Cone-beam Computed Tomography Quality Research Articles

Joint denoising and interpolating network for low-dose cone-beam CT reconstruction under hybrid dose-reduction strategy

Cone-beam computed tomography (CBCT) is generally reconstructed with hundreds of two-dimensional X-Ray projections through the FDK algorithm, and its excessive ionizing radiation of X-Ray may impair patients’ health. Two common dose-reduction strategies are to either lower the intensity of X-Ray, i.e., low-intensity CBCT, or reduce the number of projections, i.e., sparse-view CBCT. Existing efforts improve the low-dose CBCT images only under a single dose-reduction strategy. In this paper, we argue that applying the two strategies simultaneously can reduce dose in a gentle manner and avoid the extreme degradation of the projection data in a single dose-reduction strategy, especially under ultra-low-dose situations. Therefore, we develop a Joint Denoising and Interpolating Network (JDINet) in projection domain to improve the CBCT quality with the hybrid low-intensity and sparse-view projections. Specifically, JDINet mainly includes two important components, i.e., denoising module and interpolating module, to respectively suppress the noise caused by the low-intensity strategy and interpolate the missing projections caused by the sparse-view strategy. Because FDK actually utilizes the projection information after ramp-filtering, we develop a filtered structural similarity constraint to help JDINet focus on the reconstruction-required information. Afterward, we employ a Postprocessing Network (PostNet) in the reconstruction domain to refine the CBCT images that are reconstructed with denoised and interpolated projections. In general, a complete CBCT reconstruction framework is built with JDINet, FDK, and PostNet. Experiments demonstrate that our framework decreases RMSE by approximately 8 %, 15 %, and 17 %, respectively, on the 1/8, 1/16, and 1/32 dose data, compared to the latest methods. In conclusion, our learning-based framework can be deeply imbedded into the CBCT systems to promote the development of CBCT. Source code is available at https://github.com/LianyingChao/FusionLowDoseCBCT.

Proton Online Adaptation Using Novel Cone-Beam Computed Tomography System

Cone beam computed tomography (CBCT) has been used in clinic frequently to provide quick, online, and handy three-dimensional images. However, due to its significant artifacts and inaccurate Hounsfield values (HU), CBCT based online proton adaptation is infeasible. Recently, a novel in-room imaging solution was developed to allow larger images, better contrast and faster CBCT imaging acquisition. Our work demonstrated for the first time the feasibility of using this novel CBCT images for direct intensity modulated proton planning. Three patients and three CIRS phantoms were scanned using novel imaging technique (CBCTp) for this work. CT curves were acquired by scanning a CIRS electron density phantom. Stopping power ratio was computed using the stoichiometric method. Proton plans were made on treatment simulation CT and further evaluated on CBCTp to compare the differences. The table below shows three patients and three phantoms plan comparisons. Multiple sites and dose levels were studied. The planning target volume (PTV) coverage (D95%) and mean dose difference between simulation CT and CBCTp are 0.8% and 0.3%, correspondingly. Dosimetrically, phantom and patient plans are almost identical between two imaging techniques. Lung patient plan show the largest variation due to patient tumor change and the quality of breathing. The novel and high quality of CBCT is feasible for direct proton planning with accurate CT calibration. It provides a reliable alternative solution for proton online adaptive therapy.

More Efficient Auto-Segmentation Framework Using Patient-Specific Information for CBCT-Guided Adaptive Radiotherapy

