Megavoltage CT Images Research Articles

Estimation of weighted computed tomography dose index (CTDIw) in megavoltage computed tomography (MVCT)

Abstract Introduction Advanced technologies in radiation treatment deliveries require high accuracy and precision in setup before treatment to get accurate dose delivery for planned target volumes. During imaging procedures, patients are getting extra doses from kilovoltage computed tomography (KVCT) and megavoltage computed tomography (MVCT). Computed tomography dose index is an important dosimetric parameter to know the degree of radiation-induced side effects. Aim The main aim of this study is to quantify the dose received by patients during MVCT imaging procedures, weighted computed tomography dose index (CTDIw), volume computed tomography dose index (CTDIvol) and dose length product (DLP). Methods and materials This study has been performed on Radixact X9 LINAC machine with 6MV flattening filter-free (FFF) photon beam for treatment and imaging with 3.5 MeV average photon energy. CTDIw and dose length product (DLP) have been calculated for different scanning lengths 18 cm, 10 cm and 0.7 cm with fine, normal and coarse modes with standard imaging A1SL1 ion chamber, cheese phantom and tomoelectrometer. Results CTDIw, CTDIvol, and DLP have been calculated with cheese phantom using A1SL1 ion chamber. CTDIw for 0.7 cm is 2.16 cGy, 1.14 cGy and 0.77 cGy for fine normal and coarse modes, respectively. Conclusion CTDIw is a dosimetric index to know the dose levels at center and periphery of patient according to scan length. It gives an idea about how much additional dose is received by the patient during the imaging procedures. High pitch gives the lesser absorbed dose. DLP gives the knowledge about stochastic effects from radiation to patients after MVCT.

DeCoGAN: MVCT image denoising via coupled generative adversarial network

Objective. In helical tomotherapy, image-guided radiotherapy employs megavoltage computed tomography (MVCT) for precise targeting. However, the high voltage of megavoltage radiation introduces substantial noise, significantly compromising MVCT image clarity. This study aims to enhance MVCT image quality using a deep learning-based denoising method. Approach. We propose an unpaired MVCT denoising network using a coupled generative adversarial network framework (DeCoGAN). Our approach assumes that a universal latent code within a shared latent space can reconstruct any given pair of images. By employing an encoder, we enforce this shared-latent space constraint, facilitating the conversion of low-quality (noisy) MVCT images into high-quality (denoised) counterparts. The network learns the joint distribution of images from both domains by leveraging samples from their respective marginal distributions, enhanced by adversarial training for effective denoising. Main Results. Compared to an analytical algorithm (BM3D) and three deep learning-based methods (RED-CNN, WGAN-VGG and CycleGAN), the proposed method excels in preserving image details and enhancing human visual perception by removing most noise and retaining structural features. Quantitative analysis demonstrates that our method achieves the highest peak signal-to-noise ratio and Structural Similarity Index Measurement values, indicating superior denoising performance. Significance. The proposed DeCoGAN method shows remarkable MVCT denoising performance, making it a promising tool in the field of radiation therapy.

A Novel Approach for Position Verification and Dose Calculation through Local MVCT Reconstruction.

Traditional positioning verification using cone-beam computed tomography (CBCT) may incur errors due to potential misalignments between the isocenter of CBCT and the treatment beams in radiotherapy. This study introduces an innovative method for verifying patient positioning in radiotherapy. Initially, the transmission images from an electronic portal imaging device (EPID) are acquired from 10 distinct angles. Utilizing the ART-TV algorithm, a sparse reconstruction of local megavoltage computed tomography (MVCT) is performed. Subsequently, this MVCT is aligned with the planning CT via a three-dimensional mutual information registration technique, pinpointing any patient-positioning discrepancies and facilitating corrective adjustments to the treatment setup. Notably, this approach employs the same radiation source as used in treatment to obtain three-dimensional images, thereby circumventing errors stemming from misalignment between the isocenter of CBCT and the accelerator. The registration process requires only 10 EPID images, and the dose absorbed during this process is included in the total dose calculation. The results show that our method's reconstructed MVCT images fulfill the requirements for registration, and the registration algorithm accurately detects positioning errors, thus allowing for adjustments in the patient's treatment position and precise calculation of the absorbed dose.

The effects of mega-voltage CT scan parameters on offline adaptive radiation therapy.

