CBCT-guided Adaptive Radiotherapy Research Articles

Using a patient-specific diffusion model to generate CBCT-based synthetic CTs for CBCT-guided adaptive radiotherapy.

Cone beam computed tomography (CBCT) can be used to evaluate the inter-fraction anatomical changes during the entire course for image-guided radiotherapy (IGRT). However, CBCT artifacts from various sources restrict the full application of CBCT-guided adaptive radiation therapy (ART). Inter-fraction anatomical changes during ART, including variations in tumor size and normal tissue anatomy, can affect radiation therapy (RT) efficacy. Acquiring high-quality CBCT images that accurately capture patient- and fraction-specific (PFS) anatomical changes is crucial for successful IGRT. To enhance CBCT image quality, we proposed PFS lung diffusion models (PFS-LDMs). The proposed PFS models use a pre-trained general lung diffusion model (GLDM) as a baseline, which is trained on historical deformed CBCT (dCBCT)-planning CT (pCT) paired data. For a given patient, a new PFS model is fine-tuned on a CBCT-deformed pCT (dpCT) pair after each fraction to learn the PFS knowledge for generating personalized synthetic CT (sCT) with quality comparable to pCT or dpCT. The learned PFS knowledge is the specific mapping relationships, including personal inter-fraction anatomical changes between personalized CBCT-dpCT pairs. The PFS-LDMs were evaluated on an institutional lung cancer dataset, quantified by mean absolute error (MAE), peak signal-to-noise ratio (PSNR), normalized cross-correlation (NCC), and structural similarity index measure (SSIM) metrics. We also compared our PFS-LDMs with a mainstream GAN-based model, demonstrating that our PFS fine-tuning strategy could be applied to existing generative models. Our models showed remarkable improvements across all four evaluation metrics. The proposed PFS-LDMs outperformed the GLDM, demonstrating the effectiveness of our proposed fine-tuning strategy. The PFS model fine-tuned with CBCT images from four prior fractions, reduced the MAE from 103.95 to 15.96 Hounsfield units (HU), and increased the mean PSNR, NCC, and SSIM from 25.36dB to 33.57dB, 0.77 to 0.98, and 0.75 to 0.97, respectively. Applying our PFS fine-tuning strategy to a Cycle GAN model also showed improvements, with all four fine-tuned PFS Cycle GAN (PFS-CG) models outperforming the general Cycle GAN model. Overall, our proposed PFS fine-tuning strategy improved CBCT image quality compared to both the pre-correction and non-fine-tuned general models, with our proposed PFS-LDMs yielding better performance than the GAN-based model across all metrics. Our proposed PFS-LDMs significantly improve CBCT image quality with increased HU accuracy and fewer artifacts, thus better capturing inter-fraction anatomical changes. This lays the groundwork for enabling CBCT-based ART, which could enhance clinical efficiency and achieve personalized high-precision treatment by accounting for inter-fraction anatomical changes.

Evaluation of the geometric and dosimetric accuracies of deformable image registration of targets and critical organs in prostate CBCT-guided adaptive radiotherapy.

