Enhancement of Hounsfield unit distribution in cone-beam CT images for adaptive radiation therapy: Evaluation of a hybrid correction approach

Comparative analysis of delivered and planned doses in target volumes for lung stereotactic ablative radiotherapy

BACKGROUND

Adaptive therapy has been enormously improved based on the art of generating adaptive computed tomography (ACT) from planning CT (PCT) and the on-board image used for the patient setup. Exploiting the ACT, this study evaluated the dose delivered to patients with non-small-cell lung cancer (NSCLC) patients treated with stereotactic ablative radiotherapy (SABR) and derived relationship between the delivered dose and the parameters obtained through the evaluation procedure.

METHODS

SABR treatment records of 72 patients with NSCLC who were prescribed a dose of 60 Gy (Dprescribed) to the 95% volume of the planning target volume (PTV) in four fractions were analysed in this retrospective study; 288 ACTs were generated by rigid and deformable registration of a PCT to a cone-beam computed tomography (CBCT) per fraction. Each ACT was sent to the treatment planning system (TPS) and treated as an individual PCT to calculate the dose. Delivered dose to a patient was estimated by averaging four doses calculated from four ACTs per treatment. Through the process, each ACT provided the geometric parameters, such as mean displacement of the deformed PTV voxels (Warpmean) and Dice similarity coefficient (DSC) from deformation vector field, and dosimetric parameters, e.g. difference of homogeneity index (ΔHI, HI defined as (D2%-D98%)/Dprescribed*100) and mean delivered dose to the PTV (Dmean), obtained from the dose statistics in the TPS. Those parameters were analyzed using multiple linear regression and one-way-ANOVA of SPSS® (version 27).

RESULTS

The prescribed dose was confirmed to be fully delivered to internal target volume (ITV) within maximum difference of 1%, and the difference between the planned and delivered doses to the PTV was agreed within 6% for more than 95% of the ACT cases. Volume changes of the ITV during the treatment course were observed to be minor in comparison of their standard deviations. Multiple linear regression analysis between the obtained parameters and the dose delivered to 95% volume of the PTV (D95%) revealed four PTV parameters [Warpmean, DSC, ΔHI between the PCT and ACT, Dmean] and the PTV D95% to be significantly related with P-values < 0.05. The ACT cases of high ΔHI were caused by higher values of the Warpmean and DSC from the deformable image registration, resulting in lower PTV D95% delivered. The mean values of PTV D95% and Warpmean showed significant differences depending on the lung lobe where the tumour was located.

CONCLUSIONS

Evaluation of the dose delivered to patients with NSCLC treated with SABR using ACTs confirmed that the prescribed dose was accurately delivered to the ITV. However, for the PTV, certain ACT cases characterised by high HI deviations from the original plan demonstrated variations in the delivered dose. These variations may potentially arise from factors such as patient setup during treatment, as suggested by the statistical analyses of the parameters obtained from the dose evaluation process.

A conservative approach to localize loose implant screw through cemented crown: an in vitro experimental study

BACKGROUND

Retrieval of cement-retained implant-supported restorations is intriguing in cases of screw loosening. Detecting the estimated size of the screw access hole (SAH) could decrease destruction to the prosthesis and preserve the crown.

OBJECTIVES

To precisely localize loose implant screws through cemented crowns to reduce crown damage after screw loosening.

MATERIALS AND METHODS

In this in vitro study, 60 cement-retained implants supported 30 zirconia-based, and 30 ceramics fused to metal (CFM) lower molar crowns were invented, and each was subdivided into three subgroups (10 each). In group I (AI/BI) (control), SAH was created with the aid of orthopantomography (OPG). In contrast, in group II (zirconia-crown), SAH was created with the aid of CBCT + 3D printed surgical guide with a 2 mm metal sleeve in subgroups IIA/IIIA and CBCT + MAR was used to develop SAH in subgroups IIB/IIIB. SEM and Micro-CT scanned the SAH openings to determine the diameter of the hole, cracking, chipping, and chipping volume.

RESULTS

Regarding the effect of plane CBCT and CBCT + MAR on prepared crowns, a highly significant association between group I with group II (p = 0.001) and group III (p = 0.002) was detected. Regarding the cracking of SAH, significant differences between the zirconium crown and CFM restoration (p = 0.009) were found, while for the chipping, no significant association was seen between groups (p = 0.19).

CONCLUSIONS

CBCT, either as a plane CBCT or with MAR, significantly improved the accuracy of drilling the screw channel and decreased injury to the existing restoration and abutment, aiding in better localization of SAH in loosened implant abutment screws.

