Fully automated radiotherapy treatment planning: A scan to plan challenge.

BACKGROUND AND PURPOSE: Over the past decade, tools for automation of various sub-tasks in radiotherapy planning have been introduced, such as auto-contouring and auto-planning. The purpose of this study was to benchmark what degree of automation is possible. MATERIALS AND METHODS: A challenge to perform automated treatment planning for prostate and prostate bed radiotherapy was set up. Participants were provided with simulation CTs and a treatment prescription and were asked to use automated tools to produce a deliverable radiotherapy treatment plan with as little human intervention as possible. Plans were scored for their adherence to the protocol when assessed using consensus expert contours. RESULTS: Thirteen entries were received. The top submission adhered to 81.8% of the minimum objectives across all cases using the consensus contour, meeting all objectives in one of the ten cases. The same system met 89.5% of objectives when assessed with their own auto-contours, meeting all objectives in four of the ten cases. The majority of systems used in the challenge had regulatory clearance (Auto-contouring: 82.5%, Auto-planning: 77%). Despite the 'hard' rule that participants should not check or edit contours or plans, 69% reported looking at their results before submission. CONCLUSIONS: Automation of the full planning workflow from simulation CT to deliverable treatment plan is possible for prostate and prostate bed radiotherapy. While many generated plans were found to require none or minor adjustment to be regarded as clinically acceptable, the result indicated there is still a lack of trust in such systems preventing full automation.

Analyzing the Relationship between Dose and Geometric Agreement Metrics for Auto-Contouring in Head and Neck Normal Tissues.

This study aimed to determine the relationship between geometric and dosimetric agreement metrics in head and neck (H&N) cancer radiotherapy plans. A total 287 plans were retrospectively analyzed, comparing auto-contoured and clinically used contours using a Dice similarity coefficient (DSC), surface DSC (sDSC), and Hausdorff distance (HD). Organs-at-risk (OARs) with ≥200 cGy dose differences from the clinical contour in terms of Dmax (D0.01cc) and Dmean were further examined against proximity to the planning target volume (PTV). A secondary set of 91 plans from multiple institutions validated these findings. For 4995 contour pairs across 19 OARs, 90% had a DSC, sDSC, and HD of at least 0.75, 0.86, and less than 7.65 mm, respectively. Dosimetrically, the absolute difference between the two contour sets was <200 cGy for 95% of OARs in terms of Dmax and 96% in terms of Dmean. In total, 97% of OARs exhibiting significant dose differences between the clinically edited contour and auto-contour were within 2.5 cm PTV regardless of geometric agreement. There was an approximately linear trend between geometric agreement and identifying at least 200 cGy dose differences, with higher geometric agreement corresponding to a lower fraction of cases being identified. Analysis of the secondary dataset validated these findings. Geometric indices are approximate indicators of contour quality and identify contours exhibiting significant dosimetric discordance. For a small subset of OARs within 2.5 cm of the PTV, geometric agreement metrics can be misleading in terms of contour quality.

Quantifying the Risk of Technology-Driven Health Disparities in Radiation Oncology.

PURPOSE: New technologies are continuously emerging in radiation oncology. Inherent technological limitations can result in health care disparities in vulnerable patient populations. These limitations must be considered for existing and new technologies in the clinic to provide equitable care. MATERIALS AND METHODS: We created a health disparity risk assessment metric inspired by failure mode and effects analysis. We provide sample patient populations and their potential associated disparities, guidelines for clinics and vendors, and example applications of the methodology. RESULTS: A disparity risk priority number can be calculated from the product of 3 quantifiable metrics: the percentage of patients impacted, the severity of the impact of dosimetric uncertainty or quality of the radiation plan, and the clinical dependence on the evaluated technology. The disparity risk priority number can be used to rank the risk of suboptimal care due to technical limitations when comparing technologies and to plan interventions when technology is shown to have inequitable performance in the patient population of a clinic. CONCLUSIONS: The proposed methodology may simplify the evaluation of how new technology impacts vulnerable populations, help clinics quantify the limitations of their technological resources, and plan appropriate interventions to improve equity in radiation treatments.