Kilovoltage cone beam computed tomography (kVCBCT)-guided adaptive radiation therapy (ART) uses daily deformed CT (dCT), which is generated automatically through deformable registration methods. These registration methods may perform poorly in reproducing volumes of the target organ, rectum, and bladder during treatment. We analyzed the registration errors between the daily kVCBCTs and corresponding dCTs for these organs using the default optical flow algorithm and two registration procedures. We validated the effectiveness of these registration methods in replicating the geometry for dose calculation on kVCBCT for ART. We evaluated three deformable image registration (DIR) methods to assess their registration accuracy and dose calculation effeciency in mapping target and critical organs. The DIR methods include (1) default intensity-based deformable registration, (2) hybrid deformable registration, and (3) a two-step deformable registration process. Each technique was applied to a computerized imaging reference system (CIRS) phantom (Model 062M) and to five patients who received volumetric modulated arc therapy to the prostate. Registration accuracy was assessed using the 95% Hausdorff distance (HD95) and Dice similarity coefficient (DSC), and each method was compared with the intensity-based registration method. The improvement in the dCT image quality of the CIRS phantom and five patients was assessed by comparing dCT with kVCBCT. Image quality quantitative metrics for the phantom included the signal-to-noise ratio (SNR), uniformity, and contrast-to-noise ratio (CNR), whereas those for the patients included the mean absolute error (MAE), mean error, peak signal-to-noise ratio (PSNR), and structural similarity index measure (SSIM). To determine dose metric differences, we used a dose-volume histogram (DVH) and 3.0%/0.3mm gamma analysis to compare planning computed tomography (pCT) and kVCBCT recalculations with restimulated CT images used as a reference. The dCT images generated by the hybrid (dCTH) and two-step (dCTC) registration methods resulted in significant improvements compared to kVCBCT in the phantom model. Specifically, the SNR improved by 107% and 107.2%, the uniformity improved by 90% and 75%, and the CNR improved by 212.2% and 225.6 for dCTH and dCTC methods, respectively. For the patient images, the MAEs improved by 98% and 94%, the PSNRs improved by 16.3% and 22.9%, and the SSIMs improved by 1% and 1% in the dCTH and dCTC methods, respectively. For the geometric evaluation, only the two-step registration method improved registration accuracy. The dCTH method yielded an average HD95 of 12mm and average DSC of 0.73, whereas dCTC yielded an average HD95 of 2.9mm and average DSC of 0.902. The DVH showed that the dCTC-based dose calculations differed by <2% from the expected results for treatment targets and volumes of organs at risk. Additionally, gamma indices for dCTC-based treatment plans were >95% at all points, whereas they were <95% for kVCBCT-based treatment plans. The two-step registration method outperforms the intensity-based and hybrid registration methods. While the hybrid and two-step-based methods improved the image quality of kVCBCT in a linear accelerator, only the two-step method improved the registration accuracy of the corresponding structures among the pCT and kVCBCT datasets. A two-step registration process is recommended for applying kVCBCT to ART, which achieves better registration accuracy for local and global image structures. This method appears to be beneficial for radiotherapy dose calculation in patients with pelvic cancer.

Empirical analysis of a plan-of-the-day strategy to approximate daily online reoptimization for prostate CBCT-guided adaptive radiotherapy.

Adaptive radiotherapy (ART) can improve the dose delivered to the patient in the presence of anatomic variations. However, the required time, effort, and clinical resources are intensive. This work analyzed a plan-of-the-day (POD) approach on clinical patients treated with online ART to explore implementations that balance dosimetric benefit and clinical resource cost. Eight patients treated to the prostate and proximal seminal vesicles with 26 fractions of CBCT-guided, daily online ART were retrospectively analyzed. With a plan library composed of daily adaptive plans from the initial week of treatment and the original plan, the effect of a POD approach starting the following week was investigated by simulating use of these previously generated plans under 3- and 6-degree-of-freedom patient alignment. The plan selected for each treatment was that from the library that maximized the Dice similarity coefficient of the clinical target volume with that of the current treatment fraction. The resulting distribution of several target coverage and organ-at-risk dose metrics are described relative to those achieved with the daily online reoptimized adaptive technique. The values of target coverage and organ-at-risk dose metrics varied across patients and metrics. The POD schemas closely approximated the reference values from a fully reoptimized adaptive plan yet required less than 20% of the reoptimization effort. The POD schemas also had a much greater effect on target coverage metrics than 6-degree-of-freedom registration did. Organ-at-risk dose metrics also varied considerably across patients but did not exhibit a consistent dependence on the particular schema. POD schemas were able to achieve the vast majority of the dosimetric benefit of daily online ART with a small fraction of the online reoptimization effort. Strategies like this might allow for more practical and strategic implementation of ART so as to benefit a greater number of patients.

More Efficient Auto-Segmentation Framework Using Patient-Specific Information for CBCT-Guided Adaptive Radiotherapy

Accurate and fast delineation of regions of interest (ROIs) is critical for online adaptive radiotherapy (ART). Due to the noise, artifacts and low contrast of soft tissue on cone-beam computed tomography (CBCT), the CBCT-guided ART (CBCTgART) needs an effective tool to assist the segmentation of ROIs. This study aims to develop an efficient auto-segmentation workflow with personalized modeling for CBCTgART based on deep learning (DL) model. Five hundred fifty-two patients with nasopharyngeal carcinoma were enrolled in this study. At the beginning, a cycle consistent adversarial network (CycleGAN) model was trained on 172 patients' CBCT/CT paired images to improve the image quality of CBCT to CT level. A generalized segmentation model was trained using the planning CT (pCT) and contour data from 530 patients. For personalized modeling, the generalized segmentation model was fine-tuned on the specific patient's pCT and contour to get the personalized model. When CBCT images were available, the trained CycleGAN model transformed the CBCT to synthetic pCT. Then the personalized auto-segmentation model generated the contour of ROIs on the synthetic pCT. We randomly selected 22 patients for model test. The proposed method (DL-personalized) was compared with other DL methods based on the same architecture of network: regular deep learning method (DL-regular), which was trained on the CBCT and corresponding contours, and generalized model in our framework (DL-generalized). So, 22 personalized, 1 generalized and 1 regular DL models were tested. The paired t-test was performed to test the significance for mean dice similarity coefficient (DSC), mean distance to agreement (MDA), and Hausdorff distance (HD) between the alternative and proposed methods. Two ROIs were included: the clinical target volume (CTV) and nasopharynx gross tumor volume (GTVnx). The proposed DL-personalized model achieved better results compared with others as shown in the table. The accuracy and robustness of our proposed framework was reliable. All of p values were under 0.01, which indicated the statistically significant difference. The proposed framework utilizing patient-specific information can improve the segmentation accuracy of ART. This method has potential to be integrated into the ART workflow to improve efficiency.

