OpenKBP: The open-access knowledge-based planning grand challenge and dataset

Rapid dose prediction for lung CyberKnife radiotherapy plans utilizing a deep learning approach by incorporating dosimetric features delivered by noncoplanar beams

Purpose. The dose distribution of lung cancer patients treated with the CyberKnife (CK) system is influenced by various factors, including tumor location and the direction of CK beams. The objective of this study is to present a deep learning approach that integrates CK beam dose characteristics into CK planning dose calculations.Methods. The inputs utilized for the geometry and dosimetry method (GDM) include the patient's CT, the PTV structure, and multiple CK noncoplanar beam dose deposition features. The dose distributions were calculated using the Monte Carlo (MC) algorithm provided with the CK system and served as the ground truth dose label. Additionally, dose prediction was conducted through the geometry method (GM) for comparative analysis. The gamma pass rateγ(1 mm,1%),γ(2 mm,2%) andγ(3 mm,3%) were calculated between the predicted model and the MC method.Results. Compared to the GDM, the GM shows a significant dose difference from the MC approach in the low-dose region (<5 Gy) outside the target created by the various CK noncoplanar beams. The GDM increased theγ(1 mm, 1%) from 49.55% to 81.69%,γ(2 mm, 2%) from 73.24% to 98.11% and theγ(3 mm, 3%) from 81.69% to 99.37% when compared with the GM's results.Conclusions. This work proposed a deep learning dose calculation method by using patient geometry and dosimetry features in CK plans. The proposed method extends the geometric and dosimetric feature-driven deep learning dose calculation method to CK application scenarios, which has a great potential to accelerate the CK planning dose calculation and improve the planning efficiency.

Deep learning-based prediction of Monte Carlo dose distribution for heavy ion therapy

Background and purpose

Current methods, like treatment planning system algorithms (TPSDose), lack accuracy, whereas Monte Carlo dose distribution (MCDose) is accurate but computationally intensive. We proposed a deep learning (DL) model for rapid prediction of Monte Carlo simulated dose distribution (MCDose) in heavy ion therapy (HIT).

Materials and methods

We developed a DL model - the Cascade Hierarchically Densely 3D U-Net (CHD U-Net) - to predict MCDose using computed tomography images and TPSDose of 67 head-and-neck patients and 30 thorax-and-abdomen patients. We also compared the results with other proton dose DL models and TPSDose.

Results

Compared to TPSDose, the gamma passing rate (GPR) improved by 16 % (1 %/1 mm). Notably, the model achieved 99 % and 97 % accuracy under clinically relevant criteria (3 %/3 mm) across the whole dose distribution in patients. For head-and-neck patients, the GPRs of the C3D and HD U-Net models in the PTV region were 97 % and 85 %, and in the body were 98 % and 97 %, respectively. For thorax-and-abdomen patients, the GPR of the C3D and HD U-Net models in the PTV region were 71 % and 51 %, and in the body were 95 % and 90 %, respectively.

Conclusions

The proposed CHD U-Net model can predict MCDose in a few seconds and outperforms two alternative DL models. The predicted dose can replace TPSDose in HIT clinical process due to its MC simulation accuracy, thus improving the accuracy of dose calculation and providing a valuable reference for quality assurance.

Deep learning techniques for proton dose prediction across multiple anatomical sites and variable beam configurations

Objective.To evaluate the impact of beam mask implementation and data aggregation on artificial intelligence-based dose prediction accuracy in proton therapy, with a focus on scenarios involving limited or highly heterogeneous datasets.Approach.In this study, 541 prostate and 632 head and neck (H&N) proton therapy plans were used to train and evaluate convolutional neural networks designed for the task of dose prediction. Datasets were grouped by anatomical site and beam configuration to assess the impact of beam masks-graphical depictions of radiation paths-as a model input. We also evaluated the effect of combining datasets. Model performance was measured using dose-volume histograms (DVHs) scores, mean absolute error, mean absolute percent error, dice similarity coefficients (DSCs), and gamma passing rates.Main results.DSC analysis revealed that the inclusion of beam masks improved dose prediction accuracy, particularly in low-dose regions and for datasets with diverse beam configurations. Data aggregation alone produced mixed results, with improvements in high-dose regions but potential degradation in low-dose areas. Notably, combining beam masks and data aggregation yielded the best overall performance, effectively leveraging the strengths of both strategies. Additionally, the magnitude of the improvements was larger for datasets with greater heterogeneity, with the combined approach increasing the DSC score by as much as 0.2 for a subgroup of H&N cases characterized by small size and heterogeneity in beam arrangement. DVH scores reflected these benefits, showing statistically significant improvements (p< 0.05) for the more heterogeneous H&N datasets.Significance.Artificial intelligence-based dose prediction models incorporating beam masks and data aggregation significantly improve accuracy in proton therapy planning, especially for complex cases. This technique could accelerate the planning process, enabling more efficient and effective cancer treatment strategies.

A deep learning model to predict dose distributions for breast cancer radiotherapy

PURPOSE

In this work, we propose to develop a 3D U-Net-based deep learning model that accurately predicts the dose distribution for breast cancer radiotherapy.

METHODS

This study included 176 breast cancer patients, divided into training, validating and testing sets. A deep learning model based on the 3D U-Net architecture was developed to predict dose distribution, which employed a double encoder combination attention (DECA) module, a cross stage partial + Resnet + Attention (CRA) module, a difficulty perception and a critical regions loss. The performance and generalization ability of this model were evaluated by the voxel mean absolute error (MAE), several clinically relevant dosimetric indexes and 3D gamma passing rates.

RESULTS

Our model accurately predicted the 3D dose distributions with each dosage level mirroring the clinical reality in shape. The generated dose-volume histogram (DVH) matched with the ground truth curve. The total dose error of our model was below 1.16 Gy, complying with clinical usage standards. When compared to other exceptional models, our model optimally predicted eight out of nine regions, and the prediction errors for the first planning target volume (PTV1) and PTV2 were merely 1.03 Gy and 0.74 Gy. Moreover, the mean 3%/3 mm 3D gamma passing rates for PTV1, PTV2, Heart and Lung L achieved 91.8%, 96.4%, 91.5%, and 93.2%, respectively, surpassing the other models and meeting clinical standards.

CONCLUSIONS

This study developed a new deep learning model based on 3D U-Net that can accurately predict dose distributions for breast cancer radiotherapy, which can improve the quality and planning efficiency.

Deep learning dose prediction to approach Erasmus-iCycle dosimetric plan quality within seconds for instantaneous treatment planning

BACKGROUND AND PURPOSE

Fast, high-quality deep learning (DL) prediction of patient-specific 3D dose distributions can enable instantaneous treatment planning (IP), in which the treating physician can evaluate the dose and approve the plan immediately after contouring, rather than days later. This would greatly benefit clinical workload, patient waiting times and treatment quality. IP requires that predicted dose distributions closely match the ground truth. This study examines how training dataset size and model size affect dose prediction accuracy for Erasmus-iCycle GT plans to enable IP.

MATERIALS AND METHODS

For 1250 prostate patients, dose distributions were automatically generated using Erasmus-iCycle. Hierarchically Densely Connected U-Nets with 2/3/4/5/6 pooling layers were trained with datasets of 50/100/250/500/1000 patients, using a validation set of 100 patients. A fixed test set of 150 patients was used for evaluations.

RESULTS

For all model sizes, prediction accuracy increased with the number of training patients, without levelling off at 1000 patients. For 4-6 level models with 1000 training patients, prediction accuracies were high and comparable. For 6 levels and 1000 training patients, the median prediction errors and interquartile ranges for PTV V95%, rectum V75Gy and bladder V65Gy were 0.01 [-0.06,0.15], 0.01 [-0.20,0.29] and -0.02 [-0.27,0.27] %-point. Dose prediction times were around 1.2 s.

CONCLUSION

Although even for 1000 training patients there was no convergence in obtained prediction accuracy yet, the accuracy for the 6-level model with 1000 training patients may be adequate for the pursued instantaneous planning, which is subject of further research.

An Efficient 3D Convolutional Neural Network for Dose Prediction in Cancer Radiotherapy from CT Images

Introduction: Cancer is a highly lethal disease with a significantly high mortality rate. One of the most commonly used methods for treatment is radiation therapy. However, cancer treatment using radiotherapy is a time-consuming process that requires significant manual work from planners and doctors. In radiation therapy treatment planning, determining the dose distribution for each of the regions of the patient's body is one of the most difficult and important tasks. Nowadays, artificial intelligence has shown promising results in improving the quality of disease treatment, particularly in cancer radiation therapy. Objectives: The main objective of this study is to build a high-performance deep learning model for predicting radiation therapy doses for cancer and to develop software to easily manipulate and use this model. Materials and Methods: In this paper, we propose a custom 3D convolutional neural network model with a U-Net-based architecture to automatically predict radiation doses during cancer radiation therapy from CT images. To ensure that the predicted doses do not have negative values, which are not valid for radiation doses, a rectified linear unit (ReLU) function is applied to the output to convert negative values to zero. Additionally, a proposed loss function based on a dose-volume histogram is used to train the model, ensuring that the predicted dose concentrations are highly meaningful in terms of radiation therapy. The model is developed using the OpenKBP challenge dataset, which consists of 200, 100, and 40 head and neck cancer patients for training, testing, and validation, respectively. Before the training phase, preprocessing and augmentation techniques, such as standardization, translation, and flipping, are applied to the training set. During the training phase, a cosine annealing scheduler is applied to update the learning rate. Results and Conclusions: Our model achieved strong performance, with a good DVH score (1.444 Gy) on the test dataset, compared to previous studies and state-of-the-art models. In addition, we developed software to display the dose maps predicted by the proposed model for each 2D slice in order to facilitate usage and observation. These results may help doctors in treating cancer with radiation therapy in terms of both time and effectiveness.

A feasibility study of dose-band prediction in radiation therapy: Predicting a spectrum of plan dose

PURPOSE

The current deep learning-based dose prediction methods only predict one dose distribution. If the predicted dose is inaccurate, no additional options can be selected. To overcome this limitation, we propose a novel dose prediction method called "dose-band prediction," which provides a spectrum of predicted dose distributions for planning and quality assurance (QA) purposes.

MATERIAL AND METHODS

We utilized Upper/Lower-band losses in 3D neural networks to establish the Upper/Lower-band models (UBM/LBM). The maximum/minimum rational dose predicted in UBM/LBM defined the ideal dose spectrum for each voxel. We enrolled 104 nasopharyngeal carcinoma cases with tomotherapy (dataset 1), 54 cervical carcinoma cases with IMRT (dataset 2), and 37 cervical carcinoma cases with VMAT (dataset 3) in the study. Moreover, a dose band-based auto planning (Auto-plandose-band) attempt was carried out in dataset 3, compared with the MSE model (Auto-planMSE).