Landmark-based auto-contouring of clinical target volumes for radiotherapy of nasopharyngeal cancer.

BACKGROUND: The delineation of clinical target volumes (CTVs) for radiotherapy for nasopharyngeal cancer is complex and varies based on the location and extent of disease. PURPOSE: The current study aimed to develop an auto-contouring solution following one protocol guidelines (NRG-HN001) that can be adjusted to meet other guidelines, such as RTOG-0225 and the 2018 International guidelines. METHODS: The study used 2-channel 3-dimensional U-Net and nnU-Net framework to auto-contour 27 normal structures in the head and neck (H&N) region that are used to define CTVs in the protocol. To define the CTV-Expansion (CTV1 and CTV2) and CTV-Overall (the outer envelope of all the CTV contours), we used adjustable morphological geometric landmarks and mimicked physician interpretation of the protocol rules by partially or fully including select anatomic structures. The results were evaluated quantitatively using the dice similarity coefficient (DSC) and mean surface distance (MSD) and qualitatively by independent reviews by two H&N radiation oncologists. RESULTS: The auto-contouring tool showed high accuracy for nasopharyngeal CTVs. Comparison between auto-contours and clinical contours for 19 patients with cancers of various stages showed a DSC of 0.94 ± 0.02 and MSD of 0.4 ± 0.4 mm for CTV-Expansion and a DSC of 0.83 ± 0.02 and MSD of 2.4 ± 0.5 mm for CTV-Overall. Upon independent review, two H&N physicians found the auto-contours to be usable without edits in 85% and 75% of cases. In 15% of cases, minor edits were required by both physicians. Thus, one physician rated 100% of the auto-contours as usable (use as is, or after minor edits), while the other physician rated 90% as usable. The second physician required major edits in 10% of cases. CONCLUSIONS: The study demonstrates the ability of an auto-contouring tool to reliably delineate nasopharyngeal CTVs based on protocol guidelines. The tool was found to be clinically acceptable by two H&N radiation oncology physicians in at least 90% of the cases.

Unlocking the adaptive advantage: correlation and machine learning classification to identify optimal online adaptive stereotactic partial breast candidates.

Objective.Online adaptive radiotherapy (OART) is a promising technique for delivering stereotactic accelerated partial breast irradiation (APBI), as lumpectomy cavities vary in location and size between simulation and treatment. However, OART is resource-intensive, increasing planning and treatment times and decreasing machine throughput compared to the standard of care (SOC). Thus, it is pertinent to identify high-yield OART candidates to best allocate resources.Approach.Reference plans (plans based on simulation anatomy), SOC plans (reference plans recalculated onto daily anatomy), and daily adaptive plans were analyzed for 31 sequential APBI targets, resulting in the analysis of 333 treatment plans. Spearman correlations between 22 reference plan metrics and 10 adaptive benefits, defined as the difference between mean SOC and delivered metrics, were analyzed to select a univariate predictor of OART benefit. A multivariate logistic regression model was then trained to stratify high- and low-benefit candidates.Main results.Adaptively delivered plans showed dosimetric benefit as compared to SOC plans for most plan metrics, although the degree of adaptive benefit varied per patient. The univariate model showed high likelihood for dosimetric adaptive benefit when the reference plan ipsilateral breast V15Gy exceeds 23.5%. Recursive feature elimination identified 5 metrics that predict high-dosimetric-benefit adaptive patients. Using leave-one-out cross validation, the univariate and multivariate models classified targets with 74.2% and 83.9% accuracy, resulting in improvement in per-fraction adaptive benefit between targets identified as high- and low-yield for 7/10 and 8/10 plan metrics, respectively.Significance.This retrospective, exploratory study demonstrated that dosimetric benefit can be predicted using only ipsilateral breast V15Gy on the reference treatment plan, allowing for a simple, interpretable model. Using multivariate logistic regression for adaptive benefit prediction led to increased accuracy at the cost of a more complicated model. This work presents a methodology for clinics wishing to triage OART resource allocation.

Longitudinal brain volumetrics in glioma survivors.