Accurate and fast delineation of regions of interest (ROIs) is critical for online adaptive radiotherapy (ART). Due to the noise, artifacts and low contrast of soft tissue on cone-beam computed tomography (CBCT), the CBCT-guided ART (CBCTgART) needs an effective tool to assist the segmentation of ROIs. This study aims to develop an efficient auto-segmentation workflow with personalized modeling for CBCTgART based on deep learning (DL) model. Five hundred fifty-two patients with nasopharyngeal carcinoma were enrolled in this study. At the beginning, a cycle consistent adversarial network (CycleGAN) model was trained on 172 patients' CBCT/CT paired images to improve the image quality of CBCT to CT level. A generalized segmentation model was trained using the planning CT (pCT) and contour data from 530 patients. For personalized modeling, the generalized segmentation model was fine-tuned on the specific patient's pCT and contour to get the personalized model. When CBCT images were available, the trained CycleGAN model transformed the CBCT to synthetic pCT. Then the personalized auto-segmentation model generated the contour of ROIs on the synthetic pCT. We randomly selected 22 patients for model test. The proposed method (DL-personalized) was compared with other DL methods based on the same architecture of network: regular deep learning method (DL-regular), which was trained on the CBCT and corresponding contours, and generalized model in our framework (DL-generalized). So, 22 personalized, 1 generalized and 1 regular DL models were tested. The paired t-test was performed to test the significance for mean dice similarity coefficient (DSC), mean distance to agreement (MDA), and Hausdorff distance (HD) between the alternative and proposed methods. Two ROIs were included: the clinical target volume (CTV) and nasopharynx gross tumor volume (GTVnx). The proposed DL-personalized model achieved better results compared with others as shown in the table. The accuracy and robustness of our proposed framework was reliable. All of p values were under 0.01, which indicated the statistically significant difference. The proposed framework utilizing patient-specific information can improve the segmentation accuracy of ART. This method has potential to be integrated into the ART workflow to improve efficiency.

Accuracy of Electron Density and Planning Dosimetry in a Novel High-Quality CBCT Imaging System

A high-quality Cone-Beam Computed Tomography (CBCT) imaging system has been FDA approved for imaging guidance and dose calculation in radiotherapy. This study aims to evaluate the accuracy of the relative electron density in CBCT images acquired in this CBCT imaging system in a phantom study and its dosimetric impact on treatment planning in a patient study. Astoichiometric CT calibration was performed with a CIRS phantom (SunNuclear, Model 062M) to generate the HU-electron density curve for two tube voltages, 125kVp and 140 kVp, respectively. The phantom has a longitudinal length of 26.5 cm and is equipped with interchangeable inserts of various compositions, supplied by the vendor. Measurements were taken with solid water plates added to both ends of the phantom to allow adequate scattering and repeated for various clinical protocols with different combinations of tube voltages and exposures. The accuracy of the relative electron density of the CBCT imaging system was verified by comparing the calculated electron density from the Hounsfield Units (HU) measurements obtained from a Gammex phantom to the relative electron densities provided in vendor's specifications. To benchmark the relative electron density of the CBCT imaging system against a standard helical CT simulator, ten clinical plans that were created on CT simulation images were copied and recalculated on the CBCT images acquired immediately after the CT simulation, the latter of which was a standard procedure in current radiotherapy care for all patients who had given their consent to participate in the IRB-approved imaging study. The dose grids used in these calculations were 2.5mm x 2.5mm x 3mm. The Gamma passing rate was calculated using a standard 3mm/3% criterion with a 10% threshold. Ourresults showed the difference between the averaged CBCT calibration curves acquired at tube voltages of 125 kVP and 140 kVp was less than 2%. The mean discrepancy of the relative electron densities from vendor's specification was 0.0045 with a range between -0.02 and 0.04. Relative electron densities in all inserts were within 2% from the vendor's specifications except the cortical bone insert. Gamma passing rate was between 96.02% and 98.49% with mean value of 97.4% and a standard deviation of 0.95%. We consider this reflects the fact that the CT simulation and CBCT imaging were performed in separated rooms, which resulted in slight anatomical deformation that could negatively impact the Gamma passing rate. The CBCT imaging system provides sufficient accuracy of electron density for dose calculation, and the dose distribution calculated on the CBCT images is clinically equivalent to those calculated on helical CT images. The enhanced imaging quality of CBCT could further extend the role of imaging guidance to planning for adaptive radiotherapy, potentially reducing the need for re-simulation and interruptions in the radiotherapy course.

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

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. 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 based on a reference digitally reconstructed radiograph (DRR). 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 nonlocal means filter 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. 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 qualityshowed 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. The proposed method can considerably improve the CBCT image quality with little anatomical distortion, improving the accuracy of radiotherapy for head and neck patients. This article is protected by copyright. All rights reserved.

Image-based scatter correction for cone-beam CT using flip swin transformer U-shape network.

