Technical Note: Investigating of dosimetric leaf gap and leaf transmission factor variations across gantry and collimator angles in volumetric modulation arc therapy.

PURPOSE: This study investigates the influence of gantry and collimator angles on the dosimetric leaf gap (DLG) and leaf transmission factor (LTF) in a Varian LINAC equipped with rounded-end multi-leaf collimators (MLCs). While Varian guidelines recommend DLG measurements at zero degrees for both gantry and collimator, this research aims to address the knowledge gap by assessing DLG and LTF variations at different gantry and collimator angles. METHODS: Measurements were conducted using a Varian TrueBeam LINAC with a Millennium 120-leaf MLC and Eclipse TPS version 16.1. The beams utilized in this study had energies of 6 MV, 10 MV, 6 FFF, and 10 FFF. LTF and DLG were determined using ionization chambers in solid water phantoms at various gantry angles (0°, 45°, 90°, 135°, 180°, 225°, 270°, and 315°). For each gantry angle, measurements were also taken at various collimator angles (0°, 45°, 90°, and 315°). Dosimetric impacts were evaluated through VMAT Picket Fence tests and patient-specific verification using portal dosimetry for 10 clinical VMAT plans. RESULTS: LTF values showed no significant variation across gantry and collimator angles. However, DLG values exhibited notable differences depending on the gantry angle and were independent of the collimator angle. The highest DLG value was observed at a gantry angle of 270 degrees, while the lowest was at 90 degrees. The AXB DLGAverage (averaging seven measurements of DLGs at different gantry angles) model demonstrated the best agreement between measured and calculated dose distributions, indicating the importance of considering averaged DLG values across multiple gantry angles for accurate dose calculations. CONCLUSION: Our study highlights the variability of DLG with gantry angle alterations, contrary to Varian guidelines recommending DLG measurements at zero gantry angle only. We advocate for utilizing an averaged DLG value from measurements across multiple gantry angles, as outlined in our methodology.

Deep Learning Approaches for Automatic Quality Assurance of Magnetic Resonance Images Using ACR Phantom.

BACKGROUND: In recent years, there has been a growing trend towards utilizing Artificial Intelligence (AI) and machine learning techniques in medical imaging, including for the purpose of automating quality assurance. In this research, we aimed to develop and evaluate various deep learning-based approaches for automatic quality assurance of Magnetic Resonance (MR) images using the American College of Radiology (ACR) standards. METHODS: The study involved the development, optimization, and testing of custom convolutional neural network (CNN) models. Additionally, popular pre-trained models such as VGG16, VGG19, ResNet50, InceptionV3, EfficientNetB0, and EfficientNetB5 were trained and tested. The use of pre-trained models, particularly those trained on the ImageNet dataset, for transfer learning was also explored. Two-class classification models were employed for assessing spatial resolution and geometric distortion, while an approach classifying the image into 10 classes representing the number of visible spokes was used for the low contrast. RESULTS: Our results showed that deep learning-based methods can be effectively used for MR image quality assurance and can improve the performance of these models. The low contrast test was one of the most challenging tests within the ACR phantom. CONCLUSIONS: Overall, for geometric distortion and spatial resolution, all of the deep learning models tested produced prediction accuracy of 80% or higher. The study also revealed that training the models from scratch performed slightly better compared to transfer learning. For the low contrast, our investigation emphasized the adaptability and potential of deep learning models. The custom CNN models excelled in predicting the number of visible spokes, achieving commendable accuracy, recall, precision, and F1 scores.

Comparison of four commercial dose calculation algorithms in different evaluation tests.

