Search

ObjectiveFast and accurate deformable image registration (DIR) that include DIR uncertainty estimation are essential for safe and reliable clinical deployment. While recent deep learning models have shown promise in predicting DIR with its uncertainty, challenges persist in proper uncertainty evaluation and hyperparameter optimization for these methods. This work aims to develop and evaluate a model that can perform fast DIR and predict its uncertainty in seconds.ApproachIn this study, we introduce a novel probabilistic multi-resolution image registration model utilizing convolutional neural networks (CNNs) to estimate a multivariate normal distributed dense displacement field (DDF) in a multimodal image registration problem. To assess the quality of the DDF distribution predicted by the model, we propose a new metric based on the Kullback-Leibler divergence. The performance of our approach was evaluated against three other DIR algorithms (VoxelMorph, Monte Carlo Drop-Out, and Monte Carlo B-splines) capable of predicting uncertainty. The evaluation of the models included not only the quality of the deformation but also the reliability of the estimated uncertainty. Our application investigated registration of a treatment planning computed tomography (CT) to follow-up cone beam CT for daily adaptive radiotherapy.Main resultsThe hyperparameter tuning of the models showed that there is a trade-off between the reliability of the estimated uncertainty and the accuracy of the deformation. In the optimal trade-off our model excelled in contour propagation and uncertainty estimation (p < 0.01) compared to existing uncertainty estimation models. We obtained an average dice similarity coefficient of 0.89 and a KL-divergence of 0.15.SignificanceBy addressing challenges in DIR uncertainty estimation and evaluation, our work showed that both the DIR and its uncertainty can be reliably predicted paving the way for safe deployment in clinical environment..

Radiotherapy, a crucial technique in cancer therapy, has traditionally relied on the premise of largely unchanging patient anatomy during the treatment course and encompassing uncertainties by target margins. This review introduces adaptive radiotherapy (ART), a notable innovation that addresses anatomy changes and optimizes the therapeutic ratio. ART utilizes advanced imaging techniques such as CT, MRI, and PET to modify the treatment plan based on observed anatomical changes and even biological changes during the course of treatment. The narrative review provides a comprehensive guide on ART for healthcare professionals and trainees in radiation oncology and anyone else interested in the topic. The incorporation of artificial intelligence in ART has played a crucial role in improving effectiveness, particularly in contour segmentation, treatment planning, and quality assurance. This has expedited the process to render online ART feasible, lowered the burden for radiation oncology practitioners, and enhanced the precision of dynamically personalized treatment. Current technical and clinical progress on ART is discussed in this review, highlighting the ongoing development of imaging technologies and AI and emphasizing their contribution to enhancing the applicability and effectiveness of ART.

This article examines the critical role of fast Monte Carlo (MC) dose calculations in advancing proton therapy techniques, particularly in the context of increasing treatment customization and precision. As adaptive radiotherapy and other patient-specific approaches evolve, the need for accurate and precise dose calculations, essential for techniques like proton-based stereotactic radiosurgery, becomes more prominent. These calculations, however, are time-intensive, with the treatment planning/optimization process constrained by the achievable speed of dose computations. Thus, enhancing the speed of MC methods is vital, as it not only facilitates the implementation of novel treatment modalities but also leads to more optimal treatment plans. Today, the state-of-the-art in MC dose calculation speeds is 106–107 protons per second. This review highlights the latest advancements in fast MC dose calculations that have led to such speeds, including emerging artificial intelligence-based techniques, and discusses their application in both current and emerging proton therapy strategies. [ABSTRACT FROM AUTHOR]

Objective. This study describes geometry-based and intensity-based tools for quality assurance (QA) of automatically generated structures for online adaptive radiotherapy, and designs an operator-independent traffic light system that identifies erroneous structure sets. Approach. A cohort of eight head and neck (HN) patients with daily CBCTs was selected for test development. Radiotherapy contours were propagated from planning computed tomography (CT) to daily cone beam CT (CBCT) using deformable image registration. These propagated structures were visually verified for acceptability. For each CBCT, several error scenarios were used to generate what were judged unacceptable structures. Ten additional HN patients with daily CBCTs and different error scenarios were selected for validation. A suite of tests based on image intensity, intensity gradient, and structure geometry was developed using acceptable and unacceptable HN planning structures. Combinations of one test applied to one structure, referred to as structure-test combinations, were selected for inclusion in the QA system based on their discriminatory power. A traffic light system was used to aggregate the structure-test combinations, and the system was evaluated on all fractions of the ten validation HN patients. Results. The QA system distinguished between acceptable and unacceptable fractions with high accuracy, labeling 294/324 acceptable fractions as green or yellow and 19/20 unacceptable fractions as yellow or red. Significance. This study demonstrates a system to supplement manual review of radiotherapy planning structures. Automated QA is performed by aggregating results from multiple intensity- and geometry-based tests. [ABSTRACT FROM AUTHOR]

Background: Magnetic resonance-guided adaptive radiotherapy (MRgART) at MR-Linac allows for plan optimisation on the MR-based synthetic CT (sCT) images, adjusting the target and organs at risk according to the patient's daily anatomy. Conversely, conventional linac image-guided radiotherapy (IGRT) involves rigid realignment of regions of interest to the daily anatomy, followed by the delivery of the reference computed tomography (CT) plan. This study aims to evaluate the effectiveness of MRgART versus IGRT for rectal cancer patients undergoing short-course radiotherapy, while also assessing the dose accumulation process to support the findings and determine its usefulness in enhancing treatment accuracy. Methods: Nineteen rectal cancer patients treated with a 1.5 Tesla MR-Linac with a prescription dose of 25 Gy (5 Gy x 5) and undergoing daily adapted radiotherapy by plan optimization based on online MR-based sCT images, were included in this retrospective study. For each adapted plan (), a second plan () was generated by recalculating the reference CT plan on the daily MR-based sCT images after rigid registration with the reference CT images to simulate the IGRT workflow. Dosimetry of and was compared for each fraction. Cumulative doses on the first and last fractions were evaluated for both workflows. The dosimetry per single fraction and the cumulative doses were compared using dose-volume histogram parameters. Results: Ninety-five fractions delivered with MRgART were compared to corresponding simulated IGRT fractions. All MRgART fractions fulfilled the target clinical requirements. IGRT treatments did not meet the expected target coverage for 63 out of 94 fractions (67.0%), with 13 fractions showing a V95 median point percentage decrease of 2.78% (range, 1.65-4.16%), and 55 fractions exceeding the V107% threshold with a median value of 15.4 cc (range, 6.0-43.8 cc). For the bladder, the median values were 18.18 Gy for the adaptive fractions and 19.60 Gy for the IGRT fractions. Similarly the median values for the small bowel were 23.40 Gy and 25.69 Gy, respectively. No statistically significant differences were observed in the doses accumulated on the first or last fraction for the adaptive workflow, with results consistent with the single adaptive fractions. In contrast, accumulated doses in the IGRT workflow showed significant variations mitigating the high dose constraint, nevertheless, more than half of the patients still did not meet clinical requirements. Conclusions: MRgART for short-course rectal cancer treatments ensures that the dose delivered matches each fraction of the planned dose and the results are confirmed by the dose accumulation process, which therefore seems redundant. In contrast, IGRT may lead to target dose discrepancies and non-compliance with organs at risk constraints and dose accumulation can still highlight notable dosimetric differences. [ABSTRACT FROM AUTHOR]

Magnetic resonance imaging (MRI)-guided radiation therapy has proven to provide many benefits such as real-time tracking, dose escalation, and the ability to perform online adaptive therapy. The objective of this study is to compare curative treatment plans for glioblastoma tumors on a low-field MR-guided linac vs a C-arm linac and evaluate if they are comparable in terms of coverage, organ at risk sparing, delivery time, and deliverability. This is a retrospective study that consisted of 15 previously treated patients who received radiation therapy for glioblastoma on a C-arm linac. The CT simulation data used for the original clinical plans was imported into the MR-linac treatment planning system (TPS) and utilized for MR-linac plan generation. The plans were evaluated utilizing the dose volumetric histogram (DVH) and isodose lines, then compared in terms of plan quality consisting of PTV coverage, dose distributions, and OAR constraints. Statistical analysis was performed to compare differences between the two planning techniques. QA was performed on a subset of the plans to verify deliverability. Plans generated on the MR-linac were more heterogenous compared to C-arm linac plans. A statistically significant difference was found in the homogeneity index (HI) and the PTV V105% volume (cc) values. The volume of the normal brain receiving 30 Gy also showed a statistically significant (p = 0.0479) difference, where on average an additional 41.5 cc's of the normal brain tissue received 30 Gy in the MR-linac plans. The maximum dose to the normal brain structure also increased in the MR-linac plans on average by 2.6 Gy (p = 0.0002). Similarly, the average maximum dose to the scalp 4 mm structure was 6.5 Gy higher in the MR-linac plans compared to C-arm linac plans (p = 0.0103). The total MU's were higher in the MR-linac plans compared to the C-arm linac plans (p = 0.0015). Both MR-linac and C-arm linac plans met constraints for PTV coverage and OAR sparing, were deliverable, and resulted to be clinically acceptable. However, our study showed that MR-linac plans were not as conformal or as homogenous as C-arm linac plans utilizing noncoplanar beams. [ABSTRACT FROM AUTHOR]