Cone beam computed tomography (CBCT) plays an increasingly important role in image-guided radiation therapy. However, the image quality of CBCT is severely degraded by excessive scatter contamination, especially in the abdominal region, hindering its further applications in radiation therapy. To restore low-quality CBCT images contaminated by scatter signals, a scatter correction algorithm combining the advantages of convolutional neural networks (CNN) and Swin Transformer is proposed. In this paper a scatter correction model for CBCT image, the Flip Swin Transformer U-shape network (FSTUNet) model, is proposed. In this model, the advantages of CNN in texture detail and Swin Transformer in global correlation are used to accurately extract shallow and deep features, respectively. Instead of using the original Swin Transformer tandem structure, we build the Flip Swin Transformer Block to achieve a more powerful inter-window association extraction. The validity and clinical relevance of the method is demonstrated through extensive experiments on a Monte Carlo (MC) simulation dataset and frequency split dataset generated by a validated method, respectively. Experimental results on the MC simulated dataset show that the root mean square error of images corrected by the method is reduced from over 100 HU to about 7 HU. Both the structural similarity index measure (SSIM) and the universal quality index (UQI) are close to 1. Experimental results on the frequency split dataset demonstrate that the method not only corrects shading artifacts but also exhibits a high degree of structural consistency. In addition, comparison experiments show that FSTUNet outperforms UNet, Deep Residual Convolutional Neural Network (DRCNN), DSENet, Pix2pixGAN, and 3DUnet methods in both qualitative and quantitative metrics. Accurately capturing the features at different levels is greatly beneficial for reconstructing high-quality scatter-free images. The proposed FSTUNet method is an effective solution to CBCT scatter correction and has the potential to improve the accuracy of CBCT image-guided radiation therapy.

New technique and application of truncated CBCT processing in adaptive radiotherapy for breast cancer

ObjectiveA generative adversarial network (TCBCTNet) was proposed to generate synthetic computed tomography (sCT) from truncated low-dose cone-beam computed tomography (CBCT) and planning CT (pCT). The sCT was applied to the dose calculation of radiotherapy for patients with breast cancer. MethodsThe low-dose CBCT and pCT images of 80 female thoracic patients were used for training. The CBCT, pCT, and replanning CT (rCT) images of 20 thoracic patients and 20 patients with breast cancer were used for testing. All patients were fixed in the same posture with a vacuum pad. The CBCT images were scanned under the Fast Chest M20 protocol with a 50% reduction in projection frames compared with the standard Chest M20 protocol. Rigid registration was performed between pCT and CBCT, and deformation registration was performed between rCT and CBCT. In the training stage of the TCBCTNet, truncated CBCT images obtained from complete CBCT images by simulation were used. The input of the CBCT→CT generator was truncated CBCT and pCT, and TCBCTNet was applied to patients with breast cancer after training. The accuracy of the sCT was evaluated by anatomy and dosimetry and compared with the generative adversarial network with UNet and ResNet as the generators (named as UnetGAN, ResGAN). ResultsThe three models could improve the image quality of CBCT and reduce the scattering artifacts while preserving the anatomical geometry of CBCT. For the chest test set, TCBCTNet achieved the best mean absolute error (MAE, 21.18±3.76 HU), better than 23.06±3.90 HU in UnetGAN and 22.47±3.57 HU in ResGAN. When applied to patients with breast cancer, TCBCTNet performance decreased, and MAE was 25.34±6.09 HU. Compared with rCT, sCT by TCBCTNet showed consistent dose distribution and subtle absolute dose differences between the target and the organ at risk. The 3D gamma pass rates were 98.98%±0.64% and 99.69%±0.22% at 2 mm/2% and 3 mm/3%, respectively. Ablation experiments confirmed that pCT and content loss played important roles in TCBCTNet. ConclusionsHigh-quality sCT images could be synthesized from truncated low-dose CBCT and pCT by using the proposed TCBCTNet model. In addition, sCT could be used to accurately calculate the dose distribution for patients with breast cancer.

Comparison of Diagnostic Quality of Multidetector Computed Tomography and Cone-Beam Computed Tomography in High and Low Resolution Modes for Assessment of the Nasal Cavity.

