Radiotherapy Plan Quality Assurance in NRG Oncology Trials for Brain and Head/Neck Cancers: An AI-Enhanced Knowledge-Based Approach.

The quality of radiation therapy (RT) treatment plans directly affects the outcomes of clinical trials. KBP solutions have been utilized in RT plan quality assurance (QA). In this study, we evaluated the quality of RT plans for brain and head/neck cancers enrolled in multi-institutional clinical trials utilizing a KBP approach. The evaluation was conducted on 203 glioblastoma (GBM) patients enrolled in NRG-BN001 and 70 nasopharyngeal carcinoma (NPC) patients enrolled in NRG-HN001. For each trial, fifty high-quality photon plans were utilized to build a KBP photon model. A KBP proton model was generated using intensity-modulated proton therapy (IMPT) plans generated on 50 patients originally treated with photon RT. These models were then applied to generate KBP plans for the remaining patients, which were compared against the submitted plans for quality evaluation, including in terms of protocol compliance, target coverage, and organ-at-risk (OAR) doses. RT plans generated by the KBP models were demonstrated to have superior quality compared to the submitted plans. KBP IMPT plans can decrease the variation of proton plan quality and could possibly be used as a tool for developing improved plans in the future. Additionally, the KBP tool proved to be an effective instrument for RT plan QA in multi-center clinical trials.

Ultrasensitive Graphene Field-Effect Biosensors Based on Ferroelectric Polarization of Lithium Niobate for Breast Cancer Marker Detection.

Herein, we present a novel ultrasensitive graphene field-effect transistor (GFET) biosensor based on lithium niobate (LiNbO3) ferroelectric substrate for the application of breast cancer marker detection. The electrical properties of graphene are varied under the electrostatic field, which is generated through the spontaneous polarization of the ferroelectric substrate. It is demonstrated that the properties of interface between graphene and solution are also altered due to the interaction between the electrostatic field and ions. Compared with the graphene field-effect biosensor based on the conventional Si/SiO2 gate structure, our biosensor achieves a higher sensitivity to 64.7 mV/decade and shows a limit of detection down to 1.7 fM (equivalent to 12 fg·mL-1) on the detection of microRNA21 (a breast cancer marker). This innovative design combining GFETs with ferroelectric substrates holds great promise for developing an ultrahigh-sensitivity biosensing platform based on graphene that enables rapid and early disease diagnosis.

A deep learning-based method for the prediction of temporal lobe injury in patients with nasopharyngeal carcinoma.

PURPOSE: To establish a deep learning-based model to predict radiotherapy-induced temporal lobe injury (TLI). MATERIALS AND METHODS: Spatial features of dose distribution within the temporal lobe were extracted using both the three-dimensional convolution (C3D) network and the dosiomics method. The Minimal Redundancy-Maximal-Relevance (mRMR) method was employed to rank the extracted features and select the most relevant ones. Four machine learning (ML) classifiers, including logistic regression (LR), k-nearest neighbors (kNN), support vector machines (SVM) and random forest (RF), were used to establish prediction models. Nested sampling and hyperparameter tuning methods were applied to train and validate the prediction models. For comparison, a prediction model base on the conventional D0.5cc of the temporal lobe obtained from dose volume (DV) histogram was established. The area under the receiver operating characteristic (ROC) curve (AUC) was utilized to compare the predictive performance of the different models. RESULTS: A total of 127 nasopharyngeal carcinoma (NPC) patients were included in the study. In the model based on C3D deep learning features, the highest AUC value of 0.843 was achieved with 5 features. For the dosiomics features model, the highest AUC value of 0.715 was attained with 1 feature. Both of these models demonstrated superior performance compared to the prediction model based on DV parameters, which yielded an AUC of 0.695. CONCLUSION: The prediction model utilizing C3D deep learning features outperformed models based on dosiomics features or traditional parameters in predicting the onset of TLI. This approach holds promise for predicting radiation-induced toxicities and guide individualized radiotherapy.

MR radiomics predicts pathological complete response of esophageal squamous cell carcinoma after neoadjuvant chemoradiotherapy: a multicenter study.