OBJECTIVE: Radiation therapy (RT) is used selectively for patients with low-grade glioma (LGG) given the concerns for potential cognitive effects in survivors, but prior cognitive outcome studies among LGG survivors have had inconsistent findings. Translational studies that characterize changes in brain anatomy and physiology after treatment of LGG may help to both contextualize cognitive findings and improve the overall understanding of radiation effects in normal brain tissue. This study aimed to investigate the hypothesis that patients with LGG who are treated with RT will experience greater brain volume loss than those who do not receive RT. METHODS: This retrospective longitudinal study included all patients with WHO grade 2 glioma who received posttreatment surveillance MRI at the University of Alabama at Birmingham. Volumetric analysis of contralateral cortical white matter (WM), cortical gray matter (GM), and hippocampus was performed on all posttreatment T1-weighted MRI sequences using the SynthSeg script. The effect of clinical and treatment variables on brain volumes was assessed using two-level hierarchical linear models. RESULTS: The final study cohort consisted of 105 patients with 1974 time points analyzed. The median length of imaging follow-up was 4.6 years (range 0.36-18.9 years), and the median number of time points analyzed per patient was 12 (range 2-40). Resection was performed in 79 (75.2%) patients, RT was administered to 61 (58.1%) patients, and chemotherapy was administered to 66 (62.9%) patients. Age at diagnosis (ß = -0.06, p < 0.001) and use of RT (ß = -1.12, p = 0.002) were associated with the slope of the contralateral cortical GM volume model (i.e., change in GM over time). Age at diagnosis (ß = -0.08, p < 0.001), midline involvement (ß = 1.31, p = 0.006), and use of RT (ß = -1.45, p = 0.001) were associated with slope of the contralateral cortical WM volume model. Age (ß = -0.0027, p = 0.001), tumor resection (ß = -0.069, p < 0.001), use of chemotherapy (ß = -0.0597, p = 0.003), and use of RT (ß = -0.0589, p < 0.001) were associated with the slope of the contralateral hippocampus volume model. CONCLUSIONS: This study demonstrated volume loss in contralateral brain structures among LGG survivors, and patients who received RT experienced greater volume loss than those who did not. The results of this study may help to provide context for cognitive outcome research in LGG survivors and inform the design of future strategies to preserve cognition.

Prospective Evaluation of Automated Contouring for CT-Based Brachytherapy for Gynecologic Malignancies.

Purpose: The use of deep learning to auto-contour organs at risk (OARs) in gynecologic radiation treatment is well established. Yet, there is limited data investigating the prospective use of auto-contouring in clinical practice. In this study, we assess the accuracy and efficiency of auto-contouring OARs for computed tomography-based brachytherapy treatment planning of gynecologic malignancies. Methods and Materials: An inhouse contouring tool automatically delineated 5 OARs in gynecologic radiation treatment planning: the bladder, small bowel, sigmoid, rectum, and urethra. Accuracy of each auto-contour was evaluated using a 5-point Likert scale: a score of 5 indicated the contour could be used without edits, while a score of 1 indicated the contour was unusable. During scoring, automated contours were edited and subsequently used for treatment planning. Dice similarity coefficient, mean surface distance, 95% Hausdorff distance, Hausdorff distance, and dosimetric changes between original and edited contours were calculated. Contour approval time and total planning time of a prospective auto-contoured (AC) cohort were compared with times from a retrospective manually contoured (MC) cohort. Results: Thirty AC cases from January 2022 to July 2022 and 31 MC cases from July 2021 to January 2022 were included. The mean (±SD) Likert score for each OAR was the following: bladder 4.77 (±0.58), small bowel 3.96 (±0.91), sigmoid colon 3.92 (±0.81), rectum 4.6 (±0.71), and urethra 4.27 (±0.78). No ACs required major edits. All OARs had a mean Dice similarity coefficient > 0.86, mean surface distance < 0.48 mm, 95% Hausdorff distance < 3.2 mm, and Hausdorff distance < 10.32 mm between original and edited contours. There was no significant difference in dose-volume histogram metrics (D2.0 cc/D0.1 cc) between original and edited contours (P values > .05). The average time to plan approval in the AC cohort was 19% less than the MC cohort. (AC vs MC, 117.0 + 18.0 minutes vs 144.9 ± 64.5 minutes, P = .045). Conclusions: Automated contouring is useful and accurate in clinical practice. Auto-contouring OARs streamlines radiation treatment workflows and decreases time required to design and approve gynecologic brachytherapy plans.