Rapid Unpaired CBCT-Based Synthetic CT for CBCT-Guided Adaptive Radiotherapy

Quantitative cone beam CT (CBCT) is the foundation for image-guided radiation therapy, improving treatment setup, tumor delineation and dose calculation. However, CBCT images suffer from severe artifacts, limiting clinical utility. Deep learning can overcome these limitations, boosting radiographic and dosimetric quality critical for online adaptive radiotherapy (ART). We hypothesize adapted contrastive unpaired translation (CUT), a recent method for image-to-image translation of photographic images, can improve CBCT quality while reducing compute time, demonstrating utility for ART. Same-day CBCT and quality assurance CT (QACT) images acquired from 79 patients receiving proton therapy for prostate cancer between 2019 and 2020 at a single institution were retrospectively collected. QACT images were acquired for quality assurance in accordance with institutional policy. Seventy-nine patients yielded 102 non-contrast CBCT-QACT image sets. Each QACT image was rigidly registered to the corresponding CBCT and resampled to 1 × 1 × 2 mm to establish uniform voxel size and spacing. CBCT images were randomly shuffled prior to input to the CUT model for unsupervised training and QACT-quality synthetic CT images were generated as outputs. We compared mean absolute error (MAE), structural similarity index measure (SSIM), and Fréchet inception distance (FID) against same-day QACT. MAE, SSIM, and FID were compared for the CycleGAN and CUT data relative to input QACT and are reported as the mean across five-fold cross-validation ± standard error. CUT achieved superior performance in MAE (19.5 ± 3.9 HU vs. cycleGAN 47.1 ± 25.4) and FID (31.5 ± 6.6 vs cycleGAN 75.9 ± 41.3). MAE indicates pixel-level correspondence to QACT HU intensity values, making the synthetic outputs of CUT useful for dose calculations during ART. FID further demonstrates perceptual visual similarity. SSIM for CycleGAN (0.7 ± 0.2) and CUT (0.8 ± 0.0) were similar, indicating acceptable reproducibility of global structure. CUT was faster and lighter than CycleGAN. CycleGAN contained a total of 28,286,000 parameters; CUT contained 14,703,000, approximately half that of CycleGAN. As a result, CycleGAN computes on a single CT image slice over 0.33s while CUT requires just 0.18s. The contrastive method investigated here was demonstrated to be faster and more accurate than CycleGAN, requiring fewer networks and parameters to achieve superior performance. We demonstrated anatomic boundary preservation and HU fidelity superior to cycleGAN while significantly reducing compute time. We plan to investigate the use of these synthetic CT images in automated segmentation prior to exploration of CUT in a prospective setting.

CBCT-guided adaptive radiotherapy using self-supervised sequential domain adaptation with uncertainty estimation.

Adaptive radiotherapy (ART) is an advanced technology in modern cancer treatment that incorporates progressive changes in patient anatomy into active plan/dose adaption during the fractionated treatment. However, the clinical application relies on the accurate segmentation of cancer tumors on low-quality on-board images, which has posed challenges for both manual delineation and deep learning-based models. In this paper, we propose a novel sequence transduction deep neural network with an attention mechanism to learn the shrinkage of the cancer tumor based on patients' weekly cone-beam computed tomography (CBCT). We design a self-supervised domain adaption (SDA) method to learn and adapt the rich textural and spatial features from pre-treatment high-quality computed tomography (CT) to CBCT modality in order to address the poor image quality and lack of labels. We also provide uncertainty estimation for sequential segmentation, which aids not only in the risk management of treatment planning but also in the calibration and reliability of the model. Our experimental results based on a clinical non-small cell lung cancer (NSCLC) dataset with sixteen patients and ninety-six longitudinal CBCTs show that our model correctly learns weekly deformation of the tumor over time with an average dice score of 0.92 on the immediate next step, and is able to predict multiple steps (up to 5 weeks) for future patient treatments with an average dice score reduction of 0.05. By incorporating the tumor shrinkage predictions into a weekly re-planning strategy, our proposed method demonstrates a significant decrease in the risk of radiation-induced pneumonitis up to 35% while maintaining the high tumor control probability.