BACKGROUND: More than 40% of patients with resectable esophageal squamous cell cancer (ESCC) achieve pathological complete response (pCR) after neoadjuvant chemoradiotherapy (nCRT), who have favorable prognosis and may benefit from an organ-preservation strategy. Our study aims to develop and validate a machine learning model based on MR radiomics to accurately predict the pCR of ESCC patients after nCRT. METHODS: In this retrospective multicenter study, eligible patients with ESCC who underwent baseline MR (T2-weighted imaging) and nCRT plus surgery were enrolled between September 2014 and September 2022 at institution 1 (training set) and between December 2017 and August 2021 at institution 2 (testing set). Models were constructed using machine learning algorithms based on clinical factors and MR radiomics to predict pCR after nCRT. The area under the curve (AUC) and cutoff analysis were used to evaluate model performance. RESULTS: A total of 155 patients were enrolled in this study, 82 in the training set and 73 in the testing set. The radiomics model was constructed based on two radiomics features, achieving AUCs of 0.968 (95%CI 0.933-0.992) in the training set and 0.885 (95%CI 0.800-0.958) in the testing set. The cutoff analysis resulted in an accuracy of 82.2% (95%CI 72.6-90.4%), a sensitivity of 75.0% (95%CI 58.3-91.7%), and a specificity of 85.7% (95%CI 75.5-96.0%) in the testing set. CONCLUSION: A machine learning model based on MR radiomics was developed and validated to accurately predict pCR after nCRT in patients with ESCC.

Pretreatment information-aided automatic segmentation for online magnetic resonance imaging-guided prostate radiotherapy.

BACKGROUND: It is necessary to contour regions of interest (ROIs) for online magnetic resonance imaging (MRI)-guided adaptive radiotherapy (MRIgART). These updated contours are used for online replanning to obtain maximum dosimetric benefits. Contouring can be accomplished using deformable image registration (DIR) and deep learning (DL)-based autosegmentation methods. However, these methods may require considerable manual editing and thus prolong treatment time. PURPOSE: The present study aimed to improve autosegmentation performance by integrating patients' pretreatment information in a DL-based segmentation algorithm. It is expected to improve the efficiency of current MRIgART process. METHODS: Forty patients with prostate cancer were enrolled retrospectively. The online adaptive MR images, patient-specific planning computed tomography (CT), and contours in CT were used for segmentation. The deformable registration of planning CT and MR images was performed first to obtain a deformable CT and corresponding contours. A novel DL network, which can integrate such patient-specific information (deformable CT and corresponding contours) into the segmentation task of MR images was designed. We performed a four-fold cross-validation for the DL models. The proposed method was compared with DIR and DL methods on segmentation of prostate cancer. The ROIs included the clinical target volume (CTV), bladder, rectum, left femur head, and right femur head. Dosimetric parameters of automatically generated ROIs were evaluated using a clinical treatment planning system. RESULTS: The proposed method enhanced the segmentation accuracy of conventional procedures. Its mean value of the dice similarity coefficient (93.5%) over the five ROIs was higher than both DIR (87.5%) and DL (87.2%). The number of patients (n = 40) that required major editing using DIR, DL, and our method were 12, 18, and 7 (CTV); 17, 4, and 1 (bladder); 8, 11, and 5 (rectum); 2, 4, and 1 (left femur head); and 3, 7, and 1 (right femur head), respectively. The Spearman rank correlation coefficient of dosimetry parameters between the proposed method and ground truth was 0.972 ± 0.040, higher than that of DIR (0.897 ± 0.098) and DL (0.871 ± 0.134). CONCLUSION: This study proposed a novel method that integrates patient-specific pretreatment information into DL-based segmentation algorithm. It outperformed baseline methods, thereby improving the efficiency and segmentation accuracy in adaptive radiotherapy.

A novel loss function to reproduce texture features for deep learning-based MRI-to-CT synthesis.

BACKGROUND: Studies on computed tomography (CT) synthesis based on magnetic resonance imaging (MRI) have mainly focused on pixel-wise consistency, but the texture features of regions of interest (ROIs) have not received appropriate attention. PURPOSE: This study aimed to propose a novel loss function to reproduce texture features of ROIs and pixel-wise consistency for deep learning-based MRI-to-CT synthesis. The method was expected to assist the multi-modality studies for radiomics. METHODS: The study retrospectively enrolled 127 patients with nasopharyngeal carcinoma. CT and MRI images were collected for each patient, and then rigidly registered as pre-procession. We proposed a gray-level co-occurrence matrix (GLCM)-based loss function to improve the reproducibility of texture features. This novel loss function could be embedded into the present deep learning-based framework for image synthesis. In this study, a typical image synthesis model was selected as the baseline, which contained a Unet trained mean square error (MSE) loss function. We embedded the proposed loss function and designed experiments to supervise different ROIs to prove its effectiveness. The concordance correlation coefficient (CCC) of the GLCM feature was employed to evaluate the reproducibility of GLCM features, which are typical texture features. Besides, we used a publicly available dataset of brain tumors to verify our loss function. RESULTS: Compared with the baseline, the proposed method improved the pixel-wise image quality metrics (MAE: 107.5 to 106.8 HU; SSIM: 0.9728 to 0.9730). CCC values of the GLCM features in GTVnx were significantly improved from 0.78 ± 0.12 to 0.82 ± 0.11 (p < 0.05 for paired t-test). Generally, > 90% (22/24) of the GLCM-based features were improved compared with the baseline, where the Informational Measure of Correlation feature was improved the most (CCC: 0.74 to 0.83). For the public dataset, the loss function also shows its effectiveness. With our proposed loss function added, the ability to reproduce texture features was improved in the ROIs. CONCLUSIONS: The proposed method reproduced texture features for MRI-to-CT synthesis, which would benefit radiomics studies based on image multi-modality synthesis.

