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.

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.

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.

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.

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.

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.

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.

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.

Automatic segmentation of magnetic resonance images for high-dose-rate cervical cancer brachytherapy using deep learning.

PURPOSE: Magnetic resonance (MR) imaging is the gold standard in image-guided brachytherapy (IGBT) due to its superior soft-tissue contrast for target and organs-at-risk (OARs) delineation. Accurate and fast segmentation of MR images are very important for high-quality IGBT treatment planning. The purpose of this work is to implement and evaluate deep learning (DL) models for the automatic segmentation of targets and OARs in MR image-based high-dose-rate (HDR) brachytherapy for cervical cancer. METHODS: A 2D DL model using residual neural network architecture (ResNet50) was developed to contour the targets (gross tumor volume (GTV), high-risk clinical target volume (HR CTV), and intermediate-risk clinical target volume (IR CTV)) and OARs (bladder, rectum, sigmoid, and small intestine) automatically on axial MR slices of HDR brachytherapy patients. Furthermore, two additional 2D DL models using sagittal and coronal images were also developed. A 2.5D model was generated by combining the outputs from axial, sagittal, and coronal DL models. Similarly, a 2D and 2.5D DL models were also generated for the inception residual neural network (InceptionResNetv2 (InRN)) architecture. The geometric (Dice similarity coefficient (DSCs) and 95th percentile of Hausdorff distance (HD)) and dosimetric accuracy of 2D (axial only) and 2.5D (axial + sagittal + coronal) DL model generated contours were calculated and compared. RESULTS: The mean (range) DSCs of ResNet50 across all contours were 0.674 (0.05-0.96) and 0.715 (0.26-0.96) for the 2D and 2.5D models, respectively. For InRN, these were 0.676 (0.11-0.96) and 0.723 (0.35-0.97) for the 2D and 2.5D models, respectively. The mean HD of ResNet50 across all contours was 15.6 mm (1.8-69 mm) and 12.1 mm (1.7-44 mm) for the 2D and 2.5D models, respectively. The similar results for InRN were 15.4 mm (2-68 mm) and 10.3 mm (2.7-39 mm) for the 2D and 2.5D models, respectively. The dosimetric parameters (D90) of GTV and HR CTV for manually contoured plans matched better with the 2.5D model (p > 0.6) and the results from the 2D model were slightly lower (p < 0.08). On the other hand, the IR CTV doses (D90) for all of the models were slightly lower (2D: -1.3 to -1.5 Gy and 2.5D: -0.5 to -0.6 Gy) and the differences were statistically significant for the 2D model (2D: p < 0.000002 and 2.5D: p > 0.06). In case of OARs, the 2.5D model segmentations resulted in closer dosimetry than 2D models (2D: p = 0.07-0.91 and 2.5D: p = 0.16-1.0). CONCLUSIONS: The 2.5D DL models outperformed their respective 2D models for the automatic contouring of targets and OARs in MR image-based HDR brachytherapy for cervical cancer. The InceptionResNetv2 model performed slightly better than ResNet50.

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.

Generation of synthetic CT using multi-scale and dual-contrast patches for brain MRI-only external beam radiotherapy.

PURPOSE: To create a synthetic CT (sCT) from conventional brain MRI using a patch-based method for MRI-only radiotherapy planning and verification. METHODS: Conventional T1 and T2-weighted MRI and CT datasets from 13 patients who underwent brain radiotherapy were included in a retrospective study whereas 6 patients were tested prospectively. A new contribution to the Non-local Means Patch-Based Method (NMPBM) framework was done with the use of novel multi-scale and dual-contrast patches. Furthermore, the training dataset was improved by pre-selecting the closest database patients to the target patient for computation time/accuracy balance. sCT and derived DRRs were assessed visually and quantitatively. VMAT planning was performed on CT and sCT for hypothetical PTVs in homogeneous and heterogeneous regions. Dosimetric analysis was done by comparing Dose Volume Histogram (DVH) parameters of PTVs and organs at risk (OARs). Positional accuracy of MRI-only image-guided radiation therapy based on CBCT or kV images was evaluated. RESULTS: The retrospective (respectively prospective) evaluation of the proposed Multi-scale and Dual-contrast Patch-Based Method (MDPBM) gave a mean absolute error MAE=99.69±11.07HU (98.95±8.35HU), and a Dice in bones DIbone=83±0.03 (0.82±0.03). Good agreement with conventional planning techniques was obtained; the highest percentage of DVH metric deviations was 0.43% (0.53%) for PTVs and 0.59% (0.75%) for OARs. The accuracy of sCT/CBCT or DRRsCT/kV images registration parameters was <2mm and <2°. Improvements with MDPBM, compared to NMPBM, were significant. CONCLUSION: We presented a novel method for sCT generation from T1 and T2-weighted MRI potentially suitable for MRI-only external beam radiotherapy in brain sites.

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.