BACKGROUND: Accurate and fast dose calculation is crucial in modern radiation therapy. Four dose calculation algorithms (AAA, AXB, CCC, and MC) are available in Varian Eclipse and RaySearch Laboratories RayStation Treatment Planning Systems (TPSs). OBJECTIVES: This study aims to evaluate and compare dosimetric accuracy of the four dose calculation algorithms applying to homogeneous and heterogeneous media, VMAT plans (based on AAPM TG-119 test cases), and the surface and buildup regions. METHODS: The four algorithms are assessed in homogeneous (IAEA-TECDOCE 1540) and heterogeneous (IAEA-TECDOC 1583) media. Dosimetric evaluation accuracy for VMAT plans is then analyzed, along with the evaluation of the accuracy of algorithms applying to the surface and buildup regions. RESULTS: Tests conducted in homogeneous media revealed that all algorithms exhibit dose deviations within 5% for various conditions, with pass rates exceeding 95% based on recommended tolerances. Additionally, the tests conducted in heterogeneous media demonstrate high pass rates for all algorithms, with a 100% pass rate observed for 6 MV and mostly 100% pass rate for 15 MV, except for CCC, which achieves a pass rate of 94%. The results of gamma index pass rate (GIPR) for dose calculation algorithms in IMRT fields show that GIPR (3% /3 mm) for all four algorithms in all evaluated tests based on TG119, are greater than 97%. The results of the algorithm testing for the accuracy of superficial dose reveal variations in dose differences, ranging from -11.9% to 7.03% for 15 MV and -9.5% to 3.3% for 6 MV, respectively. It is noteworthy that the AXB and MC algorithms demonstrate relatively lower discrepancies compared to the other algorithms. CONCLUSIONS: This study shows that generally, two dose calculation algorithms (AXB and MC) that calculate dose in medium have better accuracy than other two dose calculation algorithms (CCC and AAA) that calculate dose to water.

Design and construction of a customizable phantom for the characterization of the three-dimensional magnetic resonance imaging geometric distortion.

One of the main challenges to using magnetic resonance imaging (MRI) in radiotherapy is the existence of system-related geometric inaccuracies caused mainly by the inhomogeneity in the main magnetic field and the nonlinearities of the gradient coils. Several physical phantoms, with fixed configuration, have been developed and commercialized for the assessment of the MRI geometric distortion. In this study, we propose a new design of a customizable phantom that can fit any type of radio frequency (RF) coil. It is composed of 3D printed plastic blocks containing holes that can hold glass tubes which can be filled with any liquid. The blocks can be assembled to construct phantoms with any dimension. The feasibility of this design has been demonstrated by assembling four phantoms with high robustness allowing the assessment of the geometric distortion for the GE split head coil, the head and neck array coil, the anterior array coil, and the body coil. Phantom reproducibility was evaluated by analyzing the geometric distortion on CT acquisition of five independent assemblages of the phantom. This solution meets all expectations in terms of having a robust, lightweight, modular, and practical tool for measuring distortion in three dimensions. Mean error in the position of the tubes was less than 0.2 mm. For the geometric distortion, our results showed that for all typical MRI sequences used for radiotherapy, the mean geometric distortion was less than 1 mm and less than 2.5 mm over radial distances of 150 mm and 250 mm, respectively. These tools will be part of a quality assurance program aimed at monitoring the image quality of MRI scanners used to guide radiation therapy.

Geometric accuracy of the MR imaging techniques in the presence of motion.

Magnetic Resonance Imaging (MRI) is increasingly being used for improving tumor delineation and tumor tracking in the presence of respiratory motion. The purpose of this work is to design and build an MR compatible motion platform and to use it for evaluating the geometric accuracy of MR imaging techniques during respiratory motion. The motion platform presented in this work is composed of a mobile base made up of a flat plate and four wheels. The mobile base is attached from one end and through a rigid rod to a synchrony motion table by Accuray® placed at the end of the MRI table and from the other end to an elastic rod. The geometric accuracy was measured by placing a control point-based phantom on top of the mobile base. In-house software module was used to automatically assess the geometric distortion. The blurring artifact was also assessed by measuring the Full Width Half Maximum (FWHM) of each control point. Our results were assessed for 50, 100, and 150 mm radial distances, with a mean geometric distortion during the superior-inferior motion of 0.27, 0.41, and 0.55 mm, respectively. Adding the anterior-posterior motion, the mean geometric distortions increased to 0.4, 0.6, and 0.8 mm. Blurring was observed during motion causing an increase in the FWHM of ≈30%. The platform presented in this work provides a valuable tool for the assessment of the geometric accuracy and blurring artifact for MR during motion. Although the main objective was to test the spatial accuracy of an MR system during motion, the modular aspect of the presented platform enables the use of any commercially available phantom for a full quality control of the MR system during motion.

Voluntary Deep Inspiration Breath-hold Reduces the Heart Dose Without Compromising the Target Volume Coverage During Radiotherapy for Left-sided Breast Cancer.