Surface Imaging-Guided Adaptive Radiotherapy (ART): Imaging Protocol and Verification for Inter- and Intra-Fractional Motion Managements

<h3>Purpose/Objective(s)</h3> CBCT-guided adaptive radiotherapy (CT-ART) has proven to account for inter-fractional anatomic changes. However, no commercial solution for intra-fractional respiratory motion management existed in CT-ART. Therefore, we investigated intra-fractional motion-managed treatment strategies in CT-ART (or Image-guided RT: IGRT) using optical surface imaging combined with <i>in-situ</i> CBCTs. <h3>Materials/Methods</h3> Eight of ten cancer patients with mobile lower lung or upper abdominal malignancies have completed CT-ART (or IGRT) using Ethos radiotherapy in an IRB-approved clinical trial (phase I). Four of the eight patients underwent additional MR simulations, including cine MRIs, to estimate the tumor/abdominal skin motion correlation. In CT-ART, a pre-configured gating window (±2 mm in vertical position) on surface imaging was used for manually triggering intra-fractional CBCT and treatment beam irradiation in breath-hold (six patients for exhalation and two patients for inhalation). A donut-shaped imaging-surrogate on patient's abdomen was used to localize Region of Interest (ROI) of the abdomen on CBCTs. Two inter-fractional CBCTs at exhalation and inhalation in each fraction were acquired to verify the tumor/imaging-surrogate motions range. Intra-fractional CBCTs were used to quantify the residual motion of the tumor/imaging-surrogate within the pre-configured gating window. Tumor motion was measured by tracking the position of a tumor centroid. <h3>Results</h3> The pre-configured gating window on surface imaging was confirmed based on the correlation of the tumor and abdominal skin motions using cine MRIs. Thirty-six pairs of inter-fractional CBCTs and 100 intra-fractional CBCTs from 36 fractions were analyzed (13 fractions for IGRT and 23 for ART). In the motion range study from the inter-fractional CBCTs, the values of mean ± std were 17.3 ± 12.0 mm in the SI direction for the tumors and 16.6 ± 7.9 mm in the AP direction for the imaging-surrogate, respectively. The residual motion of the imaging-surrogate from intra-fractional CBCTs was 2.5 ± 2.1 mm for SI, 2.3 ± 2.0 mm for RL, and 2.1 ± 2.3 mm for AP directions. The residual motion of the tumors was 2.4 ± 3.1 mm for SI, 2.3 ± 2.3 mm for RL, and 2.2 ± 2.0 mm for AP directions. Trends in the residual motion during the course of treatment fractions and over time through delivery were stable. Inter-fractional residual motions of the tumors were 2.3 ± 0.8 mm for SI, 2.3 ± 0.6 mm for RL, and 2.2 ± 0.7 mm for AP directions. Intra-fractional residual motions of the tumors were 2.4 ± 0.9 mm for SI, 2.4 ± 0.7 mm for RL, and 3.2 ± 1.3 mm for AP. <h3>Conclusion</h3> We demonstrated the inter- and intra-fractional motion-managed treatment strategies in CBCT-guided ART using surface imaging. Since the residual motion of the tumors was comparable to that of the imaging-surrogate within clinical PTV margins (5 mm) but a bit larger than the pre-configured gating window, additional caution is needed with intra-fractional motion management in CT-ART using the optical surface imaging.