We proposed a deep learning approach to classify various error types in daily VMAT treatment of head and neck cancer patients based on EPID dosimetry, which could provide additional information to support clinical decisions for adaptive planning. 146 arcs from 42 head and neck patients were analyzed. Anatomical changes and setup errors were simulated in 17,820 EPID images of 99 arcs obtained from 30 patients using in-house software for model training, validation, and testing. Subsequently, 141 clinical EPID images from 47 arcs belonging to the remaining 12 patients were utilized for clinical testing. The hierarchical convolutional neural network (HCNN) model was trained to classify error types and magnitudes using EPID dose difference maps. Gamma analysis with 3%/2 mm (dose difference/distance to agreement) criteria was also performed. The F1 score, a combination of precision and recall, was utilized to evaluate the performance of the HCNN model and gamma analysis. The adaptive fractioned doses were calculated to verify the HCNN classification results. For error type identification, the overall F1 score of the HCNN model was 0.99 and 0.91 for primary type and subtype identification, respectively. For error magnitude identification, the overall F1 score in the simulation dataset was 0.96 and 0.70 for the HCNN model and gamma analysis, respectively; while the overall F1 score in the clinical dataset was 0.79 and 0.20 for the HCNN model and gamma analysis, respectively. The HCNN model-based EPID dosimetry can identify changes in patient transmission doses and distinguish the treatment error category, which could potentially provide information for head and neck cancer treatment adaption. [ABSTRACT FROM AUTHOR]

Objective. Online adaptive radiation therapy requires fast and automated contouring of daily scans for treatment plan re-optimization. However, automated contouring is imperfect and introduces contour uncertainties. This work aims at developing and comparing robust optimization strategies accounting for such uncertainties. Approach. A deep-learning method was used to predict the uncertainty of deformable image registration, and to generate a finite set of daily contour samples. Ten optimization strategies were compared: two baseline methods, five methods that convert contour samples into voxel-wise probabilities, and three methods accounting explicitly for contour samples as scenarios in robust optimization. Target coverage and organ-at-risk (OAR) sparing were evaluated robustly for simplified proton therapy plans for five head-and-neck cancer patients. Results. We found that explicitly including target contour uncertainty in robust optimization provides robust target coverage with better OAR sparing than the baseline methods, without increasing the optimization time. Although OAR doses first increased when increasing target robustness, this effect could be prevented by additionally including robustness to OAR contour uncertainty. Compared to the probability-based methods, the scenario-based methods spared the OARs more, but increased integral dose and required more computation time. Significance. This work proposed efficient and beneficial strategies to mitigate contour uncertainty in treatment plan optimization. This facilitates the adoption of automatic contouring in online adaptive radiation therapy and, more generally, enables mitigation also of other sources of contour uncertainty in treatment planning. [ABSTRACT FROM AUTHOR]

Background: Cone-beam computed tomography (CBCT) is a convenient method for adaptive radiation therapy (ART), but its application is often hindered by its image quality. We aim to develop a unified deep learning model that can consistently enhance the quality of CBCT images across various anatomical sites by generating synthetic CT (sCT) images. Methods: A dataset of paired CBCT and planning CT images from 135 cancer patients, including head and neck, chest and abdominal tumors, was collected. This dataset, with its rich anatomical diversity and scanning parameters, was carefully selected to ensure comprehensivemodel training. Due to the imperfect registration, the inherent challenge of local structural misalignment of paired datasetmay lead to suboptimal model performance. To address this limitation, we propose SynREG, a supervised learning framework. SynREG integrates a hybrid CNN-transformer architecture designed for generating high-fidelity sCT images and a registration network designed to correct local structural misalignment dynamically during training. An independent test set of 23 additional patients was used to evaluate the image quality, and the results were compared with those of several benchmark models (pix2pix, cycleGAN and SwinIR). Furthermore, the performance of an autosegmentation application was also assessed. Results: The proposed model disentangled sCT generation from anatomical correction, leading to a more rational optimization process. As a result, the model effectively suppressed noise and artifacts in multisite applications, significantly enhancing CBCT image quality. Specifically, the mean absolute error (MAE) of SynREG was reduced to 16.81 ± 8.42 HU, whereas the structural similarity index (SSIM) increased to 94.34 ± 2.85%, representing improvements over the raw CBCT data, which had the MAE of 26.74 ± 10.11 HU and the SSIM of 89.73 ± 3.46%. The enhanced image quality was particularly beneficial for organs with low contrast resolution, significantly increasing the accuracy of automatic segmentation in these regions. Notably, for the brainstem, the mean Dice similarity coefficient (DSC) increased from 0.61 to 0.89, and the MDA decreased from 3.72 mm to 0.98 mm, indicating a substantial improvement in segmentation accuracy and precision. Conclusions: SynREG can effectively alleviate the differences in residual anatomy between paired datasets and enhance the quality of CBCT images. [ABSTRACT FROM AUTHOR]

Background and purpose: Cone beam computed tomography (CBCT) is routinely used in radiotherapy to localize target volume. The aim of our study was to determine the biological effects of CBCT dose compared to subsequent therapeutic dose by using in vitro chromosome dosimetry. Materials and methods: Peripheral blood samples from five healthy volunteers were irradiated in two phantoms (water filled in-house made cylindrical, and Pure Image CTDI phantoms) with 6 MV FFF X-ray photons, the dose rate was 800 MU/min and the absorbed doses ranged from 0.5 to 8 Gy. Irradiation was performed with a 6 MV linear accelerator (LINAC) to generate a dose–response calibration curve. In the first part of the investigation, 1–5 CBCT imaging was used, in the second, only 2 Gy doses were delivered with a LINAC, and then, in the third part, a combination of CBCT and 2 Gy irradiation was performed mimicking online adapted radiotherapy treatment. Metaphases were prepared from lymphocyte cultures, using standard cytogenetic techniques, and chromosomal aberrations were evaluated. Estimate doses were calculated from chromosome aberrations using dose–response curves. Results: Samples exposed to X-ray from CBCT imaging prior to treatment exhibited higher chromosomal aberrations and Estimate dose than the 2 Gy therapeutic (real) dose, and the magnitude of the increase depended on the number of CBCTs: 1–5 CBCT corresponded to 0.04–0.92 Gy, 1 CBCT + 2 Gy to 2.32 Gy, and 5 CBCTs + 2 Gy to 3.5 Gy. Conclusion: The estimated dose based on chromosomal aberrations is 24.8% higher than the physical dose, for the combination of 3 CBCTs and the therapeutic 2 Gy dose, which should be taken into account when calculating the total therapeutic dose that could increase the risk of a second cancer. The clinical implications of the combined radiation effect may require further investigation. [ABSTRACT FROM AUTHOR]

Simple Summary: MR-Linac is a novel magnetic resonance imaging (MRI)-guided radiotherapy (IGRT) that combines MRI with a linear accelerator (Linac). Although radiation therapy (RT) for lung cancer has traditionally been managed with a computed tomography (CT)-based workflow, an MR-Linac-based workflow would be able to address the many limitations of current practice. This narrative review summarizes the latest developments in MR-Linac lung cancer treatment, as well as its boundaries. Future research directions are also highlighted. Compared with computed tomography (CT), magnetic resonance imaging (MRI) traditionally plays a very limited role in lung cancer management, although there is plenty of room for improvement in the current CT-based workflow, for example, in structures such as the brachial plexus and chest wall invasion, which are difficult to visualize with CT alone. Furthermore, in the treatment of high-risk tumors such as ultracentral lung cancer, treatment-associated toxicity currently still outweighs its benefits. The advent of MR-Linac, an MRI-guided radiotherapy (RT) that combines MRI with a linear accelerator, could potentially address these limitations. Compared with CT-based technologies, MR-Linac could offer superior soft tissue visualization, daily adaptive capability, real-time target tracking, and an early assessment of treatment response. Clinically, it could be especially advantageous in the treatment of central/ultracentral lung cancer, early-stage lung cancer, and locally advanced lung cancer. Increasing demands for stereotactic body radiotherapy (SBRT) for lung cancer have led to MR-Linac adoption in some cancer centers. In this review, a broad overview of the latest research on imaging-guided radiotherapy (IGRT) with MR-Linac for lung cancer management is provided, and development pertaining to artificial intelligence is also highlighted. New avenues of research are also discussed. [ABSTRACT FROM AUTHOR]