RESULTS

The UBM/LBM doses tend to be higher/lower than the clinical dose, forming a predicted dose spectrum. The Middle-line dose represents the average of the Upper/Lower-band, which was consistent with the clinical dose. The mean differences of the planning target volumes (PTVs) and organs at risk (OARs) for the Upper-band, Middle-line, and Lower-band in Dataset 1 were 3.66 %, -0.40 %, and -4.48 % in Dataset 2, they were 2.40 %, -1.62 %, and -5.57 %; in Dataset 3, they were 2.18 %, -0.59 %, and -3.31 %. When PTVs meet prescription, the mean difference between Auto-plandose-band and Auto-planMSE in OARs was -2.67 %.

CONCLUSION

The dose-band prediction successfully predicted a spectrum of doses, making auto-planning and QA flexible and high quality.

Knowledge-based planning, multicriteria optimization, and plan scorecards: A winning combination

BACKGROUND AND PURPOSE

The ESTRO 2023 Physics Workshop hosted the Fully-Automated Radiotherapy Treatment Planning (Auto-RTP) Challenge, where participants were provided with CT images from 16 prostate cancer patients (6 prostate only, 6 prostate + nodes, and 4 prostate bed + nodes) across 3 challenge phases with the goal of automatically generating treatment plans with minimal user intervention. Here, we present our team's winning approach developed to swiftly adapt to both different contouring guidelines and treatment prescriptions than those used in our clinic.

MATERIALS AND METHODS

Our planning pipeline comprises two main components: 1) auto-contouring and 2) auto-planning engines, both internally developed and activated via DICOM operations. The auto-contouring engine employs 3D U-Net models trained on a dataset of 600 prostate cancer patients for normal tissues, 253 cases for pelvic lymph node, and 32 cases for prostate bed. The auto-planning engine, utilizing the Eclipse Scripting Application Programming Interface, automates target volume definition, field geometry, planning parameters, optimization, and dose calculation. RapidPlan models, combined with multicriteria optimization and scorecards defined on challenge scoring criteria, were employed to ensure plans met challenge objectives. We report leaderboard scores (0-100, where 100 is a perfect score) which combine organ-at-risk and target dose-metrics on the provided cases.

RESULTS

Our team secured 1st place across all three challenge phases, achieving leaderboard scores of 79.9, 77.3, and 78.5 outperforming 2nd place scores by margins of 6.4, 0.4, and 2.9 points for each phase, respectively. Highest plan scores were for prostate only cases, with an average score exceeding 90. Upon challenge completion, a "Plan Only" phase was opened where organizers provided contours for planning. Our current score of 90.0 places us at the top of the "Plan Only" leaderboard.

CONCLUSIONS

Our automated pipeline demonstrates adaptability to diverse guidelines, indicating progress towards fully automated radiotherapy planning. Future studies are needed to assess the clinical acceptability and integration of automatically generated plans.

Ultra-fast, one-click radiotherapy treatment planning outside a treatment planning system

We present an automated radiation oncology treatment planning pipeline that operates between segmentation and plan review, minimizing manual interaction and reliance on traditional planning systems. Two AI models work in sequence: the first generates a dose distribution, and the second creates a deliverable DICOM-RT plan. Trained and validated on 276 plans, and tested on 151 datasets, the system produced clinically deliverable plans-complete with all VMAT parameters-in about 38 s. These plans met target coverage and most organ-at-risk constraints. This proof-of-concept demonstrates the feasibility of generating high-quality, deliverable DICOM plans within seconds.

Evaluation of deep learning based dose prediction in head and neck cancer patients using two different types of input contours

PURPOSE

This study evaluates deep learning (DL) based dose prediction methods in head and neck cancer (HNC) patients using two types of input contours.

MATERIALS AND METHODS

Seventy-five HNC patients undergoing two-step volumetric-modulated arc therapy were included. Dose prediction was performed using the AIVOT prototype (AiRato.Inc, Sendai, Japan), a commercial software with an HD U-net-based dose distribution prediction system. Models were developed for the initial plan (46 Gy/23Fr) and boost plan (24 Gy/12Fr), trained with 65 cases and tested with 10 cases. The 8-channel model used one target (PTV) and seven organs at risk (OARs), while the 10-channel model added two dummy contours (PTV ring and spinal cord PRV). Predicted and deliverable doses, obtained through dose mimicking on another radiation treatment planning system, were evaluated using dose-volume indices for PTV and OARs.

RESULTS

For the initial plan, both models achieved approximately 2% prediction accuracy for the target dose and maintained accuracy within 3.2 Gy for OARs. The 10-channel model outperformed the 8-channel model for certain dose indices. For the boost plan, both models exhibited prediction accuracies of approximately 2% for the target dose and 1 Gy for OARs. The 10-channel model showed significantly closer predictions to the ground truth for D50% and Dmean. Deliverable plans based on prediction doses showed little significant difference compared to the ground truth, especially for the boost plan.

CONCLUSION

DL-based dose prediction using the AIVOT prototype software in HNC patients yielded promising results. While additional contours may enhance prediction accuracy, their impact on dose mimicking is relatively small.

Automatic Treatment Planning for Radiation Therapy: A Cross-Modality and Protocol Study

Purpose

This study investigated the applicability of 3-dimensional dose predictions from a model trained on one modality to a cross-modality automated planning workflow. Additionally, we explore the impact of integrating a multicriteria optimizer (MCO) on adapting predictions to different clinical preferences.

Methods and Materials

Using a previously created 3-stage U-Net in-house model trained on the 2020 American Association of Physicists in Medicine OpenKBP challenge data set (340 head and neck plans, all planned using 9-field static intensity modulated radiation therapy [IMRT]), we retrospectively generated dose predictions for 20 patients. These dose predictions were, in turn, used to generate deliverable IMRT, VMAT, and tomotherapy plans using the fallback plan functionality in Raystation. The deliverable plans were evaluated against the dose predictions based on primary clinical goals. A new set of plans was also generated using MCO-based optimization with predicted dose values as constraints. Delivery QA was performed on a subset of the plans to assure clinical deliverability.

Results

The mimicking approach accurately replicated the predicted dose distributions across different modalities, with slight deviations in the spinal cord and external contour maximum doses. MCO optimization significantly reduced doses to organs at risk, which were prioritized by our institution while maintaining target coverage. All tested plans met clinical deliverability standards, evidenced by a gamma analysis passing rate >98%.

Conclusions

Our findings show that a model trained only on IMRT plans can effectively contribute to planning across various modalities. Additionally, integrating predictions as constraints in an MCO-based workflow, rather than direct dose mimicking, enables a flexible, warm-start approach for treatment planning, although the benefit is reduced when the training set differs significantly from an institution's preference. Together, these approaches have the potential to significantly decrease plan turnaround time and quality variance, both at high-resource medical centers that can train in-house models and smaller centers that can adapt a model from another institution with minimal effort.

Deep evidential learning for radiotherapy dose prediction

BACKGROUND

As we navigate towards integrating deep learning methods in the real clinic, a safety concern lies in whether and how the model can express its own uncertainty when making predictions. In this work, we present a novel application of an uncertainty-quantification framework called Deep Evidential Learning in the domain of radiotherapy dose prediction.

METHOD

Using medical images of the Open Knowledge-Based Planning Challenge dataset, we found that this model can be effectively harnessed to yield uncertainty estimates that inherited correlations with prediction errors upon completion of network training. This was achieved only after reformulating the original loss function for a stable implementation.

RESULTS

We found that (i) epistemic uncertainty was highly correlated with prediction errors, with various association indices comparable or stronger than those for Monte-Carlo Dropout and Deep Ensemble methods, (ii) the median error varied with uncertainty threshold much more linearly for epistemic uncertainty in Deep Evidential Learning relative to these other two conventional frameworks, indicative of a more uniformly calibrated sensitivity to model errors, (iii) relative to epistemic uncertainty, aleatoric uncertainty demonstrated a more significant shift in its distribution in response to Gaussian noise added to CT intensity, compatible with its interpretation as reflecting data noise.

CONCLUSION

Collectively, our results suggest that Deep Evidential Learning is a promising approach that can endow deep-learning models in radiotherapy dose prediction with statistical robustness. We have also demonstrated how this framework leads to uncertainty heatmaps that correlate strongly with model errors, and how it can be used to equip the predicted Dose-Volume-Histograms with confidence intervals.

Deep learning-based prediction of the dose-volume histograms for volumetric modulated arc therapy of left-sided breast cancer

BACKGROUND

The advancements in artificial intelligence and computational power have made deep learning an attractive tool for radiotherapy treatment planning. Deep learning has the potential to significantly simplify the trial-and-error process involved in inverse planning required by modern treatment techniques such as volumetric modulated arc therapy (VMAT). In this study, we explore the ability of deep learning to predict organ-at-risk (OAR) dose-volume histograms (DVHs) of left-sided breast cancer patients undergoing VMAT treatment based solely on their anatomical characteristics. The predicted DVHs could be used to derive patient-specific dose constraints and dose objectives, streamlining the treatment planning process, standardizing the quality of the plans, and personalizing the treatment planning.

PURPOSE

This study aimed to develop a deep learning-based framework for the prediction of organ-specific dose-volume histograms (DVH) based on structures delineated for left-sided breast cancer treatment.

METHODS

We used a dataset of 249 left-sided breast cancer patients treated with tangential VMAT fields. We extracted delineated structures and dose distributions for each patient and derived slice-by-slice DVHs for planning target volume (PTV) and organs-at-risk. The patients were divided into training (70%, n = 174), validation (10%, n = 24), and test (20%, n = 51) sets. Collected data were used to train a deep learning model for the prediction of the DVHs based on the delineated structures. The developed deep learning model comprised a modified DenseNet architecture followed by a recurrent neural network.

RESULTS

In the independent test set (n = 51), the point-wise differences in the slice-by-slice DVHs between the clinical and predicted DVHs were small; the mean squared errors were 3.53, 1.58, 2.28, 3.37, and 1.44 [×10-4] for PTV, heart, ipsilateral lung, contralateral lung, and contralateral breast, respectively. With the derived cumulative DVHs, the mean absolute difference ± standard deviation of mean doses between the clinical and the predicted DVH were 0.08 ± 0.04 Gy, 0.24 ± 0.22 Gy, 0.73 ± 0.46 Gy, 0.07 ± 0.06 Gy, and 0.14 ± 0.14 Gy for PTV, heart, ipsilateral lung, contralateral lung, and contralateral breast, respectively.

