Cardiac Sparing with Volumetric Modulated Arc Therapy Enabled Total Body Irradiation (CS VMAT-TBI)

PURPOSE/OBJECTIVE(S)

Volumetric modulated arc therapy (VMAT) enabled total body irradiation (TBI) has replaced conventional TBI in our institution given the improved treatment accuracy, patient comfort, and dose modulation ability. The risk of cardiovascular disease is several folds higher among transplant patients who receive TBI, likely related to dose to the heart. We hypothesize that a cardiac-sparing (CS) VMAT-TBI technique is feasible and can meaningfully reduce dose to the heart while still adequately covering nearby lymphatic tissue.

MATERIALS/METHODS

VMAT-TBI is delivered via multi-isocentric external beams in a frame-based setup. Heart is contoured as per published guidelines. A lymph node contour, which includes tonsils, neck nodal stations, mediastinal, abdominal, retroperitoneal, and pelvic nodes is created. Coverage of the lymph node contour is prioritized over organ-sparing during inverse optimization; with a goal of V90% greater than 99.5% and mean dose less than 800 cGy for the lymph nodes and heart, respectively. An IRB-approved retrospective review was performed with mean heart dose collected for all patients treated with CS VMAT-TBI and compared to a representative cohort of five patients treated with VMAT-TBI without cardiac sparing.

RESULTS

Thirty-one patients were treated with CS VMAT-TBI between 2020-2022 with a median follow up time of 11.5 months. Mean heart dose was 796 ± 71 cGy in the CS VMAT-TBI compared to 1247 ± 29 cGy in the VMAT-TBI group without cardiac sparing (p < 0.001). Of those treated with CS VMAT-TBI, three patients relapsed; one relapse occurred in bone marrow only, one relapse occurred in bone marrow and cervical, thoracic, and intra-abdominal lymphoid tissues, and one patient was simulated but never received induction therapy due to overt progression. 100-day relapse-free survival and overall survival were 82.5% and 86.2%, respectively. Median survival time has not been met.

CONCLUSION

Cardiac sparing is feasible in VMAT-TBI and is associated with significant decrease in mean heart dose of ∼450 cGy. This is estimated to confer a 33.3% decreased absolute risk for lifetime major coronary events compared to patients treated with VMAT-TBI without cardiac sparing. Although limited by short follow-up time, there does not appear to be a significant risk for early relapse despite de-escalating cardiac tissue, likely due to prioritizing coverage of lymph nodes. Prospective clinical studies are needed to further validate cardiac and other organ at risk sparing VMAT-TBI techniques.

Towards Biology-Guided Radiotherapy Planning and Delivery on a Novel O-Ring PET-Linac Platform: Extended Beyond Bone and Lung Lesions

PURPOSE/OBJECTIVE(S)

Biology-guided radiotherapy (BgRT) with FDG signal collected via an on-board positron emission tomography (PET) system integrated in an O-ring gantry Linac was recently cleared by the FDA for lung and bone lesions. This study aims to determine if BgRT plans, guided via PET signal, are clinically acceptable for FDG-avid lesions in disease sites beyond bone and lung.

MATERIALS/METHODS

Ten patients previously treated for lesions in the liver, head and neck (HN), pancreas, renal and pelvic-abdominal lymph nodes were identified. Diagnostic PET/CT images of these treatment sites were first collected and processed/converted to mimic PET images that are acquired on PET-Linac and would be used to guide the delivery. For BgRT planning, the PTV was generated with 5 mm margin from GTV and a Biology Tracking Zone was generated including the anticipated full range of target motion. BgRT plans, guided by the emulated PET signal, were generated with 46Gy in 3 fractions for liver and 40Gy in 5 fractions for all other sites. BgRT plan deliverability was first assessed by evaluating the Activity Concentration (AC) and Normalized Target Signals (NTS) on converted PET images with the goal to meet NTS >2 (hard constraint) and AC >5kBq/ml (goal). BgRT plan quality was then evaluated with institutional guidelines on PTV coverage, OAR doses, conformity index (CI) and Heterogeneity index (HI).

RESULTS

BgRT plans were successfully generated for 11 target lesions among ten patients. The average diagnostic PET SUV, derived NTS and AC on converted PET images were 12.62, 9.33 and 12.10 kBq/ml, respectively. All images met the NTS constraints, and 8/11 plans met the AC goal for deliverability. All plans met the OAR hard constraints such as max dose on duodenum, small bowel, large bowel and spinal cord. Five of 11 plans had a limiting GI structure that resulted in an expected reduction in PTV coverage with an average PTV V100% = 77.9%, CI of 1.4, HI of 1.36 and max dose of 133.8%. The other 6 of 11 cases met the PTV V100% = 95%, had an average CI of 1.1, HI of 1.28 and Dmax of 127.67%. The estimated average time for BgRT delivery was 17 mins 25 secs. Although these plan parameters are deemed to be clinically acceptable, heterogeneity was detected inside the target region and suboptimal dose fall off was observed in some cases that may be caused by current implementation.

CONCLUSION

This preliminary study showed that BgRT plans were generated successfully with emulated PET images on 11 treatment sites covering HN, abdominal and pelvic regions. All plans met NTS constraints and 8 out of 11 met AC goals for deliverability. The plan quality of all BgRT plans were clinically acceptable based on institutional constraints. Further investigations are required to test more patients/sites for BgRT plan feasibility. Dosimetric benefit from margin reduction of BgRT target should also be investigated in future study.

Simulation-Omitted Replan with Cone Beam Computed Tomography based Adaptive Online Radiotherapy System - Transferring Adapted Plan to Non-Adaptive Ring Gantry Linear Accelerator for Image Guided Radiotherapy

PURPOSE/OBJECTIVE(S)

Artificial intelligence powered cone beam computed tomography (CBCT) based online adaptive radiotherapy (oART) system offers a streamlined and efficient process for daily ART as the default. In our prior work, we developed a workflow to utilize this oART system as simulation-omitted replan platform and treat the adapted plan on the oART system with image guided radiotherapy (IGRT) until next adaptation. However, the IGRT fractions will occupy the treatment slots of the machine. In this work, we aim to develop a semi-automatic workflow to allow the adapted plan to be treat on the non-adaptive ring gantry linear accelerator (non-ART Linac) and dedicate the oART system for adaptive treatments.

MATERIALS/METHODS

The oART system and the non-ART Linac were machine-matched to the same representative beam data. In the oART system, the initial plan is setup as 'adaptive' treatment and patients are only treated on the oART system for adaptive replan. The IGRT fractions are all treated on the non-ART Linac. An API script was developed to automatically (1) grab the adapted DICOM plan files from the secondary calculation system and write directly back to the database of the treatment management system (TMS), (2) change the DICOM tags to make the files compatible in the TMS system, (3) insert the kV-CBCT field to make the plan deliverable in the non-ART Linac. There are minimum remaining manual steps to setup the number of fractions to the intended number of IGRT fractions and to link plan to the prescription in TMS. We compare the required resources and the percentage of ART treatments on the oART system before and after the implementation of the proposed workflow to quantify the improvement of service.

RESULTS

The proposed workflow and automation eliminated the need to convert between IGRT/ART fractions in the Ethos system and reduced manual work by 25 minutes each adapted plan transfer. Table 1 summarizes the number of physics tasks and the percentage of ART fractions in oART system per month before and after the proposed workflow. This workflow reduced the physics IGRT/ART tasks from 107±31 to 65±21 tasks per month (p<0.05). Percentage of ART treatments on oART system increase from 30%±3% to 57%±13% (p<0.05). We also observed increased utilization of ART from 46% in the 1st month to 71% in the 6th month since it is easier to find a feasible time slot for the clinical team. The majority of the remaining IGRT on oART system are lung SBRT where the first fraction is not adapted due to being within a week of the simulation.

CONCLUSION

Leveraging CBCT based ART system as replan platform and non-ART Linac as IGRT platform is clinically feasible. This process significantly improved the turnaround time for replan, reduced the required resource and promotes the utilization of oART.

Intelligent Interactive Deformable Image Registration for Online Adaptive Radiotherapy

PURPOSE/OBJECTIVE(S)

The goal of this study is to streamline the time-consuming contouring process in online adaptive radiotherapy (ART) by utilizing a deep learning-based interactive deformable image registration (DIR) algorithm. The objective is to minimize manual review and editing of automatically generated initial contours of organs-at-risk (OARs) and targets, thereby improving the efficiency and effectiveness of the treatment process.

MATERIALS/METHODS

Our proposed method reforms the current DIR-based contour propagation method in clinical practice through the implementation of a deep learning-based interactive approach. The steps include: 1) generation of an initial deformable vector field (DVF) using a DL model, based on fixed and moving image pairs, resulting in the initial contours of OARs and targets; 2) clinician review/edit one the OAR/target contours as needed; 3) updated contour is sent to DL model to update the DVF and the remaining OARs/targets contours. Repeat this process until satisfactory contour qualities are achieved. We used the Open Access Series of Imaging Studies (OASIS) as the testbed, including 394 (train) and 20 (test) brain T1-weighted MRI scans, each containing 35 annotated organs. The U-Net architecture was employed to update the DVF from fixed/moving images, initial contours, and updated contours. We compared our approach to traditional manual editing without interaction and quantified the effort reduction using the added path length (APL) metric which is supposed to be proportional to the absolute time spent on the contour editing. We conducted paired t-test to show the significance. For comparison purpose, we assumed the clinicians edit the contours with the largest APL, i.e., the contours that require the most editing efforts.

