AAPM Task Group Report 267: A joint AAPM GEC-ESTRO report on biophysical models and tools for the planning and evaluation of brachytherapy.

Brachytherapy utilizes a multitude of radioactive sources and treatment techniques that often exhibit widely different spatial and temporal dose delivery patterns. Biophysical models, capable of modeling the key interacting effects of dose delivery patterns with the underlying cellular processes of the irradiated tissues, can be a potentially useful tool for elucidating the radiobiological effects of complex brachytherapy dose delivery patterns and for comparing their relative clinical effectiveness. While the biophysical models have been used largely in research settings by experts, it has also been used increasingly by clinical medical physicists over the last two decades. A good understanding of the potentials and limitations of the biophysical models and their intended use is critically important in the widespread use of these models. To facilitate meaningful and consistent use of biophysical models in brachytherapy, Task Group 267 (TG-267) was formed jointly with the American Association of Physics in Medicine (AAPM) and The Groupe Européen de Curiethérapie and the European Society for Radiotherapy & Oncology (GEC-ESTRO) to review the existing biophysical models, model parameters, and their use in selected brachytherapy modalities and to develop practice guidelines for clinical medical physicists regarding the selection, use, and interpretation of biophysical models. The report provides an overview of the clinical background and the rationale for the development of biophysical models in radiation oncology and, particularly, in brachytherapy; a summary of the results of literature review of the existing biophysical models that have been used in brachytherapy; a focused discussion of the applications of relevant biophysical models for five selected brachytherapy modalities; and the task group recommendations on the use, reporting, and implementation of biophysical models for brachytherapy treatment planning and evaluation. The report concludes with discussions on the challenges and opportunities in using biophysical models for brachytherapy and with an outlook for future developments.

Changes in Daily Apparent Diffusion Coefficient on Fully Quantitative Magnetic Resonance Imaging Correlate With Established Genomic Pathways of Radiation Sensitivity and Reveal Novel Biologic Associations.

PURPOSE: Changes in quantitative magnetic resonance imaging (qMRI) are frequently observed during chemotherapy or radiation therapy (RT). It is hypothesized that qMRI features are reflective of underlying tissue responses. It's unknown what underlying genomic characteristics underly qMRI changes. We hypothesized that qMRI changes may correlate with DNA damage response (DDR) capacity within human tumors. Therefore, we designed the current study to correlate qMRI changes from daily RT treatment with underlying tumor transcriptomic profiles. METHODS AND MATERIALS: Study participants were prospectively enrolled (National Clinical Trial 03500081). RNA expression levels for 757 genes from pretreatment biopsies were obtained using a custom panel that included signatures of radiation sensitivity and DDR. Daily qMRI data were obtained from a 1.5 Tesla MR linear accelerator. Using these images, d-slow, d-star, perfusion, and apparent diffusion coefficient-mean values in tumors were plotted per-fraction, over time, and associated with genomic pathways. RESULTS: A total of 1022 qMRIs were obtained from 39 patients and both genomic data and qMRI data from 27 total patients. For 20 of those patients, we also generated normal tissue transcriptomic data. Radio sensitivity index values most closely associated with tissue of origin. Multiple genomic pathways including DNA repair, peroxisome, late estrogen receptor responses, KRAS signaling, and UV response were significantly associated with qMRI feature changes (P < .001). CONCLUSIONS: Genomic pathway associations across metabolic, RT sensitivity, and DDR pathways indicate common tumor biology that may correlate with qMRI changes during a course of treatment. Such data provide hypothesis-generating novel mechanistic insight into the biologic meaning of qMRI changes during treatment and enable optimal selection of imaging biomarkers for biologically MR-guided RT.

Development of a comprehensive cardiac atlas on a 1.5 Tesla Magnetic Resonance Linear Accelerator.

Background and purpose: The 1.5 Tesla (T) Magnetic Resonance Linear Accelerator (MRL) provides an innovative modality for improved cardiac imaging when planning radiation treatment. No MRL based cardiac atlases currently exist, thus, we sought to comprehensively characterize cardiac substructures, including the conduction system, from cardiac images acquired using a 1.5 T MRL and provide contouring guidelines. Materials and methods: Five volunteers were enrolled in a prospective protocol (NCT03500081) and were imaged on the 1.5 T MRL with Half Fourier Single-Shot Turbo Spin-Echo (HASTE) and 3D Balanced Steady-State Free Precession (bSSFP) sequences in axial, short axis, and vertical long axis. Cardiac anatomy was contoured by (AS) and confirmed by a board certified cardiologist (JR) with expertise in cardiac MR imaging. Results: A total of five volunteers had images acquired with the HASTE sequence, with 21 contours created on each image. One of these volunteers had additional images obtained with 3D bSSFP sequences in the axial plane and additional images obtained with HASTE sequences in the key cardiac planes. Contouring guidelines were created and outlined. 15-16 contours were made for the short axis and vertical long axis. The cardiac conduction system was demonstrated with eleven representative contours. There was reasonable variation of contour volume across volunteers, with structures more clearly delineated on the 3D bSSFP sequence. Conclusions: We present a comprehensive cardiac atlas using novel images acquired prospectively on a 1.5 T MRL. This cardiac atlas provides a novel resource for radiation oncologists in delineating cardiac structures for treatment with radiotherapy, with special focus on the cardiac conduction system.