CONCLUSIONS

The deep learning-based approach enabled automatic and reliable prediction of the DVH based on delineated structures. The predicted DVHs could potentially serve as patient-specific clinical goals used to aid treatment planning and avoid suboptimal plans or to derive optimization objectives and constraints for automated treatment planning.

Transformer-Integrated Hybrid Convolutional Neural Network for Dose Prediction in Nasopharyngeal Carcinoma Radiotherapy

Radiotherapy is recognized as the major treatment of nasopharyngeal carcinoma. Rapid and accurate dose prediction can improve the efficiency of the treatment planning process and the quality of radiotherapy plans. Currently, deep learning-based methods have been widely applied to dose prediction for radiotherapy treatment planning. However, it is important to note that existing models based on Convolutional Neural Networks (CNN) often overlook long-distance information. Although some studies try to use Transformer to solve the problem, it lacks the ability of CNN to process the spatial information inherent in images. Therefore, we propose a novel CNN and Transformer hybrid dose prediction model. To enhance the transmission ability of features between CNN and Transformer, we design a hierarchical dense recurrent encoder with a channel attention mechanism. Additionally, we propose a progressive decoder that preserves richer texture information through layer-wise reconstruction of high-dimensional feature maps. The proposed model also introduces object-driven skip connections, which facilitate the flow of information between the encoder and decoder. Experiments are conducted on in-house datasets, and the results show that the proposed model is superior to baseline methods in most dosimetric criteria. In addition, the image analysis metrics including PSNR, SSIM, and NRMSE demonstrate that the proposed model is consistent with ground truth and produces promising visual effects compared to other advanced methods. The proposed method could be taken as a powerful clinical guidance tool for physicists, significantly enhancing the efficiency of radiotherapy planning. The source code is available at https://github.com/CDUTJ102/THCN-Net .

DoseDiff: Distance-Aware Diffusion Model for Dose Prediction in Radiotherapy

Treatment planning, which is a critical component of the radiotherapy workflow, is typically carried out by a medical physicist in a time-consuming trial-and-error manner. Previous studies have proposed knowledge-based or deep-learning-based methods for predicting dose distribution maps to assist medical physicists in improving the efficiency of treatment planning. However, these dose prediction methods usually fail to effectively utilize distance information between surrounding tissues and targets or organs-at-risk (OARs). Moreover, they are poor at maintaining the distribution characteristics of ray paths in the predicted dose distribution maps, resulting in a loss of valuable information. In this paper, we propose a distance-aware diffusion model (DoseDiff) for precise prediction of dose distribution. We define dose prediction as a sequence of denoising steps, wherein the predicted dose distribution map is generated with the conditions of the computed tomography (CT) image and signed distance maps (SDMs). The SDMs are obtained by distance transformation from the masks of targets or OARs, which provide the distance from each pixel in the image to the outline of the targets or OARs. We further propose a multi-encoder and multi-scale fusion network (MMFNet) that incorporates multi-scale and transformer-based fusion modules to enhance information fusion between the CT image and SDMs at the feature level. We evaluate our model on two in-house datasets and a public dataset, respectively. The results demonstrate that our DoseDiff method outperforms state-of-the-art dose prediction methods in terms of both quantitative performance and visual quality.

Improving 3D dose prediction for breast radiotherapy using novel glowing masks and gradient-weighted loss functions

BACKGROUND

The quality of treatment plans for breast cancer can vary greatly. This variation could be reduced by using dose prediction to automate treatment planning. Our work investigates novel methods for training deep-learning models that are capable of producing high-quality dose predictions for breast cancer treatment planning.

PURPOSE

The goal of this work was to compare the performance impact of two novel techniques for deep learning dose prediction models for tangent field treatments for breast cancer. The first technique, a "glowing" mask algorithm, encodes the distance from a contour into each voxel in a mask. The second, a gradient-weighted mean squared error (MSE) loss function, emphasizes the error in high-dose gradient regions in the predicted image.

METHODS

Four 3D U-Net deep learning models were trained using the planning CT and contours of the heart, lung, and tumor bed as inputs. The dataset consisted of 305 treatment plans split into 213/46/46 training/validation/test sets using a 70/15/15% split. We compared the impact of novel "glowing" anatomical mask inputs and a novel gradient-weighted MSE loss function to their standard counterparts, binary anatomical masks, and MSE loss, using an ablation study methodology. To assess performance, we examined the mean error and mean absolute error (ME/MAE) in dose across all within-body voxels, the error in mean dose to heart, ipsilateral lung, and tumor bed, dice similarity coefficient (DSC) across isodose volumes defined by 0%-100% prescribed dose thresholds, and gamma analysis (3%/3 mm).

RESULTS

The combination of novel glowing masks and gradient weighted loss function yielded the best-performing model in this study. This model resulted in a mean ME of 0.40%, MAE of 2.70%, an error in mean dose to heart and lung of -0.10 and 0.01 Gy, and an error in mean dose to the tumor bed of -0.01%. The median DSC at 50/95/100% isodose levels were 0.91/0.87/0.82. The mean 3D gamma pass rate (3%/3 mm) was 93%.

CONCLUSIONS

This study found the combination of novel anatomical mask inputs and loss function for dose prediction resulted in superior performance to their standard counterparts. These results have important implications for the field of radiotherapy dose prediction, as the methods used here can be easily incorporated into many other dose prediction models for other treatment sites. Additionally, this dose prediction model for breast radiotherapy has sufficient performance to be used in an automated planning pipeline for tangent field radiotherapy and has the major benefit of not requiring a PTV for accurate dose prediction.

Automated plan generation for prostate radiotherapy patients using deep learning and scripted optimization

Background and Purpose

Treatment planning is a time-intensive task that could be automated. We aimed to develop a "single-click" workflow, fully deployed within a commercial treatment planning system (TPS), for autoplanning prostate radiotherapy treatment plans using predictions from a deep learning model (DLM).

Materials and Methods

Automatically generated treatment plans were created with a single script, executed from within a commercial TPS scripting environment, that performed two stages sequentially. Initially, a 3D dose distribution was predicted with a ResUNet DLM. The DLM was trained and validated using previously treated datasets (n = 120) which used 3D contours as inputs. Following this, dose predictions were converted into treatment plans by extracting dose-volume metrics from the predictions to use as objectives for the inverse optimizer within the TPS. An independent test dataset (n = 20) was used to evaluate the similarity between automated and clinical plans.

Results

For planning target volumes, the median percentage difference and interquartile range between the automatically generated plans and clinical plans were 0.4% [0.2-1.1%] for the V100%, -0.5% [(-1.0)-(-0.2)%] for D99% and -0.5% [(-1.0)-(-0.2)%] for D95%. Bladder and rectum volume-at-dose objectives agreed within -6.1% [(-12.5)-0.9%]. The conversion of the DLM prediction into a treatment plan took 15 min [13-16 min].

Conclusions

An automatic plan generation workflow that uses a DL model with scripted optimization was fully deployed in a commercial TPS. Autoplans were compared to previously treated clinical plans and were found to be non-inferior.

Prediction of dose distributions for non-small cell lung cancer patients using MHA-ResUNet

BACKGROUND

The current level of automation in the production of radiotherapy plans for lung cancer patients is relatively low. With the development of artificial intelligence, it has become a reality to use neural networks to predict dose distributions and provide assistance for radiation therapy planning. However, due to the significant individual variability in the distribution of non-small cell lung cancer (NSCLC) planning target volume (PTV) and the complex spatial relationships between the PTV and organs at risk (OARs), there is still a lack of a high-precision dose prediction network tailored to the characteristics of NSCLC.

PURPOSE

To assist in the development of volumetric modulated arc therapy (VMAT) plans for non-small cell lung cancer patients, a deep neural network is proposed to predict high-precision dose distribution.

METHODS

This study has developed a network called MHA-ResUNet, which combines a large-kernel dilated convolution module and multi-head attention (MHA) modules. The network was trained based on 80 VMAT plans of NSCLC patients. CT images, PTV, and OARs were fed into the independent input channel. The dose distribution was taken as the output to train the model. The performance of this network was compared with that of several commonly used networks, and the networks' performance was evaluated based on the voxel-level mean absolute error (MAE) within the PTV and OARs, as well as the error in clinical dose-volume metrics.

RESULTS

The MAE between the predicted dose distribution and the manually planned dose distribution within the PTV is 1.43 Gy, and the D95 error is less than 1 Gy. Compared with the other three commonly used networks, the dose error of the MHA-ResUNet is the smallest in PTV and OARs.

CONCLUSIONS

The proposed MHA-ResUNet network improves the receptive field and filters the shallow features to learn the relative spatial relation between the PTV and the OARs, enabling accurate prediction of dose distributions in NSCLC patients undergoing VMAT radiotherapy.

The STOIC2021 COVID-19 AI challenge: Applying reusable training methodologies to private data

Challenges drive the state-of-the-art of automated medical image analysis. The quantity of public training data that they provide can limit the performance of their solutions. Public access to the training methodology for these solutions remains absent. This study implements the Type Three (T3) challenge format, which allows for training solutions on private data and guarantees reusable training methodologies. With T3, challenge organizers train a codebase provided by the participants on sequestered training data. T3 was implemented in the STOIC2021 challenge, with the goal of predicting from a computed tomography (CT) scan whether subjects had a severe COVID-19 infection, defined as intubation or death within one month. STOIC2021 consisted of a Qualification phase, where participants developed challenge solutions using 2000 publicly available CT scans, and a Final phase, where participants submitted their training methodologies with which solutions were trained on CT scans of 9724 subjects. The organizers successfully trained six of the eight Final phase submissions. The submitted codebases for training and running inference were released publicly. The winning solution obtained an area under the receiver operating characteristic curve for discerning between severe and non-severe COVID-19 of 0.815. The Final phase solutions of all finalists improved upon their Qualification phase solutions.

Neural network dose prediction for cervical brachytherapy: Overcoming data scarcity for applicator-specific models

BACKGROUND

3D neural network dose predictions are useful for automating brachytherapy (BT) treatment planning for cervical cancer. Cervical BT can be delivered with numerous applicators, which necessitates developing models that generalize to multiple applicator types. The variability and scarcity of data for any given applicator type poses challenges for deep learning.

PURPOSE

The goal of this work was to compare three methods of neural network training-a single model trained on all applicator data, fine-tuning the combined model to each applicator, and individual (IDV) applicator models-to determine the optimal method for dose prediction.