BACKGROUND: During radiotherapy of left-sided breast cancer, parts of the heart are irradiated, which may lead to late toxicity. We report on the experience of single institution with cardiac-sparing radiotherapy using voluntary deep inspiration breath hold (V-DIBH) and compare its dosimetric outcome with free breathing (FB) technique. PATIENTS AND METHODS: Left-sided breast cancer patients, treated at our department with postoperative radiotherapy of breast/chest wall +/- regional lymph nodes between May 2015 and January 2017, were considered for inclusion. FB-computed tomography (CT) was obtained and dose-planning performed. Cases with cardiac V25Gy ≥ 5% or risk factors for heart disease were coached for V-DIBH. Compliant patients were included. They underwent additional CT in V-DIBH for planning, followed by V-DIBH radiotherapy. Dose volume histogram parameters for heart, lung and optimized planning target volume (OPTV) were compared between FB and BH. Treatment setup shifts and systematic and random errors for V-DIBH technique were compared with FB historic control. RESULTS: Sixty-three patients were considered for V-DIBH. Nine (14.3%) were non-compliant at coaching, leaving 54 cases for analysis. When compared with FB, V-DIBH resulted in a significant reduction of mean cardiac dose from 6.1 +/- 2.5 to 3.2 +/- 1.4 Gy (p < 0.001), maximum cardiac dose from 51.1 +/- 1.4 to 48.5 +/- 6.8 Gy (p = 0.005) and cardiac V25Gy from 8.5 +/- 4.2 to 3.2 +/- 2.5% (p < 0.001). Heart volumes receiving low (10-20 Gy) and high (30-50 Gy) doses were also significantly reduced. Mean dose to the left anterior coronary artery was 23.0 (+/- 6.7) Gy and 14.8 (+/- 7.6) Gy on FB and V-DIBH, respectively (p < 0.001). Differences between FB- and V-DIBH-derived mean lung dose (11.3 +/- 3.2 vs. 10.6 +/- 2.6 Gy), lung V20Gy (20.5 +/- 7 vs. 19.5 +/- 5.1 Gy) and V95% for the OPTV (95.6 +/- 4.1 vs. 95.2 +/- 6.3%) were non-significant. V-DIBH-derived mean shifts for initial patient setup were ≤ 2.7 mm. Random and systematic errors were ≤ 2.1 mm. These results did not differ significantly from historic FB controls. CONCLUSIONS: When compared with FB, V-DIBH demonstrated high setup accuracy and enabled significant reduction of cardiac doses without compromising the target volume coverage. Differences in lung doses were non-significant.

MRI Reduces Variation of Contouring for Boost Clinical Target Volume in Breast Cancer Patients Without Surgical Clips in the Tumour Bed.

BACKGROUND: Omitting the placement of clips inside tumour bed during breast cancer surgery poses a challenge for delineation of lumpectomy cavity clinical target volume (CTVLC). We aimed to quantify inter-observer variation and accuracy for CT- and MRI-based segmentation of CTVLC in patients without clips. PATIENTS AND METHODS: CT- and MRI-simulator images of 12 breast cancer patients, treated by breast conserving surgery and radiotherapy, were included in this study. Five radiation oncologists recorded the cavity visualization score (CVS) and delineated CTVLC on both modalities. Expert-consensus (EC) contours were delineated by a senior radiation oncologist, respecting opinions of all observers. Inter-observer volumetric variation and generalized conformity index (CIgen) were calculated. Deviations from EC contour were quantified by the accuracy index (AI) and inter-delineation distances (IDD). RESULTS: Mean CVS was 3.88 +/- 0.99 and 3.05 +/- 1.07 for MRI and CT, respectively (p = 0.001). Mean volumes of CTVLC were similar: 154 +/- 26 cm3 on CT and 152 +/- 19 cm3 on MRI. Mean CIgen and AI were superior for MRI when compared with CT (CIgen: 0.74 +/- 0.07 vs. 0.67 +/- 0.12, p = 0.007; AI: 0.81 +/- 0.04 vs. 0.76 +/- 0.07; p = 0.004). CIgen and AI increased with increasing CVS. Mean IDD was 3 mm +/- 1.5 mm and 3.6 mm +/- 2.3 mm for MRI and CT, respectively (p = 0.017). CONCLUSIONS: When compared with CT, MRI improved visualization of post-lumpectomy changes, reduced interobserver variation and improved the accuracy of CTVLC contouring in patients without clips in the tumour bed. Further studies with bigger sample sizes are needed to confirm our findings.

Ultra-Hypofractionated Prostate Radiotherapy With Online Adaptive Technique: A Case Report.