We aimed to compare the diagnostic quality of multidetector computed tomography (MDCT) and cone-beam computed tomography (CBCT) in high (HR) and low (LR) resolution modes for assessment of the nasal cavity and paranasal sinuses. This in vitro study was conducted on 5 dry human skulls by using a CBCT and a MDCT scanner in HR and LR modes to assess their diagnostic quality for 21 anatomical landmarks of the nose. The quality of images was evaluated by two oral and maxillofacial radiologists and a dentist using a four-point Likert scale of (I) poor, (II) decreased, (III) good, and (IV) excellent. Data were analyzed by STATA at 95% confidence interval. The Chi-square test was applied to compare the quality of visualization of landmarks based on the type of scanner. The diagnostic quality of HR CBCT and CT for the majority of landmarks was higher than that of LR CBCT and CT (P<0.05). The diagnostic quality of HR CBCT for agger nasi cells (P=0.010), olfactory cleft (P=0.032), sphenoethmoidal recess (P=0.032), and nasolacrimal duct (P=0.014) and LR CBCT for the middle turbinate (P=0.046) and middle meatus (P=0.031) was significantly higher than that of MDCT. The diagnostic quality of HR CBCT and CT for the majority of the landmarks in the nasal cavity and paranasal sinuses was higher than that of LR CBCT and CT. For the majority of landmarks, the diagnostic quality of CBCT and CT was the same; while for some landmarks, the diagnostic quality of HR and LR CBCT was higher than HR and LR CT. In general, CBCT has high efficacy for evaluation of the paranasal sinuses and the nasal cavity, and provides diagnostic information comparable to those provided by CT, but with a much lower radiation dose.

A two-step method to improve image quality of CBCT with phantom-based supervised and patient-based unsupervised learning strategies

Objective. In this study, we aimed to develop deep learning framework to improve cone-beam computed tomography (CBCT) image quality for adaptive radiation therapy (ART) applications. Approach. Paired CBCT and planning CT images of 2 pelvic phantoms and 91 patients (15 patients for testing) diagnosed with prostate cancer were included in this study. First, well-matched images of rigid phantoms were used to train a U-net, which is the supervised learning strategy to reduce serious artifacts. Second, the phantom-trained U-net generated intermediate CT images from the patient CBCT images. Finally, a cycle-consistent generative adversarial network (CycleGAN) was trained with intermediate CT images and deformed planning CT images, which is the unsupervised learning strategy to learn the style of the patient images for further improvement. When testing or applying the trained model on patient CBCT images, the intermediate CT images were generated from the original CBCT image by U-net, and then the synthetic CT images were generated by the generator of CycleGAN with intermediate CT images as input. The performance was compared with conventional methods (U-net/CycleGAN alone trained with patient images) on the test set. Results. The proposed two-step method effectively improved the CBCT image quality to the level of CT scans. It outperformed conventional methods for region-of-interest contouring and HU calibration, which are important to ART applications. Compared with the U-net alone, it maintained the structure of CBCT. Compared with CycleGAN alone, our method improved the accuracy of CT number and effectively reduced the artifacts, making it more helpful for identifying the clinical target volume. Significance. This novel two-step method improves CBCT image quality by combining phantom-based supervised and patient-based unsupervised learning strategies. It has immense potential to be integrated into the ART workflow to improve radiotherapy accuracy.

Diagnostic usefulness of cone-beam computed tomography versus multi-detector computed tomography for sinonasal structure evaluation.

ObjectiveCone‐beam computed tomography (CBCT) is a promising imaging modality for sinonasal evaluation, with advantages of relatively low radiation dose, low cost, and quick outpatient imaging. Our study aimed to compare the diagnostic performance and image quality of CBCT with those of multi‐detector computed tomography (MDCT) with different slice thickness.MethodsWe retrospectively reviewed 60 consecutive patients who had undergone both CBCT and MDCT. MDCT images was reconstructed with 1 and 3 mm slice thickness. The quantitative image quality parameters (image noise, signal‐to‐noise ratio [SNR], and contrast‐to noise ratio [CNR] were calculated and compared between the two imaging modalities. Two observers (ENT surgeon and neuroradiologist) evaluated the presence of seven sinonasal anatomic variations in each patient and interobserver agreements were analyzed. The diagnostic performance of CBCT (0.3 mm) and MDCT (3 mm) was assessed and compared with that of high resolution MDCT (1 mm), which is considered as the gold standard.ResultsThe image noise was significantly higher and SNR and CNR values were lower in the CBCT (0.3 mm) group than in the MDCT groups (1 and 3 mm). The diagnostic performance of CBCT (0.3 mm) was similar to that of MDCT (1 mm) and superior to that of MDCT (3 mm). The highest interobserver agreement was for high resolution MDCT (1 mm), followed by CBCT (0.3 mm), and MDCT (3 mm).ConclusionConsidering its low radiation dose, low cost, and ease of clinical access, CBCT may be a useful imaging modality for as first line sinonasal evaluation and repeated follow up.Study design: Retrospective study in a tertiary referral university center.Level of evidence: NA.