METHODS

Models were produced for four applicator types-tandem-and-ovoid (T&O), T&O with 1-7 needles (T&ON), tandem-and-ring (T&R) and T&R with 1-4 needles (T&RN). First, the combined model was trained on 859 treatment plans from 266 cervical cancer patients treated from 2010 onwards. The train/validation/test split was 70%/16%/14%, with approximately 49%/10%/19%/22% T&O/T&ON/T&R/T&RN in each dataset. Inputs included four channels for anatomical masks (high-risk clinical target volume [HRCTV], bladder, rectum, and sigmoid), a mask indicating dwell position locations, and applicator channels for each applicator component. Applicator channels were created by mapping the 3D dose for a single dwell position to each dwell position and summing over each applicator component with uniform dwell time weighting. A 3D Cascade U-Net, which consists of two U-Nets in sequence, and mean squared error loss function were used. The combined model was then fine-tuned to produce four applicator-specific models by freezing the first U-Net and encoding layers of the second and resuming training on applicator-specific data. Finally, four IDV models were trained using only data from each applicator type. Performance of these three model types was compared using the following metrics for the test set: mean error (ME, representing model bias) and mean absolute error (MAE) over all dose voxels and ME of clinical metrics (HRCTV D90% and D2cc of bladder, rectum, and sigmoid), averaged over all patients. A positive ME indicates the clinical dose was higher than predicted. 3D global gamma analysis with the prescription dose as reference value was performed. Dice similarity coefficients (DSC) were computed for each isodose volume.

RESULTS

Fine-tuned and combined models showed better performance than IDV applicator training. Fine-tuning resulted in modest improvements in about half the metrics, compared to the combined model, while the remainder were mostly unchanged. Fine-tuned MAE = 3.98%/2.69%/5.36%/3.80% for T&O/T&R/T&ON/T&RN, and ME over all voxels = -0.08%/-0.89%/-0.59%/1.42%. ME D2cc were bladder = -0.77%/1.00%/-0.66%/-1.53%, rectum = 1.11%/-0.22%/-0.29%/-3.37%, sigmoid = -0.47%/-0.06%/-2.37%/-1.40%, and ME D90 = 2.6%/-4.4%/4.8%/0.0%. Gamma pass rates (3%/3 mm) were 86%/91%/83%/89%. Mean DSCs were 0.92%/0.92%/0.88%/0.91% for isodoses ≤ 150% of prescription.

CONCLUSIONS

3D BT dose was accurately predicted for all applicator types, as indicated by the low MAE and MEs, high gamma scores and high DSCs. Training on all treatment data overcomes challenges with data scarcity in each applicator type, resulting in superior performance than can be achieved by training on IDV applicators alone. This could presumably be explained by the fact that the larger, more diverse dataset allows the neural network to learn underlying trends and characteristics in dose that are common to all treatment applicators. Accurate, applicator-specific dose predictions could enable automated, knowledge-based planning for any cervical brachytherapy treatment.

Multi-center Dose Prediction Using Attention-aware Deep learning Algorithm Based on Transformers for Cervical Cancer Radiotherapy

AIMS

Accurate dose delivery is crucial for cervical cancer volumetric modulated arc therapy (VMAT). We aimed to develop a robust deep-learning (DL) algorithm for fast and accurate dose prediction of cervical cancer VMAT in multicenter datasets and then explore the feasibility of the DL algorithm to endometrial cancer VMAT with different prescriptions.

MATERIALS AND METHODS

We proposed the AtTranNet algorithm for three-dimensional dose prediction. A total of 367 cervical patients were enrolled in this study. Three hundred twenty-two cervical patients from 3 centers were randomly divided into 70%, 10%, and 20% as training, validation, and testing sets, respectively. Forty-five cervical patients from another center were selected for external testing. Moreover, 70 patients of endometrial cancer with different prescriptions were further selected to test the model. Prediction precision was evaluated by dosimetric difference, dose map, and dose-volume histogram metrics.

RESULTS

The prediction results were all clinically acceptable. The mean absolute error within the body in internal testing was 0.66 ± 0.63%. The maximum |δD| for planning target volume was observed in D98, which is 1.24 ± 2.73 Gy. The maximum |δD| for organs at risk was observed in Dmean of bladder, which is 4.79 ± 3.14 Gy. The maximum |δV| were observed in V40 of pelvic bones, which is 4.77 ± 4.48%.

CONCLUSION

AtTranNet showed the feasibility and reasonable accuracy in the dose prediction for cervical cancer in multiple centers. The model can also be generalized for endometrial cancer with different prescriptions without any transfer learning.

Knowledge-based planning for Gamma Knife

BACKGROUND

Current methods for Gamma Knife (GK) treatment planning utilizes either manual forward planning, where planners manually place shots in a tumor to achieve a desired dose distribution, or inverse planning, whereby the dose delivered to a tumor is optimized for multiple objectives based on established metrics. For other treatment modalities like IMRT and VMAT, there has been a recent push to develop knowledge-based planning (KBP) pipelines to address the limitations presented by forward and inverse planning. However, no complete KBP pipeline has been created for GK.

PURPOSE

To develop a novel (KBP) pipeline, using inverse optimization (IO) with 3D dose predictions for GK.

METHODS

Data were obtained for 349 patients from Sunnybrook Health Sciences Centre. A 3D dose prediction model was trained using 322 patients, based on a previously published deep learning methodology, and dose predictions were generated for the remaining 27 out-of-sample patients. A generalized IO model was developed to learn objective function weights from dose predictions. These weights were then used in an inverse planning model to generate deliverable treatment plans. A dose mimicking (DM) model was also implemented for comparison. The quality of the resulting plans was compared to their clinical counterparts using standard GK quality metrics. The performance of the models was also characterized with respect to the dose predictions.

RESULTS

Across all quality metrics, plans generated using the IO pipeline performed at least as well as or better than the respective clinical plans. The average conformity and gradient indices of IO plans was 0.737 ± $\pm$ 0.158 and 3.356 ± $\pm$ 1.030 respectively, compared to 0.713 ± $\pm$ 0.124 and 3.452 ± $\pm$ 1.123 for the clinical plans. IO plans also performed better than DM plans for five of the six quality metrics. Plans generated using IO also have average treatment times comparable to that of clinical plans. With regards to the dose predictions, predictions with higher conformity tend to result in higher quality KBP plans.

CONCLUSIONS

Plans resulting from an IO KBP pipeline are, on average, of equal or superior quality compared to those obtained through manual planning. The results demonstrate the potential for the use of KBP to generate GK treatment with minimal human intervention.

Inter-institutional variability of knowledge-based plan prediction of left whole breast irradiation

PURPOSE

Within a multi-institutional project, we aimed to assess the transferability of knowledge-based (KB) plan prediction models in the case of whole breast irradiation (WBI) for left-side breast irradiation with tangential fields (TF).

METHODS

Eight institutions set KB models, following previously shared common criteria. Plan prediction performance was tested on 16 new patients (2 pts per centre) extracting dose-volume-histogram (DVH) prediction bands of heart, ipsilateral lung, contralateral lung and breast. The inter-institutional variability was quantified by the standard deviations (SDint) of predicted DVHs and mean-dose (Dmean). The transferability of models, for the heart and the ipsilateral lung, was evaluated by the range of geometric Principal Component (PC1) applicability of a model to test patients of the other 7 institutions.

RESULTS

SDint of the DVH was 1.8 % and 1.6 % for the ipsilateral lung and the heart, respectively (20 %-80 % dose range); concerning Dmean, SDint was 0.9 Gy and 0.6 Gy for the ipsilateral lung and the heart, respectively (<0.2 Gy for contralateral organs). Mean predicted doses ranged between 4.3 and 5.9 Gy for the ipsilateral lung and 1.1-2.3 Gy for the heart. PC1 analysis suggested no relevant differences among models, except for one centre showing a systematic larger sparing of the heart, concomitant to a worse PTV coverage, due to high priority in sparing the left anterior descending coronary artery.

CONCLUSIONS

Results showed high transferability among models and low inter-institutional variability of 2% for plan prediction. These findings encourage the building of benchmark models in the case of TF-WBI.

vOARiability: Interobserver and intermodality variability analysis in OAR contouring from head and neck CT and MR images

BACKGROUND

Accurate and consistent contouring of organs-at-risk (OARs) from medical images is a key step of radiotherapy (RT) cancer treatment planning. Most contouring approaches rely on computed tomography (CT) images, but the integration of complementary magnetic resonance (MR) modality is highly recommended, especially from the perspective of OAR contouring, synthetic CT and MR image generation for MR-only RT, and MR-guided RT. Although MR has been recognized as valuable for contouring OARs in the head and neck (HaN) region, the accuracy and consistency of the resulting contours have not been yet objectively evaluated.

PURPOSE

To analyze the interobserver and intermodality variability in contouring OARs in the HaN region, performed by observers with different level of experience from CT and MR images of the same patients.

METHODS

In the final cohort of 27 CT and MR images of the same patients, contours of up to 31 OARs were obtained by a radiation oncology resident (junior observer, JO) and a board-certified radiation oncologist (senior observer, SO). The resulting contours were then evaluated in terms of interobserver variability, characterized as the agreement among different observers (JO and SO) when contouring OARs in a selected modality (CT or MR), and intermodality variability, characterized as the agreement among different modalities (CT and MR) when OARs were contoured by a selected observer (JO or SO), both by the Dice coefficient (DC) and 95-percentile Hausdorff distance (HD95 $_{95}$ ).

RESULTS

The mean (±standard deviation) interobserver variability was 69.0 ± 20.2% and 5.1 ± 4.1 mm, while the mean intermodality variability was 61.6 ± 19.0% and 6.1 ± 4.3 mm in terms of DC and HD95 $_{95}$ , respectively, across all OARs. Statistically significant differences were only found for specific OARs. The performed MR to CT image registration resulted in a mean target registration error of 1.7 ± 0.5 mm, which was considered as valid for the analysis of intermodality variability.

CONCLUSIONS

The contouring variability was, in general, similar for both image modalities, and experience did not considerably affect the contouring performance. However, the results indicate that an OAR is difficult to contour regardless of whether it is contoured in the CT or MR image, and that observer experience may be an important factor for OARs that are deemed difficult to contour. Several of the differences in the resulting variability can be also attributed to adherence to guidelines, especially for OARs with poor visibility or without distinctive boundaries in either CT or MR images. Although considerable contouring differences were observed for specific OARs, it can be concluded that almost all OARs can be contoured with a similar degree of variability in either the CT or MR modality, which works in favor of MR images from the perspective of MR-only and MR-guided RT.