Ultra-hypofractionated radiotherapy (UHF RT) is revolutionizing the treatment approach for low- and intermediate-risk prostate cancer patients. This study reports the planning process of UHF RT utilizing the cone beam computed tomography (CBCT)-based online adaptive radiotherapy (OART) treatment with the Ethos system, focusing on a comparative analysis between OART and image-guided radiotherapy (IGRT) plans. We also assessed the pre-planning capabilities of the Ethos system against the CyberKnife (CK) (Accuray, Sunnyvale, CA) system. A 66-year-old patient, diagnosed with prostatic acinar adenocarcinoma confirmed via biopsy and presenting with elevated prostate-specific antigen (PSA) levels, underwent UHF OART treatment using the Ethos system. The planning encompassed delineating the gross target volume (GTV) as the prostate, while the clinical target volume (CTV) comprised the prostate and proximal seminal vesicle. The planning target volume (PTV) was derived from the CTV with a 5 mm external margin except for a 3 mm posterior margin. A simultaneous integrated boost (SIB) technique was employed, delivering 40 Gy in five fractions (8 Gy per fraction) to the gross tumor volume (GTV) and 36.25 Gy in five fractions (7.25 Gy per fraction) to the remaining part of the planning target volume (PTV), with treatments scheduled biweekly. We compared OART and IGRT plans and conducted a comparative analysis between Ethos planning and the CK system for pre-planning assessment. When comparing Ethos planning and CK plans, Ethos demonstrated slightly better target coverage and organ-at-risk (OAR) sparing. However, CK plans showed superior containment of low-dose spillage, particularly at 50% and 25% iso-doses, due to non-coplanar beam arrangements. Our results demonstrated that OART plans yielded superior target coverage and improved OAR sparing compared to IGRT plans. Notably, the entire OART process, from planning to delivery, was accomplished within 27 minutes. The Ethos OART system's ability to adapt to daily anatomical changes, efficient workflow, and superior OAR-sparing capabilities make it a promising option for prostate cancer treatment using UHF RT.

Cancerous and Non-Cancerous MRI Classification Using Dual DCNN Approach.

Brain cancer is a life-threatening disease requiring close attention. Early and accurate diagnosis using non-invasive medical imaging is critical for successful treatment and patient survival. However, manual diagnosis by radiologist experts is time-consuming and has limitations in processing large datasets efficiently. Therefore, efficient systems capable of analyzing vast amounts of medical data for early tumor detection are urgently needed. Deep learning (DL) with deep convolutional neural networks (DCNNs) emerges as a promising tool for understanding diseases like brain cancer through medical imaging modalities, especially MRI, which provides detailed soft tissue contrast for visualizing tumors and organs. DL techniques have become more and more popular in current research on brain tumor detection. Unlike traditional machine learning methods requiring manual feature extraction, DL models are adept at handling complex data like MRIs and excel in classification tasks, making them well-suited for medical image analysis applications. This study presents a novel Dual DCNN model that can accurately classify cancerous and non-cancerous MRI samples. Our Dual DCNN model uses two well-performed DL models, i.e., inceptionV3 and denseNet121. Features are extracted from these models by appending a global max pooling layer. The extracted features are then utilized to train the model with the addition of five fully connected layers and finally accurately classify MRI samples as cancerous or non-cancerous. The fully connected layers are retrained to learn the extracted features for better accuracy. The technique achieves 99%, 99%, 98%, and 99% of accuracy, precision, recall, and f1-scores, respectively. Furthermore, this study compares the Dual DCNN's performance against various well-known DL models, including DenseNet121, InceptionV3, ResNet architectures, EfficientNetB2, SqueezeNet, VGG16, AlexNet, and LeNet-5, with different learning rates. This study indicates that our proposed approach outperforms these established models in terms of performance.

Investigating the impact of rapidplan on ethos automated planning.

Automated planning has surged in popularity within external beam radiation therapy in recent times. Leveraging insights from previous clinical knowledge could enhance auto-planning quality. In this work, we evaluated the performance of Ethos automated planning with knowledge-based guidance, specifically using Rapidplan (RP). Seventy-four patients with head-and-neck (HN) cancer and 37 patients with prostate cancer were used to construct separate RP models. Additionally, 16 patients from each group (HN and prostate) were selected to assess the performance of Ethos auto-planning results. Initially, a template-based Ethos plan (Non-RP plan) was generated, followed by integrating the corresponding RP model's DVH estimates into the optimization process to generate another plan (RP plan). We compared the target coverage, OAR doses, and total monitor units between the non-RP and RP plans. Both RP and non-RP plans achieved comparable target coverage in HN and Prostate cases, with a negligible difference of less than 0.5% (p > 0.2). RP plans consistently demonstrated lower doses of OARs in both HN and prostate cases. Specifically, the mean doses of OARs were significantly reduced by 9% (p < 0.05). RP plans required slightly higher monitor units in both HN and prostate sites (p < 0.05), however, the plan generation time was almost similar (p > 0.07). The inclusion of the RP model reduced the OAR doses, particularly reducing the mean dose to critical organs compared to non-RP plans while maintaining similar target coverage. Our findings provide valuable insights for clinics adopting Ethos planning, potentially enhancing the auto-planning to operate optimally.