Testicular Dysfunction in Male Childhood Cancer Survivors Treated With Radiation Therapy: A PENTEC Comprehensive Review.

PURPOSE: The male reproductive task force of the Pediatric Normal Tissue Effects in the Clinic (PENTEC) initiative performed a comprehensive review that included a meta-analysis of publications reporting radiation dose-volume effects for risk of impaired fertility and hormonal function after radiation therapy for pediatric malignancies. METHODS AND MATERIALS: The PENTEC task force conducted a comprehensive literature search to identify published data evaluating the effect of testicular radiation dose on reproductive complications in male childhood cancer survivors. Thirty-one studies were analyzed, of which 4 had testicular dose data to generate descriptive scatter plots. Two cohorts were identified. Cohort 1 consisted of pediatric and young adult patients with cancer who received scatter radiation therapy to the testes. Cohort 2 consisted of pediatric and young adult patients with cancer who received direct testicular radiation therapy as part of their cancer therapy. Descriptive scatter plots were used to delineate the relationship between the effect of mean testicular dose on sperm count reduction, testosterone, follicle stimulating hormone (FSH), and luteinizing hormone (LH) levels. RESULTS: Descriptive scatter plots demonstrated a 44% to 80% risk of oligospermia when the mean testicular dose was <1 Gy, but this was recovered by >12 months in 75% to 100% of patients. At doses >1 Gy, the rate of oligospermia increased to >90% at 12 months. Testosterone levels were generally not affected when the mean testicular dose was <0.2 Gy but were abnormal in up to 25% of patients receiving between 0.2 and 12 Gy. Doses between 12 and 19 Gy may be associated with abnormal testosterone in 40% of patients, whereas doses >20 Gy to the testes were associated with a steep increase in abnormal testosterone in at least 68% of patients. FSH levels were unaffected by a mean testicular dose <0.2 Gy, whereas at doses >0.5 Gy, the risk was between 40% and 100%. LH levels were affected at doses >0.5 Gy in 33% to 75% of patients between 10 and 24 months after radiation. Although dose modeling could not be performed in cohort 2, the risk of reproductive toxicities was escalated with doses >10 Gy. CONCLUSIONS: This PENTEC comprehensive review demonstrates important relationships between scatter or direct radiation dose on male reproductive endpoints including semen analysis and levels of FSH, LH, and testosterone.

Obtaining organ-specific radiobiological parameters from clinical data for radiation therapy planning of head and neck cancers.

Objective.Different radiation therapy (RT) strategies, e.g. conventional fractionation RT (CFRT), hypofractionation RT (HFRT), stereotactic body RT (SBRT), adaptive RT, and re-irradiation are often used to treat head and neck (HN) cancers. Combining and/or comparing these strategies requires calculating biological effective dose (BED). The purpose of this study is to develop a practical process to estimate organ-specific radiobiologic model parameters that may be used for BED calculations in individualized RT planning for HN cancers.Approach.Clinical dose constraint data for CFRT, HFRT and SBRT for 5 organs at risk (OARs) namely spinal cord, brainstem, brachial plexus, optic pathway, and esophagus obtained from literature were analyzed. These clinical data correspond to a particular endpoint. The linear-quadratic (LQ) and linear-quadratic-linear (LQ-L) models were used to fit these clinical data and extract relevant model parameters (alpha/beta ratio, gamma/alpha,dTand BED) from the iso-effective curve. The dose constraints in terms of equivalent physical dose in 2 Gy-fraction (EQD2) were calculated using the obtained parameters.Main results.The LQ-L and LQ models fitted clinical data well from the CFRT to SBRT with the LQ-L representing a better fit for most of the OARs. The alpha/beta values for LQ-L (LQ) were found to be 2.72 (2.11) Gy, 0.55 (0.30) Gy, 2.82 (2.90) Gy, 6.57 (3.86) Gy, 5.38 (4.71) Gy, and the dose constraint EQD2 were 55.91 (54.90) Gy, 57.35 (56.79) Gy, 57.54 (56.35) Gy, 60.13 (59.72) Gy and 65.66 (64.50) Gy for spinal cord, optic pathway, brainstem, brachial plexus, and esophagus, respectively. Additional two LQ-L parametersdTwere 5.24 Gy, 5.09 Gy, 7.00 Gy, 5.23 Gy, and 6.16 Gy, and gamma/alpha were 7.91, 34.02, 8.67, 5.62 and 4.95.Significance.A practical process was developed to extract organ-specific radiobiological model parameters from clinical data. The obtained parameters can be used for biologically based radiation planning such as calculating dose constraints of different fractionation regimens.

Kidney Disease in Childhood Cancer Survivors Treated With Radiation Therapy: A Comprehensive PENTEC Genitourinary Review.