A cascade transformer-based model for 3D dose distribution prediction in head and neck cancer radiotherapy

Objective. Radiation therapy is one of the primary methods used to treat cancer in the clinic. Its goal is to deliver a precise dose to the planning target volume while protecting the surrounding organs at risk (OARs). However, the traditional workflow used by dosimetrists to plan the treatment is time-consuming and subjective, requiring iterative adjustments based on their experience. Deep learning methods can be used to predict dose distribution maps to address these limitations.Approach. The study proposes a cascade model for OARs segmentation and dose distribution prediction. An encoder-decoder network has been developed for the segmentation task, in which the encoder consists of transformer blocks, and the decoder uses multi-scale convolutional blocks. Another cascade encoder-decoder network has been proposed for dose distribution prediction using a pyramid architecture. The proposed model has been evaluated using an in-house head and neck cancer dataset of 96 patients and OpenKBP, a public head and neck cancer dataset of 340 patients.Main results. The segmentation subnet achieved 0.79 and 2.71 for Dice and HD95 scores, respectively. This subnet outperformed the existing baselines. The dose distribution prediction subnet outperformed the winner of the OpenKBP2020 competition with 2.77 and 1.79 for dose and dose-volume histogram scores, respectively. Besides, the end-to-end model, including both subnets simultaneously, outperformed the related studies.Significance. The predicted dose maps showed good coincidence with ground-truth, with a superiority after linking with the auxiliary segmentation task. The proposed model outperformed state-of-the-art methods, especially in regions with low prescribed doses. The codes are available athttps://github.com/GhTara/Dose_Prediction.

Beam-wise dose composition learning for head and neck cancer dose prediction in radiotherapy

Automatic and accurate dose distribution prediction plays an important role in radiotherapy plan. Although previous methods can provide promising performance, most methods did not consider beam-shaped radiation of treatment delivery in clinical practice. This leads to inaccurate prediction, especially on beam paths. To solve this problem, we propose a beam-wise dose composition learning (BDCL) method for dose prediction in the context of head and neck (H&N) radiotherapy plan. Specifically, a global dose network is first utilized to predict coarse dose values in the whole-image space. Then, we propose to generate individual beam masks to decompose the coarse dose distribution into multiple field doses, called beam voters, which are further refined by a subsequent beam dose network and reassembled to form the final dose distribution. In particular, we design an overlap consistency module to keep the similarity of high-level features in overlapping regions between different beam voters. To make the predicted dose distribution more consistent with the real radiotherapy plan, we also propose a dose-volume histogram (DVH) calibration process to facilitate feature learning in some clinically concerned regions. We further apply an edge enhancement procedure to enhance the learning of the extracted feature from the dose falloff regions. Experimental results on a public H&N cancer dataset from the AAPM OpenKBP challenge show that our method achieves superior performance over other state-of-the-art approaches by significant margins. Source code is released at https://github.com/TL9792/BDCLDosePrediction.

Enhancing dosimetric practices through knowledge-based predictive models: a case study on VMAT prostate irradiation

Introduction

Acquisition of dosimetric knowledge by radiation therapy planners is a protracted and complex process. This study delves into the impact of empirical predictive models based on the knowledge-based planning (KBP) methodology, aimed at detecting suboptimal results and homogenizing and improving existing practices for prostate cancer. Moreover, the dosimetric effect of implementing these models into routine clinical practice was also assessed.

Materials and methods

Based on the KBP method, we analyzed 25 prostate treatment plans performed using VMAT by expert operators, aiming to correlate dose indicators with patient geometry. The D a v g C a v ( G y ) , V 45 G y C a v ( c c ) , and V 15 G y C a v ( c c ) of the peritoneal cavity and the V 60 G y ( % ) and V 70 G y ( % ) of the rectum and bladder were linked to geometric characteristics such as the distance from the planning target volume (PTV) to the organs at risk (OAR), the volume of the OAR, or the overlap between the PTV and the OAR. In the second phase, the KBP was used in routine clinical practice in a prospective cohort of 25 patients and compared with the 41 patient plans calculated before implementing the tool.

Results

Using linear regression, we identified strong geometric predictive factors for the peritoneal cavity, rectum, and bladder (R 2   > 0.8), with an average prescribed dose of 97.8%, covering 95% of the target volume. The use of the model led to a significant dose reduction ( Δ ) for all evaluated OARs. This trend was most notable for Δ V 15 G y C a v = - 171.5  cc  ( p = 0.003 ) . Significant reductions were also obtained in average doses to the rectum and bladder, Δ D a v g R e c t =   - 2.3   G y   ( p = 0.040 ) , and Δ D a v g V e s s =   - 3.3   G y   ( p = 0.039 ) respectively. Based on this model, we reduced the number of plans with OAR constraints above the clinical recommendations from 19% to 8%.

Conclusions

The KBP methodology established a robust and personalized predictive model for dose estimation to organs at risk in prostate cancer. Implementing the model resulted in improved sparing of these organs. Notably, it yields a solid foundation for harmonizing dosimetric practices, alerting us to suboptimal results, and improving our knowledge.

A Comparative Study of Deep Learning Dose Prediction Models for Cervical Cancer Volumetric Modulated Arc Therapy

Purpose: Deep learning (DL) is widely used in dose prediction for radiation oncology, multiple DL techniques comparison is often lacking in the literature. To compare the performance of 4 state-of-the-art DL models in predicting the voxel-level dose distribution for cervical cancer volumetric modulated arc therapy (VMAT). Methods and Materials: A total of 261 patients' plans for cervical cancer were retrieved in this retrospective study. A three-channel feature map, consisting of a planning target volume (PTV) mask, organs at risk (OARs) mask, and CT image was fed into the three-dimensional (3D) U-Net and its 3 variants models. The data set was randomly divided into 80% as training-validation and 20% as testing set, respectively. The model performance was evaluated on the 52 testing patients by comparing the generated dose distributions against the clinical approved ground truth (GT) using mean absolute error (MAE), dose map difference (GT-predicted), clinical dosimetric indices, and dice similarity coefficients (DSC). Results: The 3D U-Net and its 3 variants DL models exhibited promising performance with a maximum MAE within the PTV 0.83% ± 0.67% in the UNETR model. The maximum MAE among the OARs is the left femoral head, which reached 6.95% ± 6.55%. For the body, the maximum MAE was observed in UNETR, which is 1.19 ± 0.86%, and the minimum MAE was 0.94 ± 0.85% for 3D U-Net. The average error of the Dmean difference for different OARs is within 2.5 Gy. The average error of V40 difference for the bladder and rectum is about 5%. The mean DSC under different isodose volumes was above 90%. Conclusions: DL models can predict the voxel-level dose distribution accurately for cervical cancer VMAT treatment plans. All models demonstrated almost analogous performance for voxel-wise dose prediction maps. Considering all voxels within the body, 3D U-Net showed the best performance. The state-of-the-art DL models are of great significance for further clinical applications of cervical cancer VMAT.

Deep learning-based dose prediction to improve the plan quality of volumetric modulated arc therapy for gynecologic cancers

BACKGROUND

In recent years, deep-learning models have been used to predict entire three-dimensional dose distributions. However, the usability of dose predictions to improve plan quality should be further investigated.

PURPOSE

To develop a deep-learning model to predict high-quality dose distributions for volumetric modulated arc therapy (VMAT) plans for patients with gynecologic cancer and to evaluate their usability in driving plan quality improvements.

METHODS

A total of 79 VMAT plans for the female pelvis were used to train (47 plans), validate (16 plans), and test (16 plans) 3D dense dilated U-Net models to predict 3D dose distributions. The models received the normalized CT scan, dose prescription, and target and normal tissue contours as inputs. Three models were used to predict the dose distributions for plans in the test set. A radiation oncologist specializing in the treatment of gynecologic cancers scored the test set predictions using a 5-point scale (5, acceptable as-is; 4, prefer minor edits; 3, minor edits needed; 2, major edits needed; and 1, unacceptable). The clinical plans for which the dose predictions indicated that improvements could be made were reoptimized with constraints extracted from the predictions.

RESULTS

The predicted dose distributions in the test set were of comparable quality to the clinical plans. The mean voxel-wise dose difference was -0.14 ± 0.46 Gy. The percentage dose differences in the predicted target metrics ofD 1 % ${D}_{1{\mathrm{\% }}}$ andD 98 % ${D}_{98{\mathrm{\% }}}$ were -1.05% ± 0.59% and 0.21% ± 0.28%, respectively. The dose differences in the predicted organ at risk mean and maximum doses were -0.30 ± 1.66 Gy and -0.42 ± 2.07 Gy, respectively. A radiation oncologist deemed all of the predicted dose distributions clinically acceptable; 12 received a score of 5, and four received a score of 4. Replanning of flagged plans (five plans) showed that the original plans could be further optimized to give dose distributions close to the predicted dose distributions.

CONCLUSIONS

Deep-learning dose prediction can be used to predict high-quality and clinically acceptable dose distributions for VMAT female pelvis plans, which can then be used to identify plans that can be improved with additional optimization.

Clinical Implementation and Evaluation of Auto-Segmentation Tools for Multi-Site Contouring in Radiotherapy

Background and purpose

Tools for auto-segmentation in radiotherapy are widely available, but guidelines for clinical implementation are missing. The goal was to develop a workflow for performance evaluation of three commercial auto-segmentation tools to select one candidate for clinical implementation.

Materials and Methods

One hundred patients with six treatment sites (brain, head-and-neck, thorax, abdomen, and pelvis) were included. Three sets of AI-based contours for organs-at-risk (OAR) generated by three software tools and manually drawn expert contours were blindly rated for contouring accuracy. The dice similarity coefficient (DSC), the Hausdorff distance, and a dose/volume evaluation based on the recalculation of the original treatment plan were assessed. Statistically significant differences were tested using the Kruskal-Wallis test and the post-hoc Dunn Test with Bonferroni correction.

Results

The mean DSC scores compared to expert contours for all OARs combined were 0.80 ± 0.10, 0.75 ± 0.10, and 0.74 ± 0.11 for the three software tools. Physicians' rating identified equivalent or superior performance of some AI-based contours in head (eye, lens, optic nerve, brain, chiasm), thorax (e.g., heart and lungs), and pelvis and abdomen (e.g., kidney, femoral head) compared to manual contours. For some OARs, the AI models provided results requiring only minor corrections. Bowel-bag and stomach were not fit for direct use. During the interdisciplinary discussion, the physicians' rating was considered the most relevant.

Conclusion

A comprehensive method for evaluation and clinical implementation of commercially available auto-segmentation software was developed. The in-depth analysis yielded clear instructions for clinical use within the radiotherapy department.