TomoTherapy involves image-guided radiation therapy (IGRT) using Mega-voltage CT (MVCT) for each treatment session. The acquired MVCT images can be utilized for the retrospective assessment of dose distribution. The TomoTherapy provides 18 distinct imaging conditions that can be selected based on a combination of algorithms, acquisition pitch, and slice interval. We investigated the accuracy of dose calculation and deformable image registration (DIR) depending on MVCT scan parameters and their effects on adaptive radiation therapy (ART). We acquired image values for density calibration tables (IVDTs) under 18 different MVCT conditions and compared them. The planning CT (pCT) was performed using a thoracic phantom, and an esophageal intensity-modulated radiation therapy (IMRT) plan was created. MVCT images of the thoracic phantom were acquired under each of the 18 conditions, and dose recalculation was performed. DIR was performed on the MVCT images acquired under each condition. The accuracy of DIR, depending on the MVCT scan parameters, was compared using the mean distance to agreement (MDA) and Dice similarity coefficient (DSC). The dose distribution calculated on the MVCT images was deformed using deformed vector fields (DVF). No significant differences were observed in the results of the 18 IVDTs. The esophageal IMRT plan also showed a small dose difference. Regarding verifying the DIR accuracy, the MDA increased, and the DSC decreased as the acquisition pitch and slice interval increased. The difference between the dose distributions after dose mapping was comparable to that before DIR. The MVCT scan parameters had little effect on ART.

MVCT image enhancement using reference-based encoder–decoder convolutional neural network

Daily MVCT (Megavoltage Computed Tomography) in TomoTherapy plays a crucial role in patients’ setup and dose reconstruction. However, MVCT images suffer from high noise and low tissue contrast due to the limited number of X-ray photons and low detector quantum efficiency (DQE). In this study, we propose an approach to enhance MVCT images using the KVCT (Kilovoltage Computed Tomography) image of the same patient as an auxiliary reference image. Specifically, we propose a reference-based encoder–decoder convolutional neural network (RefED-CNN) by incorporating a feature extraction and alignment (FEA) module to introduce the features of the reference image as side information into the MVCT image enhancement process. The FEA module automatically searches and aligns relevant features between the reference image and the noisy image, and transfers the relevant texture from reference KVCT images to the target MVCT image. Evaluations conducted on both phantom and real patient data show that our method outperforms other denoising methods by effectively reducing noise and preserving intricate structural details.

Reproducibility and Repeatability of Pelvic Radiomics Features with Daily Imaging on a Novel Biology-Guided Radiotherapy Machine Compared to Daily Imaging on Other Radiotherapy Delivery Systems

Radiomics features extracted from computed tomography (CT) images have the potential of being used as imaging biomarkers in diagnostic and prognostic models for prostate cancer. This study aims to evaluate reproducibility and repeatability of radiomics features using daily CT images from a novel biology-guided radiotherapy (BgRT)-capable machine compared to other daily CT imaging modalities from other radiotherapy delivery systems. We retrospectively selected thirty male patients treated to the pelvic region at our institution. Among the thirty patients, ten patients were treated on a BgRT-capable machine with daily fan-beam kilovoltage CT (kVCT) scans, ten patients were treated on a conventional Linac with daily cone-beam CT (CBCT) scans, and ten patients were treated on helical tomotherapy with daily megavoltage CT (MVCT) scans. The prostate gland and femoral heads were delineated on planning CT images and on CT images at the first treatment fraction. After all the CT images were resampled to have an identical voxel size, 107 radiomics features were calculated for the prostate and femoral heads on both the planning CT images and daily CT images. Concordance correlation coefficient (CCC) and Pearson's correlation coefficient (r) were calculated to evaluate radiomics feature reproducibility between the planning CT images and pre-treatment CT scans at the first treatment fraction. To evaluate radiomics feature repeatability with kVCT images on the BgRT-capable machine, a total of ten treatment sessions with repeat kVCT scans on the BgRT-capable machine were retrieved and radiomics features were compared between the first and the second kVCT scans. For the prostate gland volume, high concordance (CCC > 0.8) was found for 8, 6, and 0 radiomics features with the first-fraction kVCT, CBCT, and MVCT images, respectively, in relation to the planning CT images, while there was strong correlation (r > 0.8) in 38, 22, and 5 radiomics features in the kVCT, CBCT, and MVCT images, respectively, in relation to the planning CT images. For the left and right femoral heads, high concordance (CCC > 0.8) was found in no more than two radiomics features with the kVCT, CBCT, or MVCT images in relation to the planning CT images. Based on repeat kVCT scan data, 97 (90.7%) radiomics features showed high concordance (CCC > 0.8) in repeat kVCT scans with 98 (91.6%) features showing high correlation (Pearson's correlation coefficient > 0.8) for the prostate gland volume. A small number of radiomics features were found to show strong reproducibility between daily CT images on the first fraction and the planning CT images with the three IGRT CT modalities in this study. On the other hand, our results indicated that strong reproducibility was found with more radiomics features with daily kVCT images from the BgRT-capable machine compared to daily CBCT and MVCT images for the prostate gland. Strong repeatability was also found with most radiomics features in daily kVCT images.