Is it necessary to perform measurement-based patient-specific quality assurance for online adaptive radiotherapy with Elekta Unity MR-Linac?

This study aimed to investigate the necessity of measurement-based patient-specific quality assurance (PSQA) for online adaptive radiotherapy by analyzing measurement-based PSQA results and calculation-based 3D independent dose verification results with Elekta Unity MR-Linac. There are two workflows for Elekta Unity enabled in the treatment planning system: adapt to position (ATP) and adapt to shape (ATS). ATP plans are those which have relatively slighter shifts from reference plans by adjusting beam shapes or weights, whereas ATS plans are the new plans optimized from the beginning with probable re-contouring targets and organs-at-risk. PSQA gamma passing rates were measured using an MR-compatible ArcCHECK diode array for 78 reference plans and corresponding 208 adaptive plans (129 ATP plans and 79 ATS plans) of Elekta Unity. Subsequently, the relationships between ATP, or ATS plans and reference plans were evaluated separately. The Pearson's r correlation coefficients between ATP or ATS adaptive plans and corresponding reference plans were also characterized using regression analysis. Moreover, the Bland-Altman plot method was used to describe the agreement of PSQA results between ATP or ATS adaptive plans and reference plans. Additionally, Monte Carlo-based independent dose verification software ArcherQA was used to perform secondary dose check for adaptive plans. For ArcCHECK measurements, the average gamma passing rates (ArcCHECK vs. TPS) of PSQA (3%/2 mm criterion) were 99.51% ± 0.88% and 99.43% ± 0.54% for ATP and ATS plans, respectively, which were higher than the corresponding reference plans 99.34% ± 1.04% (p < 0.05) and 99.20% ± 0.71% (p < 0.05), respectively. The Pearson's r correlation coefficients were 0.720 between ATP and reference plans and 0.300 between ATS and reference plans with ArcCHECK, respectively. Furthermore, >95% of data points of differences between both ATP and ATS plans and reference plans were within ±2σ (standard deviation) of the mean difference between adaptive and reference plans with ArcCHECK measurements. With ArcherQA calculation, the average gamma passing rates (ArcherQA vs. TPS) were 98.23% ± 1.64% and 98.15% ± 1.07% for ATP and ATS adaptive plans, separately. It might be unnecessary to perform measurement-based PSQA for both ATP and ATS adaptive plans for Unity if the gamma passing rates of both measurements of corresponding reference plans and independent dose verification of adaptive plans have high gamma passing rates. Periodic machine QA and verification of adaptive plans were recommended to ensure treatment safety.

Determination of internal target volume with Cine-MRI sequence for prostate MRI-Guided radiotherapy.

BACKGROUND: In prostate radiotherapy, the intrafractional target motion negatively affects treatment accuracy. Generating internal target volume (ITV) using four-dimensional (4D) images may resolve the issue of intrafractional target motion induced by bladder filling and bowel movement. However, no 4D imaging techniques suitable for the prostate are currently available in clinical practice. PURPOSE: This study aimed to determine the ITV based on cine magnetic resonance imaging (MRI) sequence for intrafractional target motion management in prostate MRI-guided radiotherapy. MATERIALS AND METHODS: A reference ITV was generated in simulation process. Then, the reference ITV was adapted with cine MRI sequence before online planning in each fraction. Finally, the reference ITV was updated with the cine MRI sequence acquired during beam delivery after each fraction. Cine MRI sequences and positioning three-dimensional (3D) MRI from 35 patients were retrospectively collected. Clinical target volume (CTV) coverage was computed according to the two-dimensional contour of CTV and ITV on cine MRI images. Relative target size was calculated as the ratio of the volume of ITV and CTV. Isotropic planning target volume (PTV; 5 mm margin) and anisotropic PTV (3 mm margin in the posterior direction and 5 mm margin in other directions) were generated for comparison. RESULTS: The CTV coverage rate of the proposed ITV had a mean value of 98.61% ± 0.51%, whereas the CTV coverage rates of the isotropic and anisotropic PTVs were 97.43% ± 0.41% and 96.58% ± 0.73%, respectively. The proposed ITV had a relative target size of 1.79 ± 0.17, whereas the anisotropic and isotropic PTVs had relative target sizes of 1.92 ± 0.12 and 2.21 ± 0.19, respectively. For both the CTV coverage rate and target relative size, significant differences were observed between the proposed ITV and the other two PTVs (p < 0.05). CONCLUSION: The ITV achieved higher CTV coverage with smaller size than conventional isotropic and anisotropic PTVs, indicating that it can effectively deal with the intrafractional movement of the prostate.