PURPOSE: Kidney injury is a known late and potentially devastating complication of abdominal radiation therapy (RT) in pediatric patients. A comprehensive Pediatric Normal Tissue Effects in the Clinic review by the Genitourinary (GU) Task Force aimed to describe RT dose-volume relationships for GU dysfunction, including kidney, bladder, and hypertension, for pediatric malignancies. The effect of chemotherapy was also considered. METHODS AND MATERIALS: We conducted a comprehensive PubMed search of peer-reviewed manuscripts published from 1990 to 2017 for investigations on RT-associated GU toxicities in children treated for cancer. We retrieved 3271 articles with 100 fulfilling criteria for full review, 24 with RT dose data and 13 adequate for modeling. Endpoints were heterogenous and grouped according to National Kidney Foundation: grade ≥1, grade ≥2, and grade ≥3. We modeled whole kidney exposure from total body irradiation (TBI) for hematopoietic stem cell transplant and whole abdominal irradiation (WAI) for patients with Wilms tumor. Partial kidney tolerance was modeled from a single publication from 2021 after the comprehensive review revealed no usable partial kidney data. Inadequate data existed for analysis of bladder RT-associated toxicities. RESULTS: The 13 reports with long-term GU outcomes suitable for modeling included 4 on WAI for Wilms tumor, 8 on TBI, and 1 for partial renal RT exposure. These reports evaluated a total of 1191 pediatric patients, including: WAI 86, TBI 666, and 439 partial kidney. The age range at the time of RT was 1 month to 18 years with medians of 2 to 11 years in the various reports. In our whole kidney analysis we were unable to include chemotherapy because of the heterogeneity of regimens and paucity of data. Age-specific toxicity data were also unavailable. Wilms studies occurred from 1968 to 2011 with mean follow-ups 8 to 15 years. TBI studies occurred from 1969 to 2004 with mean follow-ups of 4 months to 16 years. We modeled risk of dysfunction by RT dose and grade of toxicity. Normal tissue complication rates ≥5%, expressed as equivalent doses, 2 Gy/fx for whole kidney exposures occurred at 8.5, 10.2, and 14.5 Gy for National Kidney Foundation grades ≥1, ≥2, and ≥3, respectively. Conventional Wilms WAI of 10.5 Gy in 6 fx had risks of ≥grade 2 toxicity 4% and ≥grade 3 toxicity 1%. For fractionated 12 Gy TBI, those risks were 8% and <3%, respectively. Data did not support whole kidney modeling with chemotherapy. Partial kidney modeling from 439 survivors who received RT (median age, 7.3 years) demonstrated 5 or 10 Gy to 100% kidney gave a <5% risk of grades 3 to 5 toxicity with 1500 mg/m2 carboplatin or no chemo. With 480 mg/m2 cisplatin, a 3% risk of ≥grade 3 toxicity occurred without RT and a 5% risk when 26% kidney received ≥10 Gy. With 63 g/m2 of ifosfamide, a 5% risk of ≥grade 3 toxicity occurred with no RT, and a 10% toxicity risk occurred when 42% kidney received ≥10 Gy. CONCLUSIONS: In patients with Wilms tumor, the risk of toxicity from 10.5 Gy of WAI is low. For 12 Gy fractionated TBI with various mixtures of chemotherapy, the risk of severe toxicity is low, but low-grade toxicity is not uncommon. Partial kidney data are limited and toxicity is associated heavily with the use of nephrotoxic chemotherapeutic agents. Our efforts demonstrate the need for improved data gathering, systematic follow-up, and reporting in future clinical studies. Current radiation dose used for Wilms tumor and TBI appear to be safe; however, efforts in effective kidney-sparing TBI and WAI regimens may reduce the risks of renal injury without compromising cure.

Deep learning auto-segmentation on multi-sequence magnetic resonance images for upper abdominal organs.

Introduction: Multi-sequence multi-parameter MRIs are often used to define targets and/or organs at risk (OAR) in radiation therapy (RT) planning. Deep learning has so far focused on developing auto-segmentation models based on a single MRI sequence. The purpose of this work is to develop a multi-sequence deep learning based auto-segmentation (mS-DLAS) based on multi-sequence abdominal MRIs. Materials and methods: Using a previously developed 3DResUnet network, a mS-DLAS model using 4 T1 and T2 weighted MRI acquired during routine RT simulation for 71 cases with abdominal tumors was trained and tested. Strategies including data pre-processing, Z-normalization approach, and data augmentation were employed. Additional 2 sequence specific T1 weighted (T1-M) and T2 weighted (T2-M) models were trained to evaluate performance of sequence-specific DLAS. Performance of all models was quantitatively evaluated using 6 surface and volumetric accuracy metrics. Results: The developed DLAS models were able to generate reasonable contours of 12 upper abdomen organs within 21 seconds for each testing case. The 3D average values of dice similarity coefficient (DSC), mean distance to agreement (MDA mm), 95 percentile Hausdorff distance (HD95% mm), percent volume difference (PVD), surface DSC (sDSC), and relative added path length (rAPL mm/cc) over all organs were 0.87, 1.79, 7.43, -8.95, 0.82, and 12.25, respectively, for mS-DLAS model. Collectively, 71% of the auto-segmented contours by the three models had relatively high quality. Additionally, the obtained mS-DLAS successfully segmented 9 out of 16 MRI sequences that were not used in the model training. Conclusion: We have developed an MRI-based mS-DLAS model for auto-segmenting of upper abdominal organs on MRI. Multi-sequence segmentation is desirable in routine clinical practice of RT for accurate organ and target delineation, particularly for abdominal tumors. Our work will act as a stepping stone for acquiring fast and accurate segmentation on multi-contrast MRI and make way for MR only guided radiation therapy.

Automated deep learning auto-segmentation of air volumes for MRI-guided online adaptive radiation therapy of abdominal tumors.