Generation and Evaluation of Synthetic Computed Tomography (CT) from Cone-Beam CT (CBCT) by Incorporating Feature-Driven Loss into Intensity-Based Loss Functions in Deep Convolutional Neural Network

Simple Summary

Despite numerous benefits of cone-beam computed tomography (CBCT), its applications to radiotherapy were limited mainly due to degraded image quality. Recently, enhancing the CBCT image quality by generating synthetic CT image by deep convolutional neural network (CNN) has become frequent. Most of the previous works, however, generated synthetic CT with simple, classical intensity-driven loss in network training, while not specifying a full-package of verifications. This work trained the network by combining feature- and intensity-driven losses and attempted to demonstrate clinical relevance of the synthetic CT images by assessing both image similarity and dose calculating accuracy throughout a commercial Monte-Carlo.

Abstract

Deep convolutional neural network (CNN) helped enhance image quality of cone-beam computed tomography (CBCT) by generating synthetic CT. Most of the previous works, however, trained network by intensity-based loss functions, possibly undermining to promote image feature similarity. The verifications were not sufficient to demonstrate clinical applicability, either. This work investigated the effect of variable loss functions combining feature- and intensity-driven losses in synthetic CT generation, followed by strengthening the verification of generated images in both image similarity and dosimetry accuracy. The proposed strategy highlighted the feature-driven quantification in (1) training the network by perceptual loss, besides L1 and structural similarity (SSIM) losses regarding anatomical similarity, and (2) evaluating image similarity by feature mapping ratio (FMR), besides conventional metrics. In addition, the synthetic CT images were assessed in terms of dose calculating accuracy by a commercial Monte-Carlo algorithm. The network was trained with 50 paired CBCT-CT scans acquired at the same CT simulator and treatment unit to constrain environmental factors any other than loss functions. For 10 independent cases, incorporating perceptual loss into L1 and SSIM losses outperformed the other combinations, which enhanced FMR of image similarity by 10%, and the dose calculating accuracy by 1–2% of gamma passing rate in 1%/1mm criterion.

Image-based shading correction for narrow-FOV truncated pelvic CBCT with deep convolutional neural networks and transfer learning

Purpose: Cone beam computed tomography (CBCT) is a standard solution for in‐room image guidance for radiation therapy. It is used to evaluate and compensate for anatomopathological changes between the dose delivery plan and the fraction delivery day. CBCT is a fast and versatile solution, but it suffers from drawbacks like low contrast and requires proper calibration to derive density values. Although these limitations are even more prominent with in‐room customized CBCT systems, strategies based on deep learning have shown potential in improving image quality. As such, this article presents a method based on a convolutional neural network and a novel two‐step supervised training based on the transfer learning paradigm for shading correction in CBCT volumes with narrow field of view (FOV) acquired with an ad hoc in‐room system.

Methods: We designed a U‐Net convolutional neural network, trained on axial slices of corresponding CT/CBCT couples. To improve the generalization capability of the network, we exploited two‐stage learning using two distinct data sets. At first, the network weights were trained using synthetic CBCT scans generated from a public data set, and then only the deepest layers of the network were trained again with real‐world clinical data to fine‐tune the weights. Synthetic data were generated according to real data acquisition parameters. The network takes a single grayscale volume as input and outputs the same volume with corrected shading and improved HU values.

Results: Evaluation was carried out with a leave‐one‐out cross‐validation, computed on 18 unique CT/CBCT pairs from six different patients from a real‐world dataset. Comparing original CBCT to CT and improved CBCT to CT, we obtained an average improvement of 6 dB on peak signal‐to‐noise ratio (PSNR), +2% on structural similarity index measure (SSIM). The median interquartile range (IQR) Hounsfield unit (HU) difference between CBCT and CT improved from 161.37 (162.54) HU to 49.41 (66.70) HU. Region of interest (ROI)‐based HU difference was narrowed by 75% in the spongy bone (femoral head), 89% in the bladder, 85% for fat, and 83% for muscle. The improvement in contrast‐to‐noise ratio for these ROIs was about 67%.

Conclusions: We demonstrated that shading correction obtaining CT‐compatible data from narrow‐FOV CBCTs acquired with a customized in‐room system is possible. Moreover, the transfer learning approach proved particularly beneficial for such a shading correction approach.