Ensemble learning and tensor regularization for cone-beam computed tomography-based pelvic organ segmentation.

Cone-beam computed tomography (CBCT) is a widely accessible low-dose imaging approach compatible with on-table patient anatomy observation for radiotherapy. However, its use in comprehensive anatomy monitoring is hindered by low contrast and low signal-to-noise ratio and a large presence of artifacts, resulting in difficulty in identifying organ and structure boundaries either manually or automatically. In this study, we propose and develop an ensemble deep-learning model to segment post-prostatectomy organsautomatically. We utilize the ensemble logic in various modules during the segmentation process to alleviate the impact of low image quality of CBCT. Specifically, (1) semantic attention was obtained from an ensemble 2.5D You-only-look-once detector to consistently define regions of interest, (2) multiple view-specific two-stream 2.5D segmentation networks were developed, using auxiliary high-quality CT data to aid CBCT segmentation, and (3) a novel tensor-regularized ensemble scheme was proposed to aggregate the estimates from multiple views and regularize the spatial integrity of the finalsegmentation. A cross-validation study achieved Dice similarity coefficient and mean surface distance of 0.779 0.069 and 2.895 1.496 mm for the rectum, and 0.915 0.055 and 1.675 1.311 mm for the bladder. The proposed ensemble scheme manages to enhance the geometric integrity and robustness of the contours derived from CBCT with light network components. The tensor regularization approach generates organ results conforming to anatomy and physiology, without compromising typical quantitative performance in Dice similarity coefficient and mean surface distance, to support further clinical interpretation and decisionmaking.

Assessing the reproducibility of CBCT-derived radiomics features using a novel three-dimensional printed phantom.

Radiomics modeling is an exciting avenue for enhancing clinical decision making and personalized treatment. Radiation oncology patients often undergo routine imaging for position verification, particularly using LINAC-mounted cone beam computed tomography (CBCT). The wealth of imaging data collected in modern radiation therapy presents an ideal use case for radiomics modeling. Despite this, texture feature (TF) calculation can be limited by concerns over feature stability and reproducibility; in theory, this issue is compounded by the relatively poor image quality of CBCT, as well as variation of acquisition and reconstruction parameters. In this study, we developed and validated a novel three-dimensional (3D) printed phantom for evaluating CBCT-based TF reliability. The phantom has a cylindrical shape (22cm diameter and 25.5cm height) with five inner inserts designed to hold custom-printed rods (1cm diameter and 10-20cm height) of various materials, infill shapes, and densities. TF reproducibility was evaluated across and within three LINACs from a single vendor using sets of three consecutive CBCT taken with the head, thorax, and pelvis clinical imaging protocols. PyRadiomics was used to extract a standard set of TFs from regions of interest centered on each rod. Two-way mixed effects absolute agreement intra-class correlation coefficient (ICC) was used to evaluate TF reproducibility, with features showing ICC values above 0.9 considered robust if their Bonferroni-corrected p-value was below 0.05. A total of 63, 87, and 83 features exhibited test-retest reliability for the head, thorax, and pelvis imaging protocols respectively. When assessing stability between discreet imaging sessions on the same LINAC, these numbers were reduced to 5, 63, and 70 features, respectively. The thorax and pelvis protocols maintained a rich candidate feature space in inter-LINAC analysis with 61 and 65 features, respectively, exceeding the ICC criteria. Crucially, no features were deemed reproducible when compared between protocols. We have developed a 3D phantom for consistent evaluation of TF stability and reproducibility. For LINACs from a single vendor, our study found a substantial number of features available for robust radiomics modeling from CBCT imaging. However, some features showed variations across LINACs. Studies involving CBCT-based radiomics must preselect features prior to their use in clinical-based models.

Improving CBCT quality to CT level using deep learning with generative adversarial network.