Objective. In the current MR-Linac online adaptive workflow, air regions on the MR images need to be manually delineated for abdominal targets, and then overridden by air density for dose calculation. Auto-delineation of these regions is desirable for speed purposes, but poses a challenge, since unlike computed tomography, they do not occupy all dark regions on the image. The purpose of this study is to develop an automated method to segment the air regions on MRI-guided adaptive radiation therapy (MRgART) of abdominal tumors.Approach. A modified ResUNet3D deep learning (DL)-based auto air delineation model was trained using 102 patients' MR images. The MR images were acquired by a dedicated in-house sequence named 'Air-Scan', which is designed to generate air regions that are especially dark and accentuated. The air volumes generated by the newly developed DL model were compared with the manual air contours using geometric similarity (Dice Similarity Coefficient (DSC)), and dosimetric equivalence using Gamma index and dose-volume parameters.Main results. The average DSC agreement between the DL generated and manual air contours is 99% ± 1%. The gamma index between the dose calculations with overriding the DL versus manual air volumes with density of 0.01 is 97% ± 2% for a local gamma calculation with a tolerance of 2% and 2 mm. The dosimetric parameters from planning target volume-PTV and organs at risk-OARs were all within 1% between when DL versus manual contours were overridden by air density. The model runs in less than five seconds on a PC with 28 Core processor and NVIDIA Quadro®P2000 GPU.Significance: a DL based automated segmentation method was developed to generate air volumes on specialized abdominal MR images and generate results that are practically equivalent to the manual contouring of air volumes.

Real-time motion monitoring using orthogonal cine MRI during MR-guided adaptive radiation therapy for abdominal tumors on 1.5T MR-Linac.

BACKGROUND: Real-time motion monitoring (RTMM) is necessary for accurate motion management of intrafraction motions during radiation therapy (RT). PURPOSE: Building upon a previous study, this work develops and tests an improved RTMM technique based on real-time orthogonal cine magnetic resonance imaging (MRI) acquired during magnetic resonance-guided adaptive RT (MRgART) for abdominal tumors on MR-Linac. METHODS: A motion monitoring research package (MMRP) was developed and tested for RTMM based on template rigid registration between beam-on real-time orthogonal cine MRI and pre-beam daily reference 3D-MRI (baseline). The MRI data acquired under free-breathing during the routine MRgART on a 1.5T MR-Linac for 18 patients with abdominal malignancies of 8 liver, 4 adrenal glands (renal fossa), and 6 pancreas cases were used to evaluate the MMRP package. For each patient, a 3D mid-position image derived from an in-house daily 4D-MRI was used to define a target mask or a surrogate sub-region encompassing the target. Additionally, an exploratory case reviewed for an MRI dataset of a healthy volunteer acquired under both free-breathing and deep inspiration breath-hold (DIBH) was used to test how effectively the RTMM using the MMRP can address through-plane motion (TPM). For all cases, the 2D T2/T1-weighted cine MRIs were captured with a temporal resolution of 200 ms interleaved between coronal and sagittal orientations. Manually delineated contours on the cine frames were used as the ground-truth motion. Common visible vessels and segments of target boundaries in proximity to the target were used as anatomical landmarks for reproducible delineations on both the 3D and the cine MRI images. Standard deviation of the error (SDE) between the ground-truth and the measured target motion from the MMRP package were analyzed to evaluate the RTMM accuracy. The maximum target motion (MTM) was measured on the 4D-MRI for all cases during free-breathing. RESULTS: The mean (range) centroid motions for the 13 abdominal tumor cases were 7.69 (4.71-11.15), 1.73 (0.81-3.05), and 2.71 (1.45-3.93) mm with an overall accuracy of <2 mm in the superior-inferior (SI), the left-right (LR), and the anterior-posterior (AP) directions, respectively. The mean (range) of the MTM from the 4D-MRI was 7.38 (2-11) mm in the SI direction, smaller than the monitored motion of centroid, demonstrating the importance of the real-time motion capture. For the remaining patient cases, the ground-truth delineation was challenging under free-breathing due to the target deformation and the large TPM in the AP direction, the implant-induced image artifacts, and/or the suboptimal image plane selection. These cases were evaluated based on visual assessment. For the healthy volunteer, the TPM of the target was significant under free-breathing which degraded the RTMM accuracy. RTMM accuracy of <2 mm was achieved under DIBH, indicating DIBH is an effective method to address large TPM. CONCLUSIONS: We have successfully developed and tested the use of a template-based registration method for an accurate RTMM of abdominal targets during MRgART on a 1.5T MR-Linac without using injected contrast agents or radio-opaque implants. DIBH may be used to effectively reduce or eliminate TPM of abdominal targets during RTMM.

Deep learning based automatic contour refinement for inaccurate auto-segmentation in MR-guided adaptive radiotherapy.

