Fraction-variant beam orientation optimization for non-coplanar IMRT

O’Connor, Daniel; Yu, Victoria; Nguyen, Dan; Ruan, Dan; Sheng, Ke

doi:10.1088/1361-6560/aaa94f

Cited by 22 publications

(32 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Much work over the years has been focused on reducing the treatment complexity by simplifying certain aspects in the planning workflow, such as feasibility seeking, 16 multicriteria optimization for tradeoff navigation on the Pareto surface, [17][18][19] and other algorithms for performance improvements. [20][21][22][23][24][25][26] While effective, these methods still require a large amount of intelligent input from the dosimetrist and physician, such as in weight tuning and deciding appropriate dose-volume constraints and tradeoffs.…”

Section: Introductionmentioning

confidence: 99%

Incorporating human and learned domain knowledge into training deep neural networks: A differentiable dose‐volume histogram and adversarial inspired framework for generating Pareto optimal dose distributions in radiation therapy

et al. 2019

Self Cite

View full text Add to dashboard Cite

Purpose We propose a novel domain‐specific loss, which is a differentiable loss function based on the dose‐volume histogram (DVH), and combine it with an adversarial loss for the training of deep neural networks. In this study, we trained a neural network for generating Pareto optimal dose distributions, and evaluate the effects of the domain‐specific loss on the model performance. Methods In this study, three loss functions — mean squared error (MSE) loss, DVH loss, and adversarial (ADV) loss — were used to train and compare four instances of the neural network model: (a) MSE, (b) MSE + ADV, (c) MSE + DVH, and (d) MSE + DVH+ADV. The data for 70 prostate patients, including the planning target volume (PTV), and the organs at risk (OAR) were acquired as 96 × 96 × 24 dimension arrays at 5 mm3 voxel size. The dose influence arrays were calculated for 70 prostate patients, using a 7 equidistant coplanar beam setup. Using a scalarized multicriteria optimization for intensity‐modulated radiation therapy, 1200 Pareto surface plans per patient were generated by pseudo‐randomizing the PTV and OAR tradeoff weights. With 70 patients, the total number of plans generated was 84 000 plans. We divided the data into 54 training, 6 validation, and 10 testing patients. Each model was trained for a total of 100,000 iterations, with a batch size of 2. All models used the Adam optimizer, with a learning rate of 1 × 10−3. Results Training for 100 000 iterations took 1.5 days (MSE), 3.5 days (MSE+ADV), 2.3 days (MSE+DVH), and 3.8 days (MSE+DVH+ADV). After training, the prediction time of each model is 0.052 s. Quantitatively, the MSE+DVH+ADV model had the lowest prediction error of 0.038 (conformation), 0.026 (homogeneity), 0.298 (R50), 1.65% (D95), 2.14% (D98), and 2.43% (D99). The MSE model had the worst prediction error of 0.134 (conformation), 0.041 (homogeneity), 0.520 (R50), 3.91% (D95), 4.33% (D98), and 4.60% (D99). For both the mean dose PTV error and the max dose PTV, Body, Bladder and rectum error, the MSE+DVH+ADV outperformed all other models. Regardless of model, all predictions have an average mean and max dose error <2.8% and 4.2%, respectively. Conclusion The MSE+DVH+ADV model performed the best in these categories, illustrating the importance of both human and learned domain knowledge. Expert human domain‐specific knowledge can be the largest driver in the performance improvement, and adversarial learning can be used to further capture nuanced attributes in the data. The real‐time prediction capabilities allow for a physician to quickly navigate the tradeoff space for a patient, and produce a dose distribution as a tangible endpoint for the dosimetrist to use for planning. This is expected to considerably reduce the treatment planning time, allowing for clinicians to focus their efforts on the difficult and demanding cases.

show abstract

Section: Introductionmentioning

confidence: 99%

Incorporating human and learned domain knowledge into training deep neural networks: A differentiable dose‐volume histogram and adversarial inspired framework for generating Pareto optimal dose distributions in radiation therapy