Comparison of AI-Based Auto-Segmentation Quality with Different Daily IGRT Imaging Modalities for Adaptive Radiotherapy Treatment Planning

Modern AI-based auto-segmentation tools may be applied on daily computed tomography (CT) images in an image-guided radiotherapy course to facilitate evaluation of anatomical variations and adaptive treatment planning. However, most AI auto-segmentation models are trained with high-quality diagnostic or planning CT image data. This study aims to evaluate and compare auto-segmentation quality by an AI-based auto-segmentation system with different daily CT imaging modalities. We retrospectively retrieved daily IGRT images for sixty patients. Among them, twenty patients were treated on helical tomotherapy with daily megavoltage CT (MVCT) scans, twenty patients were treated on conventional Linacs with daily cone-beam CT (CBCT) scans, and twenty patients were treated on a biology-guided radiotherapy (BgRT)-capable machine with daily fan-beam kilovoltage CT (kVCT) scans. With each treatment modality, ten patients received daily CT scans in the pelvic region while the other ten received daily CT scans in the thoracic region. An auto-segmentation system using a convolutional neural network algorithm was trained in-house with historical treatment planning CT and contouring data to generate auto-segmentation models for the pelvic and thoracic regions, respectively. Normal organs were first delineated the auto-segmentation system on the daily CT images and then drawn by an experienced planner. A set of metrics including the dice similarity coefficient (DSC), Jaccard similarity index, and Hausdorff distance were used to evaluate the quality of the auto segmentation results compared with manual contours. The auto-segmentation contours on the kVCT images showed higher average DSC compared to those on the MVCT and CBCT images for all the major organs in both the pelvic and thoracic regions including the bladder, rectum, bowel, left and right femurs, esophagus, heart, left and right lung, and spinal cord. With the kVCT images, the average DSC ranged from 0.52±0.22 to 0.996±0.005. In the pelvic region, the largest absolute difference in DSC was observed for the bowel volume with an average DSC of 0.69±0.16, 0.49±0.27, and 0.32±0.25 for the kVCT, MVCT, and CBCT images, respectively (p-value < 0.05 with unpaired t-tests between kVCT and the other two imaging modalities); in the thoracic region, the largest absolute difference in DSC was observed for the esophagus with an average DSC of 0.63±0.16, 0.22±0.08, and 0.15±0.18 for the kVCT, MVCT, and CBCT images, respectively (p-value < 0.05 with unpaired t-tests between kVCT and the other two imaging modalities). Similar results were observed with other metrics. The AI-based auto segmentation system showed improved agreement with manual contouring when using kVCT images from the BgRT capable machine compared to MVCT or CBCT images. However, manual correction is necessary on auto-segmentation results from all imaging modalities especially for organs with limited contrast from surrounding tissues.

Feasibility of artificial intelligence-driven interfractional monitoring of organ changes by mega-voltage computed tomography in intensity-modulated radiotherapy of prostate cancer.

High-dose radiotherapy (RT) for localized prostate cancer requires careful consideration of target position changes and adjacent organs-at-risk (OARs), such as the rectum and bladder. Therefore, daily monitoring of target position and OAR changes is crucial in minimizing interfractional dosimetric uncertainties. For efficient monitoring of the internal condition of patients, we assessed the feasibility of an auto-segmentation of OARs on the daily acquired images, such as megavoltage computed tomography (MVCT), via a commercial artificial intelligence (AI)-based solution in this study. We collected MVCT images weekly during the entire course of RT for 100 prostate cancer patients treated with the helical TomoTherapy system. Based on the manually contoured body outline, the bladder including prostate area, and rectal balloon regions for the 100 MVCT images, we trained the commercially available fully convolutional (FC)-DenseNet model and tested its auto-contouring performance. Based on the optimally determined hyperparameters, the FC-DenseNet model successfully auto-contoured all regions of interest showing high dice similarity coefficient (DSC) over 0.8 and a small mean surface distance (MSD) within 1.43 mm in reference to the manually contoured data. With this well-trained AI model, we have efficiently monitored the patient's internal condition through six MVCT scans, analyzing DSC, MSD, centroid, and volume differences. We have verified the feasibility of utilizing a commercial AI-based model for auto-segmentation with low-quality daily MVCT images. In the future, we will establish a fast and accurate auto-segmentation and internal organ monitoring system for efficiently determining the time for adaptive replanning.