The ability to visualise cancer with imaging has been crucial to the evolution of modern radiotherapy (RT) planning and delivery. And as evolving RT technologies deliver increasingly precise treatment, the importance of accurate identification and delineation of disease assumes ever greater significance. However, innovation in imaging technology has matched that seen with RT delivery platforms, and novel imaging techniques are a focus of much research activity. How these imaging modalities may alter and improve the diagnosis and staging of cancer is an important question, but already well served by the literature. What is less clear is how novel imaging techniques may influence and improve practical and technical aspects of RT planning and delivery. In this review, current gold standard approaches to integration of imaging, and potential future applications of bleeding-edge imaging technology into RT planning pathways are explored. • Novel imaging techniques could influence and improve radiotherapy planning. • Advances in CT relevant for adaptive radiotherapy and proton beam therapy. • Functional MRI techniques may be useful for target volume definition. • Novel PET tracers such as FAPI, PMSA and FAZA may augment radiotherapy planning. • Emerging clinical data for benefits of MR-Linac. [ABSTRACT FROM AUTHOR]

Dose calculation for MR-guided radiotherapy (MRgRT) at the 0.35 T MR-Linac is currently based on deformation of planning CTs (defCT) acquired for each patient. We present a simple and robust bulk density overwrite synthetic CT (sCT) method for abdominal treatments in order to streamline clinical workflows. Fifty-six abdominal patient treatment plans were retrospectively evaluated. All patients had been treated at the MR-Linac using MR datasets for treatment planning and plan adaption and defCT for dose calculation. Bulk density CTs (4M-sCT) were generated from MR images with four material compartments (bone, lung, air, soft tissue). The relative electron densities (RED) for bone and lung were extracted from contoured CT structure average REDs. For soft tissue, a correlation between BMI and RED was evaluated. Dose was recalculated on 4M-sCT and compared to dose distributions on defCTs assessing dose differences in the PTV and organs at risk (OAR). Mean RED of bone was 1.17 ± 0.02, mean RED of lung 0.17 ± 0.05. The correlation between BMI and RED for soft tissue was statistically significant (p < 0.01). PTV dose differences between 4M-sCT and defCT were D mean : −0.4 ± 1.0%, D 1% : −0.3 ± 1.1% and D 95% : −0.5 ± 1.0%. OARs showed D 2% : −0.3 ± 1.9% and D mean : −0.1 ± 1.4% differences. Local 3D gamma index pass rates (2%/2mm) between dose calculated using 4M-sCT and defCT were 96.8 ± 2.6% (range 89.9–99.6%). The presented method for sCT generation enables precise dose calculation for MR-only abdominal MRgRT. [ABSTRACT FROM AUTHOR]

Purpose: Online adaptive radiotherapy relies on a high degree of automation to enable rapid planning procedures. The Varian Ethos intelligent optimization engine (IOE) was originally designed for conventional treatments so it is crucial to provide clear guidance for lung SAbR plans. This study investigates using the Ethos IOE together with adaptive‐specific optimization tuning structures we designed and templated within Ethos to mitigate inter‐planner variability in meeting RTOG metrics for both online‐adaptive and offline SAbR plans. Methods: We developed a planning strategy to automate the generation of tuning structures and optimization. This was validated by retrospective analysis of 35 lung SAbR cases (total 105 fractions) treated on Ethos. The effectiveness of our planning strategy was evaluated by comparing plan quality with‐and‐without auto‐generated tuning structures. Internal target volume (ITV) contour was compared between that drawn from CT simulation and from cone‐beam CT (CBCT) at time of treatment to verify CBCT image quality and treatment effectiveness. Planning strategy robustness for lung SAbR was quantified by frequency of plans meeting reference plan RTOG constraints. Results: Our planning strategy creates a gradient within the ITV with maximum dose in the core and improves intermediate dose conformality on average by 2%. ITV size showed no significant difference between those contoured from CT simulation and first fraction, and also trended towards decreasing over course of treatment. Compared to non‐adaptive plans, adaptive plans better meet reference plan goals (37% vs. 100% PTV coverage compliance, for scheduled and adapted plans) while improving plan quality (improved GI (gradient index) by 3.8%, CI (conformity index) by 1.7%). Conclusion: We developed a robust and readily shareable planning strategy for the treatment of adaptive lung SAbR on the Ethos system. We validated that automatic online plan re‐optimization along with the formulated adaptive tuning structures can ensure consistent plan quality. With the proposed planning strategy, highly ablative treatments are feasible on Ethos. [ABSTRACT FROM AUTHOR]

Background: Radical radiotherapy for muscle-invasive bladder cancer (MIBC) is challenging due to large variations in bladder shape, size and volume during treatment, with drinking protocols often employed to mitigate geometric uncertainties. Utilising adaptive radiotherapy together with CBCT imaging to select a treatment plan that best fits the bladder target and reduce normal tissue irradiation is an attractive option to compensate for anatomical changes. The aim of this retrospective study was to compare a bladder empty (BE) protocol to a bladder filling (BF) protocol with regards to variations in target volumes, plan of the day (PoD) selection and plan dosimetry throughout treatment. Methods: Forty patients were included in the study; twenty were treated with a BE protocol and twenty with a BF protocol to a total prescribed dose of 55 Gy in 20 fractions. Small, medium and large bladder plans were generated using three different CTV to PTV margins. Bladder (CTV) volumes were delineated on planning CTs and online pre-treatment CBCTs. Differences in CTV volumes throughout treatment, plan selection, PTV volumes and resulting dose metrics were compared for both protocols. Results: Mean bladder volume differed significantly on both the planning CTs and online pre-treatment CBCTs between the protocols (p < 0.05). Significant differences in bladder volumes were observed between the planning CT and pre-treatment CBCTs for BF (p < 0.05) but not for BE (p = 0.11). Both protocols saw a significant decrease in bladder volume between first and final treatment fractions (p < 0.05). Medium plans were preferentially selected for BE whilst when using the BF protocol the small plan was chosen most frequently. With no significant change to PTV coverage between the protocols, the volume of body receiving 25.0–45.8 Gy was found to be significantly smaller for BE patients (p < 0.05). Conclusions: This work provides evidence in favour of a BE protocol compared to a BF protocol for radical radiotherapy for MIBC. The smaller treatment volumes observed in the BE protocol led to reduced OAR and total body doses and were also observed to be more consistent throughout the treatment course. These results highlight improvements in dosimetry for patients who undergo a BE protocol for MIBC. [ABSTRACT FROM AUTHOR]

Background: Chemoradiation in head and neck carcinoma (HNC) shows significant anatomical resulting in erroneous dose deposition in the target or the organ at risk (OAR). Adaptive radiotherapy (ART) can overcome this. Timing of significant target and OAR changes with dosimetric impact; thus, most suitable time and frequency of ART is unclear. Methods: This dosimetric study used prospective weekly non‐contrast CT scans in 12 HNC patients (78 scans). OARs and TVs were manually contoured after registration with simulation scan. Dose overlay done on each scan without reoptimization. Dosimetric and volumetric variations assessed. Results: Commonest site was oropharynx. Gross Tumor Volume (GTV) reduced from 47.5 ± 19.2 to 17.8 ± 10.7 cc. Nodal GTV reduced from 15.7 ± 18.8 to 4.7 ± 7.1 cc. Parotid showed mean volume loss of 35%. T stage moderately correlated with GTV regression. Conclusion: Maximum GTV changes occurred after 3 weeks. Best time to do single fixed interval ART would be by the end of 3 weeks. [ABSTRACT FROM AUTHOR]

Motivation: Online adaptive radiotherapy with Ethos is based on the anatomy determined from daily cone beam computed tomography (CBCT) images. Dose optimization and computation are performed on the density map of a synthetic CT (sCT), a deformable registration of the initial planning CT (pCT) onto the current CBCT. Large density changes as present in the lung region are challenging the system. Methods: Treatment plans for Ethos were created and delivered for 1, 2, and 3 cm diameter lung lesions in an anthropomorphic phantom, combining different insets in the pCT and during adaptive and non‐adaptive treatment sessions. Primary and secondary dose calculations as well as back‐projected dose from portal images were evaluated. Results: Density changes due to changed insets were not considered in the sCTs. This resulted in errors in the dose; for example, ‐15.9% of the mean dose for a plan when changing from a 3 cm inset in the pCT to 1 cm at the time of treatment. Secondary dose calculation is based on the sCT and could therefore not reveal these dose errors. However, dose calculation on the CBCT, either as a recalculation in the treatment planning system or as pre‐treatment quality assurance (QA) before the treatment, indicated the differences. EPID in‐vivo QA also reported discrepancies between calculated and delivered dose distributions. Conclusions: An incorrect density distribution in the sCT has an impact on the dose calculation accuracy in the adaptive treatment workflow with the Ethos system. Additional quality checks of the sCT can detect such errors. [ABSTRACT FROM AUTHOR]