Objective.Fast and accurate auto-segmentation is essential for magnetic resonance-guided adaptive radiation therapy (MRgART). Deep learning auto-segmentation (DLAS) is not always clinically acceptable, particularly for complex abdominal organs. We previously reported an automatic contour refinement (ACR) solution of using an active contour model (ACM) to partially correct the DLAS contours. This study aims to develop a DL-based ACR model to work in conjunction with ACM-ACR to further improve the contour accuracy.Approach.The DL-ACR model was trained and tested using bowel contours created by an in-house DLAS system from 160 MR sets (76 from MR-simulation and 84 from MR-Linac). The contours were classified into acceptable, minor-error and major-error groups using two approaches of contour quality classification (CQC), based on the AAPM TG-132 recommendation and an in-house classification model, respectively. For the major-error group, DL-ACR was applied subsequently after ACM-ACR to further refine the contours. For the minor-error group, contours were directly corrected by DL-ACR without applying an initial ACM-ACR. The ACR workflow was performed separately for the two CQC methods and was evaluated using contours from 25 image sets as independent testing data.Main results.The best ACR performance was observed in the MR-simulation testing set using CQC by TG-132: (1) for the major-error group, 44% (177/401) were improved to minor-error group and 5% (22/401) became acceptable by applying ACM-ACR; among these 177 contours that shifted from major-error to minor-error with ACM-ACR, DL-ACR further refined 49% (87/177) to acceptable; and overall, 36% (145/401) were improved to minor-error contours, and 30% (119/401) became acceptable after sequentially applying ACM-ACR and DL-ACR; (2) for the minor-error group, 43% (320/750) were improved to acceptable contours using DL-ACR.Significance.The obtained ACR workflow substantially improves the accuracy of DLAS bowel contours, minimizing the manual editing time and accelerating the segmentation process of MRgART.

Deep learning-based prediction of deliverable adaptive plans for MR-guided adaptive radiotherapy: A feasibility study.

Purpose: Fast and automated plan generation is desirable in radiation therapy (RT), in particular, for MR-guided online adaptive RT (MRgOART) or real-time (intrafractional) adaptive RT (MRgRART), to reduce replanning time. The purpose of this study is to investigate the feasibility of using deep learning to quickly predict deliverable adaptive plans based on a target dose distribution for MRgOART/MRgRART. Methods: A conditional generative adversarial network (cGAN) was trained to predict the MLC leaf sequence corresponding to a target dose distribution based on reference plan created prior to MRgOART using a 1.5T MR-Linac. The training dataset included 50 ground truth dose distributions and corresponding beam parameters (aperture shapes and weights) created during MRgOART for 10 pancreatic cancer patients (each with five fractions). The model input was the dose distribution from each individual beam and the output was the predicted corresponding field segments with specific shape and weight. Patient-based leave-one-out-cross-validation was employed and for each model trained, four (44 training beams) out of five fractionated plans of the left-out patient were set aside for testing purposes. We deliberately kept a single fractionated plan in the training dataset so that the model could learn to replan the patient based on a prior plan. The model performance was evaluated by calculating the gamma passing rate of the ground truth dose vs. the dose from the predicted adaptive plan and calculating max and mean dose metrics. Results: The average gamma passing rate (95%, 3mm/3%) among 10 test cases was 88%. In general, we observed 95% of the prescription dose to PTV achieved with an average 7.6% increase of max and mean dose, respectively, to OARs for predicted replans. Complete adaptive plans were predicted in ≤20 s using a GTX 1660TI GPU. Conclusion: We have proposed and demonstrated a deep learning method to generate adaptive plans automatically and rapidly for MRgOART. With further developments using large datasets and the inclusion of patient contours, the method may be implemented to accelerate MRgOART process or even to facilitate MRgRART.

AAPM Task Group Report 306: Quality control and assurance for tomotherapy: An update to Task Group Report 148.

Since the publication of AAPM Task Group (TG) 148 on quality assurance (QA) for helical tomotherapy, there have been many new developments on the tomotherapy platform involving treatment delivery, on-board imaging options, motion management, and treatment planning systems (TPSs). In response to a need for guidance on quality control (QC) and QA for these technologies, the AAPM Therapy Physics Committee commissioned TG 306 to review these changes and make recommendations related to these technology updates. The specific objectives of this TG were (1) to update, as needed, recommendations on tolerance limits, frequencies and QC/QA testing methodology in TG 148, (2) address the commissioning and necessary QA checks, as a supplement to Medical Physics Practice Guidelines (MPPG) with respect to tomotherapy TPS and (3) to provide risk-based recommendations on the new technology implemented clinically and treatment delivery workflow. Detailed recommendations on QA tests and their tolerance levels are provided for dynamic jaws, binary multileaf collimators, and Synchrony motion management. A subset of TPS commissioning and QA checks in MPPG 5.a. applicable to tomotherapy are recommended. In addition, failure mode and effects analysis has been conducted among TG members to obtain multi-institutional analysis on tomotherapy-related failure modes and their effect ranking.

Development of a multi-layer quality assurance program to evaluate the uncertainty of deformable dose accumulation in adaptive radiotherapy.

