Synthesize CT Images Research Articles

CBCT-to-CT Synthesis for Cervical Cancer Adaptive Radiotherapy via U-Net-Based Model Hierarchically Trained with Hybrid Dataset.

To develop a deep learning framework based on a hybrid dataset to enhance the quality of CBCT images and obtain accurate HU values. A total of 228 cervical cancer patients treated in different LINACs were enrolled. We developed an encoder-decoder architecture with residual learning and skip connections. The model was hierarchically trained and validated on 5279 paired CBCT/planning CT images and tested on 1302 paired images. The mean absolute error (MAE), peak signal to noise ratio (PSNR), and structural similarity index (SSIM) were utilized to access the quality of the synthetic CT images generated by our model. The MAE between synthetic CT images generated by our model and planning CT was 10.93 HU, compared to 50.02 HU for the CBCT images. The PSNR increased from 27.79 dB to 33.91 dB, and the SSIM increased from 0.76 to 0.90. Compared with synthetic CT images generated by the convolution neural networks with residual blocks, our model had superior performance both in qualitative and quantitative aspects. Our model could synthesize CT images with enhanced image quality and accurate HU values. The synthetic CT images preserved the edges of tissues well, which is important for downstream tasks in adaptive radiotherapy.

MSE-Fusion: Weakly supervised medical image fusion with modal synthesis and enhancement

Existing multi-modal image fusion methods utilize multi-modal images as input that require multiple imaging of patients causing harm to patients’ bodies and large costs, moreover, image fusion needs a large number of registered images which is time-consuming and difficult to get, and has unclear texture and structure of the fused images. Therefore, a weakly supervised medical image fusion method with modal synthesis and enhancement is proposed. In modal synthesis, a weakly supervised approach is used to train the model to decrease the requirements of registered images, and MR images are used as input to synthesize CT images through a deep-structure and shallow-detail generator by training to reduce the required input modal and make the texture and structure clearer. In image enhancement, MR images are passed through a trained generator to generate enhanced MR images which enhance the texture and structure of the MR images. And then using the synthesized CT and enhanced MR images together with the original PET images as input to achieve tri-modal image fusion. Compared with 13 state-of-the-art modal synthesis and image fusion methods on the same datasets, the performance of the proposed method on 7 objective evaluation metrics is significantly improved. The subjective visual effect and objective evaluation metrics of our method are better than those of the compared image fusion methods.

Synthetic CT generation from non-attenuation corrected PET images for whole-body PET imaging

Attenuation correction (AC) of PET/MRI faces challenges including inter-scan motion, image artifacts such as truncation and distortion, and erroneous transformation of structural voxel-intensities to PET mu-map values. We propose a deep-learning-based method to derive synthetic CT (sCT) images from non-attenuation corrected PET (NAC PET) images for AC on whole-body PET/MRI imaging.A 3D cycle-consistent generative adversarial networks (CycleGAN) framework was employed to synthesize CT images from NAC PET. The method learns a transformation that minimizes the difference between sCT, generated from NAC PET, and true CT. It also learns an inverse transformation such that cycle NAC PET image generated from the sCT is close to true NAC PET image. A self-attention strategy was also utilized to identify the most informative component and mitigate the disturbance of noise. We conducted a retrospective study on a total of 119 sets of whole-body PET/CT, with 80 sets for training and 39 sets for testing and evaluation.The whole-body sCT images generated with proposed method demonstrate great resemblance to true CT images, and show good contrast on soft tissue, lung and bony tissues. The mean absolute error (MAE) of sCT over true CT is less than 110 HU. Using sCT for whole-body PET AC, the mean error of PET quantification is less than 1% and normalized mean square error (NMSE) is less than 1.4%. Average normalized cross correlation on whole body is close to one, and PSNR is larger than 42 dB.We proposed a deep learning-based approach to generate sCT from whole-body NAC PET for PET AC. sCT generated with proposed method shows great similarity to true CT images both qualitatively and quantitatively, and demonstrates great potential for whole-body PET AC in the absence of structural information.