Purpose: Dose calculation for MR-guided radiotherapy (MRgRT) at the 0.35 T MR-Linac is currently based on deformation of planning CTs (defCT) acquired for each patient. We present a simple and robust bulk density overwrite synthetic CT (sCT) method for abdominal treatments in order to streamline clinical workflows. Method: Fifty-six abdominal patient treatment plans were retrospectively evaluated. All patients had been treated at the MR-Linac using MR datasets for treatment planning and plan adaption and defCT for dose calculation. Bulk density CTs (4M-sCT) were generated from MR images with four material compartments (bone, lung, air, soft tissue). The relative electron densities (RED) for bone and lung were extracted from contoured CT structure average REDs. For soft tissue, a correlation between BMI and RED was evaluated. Dose was recalculated on 4M-sCT and compared to dose distributions on defCTs assessing dose differences in the PTV and organs at risk (OAR). Results: Mean RED of bone was 1.17 ± 0.02, mean RED of lung 0.17 ± 0.05. The correlation between BMI and RED for soft tissue was statistically significant (p

Abstract Background Radical radiotherapy for muscle-invasive bladder cancer (MIBC) is challenging due to large variations in bladder shape, size and volume during treatment, with drinking protocols often employed to mitigate geometric uncertainties. Utilising adaptive radiotherapy together with CBCT imaging to select a treatment plan that best fits the bladder target and reduce normal tissue irradiation is an attractive option to compensate for anatomical changes. The aim of this retrospective study was to compare a bladder empty (BE) protocol to a bladder filling (BF) protocol with regards to variations in target volumes, plan of the day (PoD) selection and plan dosimetry throughout treatment. Methods Forty patients were included in the study; twenty were treated with a BE protocol and twenty with a BF protocol to a total prescribed dose of 55 Gy in 20 fractions. Small, medium and large bladder plans were generated using three different CTV to PTV margins. Bladder (CTV) volumes were delineated on planning CTs and online pre-treatment CBCTs. Differences in CTV volumes throughout treatment, plan selection, PTV volumes and resulting dose metrics were compared for both protocols. Results Mean bladder volume differed significantly on both the planning CTs and online pre-treatment CBCTs between the protocols (p

Abstract This study employed a commercial software velocity to perform deformable registration and dose calculation on deformed CT images, aiming to assess the accuracy of dose delivery during the radiotherapy for lung cancers. A total of 20 patients with lung cancer were enrolled in this study. Adaptive CT (ACT) was generated by deformed the planning CT (pCT) to the CBCT of initial radiotherapy fraction, followed by contour propagation and dose recalculation. There was not significant difference between volumes of GTV and CTV calculated from the ACT and pCT. However, significant differences in dice similarity coefficient (DSC) and coverage ratio (CR) between GTV and CTV were observed, with lower values for GTV volumes below 15 cc. The mean differences in dose corresponding to 95% of the GTV, GTV-P, CTV, and CTV-P between ACT and pCT were − 0.32%, 4.52%, 2.17%, and 4.71%, respectively. For the dose corresponding to 99%, the discrepancies were − 0.18%, 8.35%, 1.92%, and 24.96%, respectively. These differences in dose primarily appeared at the edges of the target areas. Notably, a significant enhancement of dose corresponding to 1 cc for spinal cord was observed in ACT, compared with pCT. There was no statistical difference in the mean dose of lungs and heart. In general, for lung cancer patients, anatomical motion may result in both CTV and GTV moving outside the original irradiation region. The dose difference within the original target area was small, but the difference in the planning target area was considerable.

Background and purpose: MR-guided radiotherapy (MRgRT) online plan adaptation accounts for tumor volume changes, interfraction motion and thus allows daily sparing of relevant organs at risk. Due to the high interfraction variability of bladder and rectum, patients with tumors in the pelvic region may strongly benefit from adaptive MRgRT. Currently, fast automatic annotation of anatomical structures is not available within the online MRgRT workflow. Therefore, the aim of this study was to train and validate a fast, accurate deep learning model for automatic MRI segmentation at the MR-Linac for future implementation in a clinical MRgRT workflow. Materials and methods: For a total of 47 patients, T2w MRI data were acquired on a 1.5 T MR-Linac (Unity, Elekta) on five different days. Prostate, seminal vesicles, rectum, anal canal, bladder, penile bulb, body and bony structures were manually annotated. These training data consisting of 232 data sets in total was used for the generation of a deep learning based autocontouring model and validated on 20 unseen T2w-MRIs. For quantitative evaluation the validation set was contoured by a radiation oncologist as gold standard contours (GSC) and compared in MATLAB to the automatic contours (AIC). For the evaluation, dice similarity coefficients (DSC), and 95% Hausdorff distances (95% HD), added path length (APL) and surface DSC (sDSC) were calculated in a caudal-cranial window of ± 4 cm with respect to the prostate ends. For qualitative evaluation, five radiation oncologists scored the AIC on the possible usage within an online adaptive workflow as follows: (1) no modifications needed, (2) minor adjustments needed, (3) major adjustments/ multiple minor adjustments needed, (4) not usable. Results: The quantitative evaluation revealed a maximum median 95% HD of 6.9 mm for the rectum and minimum median 95% HD of 2.7 mm for the bladder. Maximal and minimal median DSC were detected for bladder with 0.97 and for penile bulb with 0.73, respectively. Using a tolerance level of 3 mm, the highest and lowest sDSC were determined for rectum (0.94) and anal canal (0.68), respectively.Qualitative evaluation resulted in a mean score of 1.2 for AICs over all organs and patients across all expert ratings. For the different autocontoured structures, the highest mean score of 1.0 was observed for anal canal, sacrum, femur left and right, and pelvis left, whereas for prostate the lowest mean score of 2.0 was detected. In total, 80% of the contours were rated be clinically acceptable, 16% to require minor and 4% major adjustments for online adaptive MRgRT. Conclusion: In this study, an AI-based autocontouring was successfully trained for online adaptive MR-guided radiotherapy on the 1.5 T MR-Linac system. The developed model can automatically generate contours accepted by physicians (80%) or only with the need of minor corrections (16%) for the irradiation of primary prostate on the clinically employed sequences.

Purpose: During the radiation treatment planning process, one of the time-consuming procedures is the final high-resolution dose calculation, which obstacles the wide application of the emerging online adaptive radiotherapy techniques (OLART). There is an urgent desire for highly accurate and efficient dose calculation methods. This study aims to develop a dose super resolution-based deep learning model for fast and accurate dose prediction in clinical practice. Method: A Multi-stage Dose Super-Resolution Network (MDSR Net) architecture with sparse masks module and multi-stage progressive dose distribution restoration method were developed to predict high-resolution dose distribution using low-resolution data. A total of 340 VMAT plans from different disease sites were used, among which 240 randomly selected nasopharyngeal, lung, and cervix cases were used for model training, and the remaining 60 cases from the same sites for model benchmark testing, and additional 40 cases from the unseen site (breast and rectum) was used for model generalizability evaluation. The clinical calculated dose with a grid size of 2 mm was used as baseline dose distribution. The input included the dose distribution with 4 mm grid size and CT images. The model performance was compared with HD U-Net and cubic interpolation methods using Dose-volume histograms (DVH) metrics and global gamma analysis with 1%/1 mm and 10% low dose threshold. The correlation between the prediction error and the dose, dose gradient, and CT values was also evaluated. Results: The prediction errors of MDSR were 0.06–0.84% of Dmean indices, and the gamma passing rate was 83.1–91.0% on the benchmark testing dataset, and 0.02–1.03% and 71.3–90.3% for the generalization dataset respectively. The model performance was significantly higher than the HD U-Net and interpolation methods (p

This study employed a commercial software velocity to perform deformable registration and dose calculation on deformed CT images, aiming to assess the accuracy of dose delivery during the radiotherapy for lung cancers. A total of 20 patients with lung cancer were enrolled in this study. Adaptive CT (ACT) was generated by deformed the planning CT (pCT) to the CBCT of initial radiotherapy fraction, followed by contour propagation and dose recalculation. There was not significant difference between volumes of GTV and CTV calculated from the ACT and pCT. However, significant differences in dice similarity coefficient (DSC) and coverage ratio (CR) between GTV and CTV were observed, with lower values for GTV volumes below 15 cc. The mean differences in dose corresponding to 95% of the GTV, GTV-P, CTV, and CTV-P between ACT and pCT were − 0.32%, 4.52%, 2.17%, and 4.71%, respectively. For the dose corresponding to 99%, the discrepancies were − 0.18%, 8.35%, 1.92%, and 24.96%, respectively. These differences in dose primarily appeared at the edges of the target areas. Notably, a significant enhancement of dose corresponding to 1 cc for spinal cord was observed in ACT, compared with pCT. There was no statistical difference in the mean dose of lungs and heart. In general, for lung cancer patients, anatomical motion may result in both CTV and GTV moving outside the original irradiation region. The dose difference within the original target area was small, but the difference in the planning target area was considerable. [ABSTRACT FROM AUTHOR]