Assessment of four dose calculation algorithms using IAEA-TECDOC-1583 with medium dependency correction factor (Kmed) application.

PURPOSE: This study discusses the measurement of dose in clinical commissioning tests described in IAEA-TECDOC-1583. It explores the application of Monte Carlo (MC) modelled medium dependency correction factors (Kmed) for accurate dose measurement in bone and lung materials using the CIRS phantom. METHODS: BEAMnrc codes simulate radiation sources and model radiation transport for 6 MV and 15 MV photon beams. CT images of the CIRS phantom are converted to an MC compatible phantom. The PTW 30013 farmer chamber measures doses within modeled CIRS phantom. Kmed are determined by averaging values from four central voxels within the sensitive volume of the farmer chamber. Kmed is calculated for Dm.m and Dw.w algorithm types in bone and lung media for both photon beams. RESULTS: Average modelled correction factors for Dm.m calculations using the farmer chamber are 0.976 (±0.1 %) for 6 MV and 0.979 (±0.1 %) for 15 MV in bone media. Correspondingly, correction factors for Dw.w calculations are 0.99 (±0.3 %) and 0.992 (±0.4 %), respectively. For lung media, average correction factors for Dm.m calculations are 1.02 (±0.3 %) for 6 MV and 1.022 (±0.4 %) for 15 MV. Correspondingly, correction factors for Dw.w calculations are 1.01 (±0.3 %) and 1.012 (±0.2 %), respectively. CONCLUSIONS: This study highlights the significant impact of applying Kmed on dose differences between measurement and calculation during the dose audit process.

Managing Inconsistent Bladder Volumes in the Prostate Cancer Patient Using Daily Online Adaptive RT: A Case Report.

PURPOSE: We report the use of online adaptive radiotherapy (OART) aiming to improve dosimetric parameters in the prostate cancer patient who had lower urinary tract symptoms that caused him not to adhere to the standard bladder filling protocol. METHODS AND MATERIALS: The reference treatment plan for adaptive radiotherapy plan was generated for the pelvis and the solitary bony lesion using the Ethos treatment planning system. For each treatment session, high-quality iterative reconstructed cone beam CT (CBCT) images were acquired, and the system automatically generated an optimal adaptive plan after verification of contours. Image-guided RT (IGRT) plans were also created using the reference plan recalculated on the CBCT scan and were compared with adaptive plans. RESULTS: The reference bladder volume in the planning CT scan was 173 cc, and the mean bladder volume difference over the course was 25.4% ± 16.6%. The ART offered superior target coverage for PTV 70 Gy over online IGRT (V95: 90.5 ± 3.2 % Vs 97.3 ± 0.4%; p=0.000) and the bladder was also better spared from the high dose (V65 Gy: 17.9 ± 9.1% vs 14.8 ± 3.6%; p=0.03). However, the mean rectum V65 doses were very similar in both plans. CONCLUSION: Managing the inconsistent bladder volume was feasible in the prostate cancer patient using the CBCT-guided OART and our analysis confirmed that adaptive plans offered better target coverage while sparing the bladder from high radiation doses in comparison to online IGRT plans. KEY WORDS: radiotherapy, CBCT, online adaptive radiotherapy, image-guided RT.

Prediction of cervix cancer stage and grade from diffusion weighted imaging using EfficientNet.

