Metal artifact reduction in kV CT images throughout two-step sequential deep convolutional neural networks by combining multi-modal imaging (MARTIAN)

This work attempted to construct a new metal artifact reduction (MAR) framework in kilo-voltage (kV) computed tomography (CT) images by combining (1) deep learning and (2) multi-modal imaging, defined as MARTIAN (Metal Artifact Reduction throughout Two-step sequentIAl deep convolutional neural Networks). Most CNNs under supervised learning require artifact-free images to artifact-contaminated images for artifact correction. Mega-voltage (MV) CT is insensitive to metal artifacts, unlike kV CT due to different physical characteristics, which can facilitate the generation of artifact-free synthetic kV CT images throughout the first network (Network 1). The pairs of true kV CT and artifact-free kV CT images after post-processing constructed a subsequent network (Network 2) to conduct the actual MAR process. The proposed framework was implemented by GAN from 90 scans for head-and-neck and brain radiotherapy and validated with 10 independent cases against commercial MAR software. The artifact-free kV CT images following Network 1 and post-processing led to structural similarity (SSIM) of 0.997, and mean-absolute-error (MAE) of 10.2 HU, relative to true kV CT. Network 2 in charge of actual MAR successfully suppressed metal artifacts, relative to commercial MAR, while retaining the detailed imaging information, yielding the SSIM of 0.995 against 0.997 from the commercial MAR.

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

BACKGROUND

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.

PURPOSE

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.

METHODS

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).

RESULTS

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%/3 mm and 2%/2 mm, respectively.

CONCLUSIONS

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.

Improved accuracy of relative electron density and proton stopping power ratio through CycleGAN machine learning

Objective. Kilovoltage computed tomography (kVCT) is the cornerstone of radiotherapy treatment planning for delineating tissues and towards dose calculation. For the former, kVCT provides excellent contrast and signal-to-noise ratio. For the latter, kVCT may have greater uncertainty in determining relative electron density ( ρ e ) and proton stopping power ratio (SPR). Conversely, megavoltage CT (MVCT) may result in superior dose calculation accuracy. The purpose of this work was to convert kVCT HU to MVCT HU using deep learning to obtain higher accuracy ρ e and SPR. Approach. Tissue-mimicking phantoms were created to compare kVCT- and MVCT-determined ρ e and SPR to physical measurements. Using 100 head-and-neck datasets, an unpaired deep learning model was trained to learn the relationship between kVCTs and MVCTs, creating synthetic MVCTs (sMVCTs). Similarity metrics were calculated between kVCTs, sMVCTs, and MVCTs in 20 test datasets. An anthropomorphic head phantom containing bone-mimicking material with known composition was scanned to provide an independent determination of ρ e and SPR accuracy by sMVCT. Main results. In tissue-mimicking bone, ρ e errors were 2.20% versus 0.19% and SPR errors were 4.38% versus 0.22%, for kVCT versus MVCT, respectively. Compared to MVCT, in vivo mean difference (MD) values were 11 and 327 HU for kVCT and 2 and 3 HU for sMVCT in soft tissue and bone, respectively. ρ e MD decreased from 1.3% to 0.35% in soft tissue and 2.9% to 0.13% in bone, for kVCT and sMVCT, respectively. SPR MD decreased from 1.8% to 0.24% in soft tissue and 6.8% to 0.16% in bone, for kVCT and sMVCT, respectively. Relative to physical measurements, ρ e and SPR error in anthropomorphic bone decreased from 7.50% and 7.48% for kVCT to <1% for both MVCT and sMVCT. Significance. Deep learning can be used to map kVCT to sMVCT, suggesting higher accuracy ρ e and SPR is achievable with sMVCT versus kVCT.

Metal artifact reduction through MVCBCT and kVCT in radiotherapy

This study proposes a new method for removal of metal artifacts from megavoltage cone beam computed tomography (MVCBCT) and kilovoltage CT (kVCT) images. Both images were combined to obtain prior image, which was forward projected to obtain surrogate data and replace metal trace in the uncorrected kVCT image. The corrected image was then reconstructed through filtered back projection. A similar radiotherapy plan was designed using the theoretical CT image, the uncorrected kVCT image, and the corrected image. The corrected images removed most metal artifacts, and the CT values were accurate. The corrected image also distinguished the hollow circular hole at the center of the metal. The uncorrected kVCT image did not display the internal structure of the metal, and the hole was misclassified as metal portion. Dose distribution calculated based on the corrected image was similar to that based on the theoretical CT image. The calculated dose distribution also evidently differed between the uncorrected kVCT image and the theoretical CT image. The use of the combined kVCT and MVCBCT to obtain the prior image can distinctly improve the quality of CT images containing large metal implants.

Monte Carlo modeling of ultrasound probes for image guided radiotherapy

PURPOSE

To build Monte Carlo (MC) models of two ultrasound (US) probes and to quantify the effect of beam attenuation due to the US probes for radiation therapy delivered under real-time US image guidance.

METHODS

MC models of two Philips US probes, an X6-1 matrix-array transducer and a C5-2 curved-array transducer, were built based on their megavoltage (MV) CT images acquired in a Tomotherapy machine with a 3.5 MV beam in the EGSnrc, BEAMnrc, and DOSXYZnrc codes. Mass densities in the probes were assigned based on an electron density calibration phantom consisting of cylinders with mass densities between 0.2 and 8.0 g/cm(3). Beam attenuation due to the US probes in horizontal (for both probes) and vertical (for the X6-1 probe) orientation was measured in a solid water phantom for 6 and 15 MV (15 × 15) cm(2) beams with a 2D ionization chamber array and radiographic films at 5 cm depth. The MC models of the US probes were validated by comparison of the measured dose distributions and dose distributions predicted by MC. Attenuation of depth dose in the (15 × 15) cm(2) beams and small circular beams due to the presence of the probes was assessed by means of MC simulations.

