Purpose: More and more automatic segmentation tools are being introduced in routine clinical practice. However, physicians need to spend a considerable amount of time in examining the generated contours slice by slice. This greatly reduces the benefit of the tool's automaticity. In order to overcome this shortcoming, we developed an automatic quality assurance (QA) method for automatic segmentation using convolutional neural networks (CNNs). Materials and Methods: The study cohort comprised 680 patients with early-stage breast cancer who received whole breast radiation. The overall architecture of the automatic QA method for deep learning-based segmentation included the following two main parts: a segmentation CNN model and a QA network that was established based on ResNet-101. The inputs were from computed tomography, segmentation probability maps, and uncertainty maps. Two kinds of Dice similarity coefficient (DSC) outputs were tested. One predicted the DSC quality level of each slice ([0.95, 1] for “good,” [0.8, 0.95] for “medium,” and [0, 0.8] for “bad” quality), and the other predicted the DSC value of each slice directly. The performances of the method to predict the quality levels were evaluated with quantitative metrics: balanced accuracy, F score, and the area under the receiving operator characteristic curve (AUC). The mean absolute error (MAE) was used to evaluate the DSC value outputs. Results: The proposed methods involved two types of output, both of which achieved promising accuracy in terms of predicting the quality level. For the good, medium, and bad quality level prediction, the balanced accuracy was 0.97, 0.94, and 0.89, respectively; the F score was 0.98, 0.91, and 0.81, respectively; and the AUC was 0.96, 0.93, and 0.88, respectively. For the DSC value prediction, the MAE was 0.06 ± 0.19. The prediction time was approximately 2 s per patient. Conclusions: Our method could predict the segmentation quality automatically. It can provide useful information for physicians regarding further verification and revision of automatic contours. The integration of our method into current automatic segmentation pipelines can improve the efficiency of radiotherapy contouring.

Purpose: Ensuring high-quality data for clinical trials in radiotherapy requires the generation of contours that comply with protocol definitions. The current workflow includes a manual review of the submitted contours, which is time-consuming and subjective. In this study, we developed an automated quality assurance (QA) system for lung cancer based on a segmentation model trained with deep active learning. Methods: The data included a gold atlas with 36 cases and 110 cases from the “NRG Oncology/RTOG 1308 Trial”. The first 70 cases enrolled to the RTOG 1308 formed the candidate set, and the remaining 40 cases were randomly assigned to validation and test sets (each with 20 cases). The organs-at-risk included the heart, esophagus, spinal cord, and lungs. A preliminary convolutional neural network segmentation model was trained with the gold standard atlas. To address the deficiency of the limited training data, we selected quality images from the candidate set to be added to the training set for fine-tuning of the model with deep active learning. The trained robust segmentation models were used for QA purposes. The segmentation evaluation metrics derived from the validation set, including the Dice and Hausdorff distance, were used to develop the criteria for QA decision making. The performance of the strategy was assessed using the test set. Results: The QA method achieved promising contouring error detection, with the following metrics for the heart, esophagus, spinal cord, left lung, and right lung: balanced accuracy, 0.96, 0.95, 0.96, 0.97, and 0.97, respectively; sensitivity, 0.95, 0.98, 0.96, 1.0, and 1.0, respectively; specificity, 0.98, 0.92, 0.97, 0.94, and 0.94, respectively; and area under the receiving operator characteristic curve, 0.96, 0.95, 0.96, 0.97, and 0.94, respectively. Conclusions: The proposed system automatically detected contour errors for QA. It could provide consistent and objective evaluations with much reduced investigator intervention in multicenter clinical trials.

Background and Purpose: Automatic segmentation model is proven to be efficient in delineation of organs at risk (OARs) in radiotherapy; its performance is usually evaluated with geometric differences between automatic and manual delineations. However, dosimetric differences attract more interests than geometric differences in the clinic. Therefore, this study aimed to evaluate the performance of automatic segmentation with dosimetric metrics for volumetric modulated arc therapy of esophageal cancer patients. Methods: Nineteen esophageal cancer cases were included in this study. Clinicians manually delineated the target volumes and the OARs for each case. Another set of OARs was automatically generated using convolutional neural network models. The radiotherapy plans were optimized with the manually delineated targets and the automatically delineated OARs separately. Segmentation accuracy was evaluated by Dice similarity coefficient (DSC) and mean distance to agreement (MDA). Dosimetric metrics of manually and automatically delineated OARs were obtained and compared. The clinically acceptable dose difference and volume difference of OARs between manual and automatic delineations are supposed to be within 1 Gy and 1%, respectively. Results: Average DSC values were greater than 0.92 except for the spinal cord (0.82), and average MDA values were 0.05). Although there were significant differences (P < 0.05) for the spinal cord (D2%), left lung (V10, V20, V30, and mean dose), and bilateral lung (V10, V20, V30, and mean dose), their absolute differences were small and acceptable for the clinic. The maximum dosimetric metrics differences of OARs between manual and automatic delineations were ΔD2% = 0.35 Gy for the spinal cord and ΔV30 = 0.4% for the bilateral lung, which were within the clinical criteria in this study. Conclusion: Dosimetric metrics were proposed to evaluate the automatic delineation in radiotherapy planning of esophageal cancer. Consequently, the automatic delineation could substitute the manual delineation for esophageal cancer radiotherapy planning based on the dosimetric evaluation in this study.