Automatic end-to-end VMAT treatment planning for rectal cancers.

BACKGROUND: The treatment planning process from segmentation to producing a deliverable plan is time-consuming and labor-intensive. Existing solutions automate the segmentation and planning processes individually. The feasibility of combining auto-segmentation and auto-planning for volumetric modulated arc therapy (VMAT) for rectal cancers in an end-to-end process is not clear. PURPOSE: To create and clinically evaluate a complete end-to-end process for auto-segmentation and auto-planning of VMAT for rectal cancer requiring only the gross tumor volume contour and a CT scan as inputs. METHODS: Patient scans and data were retrospectively selected from our institutional records for patients treated for malignant neoplasm of the rectum. We trained, validated, and tested deep learning auto-segmentation models using nnU-Net architecture for clinical target volume (CTV), bowel bag, large bowel, small bowel, total bowel, femurs, bladder, bone marrow, and female and male genitalia. For the CTV, we identified 174 patients with clinically drawn CTVs. We used data for 18 patients for all structures other than the CTV. The structures were contoured under the guidance of and reviewed by a gastrointestinal (GI) radiation oncologist. The predicted results for CTV in 35 patients and organs at risk (OAR) in six patients were scored by the GI radiation oncologist using a five-point Likert scale. For auto-planning, a RapidPlan knowledge-based planning solution was modeled for VMAT delivery with a prescription of 25 Gy in five fractions. The model was trained and tested on 20 and 34 patients, respectively. The resulting plans were scored by two GI radiation oncologists using a five-point Likert scale. Finally, the end-to-end pipeline was evaluated on 16 patients, and the resulting plans were scored by two GI radiation oncologists. RESULTS: In 31 of 35 patients, CTV contours were clinically acceptable without necessary modifications. The CTV achieved a Dice similarity coefficient of 0.85 (±0.05) and 95% Hausdorff distance of 15.25 (±5.59) mm. All OAR contours were clinically acceptable without edits, except for large and small bowel which were challenging to differentiate. However, contours for total, large, and small bowel were clinically acceptable. The two physicians accepted 100% and 91% of the auto-plans. For the end-to-end pipeline, the two physicians accepted 88% and 62% of the auto-plans. CONCLUSIONS: This study demonstrated that the VMAT treatment planning technique for rectal cancer can be automated to generate clinically acceptable and safe plans with minimal human interventions.

nnUNet for Automatic Kidney and Cyst Segmentation in Autosomal Dominant Polycystic Kidney Disease.

BACKGROUND: Autosomal Dominant Polycystic Kidney Disease (ADPKD) is a genetic disorder that causes uncontrolled kidney cyst growth, leading to kidney volume enlargement and renal function loss over time. Total kidney volume (TKV) and cyst burdens have been used as prognostic imaging biomarkers for ADPKD. OBJECTIVE: This study aimed to evaluate nnUNet for automatic kidney and cyst segmentation in T2-weighted (T2W) MRI images of ADPKD patients. METHODS: 756 kidney images were retrieved from 95 patients in the Consortium for Radiologic Imaging Studies of Polycystic Kidney Disease (CRISP) cohort (95 patients × 2 kidneys × 4 follow-up scans). The nnUNet model was trained, validated, and tested on 604, 76, and 76 images, respectively. In contrast, all images of each patient were exclusively assigned to either the training, validation, or test sets to minimize evaluation bias. The kidney and cyst regions defined using a semi-automatic method were employed as ground truth. The model performance was assessed using the Dice Similarity Coefficient (DSC), the intersection over union (IoU) score, and the Hausdorff distance (HD). RESULTS: The test DSC values were 0.96±0.01 (mean±SD) and 0.90±0.05 for kidney and cysts, respectively. Similarly, the IoU scores were 0.91± 0.09 and 0.81±0.06, and the HD values were 12.49±8.71 mm and 12.04±10.41 mm, respectively, for kidney and cyst segmentation. CONCLUSION: The nnUNet model is a reliable tool to automatically determine kidney and cyst volumes in T2W MRI images for ADPKD prognosis and therapy monitoring.