Developing a CBCT-Guided Online Adaptive Radiotherapy Approach for Isotoxic Delivery of Ultra-High Dose to Oligometastatic Disease in Combination with Immunotherapy

<h3>Purpose/Objective(s)</h3> A Phase I trial combining stereotactic body radiotherapy (SBRT), 45Gy in 3 fractions(fx), with pembrolizumab was safe (9% grade 3 toxicity) with promising local control and overall survival. Whether SBRT dose escalation can improve outcomes is uncertain. A subsequent planning study showed feasibility of ultra-high SBRT dose, 90Gy in 3fx, while respecting organ at risk (OAR) tolerances. However, due to patient anatomy interfractional uncertainty, online adaptation may enhance safety/efficacy. This work develops a CBCT-guided adaptive radiotherapy approach for isotoxic treatment with ultra-high doses using retrospective data and simulated online adaptation. <h3>Materials/Methods</h3> Patient data from 15 subjects of the original Phase I trial were used (gross tumor volume (GTV) mean volume 177cc; range 67-390cc). Disease sites were liver(n=8), pelvis(n=4), abdomen(n=2), lung(n=1). Clinical prescriptions were to 65cc subvolumes of the GTV (SubGTV). In the adaptive setting, rather than define a specific sub volume, OAR expansions based on tolerance and achievable dose gradients were cropped from GTV so that the SubGTV receiving ultra-high dose adapts to changing anatomy. Initial 6 arc VMAT plans used a clinical goal based automated plan optimization platform. High priority was given to critical OAR Dmax goals, medium priority was given to GTV coverage (V90Gy) and OAR volumetric constraints. SubGTV core was defined with further contraction away from OARs with low priority goals to ensure a centralized GTV hot spot. For a subset of patients (8 total: 3 liver, 2 abdomen, 2 pelvis, 1 lung), an online adaptive session using clinical CBCTs was simulated with anatomy recontoured for plan reoptimization/evaluation and comparison of adapted (ADP) vs. scheduled (SCH) plans. <h3>Results</h3> For initial plans, all OAR goals were satisfied. The average GTV V90Gy was 80.8cc/57.3% (range: 53.8-144.5cc/24.4-99.1%). In online adaptive sessions, the ADP plan was preferred over the SCH plan in 7 of 8 cases. ADP plans generally had superior coverage (average GTV V90Gy of 86.3cc (ADP) vs. 77.9cc (SCH)). ADP plans nearly always met OAR tolerances (one minor exception: skin D0.03cc (tolerance of 11.0Gy/fx) was 11.5Gy (ADP) vs. 11.9Gy (SCH)). In contrast, SCH plans often did not meet OAR constraints, with critical violations to bowel (n=2), stomach(n=3), duodenum(n=1), inferior vena cava(n=1) as well as violations to skin(n=3) and bone(n=2). <h3>Conclusion</h3> This work develops a paradigm for true isotoxic radiotherapy treatment maximizing ablative target doses using CBCT-guided online adaptive SBRT. A forthcoming phase I trial will investigate potential clinical benefit from escalation to ultra-high doses. As adaptive technologies evolve, the OAR-guided isotoxic approach developed here may be an important future direction for radiation oncology. NCT02608385

Evaluation of Automated Breast Cancer Contour Propagation From Planning CT to Cone Beam CT Based on Deep Learning

<h3>Purpose/Objective(s)</h3> To evaluate a regional deformable model based on deep unsupervised learning framework for automated contour propagation in breast Cone Beam Computed Tomography (CBCT) guided radiation therapy. <h3>Materials/Methods</h3> We introduce an unsupervised learning framework to map the breast tumor bed, clinical target volume (CTV), heart, left lung, right lung, and spinal cord from planning CT (pCT) to daily CBCT. To improve the performance of the traditional deformable image registration method, the proposed unsupervised learning method uses a regional deformable model based on the narrow band mapping, which can mitigate the effect of the image artifacts. We retrospectively selected 380 anonymized CBCT volumes from 107 patients with breast cancer. The CBCTs are divided into three sets. 311 / 20 / 49 CBCTs were used for training, validation, and testing. For the testing set, one physician-generated contours were generated as the references. Through the Dice similarity coefficients (DSC), the Hausdorff distances, and the distances of the center-of-mass, the results were compared between the proposed model and physician-generated contours. <h3>Results</h3> The mean DSC between the proposed segmentations and the physician-generated segmentations for breast tumor bed, clinical target volume (CTV), heart, left lung, right lung, and spinal cord were 0.79 ± 0.09, 0.89 ± 0.08, 0.90 ± 0.08, 0.93 ± 0.03, 0.94 ± 0.04, and 0.76 ± 0.13, respectively. The average distances of the center-of-mass for the whole testing images were less than 4 mm. The proposed regional deformable model's performance was better than the traditional global deformable model, especially in the CBCT image with severe image artifacts. <h3>Conclusion</h3> This novel deep learning-based regional deformable model technique can automatically propagate contours for breast cancer patient radiotherapy. The proposed approach shows that the deep learning technique can accomplish highly precise contour propagation for breast CBCT-guided adaptive radiotherapy.