RESULTS

The editing effort, as measured by APL, was reduced by 18.5% to 25.4% with a mean of 23.3%, median of 23.6%, and standard deviation of 1.9%. The significance of the results was confirmed with a p-value of 1.47e-24.

CONCLUSION

Our study demonstrates a significant reduction in editing effort, as measured by APL, compared to traditional manual contour editing. These results demonstrate the potential of our deep learning-based interactive approach to improve the efficiency and accuracy of the contouring process in clinical practice.

Deep Learning-Based Quality Assurance for Auto-Segmentation Masks in Radiotherapy

PURPOSE/OBJECTIVE(S)

Deep learning-based auto-contouring has shown great promise in several disease sites including GU and head and neck. However, quality assurance (QA) is key to identify poor auto-contours which is time consuming. We hypothesis that training a deep learning model to predict contour quality metrics, such as Dice coefficients (DSC) and associated uncertainties for QA.

MATERIALS/METHODS

We trained a 3D U-Net-based DL model for segmenting the target and three clinical-relevant OARs (bladder and rectum). To mimic the slice-by-slice review process in clinical practice, we then trained a 2D ResNet-based DL model to predict the 2D DSC for each 2D slice's contour, generated by the 3D segmentation model. Using the Monte Carlo dropout technique, we made 20 independent predictions per slice, with the final DSC calculated as their average and uncertainty estimated as 95% prediction intervals (PI). The study cohort consisted of 912 prostate cancer patients who received definitive radiotherapy. The 3D auto-segmentation model was trained on 129 patients and validated on 20, before being tested on 763 patients. The 2D DSC prediction model was trained on 293 patients with 11116 slices, validated on 73 patients with 2804 slices, and tested on 366 patients with 14117 slices. Rectum was chosen to test the 2D contour QA model as it is the most challenging OAR. We categorized 2D slices into three groups based on the lower and upper bounds of the prediction intervals. "no/minor edits" (lower bound > = 0.9), "major edits" (lower bound < 0.9 and upper bound > = 0.8), and "not acceptable" (upper bound < 0.8). The model performance was quantified by calculating correlation coefficients between predicted and ground truth DSC and the fraction of cases that were correctly identified in each category.

RESULTS

The results of the study showed that the overall correlation coefficient between predicted, and ground truth DSC was 0.842. The model was able to correctly identify 78.3%, 60.7%, and 53.4% of the "no/minor edits", "major edits", and "not acceptable" cases, respectively.

CONCLUSION

This study provides a valuable tool for clinicians in making quick decisions on the acceptance, rejection, or revision of auto-segmented masks during the radiation therapy planning process by providing quantitative results on predicted DSC and associated uncertainties.

Dosimetric Comparison of Adaptive Radiotherapy Modalities for Stereotactic Partial Breast Irradiation

PURPOSE/OBJECTIVE(S)

An increase in the availability of adaptive radiotherapy (ART) platforms have proven to be effective in the treatment of a variety of sites. In this study, we aim to evaluate the effectiveness of non-adaptive RT and 3 different ART platforms: (1) CBCT-based, (2) CT-based, and (3) MRI-based for stereotactic partial breast irradiation (SPBI).

MATERIALS/METHODS

Data were collected from 32 patients (16 left and 16 right breast) treated at a single institution. 16 patients (8 left and 8 right) treated using the non-ART platform were re-planned onto two different ART platforms, CBCT- and MRI-based. The remaining 16 patients treated using CT-based adaptive platform were not re-planned due to the prone patient treatment position (others systems supine). All cases were planned to 30 Gy in 5 fractions. Plan quality was evaluated based on pre-defined planning goals to the OARS: ipsilateral and contralateral lungs (Dmean, Dmax, V20 Gy, V9 Gy), ipsilateral (V15 Gy, V30 Gy) and contralateral breasts (Dmax), heart (Dmean, Dmax, V3 Gy, V1.5 Gy), skin (Dmax, V36.5 Gy), and rib (Dmax, V30 Gy). Target goals were defined by Dmax, Dmin, gradient index, and paddock conformality index. Re-planned cases were compared within the cohort using a paired t-test and a 2-sided t-test was used comparing to the CT-based platform.

RESULTS

Comparing the left and right breast cohort across all platforms, the CT-based ART system showed a signification dose reduction in Dmean (p<0.001 for all platforms), Dmax (p<0.001 for left breast, p<0.03 for right breast) and V9 Gy (p<0.004 for left breast, p<0.001 for right breast) to the ipsilateral lung, V15 Gy (p<0.004 for left breast cohort) to the ipsilateral breast, and Dmax to the contralateral breast (p<0.001) and ribs (p = 0.01, p<0.001, p = 0.01 for CBCT-ART, MRI-ART, and non-ART for left breast cohort only). On average, the MR-Linac platform showed the least degree of OAR sparing across nearly all dosimetric parameters evaluated when compared to all modalities, especially for contralateral lung Dmean and Dmax (p<0.05 for all dosimetric parameters for all platforms) and contralateral breast Dmax (p<0.003 for all platforms). The CBCT-based platform showed superior dose reduction in contralateral lung mean (p<0.03 for all platforms) and heart Dmean (p = 0.065, p<0.001, p = 0.045 for non-adaptive, MRI-ART, and CT-ART for left breast and p<0.008 for right breast). PTV coverage was comparable across all platforms, averaging at approximately 95%. The CT-based ART platform showed a significantly reduced gradient index relative to the CBCT- and MRI-based platforms (p<0.001).

CONCLUSION

For SPBI treatments, the CT-based ART platforms displayed a higher degree of OAR sparing for many of the dosimetric parameters recorded relative to the other ART and non-ART platforms presented. The MRI-based system typically showed less reduced OAR sparing; however, the advantage of the system is shown if soft tissue contrast is needed. PTV coverage remained comparable across all platforms.

Real-Time 3D Liver Tumor Localization via Combined Optical Surface Imaging and an X-Ray Projection from Arbitrary Imaging Angles

PURPOSE/OBJECTIVE(S)

A deep learning (DL)-based, deformable registration-driven liver tumor localization technique was developed for onboard deformable motion tracking. The technique (Surf-X360-Bio) uses in-room optical surface imaging and an x-ray projection at an arbitrary scan angle to solve volumetric liver and liver tumor motion in real-time.

MATERIALS/METHODS

Surf-X360-Bio solves the volumetric motion of the liver and localizes the liver tumor, through deforming liver and liver tumor meshes segmented on prior 4D-CTs/CBCTs. It uses real-time onboard information from an optical surface image and a simultaneously-acquired x-ray projection (from an arbitrary scan angle). Surf-X360-Bio localizes tumors via two steps: liver boundary motion estimation and intra-liver motion derivation. Surf-X360-Bio first estimates liver boundary motion by a patient-specific surface imaging model (Surf), utilizing the correlation between the external body surface motion and the internal liver boundary motion. As the correlation can be imperfect, the residual motion estimation errors were corrected by a patient-specific, angle-agnostic x-ray imaging model (X360). X360 deformed the liver boundary to match motion-related imaging features encoded in an arbitrarily-angled x-ray projection, using the Surf output for initialization. X360 adopted a geometry-aware learning module to extract and adapt to angle-varying features of arbitrarily-angled x-ray images, by calculating the projection system matrix of each x-ray image on-the-fly during model training and inference. After the liver boundary motion estimation by Surf and X360, intra-liver deformation was solved by a biomechanical model (Bio) to propagate the liver boundary motion inside to localize the tumors. The DL-based Bio model used domain knowledge of tissue biomechanics and finite element analysis (FEA) to solve intra-liver motion, with a much faster speed than conventional FEA methods to meet the real-time constraint. Surf-X360-Bio was evaluated using a dataset of 34 liver patients. Liver tumor localization accuracy was evaluated by the Dice similarity coefficient (DSC), the 95-percentile Hausdorff distance (HD95), and the center-of-mass error (COME).

RESULTS

Using 3,306 motion scenarios spanning the 360 degree x-ray scan angles for each testing patient, Surf-X360-Bio localized the liver tumors to 0.81 (mean) ± 0.16 (s.d.) in DSC, 2.5 ± 1.7 mm in HD95, and 2.1 ± 1.8 mm in COME. In comparison, the prior reference image without deformable registration-driven localization yielded 0.42 ± 0.29 in DSC, 8.1 ± 5.2 mm in HD95, and 8.5 ± 5.2 mm in COME. Via Surf-X360-Bio, the overall inference time was below 230 ms for each case.

CONCLUSION

Combining optical surface imaging and x-ray imaging, Surf-X360-Bio achieved fast (<250 ms inference time), accurate (mean error < 2.1 mm), and robust liver tumor localizations at arbitrary x-ray scan angles for real-time, marker-less 3D deformable motion tracking.