A preliminary study of using a deep convolution neural network to generate synthesized CT images based on CBCT for adaptive radiotherapy of nasopharyngeal carcinoma

This study aims to utilize a deep convolutional neural network (DCNN) for synthesized CT image generation based on cone-beam CT (CBCT) and to apply the images to dose calculations for nasopharyngeal carcinoma (NPC).An encoder-decoder 2D U-Net neural network was produced. A total of 70 CBCT/CT paired images of NPC cancer patients were used for training (50), validation (10) and testing (10) datasets. The testing datasets were treated with the same prescription dose (70 Gy to PTVnx70, 68 Gy to PTVnd68, 62 Gy to the PTV62 and 54 Gy to the PTV54). The mean error (ME) and mean absolute error (MAE) for the true CT images were calculated for image quality evaluation of the synthesized CT. The dose-volume histogram (DVH) dose metric difference and 3D gamma pass rate for the true CT images were calculated for dose analysis, and the results were compared with those for the CBCT images (original CBCT images without any correction) and a patient-specific calibration (PSC) method.Compared with CBCT, the range of the MAE for synthesized CT images improved from (60, 120) to (6, 27) Hounsfield units (HU), and the ME improved from (−74, 51) to (−26, 4) HU. Compared with the true CT method, the average DVH dose metric differences for the CBCT, PSC and synthesized CT methods were 0.8% ± 1.9%, 0.4% ± 0.7% and 0.2% ± 0.6%, respectively. The 1%/1 mm gamma pass rates within the body for the CBCT, PSC and synthesized CT methods were 90.8% ± 6.2%, 94.1% ± 4.4% and 95.5% ± 1.6%, respectively, and the rates within the PTVnx70 were 80.3% ± 16.6%, 87.9% ± 19.7%, 98.6% ± 2.9%, respectively.The DCNN model can generate high-quality synthesized CT images from CBCT images and be used for accurate dose calculations for NPC patients. This finding has great significance for the clinical application of adaptive radiotherapy for NPC.

PET attenuation correction using synthetic CT from ultrashort echo-time MR imaging.

Integrated PET/MR systems are becoming increasingly popular in clinical and research applications. Quantitative PET reconstruction requires correction for γ-photon attenuations using an attenuation coefficient map (μ map) that is a measure of the electron density. One challenge of PET/MR, in contrast to PET/CT, lies in the accurate computation of μ maps. Unlike CT, MR imaging measures physical properties not directly related to electron density. Previous approaches have computed the attenuation coefficients using a segmentation of MR images or using deformable registration of atlas CT images to the space of the subject MR images. In this work, we propose a patch-based method to generate whole-head μ maps from ultrashort echo-time (UTE) MR imaging sequences. UTE images are preferred to other MR sequences because of the increased signal from bone. To generate a synthetic CT image, we use patches from a reference dataset, which consists of dual-echo UTE images and a coregistered CT scan from the same subject. Matching of patches between the reference and target images allows corresponding patches from the reference CT scan to be combined via a Bayesian framework. No registration or segmentation is required. For evaluation, UTE, CT, and PET data acquired from 5 patients under an institutional review board-approved protocol were used. Another patient (with UTE and CT data only) was selected to be the reference to generate synthetic CT images for these 5 patients. PET reconstructions were attenuation-corrected using the original CT, our synthetic CT, Siemens Dixon-based μ maps, Siemens UTE-based μ maps, and deformable registration-based CT. Our synthetic CT-based PET reconstruction showed higher correlation (average ρ = 0.996, R(2) = 0.991) to the original CT-based PET, as compared with the segmentation- and registration-based methods. Synthetic CT-based reconstruction had minimal bias (regression slope, 0.990), as compared with the segmentation-based methods (regression slope, 0.905). A peak signal-to-noise ratio of 35.98 dB in the reconstructed PET activity was observed, compared with 29.767, 29.34, and 27.43 dB for the Siemens Dixon-, UTE-, and registration-based μ maps. A patch-matching approach to synthesize CT images from dual-echo UTE images leads to significantly improved accuracy of PET reconstruction as compared with actual CT scans. The PET reconstruction is improved over segmentation- (Dixon and Siemens UTE) and registration-based methods, even in subjects with pathologic findings.