Comprehensive evaluation of similarity between synthetic and real CT images for nasopharyngeal carcinoma.

BACKGROUND: Although magnetic resonance imaging (MRI)-to-computed tomography (CT) synthesis studies based on deep learning have significantly progressed, the similarity between synthetic CT (sCT) and real CT (rCT) has only been evaluated in image quality metrics (IQMs). To evaluate the similarity between synthetic CT (sCT) and real CT (rCT) comprehensively, we comprehensively evaluated IQMs and radiomic features for the first time. METHODS: This study enrolled 127 patients with nasopharyngeal carcinoma who underwent CT and MRI scans. Supervised-learning (Unet) and unsupervised-learning (CycleGAN) methods were applied to build MRI-to-CT synthesis models. The regions of interest (ROIs) included nasopharynx gross tumor volume (GTVnx), brainstem, parotid glands, and temporal lobes. The peak signal-to-noise ratio (PSNR), mean absolute error (MAE), root mean square error (RMSE), and structural similarity (SSIM) were used to evaluate image quality. Additionally, 837 radiomic features were extracted for each ROI, and the correlation was evaluated using the concordance correlation coefficient (CCC). RESULTS: The MAE, RMSE, SSIM, and PSNR of the body were 91.99, 187.12, 0.97, and 51.15 for Unet and 108.30, 211.63, 0.96, and 49.84 for CycleGAN. For the metrics, Unet was superior to CycleGAN (P < 0.05). For the radiomic features, the percentage of four levels (i.e., excellent, good, moderate, and poor, respectively) were as follows: GTVnx, 8.5%, 14.6%, 26.5%, and 50.4% for Unet and 12.3%, 25%, 38.4%, and 24.4% for CycleGAN; other ROIs, 5.44% ± 3.27%, 5.56% ± 2.92%, 21.38% ± 6.91%, and 67.58% ± 8.96% for Unet and 5.16% ± 1.69%, 3.5% ± 1.52%, 12.68% ± 7.51%, and 78.62% ± 8.57% for CycleGAN. CONCLUSIONS: Unet-sCT was superior to CycleGAN-sCT for the IQMs. However, neither exhibited absolute superiority in radiomic features, and both were far less similar to rCT. Therefore, further work is required to improve the radiomic similarity for MRI-to-CT synthesis. TRIAL REGISTRATION: This study was a retrospective study, so it was free from registration.

Quality assurance in a phase III, multicenter, randomized trial of POstmastectomy radioThErapy in Node posiTive breast cancer with or without Internal mAmmary nodaL irradiation (POTENTIAL): a planning benchmark case.

PURPOSE: To report the planning benchmark case results of the POTENTIAL trial-a multicenter, randomized, phase 3 trial-to evaluate the value of internal mammary nodal (IMN) irradiation for patients with high-risk breast cancer. METHODS: All participating institutions were provided the outlines of one benchmark case, and they generated radiation therapy plans per protocol. The plans were evaluated by a quality assurance team, after which the institutions resubmitted their revised plans. The information on beams arrangement, skin flash, inhomogeneity corrections, and protocol compliance was assessed in the first and final submission. RESULTS: The plans from 26 institutions were analyzed. Some major deviations were found in the first submission. The protocol compliance rates of dose coverage for the planning target volume of chest wall, supraclavicular fossa plus axilla, and IMN region (PTVim) were all significantly improved in the final submission, which were 96.2% vs. 69.2%, 100% vs. 76.9%, and 88.4% vs. 53.8%, respectively. For OARs, the compliance rates of heart Dmean, left anterior descending coronary artery V40Gy, ipsilateral lung V5Gy, and stomach V5Gy were significantly improved. In the first and final submission, the mean values of PTVim V100% were 79.9% vs. 92.7%; the mean values of heart Dmean were 11.5 Gy vs. 9.7 Gy for hypofractionated radiation therapy and 11.5 Gy vs. 11.0 Gy for conventional fractionated radiation therapy, respectively. CONCLUSION: The major deviations were corrected and protocol compliance was significantly improved after revision, which highlighted the importance of planning benchmark case to guarantee the planning quality for multicenter trials.

Efficient segmentation using domain adaptation for MRI-guided and CBCT-guided online adaptive radiotherapy.