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…The DNN learns its behavior from the full implementation of one iteration of CG, the CP‐DL‐ST processes; therefore, the DNN also tries to predict the best beam orientation that can be added to the current state of the problem. Note that as a greedy algorithm, CG does not guarantee global optimality, but it has been shown to produce superior results for treatment planning problems in existing literature …”

Section: Discussionmentioning

confidence: 99%

“…Modern BOO methods typically solve the problem in the radiation dose domain, which requires precomputing the dose influence matrices for all candidate beam orientations and then solving the FMO. The final objective function in these works is usually a function of the differences between the actual and prescribed dosage received by healthy tissues, organs at risk (OARs), and the PTV . But computing the dose influence matrices and the FMO are both very complex and time intensive operations, taking hours to compute dose influence matrices and minutes to solve the FMO, which ultimately hampers the implementation of BOO in clinical routines .…”

Section: Introductionmentioning

confidence: 99%

“…The optimization algorithm used to find the optimal set of beams to train the model is a greedy iterative algorithm known as column generation (CG). Column generation algorithms such as the one in this study have demonstrated success in finding the best treatment plan . The proposed CG method starts with an empty set of beams, and in each iteration, CG selects one beam with the highest potential to improve the current solution and adds it to the set, and the respective FMO is solved.…”

Section: Introductionmentioning

confidence: 99%

“…The final objective function in these works is usually a function of the differences between the actual and prescribed dosage received by healthy tissues, organs at risk (OARs), and the PTV. [16][17][18][19][20][21][22][23][24][25][26][27][28] But computing the dose influence matrices and the FMO are both very complex and time intensive operations, 29 taking hours to compute dose influence matrices and minutes to solve the FMO, which ultimately hampers the implementation of BOO in clinical routines. 13,20,24,[30][31][32] Breedveld et al 13 used a "wish list" to prioritize constraints and objective functions related to OARs and the tumor and to iteratively add new beam orientations to the set of currently selected beams.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A fast deep learning approach for beam orientation optimization for prostate cancer treated with intensity‐modulated radiation therapy

et al. 2020

Self Cite

View full text Add to dashboard Cite

Purpose Beam orientation selection, whether manual or protocol‐based, is the current clinical standard in radiation therapy treatment planning, but it is tedious and can yield suboptimal results. Many algorithms have been designed to optimize beam orientation selection because of its impact on treatment plan quality, but these algorithms suffer from slow calculation of the dose influence matrices of all candidate beams. We propose a fast beam orientation selection method, based on deep learning neural networks (DNN), capable of developing a plan comparable to those developed by the state‐of‐the‐art column generation (CG) method. Our model's novelty lies in its supervised learning structure (using CG to teach the network), DNN architecture, and ability to learn from anatomical features to predict dosimetrically suitable beam orientations without using dosimetric information from the candidate beams. This may save hours of computation. Methods A supervised DNN is trained to mimic the CG algorithm, which iteratively chooses beam orientations one‐by‐one by calculating beam fitness values based on Karush‐Kush‐Tucker optimality conditions at each iteration. The DNN learns to predict these values. The dataset contains 70 prostate cancer patients — 50 training, 7 validation, and 13 test patients — to develop and test the model. Each patient’s data contains 6 contours: PTV, body, bladder, rectum, and left and right femoral heads. Column generation was implemented with a GPU‐based Chambolle‐Pock algorithm, a first‐order primal‐dual proximal‐class algorithm, to create 6270 plans. The DNN trained over 400 epochs, each with 2500 steps and a batch size of 1, using the Adam optimizer at a learning rate of 1 × 10−5 and a sixfold cross‐validation technique. Results The average and standard deviation of training, validation, and testing loss functions among the six folds were 0.62 ± 0.09%, 1.04 ± 0.06%, and 1.44 ± 0.11%, respectively. Using CG and supervised DNN, we generated two sets of plans for each scenario in the test set. The proposed method took at most 1.5 s to select a set of five beam orientations and 300 s to calculate the dose influence matrices for 5 beams and finally 20 s to solve the fluence map optimization (FMO). However, CG needed around 15 h to calculate the dose influence matrices of all beams and at least 400 s to solve both the beam orientation selection and FMO problems. The differences in the dose coverage of PTV between plans generated by CG and by DNN were 0.2%. The average dose differences received by organs at risk were between 1 and 6 percent: Bladder had the smallest average difference in dose received (0.956 ± 1.184%), then Rectum (2.44 ± 2.11%), Left Femoral Head (6.03 ± 5.86%), and Right Femoral Head (5.885 ± 5.515%). The dose received by Body had an average difference of 0.10 ± 0.1% between the generated treatment plans. Conclusions We developed a fast beam orientation selection method based on a DNN that selects beam orientations in seconds and is therefore suitable for clinical routines. In the ...