Simple Summary: Adrenal metastases are frequent in solid malignancies such as lung cancer and melanoma. Recent studies support the use of ablative local therapies in oligometastatic or oligoprogressive patients. Online adaptive MR-guided radiotherapy improves tumor coverage and organ-at-risk sparing compared to non-adaptive radiotherapy. However, clinical data, especially the long-term results of this promising technique, are still limited. We here report the long-term outcomes of a large single-center cohort. (1) Background: Recent publications foster stereotactic body radiotherapy (SBRT) in patients with adrenal oligometastases or oligoprogression. However, local control (LC) after non-adaptive SBRT shows the potential for improvement. Online adaptive MR-guided SBRT (MRgSBRT) improves tumor coverage and organ-at-risk (OAR) sparing. Long-term results of adaptive MRgSBRT are still sparse. (2) Methods: Adaptive MRgSBRT was performed on a 0.35 T MR-Linac. LC, overall survival (OS), progression-free survival (PFS), overall response rate (ORR), and toxicity were assessed. (3) Results: 35 patients with 40 adrenal metastases were analyzed. The median gross tumor volume was 30.6 cc. The most common regimen was 10 fractions at 5 Gy. The median biologically effective dose (BED10) was 75.0 Gy. Plan adaptation was performed in 98% of all fractions. The median follow-up was 7.9 months. One local failure occurred after 16.6 months, resulting in estimated LC rates of 100% at one year and 90% at two years. ORR was 67.5%. The median OS was 22.4 months, and the median PFS was 5.1 months. No toxicity > CTCAE grade 2 occurred. (4) Conclusions: LC and ORR after adrenal adaptive MRgSBRT were excellent, even in a cohort with comparably large metastases. A BED10 of 75 Gy seems sufficient for improved LC in comparison to non-adaptive SBRT. [ABSTRACT FROM AUTHOR]

Introduction: The ability to dynamically adjust target contours, derived Boolean structures, and ultimately, the optimized fluence is the end goal of online adaptive radiotherapy (ART). The purpose of this work is to describe the necessary tests to perform after a software patch installation and/or upgrade for an established online ART program. Methods: A patch upgrade on a low-field MR Linac system was evaluated for post-software upgrade quality assurance (QA) with current infrastructure of ART workflow on (1) the treatment planning system (TPS) during the initial planning stage and (2) the treatment delivery system (TDS), which is a TPS integrated into the delivery console for online ART planning. Online ART QA procedures recommended for post-software upgrade include: (1) user interface (UI) configuration; (2) TPS beam model consistency; (3) segmentation consistency; (4) dose calculation consistency; (5) optimizer robustness consistency; (6) CT density table consistency; and (7) end-to-end absolute ART dose and predicted dose measured including interruption testing. Differences of calculated doses were evaluated through DVH and/or 3D gamma comparisons. The measured dose was assessed using an MR-compatible A26 ionization chamber in a motion phantom. Segmentation differences were assessed through absolute volume and visual inspection. Results: (1) No UI configuration discrepancies were observed. (2) Dose differences on TPS pre-/post-software upgrade were within 1% for DVH metrics. (3) Differences in segmentation when observed were small in general, with the largest change noted for small-volume regions of interest (ROIs) due to partial volume impact. (4) Agreement between TPS and TDS calculated doses was 99.9% using a 2%/2-mm gamma criteria. (5) Comparison between TPS and online ART plans for a given patient plan showed agreement within 2% for targets and 0.6 cc for organs at risk. (6) Relative electron densities demonstrated comparable agreement between TPS and TDS. (7) ART absolute and predicted measured end-to-end doses were within 1% of calculated TDS. Discussion: An online ART QA program for post-software upgrade has been developed and implemented on an MR Linac system. Testing mechanics and their respective baselines may vary across institutions, but all necessary components for a post-software upgrade QA have been outlined and detailed. These outlined tests were demonstrated feasible for a low-field MR Linac system; however, the scope of this work may be applied and adapted more broadly to other online ART platforms. [ABSTRACT FROM AUTHOR]

The utility of Adaptive Radiotherapy (ART) in Head and Neck Squamous Cell Carcinoma (HNSCC) remains to be ascertained. While multiple retrospective and single-arm prospective studies have demonstrated its efficacy in decreasing parotid doses and reducing xerostomia, adequate randomized evidence is lacking. ReSTART (Reducing Salivary Toxicity with Adaptive Radiotherapy) is an ongoing phase III randomized trial of patients with previously untreated, locally advanced HNSCC of the oropharynx, larynx, and hypopharynx. Patients are randomized in a 1:1 ratio to the standard Intensity Modulated Radiotherapy (IMRT) arm {Planning Target Volume (PTV) margin 5 mm} vs. Adaptive Radiotherapy arm (standard IMRT with a PTV margin 3 mm, two planned adaptive planning at 10th and 20th fractions). The stratification factors include the primary site and nodal stage. The RT dose prescribed is 66Gy in 30 fractions for high-risk PTV and 54Gy in 30 fractions for low-risk PTV over six weeks, along with concurrent chemotherapy. The primary endpoint is to compare salivary toxicity between arms using salivary scintigraphy 12 months' post-radiation. To detect a 25% improvement in the primary endpoint at 12 months in the ART arm with a two-sided 5% alpha value and a power of 80% (and 10% attrition ratio), a sample size of 130 patients is required (65 patients in each arm). The secondary endpoints include acute and late toxicities, locoregional control, disease-free survival, overall survival, quality of life, and xerostomia scores between the two arms. The ReSTART trial aims to answer an important question in Radiation Therapy for HNSCC, particularly in a resource-limited setting. The uniqueness of this trial, compared to other ongoing randomized trials, includes the PTV margins and the xerostomia assessment by scintigraphy at 12 months as the primary endpoint. • The ReSTART, trial, evaluates the efficacy of adaptive radiotherapy in reducing xerostomia for HNSCC. • In the adaptive RT arm, two pre-planned adaptive planning are done at 10 and 20 fractions. • The CTV to PTV margin in the adaptive RT arm is 3 mm. • Primary hypothesis is a 25% benefit of 1-year salivary function, which is assessed by salivary scintigraphy. [ABSTRACT FROM AUTHOR]

Postoperative prostate radiotherapy requires large planning target volume (PTV) margins to account for motion and deformation of the prostate bed. Adaptive radiation therapy (ART) can incorporate image-guidance data to personalize PTVs that maintain coverage while reducing toxicity. We present feasibility and dosimetry results of a prospective study of postprostatectomy ART. Twenty-one patients were treated with single-adaptation ART. Conventional treatments were delivered for fractions 1 to 6 and adapted plans for the remaining 27 fractions. Clinical target volumes (CTVs) and small bowel delineated on fraction 1 to 4 CBCT were used to generate adapted PTVs and planning organ-at-risk (OAR) volumes for adapted plans. PTV volume and OAR dose were compared between ART and conventional using Wilcoxon signed-rank tests. Weekly CBCT were used to assess the fraction of CTV covered by PTV, CTV D99, and small bowel D 1cc. Clinical metrics were compared using a Student's t-test (p < 0.05 significant). Offline adaptive planning required 1.9 ± 0.4 days (mean ± SD). ART decreased mean adapted PTV volume 61 ± 37 cc and bladder wall D50 compared with conventional treatment (p < 0.01). The CTV was fully covered for 96% (97%) of fractions with ART (conventional). Reconstructing dose on weekly CBCT, a nonsignificant reduction in CTV D99 was observed with ART (94%) compared to conventional (96%). Reduced CTV D99 with ART was significantly correlated with large anterior-posterior rectal diameter on simulation CT. ART reduced the number of fractions exceeding our institution's small bowel D1c limit from 14% to 7%. This study has demonstrated the feasibility of offline ART for post-prostatectomy cancer. ART facilitates PTV volume reduction while maintaining reasonable CTV coverage and can reduce the dose to adjacent normal tissues. [ABSTRACT FROM AUTHOR]

