Search

Objective. Automatic segmentation of organs-at-risk in radiotherapy planning computed tomography (CT) scans using convolutional neural networks (CNNs) is an active research area. Very large datasets are usually required to train such CNN models. In radiotherapy, large, high-quality datasets are scarce and combining data from several sources can reduce the consistency of training segmentations. It is therefore important to understand the impact of training data quality on the performance of auto-segmentation models for radiotherapy. Approach. In this study, we took an existing 3D CNN architecture for head and neck CT auto-segmentation and compare the performance of models trained with a small, well-curated dataset (n = 34) and then a far larger dataset (n = 185) containing less consistent training segmentations. We performed 5-fold cross-validations in each dataset and tested segmentation performance using the 95th percentile Hausdorff distance and mean distance-to-agreement metrics. Finally, we validated the generalisability of our models with an external cohort of patient data (n = 12) with five expert annotators. Main results. The models trained with a large dataset were greatly outperformed by models (of identical architecture) trained with a smaller, but higher consistency set of training samples. Our models trained with a small dataset produce segmentations of similar accuracy as expert human observers and generalised well to new data, performing within inter-observer variation. Significance. We empirically demonstrate the importance of highly consistent training samples when training a 3D auto-segmentation model for use in radiotherapy. Crucially, it is the consistency of the training segmentations which had a greater impact on model performance rather than the size of the dataset used.

Background and purpose: Auto-contouring performance has been widely studied in development and commissioning studies in radiotherapy, and its impact on clinical workflow assessed in that context. This study aimed to evaluate the manual adjustment of auto-contouring in routine clinical practice and to identify improvements regarding the auto-contouring model and clinical user interaction, to improve the efficiency of auto-contouring. Materials and methods: A total of 103 clinical head and neck cancer cases, contoured using a commercial deep-learning contouring system and subsequently checked and edited for clinical use were retrospectively taken from clinical data over a twelve-month period (April 2019–April 2020). The amount of adjustment performed was calculated, and all cases were registered to a common reference frame for assessment purposes. The median, 10th and 90th percentile of adjustment were calculated and displayed using 3D renderings of structures to visually assess systematic and random adjustment. Results were also compared to inter-observer variation reported previously. Assessment was performed for both the whole structures and for regional sub-structures, and according to the radiation therapy technologist (RTT) who edited the contour. Results: The median amount of adjustment was low for all structures (

Background and purpose: Head and neck (HN) radiotherapy can benefit from automatic delineation of tumor andsurrounding organs because of the complex anatomy and the regular need for adaptation. The aim of this studywas to assess the performance of a commercially available deep learning contouring (DLC) model on an externalvalidation set.Materials and methods: The CT-based DLC model, trained at the University Medical Center Groningen (UMCG),was applied to an independent set of 58 patients from the Radboud University Medical Center (RUMC). DLCresults were compared to the RUMC manual reference using the Dice similarity coefficient (DSC) and 95thpercentile of Hausdorff distance (HD95). Craniocaudal spatial information was added by calculating binnedmeasures. In addition, a qualitative evaluation compared the acceptance of manual and DLC contours in bothgroups of observers.Results: Good correspondence was shown for the mandible (DSC 0.90; HD95 3.6 mm). Performance was reasonablefor the glandular OARs, brainstem and oral cavity (DSC 0.78–0.85, HD95 3.7–7.3 mm). The otheraerodigestive tract OARs showed only moderate agreement (DSC 0.53–0.65, HD95 around 9 mm). The binnedmeasures displayed the largest deviations caudally and/or cranially.Conclusions: This study demonstrates that the DLC model can provide a reasonable starting point for delineationwhen applied to an independent patient cohort. The qualitative evaluation did not reveal large differences in theinterpretation of contouring guidelines between RUMC and UMCG observers.