Improved Dosimetry and Plan Quality for Accelerated Partial Breast Irradiation Using Online Adaptive Radiation Therapy: A Single Institutional Study.

Purpose: Accelerated partial breast irradiation (APBI) is an attractive treatment modality for eligible patients as it has been shown to result in similar local control and improved cosmetic outcomes compared with whole breast radiation therapy. The use of online adaptive radiation therapy (OART) for APBI is promising as it allows for a reduction of planning target volume margins because breast motion and lumpectomy cavity volume changes are accounted for in daily imaging. Here we present a retrospective, single-institution evaluation on the adequacy of kV-cone beam computed tomography (CBCT) OART for APBI treatments. Methods and Materials: Nineteen patients (21 treatment sites) were treated to 30 Gy in 5 fractions between January of 2022 and May of 2023. Time between simulation and treatment, change in gross tumor (ie, lumpectomy cavity) volume, and differences in dose volume histogram metrics with adaption were analyzed. The Wilcoxon paired, nonparametric test was used to test for dose volume histogram metric differences between the scheduled plans (initial plans recalculated on daily CBCT anatomy) and delivered plans, either the scheduled or adapted plan, which was reoptimized using daily anatomy. Results: Median (interquartile range) time from simulation to first treatment was 26 days (21-32 days). During this same time, median gross tumor volume reduction was 16.0% (7.3%-23.9%) relative to simulation volume. Adaptive treatments took 31.3 minutes (27.4-36.6 minutes) from start of CBCT to treatment session end. At treatment, the adaptive plan was selected for 86% (89/103) of evaluable fractions. In evaluating plan quality, 78% of delivered plans met all target, organs at risk, and conformity metrics evaluated, compared with 34% of scheduled plans. Conclusions: Use of OART for stereotactic linac-based APBI allowed for safe, high-quality treatments in this cohort of 21 treatment courses. Although treatment delivery times were longer than traditional stereotactic body treatments, there were notable improvements in plan quality for APBI using OART.

Investigation of autosegmentation techniques on T2-weighted MRI for off-line dose reconstruction in MR-linac workflow for head and neck cancers.

BACKGROUND: In order to accurately accumulate delivered dose for head and neck cancer patients treated with the Adapt to Position workflow on the 1.5T magnetic resonance imaging (MRI)-linear accelerator (MR-linac), the low-resolution T2-weighted MRIs used for daily setup must be segmented to enable reconstruction of the delivered dose at each fraction. PURPOSE: In this pilot study, we evaluate various autosegmentation methods for head and neck organs at risk (OARs) on on-board setup MRIs from the MR-linac for off-line reconstruction of delivered dose. METHODS: Seven OARs (parotid glands, submandibular glands, mandible, spinal cord, and brainstem) were contoured on 43 images by seven observers each. Ground truth contours were generated using a simultaneous truth and performance level estimation (STAPLE) algorithm. Twenty total autosegmentation methods were evaluated in ADMIRE: 1-9) atlas-based autosegmentation using a population atlas library (PAL) of 5/10/15 patients with STAPLE, patch fusion (PF), random forest (RF) for label fusion; 10-19) autosegmentation using images from a patient's 1-4 prior fractions (individualized patient prior [IPP]) using STAPLE/PF/RF; 20) deep learning (DL) (3D ResUNet trained on 43 ground truth structure sets plus 45 contoured by one observer). Execution time was measured for each method. Autosegmented structures were compared to ground truth structures using the Dice similarity coefficient, mean surface distance (MSD), Hausdorff distance (HD), and Jaccard index (JI). For each metric and OAR, performance was compared to the inter-observer variability using Dunn's test with control. Methods were compared pairwise using the Steel-Dwass test for each metric pooled across all OARs. Further dosimetric analysis was performed on three high-performing autosegmentation methods (DL, IPP with RF and 4 fractions [IPP_RF_4], IPP with 1 fraction [IPP_1]), and one low-performing (PAL with STAPLE and 5 atlases [PAL_ST_5]). For five patients, delivered doses from clinical plans were recalculated on setup images with ground truth and autosegmented structure sets. Differences in maximum and mean dose to each structure between the ground truth and autosegmented structures were calculated and correlated with geometric metrics. RESULTS: DL and IPP methods performed best overall, all significantly outperforming inter-observer variability and with no significant difference between methods in pairwise comparison. PAL methods performed worst overall; most were not significantly different from the inter-observer variability or from each other. DL was the fastest method (33 s per case) and PAL methods the slowest (3.7-13.8 min per case). Execution time increased with a number of prior fractions/atlases for IPP and PAL. For DL, IPP_1, and IPP_RF_4, the majority (95%) of dose differences were within ± 250 cGy from ground truth, but outlier differences up to 785 cGy occurred. Dose differences were much higher for PAL_ST_5, with outlier differences up to 1920 cGy. Dose differences showed weak but significant correlations with all geometric metrics (R2 between 0.030 and 0.314). CONCLUSIONS: The autosegmentation methods offering the best combination of performance and execution time are DL and IPP_1. Dose reconstruction on on-board T2-weighted MRIs is feasible with autosegmented structures with minimal dosimetric variation from ground truth, but contours should be visually inspected prior to dose reconstruction in an end-to-end dose accumulation workflow.