Automated evaluation for rapid implementation of knowledge-based radiotherapy planning models

PURPOSE

Knowledge-based planning (KBP) offers the ability to predict dose-volume metrics based on information extracted from previous plans, reducing plan variability and improving plan quality. As clinical integration of KBP is increasing there is a growing need for quantitative evaluation of KBP models. A .NET-based application, RapidCompare, was created for automated plan creation and analysis of Varian RapidPlan models.

METHODS

RapidCompare was designed to read calculation parameters and a list of reference plans. The tool copies the reference plan field geometry and structure set, applies the RapidPlan model, optimizes the KBP plan, and generates data for quantitative evaluation of dose-volume metrics. A cohort of 85 patients, divided into training (50), testing (10), and validation (25) groups, was used to demonstrate the utility of RapidCompare. After training and tuning, the KBP model was paired with three different optimization templates to compare various planning strategies in the validation cohort. All templates used the same set of constraints for the planning target volume (PTV). For organs-at-risk, the optimization template provided constraints using the whole dose-volume histogram (DVH), fixed-dose/volume points, or generalized equivalent uniform dose (gEUD). The resulting plans from each optimization approach were compared using DVH metrics.

RESULTS

RapidCompare allowed for the automated generation of 75 total plans for comparison with limited manual intervention. In comparing optimization techniques, the Dose/Volume and Lines optimization templates generated plans with similar DVH metrics, with a slight preference for the Lines technique with reductions in heart V30Gy and spinal cord max dose. The gEUD model produced high target heterogeneity.

CONCLUSION

Automated evaluation allowed for the exploration of multiple optimization templates in a larger validation cohort than would have been feasible using a manual approach. A final KBP model using line optimization objectives produced the highest quality plans without human intervention.

A comparative study of deep learning-based knowledge-based planning methods for 3D dose distribution prediction of head and neck

PURPOSE

In this paper, we compare four novel knowledge-based planning (KBP) algorithms using deep learning to predict three-dimensional (3D) dose distributions of head and neck plans using the same patients' dataset and quantitative assessment metrics.

METHODS

A dataset of 340 oropharyngeal cancer patients treated with intensity-modulated radiation therapy was used in this study, which represents the AAPM OpenKBP - 2020 Grand Challenge dataset. Four 3D convolutional neural network architectures were built. The models were trained on 64% of the data set and validated on 16% for voxel-wise dose predictions: U-Net, attention U-Net, residual U-Net (Res U-Net), and attention Res U-Net. The trained models were then evaluated for their performance on a test data set (20% of the data) by comparing the predicted dose distributions against the ground-truth using dose statistics and dose-volume indices.

RESULTS

The four KBP dose prediction models exhibited promising performance with an averaged mean absolute dose error within the body contour <3 Gy on 68 plans in the test set. The average difference in predicting the D99 index for all targets was 0.92 Gy (p = 0.51) for attention Res U-Net, 0.94 Gy (p = 0.40) for Res U-Net, 2.94 Gy (p = 0.09) for attention U-Net, and 3.51 Gy (p = 0.08) for U-Net. For the OARs, the values for theD m a x ${D_{max}}$ andD m e a n ${D_{mean}}$ indices were 2.72 Gy (p < 0.01) for attention Res U-Net, 2.94 Gy (p < 0.01) for Res U-Net, 1.10 Gy (p < 0.01) for attention U-Net, 0.84 Gy (p < 0.29) for U-Net.

CONCLUSION

All models demonstrated almost comparable performance for voxel-wise dose prediction. KBP models that employ 3D U-Net architecture as a base could be deployed for clinical use to improve cancer patient treatment by creating plans with consistent quality and making the radiotherapy workflow more efficient.

Deep-Learning-Based Dose Predictor for Glioblastoma-Assessing the Sensitivity and Robustness for Dose Awareness in Contouring

External beam radiation therapy requires a sophisticated and laborious planning procedure. To improve the efficiency and quality of this procedure, machine-learning models that predict these dose distributions were introduced. The most recent dose prediction models are based on deep-learning architectures called 3D U-Nets that give good approximations of the dose in 3D almost instantly. Our purpose was to train such a 3D dose prediction model for glioblastoma VMAT treatment and test its robustness and sensitivity for the purpose of quality assurance of automatic contouring. From a cohort of 125 glioblastoma (GBM) patients, VMAT plans were created according to a clinical protocol. The initial model was trained on a cascaded 3D U-Net. A total of 60 cases were used for training, 15 for validation and 20 for testing. The prediction model was tested for sensitivity to dose changes when subject to realistic contour variations. Additionally, the model was tested for robustness by exposing it to a worst-case test set containing out-of-distribution cases. The initially trained prediction model had a dose score of 0.94 Gy and a mean DVH (dose volume histograms) score for all structures of 1.95 Gy. In terms of sensitivity, the model was able to predict the dose changes that occurred due to the contour variations with a mean error of 1.38 Gy. We obtained a 3D VMAT dose prediction model for GBM with limited data, providing good sensitivity to realistic contour variations. We tested and improved the model's robustness by targeted updates to the training set, making it a useful technique for introducing dose awareness in the contouring evaluation and quality assurance process.

Generating deliverable DICOM RT treatment plans for prostate VMAT by predicting MLC motion sequences with an encoder-decoder network

BACKGROUND

Deep learning-based auto-planning is an active research field; however, for some tasks a treatment planning system (TPS) is still required.

PURPOSE

To introduce a deep learning-based model generating deliverable DICOM RT treatment plans that can be directly irradiated by a linear accelerator (LINAC). The model was based on an encoder-decoder network and can predict multileaf collimator (MLC) motion sequences for prostate VMAT radiotherapy.

METHODS

A total of 619 treatment plans from 460 patients treated for prostate cancer with single-arc VMAT were included in this study. An encoder-decoder network was trained using 465 clinical treatment plans and validated on 77 plans. The performance was analyzed on a separate test set of 77 treatment plans. Separate L1 losses were computed for the leaf and jaw positions as well as the monitor units, with the leaf loss being weighted by a factor of 100 before being added to the other losses. The generated treatment plans were recalculated in a treatment planning system and the dose-volume metrics and gamma passing rates were compared to the original dose.

RESULTS

All generated treatment plans showed good agreement with the original data, with an average gamma passing rate (3%/3 mm) of 91.9 ± 7.1%. However, the coverage of the PTVs. was slightly lower for the generated plans (D98%  = 92.9 ± 2.6%) in comparison to the original plans (D98%  = 95.7 ± 2.2%). There was no significant difference in mean dose to the bladder between the predicted and original plan (Dmean of 28.0 ± 13.5 vs. 28.1 ± 13.3% of prescribed dose) or rectum (Dmean of 42.3 ± 7.4 vs. 42.6 ± 7.5%). The maximum dose to bladder was only slightly higher in the predicted plans (D2% of 100.7 ± 5.3 vs. 99.8 ± 4.0%) and for the rectum it was even lower (D2% of 100.5 ± 3.7 vs. 100.1 ± 4.3).

CONCLUSIONS

The deep learning-based model could predict MLC motion sequences in prostate VMAT plans, eliminating the need for sequencing inside a TPS, thus revolutionizing autonomous treatment planning workflows. This research completes the loop in deep learning-based treatment planning processes, enabling more efficient workflows for real-time or online adaptive radiotherapy.

Increasing Quality and Efficiency of the Radiotherapy Treatment Planning Process by Constructing and Implementing a Workflow-Monitoring Application

PURPOSE

To demonstrate how the efficiency of the treatment planning processes of a university radiation oncology department (2,500 new patients/year) could be improved by constructing and implementing a workflow-monitoring application.

METHODS

A web-based application was developed in house, which enhanced the process management tools of the clinic's oncology information system. The application calculates the days left for the next task in the treatment planning process and visualizes the information on a browser-based whiteboard. Workflow monitoring considers tumor types (breast, prostate, lung, etc) and treatment techniques and is backward planned from the planned start of treatment. The effect of introducing this application was analyzed over four phases: (1) baseline data without the workflow-monitoring application, (2) after introducing workflow visualization via a browser-based whiteboard, (3) after upgrading the whiteboard and introducing backend rules, and (4) after updating these rules on the basis of data from the previous phase.

RESULTS

Implementing the workflow-monitoring application and the introduced measures significantly reduced delays and, consequently, stress and a negative working atmosphere in the treatment planning process. Most notably, the amount of last-minute physics checks (on the day of the treatment start) could be reduced by 50%.

CONCLUSION

The study showed what measures can help organize and prioritize the treatment planning workflow. The increased efficiency is believed to improve the quality and reduce the risk of human error.

TrDosePred: A deep learning dose prediction algorithm based on transformers for head and neck cancer radiotherapy

BACKGROUND

Intensity-Modulated Radiation Therapy (IMRT) has been the standard of care for many types of tumors. However, treatment planning for IMRT is a time-consuming and labor-intensive process.

PURPOSE

To alleviate this tedious planning process, a novel deep learning based dose prediction algorithm (TrDosePred) was developed for head and neck cancers.

METHODS

The proposed TrDosePred, which generated the dose distribution from a contoured CT image, was a U-shape network constructed with a convolutional patch embedding and several local self-attention based transformers. Data augmentation and ensemble approach were used for further improvement. It was trained based on the dataset from Open Knowledge-Based Planning Challenge (OpenKBP). The performance of TrDosePred was evaluated with two mean absolute error (MAE) based scores utilized by OpenKBP challenge (i.e., Dose score and DVH score) and compared to the top three approaches of the challenge. In addition, several state-of-the-art methods were implemented and compared to TrDosePred.

RESULTS

The TrDosePred ensemble achieved the dose score of 2.426 Gy and the DVH score of 1.592 Gy on the test dataset, ranking at 3rd and 9th respectively in the leaderboard on CodaLab as of writing. In terms of DVH metrics, on average, the relative MAE against the clinical plans was 2.25% for targets and 2.17% for organs at risk.

CONCLUSIONS

A transformer-based framework TrDosePred was developed for dose prediction. The results showed a comparable or superior performance as compared to the previous state-of-the-art approaches, demonstrating the potential of transformer to boost the treatment planning procedures.

