truth. The quantitative evaluation showed that the pixel-based mean absolute error between predicted fluence and ground truth fluence was 0.53% ± 0.13%. The structural similarity index also showed high fluence similarity with values of 0.96 ± 0.02. Meanwhile, the difference in the clinical dose indices for most structures between predicted dose, predicted fluence generated dose and ground truth dose were less than 1 Gy. As a comparison, the predicted dose achieved better target dose coverage and dose hot spot than predicted fluence generated dose compared with ground truth dose.ConclusionWe proposed an approach to predict 3D dose distribution and fluence maps simultaneously for nasopharyngeal carcinoma patients. Hence, the proposed method can be potentially integrated in a fast automatic plan generation scheme by using predicted dose as dose objectives and predicted fluence as a warm start.

Detection of suboptimal IMRT treatment plan using machine learning on radiomics features of dose distribution for lung cancers

Quality assurance in radiotherapy is an important process so that the use of radiation provides maximum benefits. Currently, the implementation of machine learning (ML) in the quality assurance of treatment planning is growing. In this study, 34 optimal intensity-modulated radiation therapy (IMRT) treatment plans and 10 suboptimal IMRT treatment plans obtained from Siloam MRCCC Semanggi Hospital were used to train an ML model of the type autoencoder for anomaly detection developed using PyTorch. There were four stages in this study, namely the preparation stage, development stage, validation stage, and evaluation stage. At the development stage, the raw data was prepared so that it is ready to be used for training. At the development stage, the model was developed and a hyperparameters optimization was performed. The accuracy of the model was analyzed at the validation stage. Finally, at the evaluation stage, the model performance was evaluated along with a Mann-Whitney U test on dose-volume histogram (DVH) parameters, radiomics features, and DVH metrics (conformity index and homogeneity index) to show the difference between treatment plans. The model used 161 radiomics features with an epoch of 1250 iterations, 150-50-17 hidden layers configuration, and a learning rate of 0.2 as the most optimal configuration. The results showed an accuracy of 36% with 7% of radiomics features, 50% of DVH parameters, and the homogeneity index being different significantly. After refinement, that is removing data with conformity index below one, the accuracy became 50% with 12% of radiomics features, 45% of DVH parameters, and both DVH metrics being different significantly. If the radiomics features used are those that were significantly different, the accuracy increased to 93%. From these results, it can be concluded that most radiomics features are noise that need to be removed so the model can detect suboptimal treatment plans. In addition, the planning target volume segment along with the firstorder radiomics features group is the main differentiator between the optimal and suboptimal treatment plans.

Comparison of 3D-conformal and intensity-modulated radiation therapy for left-sided breast cancer

Background: We compared 3D-conformal radiation therapy (3D-CRT) and intensity-modulated radiation therapy (IMRT) planning for left-sided post-mastectomy patients. Aim: To compare the dose coverage of the planning target volume (PTV) and dose delivered to organs at risk (OAR) of 3D-CRT and IMRT plans. Setting: Department of Oncology, central South Africa. Methods: Twenty-six archived CT scans of patients with left-sided breast cancer were included. The 3D-CRT and IMRT plans were designed for each patient and compared using the Monaco © planning system (version 5.11.02). Statistical analysis was performed for PTV coverage (V 95% , V 98% , V 105% ) and radiation doses to the heart, ipsilateral lung, combined lungs, contralateral breast, and oesophagus. Results: The V 98% and V 105% target volume dose coverage for the 3D-CRT plans were 67.07% and 0.21%, respectively, compared to 92.32% and 1.10% of the IMRT plans. However, the IMRT plans’ mean volume of PTV, receiving 95% of the prescribed dose (PD), was 7.68% compared to the 3D-CRT’s 32.93%. The IMRT plans resulted in a V 22 Gy &lt; 10% for the heart, with a value of 4.15%. The V 18.87 Gy &lt; 45% values for the ipsilateral and combined lungs were 28.09% and 13.70%, respectively. The 3D-CRT plans showed a lower dose to the oesophagus (5.07 Gy) and contralateral breast (V 5 Gy &lt; 15% = 3.51%). Conclusion: It was shown that 3D-CRT and IMRT treatment planning can effectively achieve clinical goals for post-mastectomy left-sided breast cancer radiotherapy. Contribution: The findings underscore the continuing relevance of 3D-CRT planning in oncology for optimal PTV dose coverage and low OAR dose.

Advancing Precision in Post-mastectomy Chest Wall Radiotherapy: A Comparative Dosimetric Analysis of Volumetric-Modulated Arc Therapy (VMAT) and Intensity-Modulated Radiotherapy (IMRT) Based on Institutional Experience.

Post-mastectomy radiation therapy (PMRT) is an important component in the management of breast cancer patients who have undergone mastectomy. Intensity-modulated radiation therapy (IMRT) and volumetric-modulated arc therapy (VMAT) are two popular methods of delivering PMRT. With IMRT, high radiation doses are directed at the tumor, while exposure to healthy tissue is kept to a minimum. VMAT, on the other hand, is a more advanced version of IMRT that allows for faster radiation dose delivery while maintaining precision. The complexity of the VMAT treatment planning and delivery process, on the other hand, may increase the risk of technical errors, which can reduce treatment effectiveness. Studies have compared VMAT and IMRT in PMRT for breast cancer patients, but most have found no significant differences in treatment outcomes between the two methods. Individual patient factors such as treatment goals, available resources, and other characteristics may influence the choice between the two techniques. This prospective observational study aimed to compare the dosimetry of two cutting-edge modern radiotherapy techniques for post-mastectomy breast cancer patients receiving hypofractionated doses. For 58 patients with breast cancer, 116 plans for radiotherapy treatment were generated by both VMAT and IMRT. To maintain the uniformity of contouring, every CT image was contoured by the same physician, and Radiotherapy Oncology Group (RTOG) contouring guidelines were strictly followed during contouring. Both techniques had comparable target volume coverage, but VMAT produced a significantly better conformity index than IMRT for both the left (0.71 vs. 0.65) and right (0.72 vs. 0.66) breasts (p-value < 0.05). VMAT plans had significantly higher low-dose spillage to the ipsilateral lung (V5Gy and V10Gy) but significantly lower high-dose spillage (V20Gy, V30Gy, and V40Gy) than IMRT plans (p-value < 0.05). Dmax and Dmean for the ipsilateral lung were comparable for both techniques. When compared to alternative treatment approaches, IMRT treatment plans were found to be more effective in minimizing radiation exposure to the heart for all patients with right-sided breast cancer, resulting in considerably lower levels of Dmean, V5Gy, V10Gy, V20Gy, and V35Gy. Plans for VMAT treatment were found to be significantly superior to left-side chest wall radiotherapy in terms of lower exposure to the heart for higher doses. IMRT plans, on the other hand, were successful in dramatically lowering the levels of Dmax that reached the spinal cord for both right- and left-sided breast cancers. Apart from similar planning target volume (PTV) coverage to IMRT plans, VMAT produced significantly better conformity. VMAT plans have more low-dose spillage to normal tissues, while IMRT plans spare various organs at risk significantly better at lower doses in both right and left-sided breast cancer. VMAT was found to be better at sparing the heart (in left-sided breast cancer only) and ipsilateral lung at a high dose range. The best radiotherapy approach for breast cancer should be established on an individual basis, taking into account tumor laterality and the risk-benefit ratio.