Volumetric brain assessment of long-term head and neck cancer survivors.

BACKGROUND: Radiation therapy (RT) for locally advanced head and neck cancer (HNC) often exposes subcortical brain structures to radiation. We performed this study to assess region-specific brain volumetrics in a population of long term HNC survivors. METHODS AND MATERIALS: Forty HNC survivors were enrolled at a mean of 6.4 years from completion of RT. Patients underwent a research MRI protocol that included a 3D T1- weighted whole-brain scan on a 3 Tesla MRI scanner. Voxel based morphometry was performed using the Computational Anatomy Toolbox with the Neuromorphometrics atlas. Healthy controls from the Human Connectome Project were used as a comparison cohort. Study participants also completed a comprehensive neurocognitive assessment. RESULTS: The final study cohort consisted of 38 participants after excluding 2 participants due to image quality. HNC survivors displayed widespread reduction in gray matter (GM) brain region volumes that included bilateral medial frontal cortex, temporal lobe, hippocampus, supplemental motor area, and cerebellum. Greater radiation exposure was associated with reduced GM volume in the left ventral diencephalon (r = -0.512, p = 0.003). Associations between cognition and regional GM volumes were identified for motor coordination and bilateral cerebellum (left, r = 0.444, p = 0.009; right, r = 0.372, p = 0.030), confrontation naming and left amygdala (r = 0.382, p = 0.026), verbal memory and bilateral thalamus (left, r = 0.435, p = 0.010; right, r = 0.424, p = 0.012), right amygdala (r = 0.339, p = 0.050), and right putamen (r = 0.364, p = 0.034). CONCLUSIONS: Reductions in GM were observed within this cohort of primarily non-nasopharyngeal HNC survivors as compared to a control sample. GM volumes were associated with performance in multiple cognitive domains. Results of this exploratory study support the need for investigation of anatomic brain changes as an important translational corollary to cognitive problems among HNC survivors.

Fully-automated, CT-only GTV contouring for palliative head and neck radiotherapy.

Planning for palliative radiotherapy is performed without the advantage of MR or PET imaging in many clinics. Here, we investigated CT-only GTV delineation for palliative treatment of head and neck cancer. Two multi-institutional datasets of palliative-intent treatment plans were retrospectively acquired: a set of 102 non-contrast-enhanced CTs and a set of 96 contrast-enhanced CTs. The nnU-Net auto-segmentation network was chosen for its strength in medical image segmentation, and five approaches separately trained: (1) heuristic-cropped, non-contrast images with a single GTV channel, (2) cropping around a manually-placed point in the tumor center for non-contrast images with a single GTV channel, (3) contrast-enhanced images with a single GTV channel, (4) contrast-enhanced images with separate primary and nodal GTV channels, and (5) contrast-enhanced images along with synthetic MR images with separate primary and nodal GTV channels. Median Dice similarity coefficient ranged from 0.6 to 0.7, surface Dice from 0.30 to 0.56, and 95th Hausdorff distance from 14.7 to 19.7 mm across the five approaches. Only surface Dice exhibited statistically-significant difference across these five approaches using a two-tailed Wilcoxon Rank-Sum test (p ≤ 0.05). Our CT-only results met or exceeded published values for head and neck GTV autocontouring using multi-modality images. However, significant edits would be necessary before clinical use in palliative radiotherapy.