BACKGROUND: Delineation of regions of interest (ROIs) is important for adaptive radiotherapy (ART) but it is also time consuming and labor intensive. AIM: This study aims to develop efficient segmentation methods for magnetic resonance imaging-guided ART (MRIgART) and cone-beam computed tomography-guided ART (CBCTgART). MATERIALS AND METHODS: MRIgART and CBCTgART studies enrolled 242 prostate cancer patients and 530 nasopharyngeal carcinoma patients, respectively. A public dataset of CBCT from 35 pancreatic cancer patients was adopted to test the framework. We designed two domain adaption methods to learn and adapt the features from planning computed tomography (pCT) to MRI or CBCT modalities. The pCT was transformed to synthetic MRI (sMRI) for MRIgART, while CBCT was transformed to synthetic CT (sCT) for CBCTgART. Generalized segmentation models were trained with large popular data in which the inputs were sMRI for MRIgART and pCT for CBCTgART. Finally, the personalized models for each patient were established by fine-tuning the generalized model with the contours on pCT of that patient. The proposed method was compared with deformable image registration (DIR), a regular deep learning (DL) model trained on the same modality (DL-regular), and a generalized model in our framework (DL-generalized). RESULTS: The proposed method achieved better or comparable performance. For MRIgART of the prostate cancer patients, the mean dice similarity coefficient (DSC) of four ROIs was 87.2%, 83.75%, 85.36%, and 92.20% for the DIR, DL-regular, DL-generalized, and proposed method, respectively. For CBCTgART of the nasopharyngeal carcinoma patients, the mean DSC of two target volumes were 90.81% and 91.18%, 75.17% and 58.30%, for the DIR, DL-regular, DL-generalized, and the proposed method, respectively. For CBCTgART of the pancreatic cancer patients, the mean DSC of two ROIs were 61.94% and 61.44%, 63.94% and 81.56%, for the DIR, DL-regular, DL-generalized, and the proposed method, respectively. CONCLUSION: The proposed method utilizing personalized modeling improved the segmentation accuracy of ART.

Improving accelerated 3D imaging in MRI-guided radiotherapy for prostate cancer using a deep learning method.

PURPOSE: This study was to improve image quality for high-speed MR imaging using a deep learning method for online adaptive radiotherapy in prostate cancer. We then evaluated its benefits on image registration. METHODS: Sixty pairs of 1.5 T MR images acquired with an MR-linac were enrolled. The data included low-speed, high-quality (LSHQ), and high-speed low-quality (HSLQ) MR images. We proposed a CycleGAN, which is based on the data augmentation technique, to learn the mapping between the HSLQ and LSHQ images and then generate synthetic LSHQ (synLSHQ) images from the HSLQ images. Five-fold cross-validation was employed to test the CycleGAN model. The normalized mean absolute error (nMAE), peak signal-to-noise ratio (PSNR), structural similarity index measurement (SSIM), and edge keeping index (EKI) were calculated to determine image quality. The Jacobian determinant value (JDV), Dice similarity coefficient (DSC), and mean distance to agreement (MDA) were used to analyze deformable registration. RESULTS: Compared with the LSHQ, the proposed synLSHQ achieved comparable image quality and reduced imaging time by ~ 66%. Compared with the HSLQ, the synLSHQ had better image quality with improvement of 57%, 3.4%, 26.9%, and 3.6% for nMAE, SSIM, PSNR, and EKI, respectively. Furthermore, the synLSHQ enhanced registration accuracy with a superior mean JDV (6%) and preferable DSC and MDA values compared with HSLQ. CONCLUSION: The proposed method can generate high-quality images from high-speed scanning sequences. As a result, it shows potential to shorten the scan time while ensuring the accuracy of radiotherapy.

Deep learning framework to improve the quality of cone-beam computed tomography for radiotherapy scenarios.

BACKGROUND: The application of cone-beam computed tomography (CBCT) in image-guided radiotherapy and adaptive radiotherapy remains limited due to its poor image quality. PURPOSE: In this study, we aim to develop a deep learning framework to generate high-quality CBCT images for therapeutic applications. METHODS: The synthetic CT (sCT) generation from the CBCT was proposed using a transformer-based network with a hybrid loss function. The network was trained and validated using the data from 176 patients to produce a general model that can be extensively applied to enhance CBCT images. After the first therapy, each patient can receive paired CBCT/planning CT (pCT) scans, and the obtained data were used to fine-tune the general model for further improvement. For subsequent treatment, a patient-specific, personalized model was made available. In total, 34 patients were examined for general model testing, and another six patients who underwent rescanned pCT scan were used for personalized model training and testing. RESULTS: The general model decreased the mean absolute error (MAE) from 135 HU to 59 HU as compared to the CBCT. The hybrid loss function demonstrated superior performance in CT number correction and noise/artifacts reduction. The proposed transformer-based network also showed superior power in CT number correction compared to the classical convolutional neural network. The personalized model showed improvement based on the general model in some details, and the MAE was reduced from 59 HU (for the general model) to 57 HU (p < 0.05 Wilcoxon signed-rank test). CONCLUSION: We established a deep learning framework based on transformer for clinical needs. The deep learning model demonstrated potential for continuous improvement with the help of a suggested personalized training strategy compatible with the clinical workflow.