Cancer Informatics for Cancer Centers: Sharing Ideas on How to Build an Artificial Intelligence-Ready Informatics Ecosystem for Radiation Oncology

In August 2022, the Cancer Informatics for Cancer Centers brought together cancer informatics leaders for its biannual symposium, Precision Medicine Applications in Radiation Oncology, co-chaired by Quynh-Thu Le, MD (Stanford University), and Walter J. Curran, MD (GenesisCare). Over the course of 3 days, presenters discussed a range of topics relevant to radiation oncology and the cancer informatics community more broadly, including biomarker development, decision support algorithms, novel imaging tools, theranostics, and artificial intelligence (AI) for the radiotherapy workflow. Since the symposium, there has been an impressive shift in the promise and potential for integration of AI in clinical care, accelerated in large part by major advances in generative AI. AI is now poised more than ever to revolutionize cancer care. Radiation oncology is a field that uses and generates a large amount of digital data and is therefore likely to be one of the first fields to be transformed by AI. As experts in the collection, management, and analysis of these data, the informatics community will take a leading role in ensuring that radiation oncology is prepared to take full advantage of these technological advances. In this report, we provide highlights from the symposium, which took place in Santa Barbara, California, from August 29 to 31, 2022. We discuss lessons learned from the symposium for data acquisition, management, representation, and sharing, and put these themes into context to prepare radiation oncology for the successful and safe integration of AI and informatics technologies.

NRG-DT001 phase Ib trial of neoadjuvant navtemadlin (previously AMG232 and KRT232) concurrent with preoperative radiotherapy in wild-type p53 soft tissue sarcoma of the extremity and body wall

11521

Background: NRG-DT001 is a phase Ib trial evaluating neoadjuvant navtemadlin with preoperative radiation therapy (RT) in patients (pts) with wild-type (WT) p53 soft tissue sarcoma (STS). The primary objective is to evaluate the safety and tolerability of the MDM2 inhibitor navtemadlin in combination with standard-dose RT in STS in two cohorts (A, extremity or body wall; B, abdomen/pelvis/retroperitoneum) to determine the maximum tolerated dose/recommended phase II dose (MTD/RP2D) of navtemadlin in combination with RT. This report contains the results for cohort A. Methods: Eligible pts had grade 2-3 STS ≥ 5 cm, age ≥ 18, and Zubrod performance status 0-1. Dose levels were 120 mg 2x/week (DL-1), 120 mg 3x/week (DL1), 4x/week (DL2), and 5x/week (DL3) 1 week prior to and during RT (50Gy/5 weeks). Surgery was 5-8 weeks after RT. A 3+3 design was used to make dose escalation/de-escalation decisions at each dose level. Five additional pts were enrolled to the MTD to ensure safety (expansion cohort) with a dose limiting toxicity (DLT) rate of ≤ 1/5 considered safe. The DLT observation period was from the start of navtemadlin until 4 weeks after completion of drug+RT. Tumor Tp53 mutation status was determined by NGS sequencing. All eligible and treated p53 WT pts who experienced DLT or completed the observation period were considered DLT-evaluable. DLT included all grade 4-5 AE definitely, probably, or possibly related to navtemadlin. Any grade 3 AE definitely, probably, or possibly related to navtemadlin was also considered DLT if any of the 2 following situations occurred: a delay of treatment > 2 weeks or ≥ 2 dose reductions due to the grade 3 AE. The decision to escalate or de-escalate was made by consensus of the study team in accordance with the protocol. Results: Between 11/3/2017 and 9/10/2021, 4 (3 WT), 7 (4 WT) and 7 (4 WT) pts were enrolled at DL1, DL2, and DL3 respectively. An additional 9 (5 WT) pts were enrolled on DL3 expansion cohort. Preoperative RT was completed for all except 1 pt (pt refusal/DL3). On DL1 and DL2, 100% of pts completed navtemadlin. On DL3 (including expansion cohort), 78% (7/9) completed navtemadlin (1 AE, 1 pt refusal). On DL1, DL2, and DL3, 3/3, 3/4 (1 disease progression), and 5/6 (1 consent withdrawal; 3 pending) completed surgery. There were no DLTs in any dose level (DL1 0/3, DL2 0/4, DL3 0/9), establishing DL3 as the MTD/RP2D. Tumor necrosis rates will be reported at the time of presentation. Conclusions: Neoadjuvant navtemadlin concurrent with standard dose preoperative RT is well tolerated in patients with WT p53 STS at extremity or body wall, and the 120 mg PO daily of navtemadlin, 5 days per week dose should be used to design future trials of RT with extremity STS. Incorporating NGS sequencing results as an integral biomarker in a clinical trial of neoadjuvant radiotherapy and a radiosensitizer is feasible. Clinical trial information: NRG-DT001 NCT03217266.

Synthesizing CT images from MR images with deep learning: model generalization for different datasets through transfer learning

PURPOSE

Synthetic CT generation is the focus of many studies, however, only models on data with the same dataset were tested. Therefore, how well the trained model will work for data from different hospitals and MR protocols is still unknown. In this study, we addressed the model generalization problem for the MR-to-CT conversion task.

METHODS

Brain T2 MR and corresponding CT images were collected from one hospital and brain T1-FLAIR, T1-POST MR, and corresponding CT images were collected from another hospital. To investigate the model's generalizability ability, four potential solutions were proposed: source model, target model, combined model, and adapted model. All models were trained using the CycleGAN network. The source model was trained with a source domain dataset from scratch and tested with a target domain dataset. The target model was trained with a target domain dataset and tested with a target domain dataset. The combined model was trained with both source domain and target domain datasets, and tested with the target domain dataset. The adapted model used a transfer learning strategy to train a CycleGAN model with a source domain dataset and retrain the pre-trained model with a target domain dataset. MAE, RMSE, PSNR, and SSIM were used to quantitatively evaluate model performance on a target domain dataset.

RESULTS

The adapted model achieved best quantitative results of 74.56±8.61, 193.18±17.98, 28.30±0.83, and 0.84±0.01 for MAE, RMSE, PSNR, and SSIM using the T1-FLAIR dataset and 74.89±15.64, 195.73±31.29, 27.72±1.43, and 0.83±0.04 for MAE, RMSE, PSNR, and SSIM using the T1-POST dataset. The source model had the poorest performance.

CONCLUSIONS

This work indicates high generalization ability to generate synthetic CT images from small training datasets of MR images using pre-trained CycleGAN. The quantitative results of the test data, including different scanning protocols and different acquisition centers, indicated the proof of this concept.

Semi-automatic sigmoid colon segmentation in CT for radiation therapy treatment planning via an iterative 2.5-D deep learning approach

Automatic sigmoid colon segmentation in CT for radiotherapy treatment planning is challenging due to complex organ shape, close distances to other organs, and large variations in size, shape, and filling status. The patient bowel is often not evacuated, and CT contrast enhancement is not used, which further increase problem difficulty. Deep learning (DL) has demonstrated its power in many segmentation problems. However, standard 2-D approaches cannot handle the sigmoid segmentation problem due to incomplete geometry information and 3-D approaches often encounters the challenge of a limited training data size. Motivated by human's behavior that segments the sigmoid slice by slice while considering connectivity between adjacent slices, we proposed an iterative 2.5-D DL approach to solve this problem. We constructed a network that took an axial CT slice, the sigmoid mask in this slice, and an adjacent CT slice to segment as input and output the predicted mask on the adjacent slice. We also considered other organ masks as prior information. We trained the iterative network with 50 patient cases using five-fold cross validation. The trained network was repeatedly applied to generate masks slice by slice. The method achieved average Dice similarity coefficients of 0.82 0.06 and 0.88 0.02 in 10 test cases without and with using prior information.

On the robustness of deep learning-based lung-nodule classification for CT images with respect to image noise

Robustness is an important aspect when evaluating a method of medical image analysis. In this study, we investigated the robustness of a deep learning (DL)-based lung-nodule classification model for CT images with respect to noise perturbations. A deep neural network (DNN) was established to classify 3D CT images of lung nodules into malignant or benign groups. The established DNN was able to predict malignancy rate of lung nodules based on CT images, achieving the area under the curve of 0.91 for the testing dataset in a tenfold cross validation as compared to radiologists' prediction. We then evaluated its robustness against noise perturbations. We added to the input CT images noise signals generated randomly or via an optimization scheme using a realistic noise model based on a noise power spectrum for a given mAs level, and monitored the DNN's output. The results showed that the CT noise was able to affect the prediction results of the established DNN model. With random noise perturbations at 100 mAs, DNN's predictions for 11.2% of training data and 17.4% of testing data were successfully altered by at least once. The percentage increased to 23.4% and 34.3%, respectively, for optimization-based perturbations. We further evaluated robustness of models with different architectures, parameters, number of output labels, etc, and robustness concern was found in these models to different degrees. To improve model robustness, we empirically proposed an adaptive training scheme. It fine-tuned the DNN model by including perturbations in the training dataset that successfully altered the DNN's perturbations. The adaptive scheme was repeatedly performed to gradually improve DNN's robustness. The numbers of perturbations at 100 mAs affecting DNN's predictions were reduced to 10.8% for training and 21.1% for testing by the adaptive training scheme after two iterations. Our study illustrated that robustness may potentially be a concern for an exemplary DL-based lung-nodule classification model for CT images, indicating the needs for evaluating and ensuring model robustness when developing similar models. The proposed adaptive training scheme may be able to improve model robustness.