A qualitative study of improving megavoltage computed tomography image quality and maintaining dose accuracy using cycleGAN-based image synthesis.

Due to inconsistent positioning, tumor shrinking, and weight loss during fractionated treatment, the initial plan was no longer appropriate after a few fractional treatments, and the patient will require adaptive helical tomotherapy (HT) to overcome the issue. Patients are scanned with megavoltage computed tomography (MVCT) before each fractional treatment, which is utilized for patient setup and provides information for dose reconstruction. However, the low contrast and high noise of MVCT make it challenging to delineate treatment targets and organs at risk (OAR). This study developed a deep-learning-based approach to generate high-quality synthetic kilovoltage computed tomography (skVCT) from MVCT and meet clinical dose requirements. Data from 41 head and neck cancer patients were collected; 25 (2995 slices) were used for training, and 16 (1898 slices) for testing. A cycle generative adversarial network (cycleGAN) based on attention gate and residual blocks was used to generate MVCT-based skVCT. For the 16 patients, kVCT-based plans were transferred to skVCT images and electron density profile-corrected MVCT images to recalculate the dose. The quantitative indices and clinically relevant dosimetric metrics, including the mean absolute error (MAE), structural similarity index measure (SSIM), peak signal-to-noise ratio (PSNR), gamma passing rates, and dose-volume-histogram (DVH) parameters (Dmax , Dmean , Dmin ), were used to assess the skVCT images. The MAE, PSNR, and SSIM of MVCT were 109.6±12.3 HU, 27.5±1.1dB, and 91.9%±1.7%, respectively, while those of skVCT were 60.6±9.0 HU, 34.0±1.9dB, and 96.5%±1.1%. The image quality and contrast were enhanced, and the noise was reduced. The gamma passing rates improved from 98.31%±1.11% to 99.71%±0.20% (2mm/2%) and 99.77%±0.18% to 99.98%±0.02% (3mm/3%). No significant differences (p>0.05) were observed in DVH parameters between kVCT and skVCT. With training on a small data set (2995 slices), the model successfully generated skVCT with improved image quality, and the dose calculation accuracy was similar to that of MVCT. MVCT-based skVCT can increase treatment accuracy and offer the possibility of implementing adaptive radiotherapy.

Identification of reproducible radiomic features from on-board volumetric images: A multi-institutional phantom study.

Radiomics analysis using on-board volumetric images has attracted research attention as a method for predicting prognosis during treatment; however, the lack of standardization is still one of the main concerns. This study investigated the factors that influence the reproducibility of radiomic features extracted from on-board volumetric images using an anthropomorphic radiomics phantom. Furthermore, a phantom experiment was conducted with different treatment machines from multiple institutions as external validation to identify reproducible radiomic features. The phantom was designed to be 35 × 20 × 20cm with eight types of heterogeneous spheres (⌀=1, 2, and 3cm). On-board volumetric images were acquired using 15 treatment machines from eight institutions. Of these, kilovoltage cone-beam computed tomography (kV-CBCT) image data acquired from four treatment machines at one institution were used as an internal evaluation dataset to explore the reproducibility of radiomic features. The remaining image data, including kV-CBCT, megavoltage-CBCT (MV-CBCT), and megavoltage computed tomography (MV-CT) provided by seven different institutions (11 treatment machines), were used as an external validation dataset. A total of 1,302 radiomic features, including 18 first-order, 75 texture, 465 (i.e., 93 × 5) Laplacian of Gaussian (LoG) filter-based, and 744 (i.e., 93 × 8) wavelet filter-based features, were extracted within the spheres. The intraclass correlation coefficient (ICC) was calculated to explore feature repeatability and reproducibility using an internal evaluation dataset. Subsequently, the coefficient of variation (COV) was calculated to validate the feature variability of external institutions. An absolute ICC exceeding 0.85 or COV under 5% was considered indicative of a highly reproducible feature. For internal evaluation, ICC analysis showed that the median percentage of radiomic features with high repeatability was 95.2%. The ICC analysis indicated that the median percentages of highly reproducible features for inter-tube current, reconstruction algorithm, and treatment machine were decreased by 20.8%, 29.2%, and 33.3%, respectively. For external validation, the COV analysis showed that the median percentage of reproducible features was 31.5%. A total of 16 features, including nine LoG filter-based and seven wavelet filter-based features, were indicated as highly reproducible features. The gray-level run-length matrix (GLRLM) was classified as containing the most frequent features (N=8), followed by the gray-level dependence matrix (N=7) and gray-level co-occurrence matrix (N=1) features. We developed the standard phantom for radiomics analysis of kV-CBCT, MV-CBCT, and MV-CT images. With this phantom, we revealed that the differences in the treatment machine and image reconstruction algorithm reduce the reproducibility of radiomic features from on-board volumetric images. Specifically, the most reproducible features for external validation were LoG or wavelet filter-based GLRLM features. However, the acceptability of the identified features should be examined in advance at each institution before applying the findings to prognosis prediction.