Initial Plan Quality Evaluation Using a Novel AI-Driven Planning System and Paradigm for Adaptive Head and Neck Patients

An innovative ring gantry CBCT-guided adaptive radiotherapy system enables efficient planning and online adaptation to account for interfraction changes. Initial planning strategies and plan quality for head and neck (H&N) patients using the artificial intelligence (AI) plan optimizer for this system are evaluated.Nine previously treated H&N patients are used in this study. Planning target volumes (PTVs) have prescribed dose levels: PTVhigh = 70.0/69.3/66.0, PTVmed = 63.0/60, and PTVlow = 56.0/54.1/54.0/52.8 Gy. Clinical planning CT and structures are used for AI plan generation with the novel paradigm based on clinical goal prioritization. The system automatically optimizes 5 candidates plans: 7/9/12 equidistant field IMRT and 2/3 full arc VMAT. Three planning approaches are tested: 1) input physician goals; 2) prioritize PTV coverage by excluding organ at risk (OAR) goals when OAR intersects PTVs; 3) generate and assign goals to OAR subvolumes cropped from PTVs. Utility of generalized helper structures for controlling hot spot location and low dose spillage is investigated. The best of each patient's 5 candidate AI plans (AcurosXB) is compared to clinical (collapsed cone) plan using NRG guidelines.Strategies 1 and 2 consistently result in PTV under-coverage, high PTV hot spots, and failed OAR goals when intersecting PTVs. Strategy 3 provides acceptable PTV coverage and OAR dosimetry comparable to clinical plans. Helper structures (posterior block, distal OAR subvolumes, and PTVmed/low cropped from PTVhigh) were found to be necessary for shaping dose distributions similarly to clinical plans. Optimal plan type varied: 12 (n = 5), 9 (n = 2), 7 (n = 1) field IMRT and 3 arc VMAT (n = 1). Comparison of AI and clinical plans is based on using strategy (3) with helper structures. Compared to clinical plans, for all PTVs, average D99%, D95%, and Dmax are higher by 3.1% (103.6% vs. 100.5%), 1.3% (102.2 vs. 100.9%), and 3.3% (108.8% vs. 105.5%), respectively, but still within NRG guidelines. However, global hot spots may fall outside of PTVs and are higher by 5%. Hot spot differences may be from differences in dose calculation algorithms. The spinal cord, brainstem, parotid, oral cavity, mandible and thyroid doses are below NRG guidelines for both plans, with AI plan doses slightly lower. Larynx, Pharynx and submandibular doses on average are higher than the NRG guidelines due to PTV proximity for both plans, with AI plan doses slightly higher.The AI plan optimizer for this adaptive platform utilizes a novel planning paradigm based on clinical goals rather than direct optimization parameters and can efficiently generate H&N treatment plans. AI plans are comparable to clinical plans with slightly better PTV coverage and lower OAR doses. However, issues with higher and spatially undesirable calculated hot spots remain. All plans meet NRG guidelines.

Automated Contour Propagation of the Prostate From pCT to CBCT Images via Deep Unsupervised Learning

To develop and evaluate a deep unsupervised learning (DUL) framework based on a regional deformable model for automated prostate contour propagation from planning computed tomography (pCT) to cone-beam CT (CBCT).A DUL model was introduced to map the prostate contour from pCT to on-treatment CBCT. The DUL framework used a regional deformable model via narrow band mapping to augment the conventional strategy. 251 anonymized CBCT images from prostate cancer patients were retrospectively selected and divided into three sets: 180 were used for training, 12 for validation, and 59 for testing. The testing dataset was divided into two Groups. Group one contained 50 CBCT volumes, with one physician-generated prostate contour on CBCT image. Group two contained 9 CBCT images, each including prostate contours delineated by four independent physicians and a consensus contour generated using the STAPLE method. Results were compared between the proposed DUL and physician-generated contours through the Dice similarity coefficients (DSC), the Hausdorff distances, and the distances of the center-of-mass.The average DSCs between DUL-based prostate contours and reference contours for test data in Group one and Group two-consensus were 0.83 ± 0.04, and 0.85 ± 0.04, respectively. Correspondingly, the mean center-of-mass distances were 3.52 mm ± 1.15 mm, and 2.98 mm ± 1.42 mm, respectively.This novel DUL technique can automatically propagate the contour of the prostate from pCT to CBCT. The proposed method shows that highly accurate contour propagation for CBCT-guided adaptive radiotherapy is achievable via the deep learning technique.