Operating a treatment planning system using a deep-reinforcement learning-based virtual treatment planner for prostate cancer intensity-modulated radiation therapy treatment planning

PURPOSE

In the treatment planning process of intensity-modulated radiation therapy (IMRT), a human planner operates the treatment planning system (TPS) to adjust treatment planning parameters, for example, dose volume histogram (DVH) constraints' locations and weights, to achieve a satisfactory plan for each patient. This process is usually time-consuming, and the plan quality depends on planer's experience and available planning time. In this study, we proposed to model the behaviors of human planners in treatment planning by a deep reinforcement learning (DRL)-based virtual treatment planner network (VTPN), such that it can operate the TPS in a human-like manner for treatment planning.

METHODS AND MATERIALS

Using prostate cancer IMRT as an example, we established the VTPN using a deep neural network developed. We considered an in-house optimization engine with a weighted quadratic objective function. Virtual treatment planner network was designed to observe an intermediate plan DVHs and decide the action to improve the plan by changing weights and threshold dose in the objective function. We trained the VTPN in an end-to-end DRL process in 10 patient cases. A plan score was used to measure plan quality. We demonstrated the feasibility and effectiveness of the trained VTPN in another 64 patient cases.

RESULTS

Virtual treatment planner network was trained to spontaneously learn how to adjust treatment planning parameters to generate high-quality treatment plans. In the 64 testing cases, with initialized parameters, quality score was 4.97 (±2.02), with 9.0 being the highest possible score. Using VTPN to perform treatment planning improved quality score to 8.44 (±0.48).

CONCLUSIONS

To our knowledge, this was the first time that intelligent treatment planning behaviors of human planner in external beam IMRT are autonomously encoded in an artificial intelligence system. The trained VTPN is capable of behaving in a human-like way to produce high-quality plans.

An introduction to deep learning in medical physics: advantages, potential, and challenges

As one of the most popular approaches in artificial intelligence, deep learning (DL) has attracted a lot of attention in the medical physics field over the past few years. The goals of this topical review article are twofold. First, we will provide an overview of the method to medical physics researchers interested in DL to help them start the endeavor. Second, we will give in-depth discussions on the DL technology to make researchers aware of its potential challenges and possible solutions. As such, we divide the article into two major parts. The first part introduces general concepts and principles of DL and summarizes major research resources, such as computational tools and databases. The second part discusses challenges faced by DL, present available methods to mitigate some of these challenges, as well as our recommendations.

Beam modeling and beam model commissioning for Monte Carlo dose calculation-based radiation therapy treatment planning: Report of AAPM Task Group 157

Dose calculation plays an important role in the accuracy of radiotherapy treatment planning and beam delivery. The Monte Carlo (MC) method is capable of achieving the highest accuracy in radiotherapy dose calculation and has been implemented in many commercial systems for radiotherapy treatment planning. The objective of this task group was to assist clinical physicists with the potentially complex task of acceptance testing and commissioning MC-based treatment planning systems (TPS) for photon and electron beam dose calculations. This report provides an overview on the general approach of clinical implementation and testing of MC-based TPS with a specific focus on models of clinical photon and electron beams. Different types of beam models are described including those that utilize MC simulation of the treatment head and those that rely on analytical methods and measurements. The trade-off between accuracy and efficiency in the various source-modeling approaches is discussed together with guidelines for acceptance testing of MC-based TPS from the clinical standpoint. Specific recommendations are given on methods and practical procedures to commission clinical beam models for MC-based TPS.

Intelligent inverse treatment planning via deep reinforcement learning, a proof-of-principle study in high dose-rate brachytherapy for cervical cancer

Inverse treatment planning in radiation therapy is formulated as solving optimization problems. The objective function and constraints consist of multiple terms designed for different clinical and practical considerations. Weighting factors of these terms are needed to define the optimization problem. While a treatment planning optimization engine can solve the optimization problem with given weights, adjusting the weights to yield a high-quality plan is typically performed by a human planner. Yet the weight-tuning task is labor intensive, time consuming, and it critically affects the final plan quality. An automatic weight-tuning approach is strongly desired. The procedure of weight adjustment to improve the plan quality is essentially a decision-making problem. Motivated by the tremendous success in deep learning for decision making with human-level intelligence, we propose a novel framework to adjust the weights in a human-like manner. This study used inverse treatment planning in high-dose-rate brachytherapy (HDRBT) for cervical cancer as an example. We developed a weight-tuning policy network (WTPN) that observes dose volume histograms of a plan and outputs an action to adjust organ weighting factors, similar to the behaviors of a human planner. We trained the WTPN via end-to-end deep reinforcement learning. Experience replay was performed with the epsilon greedy algorithm. After training was completed, we applied the trained WTPN to guide treatment planning of five testing patient cases. It was found that the trained WTPN successfully learnt the treatment planning goals and was able to guide the weight tuning process. On average, the quality score of plans generated under the WTPN's guidance was improved by ~8.5% compared to the initial plan with arbitrarily set weights, and by 10.7% compared to the plans generated by human planners. To our knowledge, this was the first time that a tool was developed to adjust organ weights for the treatment planning optimization problem in a human-like fashion based on intelligence learnt from a training process, which was different from existing strategies based on pre-defined rules. The study demonstrated potential feasibility to develop intelligent treatment planning approaches via deep reinforcement learning.

MDM2 inhibitor AMG-232 and radiation therapy in treating patients with soft tissue sarcoma with wild-type TP53: A phase IB study (NRG-DT001)

TPS11076

Background: Over-expression of mdm2 is a major block to p53 activation in soft tissue sarcoma (STS). AMG-232 specifically inhibits human MDM2-p53 interactions in vitro and in vivo, thereby activating p53. Resistance mechanisms to MDM2 inhibition by AMG-232 is through accumulation of MDM2 and MDMX. However, DNA damaging agents, such as radiotherapy (RT), promote MDMX degradation and lead to sustained p53 activation, which is critically important for radiation-induced tumor cell killing in STS. Indeed, preclinical studies reveal that AMG-232 synergizes with RT in vitro and in vivo when treating STS with wild type (WT) P53 gene . Since a majority of STS harbor WT TP53 gene and RT is part of the standard of care in treating STS; it is the next rational step to test the safety and efficacy of combination of AMG-232 and RT followed by surgery. In this clinical trial, we hypothesize that preoperative AMG-232 plus RT is safe and well tolerated in 2 separate cohorts: (A) extremity and body wall; (B) abdomen/pelvis/retroperitoneum. Methods: A key feature of this trial is to incorporate TP53 NGS sequencing result as one of the eligibility criteria. Adult patients with pathologically proven grade 2-3 STS with size ≥ 5 cm are eligible to enroll if there is a planned definitive surgical resection of the primary tumor. There must be sufficient tissue to submit to central laboratory for NGS sequencing of the TP53 gene, which is an integral biomarker. Only those with tumors harboring WT p53 gene will be allowed to continue protocol treatment. The primary objective is to evaluate the safety and tolerability of AMG-232 in combination with standard-dose RT in STS in two separate cohort; as well as to determine the maximum tolerated dose (MTD) and recommended phase II dose (RP2D) of AMG-232 in combination with RT. We define DLT as grade 4-5 toxicities attributable to AMG-232 for up to 4 weeks after the completion of combination treatment (prior to surgery) as well as any AMG-232 related grade 3 toxicity that should lead to > 2 weeks treatment delay or ≥ two dose reductions. There are 3 dose levels for AMG-232: dose level 1 is 120 mg daily for 3 days per week; dose level 2 is 120 mg daily 4 days per week and dose level 5 is 120 mg daily 5 days a week. If patients did not tolerate dose level 1, AMG-232 will be de-escalated to a lower level, which is 120 mg daily 2 days a week. The trial was open since 11/2017; 3 patients have been enrolled to cohort A and 2 patients in cohort B at dose level 1. The first patient has a tumor with WT p53 gene and has completed protocol treatment safely. Clinical trial information: NCT03217266.

Intelligent Parameter Tuning in Optimization-Based Iterative CT Reconstruction via Deep Reinforcement Learning

A number of image-processing problems can be formulated as optimization problems. The objective function typically contains several terms specifically designed for different purposes. Parameters in front of these terms are used to control the relative importance among them. It is of critical importance to adjust these parameters, as quality of the solution depends on their values. Tuning parameters are a relatively straight forward task for a human, as one can intuitively determine the direction of parameter adjustment based on the solution quality. Yet manual parameter tuning is not only tedious in many cases, but also becomes impractical when a number of parameters exist in a problem. Aiming at solving this problem, this paper proposes an approach that employs deep reinforcement learning to train a system that can automatically adjust parameters in a human-like manner. We demonstrate our idea in an example problem of optimization-based iterative computed tomography (CT) reconstruction with a pixel-wise total-variation regularization term. We set up a parameter-tuning policy network (PTPN), which maps a CT image patch to an output that specifies the direction and amplitude by which the parameter at the patch center is adjusted. We train the PTPN via an end-to-end reinforcement learning procedure. We demonstrate that under the guidance of the trained PTPN, reconstructed CT images attain quality similar or better than those reconstructed with manually tuned parameters.