Influence of different air CT numbers for IVDT on the dose distribution in TomoTherapy MVCT.

This study aims to evaluate the effect of different air computed tomography (CT) numbers of the image value density table (IVDT) on the retrospective dose calculation of head-and-neck (HN) radiotherapy using TomoTherapy megavoltage CT (MVCT) images. The CT numbers of the inside and outside air and each tissue-equivalent plug of the "Cheese" phantom were obtained from TomoTherapy MVCT. Two IVDTs with different air CT numbers were created and applied to MVCT images of the HN anthropomorphic phantom and recalculated by Planned Adaptive to verify dose distribution. We defined the recalculation dose with MVCT images using both inside and outside air of the IVDT as and , respectively. Treatment planning doses calculated on kVCT images were compared with those calculated on MVCT images using two different IVDT tables, namely, and . The difference between average MVCT numbers ±1 standard deviation on inside and outside air of the calibration phantom was 65±36HU. This difference in MVCT number of air exceeded the recommendation lung tolerance for dose calculation error of 2%. The dose differences between the planning target volume (PTV): D98% , D50% , D2% and the organ at risk (OAR): Dmax , Dmean recalculated by and using MVCT images were a maximum of 0.7% and 1.2%, respectively. Recalculated doses to the PTV and OAR with MVCT showed that was 0.5%-0.7% closer to the kVCT treatment planning dose than . This study showed that was more accurate than in recalculating the dose HN cases of MVCT using TomoTherapy.

Benefits for Head and Neck Positioning Using a Surface Image Guidance System

<h3>Purpose/Objective(s)</h3> Surface-guided radiation therapy (SGRT) is an emerging technology that enables three-dimensional surfaces image guidance for radiotherapy. The technology is commonly employed in the setup for breast cancer patients; however, it would be challenging for head and neck cancer patients. The purpose of this study was to investigate the accuracy of surface image-guided positioning compared to conventional in-room-laser based positioning using marks on a thermoplastic mask for the head and neck patient. <h3>Materials/Methods</h3> Thirty patients were randomly selected for two groups: the surface image-guided positioning group (SGRT group) and the conventional in-room-laser based positioning group using marks on a thermoplastic mask (non-SGRT group). Total sixty patients received intensity-modulated radiotherapy using helical tomotherapy for head and neck cancer. For both groups, the positional errors for translational and rotational shifts were acquired by aligning the mega-voltage CT (MVCT) images in the first treatment fraction for the treatment planning kVCT images. In addition, the number of the additional MVCT scan for repositioning during all treatment fractions was counted. The reposition was performed when the rotational shift error was greater than 2.0°. The differences of the positional errors and the mean number of MVCT scans per fraction between the SGRT and non-SGRT groups were evaluated using the Mann-Whitney U test. <h3>Results</h3> The mean positional errors in the lateral, longitudinal, and vertical directions were as follows: -0.8±0.9, -3.0±2.6, and 0.7±1.4 mm in the SGRT group, and -1.3±1.5, -1.0±1.9, and 0.1±1.2 mm in the non-SGRT group. The only positional error in the longitudinal direction was significantly larger in the SGRT than non-SGRT groups (<i>p</i> < 0.05); however, the difference could be relatively small. The mean positional error of rotational shift in each axis was less than 0.2° in both groups, and there was no significant difference between the two groups. The mean number of MVCT scans per fraction in the SGRT and non-SGRT groups was 1.14 and 1.26 (total number; about 40 and 44 scans at 35 fractions), respectively. It was significantly lower in the SGRT group than in the non-SGRT group (<i>p</i> < 0.05). <h3>Conclusion</h3> Surface image-guided positioning resulted in a larger positional error in the longitudinal direction than conventional in-room-laser based positioning using marks on a thermoplastic mask. The difference could be relatively small. In addition, it contributed to the reduction of the additional MVCT scan for repositioning.

Generation of synthetic megavoltage CT for MRI-only radiotherapy treatment planning using a 3D deep convolutional neural network.