Purpose. This study aims to introduce an innovative noninvasive method that leverages a single image for both grading and staging prediction. The grade and the stage of cervix cancer (CC) are determined from diffusion-weighted imaging (DWI) in particular apparent diffusion coefficient (ADC) maps using deep convolutional neural networks (DCNN).Methods. datasets composed of 85 patients having annotated tumor stage (I, II, III, and IV), out of this, 66 were with grade (II and III) and the remaining patients with no reported grade were retrospectively collected. The study was IRB approved. For each patient, sagittal and axial slices containing the gross tumor volume (GTV) were extracted from ADC maps. These were computed using the mono exponential model from diffusion weighted images (b-values = 0, 100, 1000) that were acquired prior to radiotherapy treatment. Balanced training sets were created using the Synthetic Minority Oversampling Technique (SMOTE) and fed to the DCNN. EfficientNetB0 and EfficientNetB3 were transferred from the ImageNet application to binary and four-class classification tasks. Five-fold stratified cross validation was performed for the assessment of the networks. Multiple evaluation metrics were computed including the area under the receiver operating characteristic curve (AUC). Comparisons with Resnet50, Xception, and radiomic analysis were performed.Results. for grade prediction, EfficientNetB3 gave the best performance with AUC = 0.924. For stage prediction, EfficientNetB0 was the best with AUC = 0.931. The difference between both models was, however, small and not statistically significant EfficientNetB0-B3 outperformed ResNet50 (AUC = 0.71) and Xception (AUC = 0.89) in stage prediction, and demonstrated comparable results in grade classification, where AUCs of 0.89 and 0.90 were achieved by ResNet50 and Xception, respectively. DCNN outperformed radiomic analysis that gave AUC = 0.67 (grade) and AUC = 0.66 (stage).Conclusion.the prediction of CC grade and stage from ADC maps is feasible by adapting EfficientNet approaches to the medical context.

Generation of synthetic CT from CBCT using deep learning approaches for head and neck cancer patients.

Purpose.To create a synthetic CT (sCT) from daily CBCT using either deep residual U-Net (DRUnet), or conditional generative adversarial network (cGAN) for adaptive radiotherapy planning (ART).Methods.First fraction CBCT and planning CT (pCT) were collected from 93 Head and Neck patients who underwent external beam radiotherapy. The dataset was divided into training, validation, and test sets of 58, 10 and 25 patients respectively. Three methods were used to generate sCT, 1. Nonlocal means patch based method was modified to include multiscale patches defining the multiscale patch based method (MPBM), 2. An encoder decoder 2D Unet with imbricated deep residual units was implemented, 3. DRUnet was integrated to the generator part of cGAN whereas a convolutional PatchGAN classifier was used as the discriminator. The accuracy of sCT was evaluated geometrically using Mean Absolute Error (MAE). Clinical Volumetric Modulated Arc Therapy (VMAT) plans were copied from pCT to registered CBCT and sCT and dosimetric analysis was performed by comparing Dose Volume Histogram (DVH) parameters of planning target volumes (PTVs) and organs at risk (OARs). Furthermore, 3D Gamma analysis (2%/2mm, global) between the dose on the sCT or CBCT and that on the pCT was performed.Results. The average MAE calculated between pCT and CBCT was 180.82 ± 27.37HU. Overall, all approaches significantly reduced the uncertainties in CBCT. Deep learning approaches outperformed patch-based methods with MAE = 67.88 ± 8.39HU (DRUnet) and MAE = 72.52 ± 8.43HU (cGAN) compared to MAE = 90.69 ± 14.3HU (MPBM). The percentages of DVH metric deviations were below 0.55% for PTVs and 1.17% for OARs using DRUnet. The average Gamma pass rate was 99.45 ± 1.86% for sCT generated using DRUnet.Conclusion.DL approaches outperformed MPBM. Specifically, DRUnet could be used for the generation of sCT with accurate intensities and realistic description of patient anatomy. This could be beneficial for CBCT based ART.

Virtual pretreatment patient-specific quality assurance of volumetric modulated arc therapy using deep learning.

BACKGROUND: Automatic patient-specific quality assurance (PSQA) is recently explored using artificial intelligence approaches, and several studies reported the development of machine learning models for predicting the gamma pass rate (GPR) index only. PURPOSE: To develop a novel deep learning approach using a generative adversarial network (GAN) to predict the synthetic measured fluence. METHODS AND MATERIALS: A novel training method called "dual training," which involves the training of the encoder and decoder separately, was proposed and evaluated for cycle GAN (cycle-GAN) and conditional GAN (c-GAN). A total of 164 VMAT treatment plans, including 344 arcs (training data: 262, validation data: 30, and testing data: 52) from various treatment sites, were selected for prediction model development. For each patient, portal-dose-image-prediction fluence from TPS was used as input, and measured fluence from EPID was used as output/response for model training. Predicted GPR was derived by comparing the TPS fluence with the synthetic measured fluence generated by the DL models using gamma evaluation of criteria 2%/2 mm. The performance of dual training was compared against the traditional single-training approach. In addition, we also developed a separate classification model specifically designed to detect automatically three types of errors (rotational, translational, and MU-scale) in the synthetic EPID-measured fluence. RESULTS: Overall, the dual training improved the prediction accuracy of both cycle-GAN and c-GAN. Predicted GPR results of single training were within 3% for 71.2% and 78.8% of test cases for cycle-GAN and c-GAN, respectively. Moreover, similar results for dual training were 82.7% and 88.5% for cycle-GAN and c-GAN, respectively. The error detection model showed high classification accuracy (>98%) for detecting errors related to rotational and translational errors. However, it struggled to differentiate the fluences with "MU scale error" from "error-free" fluences. CONCLUSION: We developed a method to automatically generate the synthetic measured fluence and identify errors within them. The proposed dual training improved the PSQA prediction accuracy of both the GAN models, with c-GAN demonstrating superior performance over the cycle-GAN. Our results indicate that the c-GAN with dual training approach combined with error detection model, can accurately generate the synthetic measured fluence for VMAT PSQA and identify the errors. This approach has the potential to pave the way for virtual patient-specific QA of VMAT treatments.