ASTRA: Atomic Surface Transformations for Radiotherapy Quality Assurance

Treatment for glioblastoma, an aggressive brain tumour usually relies on radiotherapy. This involves planning how to achieve the desired radiation dose distribution, which is known as treatment planning. Treatment planning is impacted by human errors, inter-expert variability in segmenting (or outlining) the tumor target and organs-at-risk, and differences in segmentation protocols. Erroneous segmentations translate to erroneous dose distributions, and hence sub-optimal clinical outcomes. Reviewing segmentations is time-intensive, significantly reduces the efficiency of radiation oncology teams, and hence restricts timely radiotherapy interventions to limit tumor growth. Moreover, to date, radiation oncologists review and correct segmentations without information on how potential corrections might affect radiation dose distributions, leading to an ineffective and suboptimal segmentation correction workflow. In this paper, we introduce an automated deep-learning based method: atomic surface transformations for radiotherapy quality assurance (ASTRA), that predicts the potential impact of local segmentation variations on radiotherapy dose predictions, thereby serving as an effective dose-aware sensitivity map of segmentation variations. On a dataset of 100 glioblastoma patients, we show how the proposed approach enables assessment and visualization of areas of organs-at-risk being most susceptible to dose changes, providing clinicians with a dose-informed mechanism to review and correct segmentations for radiation therapy planning. These initial results suggest strong potential for employing such methods within a broader automated quality assurance system in the radiotherapy planning workflow. Code to reproduce this is available at https://github.com/amithjkamath/astraClinical Relevance: ASTRA shows promise in indicating what regions of the OARs are more likely to impact the distribution of radiation dose.

An Efficient Multi-Scale Fusion Network for 3D Organs at Risk (OARs) Segmentation

Accurate segmentation of organs-at-risks (OARs) is a precursor for optimizing radiation therapy planning. Existing deep learning-based multi-scale fusion architectures have demonstrated a tremendous capacity for 2D medical image segmentation. The key to their success is aggregating global context and maintaining high resolution representations. However, when translated into 3D segmentation problems, existing multi-scale fusion architectures might underperform due to their heavy computation overhead and substantial data diet. To address this issue, we propose a new OAR segmentation framework, called OARFocalFuseNet, which fuses multi-scale features and employs focal modulation for capturing global-local context across multiple scales. Each resolution stream is enriched with features from different resolution scales, and multi-scale information is aggregated to model diverse contextual ranges. As a result, feature representations are further boosted. The comprehensive comparisons in our experimental setup with OAR segmentation as well as multi-organ segmentation show that our proposed OARFocalFuseNet outperforms the recent state-of-the-art methods on publicly available OpenKBP datasets and Synapse multi-organ segmentation. Both of the proposed methods (3D-MSF and OARFocalFuseNet) showed promising performance in terms of standard evaluation metrics. Our best performing method (OARFocalFuseNet) obtained a dice coefficient of 0.7995 and hausdorff distance of 5.1435 on OpenKBP datasets and dice coefficient of 0.8137 on Synapse multi-organ segmentation dataset. Our code is available at https://github.com/NoviceMAn-prog/OARFocalFuse.

Learning image representations for content-based image retrieval of radiotherapy treatment plans

Objective.In this work, we propose a content-based image retrieval (CBIR) method for retrieving dose distributions of previously planned patients based on anatomical similarity. Retrieved dose distributions from this method can be incorporated into automated treatment planning workflows in order to streamline the iterative planning process. As CBIR has not yet been applied to treatment planning, our work seeks to understand which current machine learning models are most viable in this context.Approach.Our proposed CBIR method trains a representation model that produces latent space embeddings of a patient's anatomical information. The latent space embeddings of new patients are then compared against those of previous patients in a database for image retrieval of dose distributions. All source code for this project is available on github.Main results.The retrieval performance of various CBIR methods is evaluated on a dataset consisting of both publicly available image sets and clinical image sets from our institution. This study compares various encoding methods, ranging from simple autoencoders to more recent Siamese networks like SimSiam, and the best performance was observed for the multitask Siamese network.Significance.Our current results demonstrate that excellent image retrieval performance can be obtained through slight changes to previously developed Siamese networks. We hope to integrate CBIR into automated planning workflow in future works.

Deep Learning-Based Dose Prediction for Automated, Individualized Quality Assurance of Head and Neck Radiation Therapy Plans

PURPOSE

This study aimed to use deep learning-based dose prediction to assess head and neck (HN) plan quality and identify suboptimal plans.

METHODS AND MATERIALS

A total of 245 volumetric modulated arc therapy HN plans were created using RapidPlan knowledge-based planning (KBP). A subset of 112 high-quality plans was selected under the supervision of an HN radiation oncologist. We trained a 3D Dense Dilated U-Net architecture to predict 3-dimensional dose distributions using 3-fold cross-validation on 90 plans. Model inputs included computed tomography images, target prescriptions, and contours for targets and organs at risk (OARs). The model's performance was assessed on the remaining 22 test plans. We then tested the application of the dose prediction model for automated review of plan quality. Dose distributions were predicted on 14 clinical plans. The predicted versus clinical OAR dose metrics were compared to flag OARs with suboptimal normal tissue sparing using a 2 Gy dose difference or 3% dose-volume threshold. OAR flags were compared with manual flags by 3 HN radiation oncologists.

RESULTS

The predicted dose distributions were of comparable quality to the KBP plans. The differences between the predicted and KBP-planned D1%,D95%, and D99% across the targets were within -2.53% ± 1.34%, -0.42% ± 1.27%, and -0.12% ± 1.97%, respectively, and the OAR mean and maximum doses were within -0.33 ± 1.40 Gy and -0.96 ± 2.08 Gy, respectively. For the plan quality assessment study, radiation oncologists flagged 47 OARs for possible plan improvement. There was high interphysician variability; 83% of physician-flagged OARs were flagged by only one of 3 physicians. The comparative dose prediction model flagged 63 OARs, including 30 of 47 physician-flagged OARs.

CONCLUSIONS

Deep learning can predict high-quality dose distributions, which can be used as comparative dose distributions for automated, individualized assessment of HN plan quality.

Automated treatment planning framework for brachytherapy of cervical cancer using 3D dose predictions

Objective. To lay the foundation for automated knowledge-based brachytherapy treatment planning using 3D dose estimations, we describe an optimization framework to convert brachytherapy dose distributions directly into dwell times (DTs).Approach. A dose rate kernelḋ(r,θ,φ)was produced by exporting 3D dose for one dwell position from the treatment planning system and normalizing by DT. By translating and rotating this kernel to each dwell position, scaling by DT and summing over all dwell positions, dose was computed (Dcalc). We used a Python-coded COBYLA optimizer to iteratively determine the DTs that minimize the mean squared error betweenDcalcand reference doseDref, computed using voxels withDref80%-120% of prescription. As validation of the optimization, we showed that the optimizer replicates clinical plans whenDref= clinical dose in 40 patients treated with tandem-and-ovoid (T&O) or tandem-and-ring (T&R) and 0-3 needles. Then we demonstrated automated planning in 10 T&O usingDref= dose predicted from a convolutional neural network developed in past work. Validation and automated plans were compared to clinical plans using mean absolute differences (MAD=1N∑n=1Nabsxn-xn') over all voxels (xn= Dose,N= #voxels) and DTs (xn= DT,N= #dwell positions), mean differences (MD) in organD2ccand high-risk CTV D90 over all patients (where positive indicates higher clinical dose), and mean Dice similarity coefficients (DSC) for 100% isodose contours.Main results. Validation plans agreed well with clinical plans (MADdose= 1.1%, MADDT= 4 s or 0.8% of total plan time,D2ccMD = -0.2% to 0.2% and D90 MD = -0.6%, DSC = 0.99). For automated plans, MADdose= 6.5% and MADDT= 10.3 s (2.1%). The slightly higher clinical metrics in automated plans (D2ccMD = -3.8% to 1.3% and D90 MD = -5.1%) were due to higher neural network dose predictions. The overall shape of the automated dose distributions were similar to clinical doses (DSC = 0.91).Significance. Automated planning with 3D dose predictions could provide significant time savings and standardize treatment planning across practitioners, regardless of experience.

Domain knowledge driven 3D dose prediction using moment-based loss function

Objective. To propose a novel moment-based loss function for predicting 3D dose distribution for the challenging conventional lung intensity modulated radiation therapy plans. The moment-based loss function is convex and differentiable and can easily incorporate clinical dose volume histogram (DVH) domain knowledge in any deep learning (DL) framework without computational overhead. Approach. We used a large dataset of 360 (240 for training, 50 for validation and 70 for testing) conventional lung patients with 2 Gy × 30 fractions to train the DL model using clinically treated plans at our institution. We trained a UNet like convolutional neural network architecture using computed tomography, planning target volume and organ-at-risk contours as input to infer corresponding voxel-wise 3D dose distribution. We evaluated three different loss functions: (1) the popular mean absolute error (MAE) loss, (2) the recently developed MAE + DVH loss, and (3) the proposed MAE + moments loss. The quality of the predictions was compared using different DVH metrics as well as dose-score and DVH-score, recently introduced by the AAPM knowledge-based planning grand challenge. Main results. Model with (MAE + moment) loss function outperformed the model with MAE loss by significantly improving the DVH-score (11%, p < 0.01) while having similar computational cost. It also outperformed the model trained with (MAE + DVH) by significantly improving the computational cost (48%) and the DVH-score (8%, p < 0.01). Significance. DVH metrics are widely accepted evaluation criteria in the clinic. However, incorporating them into the 3D dose prediction model is challenging due to their non-convexity and non-differentiability. Moments provide a mathematically rigorous and computationally efficient way to incorporate DVH information in any DL architecture. The code, pretrained models, docker container, and Google Colab project along with a sample dataset are available on our DoseRTX GitHub (https://github.com/nadeemlab/DoseRTX)

OpenKBP-Opt: an international and reproducible evaluation of 76 knowledge-based planning pipelines

Objective.To establish an open framework for developing plan optimization models for knowledge-based planning (KBP).Approach.Our framework includes radiotherapy treatment data (i.e. reference plans) for 100 patients with head-and-neck cancer who were treated with intensity-modulated radiotherapy. That data also includes high-quality dose predictions from 19 KBP models that were developed by different research groups using out-of-sample data during the OpenKBP Grand Challenge. The dose predictions were input to four fluence-based dose mimicking models to form 76 unique KBP pipelines that generated 7600 plans (76 pipelines × 100 patients). The predictions and KBP-generated plans were compared to the reference plans via: the dose score, which is the average mean absolute voxel-by-voxel difference in dose; the deviation in dose-volume histogram (DVH) points; and the frequency of clinical planning criteria satisfaction. We also performed a theoretical investigation to justify our dose mimicking models.Main results.The range in rank order correlation of the dose score between predictions and their KBP pipelines was 0.50-0.62, which indicates that the quality of the predictions was generally positively correlated with the quality of the plans. Additionally, compared to the input predictions, the KBP-generated plans performed significantly better (P< 0.05; one-sided Wilcoxon test) on 18 of 23 DVH points. Similarly, each optimization model generated plans that satisfied a higher percentage of criteria than the reference plans, which satisfied 3.5% more criteria than the set of all dose predictions. Lastly, our theoretical investigation demonstrated that the dose mimicking models generated plans that are also optimal for an inverse planning model.Significance.This was the largest international effort to date for evaluating the combination of KBP prediction and optimization models. We found that the best performing models significantly outperformed the reference dose and dose predictions. In the interest of reproducibility, our data and code is freely available.