Megavoltage computed tomography (MVCT) has been implemented on many radiotherapy treatment machines for on-board anatomical visualization, localization, and adaptive dose calculation. Implementing an MR-only workflow by synthesizing MVCT from magnetic resonance imaging (MRI) would offer numerous advantages for treatment planning and online adaptation. In this work, we sought to synthesize MVCT (sMVCT) datasets from MRI using deep learning to demonstrate the feasibility of MRI-MVCT only treatment planning. MVCTs and T1-weighted MRIs for 120 patients treated for head-and-neck cancer were retrospectively acquired and co-registered. A deep neural network based on a fully-convolutional 3D U-Net architecture was implemented to map MRI intensity to MVCT HU. Input to the model were volumetric patches generated from paired MRI and MVCT datasets. The U-Net was initialized with random parameters and trained on a mean absolute error (MAE) objective function. Model accuracy was evaluated on 18 withheld test exams. sMVCTs were compared to respective MVCTs. Intensity-modulated volumetric radiotherapy (IMRT) plans were generated on MVCTs of four different disease sites and compared to plans calculated onto corresponding sMVCTs using the gamma metric and dose-volume-histograms (DVHs). MAE values between sMVCT and MVCT datasets were 93.3±27.5, 78.2±27.5, and 138.0±43.4 HU for whole body, soft tissue, and bone volumes, respectively. Overall, there was good agreement between sMVCT and MVCT, with bone and air posing the greatest challenges. The retrospective dataset introduced additional deviations due to sinus filling or tumor growth/shrinkage between scans, differences in external contours due to variability in patient positioning, or when immobilization devices were absent from diagnostic MRIs. Dose distributions of IMRT plans evaluated for four test cases showed close agreement between sMVCT and MVCT images when evaluated using DVHs and gamma dose metrics, which averaged to 98.9±1.0% and 96.8±2.6% analyzed at 3%/3mm and 2%/2mm, respectively. MVCT datasets can be generated from T1-weighted MRI using a 3D deep convolutional neural network with dose calculation on a sample sMVCT in close agreement with the MVCT. These results demonstrate the feasibility of using MRI-derived sMVCT in an MR-only treatment planning workflow.

Training of deep cross-modality conversion models with a small data set, and their application in megavoltage CT to kilovoltage CT conversion.

In recent years, deep learning-based image processing has emerged as a valuable tool for medical imaging owing to its high performance. However, the quality of deep learning-based methods heavily relies on the amount of training data; the high cost of acquiring a large data set is a limitation to their utilization in medical fields. Herein, based on deep learning, we developed a computed tomography (CT) modality conversion method requiring only a few unsupervisedimages. The proposed method is based on cycle-consistency generative adversarial network (CycleGAN) with several extensions tailored for CT images, which aims at preserving the structure in the processed images and reducing the amount of training data. This method was applied to realize the conversion of megavoltage computed tomography (MVCT) to kilovoltage computed tomography (kVCT) images. Training was conducted using several data sets acquired from patients with head and neck cancer. The size of the data sets ranged from 16 slices (two patients) to 2745 slices (137 patients) for MVCT and 2824 slices (98 patients) forkVCT. The required size of the training data was found to be as small as a few hundred slices. By statistical and visual evaluations, the quality improvement and structure preservation of the MVCT images converted by the proposed model were investigated. As a clinical benefit, it was observed by medical doctors that the converted images enhanced the precision ofcontouring. We developed an MVCT to kVCT conversion model based on deep learning, which can be trained using only a few hundred unpaired images. The stability of the model against changes in data size was demonstrated. This study promotes the reliable use of deep learning in clinical medicine by partially answering commonly asked questions, such as "Is our data sufficient?" and "How much data should we acquire?"

Repetitive Painting (REPEAT) Irradiation in Stereotactic Radiotherapy Using Helical Tomotherapy.

Stereotactic radiotherapy (SRT) for spine metastases with helical tomotherapy requires a long irradiation time due to the high dose per fraction. Since helical tomotherapy can neither confirm nor correct the position during irradiation, a plan with a long irradiation time cannot be used in actual clinical practice, given the intra-fractional motion error. To address this problem, we devised a method called REPEAT irradiation. REPEtitive pAinTing (REPEAT) irradiation is a method of dividing the irradiation for a given fraction per day into several sessions and performing the irradiation after position correction using mega-voltage computed tomography images for each session. In order to evaluate how REPEAT irradiation changes irradiation time and the dose-volume histogram (DVH), a planning study with helical tomotherapy was conducted using CT images of a patient with lumbar spine metastasis. In this case, we found that dividing 3 irradiation fractions into 3 sessions per day (i.e., 9 fractions=9 sessions in 3 days) using REPEAT irradiation shortened the irradiation time per session and simultaneously improved dose-volume histogram parameters. Although the optimal number of sessions may differ depending on the patient's condition, the fixing method, the irradiation site, and the calculation parameters, REPEAT irradiation does not require any special equipment and is a simple practical treatment method.