RESULTS

The 3.5 MV CT number to mass density calibration curve was found to be linear with R(2) > 0.99. The maximum mass densities in the X6-1 and C5-2 probes were found to be 4.8 and 5.2 g/cm(3), respectively. Dose profile differences between MC simulations and measurements of less than 3% for US probes in horizontal orientation were found, with the exception of the penumbra region. The largest 6% dose difference was observed in dose profiles of the X6-1 probe placed in vertical orientation, which was attributed to inadequate modeling of the probe cable. Gamma analysis of the simulated and measured doses showed that over 96% of measurement points passed the 3%/3 mm criteria for both probes placed in horizontal orientation and for the X6-1 probe in vertical orientation. The X6-1 probe in vertical orientation caused the highest attenuation of the 6 and 15 MV beams, which at 10 cm depth accounted for 33% and 43% decrease compared to the respective (15 × 15) cm(2) open fields. The C5-2 probe in horizontal orientation, on the other hand, caused a dose increase of 10% and 53% for the 6 and 15 MV beams, respectively, in the buildup region at 0.5 cm depth. For the X6-1 probe in vertical orientation, the dose at 5 cm depth for the 3-cm diameter 6 MV and 5-cm diameter 15 MV beams was attenuated compared to the corresponding open fields to a greater degree by 65% and 43%, respectively.

CONCLUSIONS

MC models of two US probes used for real-time image guidance during radiotherapy have been built. Due to the high beam attenuation of the US probes, the authors generally recommend avoiding delivery of treatment beams that intersect the probe. However, the presented MC models can be effectively integrated into US-guided radiotherapy treatment planning in cases for which beam avoidance is not practical due to anatomy geometry.

Metal artifact correction for x-ray computed tomography using kV and selective MV imaging

PURPOSE

The overall goal of this work is to improve the computed tomography (CT) image quality for patients with metal implants or fillings by completing the missing kilovoltage (kV) projection data with selectively acquired megavoltage (MV) data that do not suffer from photon starvation. When both of these imaging systems, which are available on current radiotherapy devices, are used, metal streak artifacts are avoided, and the soft-tissue contrast is restored, even for regions in which the kV data cannot contribute any information.

METHODS

Three image-reconstruction methods, including two filtered back-projection (FBP)-based analytic methods and one iterative method, for combining kV and MV projection data from the two on-board imaging systems of a radiotherapy device are presented in this work. The analytic reconstruction methods modify the MV data based on the information in the projection or image domains and then patch the data onto the kV projections for a FBP reconstruction. In the iterative reconstruction, the authors used dual-energy (DE) penalized weighted least-squares (PWLS) methods to simultaneously combine the kV/MV data and perform the reconstruction.

RESULTS

The authors compared kV/MV reconstructions to kV-only reconstructions using a dental phantom with fillings and a hip-implant numerical phantom. Simulation results indicated that dual-energy sinogram patch FBP and the modified dual-energy PWLS method can successfully suppress metal streak artifacts and restore information lost due to photon starvation in the kV projections. The root-mean-square errors of soft-tissue patterns obtained using combined kV/MV data are 10-15 Hounsfield units smaller than those of the kV-only images, and the structural similarity index measure also indicates a 5%-10% improvement in the image quality. The added dose from the MV scan is much less than the dose from the kV scan if a high efficiency MV detector is assumed.

CONCLUSIONS

The authors have shown that it is possible to improve the image quality of kV CTs for patients with metal implants or fillings by completing the missing kV projection data with selectively acquired MV data that do not suffer from photon starvation. Numerical simulations demonstrated that dual-energy sinogram patch FBP and a modified kV/MV PWLS method can successfully suppress metal streak artifacts and restore information lost due to photon starvation in kV projections. Combined kV/MV images may permit the improved delineation of structures of interest in CT images for patients with metal implants or fillings.

Calibration of CT Hounsfield units for proton therapy treatment planning: use of kilovoltage and megavoltage images and comparison of parameterized methods

Proton beam range is of major concern, in particular, when images used for dose computations are artifacted (for example in patients with surgically treated bone tumors). We investigated several conditions and methods for determination of computed tomography Hounsfield unit (CT-HU) calibration curves, using two different conversion schemes. A stoichiometric methodology was used on either kilovoltage (kV) or megavoltage (MV) CT images and the accuracy of the calibration methods was evaluated. We then studied the effects of metal artifacts on proton dose distributions using metallic implants in rigid phantom mimicking clinical conditions. MV-CT images were used to evaluate relative proton stopping power in certain high density implants, and a methodology is proposed for accurate delineation and dose calculation, using a combined set of kV- and MV-CT images. Our results show good agreement between measurements and dose calculations or relative proton stopping power determination (<5%). The results also show that range uncertainty increases when only kV-CT images are used or when no correction is made on artifacted images. However, differences between treatment plans calculated on corrected kV-CT data and MV-CT data remained insignificant in the investigated patient case, even with streak artifacts and volume effects that reduce the accuracy of manual corrections.