PURPOSE: Deformable dose accumulation (DDA) has uncertainties which impede the implementation of DDA-based adaptive radiotherapy (ART) in clinic. The purpose of this study is to develop a multi-layer quality assurance (MLQA) program to evaluate uncertainties in DDA. METHODS: A computer program is developed to generate a pseudo-inverse displacement vector field (DVF) for each deformable image registration (DIR) performed in Accuray's PreciseART. The pseudo-inverse DVF is first used to calculate a pseudo-inverse consistency error (PICE) and then implemented in an energy and mass congruent mapping (EMCM) method to reconstruct a deformed dose. The PICE is taken as a metric to estimate DIR uncertainties. A pseudo-inverse dose agreement rate (PIDAR) is used to evaluate the consequence of the DIR uncertainties in DDA and the principle of energy conservation is used to validate the integrity of dose mappings. The developed MLQA program was tested using the data collected from five representative cancer patients treated with tomotherapy. RESULTS: DIRs were performed in PreciseART to generate primary DVFs for the five patients. The fidelity index and PICE of these DVFs on average are equal to 0.028 mm and 0.169 mm, respectively. With the criteria of 3 mm/3% and 5 mm/5%, the PIDARs of the PreciseART-reconstructed doses are 73.9 ± 4.4% and 87.2 ± 3.3%, respectively. The PreciseART and EMCM-based dose reconstructions have their deposited energy changed by 5.6 ± 3.9% and 2.6 ± 1.5% in five GTVs, and by 9.2 ± 7.8% and 4.7 ± 3.6% in 30 OARs, respectively. CONCLUSIONS: A pseudo-inverse map-based EMCM program has been developed to evaluate DIR and dose mapping uncertainties. This program could also be used as a sanity check tool for DDA-based ART.

Auto-detection of necessity for MRI-guided online adaptive replanning using a machine learning classifier.

PURPOSE: MRI-guided adaptive radiation therapy (MRgART), particularly daily online adaptive replanning (OLAR) can substantially improve radiation therapy delivery, however, it can be labor-intensive and time-consuming. Currently, the decision to perform OLAR for a treatment fraction is determined subjectively. In this work, we develop a machine learning algorithm based on structural similarity index measure (SSIM) and change in entropy to quickly and objectively determine whether OLAR is necessary for a daily MRI set. METHODS: A total of 109 daily MRI sets acquired on a 1.5T MR-Linac during MRgART for 22 pancreatic cancer patients each treated with five fractions were retrospectively analyzed. For each daily MRI set, OLAR and reposition (No-OLAR) plans were created and the superior plan with the daily fraction determined per clinical dose-volume criteria. SSIM and entropy maps were extracted from each daily MRI set, with respect to its reference (e.g., dry-run) MRI in the region enclosed by 50-100% isodose surfaces. A total of six common features were extracted from SSIM maps. Pearson's rank correlation coefficient was utilized to rule out redundant SSIM features. A t-test was used to determine significant SSIM features which were combined with the change in entropy to develop anensemble machine classifier with fivefold cross validation. The performance of the classifier was evaluated using the area under the curve (AUC) of the receiver operating characteristic curve. RESULTS: A machine learning classifier model using two SSIM features (mean and full width at half maximum) and change in entropy was determined to be able to significantly discriminate between No-OLAR and OLAR groups. The obtained machine learning ensemble classifier can predict OLAR necessity with a cross validated AUC of 0.93. Misclassification was found primarily for No-OLAR cases with dosimetric plan quality closely comparable to the corresponding OLAR plans, thus, are not a major practical concern. CONCLUSION: A machine learning classifier based on simple first-order image features, that is, SSIM features and change in entropy, was developed to determine when OLAR is necessary for a daily MRI set with practical acceptable prediction accuracy. This classifier may be implemented in the MRgART process to automatically and objectively determine if OLAR is required following daily MRI.

Incorporating uncertainty bounds in daily deformable dose accumulation for adaptive radiation therapy of head-and-neck cancer.

BACKGROUND: The widespread use of deformable dose accumulation (DDA) in adaptive radiotherapy (ART) has been limited due to the lack of clinically compatible methods to consider its related uncertainties. PURPOSE: We estimate dose reconstruction uncertainties in daily DDA during CT-guided radiotherapy of head-and-neck cancer (HNC). We project confidence intervals of cumulative dose-volume parameters to the parotids and determine threshold values to guide clinical decision-making in ART. METHODS: Doses from daily images (megavoltage CTs [MVCTs]) of 20 HNC patients treated with tomotherapy were reconstructed and accumulated in the planning CT (PCT) utilizing a commercial DDA algorithm (PreciseART, Accuray, Inc.). For each mapped fraction, we warped the planning contours to the MVCT. Dose-volume histograms (DVHs) calculated in the MVCT (with warped contour and native dose) and the PCT (with native contour and mapped dose) were compared; the observed inconsistencies were associated with dose reconstruction errors. We derived uncertainty bounds for the transferred dose to voxels within the structure of interest in the PCT. The confidence intervals of cumulative dose-volume parameters were mid-treatment projected and evaluated as predictors of the end of treatment cumulative metrics. The need for plan adaptation was tested by comparing the projected uncertainty bounds with the treatment constraint points. RESULTS: Among all cases, the uncertainty in mean values of daily dose distributions mapped to the reference parotid's contours averaged between 2.8% and 3.8% of typical single fraction planning values and less than 1% for the planning target volume (PTV) D95%. These daily inconsistencies were higher in the ipsilateral compared to the contralateral parotid and increased toward the end of treatment. The magnitude of the uncertainty bounds for the cumulative treatment mean dose, D50%, and V20 Gy to the parotids, and PTV D95% were on average 3.5%, 6.6%, 4.6%, and 0.4% of the planned or prescribed values, with confidence intervals of 97.1%-107.0%, 98.2%-110.4%, 95.6%-111.1%, and 98.2%-100.2% respectively. The uncertainty intervals projected at mid-treatment intersected with the end of treatment bounds in 82% of the parotid's metrics; half of them presented an overlapping percentage greater than 60%. In five patients, the cumulative mean doses were projected at mid-treatment to exceed the total treatment constraint point by at least 3%; this threshold was exceeded at the end of treatment in the five cases. Underdosing was projected in only one case; the cumulative PTV D95% at the end of treatment was below the clinical threshold. CONCLUSION: Uncertainty bounds were incorporated into the results of a commercial DDA tool. The cohort's statistics showed that the parotids' cumulative DVH metrics frequently exceeded the planning values if confidence intervals were included. Most of the uncertainty bounds of the PTV metrics were kept within the clinical thresholds. We verified that mid-treatment violation projections led to exceeding the constraint point at the end of the treatment. Based on a 3% threshold, approximately one fourth of the patients are expected to be replanned at mid-treatment for parotids sparing during HNC radiotherapy.