Deep learning-based dose prediction to improve the plan quality of volumetric modulated arc therapy for gynecologic cancers.

BACKGROUND: In recent years, deep-learning models have been used to predict entire three-dimensional dose distributions. However, the usability of dose predictions to improve plan quality should be further investigated. PURPOSE: To develop a deep-learning model to predict high-quality dose distributions for volumetric modulated arc therapy (VMAT) plans for patients with gynecologic cancer and to evaluate their usability in driving plan quality improvements. METHODS: A total of 79 VMAT plans for the female pelvis were used to train (47 plans), validate (16 plans), and test (16 plans) 3D dense dilated U-Net models to predict 3D dose distributions. The models received the normalized CT scan, dose prescription, and target and normal tissue contours as inputs. Three models were used to predict the dose distributions for plans in the test set. A radiation oncologist specializing in the treatment of gynecologic cancers scored the test set predictions using a 5-point scale (5, acceptable as-is; 4, prefer minor edits; 3, minor edits needed; 2, major edits needed; and 1, unacceptable). The clinical plans for which the dose predictions indicated that improvements could be made were reoptimized with constraints extracted from the predictions. RESULTS: The predicted dose distributions in the test set were of comparable quality to the clinical plans. The mean voxel-wise dose difference was -0.14 ± 0.46 Gy. The percentage dose differences in the predicted target metrics of D 1 % ${D}_{1{\mathrm{\% }}}$ and D 98 % ${D}_{98{\mathrm{\% }}}$ were -1.05% ± 0.59% and 0.21% ± 0.28%, respectively. The dose differences in the predicted organ at risk mean and maximum doses were -0.30 ± 1.66 Gy and -0.42 ± 2.07 Gy, respectively. A radiation oncologist deemed all of the predicted dose distributions clinically acceptable; 12 received a score of 5, and four received a score of 4. Replanning of flagged plans (five plans) showed that the original plans could be further optimized to give dose distributions close to the predicted dose distributions. CONCLUSIONS: Deep-learning dose prediction can be used to predict high-quality and clinically acceptable dose distributions for VMAT female pelvis plans, which can then be used to identify plans that can be improved with additional optimization.

Identifying the optimal deep learning architecture and parameters for automatic beam aperture definition in 3D radiotherapy.

PURPOSE: Two-dimensional radiotherapy is often used to treat cervical cancer in low- and middle-income countries, but treatment planning can be challenging and time-consuming. Neural networks offer the potential to greatly decrease planning time through automation, but the impact of the wide range of hyperparameters to be set during training on model accuracy has not been exhaustively investigated. In the current study, we evaluated the effect of several convolutional neural network architectures and hyperparameters on 2D radiotherapy treatment field delineation. METHODS: Six commonly used deep learning architectures were trained to delineate four-field box apertures on digitally reconstructed radiographs for cervical cancer radiotherapy. A comprehensive search of optimal hyperparameters for all models was conducted by varying the initial learning rate, image normalization methods, and (when appropriate) convolutional kernel size, the number of learnable parameters via network depth and the number of feature maps per convolution, and nonlinear activation functions. This yielded over 1700 unique models, which were all trained until performance converged and then tested on a separate dataset. RESULTS: Of all hyperparameters, the choice of initial learning rate was most consistently significant for improved performance on the test set, with all top-performing models using learning rates of 0.0001. The optimal image normalization was not consistent across architectures. High overlap (mean Dice similarity coefficient = 0.98) and surface distance agreement (mean surface distance < 2 mm) were achieved between the treatment field apertures for all architectures using the identified best hyperparameters. Overlap Dice similarity coefficient (DSC) and distance metrics (mean surface distance and Hausdorff distance) indicated that DeepLabv3+ and D-LinkNet architectures were least sensitive to initial hyperparameter selection. CONCLUSION: DeepLabv3+ and D-LinkNet are most robust to initial hyperparameter selection. Learning rate, nonlinear activation function, and kernel size are also important hyperparameters for improving performance.