To improve image quality and computed tomography (CT) number accuracy of daily cone beam CT (CBCT) through a deep learning methodology with generative adversarial network. One hundred and fifty paired pelvic CT and CBCT scans were used for model training and validation. An unsupervised deep learning method, 2.5D pixel-to-pixel generative adversarial network (GAN) model with feature mapping was proposed. A total of 12000 slice pairs of CT and CBCT were used for model training, while ten-fold cross validation was applied to verify model robustness. Paired CT-CBCT scans from an additional 15 pelvic patients and 10 head-and-neck (HN) patients with CBCT images collected at a different machine were used for independent testing purpose. Besides the proposed method above, other network architectures were also tested as: 2D vs 2.5D; GAN model with vs without feature mapping; GAN model with vs without additional perceptual loss; and previously reported models as U-net and cycleGAN with or without identity loss. Image quality of deep-learning generated synthetic CT (sCT) images was quantitatively compared against the reference CT (rCT) image using mean absolute error (MAE) of Hounsfield units (HU) and peak signal-to-noise ratio (PSNR). The dosimetric calculation accuracy was further evaluated with both photon and proton beams. The deep-learning generated sCTs showed improved image quality with reduced artifact distortion and improved soft tissue contrast. The proposed algorithm of 2.5 Pix2pix GAN with feature matching (FM) was shown to be the best model among all tested methods producing the highest PSNR and the lowest MAE to rCT. The dose distribution demonstrated a high accuracy in the scope of photon-based planning, yet more work is needed for proton-based treatment. Once the model was trained, it took 11-12ms to process one slice, and could generate a 3D volume of dCBCT (80 slices) in less than a second using a NVIDIA GeForce GTX Titan X GPU (12GB, Maxwell architecture). The proposed deep learning algorithm is promising to improve CBCT image quality in an efficient way, thus has a potential to support online CBCT-based adaptive radiotherapy.

Cone-beam CT image quality improvement using Cycle-Deblur consistent adversarial networks (Cycle-Deblur GAN) for chest CT imaging in breast cancer patients

Cone-beam computed tomography (CBCT) integrated with a linear accelerator is widely used to increase the accuracy of radiotherapy and plays an important role in image-guided radiotherapy (IGRT). For comparison with fan-beam computed tomography (FBCT), the image quality of CBCT is indistinct due to X-ray scattering, noise, and artefacts. We proposed a deep learning model, “Cycle-Deblur GAN”, combined with CycleGAN and Deblur-GAN models to improve the image quality of chest CBCT images. The 8706 CBCT and FBCT image pairs were used for training, and 1150 image pairs were used for testing in deep learning. The generated CBCT images from the Cycle-Deblur GAN model demonstrated closer CT values to FBCT in the lung, breast, mediastinum, and sternum compared to the CycleGAN and RED-CNN models. The quantitative evaluations of MAE, PSNR, and SSIM for CBCT generated from the Cycle-Deblur GAN model demonstrated better results than the CycleGAN and RED-CNN models. The Cycle-Deblur GAN model improved image quality and CT-value accuracy and preserved structural details for chest CBCT images.

Feasibility of image quality improvement for high-speed CBCT imaging using deep convolutional neural network for image-guided radiotherapy in prostate cancer.

High-speed cone-beam computed tomography (CBCT) scan for image-guided radiotherapy (IGRT) can reduce both the scan time and the exposure dose. However, it causes noise and artifacts in the reconstructed images due to the lower number of acquired projection data. The purpose of this study is to improve the image quality of high-speed CBCT using a deep convolutional neural network (DCNN). CBCT images of 36 prostate cancer patients were selected. The CBCT images acquired at normal scan speed were defined as CBCT100%. Simulated high-speed CBCT images acquired at twofold and fourfold scan speed were created, which were defined as CBCT50% and CBCT25%, respectively. The image quality of the CBCT50% was treated as the requirement for IGRT in this study because previous studies reported that its image is sufficient with respect to IGRT. The DCNN model was trained to learn direct mapping from CBCT25% to the corresponding CBCT100%. The performance of the DCNN model was evaluated using the sixfold cross-validation method. CBCT images generated by DCNN (CBCT25%+DCNN) were evaluated for voxel value accuracy and image quality. The DCNN model can process CBCT25% of a new patient within 0.06s/slice. The CBCT25%+DCNN was comparable to the CBCT50% in terms of both voxel value accuracy and image quality. We developed a DCNN model to remove noise and artifacts from high-speed CBCT. We emphasize that it is possible to reduce exposure to one quarter and to increase the CBCT scan speed by a factor of four.