Modeling respiratory motion for reducing motion artifacts in 4D CT images

Four-dimensional computed tomography (4D CT) images have been recently adopted in radiation treatment planning for thoracic and abdominal cancers to explicitly define respiratory motion and anatomy deformation. However, significant image distortions (artifacts) exist in 4D CT images that may affect accurate tumor delineation and the shape representation of normal anatomy. In this study, the authors present a patient-specific respiratory motion model, based on principal component analysis (PCA) of motion vectors obtained from deformable image registration, with the main goal of reducing image artifacts caused by irregular motion during 4D CT acquisition. For a 4D CT image set of a specific patient, the authors calculated displacement vector fields relative to a reference phase, using an in-house deformable image registration method. The authors then used PCA to decompose each of the displacement vector fields into linear combinations of principal motion bases. The authors have demonstrated that the regular respiratory motion of a patient can be accurately represented by a subspace spanned by three principal motion bases and their projections. These projections were parameterized using a spline model to allow the reconstruction of the displacement vector fields at any given phase in a respiratory cycle. Finally, the displacement vector fields were used to deform the reference CT image to synthesize CT images at the selected phase with much reduced image artifacts. The authors evaluated the performance of the in-house deformable image registration method using benchmark datasets consisting of ten 4D CT sets annotated with 300 landmark pairs that were approved by physicians. The initial large discrepancies across the landmark pairs were significantly reduced after deformable registration, and the accuracy was similar to or better than that reported by state-of-the-art methods. The proposed motion model was quantitatively validated on 4D CT images of a phantom and a lung cancer patient by comparing the synthesized images and the original images at different phases. The synthesized images matched well with the original images. The motion model was used to reduce irregular motion artifacts in the 4D CT images of three lung cancer patients. Visual assessment indicated that the proposed approach could reduce severe image artifacts. The shape distortions around the diaphragm and tumor regions were mitigated in the synthesized 4D CT images. The authors have derived a mathematical model to represent the regular respiratory motion from a patient-specific 4D CT set and have demonstrated its application in reducing irregular motion artifacts in 4D CT images. The authors' approach can mitigate shape distortions of anatomy caused by irregular breathing motion during 4D CT acquisition.

SU‐E‐J‐121: A New 4D Radiotherapy Planning Strategy Using Synthesized Tumor‐Motion‐Compensated Computed Tomography

To validate a four-dimensional radiotherapy (4DRT) planning based on a synthesized CT image compensating for tumor motion, accounting for tumor rotation, deformation and distortion due to motion artifacts, and producing realistic normal tissue density in motion-tracking beam eye view. 4D computed tomography (4DCT) images of six patients with peripheral lung lesions in mid or lower lobs (motion range: 0.5-3.5cm and size: 1.5±2.0 cm3 ) were used. A customized program was used for simulating the patient anatomy with a motion-compensated tumor using 4DCT by aligning the tumor and averaging the 4DCT into a static 3.5DCT image. The gross tumor volume (GTV) was delineated semiautomatically using a threshold algorithm. Variation of GTV in each phase CT was assessed across all phases. 3DRT plans were generated using 3.5DCT and 4DCT and compared for validation. An integrated dose volume histogram (iDVH) from all phase plans and dose warping using deformable image registration (DIR) were used for evaluating 4D plans and comparing with 3.5D plans. The range of tumor volume variation over the mean within a breathing cycle was 87%±46%. The 3.5DCT produced an 'averaged' GTV, more reliable than that from any phase CT. The results show that the 3.5D plan is equivalent to the 4D plan, except for low dose area, using the iDVH evaluation. On average, the percentage difference for the areas under the DVH and iDVH is 4.3%±2.7%, while 2/3 of the difference results from low dose region blow D20%. Using DIR-based dose warping, PTV coverage varies due to DIR uncertainty for the small lesions. The 3.5D plan is equivalent to the 4D plan for peripheral lung lesions, yet requires much less clinical workload. The 3.5D plan accounts for tumor motion and tumor variation for a more reliable delineation, and for realistic normal tissue representation for motion tracking.