Automated contour propagation of the prostate from pCT to CBCT images via deep unsupervised learning.

To develop and evaluate a deep unsupervised learning (DUL) framework based on a regional deformable model for automated prostate contour propagation from planning computed tomography (pCT) to cone-beam CT (CBCT). We introduce a DUL model to map the prostate contour from pCT to on-treatment CBCT. The DUL framework used a regional deformable model via narrow-band mapping to augment the conventional strategy. Two hundred and fifty-one anonymized CBCT images from prostate cancer patients were retrospectively selected and divided into three sets: 180 were used for training, 12 for validation, and 59 for testing. The testing dataset was divided into two groups. Group 1 contained 50 CBCT volumes, with one physician-generated prostate contour on CBCT image. Group 2 contained nine CBCT images, each including prostate contours delineated by four independent physicians and a consensus contour generated using the STAPLE method. Results were compared between the proposed DUL and physician-generated contours through the Dice similarity coefficients (DSCs), the Hausdorff distances, and the distances of the center-of-mass. The average DSCs between DUL-based prostate contours and reference contours for test data in group 1 and group 2 consensus were 0.83±0.04, and 0.85±0.04, respectively. Correspondingly, the mean center-of-mass distances were 3.52mm±1.15mm, and 2.98mm±1.42mm, respectively. This novel DUL technique can automatically propagate the contour of the prostate from pCT to CBCT. The proposed method shows that highly accurate contour propagation for CBCT-guided adaptive radiotherapy is achievable via the deep learning technique.

Head-and-neck organs-at-risk auto-delineation using dual pyramid networks for CBCT-guided adaptive radiotherapy

Organ-at-risk (OAR) delineation is a key step for cone-beam CT (CBCT) based adaptive radiotherapy planning that can be a time-consuming, labor-intensive, and subject-to-variability process. We aim to develop a fully automated approach aided by synthetic MRI for rapid and accurate CBCT multi-organ contouring in head-and-neck (HN) cancer patients. MRI has superb soft-tissue contrasts, while CBCT offers bony-structure contrasts. Using the complementary information provided by MRI and CBCT is expected to enable accurate multi-organ segmentation in HN cancer patients. In our proposed method, MR images are firstly synthesized using a pre-trained cycle-consistent generative adversarial network given CBCT. The features of CBCT and synthetic MRI (sMRI) are then extracted using dual pyramid networks for final delineation of organs. CBCT images and their corresponding manual contours were used as pairs to train and test the proposed model. Quantitative metrics including Dice similarity coefficient (DSC), Hausdorff distance 95% (HD95), mean surface distance, and residual mean square distance (RMS) were used to evaluate the proposed method. The proposed method was evaluated on a cohort of 65 HN cancer patients. CBCT images were collected from those patients who received proton therapy. Overall, DSC values of 0.87 ± 0.03, 0.79 ± 0.10/0.79 ± 0.11, 0.89 ± 0.08/0.89 ± 0.07, 0.90 ± 0.08, 0.75 ± 0.06/0.77 ± 0.06, 0.86 ± 0.13, 0.66 ± 0.14, 0.78 ± 0.05/0.77 ± 0.04, 0.96 ± 0.04, 0.89 ± 0.04/0.89 ± 0.04, 0.83 ± 0.02, and 0.84 ± 0.07 for commonly used OARs for treatment planning including brain stem, left/right cochlea, left/right eye, larynx, left/right lens, mandible, optic chiasm, left/right optic nerve, oral cavity, left/right parotid, pharynx, and spinal cord, respectively, were achieved. This study provides a rapid and accurate OAR auto-delineation approach, which can be used for adaptive radiation therapy.

Synthetic CT-aided Online CBCT Multi-Organ Segmentation for CBCT-guided Adaptive Radiotherapy of Pancreatic Cancer

Synthetic CT-aided Online CBCT Multi-Organ Segmentation for CBCT-guided Adaptive Radiotherapy of Pancreatic Cancer

Dosimetric study on learning-based cone-beam CT correction in adaptive radiation therapy