Cone-Beam CT Compared to Multi-Slice CT for the Diagnostic Analysis of Conductive Hearing Loss: A Feasibility Study

Multislice computed tomography (MSCT) is commonly used as a diagnostic tool for patients with a conductive hearing loss. Recent studies indicate that cone-beam computed tomography (CBCT) may be used as a low-radiation dose alternative for temporal bone imaging. This study compares image quality and radiation dose between CBCT and MSCT when assessing anatomical landmarks related to conductive hearing loss. Five human cadaver heads (10 ears) were imaged on the NewTom 5G CBCT and the Discovery CT750 HD MSCT. Visibility of 16 anatomical landmarks of the middle and inner ear was assessed by two observers on a 4-point Likert scale. Furthermore, effective radiation dose was compared, and contrast-to-noise ratio and spatial resolution were measured with a phantom head. Image quality of CBCT was assessed as superior to MSCT. Effective radiation dose of the high-resolution CBCT protocol was 30.5% of the clinical MSCT dose. High-resolution CBCT was reported as having a higher spatial resolution and superior contrast-to-noise perception in comparison with MSCT. High-resolution CBCT was evaluated as superior to MSCT in the assessment of structures related to conductive hearing loss. Furthermore, CBCT imaging resulted in a considerably lower effective radiation dose.

Cone-beam CT-derived relative stopping power map generation via deep learning for proton radiotherapy.

In intensity-modulated proton therapy (IMPT), protons are used to deliver highly conformal dose distributions, targeting tumors, and sparing organs-at-risk. However, due to uncertainties in both patient setup and relative stopping power (RSP) calculation, margins are added to the treatment volume during treatment planning, leading to higher doses to normal tissues. Cone-beam computed tomography (CBCT) images are taken daily before treatment; however, the poor image quality of CBCT limits the use of these images for online dose calculation. In this work, we use a deep-learning-based method to predict RSP maps from daily CBCT images, allowing for online dose calculation in a step toward adaptive radiation therapy. Twenty-three head-and-neck cancer patients were simulated using a Siemens TwinBeam dual-energy CT (DECT) scanner. Mixed-energy scans (equivalent to a 120kVp single-energy CT scan) were converted to RSP maps for treatment planning. Cone-beam computed tomography images were taken on the first day of treatment, and the planning RSP maps were registered to these images. A deep learning network based on a cycle-GAN architecture, relying on a compound loss function designed for structural and contrast preservation, was then trained to create an RSP map from a CBCT image. Leave-one-out and holdout cross validations were used for evaluation, and mean absolute error (MAE), mean error (ME), peak signal-to-noise ratio (PSNR), and structural similarity (SSIM) were used to quantify the differences between the CT-based and CBCT-based RSP maps. The proposed method was compared to a deformable image registration-based method which was taken as the ground truth and two other deep learning methods. For one patient who underwent resimulation, the new planning RSP maps and CBCT images were used for further evaluation and validation. The CBCT-based RSP generation method was evaluated on 23 head-and-neck cancer patients. From leave-one-out testing, the MAE between CT-based and CBCT-based RSP was 0.06±0.01 and the ME was -0.01±0.01. The proposed method statistically outperformed the comparison DL methods in terms of MAE and ME when compared to the planning CT. In terms of dose comparison, the mean gamma passing rate at 3%/3mm was 94% when three-dimensional (3D) gamma index was calculated per plan and 96% when gamma index was calculated per field. The proposed method provides sufficiently accurate RSP map generation from CBCT images, allowing for evaluation of daily dose based on CBCT and possibly allowing for CBCT-guided adaptive treatment planning for IMPT.

Visual enhancement of Cone-beam CT by use of CycleGAN.

Cone-beam computed tomography (CBCT) offers advantages over conventional fan-beam CT in that it requires a shorter time and less exposure to obtain images. However, CBCT images suffer from low soft-tissue contrast, noise, and artifacts compared to conventional fan-beam CT images. Therefore, it is essential to improve the image quality of CBCT. In this paper, we propose a synthetic approach to translate CBCT images with deep neural networks. Our method requires only unpaired and unaligned CBCT images and planning fan-beam CT (PlanCT) images for training. The CBCT images and PlanCT images may be obtained from other patients as long as they are acquired with the same scanner settings. Once trained, three-dimensionally reconstructed CBCT images can be directly translated into high-quality PlanCT-like images. We demonstrate the effectiveness of our method with images obtained from 20 prostate patients, and provide a statistical and visual comparison. The image quality of the translated images shows substantial improvement in voxel values, spatial uniformity, and artifact suppression compared to those of the original CBCT. The anatomical structures of the original CBCT images were also well preserved in the translated images. Our method produces visually PlanCT-like images from CBCT images while preserving anatomical structures.