Artificial intelligence in radiotherapy

Radiotherapy is a discipline closely integrated with computer science. Artificial intelligence (AI) has developed rapidly over the past few years. With the explosive growth of medical big data, AI promises to revolutionize the field of radiotherapy through highly automated workflow, enhanced quality assurance, improved regional balances of expert experiences, and individualized treatment guided by multi-omics. In addition to independent researchers, the increasing number of large databases, biobanks, and open challenges significantly facilitated AI studies on radiation oncology. This article reviews the latest research, clinical applications, and challenges of AI in each part of radiotherapy including image processing, contouring, planning, quality assurance, motion management, and outcome prediction. By summarizing cutting-edge findings and challenges, we aim to inspire researchers to explore more future possibilities and accelerate the arrival of AI radiotherapy.

Domain adaptation of automated treatment planning from computed tomography to magnetic resonance

Objective. Machine learning (ML) based radiation treatment planning addresses the iterative and time-consuming nature of conventional inverse planning. Given the rising importance of magnetic resonance (MR) only treatment planning workflows, we sought to determine if an ML based treatment planning model, trained on computed tomography (CT) imaging, could be applied to MR through domain adaptation. Methods. In this study, MR and CT imaging was collected from 55 prostate cancer patients treated on an MR linear accelerator. ML based plans were generated for each patient on both CT and MR imaging using a commercially available model in RayStation 8B. The dose distributions and acceptance rates of MR and CT based plans were compared using institutional dose-volume evaluation criteria. The dosimetric differences between MR and CT plans were further decomposed into setup, cohort, and imaging domain components. Results. MR plans were highly acceptable, meeting 93.1% of all evaluation criteria compared to 96.3% of CT plans, with dose equivalence for all evaluation criteria except for the bladder wall, penile bulb, small and large bowel, and one rectum wall criteria (p < 0.05). Changing the input imaging modality (domain component) only accounted for about half of the dosimetric differences observed between MR and CT plans. Anatomical differences between the ML training set and the MR linac cohort (cohort component) were also a significant contributor. Significance. We were able to create highly acceptable MR based treatment plans using a CT-trained ML model for treatment planning, although clinically significant dose deviations from the CT based plans were observed. Future work should focus on combining this framework with atlas selection metrics to create an interpretable quality assurance QA framework for ML based treatment planning.

Combining dense elements with attention mechanisms for 3D radiotherapy dose prediction on head and neck cancers

Purpose

External radiation therapy planning is a highly complex and tedious process as it involves treating large target volumes, prescribing several levels of doses, as well as avoiding irradiating critical structures such as organs at risk close to the tumor target. This requires highly trained dosimetrists and physicists to generate a personalized plan and adapt it as treatment evolves, thus affecting the overall tumor control and patient outcomes. Our aim is to achieve accurate dose predictions for head and neck (H&N) cancer patients on a challenging in‐house dataset that reflects realistic variability and to further compare and validate the method on a public dataset.

Methods

We propose a three‐dimensional (3D) deep neural network that combines a hierarchically dense architecture with an attention U‐net (HDA U‐net). We investigate a domain knowledge objective, incorporating a weighted mean squared error (MSE) with a dose‐volume histogram (DVH) loss function. The proposed HDA U‐net using the MSE‐DVH loss function is compared with two state‐of‐the‐art U‐net variants on two radiotherapy datasets of H&N cases. These include reference dose plans, computed tomography (CT) information, organs at risk (OARs), and planning target volume (PTV) delineations. All models were evaluated using coverage, homogeneity, and conformity metrics as well as mean dose error and DVH curves.

Results

Overall, the proposed architecture outperformed the comparative state‐of‐the‐art methods, reaching 0.95 (0.98) on D95 coverage, 1.06 (1.07) on the maximum dose value, 0.10 (0.08) on homogeneity, 0.53 (0.79) on conformity index, and attaining the lowest mean dose error on PTVs of 1.7% (1.4%) for the in‐house (public) dataset. The improvements are statistically significant (p<0.05) for the homogeneity and maximum dose value compared with the closest baseline. All models offer a near real‐time prediction, measured between 0.43 and 0.88 s per volume.

Conclusion

The proposed method achieved similar performance on both realistic in‐house data and public data compared to the attention U‐net with a DVH loss, and outperformed other methods such as HD U‐net and HDA U‐net with standard MSE losses. The use of the DVH objective for training showed consistent improvements to the baselines on most metrics, supporting its added benefit in H&N cancer cases. The quick prediction time of the proposed method allows for real‐time applications, providing physicians a method to generate an objective end goal for the dosimetrist to use as reference for planning. This could considerably reduce the number of iterations between the two expert physicians thus reducing the overall treatment planning time.

Attention-aware 3D U-Net convolutional neural network for knowledge-based planning 3D dose distribution prediction of head-and-neck cancer

PURPOSE

Deep learning-based knowledge-based planning (KBP) methods have been introduced for radiotherapy dose distribution prediction to reduce the planning time and maintain consistent high-quality plans. This paper presents a novel KBP model using an attention-gating mechanism and a three-dimensional (3D) U-Net for intensity-modulated radiation therapy (IMRT) 3D dose distribution prediction in head-and-neck cancer.

METHODS

A total of 340 head-and-neck cancer plans, representing the OpenKBP-2020 AAPM Grand Challenge data set, were used in this study. All patients were treated with the IMRT technique and a dose prescription of 70 Gy. The data set was randomly divided into 64%/16%/20% as training/validation/testing cohorts. An attention-gated 3D U-Net architecture model was developed to predict full 3D dose distribution. The developed model was trained using the mean-squared error loss function, Adam optimization algorithm, a learning rate of 0.001, 120 epochs, and batch size of 4. In addition, a baseline U-Net model was also similarly trained for comparison. The model performance was evaluated on the testing data set by comparing the generated dose distributions against the ground-truth dose distributions using dose statistics and clinical dosimetric indices. Its performance was also compared to the baseline model and the reported results of other deep learning-based dose prediction models.

RESULTS

The proposed attention-gated 3D U-Net model showed high capability in accurately predicting 3D dose distributions that closely replicated the ground-truth dose distributions of 68 plans in the test set. The average value of the mean absolute dose error was 2.972 ± 1.220 Gy (vs. 2.920 ± 1.476 Gy for a baseline U-Net) in the brainstem, 4.243 ± 1.791 Gy (vs. 4.530 ± 2.295 Gy for a baseline U-Net) in the left parotid, 4.622 ± 1.975 Gy (vs. 4.223 ± 1.816 Gy for a baseline U-Net) in the right parotid, 3.346 ± 1.198 Gy (vs. 2.958 ± 0.888 Gy for a baseline U-Net) in the spinal cord, 6.582 ± 3.748 Gy (vs. 5.114 ± 2.098 Gy for a baseline U-Net) in the esophagus, 4.756 ± 1.560 Gy (vs. 4.992 ± 2.030 Gy for a baseline U-Net) in the mandible, 4.501 ± 1.784 Gy (vs. 4.925 ± 2.347 Gy for a baseline U-Net) in the larynx, 2.494 ± 0.953 Gy (vs. 2.648 ± 1.247 Gy for a baseline U-Net) in the PTV_70, and 2.432 ± 2.272 Gy (vs. 2.811 ± 2.896 Gy for a baseline U-Net) in the body contour. The average difference in predicting the D₉₉ value for the targets (PTV_70, PTV_63, and PTV_56) was 2.50 ± 1.77 Gy. For the organs at risk, the average difference in predicting the D m a x ${D_{max}}$ (brainstem, spinal cord, and mandible) and D m e a n ${D_{mean}}$ (left parotid, right parotid, esophagus, and larynx) values was 1.43 ± 1.01 and 2.44 ± 1.73 Gy, respectively. The average value of the homogeneity index was 7.99 ± 1.45 for the predicted plans versus 5.74 ± 2.95 for the ground-truth plans, whereas the average value of the conformity index was 0.63 ± 0.17 for the predicted plans versus 0.89 ± 0.19 for the ground-truth plans. The proposed model needs less than 5 s to predict a full 3D dose distribution of 64 × 64 × 64 voxels for a new patient that is sufficient for real-time applications.

CONCLUSIONS

The attention-gated 3D U-Net model demonstrated a capability in predicting accurate 3D dose distributions for head-and-neck IMRT plans with consistent quality. The prediction performance of the proposed model was overall superior to a baseline standard U-Net model, and it was also competitive to the performance of the best state-of-the-art dose prediction method reported in the literature. The proposed model could be used to obtain dose distributions for decision-making before planning, quality assurance of planning, and guiding-automated planning for improved plan consistency, quality, and planning efficiency.

Emerging technologies in brachytherapy

Brachytherapy is a mature treatment modality. The literature is abundant in terms of review articles and comprehensive books on the latest established as well as evolving clinical practices. The intent of this article is to part ways and look beyond the current state-of-the-art and review emerging technologies that are noteworthy and perhaps may drive the future innovations in the field. There are plenty of candidate topics that deserve a deeper look, of course, but with practical limits in this communicative platform, we explore four topics that perhaps is worthwhile to review in detail at this time. First, intensity modulated brachytherapy (IMBT) is reviewed. The IMBT takes advantage ofanisotropicradiation profile generated through intelligent high-density shielding designs incorporated onto sources and applicators such to achieve high quality plans. Second, emerging applications of 3D printing (i.e. additive manufacturing) in brachytherapy are reviewed. With the advent of 3D printing, interest in this technology in brachytherapy has been immense and translation swift due to their potential to tailor applicators and treatments customizable to each individual patient. This is followed by, in third, innovations in treatment planning concerning catheter placement and dwell times where new modelling approaches, solution algorithms, and technological advances are reviewed. And, fourth and lastly, applications of a new machine learning technique, called deep learning, which has the potential to improve and automate all aspects of brachytherapy workflow, are reviewed. We do not expect that all ideas and innovations reviewed in this article will ultimately reach clinic but, nonetheless, this review provides a decent glimpse of what is to come. It would be exciting to monitor as IMBT, 3D printing, novel optimization algorithms, and deep learning technologies evolve over time and translate into pilot testing and sensibly phased clinical trials, and ultimately make a difference for cancer patients. Today's fancy is tomorrow's reality. The future is bright for brachytherapy.