Automatic Parotid Gland Segmentation in MVCT Using Deep Convolutional Neural Networks

Radiation-induced xerostomia, as a major problem in radiation treatment of the head and neck cancer, is mainly due to the overdose irradiation injury to the parotid glands. Helical Tomotherapy-based megavoltage computed tomography (MVCT) imaging during the Tomotherapy treatment can be applied to monitor the successive variations in the parotid glands. While manual segmentation is time consuming, laborious, and subjective, automatic segmentation is quite challenging due to the complicated anatomical environment of head and neck as well as noises in MVCT images. In this article, we propose a localization-refinement scheme to segment the parotid gland in MVCT. After data pre-processing we use mask region convolutional neural network (Mask R-CNN) in the localization stage after data pre-processing, and design a modified U-Net in the following fine segmentation stage. To the best of our knowledge, this study is a pioneering work of deep learning on MVCT segmentation. Comprehensive experiments based on different data distribution of head and neck MVCTs and different segmentation models have demonstrated the superiority of our approach in terms of accuracy, effectiveness, flexibility, and practicability. Our method can be adopted as a powerful tool for radiation-induced injury studies, where accurate organ segmentation is crucial.

A deep-learning method for generating synthetic kV-CT and improving tumor segmentation for helical tomotherapy of nasopharyngeal carcinoma

Objective: Megavoltage computed tomography (MV-CT) is used for setup verification and adaptive radiotherapy in tomotherapy. However, its low contrast and high noise lead to poor image quality. This study aimed to develop a deep-learning-based method to generate synthetic kilovoltage CT (skV-CT) and then evaluate its ability to improve image quality and tumor segmentation. Approach: The planning kV-CT and MV-CT images of 270 patients with nasopharyngeal carcinoma (NPC) treated on an Accuray TomoHD system were used. An improved cycle-consistent adversarial network which used residual blocks as its generator was adopted to learn the mapping between MV-CT and kV-CT and then generate skV-CT from MV-CT. A Catphan 700 phantom and 30 patients with NPC were used to evaluate image quality. The quantitative indices included contrast-to-noise ratio (CNR), uniformity and signal-to-noise ratio (SNR) for the phantom and the structural similarity index measure (SSIM), mean absolute error (MAE), and peak signal-to-noise ratio (PSNR) for patients. Next, we trained three models for segmentation of the clinical target volume (CTV): MV-CT, skV-CT, and MV-CT combined with skV-CT. The segmentation accuracy was compared with indices of the dice similarity coefficient (DSC) and mean distance agreement (MDA). Main results: Compared with MV-CT, skV-CT showed significant improvement in CNR (184.0%), image uniformity (34.7%), and SNR (199.0%) in the phantom study and improved SSIM (1.7%), MAE (24.7%), and PSNR (7.5%) in the patient study. For CTV segmentation with only MV-CT, only skV-CT, and MV-CT combined with skV-CT, the DSCs were 0.75 ± 0.04, 0.78 ± 0.04, and 0.79 ± 0.03, respectively, and the MDAs (in mm) were 3.69 ± 0.81, 3.14 ± 0.80, and 2.90 ± 0.62, respectively. Significance: The proposed method improved the image quality of MV-CT and thus tumor segmentation in helical tomotherapy. The method potentially can benefit adaptive radiotherapy.

Improvement of megavoltage computed tomography image quality for adaptive helical tomotherapy using cycleGAN-based image synthesis with small datasets.

Megavoltage computed tomography (MVCT) offers an opportunity for adaptive helical tomotherapy. However, high noise and reduced contrast in the MVCT images due to a decrease in the imaging dose to patients limits its usability. Therefore, we propose an algorithm to improve the image quality of MVCT. The proposed algorithm generates kilovoltage CT (kVCT)-like images from MVCT images using a cycle-consistency generative adversarial network (cycleGAN)-based image synthesis model. Data augmentation using an affine transformation was applied to the training data to overcome the lack of data diversity in the network training. The mean absolute error (MAE), root-mean-square error (RMSE), peak signal-to-noise ratio (PSNR), and structural similarity index measure (SSIM) were used to quantify the correction accuracy of the images generated by the proposed algorithm. The proposed method was validated by comparing the images generated with those obtained from conventional and deep learning-based image processing method through non-augmented datasets. The average MAE, RMSE, PSNR, and SSIM values were 18.91 HU, 69.35 HU, 32.73dB, and 95.48 using the proposed method, respectively, whereas cycleGAN with non-augmented data showed inferior results (19.88 HU, 70.55 HU, 32.62dB, 95.19, respectively). The voxel values of the image obtained by the proposed method also indicated similar distributions to those of the kVCT image. The dose-volume histogram of the proposed method was also similar to that of electron density corrected MVCT. The proposed algorithm generates synthetic kVCT images from MVCT images using cycleGAN with small patient datasets. The image quality achieved by the proposed method was correspondingly improved to the level of a kVCT image while maintaining the anatomical structure of an MVCT image. The evaluation of dosimetric effectiveness of the proposed method indicates the applicability of accurate treatment planning in adaptive radiation therapy.