A projection-domain correction method in CBCT reconstruction for head and neck radiotherapy using cycle-GAN and nonlocal means filter.

BACKGROUND: In radiation treatments for head and neck tumors, cone-beam computed tomography (CBCT) is employed for patient positioning and dose calculation of adaptive radiotherapy. However, the quality of CBCT is degraded by the scatter and noise, majorly impacting the accuracy of patient positioning and dose calculation. PURPOSE: To improve the quality of CBCT for patients with head and neck cancer, a projection-domain CBCT correction method was proposed using a cycle-consistent generative adversarial network (cycle-GAN) and a nonlocal means filter (NLMF) based on a reference digitally reconstructed radiograph (DRR). METHODS: A cycle-GAN was initially trained to learn mapping from CBCT projections to a DRR using the data obtained from 30 patients. For each patient, 671 CBCT projections were measured for CBCT reconstruction. Moreover, 360 Digital Reconstructed Radiographs (DRR) were computed from each patient's planning computed tomography (CT), whose projection angles ranged from 0° to 359° with an interval of 1°. By applying the trained generator of the cycle-GAN to the unseen CBCT projection, a synthetic DRR with considerably less scatter was obtained. However, annular artifacts were observed in the CBCT reconstructed with synthetic DRR. To address this issue, a NLMF based on reference DRR was used to further correct the synthetic DRR, which corrected the synthetic DRR using the calculated DRR as a reference image. Finally, the CBCT with no annular artifact and little noise was reconstructed with the corrected synthetic DRR. The proposed method was tested using the data of six patients. The corrected synthetic DRR and CBCT were compared with the corresponding real DRR and CT images. The structural preservation ability of the proposed method was evaluated using the Dice coefficients of the automatically extracted nasal cavity. Moreover, the image quality of CBCT corrected with the proposed method was objectively assessed with an five-point human scoring system and compared with CT, original CBCT and CBCT corrected with other strategies. RESULTS: The mean absolute value (MAE) of the relative error between the corrected synthetic and real DRR was <8%. The MAE between the corrected CBCT and corresponding CT was <30 HU. Moreover, the Dice coefficient of nasal cavity between the corrected CBCT image and the original image exceeded 98.8 for all the patients. Last but not least, the objective assessment of image quality showed the proposed method had an average score of 4.2 in overall image quality, which was higher than that of the original CBCT, CBCT reconstructed with synthetic DRR, and CBCT reconstructed with projections filtered with NLMF only. CONCLUSIONS: The proposed method can considerably improve the CBCT image quality with little anatomical distortion, improving the accuracy of radiotherapy for head and neck patients.

A subregion-based prediction model for local-regional recurrence risk in head and neck squamous cell carcinoma.

BACKGROUND AND PURPOSE: Given that the intratumoral heterogeneity of head and neck squamous cell carcinoma may be related to the local control rate of radiotherapy, the aim of this study was to construct a subregion-based model that can predict the risk of local-regional recurrence, and to quantitatively assess the relative contribution of subregions. MATERIALS AND METHODS: The CT images, PET images, dose images and GTVs of 228 patients with head and neck squamous cell carcinoma from four different institutions of the The Cancer Imaging Archive(TCIA) were included in the study. Using a supervoxel segmentation algorithm called maskSLIC to generate individual-level subregions. After extracting 1781 radiomics and 1767 dosiomics features from subregions, an attention-based multiple instance risk prediction model (MIR) was established. The GTV model was developed based on the whole tumour area and was used to compare the prediction performance with the MIR model. Furthermore, the MIR-Clinical model was constructed by integrating the MIR model with clinical factors. Subregional analysis was carried out through the Wilcoxon test to find the differential radiomic features between the highest and lowest weighted subregions. RESULTS: Compared with the GTV model, the C-index of MIR model was significantly increased from 0.624 to 0.721(Wilcoxon test, p value < 0.0001). When MIR model was combined with clinical factors, the C-index was further increased to 0.766. Subregional analysis showed that for LR patients, the top three differential radiomic features between the highest and lowest weighted subregions were GLRLM_ShortRunHighGrayLevelEmphasis, GRLM_HghGrayLevelRunEmphasis and GLRLM_LongRunHighGrayLevelEmphasis. CONCLUSION: This study developed a subregion-based model that can predict the risk of local-regional recurrence and quantitatively assess relevant subregions, which may provide technical support for the precision radiotherapy in head and neck squamous cell carcinoma.