INTRODUCTION: Adequate head and neck (HN) organ-at-risk (OAR) delineation is crucial for HN radiotherapy and for investigating the relationships between radiation dose to OARs and radiation-induced side effects. The automatic contouring algorithms that are currently in clinical use, such as atlas-based contouring (ABAS), leave room for improvement. The aim of this study was to use a comprehensive evaluation methodology to investigate the performance of HN OAR auto-contouring when using deep learning contouring (DLC), compared to ABAS.METHODS: The DLC neural network was trained on 589 HN cancer patients. DLC was compared to ABAS by providing each method with an independent validation cohort of 104 patients, which had also been manually contoured. For each of the 22 OAR contours - glandular, upper digestive tract and central nervous system (CNS)-related structures - the dice similarity coefficient (DICE), and absolute mean and max dose differences (|Δmean-dose| and |Δmax-dose|) performance measures were obtained. For a subset of 7 OARs, an evaluation of contouring time, inter-observer variation and subjective judgement was performed.RESULTS: DLC resulted in equal or significantly improved quantitative performance measures in 19 out of 22 OARs, compared to the ABAS (DICE/|Δmean dose|/|Δmax dose|: 0.59/4.2/4.1 Gy (ABAS); 0.74/1.1/0.8 Gy (DLC)). The improvements were mainly for the glandular and upper digestive tract OARs. DLC significantly reduced the delineation time for the inexperienced observer. The subjective evaluation showed that DLC contours were more often preferable to the ABAS contours overall, were considered to be more precise, and more often confused with manual contours. Manual contours still outperformed both DLC and ABAS; however, DLC results were within or bordering the inter-observer variability for the manual edited contours in this cohort.CONCLUSION: The DLC, trained on a large HN cancer patient cohort, outperformed the ABAS for the majority of HN OARs.

Background and purpose: Accurate segmentation of organs-at-risk (OARs) is crucial but tedious and time-consuming in adaptive radiotherapy (ART). The purpose of this work was to automate head and neck OAR-segmentation on repeat CT (rCT) by an optimal combination of human and auto-segmentation for accurate prediction of Normal Tissue Complication Probability (NTCP).Materials and methods: Human segmentation (HS) of 3 observers, deformable image registration (DIR) based contour propagation and deep learning contouring (DLC) were carried out to segment 15 OARs on 15 rCTs. The original treatment plan was re-calculated on rCT to obtain mean dose (D-mean) and con-sequent NTCP-predictions. The average Dmean and NTCP-predictions of the three observers were referred to as the gold standard to calculate the absolute difference of D-mean and NTCP-predictions (vertical bar AD(mean)vertical bar and vertical bar ANTCP vertical bar).Results: The average vertical bar AD(mean)vertical bar of parotid glands in HS was 1.40 Gy, lower than that obtained with DIR and DLC (3.64 Gy, p < 0.001 and 3.72 Gy, p < 0.001, respectively). DLC showed the highest vertical bar AD(mean)vertical bar in middle Pharyngeal Constrictor Muscle (PCM) (5.13 Gy, p = 0.01). DIR showed second highest vertical bar AD(mean)vertical bar in the cricopharyngeal inlet (2.85 Gy, p = 0.01). The semi auto-segmentation (SAS) adopted HS, DIR and DLC for segmentation of parotid glands, PCM and all other OARs, respectively. The 90th percentile vertical bar ANTCP vertical bar was 2.19%, 2.24%, 1.10% and 1.50% for DIR, DLC, HS and SAS respectively.Conclusions: Human segmentation of the parotid glands remains necessary for accurate interpretation of mean dose and NTCP during ART. Proposed semi auto-segmentation allows NTCP-predictions within 1.5% accuracy for 90% of the cases. (C) 2021 The Author(s). Published by Elsevier B.V.