Prototype volumetric ultrasound tomography image guidance system for prone stereotactic partial breast irradiation: proof-of-concept

Accurate dose delivery in stereotactic partial breast irradiation (S-PBI) is challenging because of the target position uncertainty caused by breast deformation, the target volume changes caused by lumpectomy cavity shrinkage, and the target delineation uncertainty on simulation computed tomography (CT) images caused by poor soft tissue contrast. We have developed a volumetric ultrasound tomography (UST) image guidance system for prone position S-PBI. The system is composed of a novel 3D printed rotation water tank, a patient-specific resin breast immobilization cup, and a 1D array ultrasound transducer. Coronal 2D US images were acquired in 5° increments over a 360° range, and planes were acquired every 2 mm in elevation. A super-compounding technique was used to reconstruct the image volume. The image quality of UST was evaluated with a BB-1 breast phantom and BioZorb surgical marker, and the results revealed that UST offered better soft tissue contrast than CT and similar image quality to MR. In the evaluated plane, the size and location of five embedded objects were measured and compared to MR, which is considered as the ground truth. Objects' diameters and the distances between objects in UST differ by approximately 1 to 2 mm from those in MR, which showed that UST offers the image quality required for S-PBI. In future work we will develop a robotic system that will be ultimately implemented in the clinic.

Full Monte Carlo-Based Biologic Treatment Plan Optimization System for Intensity Modulated Carbon Ion Therapy on Graphics Processing Unit

PURPOSE

One of the major benefits of carbon ion therapy is enhanced biological effectiveness at the Bragg peak region. For intensity modulated carbon ion therapy (IMCT), it is desirable to use Monte Carlo (MC) methods to compute the properties of each pencil beam spot for treatment planning, because of their accuracy in modeling physics processes and estimating biological effects. We previously developed goCMC, a graphics processing unit (GPU)-oriented MC engine for carbon ion therapy. The purpose of the present study was to build a biological treatment plan optimization system using goCMC.

METHODS AND MATERIALS

The repair-misrepair-fixation model was implemented to compute the spatial distribution of linear-quadratic model parameters for each spot. A treatment plan optimization module was developed to minimize the difference between the prescribed and actual biological effect. We used a gradient-based algorithm to solve the optimization problem. The system was embedded in the Varian Eclipse treatment planning system under a client-server architecture to achieve a user-friendly planning environment. We tested the system with a 1-dimensional homogeneous water case and 3 3-dimensional patient cases.

RESULTS

Our system generated treatment plans with biological spread-out Bragg peaks covering the targeted regions and sparing critical structures. Using 4 NVidia GTX 1080 GPUs, the total computation time, including spot simulation, optimization, and final dose calculation, was 0.6 hour for the prostate case (8282 spots), 0.2 hour for the pancreas case (3795 spots), and 0.3 hour for the brain case (6724 spots). The computation time was dominated by MC spot simulation.

CONCLUSIONS

We built a biological treatment plan optimization system for IMCT that performs simulations using a fast MC engine, goCMC. To the best of our knowledge, this is the first time that full MC-based IMCT inverse planning has been achieved in a clinically viable time frame.

A deep convolutional neural network-based automatic delineation strategy for multiple brain metastases stereotactic radiosurgery

Accurate and automatic brain metastases target delineation is a key step for efficient and effective stereotactic radiosurgery (SRS) treatment planning. In this work, we developed a deep learning convolutional neural network (CNN) algorithm for segmenting brain metastases on contrast-enhanced T1-weighted magnetic resonance imaging (MRI) datasets. We integrated the CNN-based algorithm into an automatic brain metastases segmentation workflow and validated on both Multimodal Brain Tumor Image Segmentation challenge (BRATS) data and clinical patients' data. Validation on BRATS data yielded average DICE coefficients (DCs) of 0.75±0.07 in the tumor core and 0.81±0.04 in the enhancing tumor, which outperformed most techniques in the 2015 BRATS challenge. Segmentation results of patient cases showed an average of DCs 0.67±0.03 and achieved an area under the receiver operating characteristic curve of 0.98±0.01. The developed automatic segmentation strategy surpasses current benchmark levels and offers a promising tool for SRS treatment planning for multiple brain metastases.

Multistage stereotactic radiosurgery for large cerebral arteriovenous malformations using the Gamma Knife platform

PURPOSE

Radiosurgery is an established technique to treat cerebral arteriovenous malformations (AVMs). Obliteration of larger AVMs (> 10-15 cm³ or diameter > 3 cm) in a single session is challenging with current radiosurgery platforms due to toxicity. We present a novel technique of multistage stereotactic radiosurgery (SRS) for large intracranial arteriovenous malformations (AVM) using the Gamma Knife system.

MATERIALS/METHODS

Eighteen patients with large (> 10-15 cm³ or diameter > 3 cm) AVMs, which were previously treated using a staged SRS technique on the Cyberknife platform, were retrospectively selected for this study. The AVMs were contoured and divided into 3-8 subtargets to be treated sequentially in a staged approach at half to 4 week intervals. The prescription dose ranged from 15 Gy to 20 Gy, depending on the subtarget number, volume, and location. Gamma Knife plans using multiple collimator settings were generated and optimized. The coordinates of each shot from the initial plan covering the total AVM target were extracted based on their relative positions within the frame system. The shots were regrouped based on their location with respect to the subtarget contours to generate subplans for each stage. The delivery time of each shot for a subtarget was decay corrected with ⁶⁰ Co for staging the treatment course to generate the same dose distribution as that planned for the total AVM target. Conformality indices and dose-volume analysis were performed to evaluate treatment plans.

RESULTS

With the shot redistribution technique, the composite dose for the multistaged treatment of multiple subtargets is equivalent to the initial plan for total AVM target. Gamma Knife plans resulted in an average PTV coverage of 96.3 ± 0.9% and a PITV of 1.23 ± 0.1. The resulting Conformality indices, V_12Gy and R₅₀ dose spillage values were 0.76 ± 0.05, 3.4 ± 1.8, and 3.1 ± 0.5 respectively.

CONCLUSION

The Gamma Knife system can deliver a multistaged conformal dose to treat large AVMs when correcting for translational setup errors of each shot at each staged treatment.

[The efficacy and safety of ultrafiltration for patients with heart failure: results from a single-center randomized controlled study]

Objective: To evaluate the efficacy and safety of ultrafiltration in patients with heart failure. Methods: One hundred and thirty four cases of patients with heart failure, who hospitalized in our hospital from June 2010 to June 2016 were enrolled in this study. Random serial number was generated using SPSS 22.0 software, patients were then randomly divided into control group and ultrafiltration group with the proportion of 1∶1 (67 cases in each group). Patients in the control group received standard therapy. Patients in the ultrafiltration group received ultrafiltration therapy for 8 hours. Curative effect was evaluated after 8 hours treatment in the control group and after 12 hours in the ultrafiltration group. Following parameters were compared between the two groups: body weight, dyspnea score and 6 minutes walking distance as well as blood pressure, heart rate, Na(+) , K(+) , Cl(-), pH, HCO(3)(-), Hb, PLT, Cr, BUN levels. Results: (1)Two patients died during run-in process and eventually 132 cases were chosen for final analysis (65 cases in control group and 67 cases in the ultrafiltration group). Gender, age, type of heart failure, dyspnea score, body weight at baseline were similar between the two groups. (2)Post therapy, patients' body weight decreased obviously, while dyspnea score and 6 minutes walking distance increased significantly in the ultrafiltration group compared to baseline(all P<0.05), and the improvement was significantly greater compared to control group(all P<0.05). (3)The safety index comparison of two groups: blood pressure, heart rate, Na(+) , K(+) , Cl(-), pH, HCO(3)(-), Hb, PLT, Cr, and BUN were similar between the two groups at baseline and post therapy. Conclusion: Ultrafiltration therapy is safe and effective to treat patients with heart failure.

Monte Carlo dose calculations for high-dose-rate brachytherapy using GPU-accelerated processing

PURPOSE

Current clinical brachytherapy dose calculations are typically based on the Association of American Physicists in Medicine Task Group report 43 (TG-43) guidelines, which approximate patient geometry as an infinitely large water phantom. This ignores patient and applicator geometries and heterogeneities, causing dosimetric errors. Although Monte Carlo (MC) dose calculation is commonly recognized as the most accurate method, its associated long computational time is a major bottleneck for routine clinical applications. This article presents our recent developments of a fast MC dose calculation package for high-dose-rate (HDR) brachytherapy, gBMC, built on a graphics processing unit (GPU) platform.

METHODS AND MATERIALS

gBMC-simulated photon transport in voxelized geometry with physics in (192)Ir HDR brachytherapy energy range considered. A phase-space file was used as a source model. GPU-based parallel computation was used to simultaneously transport multiple photons, one on a GPU thread. We validated gBMC by comparing the dose calculation results in water with that computed TG-43. We also studied heterogeneous phantom cases and a patient case and compared gBMC results with Acuros BV results.

RESULTS

Radial dose function in water calculated by gBMC showed <0.6% relative difference from that of the TG-43 data. Difference in anisotropy function was <1%. In two heterogeneous slab phantoms and one shielded cylinder applicator case, average dose discrepancy between gBMC and Acuros BV was <0.87%. For a tandem and ovoid patient case, good agreement between gBMC and Acruos BV results was observed in both isodose lines and dose-volume histograms. In terms of the efficiency, it took ∼47.5 seconds for gBMC to reach 0.15% statistical uncertainty within the 5% isodose line for the patient case.

CONCLUSIONS

The accuracy and efficiency of a new GPU-based MC dose calculation package, gBMC, for HDR brachytherapy make it attractive for clinical applications.