Automated evaluation for rapid implementation of knowledge-based radiotherapy planning models.

PURPOSE: Knowledge-based planning (KBP) offers the ability to predict dose-volume metrics based on information extracted from previous plans, reducing plan variability and improving plan quality. As clinical integration of KBP is increasing there is a growing need for quantitative evaluation of KBP models. A .NET-based application, RapidCompare, was created for automated plan creation and analysis of Varian RapidPlan models. METHODS: RapidCompare was designed to read calculation parameters and a list of reference plans. The tool copies the reference plan field geometry and structure set, applies the RapidPlan model, optimizes the KBP plan, and generates data for quantitative evaluation of dose-volume metrics. A cohort of 85 patients, divided into training (50), testing (10), and validation (25) groups, was used to demonstrate the utility of RapidCompare. After training and tuning, the KBP model was paired with three different optimization templates to compare various planning strategies in the validation cohort. All templates used the same set of constraints for the planning target volume (PTV). For organs-at-risk, the optimization template provided constraints using the whole dose-volume histogram (DVH), fixed-dose/volume points, or generalized equivalent uniform dose (gEUD). The resulting plans from each optimization approach were compared using DVH metrics. RESULTS: RapidCompare allowed for the automated generation of 75 total plans for comparison with limited manual intervention. In comparing optimization techniques, the Dose/Volume and Lines optimization templates generated plans with similar DVH metrics, with a slight preference for the Lines technique with reductions in heart V30Gy and spinal cord max dose. The gEUD model produced high target heterogeneity. CONCLUSION: Automated evaluation allowed for the exploration of multiple optimization templates in a larger validation cohort than would have been feasible using a manual approach. A final KBP model using line optimization objectives produced the highest quality plans without human intervention.

A real-time contouring feedback tool for consensus-based contour training.

Purpose: Variability in contouring structures of interest for radiotherapy continues to be challenging. Although training can reduce such variability, having radiation oncologists provide feedback can be impractical. We developed a contour training tool to provide real-time feedback to trainees, thereby reducing variability in contouring. Methods: We developed a novel metric termed localized signed square distance (LSSD) to provide feedback to the trainee on how their contour compares with a reference contour, which is generated real-time by combining trainee contour and multiple expert radiation oncologist contours. Nine trainees performed contour training by using six randomly assigned training cases that included one test case of the heart and left ventricle (LV). The test case was repeated 30 days later to assess retention. The distribution of LSSD maps of the initial contour for the training cases was combined and compared with the distribution of LSSD maps of the final contours for all training cases. The difference in standard deviations from the initial to final LSSD maps, ΔLSSD, was computed both on a per-case basis and for the entire group. Results: For every training case, statistically significant ΔLSSD were observed for both the heart and LV. When all initial and final LSSD maps were aggregated for the training cases, before training, the mean LSSD ([range], standard deviation) was -0.8 mm ([-37.9, 34.9], 4.2) and 0.3 mm ([-25.1, 32.7], 4.8) for heart and LV, respectively. These were reduced to -0.1 mm ([-16.2, 7.3], 0.8) and 0.1 mm ([-6.6, 8.3], 0.7) for the final LSSD maps during the contour training sessions. For the retention case, the initial and final LSSD maps of the retention case were aggregated and were -1.5 mm ([-22.9, 19.9], 3.4) and -0.2 mm ([-4.5, 1.5], 0.7) for the heart and 1.8 mm ([-16.7, 34.5], 5.1) and 0.2 mm ([-3.9, 1.6],0.7) for the LV. Conclusions: A tool that uses real-time contouring feedback was developed and successfully used for contour training of nine trainees. In all cases, the utility was able to guide the trainee and ultimately reduce the variability of the trainee's contouring.