Background and purpose: Developing NTCP-models for cardiac complications after breast cancer (BC) radiotherapy requires cardiac dose-volume parameters for many patients. These can be obtained by using multi-atlas based automatic segmentation (MABAS) of cardiac structures in planning CT scans. We investigated the relevance of separate multi-atlases for deep inspiration breath hold (DIBH) and free breathing (FB) CT scans.Materials and methods: BC patients scanned in DIBH (n = 10) and in FB (n = 20) were selected to create separate multi-atlases consisting of expert panel delineations of the whole heart, atria and ventricles. The accuracy of atlas-generated contours was validated with expert delineations in independent datasets (n = 10 for DIBH and FB) and reported as Dice coefficients, contour distances and dose-volume differences in relation to interobserver variability of manual contours. Dependency of MABAS contouring accuracy on breathing technique was assessed by validation of a FB atlas in DIBH patients and vice versa (cross validation).Results: For all structures the FB and DIBH atlases resulted in Dice coefficients with their respective reference contours > 0.8 and average contour distances < 2 mm smaller than slice thickness of (CTs). No significant differences were found for dose-volume parameters in volumes receiving relevant dose levels (WH, LV and RV). Accuracy of the DIBH atlas was at least similar to, and for the ventricles better than, the interobserver variation in manual delineation. Cross-validation between breathing techniques showed a reduced MABAS performance.Conclusion: Multi-atlas accuracy was at least similar to interobserver delineation variation. Separate atlases for scans made in DIBH and FB could benefit atlas performance because accuracy depends on breathing technique.

BACKGROUND AND PURPOSE: The use of artificial intelligence (AI)/ machine learning (ML) applications in radiation oncology is emerging, however no clear guidelines on commissioning of ML-based applications exist. The purpose of this study was therefore to investigate the current use and needs to support implementation of ML-based applications in routine clinical practice. MATERIALS AND METHODS: A survey was conducted among medical physicists in radiation oncology, consisting of four parts: clinical applications (1), model training, acceptance and commissioning (2), quality assurance (QA) in clinical practice and General Data Protection Regulation (GDPR) (3), and need for education and guidelines (4). Survey answers of medical physicists of the same radiation oncology centre were treated as a separate unique responder in case reporting on different AI applications. RESULTS: In total, 213 medical physicists from 202 radiation oncology centres were included in the analysis. Sixty-nine percent (1 4 7) was using (37%) or preparing (32%) to use ML in clinic, mostly for contouring and treatment planning. In 86%, human observers were still involved in daily clinical use for quality check of the output of the ML algorithm. Knowledge on ethics, legislation and data sharing was limited and scattered among responders. Besides the need for (implementation) guidelines, training of medical physicists and larger databases containing multicentre data was found to be the top priority to accommodate the further introduction of ML in clinical practice. CONCLUSION: The results of this survey indicated the need for education and guidelines on the implementation and quality assurance of ML-based applications to benefit clinical introduction. ispartof: PHYSICS & IMAGING IN RADIATION ONCOLOGY vol:16 pages:144-148 ispartof: location:Netherlands status: published

Purpose: Large-field photon radiotherapy is current standard in the treatment of cervical cancer patients. However, with the increasing availability of Pencil Beam Scanning Proton Therapy (PBS-PT) and robust treatment planning techniques, protons may have significant advantages for cervical cancer patients in the reduction of toxicity. In this study, PBS-PT and photon Volumetric Modulated Arc Therapy (VMAT) were compared, examining target coverage and organ at risk (OAR) dose, taking inter- and intra-fraction motion into account. Materials and methods: Twelve cervical cancer patients were included in this in-silico planning study. In all cases, a planning CT scan, five weekly repeat CT scans (reCTs) and an additional reCT 10 min after the first reCT were available. Two-arc VMAT and robustly optimised two- and four-field (2F and 4F) PBS-PT plans were robustly evaluated on planCTs and reCTs using set-up and range uncertainty. Nominal OAR doses and voxel-wise minimum target coverage robustness were compared. Results: Average voxel-wise minimum accumulated doses for pelvic target structures over all patients were adequate for both photon and proton treatment techniques (D98 > 95%, [91.7–99.3%]). Average accumulated dose of the para-aortic region was lower than the required 95%, D98 > 94.4% [91.1–98.2%]. With PBS-PT 4F, dose to all OARs was significantly lower than with VMAT. Major differences were observed for mean bowel bag V15Gy: 60% [39–70%] for VMAT vs 30% [10–52%] and 32% [9–54%] for PBS-PT 2F and 4F and for mean bone marrow V10Gy: 88% [82–97%] for VMAT vs 66% [60–73%] and 67% [60–75%] for PBS-PT 2F and 4F. Conclusion: Robustly optimised PBS-PT for cervical cancer patients shows equivalent target robustness against inter- and intra-fraction variability compared to VMAT, and offers significantly better OAR sparing.