Investigation of radiomics and deep convolutional neural networks approaches for glioma grading.

Purpose.To determine glioma grading by applying radiomic analysis or deep convolutional neural networks (DCNN) and to benchmark both approaches on broader validation sets.Methods.Seven public datasets were considered: (1) low-grade glioma or high-grade glioma (369 patients, BraTS'20) (2) well-differentiated liposarcoma or lipoma (115, LIPO); (3) desmoid-type fibromatosis or extremity soft-tissue sarcomas (203, Desmoid); (4) primary solid liver tumors, either malignant or benign (186, LIVER); (5) gastrointestinal stromal tumors (GISTs) or intra-abdominal gastrointestinal tumors radiologically resembling GISTs (246, GIST); (6) colorectal liver metastases (77, CRLM); and (7) lung metastases of metastatic melanoma (103, Melanoma). Radiomic analysis was performed on 464 (2016) radiomic features for the BraTS'20 (others) datasets respectively. Random forests (RF), Extreme Gradient Boosting (XGBOOST) and a voting algorithm comprising both classifiers were tested. The parameters of the classifiers were optimized using a repeated nested stratified cross-validation process. The feature importance of each classifier was computed using the Gini index or permutation feature importance. DCNN was performed on 2D axial and sagittal slices encompassing the tumor. A balanced database was created, when necessary, using smart slices selection. ResNet50, Xception, EficientNetB0, and EfficientNetB3 were transferred from the ImageNet application to the tumor classification and were fine-tuned. Five-fold stratified cross-validation was performed to evaluate the models. The classification performance of the models was measured using multiple indices including area under the receiver operating characteristic curve (AUC).Results.The best radiomic approach was based on XGBOOST for all datasets; AUC was 0.934 (BraTS'20), 0.86 (LIPO), 0.73 (LIVER), (0.844) Desmoid, 0.76 (GIST), 0.664 (CRLM), and 0.577 (Melanoma) respectively. The best DCNN was based on EfficientNetB0; AUC was 0.99 (BraTS'20), 0.982 (LIPO), 0.977 (LIVER), (0.961) Desmoid, 0.926 (GIST), 0.901 (CRLM), and 0.89 (Melanoma) respectively.Conclusion.Tumor classification can be accurately determined by adapting state-of-the-art machine learning algorithms to the medical context.

Generating synthetic images from cone beam computed tomography using self-attention residual UNet for head and neck radiotherapy.

Background and purpose: Accurate CT numbers in Cone Beam CT (CBCT) are crucial for precise dose calculations in adaptive radiotherapy (ART). This study aimed to generate synthetic CT (sCT) from CBCT using deep learning (DL) models in head and neck (HN) radiotherapy. Materials and methods: A novel DL model, the 'self-attention-residual-UNet' (ResUNet), was developed for accurate sCT generation. ResUNet incorporates a self-attention mechanism in its long skip connections to enhance information transfer between the encoder and decoder. Data from 93 HN patients, each with planning CT (pCT) and first-day CBCT images were used. Model performance was evaluated using two DL approaches (non-adversarial and adversarial training) and two model types (2D axial only vs. 2.5D axial, sagittal, and coronal). ResUNet was compared with the traditional UNet through image quality assessment (Mean Absolute Error (MAE), Peak-Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM)) and dose calculation accuracy evaluation (DVH deviation and gamma evaluation (1 %/1mm)). Results: Image similarity evaluation results for the 2.5D-ResUNet and 2.5D-UNet models were: MAE: 46±7 HU vs. 51±9 HU, PSNR: 66.6±2.0 dB vs. 65.8±1.8 dB, and SSIM: 0.81±0.04 vs. 0.79±0.05. There were no significant differences in dose calculation accuracy between DL models. Both models demonstrated DVH deviation below 0.5 % and a gamma-pass-rate (1 %/1mm) exceeding 97 %. Conclusions: ResUNet enhanced CT number accuracy and image quality of sCT and outperformed UNet in sCT generation from CBCT. This method holds promise for generating precise sCT for HN ART.