Predicting necessity of daily online adaptive replanning based on wavelet image features for MRI guided adaptive radiation therapy.

PURPOSE: Online adaptive replanning (OLAR) is generally labor-intensive and time-consuming during MRI-guided adaptive radiation therapy (MRgART). This work aims to develop a method to determine OLAR necessity during MRgART. METHODS: A machine learning classifier was developed to predict OLAR necessity based on wavelet multiscale texture features extracted from daily MRIs and was trained and tested with data from 119 daily MRI datasets acquired during MRgART for 24 pancreatic cancer patients treated on a 1.5 T MR-Linac. Spearman correlations, interclass correlation (ICC), coefficient of variance (COV), t-test (p < 0.05), self-organized map (SOM) and maximum stable extremal region (MSER) algorithm were used to determine candidate features, which were used to build the prediction models using Bayesian classifiers. The model performance was judged using the AUC of the ROC curve. RESULTS: Spearman correlation identified 123 features that were not redundant (r < 0.9). Of them 82 showed high ICC for repositioning > 0.6, 67 had a COV greater than 9% for OLAR. Among the 38 features passed the t-test, 25 passed the SOM and 12 passed the MSER. These final 12 features were used to build the classifier model. The combination of 2-3 features at a time was used to build the classifier models. The best performing model was a 3-feature combination, which can predict OLAR necessity with a CV-AUC of 0.98. CONCLUSIONS: A machine learning classifier model based on the wavelet features extracted from daily MRI for pancreatic cancer was developed to automatically and objectively determine if OLAR is necessary for a treatment fraction avoiding unnecessary effort during MRgART.

Automatic Contour Refinement for Deep Learning Auto-segmentation of Complex Organs in MRI-guided Adaptive Radiation Therapy.

Purpose: Fast and accurate auto-segmentation on daily images is essential for magnetic resonance imaging (MRI)-guided adaptive radiation therapy (ART). However, the state-of-the-art auto-segmentation based on deep learning still has limited success, particularly for complex structures in the abdomen. This study aimed to develop an automatic contour refinement (ACR) process to quickly correct for unacceptable auto-segmented contours. Methods and Materials: An improved level set-based active contour model (ACM) was implemented for the ACR process and was tested on the deep learning-based auto-segmentation of 80 abdominal MRI sets along with their ground truth contours. The performance of the ACR process was evaluated using 4 contour accuracy metrics: the Dice similarity coefficient (DSC), mean distance to agreement (MDA), surface DSC, and added path length (APL) on the auto-segmented contours of the small bowel, large bowel, combined bowels, pancreas, duodenum, and stomach. Results: A portion (3%-39%) of the corrected contours became practically acceptable per the American Association of Physicists in Medicine Task Group 132 (TG-132) recommendation (DSC >0.8 and MDA <3 mm). The best correction performance was seen in the combined bowels, where for the contours with major errors (initial DSC <0.5 or MDA >8 mm), the mean DSC increased from 0.34 to 0.59, the mean MDA decreased from 7.02 mm to 5.23 mm, and the APL reduced by almost 20 mm, whereas for the contours with minor errors, the mean DSC increased from 0.72 to 0.79, the mean MDA decreased from 3.35 mm to 3.29 mm, and more than one-third (39%) of the ACR contours became clinically acceptable. The execution time for the ACR process on one subregion was less than 2 seconds using an NVIDIA GTX 1060 GPU. Conclusions: The ACR process implemented based on the ACM was able to quickly correct for some inaccurate contours produced from MRI-based deep learning auto-segmentation of complex abdominal anatomy. The ACR method may be integrated into the auto-segmentation step to accelerate the process of MRI-guided ART.

Deep Learning-Based Automatic Detection of Brain Metastases in Heterogenous Multi-Institutional Magnetic Resonance Imaging Sets: An Exploratory Analysis of NRG-CC001.