Initial development of goCMC: a GPU-oriented fast cross-platform Monte Carlo engine for carbon ion therapy

Monte Carlo (MC) simulation is considered as the most accurate method for calculation of absorbed dose and fundamental physics quantities related to biological effects in carbon ion therapy. To improve its computational efficiency, we have developed a GPU-oriented fast MC package named goCMC, for carbon therapy. goCMC simulates particle transport in voxelized geometry with kinetic energy up to 450 MeV u-1. Class II condensed history simulation scheme with a continuous slowing down approximation was employed. Energy straggling and multiple scattering were modeled. δ-electrons were terminated with their energy locally deposited. Four types of nuclear interactions were implemented in goCMC, i.e. carbon-hydrogen, carbon-carbon, carbon-oxygen and carbon-calcium inelastic collisions. Total cross section data from Geant4 were used. Secondary particles produced in these interactions were sampled according to particle yield with energy and directional distribution data derived from Geant4 simulation results. Secondary charged particles were transported following the condensed history scheme, whereas secondary neutral particles were ignored. goCMC was developed under OpenCL framework and is executable on different platforms, e.g. GPU and multi-core CPU. We have validated goCMC with Geant4 in cases with different beam energy and phantoms including four homogeneous phantoms, one heterogeneous half-slab phantom, and one patient case. For each case [Formula: see text] carbon ions were simulated, such that in the region with dose greater than 10% of maximum dose, the mean relative statistical uncertainty was less than 1%. Good agreements for dose distributions and range estimations between goCMC and Geant4 were observed. 3D gamma passing rates with 1%/1 mm criterion were over 90% within 10% isodose line except in two extreme cases, and those with 2%/1 mm criterion were all over 96%. Efficiency and code portability were tested with different GPUs and CPUs. Depending on the beam energy and voxel size, the computation time to simulate [Formula: see text] carbons was 9.9-125 s, 2.5-50 s and 60-612 s on an AMD Radeon GPU card, an NVidia GeForce GTX 1080 GPU card and an Intel Xeon E5-2640 CPU, respectively. The combined accuracy, efficiency and portability make goCMC attractive for research and clinical applications in carbon ion therapy.

Accelerated Monte Carlo simulation on the chemical stage in water radiolysis using GPU

The accurate simulation of water radiolysis is an important step to understand the mechanisms of radiobiology and quantitatively test some hypotheses regarding radiobiological effects. However, the simulation of water radiolysis is highly time consuming, taking hours or even days to be completed by a conventional CPU processor. This time limitation hinders cell-level simulations for a number of research studies. We recently initiated efforts to develop gMicroMC, a GPU-based fast microscopic MC simulation package for water radiolysis. The first step of this project focused on accelerating the simulation of the chemical stage, the most time consuming stage in the entire water radiolysis process. A GPU-friendly parallelization strategy was designed to address the highly correlated many-body simulation problem caused by the mutual competitive chemical reactions between the radiolytic molecules. Two cases were tested, using a 750 keV electron and a 5 MeV proton incident in pure water, respectively. The time-dependent yields of all the radiolytic species during the chemical stage were used to evaluate the accuracy of the simulation. The relative differences between our simulation and the Geant4-DNA simulation were on average 5.3% and 4.4% for the two cases. Our package, executed on an Nvidia Titan black GPU card, successfully completed the chemical stage simulation of the two cases within 599.2 s and 489.0 s. As compared with Geant4-DNA that was executed on an Intel i7-5500U CPU processor and needed 28.6 h and 26.8 h for the two cases using a single CPU core, our package achieved a speed-up factor of 171.1-197.2.

New concept on an integrated interior magnetic resonance imaging and medical linear accelerator system for radiation therapy

Image guidance plays a critical role in radiotherapy. Currently, cone-beam computed tomography (CBCT) is routinely used in clinics for this purpose. While this modality can provide an attenuation image for therapeutic planning, low soft-tissue contrast affects the delineation of anatomical and pathological features. Efforts have recently been devoted to several MRI linear accelerator (LINAC) projects that lead to the successful combination of a full diagnostic MRI scanner with a radiotherapy machine. We present a new concept for the development of the MRI-LINAC system. Instead of combining a full MRI scanner with the LINAC platform, we propose using an interior MRI (iMRI) approach to image a specific region of interest (RoI) containing the radiation treatment target. While the conventional CBCT component still delivers a global image of the patient's anatomy, the iMRI offers local imaging of high soft-tissue contrast for tumor delineation. We describe a top-level system design for the integration of an iMRI component into an existing LINAC platform. We performed numerical analyses of the magnetic field for the iMRI to show potentially acceptable field properties in a spherical RoI with a diameter of 15 cm. This field could be shielded to a sufficiently low level around the LINAC region to avoid electromagnetic interference. Furthermore, we investigate the dosimetric impacts of this integration on the radiotherapy beam.

Moving GPU-OpenCL-based Monte Carlo dose calculation toward clinical use: Automatic beam commissioning and source sampling for treatment plan dose calculation

We have previously developed a GPU‐based Monte Carlo (MC) dose engine on the OpenCL platform, named goMC, with a built‐in analytical linear accelerator (linac) beam model. In this paper, we report our recent improvement on goMC to move it toward clinical use. First, we have adapted a previously developed automatic beam commissioning approach to our beam model. The commissioning was conducted through an optimization process, minimizing the discrepancies between calculated dose and measurement. We successfully commissioned six beam models built for Varian TrueBeam linac photon beams, including four beams of different energies (6 MV, 10 MV, 15 MV, and 18 MV) and two flattening‐filter‐free (FFF) beams of 6 MV and 10 MV. Second, to facilitate the use of goMC for treatment plan dose calculations, we have developed an efficient source particle sampling strategy. It uses the pre‐generated fluence maps (FMs) to bias the sampling of the control point for source particles already sampled from our beam model. It could effectively reduce the number of source particles required to reach a statistical uncertainty level in the calculated dose, as compared to the conventional FM weighting method. For a head‐and‐neck patient treated with volumetric modulated arc therapy (VMAT), a reduction factor of ~2.8 was achieved, accelerating dose calculation from 150.9 s to 51.5 s. The overall accuracy of goMC was investigated on a VMAT prostate patient case treated with 10 MV FFF beam. 3D gamma index test was conducted to evaluate the discrepancy between our calculated dose and the dose calculated in Varian Eclipse treatment planning system. The passing rate was 99.82% for 2%/2 mm criterion and 95.71% for 1%/1 mm criterion. Our studies have demonstrated the effectiveness and feasibility of our auto‐commissioning approach and new source sampling strategy for fast and accurate MC dose calculations for treatment plans.

Data correlation based noise level estimation for cone beam projection data

BACKGROUND

In regularized iterative reconstruction algorithms, the selection of regularization parameter depends on the noise level of cone beam projection data.

OBJECTIVE

Our aim is to propose an algorithm to estimate the noise level of cone beam projection data.

METHODS

We first derived the data correlation of cone beam projection data in the Fourier domain, based on which, the signal and the noise were decoupled. Then the noise was extracted and averaged for estimation. An adaptive regularization parameter selection strategy was introduced based on the estimated noise level. Simulation and real data studies were conducted for performance validation.

RESULTS

There exists an approximately zero-energy double-wedge area in the 3D Fourier domain of cone beam projection data. As for the noise level estimation results, the averaged relative errors of the proposed algorithm in the analytical/MC/spotlight-mode simulation experiments were 0.8%, 0.14% and 0.24%, respectively, and outperformed the homogeneous area based as well as the transformation based algorithms. Real studies indicated that the estimated noise levels were inversely proportional to the exposure levels, i.e., the slopes in the log-log plot were -1.0197 and -1.049 with respect to the short-scan and half-fan modes. The introduced regularization parameter selection strategy could deliver promising reconstructed image qualities.

CONCLUSIONS

Based on the data correlation of cone beam projection data in Fourier domain, the proposed algorithm could estimate the noise level of cone beam projection data accurately and robustly. The estimated noise level could be used to adaptively select the regularization parameter.

Automatic metastatic brain tumor segmentation for stereotactic radiosurgery applications

The objective of this study is to develop an automatic segmentation strategy for efficient and accurate metastatic brain tumor delineation on contrast-enhanced T1-weighted (T1c) magnetic resonance images (MRI) for stereotactic radiosurgery (SRS) applications. The proposed four-step automatic brain metastases segmentation strategy is comprised of pre-processing, initial contouring, contour evolution, and contour triage. First, T1c brain images are preprocessed to remove the skull. Second, an initial tumor contour is created using a multi-scaled adaptive threshold-based bounding box and a super-voxel clustering technique. Third, the initial contours are evolved to the tumor boundary using a regional active contour technique. Fourth, all detected false-positive contours are removed with geometric characterization. The segmentation process was validated on a realistic virtual phantom containing Gaussian or Rician noise. For each type of noise distribution, five different noise levels were tested. Twenty-one cases from the multimodal brain tumor image segmentation (BRATS) challenge dataset and fifteen clinical metastases cases were also included in validation. Segmentation performance was quantified by the Dice coefficient (DC), normalized mutual information (NMI), structural similarity (SSIM), Hausdorff distance (HD), mean value of surface-to-surface distance (MSSD) and standard deviation of surface-to-surface distance (SDSSD). In the numerical phantom study, the evaluation yielded a DC of 0.98  ±  0.01, an NMI of 0.97  ±  0.01, an SSIM of 0.999  ±  0.001, an HD of 2.2  ±  0.8 mm, an MSSD of 0.1  ±  0.1 mm, and an SDSSD of 0.3  ±  0.1 mm. The validation on the BRATS data resulted in a DC of 0.89  ±  0.08, which outperform the BRATS challenge algorithms. Evaluation on clinical datasets gave a DC of 0.86  ±  0.09, an NMI of 0.80  ±  0.11, an SSIM of 0.999  ±  0.001, an HD of 8.8  ±  12.6 mm, an MSSD of 1.5  ±  3.2 mm, and an SDSSD of 1.8  ±  3.4 mm when comparing to the physician drawn ground truth. The result indicated that the developed automatic segmentation strategy yielded accurate brain tumor delineation and presented as a useful clinical tool for SRS applications.