Purpose: To identify a surrogate marker for late xerostomia 12 months after radiation therapy(Xer12m), according to information obtained shortly after treatment.Methods and Materials: Differences in parotid gland (PG) were quantified in image biomarkers (ΔIBMs) before and 6 weeks after radiation therapy in 107 patients. By performing stepwise forward selection, ΔIBMs that were associated with Xer12m were selected. Subsequently other variables, such as PG dose and acute xerostomia scores, were added to improve the prediction performance. All models were internally validated.Results: Prediction of Xer12m based on PG surface reduction (ΔPG-surface) was good (area under the receiver operating characteristic curve, 0.82). Parotid gland dose was related to ΔPG-surface (PR2 = 0.27). The addition of acute xerostomia scores to the ΔPG-surface improved the prediction of Xer12m significantly, and vice versa. The final model including ΔPG-surface and acute xerostomia had outstanding performance in predicting Xer12m early after radiation therapy (area under the receiver operating characteristic curve, 0.90).Conclusions: Parotid gland surface reduction was associated with late xerostomia. The early posttreatment model with ΔPG-surface and acute xerostomia scores can be considered as a surrogate marker for late xerostomia.

Purpose: To develop and validate prediction models of overall survival (OS) for head and neck cancer (HNC) patients based on image biomarkers (IBMs) of the primary tumor and positive lymph nodes (Ln) in combination with clinical parameters.Material and methods: The study cohort was composed of 289 nasopharyngeal cancer (NPC) patients from China and 298 HNC patients from the Netherlands. Multivariable Cox-regression analysis was performed to select clinical parameters from the NPC and HNC datasets, and IBMs from the NPC dataset. Final prediction models were based on both IBMs and clinical parameters.Results: Multivariable Cox-regression analysis identified three independent IBMs (tumor Volume density, Run Length Non-uniformity and Ln Major-axis-length). This IBM model showed a concordance (c)-index of 0.72 (95%Cl: 0.65-0.79) for the NPC dataset, which performed reasonably with a c-index of 0.67 (95%Cl: 0.62-0.72) in the external validation HNC dataset. When IBMs were added in clinical models, the c-index of the NPC and HNC datasets improved to 0.75 (95%Cl: 0.68-0.82; p = 0.019) and 0.75 (95%Cl: 0.70-0.81; p Conclusion: The addition of IBMs from the primary tumor and Ln improved the prognostic performance of the models containing clinical factors only. These combined models may improve pre-treatment individualized prediction of OS for HNC patients. (C) 2017 The Authors. Published by Elsevier Ireland Ltd.

Background and purpose: Current models for the prediction of late patient-rated moderate-to-severe xerostomia (XER12m) and sticky saliva (STIC12m) after radiotherapy are based on dose-volume parameters and baseline xerostomia (XERbase) or sticky saliva (STICbase) scores. The purpose is to improve prediction of XER12m, and STIC12m with patient-specific characteristics, based on CT image biomarkers (IBMs).Methods: Planning CT-scans and patient-rated outcome measures were prospectively collected for 249 head and neck cancer patients treated with definitive radiotherapy with or without systemic treatment. The potential IBMs represent geometric. CT intensity and textural characteristics of the parotid and sub mandibular glands. Lasso regularisation was used to create multivariable logistic regression models, which were internally validated by bootstrapping.Results: The prediction of XER12m could be improved significantly by adding the IBM "Short Run Emphasis" (SRE), which quantifies heterogeneity of parotid tissue, to a model with mean contra-lateral parotid gland dose and XERbase. For STIC12m, the IBM maximum CT intensity of the submandibular gland was selected in addition to STICbase and mean dose to submandibular glands.Conclusion: Prediction of XER12m and STIC12m was improved by including IBMs representing heterogeneity and density of the salivary glands, respectively. These IBMs could guide additional research to the patient-specific response of healthy tissue to radiation dose. (C) 2016 The Authors. Published by Elsevier Ireland Ltd.