Determining the quality control frequency of an MR-linac using risk matrix (RM) analysis.

PURPOSE: Quality control (QC) is performed routinely through professional guidelines. However, the recommended QC frequency may not be optimal among different institutional settings. Here we propose a novel method for determining the optimal QC frequency using risk matrix (RM) analysis. METHODS AND MATERIALS: A newly installed Magnetic Resonance linac (MR-linac) was chosen as the testing platform and six routine QC items were investigated. Failures of these QC items can adversely affect treatment outcome for the patient. Accordingly, each QC item with its assigned frequency forms a unique failure mode (FM). Using FM-effect analysis (FMEA), the severity (S), occurrence (O), and detection (D) of each FM was obtained. Next, S and D based on RM was used to determine the appropriate QC frequency. Finally, the performance of new frequency for each QC item was evaluated using the metric E = O/D. RESULTS: One new QC frequency was the same as the old frequency, two new QC frequencies were less than the old ones, and three new QC frequencies were higher than the old ones. For six QC items, E values at the new frequencies were not less than their values at the old frequencies. This indicates that the risk of machine failure is reduced at the new QC frequencies. CONCLUSIONS: The application of RM analysis provides a useful tool for determining the optimal frequencies for routine linac QC. This study demonstrated that linac QC can be performed in a way that maintains high performance of the treatment machine in a radiotherapy clinic.

Development and clinical application of a GPU-based Monte Carlo dose verification module and software for 1.5 T MR-LINAC.

BACKGROUND: Adaptive radiotherapy (ART) has made significant advances owing to magnetic resonance linear accelerator (MR-LINAC), which provides superior soft-tissue contrast, fast speed and rich functional magnetic resonance imaging (MRI) to guide radiotherapy. Independent dose verification plays a critical role in discovering errors, while several challenges remain in MR-LINAC. PURPOSE: A Monte Carlo-based GPU-accelerated dose verification module for Unity is proposed and integrated into the commercial software ArcherQA to achieve fast and accurate quality assurance (QA) for online ART. METHODS: Electron or positron motion in a magnetic field was implemented, and a material-dependent step-length limit method was used to trade off speed and accuracy. Transport was verified by dose comparison with EGSnrc in three A-B-A phantoms. Then, an accurate Monte Carlo-based Unity machine model was built in ArcherQA, including an MR-LINAC head, cryostat, coils, and treatment couch. In particular, a mixed model combining measured attenuation and homogeneous geometry was adopted for the cryostat. Several parameters in the LINAC model were tuned to commission it in the water tank. An alternating open-closed MLC plan on solid water measured with EBT-XD film was used to verify the LINAC model. Finally, the ArcherQA dose was compared with ArcCHECK measurements and GPUMCD in 30 clinical cases through the gamma test. RESULTS: ArcherQA and EGSnrc were well matched in three A-B-A phantom tests, and the relative dose difference (RDD) was less than 1.6% in the homogenous region. A Unity model was commissioned in the water tank, and the RDD in the homogenous region was less than 2%. In the alternating open-closed MLC plan, the gamma result (3%/3 mm) between ArcherQA and Film was 96.55%, better than the gamma result between GPUMCD and Film (92.13%). In 30 clinical cases, the mean three-dimensional (3D) gamma result (3%/2 mm) was 99.36% ± 1.28% between ArcherQA and ArcCHECK for the QA plans and 99.27% ± 1.04% between ArcherQA and GPUMCD for the clinical patient plans. The average dose calculation time was 106 s in all clinical patient plans. CONCLUSIONS: A GPU-accelerated Monte Carlo-based dose verification module was developed and built for the Unity MR-LINAC. The fast speed and high accuracy were proven by comparison with EGSnrc, commission data, the ArcCHECK measurement dose, and the GPUMCD dose. This module can achieve fast and accurate independent dose verification for Unity.

Combining distance and anatomical information for deep-learning based dose distribution predictions for nasopharyngeal cancer radiotherapy planning.