MVCT versus kV-CBCT for targets subject to respiratory motion: A phantom study.

The use of kilovoltage cone‐beam computed tomography (kV‐CBCT) or megavoltage computed tomography (MVCT) for image guidance prior to lung stereotactic body radiation therapy (SBRT) is common clinical practice. We demonstrate that under equivalent respiratory conditions, image guidance using both kV‐CBCT and MVCT may result in the inadequate estimation of the range of target motion under free‐breathing (FB) conditions when standard low‐density window and levels are used. Two spherical targets within a respiratory motion phantom were imaged using both long‐exhale (LE) and sinusoidal respiratory traces. MVCT and kV‐CBCT images were acquired and evaluated for peak‐to‐peak amplitudes of 10 or 20 mm in the cranial‐caudal direction, and with 2, 4 or 5 s periods. All images were visually inspected for artifacts and conformity to the ITV for each amplitude, period, trace‐type, and target size. All LE respiratory traces required a lower threshold HU window for MVCT and kV‐CBCT compared to sinusoidal traces to obtain 100% volume conformity compared with the theoretical ITV (ITVT). Excess volume was less than 2% for all kV‐CBCT contours regardless of trace‐type, breathing period, or amplitude, while the maximum excess volume for MVCT was 48%. Adjusting window and level to maximize conformity with the ITVT is necessary to reduce registration uncertainty to less than 5 mm. To fully capture target motion with either MVCT or kV‐CBCT, substantial changes in HU levels up to −600 HU are required which may not be feasible clinically depending on the target's location and surrounding tissue contrast. This registration method, utilizing a substantially decreased window and level compared to standard low‐density settings, was retrospectively compared to the automated registration algorithm for five lung SBRT patients exposed to pre‐treatment kV‐CBCT image guidance. Differences in registrations in the super‐inferior (SI) direction greater than the commonly used ITV to PTV margin of 5 mm were encountered for several cases. In conclusion, pre‐treatment image guidance for lung SBRT targets using MVCT or kV‐CBCT is unlikely to capture the full extent of target motion as defined by the ITVT and additional caution is warranted to avoid registration errors for small targets and patients with LE respiratory traces.

The effects of rotational setup errors in total body irradiation using helical tomotherapy.

PurposeHelical tomotherapy (HT) is a form of intensity‐modulated radiation therapy that is employed in total body irradiation (TBI). Because TBI targets the whole body, accurate setup positioning at the edge of the treatment volume is made difficult by the whole‐body rotational posture. The purpose of this study is to clarify the tolerance for rotational setup error (SE) in the vertical direction. In addition, we perform a retrospective analysis of actually irradiated dose distributions using previous patients’ irradiation data.MethodsTo clarify the effects of rotational SE on the dose distribution, the planned CT images of 10 patients were rotated by 1–5° in the vertical (pitch) direction to create a pseudo‐rotational SE image. Then, the effect of the magnitude of the rotational SE on the dose distribution was simulated. In addition, the irradiated dose to the patients was analyzed by obtaining recalculated dose distributions using megavoltage CT images acquired before treatment.ResultsThe simulation results showed that the average value of the lung volume receiving at least 10 Gy did not exceed the allowable value when the SE value was ≤2°. When the rotational SE was ≤3°, it was possible to maintain the clinical target volume dose heterogeneity within ±10% of the prescribed dose, which is acceptable according to the guidelines. A retrospective analysis of previous patients’ irradiation data showed their daily irradiation dose distribution. The dose to the clinical target volume was reduced by up to 3.4% as a result of the residual rotational SE. Although whole‐course retrospective analyses showed a statistically significant increase in high‐dose areas, the increase was only approximately 1.0%.ConclusionsDose errors induced by rotational SEs of ≤2° were acceptable in this study.