Purpose: The challenge of cone-beam computed tomography (CBCT) is its low image quality, which limits its application for adaptive radiotherapy (ART). Despite recent substantial improvement in CBCT imaging using the deep learning method, the image quality still needs to be improved for effective ART application. Spurred by the advantages of transformers, which employs multi-head attention mechanisms to capture long-range contextual relations between image pixels, we proposed a novel transformer-based network (called TransCBCT) to generate synthetic CT (sCT) from CBCT. This study aimed to further improve the accuracy and efficiency of ART.Materials and methods: In this study, 91 patients diagnosed with prostate cancer were enrolled. We constructed a transformer-based hierarchical encoder–decoder structure with skip connection, called TransCBCT. The network also employed several convolutional layers to capture local context. The proposed TransCBCT was trained and validated on 6,144 paired CBCT/deformed CT images from 76 patients and tested on 1,026 paired images from 15 patients. The performance of the proposed TransCBCT was compared with a widely recognized style transferring deep learning method, the cycle-consistent adversarial network (CycleGAN). We evaluated the image quality and clinical value (application in auto-segmentation and dose calculation) for ART need.Results: TransCBCT had superior performance in generating sCT from CBCT. The mean absolute error of TransCBCT was 28.8 ± 16.7 HU, compared to 66.5 ± 13.2 for raw CBCT, and 34.3 ± 17.3 for CycleGAN. It can preserve the structure of raw CBCT and reduce artifacts. When applied in auto-segmentation, the Dice similarity coefficients of bladder and rectum between auto-segmentation and oncologist manual contours were 0.92 and 0.84 for TransCBCT, respectively, compared to 0.90 and 0.83 for CycleGAN. When applied in dose calculation, the gamma passing rate (1%/1 mm criterion) was 97.5% ± 1.1% for TransCBCT, compared to 96.9% ± 1.8% for CycleGAN.Conclusions: The proposed TransCBCT can effectively generate sCT for CBCT. It has the potential to improve radiotherapy accuracy.

Purpose: This study proposes a method to identify plans that failed patient-specific quality assurance (QA) and attempts to establish a criterion to prescreen treatment plans for patient-specific QA measurements with independent Monte Carlo dose calculations.Materials and methods: Patient-specific QA results measured with an ArcCHECK diode array of 207 patients (head and neck: 25; thorax: 61; abdomen: 121) were retrospectively analyzed. All patients were treated with the volumetric modulated arc therapy (VMAT) technique and plans were optimized with a Pinnacle v16.2 treatment planning system using an analytical algorithm-based dose engine. Afterwards, phantom verification plans were designed and recalculated by an independent GPU-accelerated Monte Carlo (MC) dose engine, ArcherQA. Moreover, sensitivity and specificity analyzes of gamma passing rates between measurements and MC calculations were carried out to show the ability of MC to monitor failing plans (ArcCHECK 3%/3 mm,0.9) showed excellent classification accuracy for monitoring failing plans with independent MC calculations.

Background: Radical radiotherapy is the main treatment modality for early and locally advanced nasopharyngeal carcinoma (NPC). Magnetic resonance imaging (MRI) has the advantages of no ionizing radiation and high soft-tissue resolution compared to computed tomography (CT), but it does not provide electron density (ED) information for radiotherapy planning. Therefore, in this study, we developed a pseudo-CT (pCT) generation method to provide necessary ED information for MRI-only planning in NPC radiotherapy.Methods: Twenty patients with early-stage NPC who received radiotherapy in our hospital were investigated. First, 1433 sets of paired T1 weighted magnetic resonance (MR) simulation images and CT simulation images were rigidly registered and preprocessed. A 16-layer U-Net was used to train the pCT generative model and a “pix2pix” generative adversarial network (GAN) was also trained to compare with the pure U-Net regrading pCT quality. Second, the contours of all target volumes and organs at risk in the original CT were transferred to the pCT for planning, and the beams were copied back to the original CT for reference dose calculation. Finally, the dose distribution calculated on the pCT was compared with the reference dose distribution through gamma analysis and dose-volume indices.Results: The average time for pCT generation for each patient was 7.90 ± 0.47 seconds. The average mean (absolute) error was −9.3 ± 16.9 HU (102.6 ± 11.4 HU), and the mean-root-square error was 209.8 ± 22.6 HU. There was no significant difference between the pCT quality of pix2pix GAN and that of pure U-Net (p > 0.05). The dose distribution on the pCT was highly consistent with that on the original CT. The mean gamma pass rate (2 mm/3%, 10% low dose threshold) was 99.1% ± 0.3%, and the mean absolute difference of nasopharyngeal PGTV D_99% and PTV V_95% were 0.4% ± 0.2% and 0.1% ± 0.1%.Conclusion: The proposed deep learning model can accurately predict CT from MRI, and the generated pCT can be employed in precise dose calculations. It is of great significance to realize MRI-only planning in NPC radiotherapy, which can improve structure delineation and considerably reduce additional imaging dose, especially when an MR-guided linear accelerator is adopted for treatment.