Purpose: Deep-learning effectively predicts dose distributions in knowledge-based radiotherapy planning. Using anatomical information that includes a structure map and computed tomography (CT) data as input has been proven to work well. The minimum distance from each voxel in normal structures to planning target volume (DPTV) closely affects each voxel's dose. In this study, we combined DPTV and anatomical information as input for a deep-learning-based dose-prediction network to improve performance. Materials and methods: One hundred patients who underwent volumetric-modulated arc therapy for nasopharyngeal cancer were selected in this study. The prediction model based on a residual network had DPTV maps, structure maps, and CT as inputs and the corresponding dose distribution maps as outputs. The performances of the combined distance and anatomical information (COM) model and the traditional anatomical (ANAT) model with two-channel inputs (structure maps and CT) were compared. A 10-fold cross validation was performed to separately train and test the COM and ANAT models. The voxel-based mean error (ME), mean absolute error (MAE), dosimetric parameters, and dice similarity coefficient (DSC) of isodose volumes were used for modeling evaluation. Results: The mean MAE of the body volume of the COM model were 4.89 ± 1.35%, highly significantly lower than those for the ANAT model of 5.07 ± 1.37% (p<0.001). The ME values of the body for the 2-type models were similar (p >0.05). The mean DSC values of the isodose volumes in the range of 60 Gy were all better in the COM model (p<0.05), and there were highly significant differences between 10 Gy and 55 Gy (p<0.001). For most organs at risk, the ME, MAE, and dosimetric parameters predicted by both models were concurrent with the ground truth values except the MAE values of the pituitary and optic chiasm in the ANAT model and the average mean dose of the right parotid in the ANAT model. Conclusions: The COM model outperformed the ANAT model and could improve automated planning with statistically highly significant differences.

Risk stratification and prognostic value of multi-modal MRI-based radiomics for extranodal nasal-type NK/T-cell lymphoma.

BACKGROUND: Magnetic resonance imaging (MRI) performs well in the locoregional assessment of extranodal nasal-type NK/T-cell lymphoma (ENKTCL). It's important to assess the value of multi-modal MRI-based radiomics for estimating overall survival (OS) in patients with ENKTCL. METHODS: Patients with ENKTCL in a prospectively cohort were systemically reviewed and all the pretreatment MRI were acquisitioned. An unsupervised spectral clustering method was used to identify risk groups of patients and radiomic features. A nomogram-revised risk index (NRI) plus MRI radiomics signature (NRI-M) was developed, and compared with the NRI. RESULTS: The 2 distinct type I and II groups of the MRI radiomics signatures were identified. The 5-year OS rates between the type I and type II groups were 87.2% versus 67.3% (P = 0.002) in all patients, and 88.8% versus 69.2% (P = 0.003) in early-stage patients. The discrimination and calibration of the NRI-M for OS prediction demonstrated a better performance than that of either MRI radiomics or NRI, with a mean area under curve (AUC) of 0.748 and 0.717 for predicting the 5-year OS in all-stages and early-stage patients. CONCLUSIONS: The NRI-M model has good performance for predicting the prognosis of ENKTCL and may help design clinical trials and improve clinical decision making.

Prediction of adaptive strategies based on deformation vector field features for MR-guided adaptive radiotherapy of prostate cancer.

BACKGROUND: The selection of adaptive strategies in MR-guided adaptive radiotherapy (MRgART) usually relies on subjective review of anatomical changes. However, this kind of review may lead to improper selection of adaptive strategy for some fractions. PURPOSE: The purpose of this study was to develop prediction models based on deformation vector field (DVF) features for automatic and accurate strategy selection, using prostate cancer as an example. METHODS: 100 fractions of 20 prostate cancer patients were retrospectively selected in this study. Treatment plans using both adapt to position (ATP) strategy and adapt to shape (ATS) strategy were generated. Optimal adaptive strategy was determined according to dosimetric evaluation. DVFs of the deformable image registration (DIR) of daily MRI and CT simulation scans were extracted. The shape, first order statistics, and spatial features were extracted from the DVFs, subjected to further selection using the minimum redundancy maximum relevance (mRMR) method. The number of features (Fn ) was hyper-tuned using bootstrapping method, and then Fn indicating a peak area under the curve (AUC) value was used to construct three prediction models. RESULTS: According to subjective review, the ATS strategy was adopted for all 100 fractions. However, the evaluation results showed that the ATP strategy could have met the clinical requirements for 23 (23%) fractions. The three prediction models showed high prediction performance, with the best performing model achieving an AUC value of 0.896, corresponding accuracy (ACC), sensitivity (SEN) and specificity (SPC) of 0.9, 0.958, and 0.667, respectively. The features used to construct prediction models included four features extracted from y direction of DVF (DVFy ) and mask, one feature from z direction of DVF (DVFz ). It indicated that the deformation along the anterior-posterior direction had a greater impact on determining the adaptive strategy than other directions. CONCLUSIONS: DVF-feature-based models could accurately predict the adaptive strategy and avoid unnecessary selection of time-consuming ATS strategy, which consequently improves the efficiency of the MRgART process.