Purpose: We evaluate the performance of a deformable image registration (DIR) software package in registering abdominal magnetic resonance images (MRIs) and then develop a mechanical modeling method to mitigate detected DIR uncertainties. Materials and Methods: Three evaluation metrics, namely mean displacement to agreement (MDA), DICE similarity coefficient (DSC), and standard deviation of Jacobian determinants (STD‐JD), are used to assess the multi‐modality (MM), contour‐consistency (CC), and image‐intensity (II)‐based DIR algorithms in the MIM software package, as well as an in‐house developed, contour matching‐based finite element method (CM‐FEM). Furthermore, we develop a hybrid FEM registration technique to modify the displacement vector field of each MIM registration. The MIM and FEM registrations were evaluated on MRIs obtained from 10 abdominal cancer patients. One‐tailed Wilcoxon‐Mann‐Whitney (WMW) tests were conducted to compare the MIM registrations with their FEM modifications. Results: For the registrations performed with the MIM‐CC, MIM‐MM, MIM‐II, and CM‐FEM algorithms, their average MDAs are 0.62 ± 0.27, 2.39 ± 1.30, 3.07 ± 2.42, 1.04 ± 0.72 mm, and average DSCs are 0.94 ± 0.03, 0.80 ± 0.12, 0.77 ± 0.15, 0.90 ± 0.11, respectively. The p‐values of the WMW tests between the MIM registrations and their FEM modifications are less than 0.0084 for STD‐JDs and greater than 0.87 for MDA and DSC. Conclusions: Among the three MIM DIR algorithms, MIM‐CC shows the smallest errors in terms of MDA and DSC but exhibits significant Jacobian uncertainties in the interior regions of abdominal organs. The hybrid FEM technique effectively mitigates the Jacobian uncertainties in these regions. [ABSTRACT FROM AUTHOR]

Simple Summary: This study assessed hypo-fractionated radiotherapy's feasibility and acute toxicity using the first Spanish 0.35T MR-LINAC in 37 patients. Prostate tumors (59.46%) were the most treated, followed by pancreatic tumors (32.44%). Treatment adaptation was successful, with manageable acute toxicity profiles. For prostate cancer, hypo-fractionated radiotherapy yielded promising outcomes with minimal toxicity, predominantly grade I and II cystitis. Pancreatic cancer patients received ablative dose radiotherapy with acceptable toxicity. Quality assurance measures demonstrated precise dose delivery. Overall, our study highlights the safety and feasibility of hypo-fractionated radiotherapy on a 0.35T MR-LINAC, particularly for challenging anatomical sites like prostate and pancreatic tumors, supporting its potential as an effective cancer treatment strategy. This observational, descriptive, longitudinal, and prospective basket-type study (Registry #5289) prospectively evaluated the feasibility and acute toxicity of hypo-fractionated radiotherapy on the first 0.35T MR-LINAC in Spain. A total of 37 patients were included between August and December 2023, primarily with prostate tumors (59.46%), followed by pancreatic tumors (32.44%). Treatment regimens typically involved extreme hypo-fractionated radiotherapy, with precise dose delivery verified through quality assurance measures. Acute toxicity assessment at treatment completion revealed manageable cystitis, with one case persisting at the three-month follow-up. Gastrointestinal toxicity was minimal. For pancreatic tumors, daily adaptation of organ-at-risk (OAR) and gross tumor volume (GTV) was practiced, with median doses to OAR within acceptable limits. Three patients experienced gastrointestinal toxicity, mainly nausea. Overall, the study demonstrates the feasibility and safety of extreme hypo-fractionated radiotherapy on a 0.35T MR-LINAC, especially for challenging anatomical sites like prostate and pancreatic tumors. These findings support the feasibility of MR-LINAC-based radiotherapy in delivering precise treatments with minimal toxicity, highlighting its potential for optimizing cancer treatment strategies. [ABSTRACT FROM AUTHOR]

MR-guided radiotherapy (MRgRT) online plan adaptation accounts for tumor volume changes, interfraction motion and thus allows daily sparing of relevant organs at risk. Due to the high interfraction variability of bladder and rectum, patients with tumors in the pelvic region may strongly benefit from adaptive MRgRT. Currently, fast automatic annotation of anatomical structures is not available within the online MRgRT workflow. Therefore, the aim of this study was to train and validate a fast, accurate deep learning model for automatic MRI segmentation at the MR-Linac for future implementation in a clinical MRgRT workflow. For a total of 47 patients, T2w MRI data were acquired on a 1.5 T MR-Linac (Unity, Elekta) on five different days. Prostate, seminal vesicles, rectum, anal canal, bladder, penile bulb, body and bony structures were manually annotated. These training data consisting of 232 data sets in total was used for the generation of a deep learning based autocontouring model and validated on 20 unseen T2w-MRIs. For quantitative evaluation the validation set was contoured by a radiation oncologist as gold standard contours (GSC) and compared in MATLAB to the automatic contours (AIC). For the evaluation, dice similarity coefficients (DSC), and 95% Hausdorff distances (95% HD), added path length (APL) and surface DSC (sDSC) were calculated in a caudal-cranial window of ± 4 cm with respect to the prostate ends. For qualitative evaluation, five radiation oncologists scored the AIC on the possible usage within an online adaptive workflow as follows: (1) no modifications needed, (2) minor adjustments needed, (3) major adjustments/ multiple minor adjustments needed, (4) not usable. The quantitative evaluation revealed a maximum median 95% HD of 6.9 mm for the rectum and minimum median 95% HD of 2.7 mm for the bladder. Maximal and minimal median DSC were detected for bladder with 0.97 and for penile bulb with 0.73, respectively. Using a tolerance level of 3 mm, the highest and lowest sDSC were determined for rectum (0.94) and anal canal (0.68), respectively. Qualitative evaluation resulted in a mean score of 1.2 for AICs over all organs and patients across all expert ratings. For the different autocontoured structures, the highest mean score of 1.0 was observed for anal canal, sacrum, femur left and right, and pelvis left, whereas for prostate the lowest mean score of 2.0 was detected. In total, 80% of the contours were rated be clinically acceptable, 16% to require minor and 4% major adjustments for online adaptive MRgRT. In this study, an AI-based autocontouring was successfully trained for online adaptive MR-guided radiotherapy on the 1.5 T MR-Linac system. The developed model can automatically generate contours accepted by physicians (80%) or only with the need of minor corrections (16%) for the irradiation of primary prostate on the clinically employed sequences. [ABSTRACT FROM AUTHOR]

During the radiation treatment planning process, one of the time-consuming procedures is the final high-resolution dose calculation, which obstacles the wide application of the emerging online adaptive radiotherapy techniques (OLART). There is an urgent desire for highly accurate and efficient dose calculation methods. This study aims to develop a dose super resolution-based deep learning model for fast and accurate dose prediction in clinical practice. A Multi-stage Dose Super-Resolution Network (MDSR Net) architecture with sparse masks module and multi-stage progressive dose distribution restoration method were developed to predict high-resolution dose distribution using low-resolution data. A total of 340 VMAT plans from different disease sites were used, among which 240 randomly selected nasopharyngeal, lung, and cervix cases were used for model training, and the remaining 60 cases from the same sites for model benchmark testing, and additional 40 cases from the unseen site (breast and rectum) was used for model generalizability evaluation. The clinical calculated dose with a grid size of 2 mm was used as baseline dose distribution. The input included the dose distribution with 4 mm grid size and CT images. The model performance was compared with HD U-Net and cubic interpolation methods using Dose-volume histograms (DVH) metrics and global gamma analysis with 1%/1 mm and 10% low dose threshold. The correlation between the prediction error and the dose, dose gradient, and CT values was also evaluated. The prediction errors of MDSR were 0.06–0.84% of D mean indices, and the gamma passing rate was 83.1–91.0% on the benchmark testing dataset, and 0.02–1.03% and 71.3–90.3% for the generalization dataset respectively. The model performance was significantly higher than the HD U-Net and interpolation methods (p < 0.05). The mean errors of the MDSR model decreased (monotonously by 0.03–0.004%) with dose and increased (by 0.01–0.73%) with the dose gradient. There was no correlation between prediction errors and the CT values. The proposed MDSR model achieved good agreement with the baseline high-resolution dose distribution, with small prediction errors for DVH indices and high gamma passing rate for both seen and unseen sites, indicating a robust and generalizable dose prediction model. The model can provide fast and accurate high-resolution dose distribution for clinical dose calculation, particularly for the routine practice of OLART. [ABSTRACT FROM AUTHOR]