IntroductionCone-beam CT (CBCT) image quality is important for its quantitative analysis in adaptive radiation therapy. However, due to severe artifacts, the CBCTs are primarily used for verifying patient setup only so far. We have developed a learning-based image quality improvement method which could provide CBCTs with image quality comparable to planning CTs (pCTs). The accuracy of dose calculations based on these CBCTs is unknown. In this study, we aim to investigate the dosimetric accuracy of our corrected CBCT (CCBCT) in brain stereotactic radiosurgery (SRS) and pelvic radiotherapy. Materials and MethodsWe retrospectively investigated a total of 32 treatment plans from 22 patients, each of whom with both original treatment pCTs and CBCTs acquired during treatment setup. The CCBCT and original CBCT (OCBCT) were registered to the pCT for generating CCBCT-based and OCBCT-based treatment plans. The original pCT-based plans served as ground truth. Clinically-relevant dose volume histogram (DVH) metrics were extracted from the ground truth, OCBCT-based and CCBCT-based plans for comparison. Gamma analysis was also performed to compare the absorbed dose distributions between the pCT-based and OCBCT/CCBCT-based plans of each patient. ResultsCCBCTs demonstrated better image contrast and more accurate HU ranges when compared side-by-side with OCBCTs. For pelvic radiotherapy plans, the mean dose error in DVH metrics for planning target volume (PTV), bladder and rectum was significantly reduced, from 1% to 0.3%, after CBCT correction. The gamma analysis showed the average pass rate increased from 94.5% before correction to 99.0% after correction. For brain SRS treatment plans, both original and corrected CBCT images were accurate enough for dose calculation, though CCBCT featured higher image quality. ConclusionCCBCTs can provide a level of dose accuracy comparable to traditional pCTs for brain and prostate radiotherapy planning and the correction method proposed here can be useful in CBCT-guided adaptive radiotherapy.

Learning-based CBCT correction using alternating random forest based on auto-context model.

Quantitative Cone Beam CT (CBCT) imaging is increasing in demand for precise image-guided radiotherapy because it provides a foundation for advanced image-guided techniques, including accurate treatment setup, online tumor delineation, and patient dose calculation. However, CBCT is currently limited only to patient setup in the clinic because of the severe issues in its image quality. In this study, we develop a learning-based approach to improve CBCT's image quality for extended clinical applications. An auto-context model is integrated into a machine learning framework to iteratively generate corrected CBCT (CCBCT) with high-image quality. The first step is data preprocessing for the built training dataset, in which uninformative image regions are removed, noise is reduced, and CT and CBCT images are aligned. After a CBCT image is divided into a set of patches, the most informative and salient anatomical features are extracted to train random forests. Within each patch, alternating RF is applied to create a CCBCT patch as the output. Moreover, an iterative refinement strategy is exercised to enhance the image quality of CCBCT. Then, all the CCBCT patches are integrated to reconstruct final CCBCT images. The learning-based CBCT correction algorithm was evaluated using the leave-one-out cross-validation method applied on a cohort of 12 patients' brain data and 14 patients' pelvis data. The mean absolute error (MAE), peak signal-to-noise ratio (PSNR), normalized cross-correlation (NCC) indexes, and spatial nonuniformity (SNU) in the selected regions of interest (ROIs) were used to quantify the proposed algorithm's correction accuracy and generat the following results: mean MAE=12.81±2.04 and 19.94±5.44HU, mean PSNR=40.22±3.70 and 31.31±2.85dB, mean NCC=0.98±0.02 and 0.95±0.01, and SNU=2.07±3.36% and 2.07±3.36% for brain and pelvis data. Preliminary results demonstrated that the novel learning-based correction method can significantly improve CBCT image quality. Hence, the proposed algorithm is of great potential in improving CBCT's image quality to support its clinical utility in CBCT-guided adaptive radiotherapy.

Improving Image Quality of Cone-Beam CT Using Alternating Regression Forest.

We propose a CBCT image quality improvement method based on anatomic signature and auto-context alternating regression forest. Patient-specific anatomical features are extracted from the aligned training images and served as signatures for each voxel. The most relevant and informative features are identified to train regression forest. The well-trained regression forest is used to correct the CBCT of a new patient. This proposed algorithm was evaluated using 10 patients' data with CBCT and CT images. The mean absolute error (MAE), peak signal-to-noise ratio (PSNR) and normalized cross correlation (NCC) indexes were used to quantify the correction accuracy of the proposed algorithm. The mean MAE, PSNR and NCC between corrected CBCT and ground truth CT were 16.66HU, 37.28dB and 0.98, which demonstrated the CBCT correction accuracy of the proposed learning-based method. We have developed a learning-based method and demonstrated that this method could significantly improve CBCT image quality. The proposed method has great potential in improving CBCT image quality to a level close to planning CT, therefore, allowing its quantitative use in CBCT-guided adaptive radiotherapy.