The purpose of this study was to develop a method enabling synthetic computed tomography (sCT) generation of the whole abdomen using magnetic resonance imaging (MRI) images of pediatric patients with abdominal tumors. The proposed method relies on an automatic atlas-based segmentation of bone and lungs followed by an MRI intensity to synthetic Hounsfield unit conversion. Separate conversion algorithms were used for bone, lungs and soft-tissue. Rigidly registered CT and T2-weighted MR images of 30 patients in treatment position and with the same field of view were used for the evaluation of the atlas and the conversion algorithms. The dose calculation accuracy of the generated sCTs was verified for volumetric modulated arc therapy (VMAT) and pencil beam scanning (PBS). VMAT and PBS plans were robust optimized on an internal target volume (ITV) against a patient set-up uncertainty of 5 mm. Average differences between CT and sCT dose calculations for the ITV V95% were 0.5% (min 0.0%; max 5.0%) and 0.0% (min -0.1%; max 0.1%) for VMAT and PBS dose distributions, respectively. Average differences for the mean dose to the organs at risk were

BACKGROUND AND PURPOSE: Current prediction of radiation-induced xerostomia 12months after radiotherapy (Xer12m) is based on mean parotid gland dose and baseline xerostomia (Xerbaseline) scores. The hypothesis of this study was that prediction of Xer12m is improved with patient-specific characteristics extracted from (18)F-FDG PET images, quantified in PET image biomarkers (PET-IBMs).PATIENTS AND METHODS: Intensity and textural PET-IBMs of the parotid gland were collected from pre-treatment (18)F-FDG PET images of 161 head and neck cancer patients. Patient-rated toxicity was prospectively collected. Multivariable logistic regression models resulting from step-wise forward selection and Lasso regularisation were internally validated by bootstrapping. The reference model with parotid gland dose and Xerbaseline was compared with the resulting PET-IBM models.RESULTS: High values of the intensity PET-IBM (90th percentile (P90)) and textural PET-IBM (Long Run High Grey-level Emphasis 3 (LRHG3E)) were significantly associated with lower risk of Xer12m. Both PET-IBMs significantly added in the prediction of Xer12m to the reference model. The AUC increased from 0.73 (0.65-0.81) (reference model) to 0.77 (0.70-0.84) (P90) and 0.77 (0.69-0.84) (LRHG3E).CONCLUSION: Prediction of Xer12m was significantly improved with pre-treatment PET-IBMs, indicating that high metabolic parotid gland activity is associated with lower risk of developing late xerostomia. This study highlights the potential of incorporating patient-specific PET-derived functional characteristics into NTCP model development.

Background and purpose: To evaluate the dosimetric sparing and robustness against inter-fraction anatomical changes between photon and proton dose distributions for children with abdominal tumors.Material and methods: Volumetric modulated arc therapy (VMAT) and intensity-modulated pencil beam scanning (PBS) proton dose distributions were calculated for 20 abdominal pediatric cases (average 3, range 1-8 years). VMAT plans were based on a full-arc while PBS plans on 2-3 posterior-oblique irradiation fields. Plans were robustly optimized on a patient-specific internal target volume (ITV) using a uniform 5 mm set-up uncertainty. Additionally, for the PBS plans a +/- 3% proton range uncertainty was accounted for. Fractional dose re-calculations were performed using the planning computed tomography (CT) deformably registered to the daily cone-beam CT (CBCT) images. Fractional doses were accumulated rigidly. Planned and CBCT accumulated VMAT and PBS dose distributions were compared using dose-volume histogram (DVH) parameters.Results: Significant better sparing of the organs at risk with a maximum reduction in the mean dose of 40% was achieved with PBS. Mean ITV DVH parameters differences between planned and CBCT accumulated dose distributions were smaller than 0.5% for both VMAT and PBS. However, the ITV coverage (V-95% > 99%) was not reached for one patient for the accumulated VMAT dose distribution.Conclusions: For pediatric patients with abdominal tumors, improved dosimetric sparing was obtained with PBS compared to VMAT. In addition, PBS delivered by posterior-oblique irradiation fields demonstrated to be robust against anatomical inter-fraction changes. Compared to PBS, daily anatomical changes proved to affect the target coverage of VMAT dose distributions to a higher extent. (C) 2019 Elsevier B.V. All rights reserved.