Aim: This study aimed to evaluate the Deformable Image Registration (DIR) software of Mirada® and Velocity® using quantitative measures as per the recommendation based on Task Group TG- 132. Materials and Methods: Task Group 132 has provided geometric and anatomic virtual phantoms for quality assurance which were imported and contoured on the reference image, followed by DIR and then the contours were propagated on all the CT/CBCT data sets with different offsets and orientation of geometric and anatomic virtual phantom using Mirada and Velocity. Analysis was performed using the Dice Similarity Coefficient (DSC), Mean Distance to Agreement (MDA), Housdorff (HD) and Target Registration Error (TRE). The results were compared to find the difference in significance using a Paired t-test with a significance level set at 5%. Results: The obtained statistical result is listed below, considering the Geometric phantom, with regards to DSC, Mean and SD values obtained for Mirada and Velocity is 0.955 ± 0.348 and 0.965 ± 0.418 respectively with the p-value coming at 0.013 indicating a significant difference. Considering MDA, a significant difference of p-value of 0.001 was found between Mirada and Velocity with the former having 0.668 ± 0.684 compared to 0.381 ± 0.424 of the latter. For HD, Mirada obtained a mean value of 3.464 ± 2.091 in comparison with 2.202 ± 1.215 of Velocity. As for TRE, three fiducials within the phantom were considered and Mirada had a mean value of 1.037 mm while Velocity had 1.338 mm. While analyzing the Anatomic phantom, DSC values for Mirada and Velocity were 0.946 ± 0.031 and 0.944 ± 0.313 respectively indicating no significance. Likewise, no significance was found considering MDA (Mirada: 0.435 ± 0.235, Velocity: 0.449 ± 0.242). Regarding HD, Mirada obtained 4.216 mm while Velocity had 4.233 mm showing exceptional compliance. The mean value of TRE for three fiducials was found to be less than 1 mm it's also found significant difference. Conclusion: This study highlights the strengths and differences between Mirada and Velocity in DIR for adaptive radiotherapy. While both software applications excel in DSC, MDA and HD, the observed difference in TRE indicates a potential advantage of Mirada in achieving better image registration, especially in landmark-based registration scenarios. [ABSTRACT FROM AUTHOR]

Objective: Adaptive planning to accommodate anatomic changes during treatment often requires repeated segmentation. In this study, prior patient‐specific data was integrateda into a registration‐guided multi‐channel multi‐path (Rg‐MCMP) segmentation framework to improve the accuracy of repeated clinical target volume (CTV) segmentation. Methods: This study was based on CT image datasets for a total of 90 cervical cancer patients who received two courses of radiotherapy. A total of 15 patients were selected randomly as the test set. In the Rg‐MCMP segmentation framework, the first‐course CT images (CT1) were registered to second‐course CT images (CT2) to yield aligned CT images (aCT1), and the CTV in the first course (CTV1) was propagated to yield aligned CTV contours (aCTV1). Then, aCT1, aCTV1, and CT2 were combined as the inputs for 3D U‐Net consisting of a channel‐based multi‐path feature extraction network. The performance of the Rg‐MCMP segmentation framework was evaluated and compared with the single‐channel single‐path model (SCSP), the standalone registration methods, and the registration‐guided multi‐channel single‐path (Rg‐MCSP) model. The Dice similarity coefficient (DSC), 95% Hausdorff distance (HD95), and average surface distance (ASD) were used as the metrics. Results: The average DSC of CTV for the deformable image DIR‐MCMP model was found to be 0.892, greater than that of the standalone DIR (0.856), SCSP (0.837), and DIR‐MCSP (0.877), which were improvements of 4.2%, 6.6%, and 1.7%, respectively. Similarly, the rigid body DIR‐MCMP model yielded an average DSC of 0.875, which exceeded standalone RB (0.787), SCSP (0.837), and registration‐guided multi‐channel single‐path (0.848), which were improvements of 11.2%, 4.5%, and 3.2%, respectively. These improvements in DSC were statistically significant (p < 0.05). Conclusion: The proposed Rg‐MCMP framework achieved excellent accuracy in CTV segmentation as part of the adaptive radiotherapy workflow. [ABSTRACT FROM AUTHOR]

Background: Patients with head and neck cancer (HNC) may experience substantial anatomical changes during the course of radiotherapy treatment. The implementation of adaptive radiotherapy (ART) proves effective in managing the consequent impact on the planned dose distribution. Methods: This narrative literature review comprehensively discusses the diverse strategies of ART in HNC and the documented dosimetric and clinical advantages associated with these approaches, while also addressing the current challenges for integration of ART into clinical practice. Results and Conclusion: Although based on mainly non‐randomized and retrospective trials, there is accumulating evidence that ART has the potential to reduce toxicity and improve quality of life and tumor control in HNC patients treated with RT. However, several questions remain regarding accurate patient selection, the ideal frequency and timing of replanning, and the appropriate way for image registration and dose calculation. Well‐designed randomized prospective trials, with a predetermined protocol for both image registration and dose summation, are urgently needed to further investigate the dosimetric and clinical benefits of ART. [ABSTRACT FROM AUTHOR]

Purpose: The aim of this study was to assess the use of magnetic resonance guided adaptive radiotherapy (MRgART) in the post‐prostatectomy setting; comparing dose accumulation for our initial seven patients treated with fully adaptive workflow on the Unity MR‐Linac (MRL) and with non‐adaptive plans generated offline. Additionally, we analyzed toxicity in patients receiving treatment. Methods: Seven patients were treated with MRgART. The prescription was 70–72 Gy in 35–36 fractions. Patients were treated with an adapt to shape (ATS) technique. For each clinically delivered plan, a non‐adaptive plan based upon the reference plan was generated and compared to the associated clinically delivered plan. A total of 468 plans were analyzed. Concordance Index of target and Organs at Risk (OARs) for each fraction with reference contours was analyzed. Acute toxicity was then assessed at six‐months following completion of treatment with Common Terminology for Adverse Events (CTCAE) Toxicity Criteria. Results: A total of 246 fractions were clinically delivered to seven patients; 234 fractions were delivered via MRgART and 12 fractions delivered via a traditional linear accelerator due to machine issues. Pre‐treatment reference plans met CTV and OAR criteria. PTV coverage satisfaction was higher in the clinically delivered adaptive plans than non‐adaptive comparison plans; 42.93% versus 7.27% respectively. Six‐month CTCAE genitourinary and gastrointestinal toxicity was absent in most patients, and mild‐to‐moderate in a minority of patients (Grade 1 GU toxicity in one patient and Grade 2 GI toxicity in one patient). Conclusions: Daily MRgART treatment consistently met planning criteria. Target volume variability in prostate bed treatment can be mitigated by using MRgART and deliver satisfactory coverage of CTV whilst minimizing dose to adjacent OARs and reducing toxicity [ABSTRACT FROM AUTHOR]

Purpose: High‐quality CBCT and AI‐enhanced adaptive planning techniques allow CBCT‐guided stereotactic adaptive radiotherapy (CT‐STAR) to account for inter‐fractional anatomic changes. Studies of intra‐fractional respiratory motion management with a surface imaging solution for CT‐STAR have not been fully conducted. We investigated intra‐fractional motion management in breath‐hold Ethos‐based CT‐STAR and CT‐SBRT (stereotactic body non‐adaptive radiotherapy) using optical surface imaging combined with onboard CBCTs. Methods: Ten cancer patients with mobile lower lung or upper abdominal malignancies participated in an IRB‐approved clinical trial (Phase I) of optical surface image‐guided Ethos CT‐STAR/SBRT. In the clinical trial, a pre‐configured gating window (± 2 mm in AP direction) on optical surface imaging was used for manually triggering intra‐fractional CBCT acquisition and treatment beam irradiation during breath‐hold (seven patients for the end of exhalation and three patients for the end of inhalation). Two inter‐fractional CBCTs at the ends of exhalation and inhalation in each fraction were acquired to verify the primary direction and range of the tumor/imaging‐surrogate (donut‐shaped fiducial) motion. Intra‐fractional CBCTs were used to quantify the residual motion of the tumor/imaging‐surrogate within the pre‐configured breath‐hold window in the AP direction. Fifty fractions of Ethos RT were delivered under surface image‐guidance: Thirty‐two fractions with CT‐STAR (adaptive RT) and 18 fractions with CT‐SBRT (non‐adaptive RT). The residual motion of the tumor was quantified by determining variations in the tumor centroid position. The dosimetric impact on target coverage was calculated based on the residual motion. Results: We used 46 fractions for the analysis of intra‐fractional residual motion and 43 fractions for the inter‐fractional motion analysis due to study constraints. Using the image registration method, 43 pairs of inter‐fractional CBCTs and 100 intra‐fractional CBCTs attached to dose maps were analyzed. In the motion range study (image registration) from the inter‐fractional CBCTs, the primary motion (mean ± std) was 16.6 ± 9.2 mm in the SI direction (magnitude: 26.4 ± 11.3 mm) for the tumors and 15.5 ± 7.3 mm in the AP direction (magnitude: 20.4 ± 7.0 mm) for the imaging‐surrogate, respectively. The residual motion of the tumor (image registration) from intra‐fractional breath‐hold CBCTs was 2.2 ± 2.0 mm for SI, 1.4 ± 1.4 mm for RL, and 1.3 ± 1.3 mm for AP directions (magnitude: 3.5 ± 2.1 mm). The ratio of the actual dose coverage to 99%, 90%, and 50% of the target volume decreased by 0.95 ± 0.11, 0.96 ± 0.10, 0.99 ± 0.05, respectively. The mean percentage of the target volume covered by the prescribed dose decreased by 2.8 ± 4.4%. Conclusion: We demonstrated the intra‐fractional motion‐managed treatment strategy in breath‐hold Ethos CT‐STAR/SBRT using optical surface imaging and CBCT. While the controlled residual tumor motion measured at 3.5 mm exceeded the predetermined setup value of 2 mm, it is important to note that this motion still fell within the clinically acceptable range defined by the PTV margin of 5 mm. Nonetheless, additional caution is needed with intra‐fractional motion management in breath‐hold Ethos CT‐STAR/SBRT using optical surface imaging and CBCT. [ABSTRACT FROM AUTHOR]