show abstract

Automated 4π radiotherapy treatment planning with evolving knowledge‐base

et al. 2019

Self Cite

View full text Add to dashboard Cite

Purpose Non‐coplanar 4π radiotherapy generalizes intensity modulated radiation therapy (IMRT) to automate beam geometry selection but requires complicated hyperparameter tuning to attain superior plan quality, which can be tedious and inconsistent. In this study, a fully automated 4π treatment planning was developed using evolving knowledge‐base (EKB) planning guided by dose prediction. Methods Twenty 4π lung and twenty 4π head and neck (HN) cases were included. A statistical voxel dose learning model was initially trained on low‐quality plans created using generic hyperparameter templates without manual tuning. To improve the automated plan quality without being limited by the training data quality, a new 4π optimization problem was formulated to include a one‐sided penalty on the organ‐at‐risk (OAR) dose deviation from the predicted dose. This directional OAR penalty encourages superior OAR sparing. The fast iterative shrinkage‐thresholding algorithm (FISTA) was used to solve the large‐scale beam orientation optimization problem. With the improved plans, new predictions were created to guide the next loop of EKB planning for a total of 10 loops. Plan quality was evaluated using a plan quality metric (PQM) points system based on clinical dose constraints and compared with automated planning approaches guided by manual high‐quality plans using all non‐coplanar beams, automated plans using individually evolved targeted dose, and manually created 4π plans. Results For the lung cases, the final EKB plans had significantly higher PQM than manually created 4π (+2.60%). The improvements plateaued after the third loop. The final HN EKB plans and manually created 4π plans had comparable PQMs, but had lower PQM compared to automated plans using a high‐quality training set (−3.00% and −4.44%, respectively). The PQM consistently increased up to the sixth loop. Individually evolved plans were able to improve the plan quality from initial condition due to the one‐sided cost function but the 60% of them were trapped in undesired local minima that were substantially worse than their corresponding EKB plans. Conclusion Evolving knowledge‐base planning is a novel automated planning technique guided by the predicted three‐dimensional dose distribution, which can evolve from low‐quality plans. EKB allows new beams to be used in the automated planning workflow for superior plan quality.

show abstract

Fraction-variant beam orientation optimization for non-coplanar IMRT

Cited by 22 publications

References 48 publications

Incorporating human and learned domain knowledge into training deep neural networks: A differentiable dose‐volume histogram and adversarial inspired framework for generating Pareto optimal dose distributions in radiation therapy

Incorporating human and learned domain knowledge into training deep neural networks: A differentiable dose‐volume histogram and adversarial inspired framework for generating Pareto optimal dose distributions in radiation therapy

A fast deep learning approach for beam orientation optimization for prostate cancer treated with intensity‐modulated radiation therapy

Automated 4π radiotherapy treatment planning with evolving knowledge‐base

Contact Info

Product

Resources

About