Comparison of Supervised and Unsupervised Approaches for the Generation of Synthetic CT from Cone-Beam CT

Due to major artifacts and uncalibrated Hounsfield units (HU), cone-beam computed tomography (CBCT) cannot be used readily for diagnostics and therapy planning purposes. This study addresses image-to-image translation by convolutional neural networks (CNNs) to convert CBCT to CT-like scans, comparing supervised to unsupervised training techniques, exploiting a pelvic CT/CBCT publicly available dataset. Interestingly, quantitative results were in favor of supervised against unsupervised approach showing improvements in the HU accuracy (62% vs. 50%), structural similarity index (2.5% vs. 1.1%) and peak signal-to-noise ratio (15% vs. 8%). Qualitative results conversely showcased higher anatomical artifacts in the synthetic CBCT generated by the supervised techniques. This was motivated by the higher sensitivity of the supervised training technique to the pixel-wise correspondence contained in the loss function. The unsupervised technique does not require correspondence and mitigates this drawback as it combines adversarial, cycle consistency, and identity loss functions. Overall, two main impacts qualify the paper: (a) the feasibility of CNN to generate accurate synthetic CT from CBCT images, which is fast and easy to use compared to traditional techniques applied in clinics; (b) the proposal of guidelines to drive the selection of the better training technique, which can be shifted to more general image-to-image translation.

Cone beam CT based validation of neural network generated synthetic CTs for radiotherapy in the head region

PURPOSE

In the past years, many different neural network-based conversion techniques for synthesizing computed tomographys (sCTs) from MR images have been published. While the model's performance can be checked during the training against the test set, test datasets can never represent the whole population. Conversion errors can still occur for special cases, for example, for unusual anatomical situations. Therefore, the performance of sCT conversion needs to be verified on a patient specific level, especially in the absence of a planning CT (pCT). In this study, the capability of cone-beam CTs (CBCTs) for the validation of sCTs generated by a neural network was investigated.

METHODS

41 patients with tumors in the head region were selected. 20 of them were used for model training and 10 for validation. Different implementations of CycleGAN (with/without identity and feature loss) were used to generate sCTs. The pixel (MAE, RMSE, PSNR) and geometric error (DICE, Sensitivity, Specificity) values were reported to identify the best model. VMAT plans were created for the remaining 11 patients on the pCTs. These plans were re-calculated on sCTs and CBCTs. An automatic density overriding method ( C B C T RS ) and a population-based dose calculation method ( C B C T Pop ) were employed for CBCT-based dose calculation. The dose distributions were analysed using 3D global gamma analysis, applying a threshold of 10% with respect to the prescribed dose. Differences in DVH metrics for the PTV and the organs-at-risk were compared among the dose distributions based on pCTs, sCTs, and CBCTs.

RESULTS

The best model was the CycleGAN without identity and feature matching loss. Including the identity loss led to a metric decrease of 10% for DICE and a metric increase of 20-60 HU for MAE. Using the 2%/2 mm gamma criterion and pCT as reference, the mean gamma pass rates were 99.0  ±  0.4% for sCTs. Mean gamma pass rate values comparing pCT and CBCT were 99.0  ±  0.8% and 99.1  ±  0.8% for the C B C T RS and C B C T Pop , respectively. The mean gamma pass rates comparing sCT and CBCT resulted in 98.4  ±  1.6% and 99.2  ±  0.6% for C B C T RS and C B C T Pop , respectively. The differences between the gamma-pass-rates of the sCT and two CBCT-based methods were not significant. The majority of deviations of the investigated DVH metrices between sCTs and CBCTs were within 2%.

CONCLUSION

The dosimetric results demonstrate good agreement between sCT, CBCT, and pCT based calculations. A properly applied CBCT conversion method can serve as a tool for quality assurance procedures in an MR only radiotherapy workflow for head patients. Dosimetric deviations of DVH metrics between sCT and CBCTs of larger than 2% should be followed up. A systematic shift of approximately 1% should be taken into account when using the C B C T RS approach in an MR only workflow.

Quantitative analysis of metal artifact reduction using the auto-edge counting method in cone-beam computed tomography

The metal artifact reduction (MAR) algorithm is used in most CBCT unit to reduce artifact from various dental materials. The performance of MAR program of a CBCT unit according to the dental material type under different imaging mode was evaluated as introducing automatic quantification of the amount of artifact reduced. Four customized phantoms with different dental prostheses (amalgam, gold, porcelain-fused-metal, zirconia) underwent CBCT scanning with and without the MAR option. The imaging was performed under varied scanning conditions; 0.2 and 0.3 mm³ voxel sizes; 70 and 100 kVp. The amount of artifacts reduced by each prosthesis and scanning mode automatically counted using canny edge detection in MATLAB, and statistical analysis was performed. The overall artifact reduction ratio was ranged from 17.3% to 55.4%. The artifact caused by the gold crown was most effectively reduced compared to the other prostheses (p < 0.05, Welch’s ANOVA analysis). MAR showed higher performance in smaller voxel size mode for all prostheses (p < 0.05, independent t-test). Automatic quantification efficiently evaluated MAR performance in CBCT image. The impact of MAR was different according to the prostheses type and imaging mode, suggesting that thoughtful consideration is required when selecting the imaging mode of CBCT.