PURPOSE: Deep learning-based algorithms have been shown to be able to automatically detect and segment brain metastases (BMs) in magnetic resonance imaging, mostly based on single-institutional data sets. This work aimed to investigate the use of deep convolutional neural networks (DCNN) for BM detection and segmentation on a highly heterogeneous multi-institutional data set. METHODS AND MATERIALS: A total of 407 patients from 98 institutions were randomly split into 326 patients from 78 institutions for training/validation and 81 patients from 20 institutions for unbiased testing. The data set contained T1-weighted gadolinium and T2-weighted fluid-attenuated inversion recovery magnetic resonance imaging acquired on diverse scanners using different pulse sequences and various acquisition parameters. Several variants of 3-dimensional U-Net based DCNN models were trained and tuned using 5-fold cross validation on the training set. Performances of different models were compared based on Dice similarity coefficient for segmentation and sensitivity and false positive rate (FPR) for detection. The best performing model was evaluated on the test set. RESULTS: A DCNN with an input size of 64 × 64 × 64 and an equal number of 128 kernels for all convolutional layers using instance normalization was identified as the best performing model (Dice similarity coefficient 0.73, sensitivity 0.86, and FPR 1.9) in the 5-fold cross validation experiments. The best performing model demonstrated consistent behavior on the test set (Dice similarity coefficient 0.73, sensitivity 0.91, and FPR 1.7) and successfully detected 7 BMs (out of 327) that were missed during manual delineation. For large BMs with diameters greater than 12 mm, the sensitivity and FPR improved to 0.98 and 0.3, respectively. CONCLUSIONS: The DCNN model developed can automatically detect and segment brain metastases with reasonable accuracy, high sensitivity, and low FPR on a multi-institutional data set with nonprespecified and highly variable magnetic resonance imaging sequences. For large BMs, the model achieved clinically relevant results. The model is robust and may be potentially used in real-world situations.

Development and implementation of an automatic air delineation technique for MRI-guided adaptive radiation therapy.

Objective.Auto-delineation of air regions on daily MRI for MR-guided online adaptive radiotherapy (MRgOART) of abdominal tumors is challenging since the air packets occur randomly and their MR intensities can be similar to some other tissue types. This work reports a new method to auto-delineate air regions on MRI.Approach.The proposed method (named DIFF method) consists of (1) generating a combined volumeVcomb, which is a union of the air-containing organs on a reference MR image offline, (2) transferringVcombfrom the reference MR to a daily MR via DIR, (3) combining the transferredVcombwith a region of high DIR inaccuracy, and (4) applying a threshold to the obtained final combined volume to generate the air volumes. The high DIR inaccuracy region was calculated from the absolute difference between the deformed daily and the reference images. This method was tested on 36 abdominal daily MRI sets acquired from 7 patients on a 1.5 T MR-Linac. The performance of DIFF was compared with alternative auto-air generation methods that (1) does not account for DIR inaccuracies, and (2) uses rigid registration instead of DIR.Main results.The results show that the proposed DIFF method can be fully automated and can be executed within 25 s. The Dice similarity coefficient of manual and DIFF auto-generated air contours was >92% for all cases, while it was 90% for the alternative auto-delineation methods. Dosimetrically, the auto-generated air regions using DIFF resulted in practically identical DVHs as those generated by using manual air contours.Significance.The DIFF method is robust and accurate and can be implemented to automatically consider the inter- and intra- fractional air volume variations during MRgOART for abdominal tumors. The use of DIFF method improves dosimetric accuracy as compared to other methods, especially beneficial for the patients with large daily abdominal air volume variations.

A Prior Knowledge-Guided, Deep Learning-Based Semiautomatic Segmentation for Complex Anatomy on Magnetic Resonance Imaging.

PURPOSE: Despite recent substantial improvement in autosegmentation using deep learning (DL) methods, labor-intensive and time-consuming slice-by-slice manual editing is often needed, particularly for complex anatomy (eg, abdominal organs). This work aimed to develop a fast, prior knowledge-guided DL semiautomatic segmentation (DL-SAS) method for complex structures on abdominal magnetic resonance imaging (MRI) scans. METHODS AND MATERIALS: A novel application using contours on an adjacent slice as a prior knowledge informant in a 2-dimensional UNet DL model to guide autosegmentation for a subsequent slice was implemented for DL-SAS. A generalized, instead of organ-specific, DL-SAS model was trained and tested for abdominal organs on T2-weighted MRI scans collected from 75 patients (65 for training and 10 for testing). The DL-SAS model performance was compared with 3 common autocontouring methods (linear interpolation, rigid propagation, and a full 3-dimensional DL autosegmentation model trained with the same training data set) based on various quantitative metrics including the Dice similarity coefficient (DSC) and ratio of acceptable slices (ROA) using paired t tests. RESULTS: For the 10 testing cases, the DL-SAS model performed best with the slice interval (SI) of 1, resulting in an average DSC of 0.93 ± 0.02, 0.92 ± 0.02, 0.91 ± 0.02, 0.88 ± 0.03, and 0.87 ± 0.02 for the large bowel, stomach, small bowel, duodenum, and pancreas, respectively. The performance decreased with increased SIs from the guidance slice. The DL-SAS method performed significantly better (P < .05) than the other 3 methods. The ROA values were in the range of 48% to 66% for all the organs with an SI of 1 for DL-SAS, higher than those for linear interpolation (31%-57% for an SI of 1) and DL auto-segmentation (16%-51%). CONCLUSIONS: The developed DL-SAS model segmented complex abdominal structures on MRI with high accuracy and efficiency and may be implemented as an interactive manual contouring tool or a contour editing tool in conjunction with a full autosegmentation process, facilitating fast and accurate segmentation for MRI-guided online adaptive radiation therapy.