The role of radiotherapy in newly diagnosed primary CNS lymphoma: A descriptive review and a pragmatic approach to clinical practice.

Earlier, prior to the development of effective systemic therapy, monotherapy with whole-brain radiotherapy (WBRT) was widely used to treat primary central nervous system lymphoma (PCNSL). Recently, chemotherapy, especially with high dose methotrexate (HDMTX), has largely replaced WBRT as upfront treatment, and the most accepted standard of care is induction with a combination drug therapy followed by consolidation therapy with either autologous stem-cell transplantation (ASCT) or radiation. Whilst WBRT is an effective component of treatment, it is occasionally associated with risk of permanent, irreversible neurotoxicity when doses of more than 30 Gy are used. Hence, there has been a strong focus on the optimization of radiotherapy (RT) which includes dose reduction in the consolidation phase. In this comprehensive review, we have summarized the progress on clinical results and evidence considering the role and use of radiation including combined treatment modalities, low-dose radiotherapy, and neurotoxicity. Finally, we present a practical approach to low-dose WBRT and boosting higher doses to the gross tumor that can be integrated into clinical practice.

Real-world clinical outcomes with daily image-guided IMRT in extremity soft tissue sarcomas.

PURPOSE: We report the clinical outcomes of patients with soft tissue sarcomas (STS) arising in extremities treated with image-guided intensity modulated radiotherapy (IG-IMRT) at our institute. Local control of the tumors treated with RT was the primary end point of this study. Analyzing overall survival and long-term toxicities were the secondary objectives. METHODS AND MATERIALS: The database of the patients with STS who received wide local excision and IG-IMRT at our institution from January 2012 to December 2020 was reviewed. Radiation was offered either preoperatively or postoperatively as part of multi-modality treatment. RESULTS: Thirty-three consecutive patients were identified and included for analysis. Twenty-eight patients (84.8%) received postoperative adjuvant radiotherapy. Dedicated MRI simulation studies were performed in 31 patients (93.9%) in the treatment position. RapidArc IMRT technique was used in 31 patients (93.9%). A total of 2954 images were acquired during 991 treatment sessions. Errors exceeding 1 mm in the x, y and z directions were corrected online before the treatment. With a median follow-up of 36 months, two patients (6.1%) developed local recurrence. The 3-year local control was 90.9% (95% CI, 0.76 - 0.98), and the 5-year overall survival was 71.7% (95% CI, 0.44 - 0.88). One patient (3.03%) sustained a pathological fracture during the follow-up period. CONCLUSION: Our results showed that IMRT with daily imaging offered excellent local control with acceptable long-term toxicity, as well as being feasible and practical to implement in our routine clinical practice.

Predicting respiratory motion using a novel patient specific dual deep recurrent neural networks.

Real-time tracking of a target volume is a promising solution for reducing the planning margins and both dosimetric and geometric uncertainties in the treatment of thoracic and upper-abdomen cancers. Respiratory motion prediction is an integral part of real-time tracking to compensate for the latency of tracking systems. The purpose of this work was to develop a novel method for accurate respiratory motion prediction using dual deep recurrent neural networks (RNNs). The respiratory motion data of 111 patients were used to train and evaluate the method. For each patient, two models (Network1 and Network2) were trained on 80% of the respiratory wave, and the remaining 20% was used for evaluation. The first network (Network 1) is a 'coarse resolution' prediction of future points and second network (Network 2) provides a 'fine resolution' prediction to interpolate between the future predictions. The performance of the method was tested using two types of RNN algorithms : Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU). The accuracy of each model was evaluated using the root mean square error (RMSE) and mean absolute error (MAE). Overall, the RNN model with GRU- function had better accuracy than the RNN model with LSTM-function (RMSE (mm): 0.4 ± 0.2 versus 0.6 ± 0.3; MAE (mm): 0.4 ± 0.2 versus 0.6 ± 0.2). The GRU was able to predict the respiratory motion accurately (<1 mm) up to the latency period of 440 ms, and LSTM's accuracy was acceptable only up to 240 ms. The proposed method using GRU function can be used for respiratory-motion prediction up to a latency period of 440 ms.