We examined adjuvant irradiation of the scrotum in primary testicular lymphoma (PTL) by means of a literature review in MEDLINE, a telephone survey among Dutch institutes, and an in silico planning comparative study on scrotal irradiation in PTL. We did not find any uniform adjuvant irradiation technique assuring a safe planning target volume (PTV) coverage in published reports, and the definition of the clinical target volume is unclear. Histopathologic studies of PTL show a high invasion rate of the tunica albuginea, the epididymis, and the spermatic cord. In retrospective studies, a prescribed dose of at least 30 Gy involving the scrotum is associated with best survival. The majority of Dutch institutes irradiate the whole scrotum without using a planning computed tomography scan, with a single electron beam and a total dose of 30 Gy. The in silico planning comparative study showed that all evaluated approaches met a D-95% scrotal dose of at least 85% of the prescription dose, without exceeding the dose limits of critical organs. Photon irradiation with 2 oblique beams using wedges resulted in the best PTV coverage, with a mean value of 95% of the prescribed dose, with lowest maximum dose. Adjuvant photon or electron irradiation of the whole scrotum including the contralateral testicle with a minimum dose of 30 Gy is recommended in PTL. Computed tomography-based radiation therapy treatment planning with proper patient positioning and position verification guarantees optimal dose coverage. (C) 2013 Elsevier Inc.

Background and purpose: The aim of this study was to develop and validate a method to select head and neck cancer patients for adaptive radiotherapy (ART) pre-treatment. Potential pre-treatment selection criteria presented in recent literature were included in the analysis.Materials and methods: Deviations from the planned parotid gland mean dose (PG Delta Dmean) were estimated for 113 head and neck cancer patients by re-calculating plans on repeat CT scans. Uni- and multivariable linear regression analyses were performed to select pre-treatment parameters, and ROC curve analysis was used to determine cut off values, for selecting patients with a PG dose deviation larger than 3 Gy. The patient selection method was validated in a second patient cohort of 43 patients.Results: After multivariable analysis, the planned PG Dmean remained the only significant parameter for PG Delta Dmean. A sensitivity of 91% and 80% could be obtained using a threshold of PG Dmean of 22.2 Gy, for the development and validation cohorts, respectively. This would spare 38% (development cohort) and 24% (validation cohort) of patients from the labour-intensive ART procedure.Conclusions: The presented method to select patients for ART pre-treatment reduces the labour of ART, contributing to a more effective allocation of the department resources. (C) 2016 The Author(s). Published by Elsevier Ireland Ltd.

Purpose: The objective of this project was to define consensus guidelines for delineating organs at risk (OARS) for head and neck radiotherapy for routine daily, practice and for research purposes.Methods: Consensus guidelines were formulated based on in-depth discussions of a panel of European, North American, Asian and Australian radiation oncologists.Results: Twenty-five OARS in the head and neck region were defined with a concise description of their main anatomic boundaries. The Supplemental material provides an atlas of the consensus guidelines, projected on 1 mm axial slices. The atlas can also be obtained in DICOM-RT format on request.Conclusion: Consensus guidelines for head and neck OAR delineation were defined, aiming to decrease interobserver variability among clinicians and radiotherapy centers. (C) 2015 The Authors. Published by Elsevier Ireland Ltd.

In the last decade, many efforts have been made to characterize anatomic changes of head and neck organs at risk (OARs) and the dosimetric consequences during radiotherapy. This review was undertaken to provide an overview of the magnitude and frequency of these effects, and to investigate whether we could find criteria to identify head and neck cancer patients who may benefit from adaptive radiotherapy (ART). Possible relationships between anatomic and dosimetric changes and outcome were explicitly considered. A literature search according to PRISMA guidelines was performed in MEDLINE and EMBASE for studies concerning anatomic or dosimetric changes of head and neck OARs during radiotherapy. Fifty-one eligible studies were found. The majority of papers reported on parotid gland (PG) anatomic and dosimetric changes. In some patients, PG mean dose differences between planning CT and repeat CT scans up to 10Gy were reported. In other studies, only minor dosimetric effects (i.e.