Recent developments and comprehensive evaluations of a GPU-based Monte Carlo package for proton therapy

Monte Carlo (MC) simulation is commonly considered as the most accurate dose calculation method for proton therapy. Aiming at achieving fast MC dose calculations for clinical applications, we have previously developed a graphics-processing unit (GPU)-based MC tool, gPMC. In this paper, we report our recent updates on gPMC in terms of its accuracy, portability, and functionality, as well as comprehensive tests on this tool. The new version, gPMC v2.0, was developed under the OpenCL environment to enable portability across different computational platforms. Physics models of nuclear interactions were refined to improve calculation accuracy. Scoring functions of gPMC were expanded to enable tallying particle fluence, dose deposited by different particle types, and dose-averaged linear energy transfer (LETd). A multiple counter approach was employed to improve efficiency by reducing the frequency of memory writing conflict at scoring. For dose calculation, accuracy improvements over gPMC v1.0 were observed in both water phantom cases and a patient case. For a prostate cancer case planned using high-energy proton beams, dose discrepancies in beam entrance and target region seen in gPMC v1.0 with respect to the gold standard tool for proton Monte Carlo simulations (TOPAS) results were substantially reduced and gamma test passing rate (1%/1 mm) was improved from 82.7%-93.1%. The average relative difference in LETd between gPMC and TOPAS was 1.7%. The average relative differences in the dose deposited by primary, secondary, and other heavier particles were within 2.3%, 0.4%, and 0.2%. Depending on source proton energy and phantom complexity, it took 8-17 s on an AMD Radeon R9 290x GPU to simulate [Formula: see text] source protons, achieving less than [Formula: see text] average statistical uncertainty. As the beam size was reduced from 10  ×  10 cm² to 1  ×  1 cm², the time on scoring was only increased by 4.8% with eight counters, in contrast to a 40% increase using only one counter. With the OpenCL environment, the portability of gPMC v2.0 was enhanced. It was successfully executed on different CPUs and GPUs and its performance on different devices varied depending on processing power and hardware structure.

[Efficacy and safety of a novel ultrafiltration device for treating patients with refractory heart failure]

OBJECTIVE

To evaluate the efficacy and safety of a new ultrafiltration device for treating refractory heart failure patients.

METHODS

A total of 52 patients (37 male, age 29-85(33±44)years) with refractory heart failure were treated using a new ultrafiltration device (FQ-16). Body weight, dyspnea score, oxygen saturation (SatO2), left ventricular ejection fraction(LVEF), BUN, creatinine, electrolytes and blood gas analysis were assessed before and after the treatment. Hypotension event and other main adverse events were recorded.

RESULTS

Ultrafiltration duration ranged between 8-22 hours. Total ultrafiltration volume was (4 489±1 548) ml. Compared with baseline, patients' body weight decreased from (75.3±8.74) kg to (69.8±8.39) kg (P<0.01), dyspnea score improved from 2.47±1.55 to 12.87±3.61 (P<0.01) and SatO2 increased from 91.0±6.01 to 96.4±2.52 (P<0.01) and LVEF increased from (30.0±4.1)% to (36.0±4.3)% (P<0.01) after ultrafiltration. Blood creatinine, BUN, electrolytes and blood gas analysis values were similar at baseline and post ultrafiltration. No hypotension event and other main adverse events occurred during the ultrafiltration treatment.

CONCLUSIONS

The novel ultrafiltration device adequately relieved hypervolemia and dyspnea in patients with refractory heart failure and the treatment process is safe in this patient cohort.

A new scheme for real-time high-contrast imaging in lung cancer radiotherapy: a proof-of-concept study

Visualization of anatomy in real time is of critical importance for motion management in lung cancer radiotherapy. To achieve real-time, and high-contrast in-treatment imaging, we propose a novel scheme based on the measurement of Compton scatter photons. In our method, a slit x-ray beam along the superior-inferior direction is directed to the patient, (intersecting the lung region at a 2D plane) containing most of the tumor motion trajectory. X-ray photons are scattered off this plane primarily due to the Compton interaction. An imager with a pinhole or a slat collimator is placed at one side of the plane to capture the scattered photons. The resulting image, after correcting for incoming fluence inhomogeneity, x-ray attenuation, scatter angle variation, and outgoing beam geometry, represents the linear attenuation coefficient of Compton scattering. This allows the visualization of the anatomy on this plane. We performed Monte Carlo simulation studies both on a phantom and a patient for proof-of-principle purposes. In the phantom case, a small tumor-like structure could be clearly visualized. The contrast-resolution calculated using tumor/lung as foreground/background for kV fluoroscopy, cone beam computed tomography (CBCT), and scattering image were 0.037, 0.70, and 0.54, respectively. In the patient case, tumor motion could be clearly observed in the scatter images. Imaging dose to the voxels directly exposed by the slit beam was ~0.4 times of that under a single CBCT projection. These studies demonstrated the potential feasibility of the proposed imaging scheme to capture the instantaneous anatomy of a patient on a 2D plane with a high image contrast. Clear visualization of the tumor motion may facilitate marker-less tumor tracking.

A deformable head and neck phantom with in-vivo dosimetry for adaptive radiotherapy quality assurance

PURPOSE

Patients' interfractional anatomic changes can compromise the initial treatment plan quality. To overcome this issue, adaptive radiotherapy (ART) has been introduced. Deformable image registration (DIR) is an important tool for ART and several deformable phantoms have been built to evaluate the algorithms' accuracy. However, there is a lack of deformable phantoms that can also provide dosimetric information to verify the accuracy of the whole ART process. The goal of this work is to design and construct a deformable head and neck (HN) ART quality assurance (QA) phantom with in vivo dosimetry.

METHODS

An axial slice of a HN patient is taken as a model for the phantom construction. Six anatomic materials are considered, with HU numbers similar to a real patient. A filled balloon inside the phantom tissue is inserted to simulate tumor. Deflation of the balloon simulates tumor shrinkage. Nonradiopaque surface markers, which do not influence DIR algorithms, provide the deformation ground truth. Fixed and movable holders are built in the phantom to hold a diode for dosimetric measurements.

RESULTS

The measured deformations at the surface marker positions can be compared with deformations calculated by a DIR algorithm to evaluate its accuracy. In this study, the authors selected a Demons algorithm as a DIR algorithm example for demonstration purposes. The average error magnitude is 2.1 mm. The point dose measurements from the in vivo diode dosimeters show a good agreement with the calculated doses from the treatment planning system with a maximum difference of 3.1% of prescription dose, when the treatment plans are delivered to the phantom with original or deformed geometry.

CONCLUSIONS

In this study, the authors have presented the functionality of this deformable HN phantom for testing the accuracy of DIR algorithms and verifying the ART dosimetric accuracy. The authors' experiments demonstrate the feasibility of this phantom serving as an end-to-end ART QA phantom.

Patient-specific dosimetric endpoints based treatment plan quality control in radiotherapy

In intensity modulated radiotherapy (IMRT), the optimal plan for each patient is specific due to unique patient anatomy. To achieve such a plan, patient-specific dosimetric goals reflecting each patient's unique anatomy should be defined and adopted in the treatment planning procedure for plan quality control. This study is to develop such a personalized treatment plan quality control tool by predicting patient-specific dosimetric endpoints (DEs). The incorporation of patient specific DEs is realized by a multi-OAR geometry-dosimetry model, capable of predicting optimal DEs based on the individual patient's geometry. The overall quality of a treatment plan is then judged with a numerical treatment plan quality indicator and characterized as optimal or suboptimal. Taking advantage of clinically available prostate volumetric modulated arc therapy (VMAT) treatment plans, we built and evaluated our proposed plan quality control tool. Using our developed tool, six of twenty evaluated plans were identified as sub-optimal plans. After plan re-optimization, these suboptimal plans achieved better OAR dose sparing without sacrificing the PTV coverage, and the dosimetric endpoints of the re-optimized plans agreed well with the model predicted values, which validate the predictability of the proposed tool. In conclusion, the developed tool is able to accurately predict optimally achievable DEs of multiple OARs, identify suboptimal plans, and guide plan optimization. It is a useful tool for achieving patient-specific treatment plan quality control.