Improvement in Image Quality of CBCT during Treatment by Cycle Generative Adversarial Network

Volumetric modulated arc therapy (VMAT) can acquire projection images during rotational irradiation, and cone-beam computed tomography (CBCT) images during VMAT delivery can be reconstructed. The poor quality of CBCT images prevents accurate recognition of organ position during the treatment. The purpose of this study was to improve the image quality of CBCT during the treatment by cycle generative adversarial network (CycleGAN). Twenty patients with clinically localized prostate cancer were treated with VMAT, and projection images for intra-treatment CBCT (iCBCT) were acquired. Synthesis of PCT (SynPCT) with improved image quality by CycleGAN requires only unpaired and unaligned iCBCT and planning CT (PCT) images for training. We performed visual and quantitative evaluation to compare iCBCT, SynPCT and PCT deformable image registration (DIR) to confirm the clinical usefulness. We demonstrated suitable CycleGAN networks and hyperparameters for SynPCT. The image quality of SynPCT improved visually and quantitatively while preserving anatomical structures of the original iCBCT. The undesirable deformation of PCT was reduced when SynPCT was used as its reference instead of iCBCT. We have performed image synthesis with preservation of organ position by CycleGAN for iCBCT and confirmed the clinical usefulness.

Cone-beam CT dose and imaging performance evaluation with a modular, multipurpose phantom.

A modular phantom for dosimetry and imaging performance evaluation in cone-beam computed tomography (CBCT) is reported, providing a tool for quantitative technical assessment that can be adapted to a broad variety of CBCT imaging configurations and clinical applications. The phantom presents a set of modules that can be ordered in various configurations suitable to a particular CBCT system. Modules include slabs containing a uniform medium, low-contrast inserts, line-spread features, and disk features suitable to measurement of image uniformity, noise, noise-power spectrum (NPS), contrast, contrast-to-noise ratio (CNR), Hounsfield (HU) accuracy, linearity, spatial resolution modulation transfer function (MTF), and magnitude of cone-beam artifact. Automated software recognizes the phantom configuration in DICOM images and provides structured reporting of such test measures. In any modular configuration, the phantom permits measurement of air kerma in central and peripheral locations with an air ionization chamber (e.g., Farmer chamber). The utility and adaptability of the phantom were demonstrated across a spectrum of CBCT systems, including scanners for orthopaedic imaging (Carestream OnSight 3D, Rochester NY), breast imaging (Doheny prototype, UC Davis), image-guided surgery (IGS, Medtronic O-arm, Littleton MA), angiography (Siemens Artis Zeego, Forcheim Germany), and image-guided radiation therapy (IGRT, Elekta Synergy XVI, Stockholm Sweden). The phantom provided a consistent platform for quantitative assessment of dose and imaging performance compatible with a broad spectrum of CBCT systems. The purpose of the survey was not to obtain head-to-head performance comparison of systems designed for such distinct clinical applications. Rather, the survey demonstrated the suitability of the phantom to a broad spectrum of systems in a manner that provides characterization pertinent to disparate applications and imaging tasks. For example: the orthopaedic CBCT system (pertinent clinical tasks relating to high-resolution bone imaging) was shown to achieve MTF consistent with imaging of high-contrast trabecular bone structures (i.e., the MTF reduced to 10% at spatial frequency, =1.2mm-1 ); the breast system (even higher-resolution imaging of microcalcifications) exhibited =2.2mm-1 ; the IGS system (tasks including both bone and soft-tissue contrast resolution) provided =0.9mm-1 and soft-tissue =1.64; the angiography system (soft-tissue body interventions) demonstrated =1.2 in soft tissues approximating liver lesions; and the IGRT system (pertinent tasks emphasizing HU linearity and image uniformity) showed linear response with HU values ( =1), with a cupping artifact ( =5.8%) due to x-ray scatter. The phantom provides an adaptable, quantitative basis for CBCT dosimetry and imaging performance evaluation suitable to a broad variety of CBCT systems. The dosimetry and image quality metrics are consistent with up-to-date methods for rigorous, quantitative, physics testing and should be suitable to emerging standards for CBCT quality assurance.