Background and purpose: The adoption of hypo-fractionated stereotactic body radiotherapy (SBRT) for treating prostate cancer has led to an increase in specialised techniques for monitoring prostate motion. The aim of this study was to comprehensively review a radiation therapist (RTT) led treatment process in which two such systems were utilised, and present initial findings on their use within a SBRT prostate clinical trial. Materials and Methods: 18 patients were investigated, nine were fitted with the Micropos RayPilotTM (RP) system (Micropos Medical, Gothenburg, SE) and nine were fitted with the Micropos Raypilot Hypocath TM (HC) system. 36.25 Gray (Gy) was delivered in 5 fractions over 7 days with daily pre- and post-treatment cone beam computed tomography (CBCT) images acquired. Acute toxicity was reported on completion of treatment at six- and 12-weeks post-treatment, using the Radiation Therapy Oncology Group (RTOG) grading system and vertical (Vrt), longitudinal (Lng) and lateral (Lat) transmitter displacements recorded. Results: A significant difference was found in the Lat displacement between devices (P=0.003). A more consistent bladder volume was reported in the HC group (68.03 cc to 483.7 cc RP, 196.11 cc to 313.85 cc HC). No significant difference was observed in mean dose to the bladder, rectum and bladder dose maximum between the groups. Comparison of the rectal dose maximum between the groups reported a significant result (P=0.09). Comparing displacements with toxicity endpoints identified two significant correlations: Grade 2 Genitourinary (GU) at 6 weeks, P=0.029; and no toxicity, Gastrointestinal (GI) at 12 weeks P=0.013. Conclusion: Both the directly implanted RP device and the urinary catheter-based HC device are capable of real time motion monitoring. Here, the HC system was advantageous in the SBRT prostate workflow.

Background and purpose: Despite the superior dose conformity of proton therapy, the dose distribution is sensitive to daily anatomical changes, which can affect treatment accuracy. This study evaluated the dose recalculation accuracy of two synthetic computed tomography (sCT) generation algorithms in a commercial treatment planning system. Materials and methods: The evaluation was conducted for head-and-neck, thorax-and-abdomen, and pelvis sites treated with proton therapy. Thirty patients with two cone-beam computed tomography (CBCT) scans each were selected. The sCT images were generated from CBCT scans using two algorithms, Corrected CBCT (corrCBCT) and Virtual CT (vCT). Dose recalculations were performed based on these images for comparison with “ground truth” deformed CTs. Results: The choice of algorithm influenced dose recalculation accuracy, particularly in high dose regions. For head-and-neck cases, the corrCBCT method showed closer agreement with the “ground truth”, while for thorax-and-abdomen and pelvis cases, the vCT algorithm yielded better results (mean percentage dose discrepancy of 0.6 %, 1.3 % and 0.5 % for the three sites, respectively, in the high dose region). Head-and-neck and pelvis cases exhibited excellent agreement in high dose regions (2 %/2 mm gamma passing rate >98 %), while thorax-and-abdomen cases exhibited the largest differences, suggesting caution in sCT algorithm usage for this site. Significant systematic differences were observed in the clinical target volume and organ-at-risk doses in head-and-neck and pelvis cases, highlighting the importance of using the correct algorithm. Conclusions: This study provided treatment site-specific recommendations for sCT algorithm selection in proton therapy. The findings offered insights for proton beam centers implementing adaptive radiotherapy workflows.

Background and Purpose: Improved hounsfield-unit accuracy of on-board imaging may lead to direct-to-unit treatment approaches We aimed to demonstrate the feasibility of using only a diagnostic (dx) computed tomography (CT)-defined target pre-plan in an in silico study of simulation-free abdominal stereotactic adaptive radiotherapy (ART). Materials and Methods: Eight patients with abdominal treatment sites (five pancreatic cancer, three oligometastases) were treated using an integrated adaptive O-Ring gantry system. Each patient’s target was delineated on a dxCT. The target only pre-plan served primarily to seed the ART process. During the ART session, all structures were delineated. All simulated cases were treated to 50 Gy in 5 fractions to a planning target optimization structure (PTV_OPT) to allow for dose escalation within the planning target volume. Timing of steps during this workflow was recorded. Plan quality was compared between ART treatment plans and a plan created on a CT simulation scan using the traditional planning workflow. Results: The workflow was feasible in all attempts, with organ-at-risk (OAR) constraints met in all fractions despite lack of initial OAR contours. Median absolute difference between the adapted plan and simulation CT plan for the PTV_Opt V95% was 2.0 %. Median absolute difference in the D0.5 cm3 between the adapted plan and simulation CT plan was −0.9 Gy for stomach, 1.2 Gy for duodenum, −5.3 Gy for small bowel, and 0.3 Gy for large bowel. Median end-to-end workflow time was 63 min. Conclusion: The workflow was feasible for a dxCT-defined target-only pre-plan approach to stereotactic abdominal ART.

Purpose: Aim of the present study is to characterize a deep learning-based auto-segmentation software (DL) for prostate cone beam computed tomography (CBCT) images and to evaluate its applicability in clinical adaptive radiation therapy routine. Materials and methods: Ten patients, who received exclusive radiation therapy with definitive intent on the prostate gland and seminal vesicles, were selected. Femoral heads, bladder, rectum, prostate, and seminal vesicles were retrospectively contoured by four different expert radiation oncologists on patients CBCT, acquired during treatment. Consensus contours (CC) were generated starting from these data and compared with those created by DL with different algorithms, trained on CBCT (DL-CBCT) or computed tomography (DL-CT). Dice similarity coefficient (DSC), centre of mass (COM) shift and volume relative variation (VRV) were chosen as comparison metrics. Since no tolerance limit can be defined, results were also compared with the inter-operator variability (IOV), using the same metrics. Results: The best agreement between DL and CC was observed for femoral heads (DSC of 0.96 for both DL-CBCT and DL-CT). Performance worsened for low-contrast soft tissue organs: the worst results were found for seminal vesicles (DSC of 0.70 and 0.59 for DL-CBCT and DL-CT, respectively). The analysis shows that it is appropriate to use algorithms trained on the specific imaging modality. Furthermore, the statistical analysis showed that, for almost all considered structures, there is no significant difference between DL-CBCT and human operator in terms of IOV. Conclusions: The accuracy of DL-CBCT is in accordance with CC; its use in clinical practice is justified by the comparison with the inter-operator variability.

Introduction: Organ motion (OM) and volumetric changes pose challenges in radiotherapy (RT) for locally advanced cervical cancer (LACC). Magnetic Resonance-guided Radiotherapy (MRgRT) combines improved MRI contrast with adaptive RT plans for daily anatomical changes. Our goal was to analyze cervico-uterine structure (CUS) changes during RT to develop strategies for managing OM. Materials and methods: LACC patients received chemoradiation by MRIdian system with a simultaneous integrated boost (SIB) protocol. Prescription doses of 55–50.6 Gy at PTV1 and 45–39.6 Gy at PTV2 were given in 22 and 25 fractions. Daily MRI scans were co-registered with planning scans and CUS changes were assessed.Six PTVs were created by adding 0.5, 0.7, 1, 1.3, 1.5, and 2 cm margins to the CUS, based on the simulation MRI. Adequate margins were determined to include 95 % of the CUSs throughout the entire treatment in 95 % of patients. Results: Analysis of 15 LACC patients and 372 MR scans showed a 31 % median CUS volume decrease. Asymmetric margins of 2 cm cranially, 0.5 cm caudally, 1.5 cm posteriorly, 2 cm anteriorly, and 1.5 cm on both sides were optimal for PTV, adapting to CUS variations. Post-14th fraction, smaller margins of 0.7 cm cranially, 0.5 cm caudally, 1.3 cm posteriorly, 1.3 cm anteriorly, and 1.3 cm on both sides sufficed. Conclusion: CUS mobility varies during RT, suggesting reduced PTV margins after the third week. MRgRT with adaptive strategies optimizes dose delivery, emphasizing the importance of streamlined IGRT with reduced PTV margins using a tailored MRgRT workflow with hybrid MRI-guided systems.