An analytic linear accelerator source model for GPU-based Monte Carlo dose calculations

Recently, there has been a lot of research interest in developing fast Monte Carlo (MC) dose calculation methods on graphics processing unit (GPU) platforms. A good linear accelerator (linac) source model is critical for both accuracy and efficiency considerations. In principle, an analytical source model should be more preferred for GPU-based MC dose engines than a phase-space file-based model, in that data loading and CPU-GPU data transfer can be avoided. In this paper, we presented an analytical field-independent source model specifically developed for GPU-based MC dose calculations, associated with a GPU-friendly sampling scheme. A key concept called phase-space-ring (PSR) was proposed. Each PSR contained a group of particles that were of the same type, close in energy and reside in a narrow ring on the phase-space plane located just above the upper jaws. The model parameterized the probability densities of particle location, direction and energy for each primary photon PSR, scattered photon PSR and electron PSR. Models of one 2D Gaussian distribution or multiple Gaussian components were employed to represent the particle direction distributions of these PSRs. A method was developed to analyze a reference phase-space file and derive corresponding model parameters. To efficiently use our model in MC dose calculations on GPU, we proposed a GPU-friendly sampling strategy, which ensured that the particles sampled and transported simultaneously are of the same type and close in energy to alleviate GPU thread divergences. To test the accuracy of our model, dose distributions of a set of open fields in a water phantom were calculated using our source model and compared to those calculated using the reference phase-space files. For the high dose gradient regions, the average distance-to-agreement (DTA) was within 1 mm and the maximum DTA within 2 mm. For relatively low dose gradient regions, the root-mean-square (RMS) dose difference was within 1.1% and the maximum dose difference within 1.7%. The maximum relative difference of output factors was within 0.5%. Over 98.5% passing rate was achieved in 3D gamma-index tests with 2%/2 mm criteria in both an IMRT prostate patient case and a head-and-neck case. These results demonstrated the efficacy of our model in terms of accurately representing a reference phase-space file. We have also tested the efficiency gain of our source model over our previously developed phase-space-let file source model. The overall efficiency of dose calculation was found to be improved by ~1.3-2.2 times in water and patient cases using our analytical model.

Dosimetric comparison of Acuros XB with collapsed cone convolution/superposition and anisotropic analytic algorithm for stereotactic ablative radiotherapy of thoracic spinal metastases

The aim of this study is to compare the recent Eclipse Acuros XB (AXB) dose calculation engine with the Pinnacle collapsed cone convolution/superposition (CCC) dose calculation algorithm and the Eclipse anisotropic analytic algorithm (AAA) for stereotactic ablative radiotherapy (SAbR) treatment planning of thoracic spinal (T‐spine) metastases using IMRT and VMAT delivery techniques. The three commissioned dose engines (CCC, AAA, and AXB) were validated with ion chamber and EBT2 film measurements utilizing a heterogeneous slab‐geometry water phantom and an anthropomorphic phantom. Step‐and‐shoot IMRT and VMAT treatment plans were developed and optimized for eight patients in Pinnacle, following our institutional SAbR protocol for spinal metastases. The CCC algorithm, with heterogeneity corrections, was used for dose calculations. These plans were then exported to Eclipse and recalculated using the AAA and AXB dose calculation algorithms. Various dosimetric parameters calculated with CCC and AAA were compared to that of the AXB calculations. In regions receiving above 50% of prescription dose, the calculated CCC mean dose is 3.1%–4.1% higher than that of AXB calculations for IMRT plans and 2.8%–3.5% higher for VMAT plans, while the calculated AAA mean dose is 1.5%–2.4% lower for IMRT and 1.2%–1.6% lower for VMAT. Statistically significant differences (p<0.05) were observed for most GTV and PTV indices between the CCC and AXB calculations for IMRT and VMAT, while differences between the AAA and AXB calculations were not statistically significant. For T‐spine SAbR treatment planning, the CCC calculations give a statistically significant overestimation of target dose compared to AXB. AAA underestimates target dose with no statistical significance compared to AXB. Further study is needed to determine the clinical impact of these findings.

PACS number: 87.55.D‐, 87.53.Ly

Towards the clinical implementation of iterative low-dose cone-beam CT reconstruction in image-guided radiation therapy: cone/ring artifact correction and multiple GPU implementation

PURPOSE

Compressed sensing (CS)-based iterative reconstruction (IR) techniques are able to reconstruct cone-beam CT (CBCT) images from undersampled noisy data, allowing for imaging dose reduction. However, there are a few practical concerns preventing the clinical implementation of these techniques. On the image quality side, data truncation along the superior-inferior direction under the cone-beam geometry produces severe cone artifacts in the reconstructed images. Ring artifacts are also seen in the half-fan scan mode. On the reconstruction efficiency side, the long computation time hinders clinical use in image-guided radiation therapy (IGRT).

METHODS

Image quality improvement methods are proposed to mitigate the cone and ring image artifacts in IR. The basic idea is to use weighting factors in the IR data fidelity term to improve projection data consistency with the reconstructed volume. In order to improve the computational efficiency, a multiple graphics processing units (GPUs)-based CS-IR system was developed. The parallelization scheme, detailed analyses of computation time at each step, their relationship with image resolution, and the acceleration factors were studied. The whole system was evaluated in various phantom and patient cases.

RESULTS

Ring artifacts can be mitigated by properly designing a weighting factor as a function of the spatial location on the detector. As for the cone artifact, without applying a correction method, it contaminated 13 out of 80 slices in a head-neck case (full-fan). Contamination was even more severe in a pelvis case under half-fan mode, where 36 out of 80 slices were affected, leading to poorer soft tissue delineation and reduced superior-inferior coverage. The proposed method effectively corrects those contaminated slices with mean intensity differences compared to FDK results decreasing from ∼497 and ∼293 HU to ∼39 and ∼27 HU for the full-fan and half-fan cases, respectively. In terms of efficiency boost, an overall 3.1 × speedup factor has been achieved with four GPU cards compared to a single GPU-based reconstruction. The total computation time is ∼30 s for typical clinical cases.

CONCLUSIONS

The authors have developed a low-dose CBCT IR system for IGRT. By incorporating data consistency-based weighting factors in the IR model, cone/ring artifacts can be mitigated. A boost in computational efficiency is achieved by multi-GPU implementation.

A method for volumetric imaging in radiotherapy using single x-ray projection

PURPOSE

It is an intriguing problem to generate an instantaneous volumetric image based on the corresponding x-ray projection. The purpose of this study is to develop a new method to achieve this goal via a sparse learning approach.

METHODS

To extract motion information hidden in projection images, the authors partitioned a projection image into small rectangular patches. The authors utilized a sparse learning method to automatically select patches that have a high correlation with principal component analysis (PCA) coefficients of a lung motion model. A model that maps the patch intensity to the PCA coefficients was built along with the patch selection process. Based on this model, a measured projection can be used to predict the PCA coefficients, which are then further used to generate a motion vector field and hence a volumetric image. The authors have also proposed an intensity baseline correction method based on the partitioned projection, in which the first and the second moments of pixel intensities at a patch in a simulated projection image are matched with those in a measured one via a linear transformation. The proposed method has been validated in both simulated data and real phantom data.

RESULTS

The algorithm is able to identify patches that contain relevant motion information such as the diaphragm region. It is found that an intensity baseline correction step is important to remove the systematic error in the motion prediction. For the simulation case, the sparse learning model reduced the prediction error for the first PCA coefficient to 5%, compared to the 10% error when sparse learning was not used, and the 95th percentile error for the predicted motion vector was reduced from 2.40 to 0.92 mm. In the phantom case with a regular tumor motion, the predicted tumor trajectory was successfully reconstructed with a 0.82 mm error for tumor center localization compared to a 1.66 mm error without using the sparse learning method. When the tumor motion was driven by a real patient breathing signal with irregular periods and amplitudes, the average tumor center error was 0.6 mm. The algorithm robustness with respect to sparsity level, patch size, and presence or absence of diaphragm, as well as computation time, has also been studied.

CONCLUSIONS

The authors have developed a new method that automatically identifies motion information from an x-ray projection, based on which a volumetric image is generated.

A practical cone-beam CT scatter correction method with optimized Monte Carlo simulations for image-guided radiation therapy

Cone-beam CT (CBCT) has become the standard image guidance tool for patient setup in image-guided radiation therapy. However, due to its large illumination field, scattered photons severely degrade its image quality. While kernel-based scatter correction methods have been used routinely in the clinic, it is still desirable to develop Monte Carlo (MC) simulation-based methods due to their accuracy. However, the high computational burden of the MC method has prevented routine clinical application. This paper reports our recent development of a practical method of MC-based scatter estimation and removal for CBCT. In contrast with conventional MC approaches that estimate scatter signals using a scatter-contaminated CBCT image, our method used a planning CT image for MC simulation, which has the advantages of accurate image intensity and absence of image truncation. In our method, the planning CT was first rigidly registered with the CBCT. Scatter signals were then estimated via MC simulation. After scatter signals were removed from the raw CBCT projections, a corrected CBCT image was reconstructed. The entire workflow was implemented on a GPU platform for high computational efficiency. Strategies such as projection denoising, CT image downsampling, and interpolation along the angular direction were employed to further enhance the calculation speed. We studied the impact of key parameters in the workflow on the resulting accuracy and efficiency, based on which the optimal parameter values were determined. Our method was evaluated in numerical simulation, phantom, and real patient cases. In the simulation cases, our method reduced mean HU errors from 44 to 3 HU and from 78 to 9 HU in the full-fan and the half-fan cases, respectively. In both the phantom and the patient cases, image artifacts caused by scatter, such as ring artifacts around the bowtie area, were reduced. With all the techniques employed, we achieved computation time of less than 30 s including the time for both the scatter estimation and CBCT reconstruction steps. The efficacy of our method and its high computational efficiency make our method attractive for clinical use.