A more effective CT synthesizer using transformers for cone-beam CT-guided adaptive radiotherapy

Kerun

et al. 2023

Sci Rep

This study aims to utilize a hybrid approach of phantom correction and deep learning for synthesized CT (sCT) images generation based on cone-beam CT (CBCT) images for nasopharyngeal carcinoma (NPC). 52 CBCT/CT paired images of NPC patients were used for model training (41), validation (11). Hounsfield Units (HU) of the CBCT images was calibrated by a commercially available CIRS phantom. Then the original CBCT and the corrected CBCT (CBCT_cor) were trained separately with the same cycle generative adversarial network (CycleGAN) to generate SCT1 and SCT2. The mean error and mean absolute error (MAE) were used to quantify the image quality. For validations, the contours and treatment plans in CT images were transferred to original CBCT, CBCT_cor, SCT1 and SCT2 for dosimetric comparison. Dose distribution, dosimetric parameters and 3D gamma passing rate were analyzed. Compared with rigidly registered CT (RCT), the MAE of CBCT, CBCT_cor, SCT1 and SCT2 were 346.11 ± 13.58 HU, 145.95 ± 17.64 HU, 105.62 ± 16.08 HU and 83.51 ± 7.71 HU, respectively. Moreover, the average dosimetric parameter differences for the CBCT_cor, SCT1 and SCT2 were 2.7% ± 1.4%, 1.2% ± 1.0% and 0.6% ± 0.6%, respectively. Using the dose distribution of RCT images as reference, the 3D gamma passing rate of the hybrid method was significantly better than the other methods. The effectiveness of CBCT-based sCT generated using CycleGAN with HU correction for adaptive radiotherapy of nasopharyngeal carcinoma was confirmed. The image quality and dose accuracy of SCT2 were outperform the simple CycleGAN method. This finding has great significance for the clinical application of adaptive radiotherapy for NPC.

Section: Discussionsupporting

confidence: 69%

CBCT-based synthetic CT generated using CycleGAN with HU correction for adaptive radiotherapy of nasopharyngeal carcinoma

Kerun

et al. 2023

Sci Rep

“…Image quality and dose distribution were evaluated on the CBCT images corrected by HU-ED curves and sCT images generated by CycleGAN model, with RCT images as ground truth.As the result of side-by-side comparison shown, sCT images generated by CycleGAN model removed most scatter artifacts on the CBCT images. The image quality of SCT1 and SCT2 was visually comparable to the RCT, as Zhang et al 28 and Chen et al29 reported. Moreover, the HU pro le of STC images in different regions, especially SCT2, was closer to that of RCT images.…”

supporting

confidence: 71%

CBCT-based synthetic CT generation using a hybrid approach for adaptive radiotherapy of nasopharyngeal carcinoma

Kerun

et al. 2023

Preprint

Objective: This study aims to utilize a hybrid approach of phantom correction and deep learning for synthesized CT (sCT) images generation based on cone-beam CT (CBCT) images for nasopharyngeal carcinoma (NPC). Methods: A total of 52 CBCT/CT paired images of NPC patients were used for training (41), validation (11) datasets. Hounsfield Units (HU) of the CBCT images was corrected by a commercial CIRS phantom. Then the original CBCT and the corrected CBCT (CBCT_cor) were trained separately with the some cycle generative adversarial network (CycleGAN) to generate SCT1 and SCT2. The mean error (ME) and mean absolute error (MAE) were used to quantify the image quality. For the patients in the validation datasets, the contours and treatment plans in CT images were transferred to original CBCT, CBCT_cor, SCT1 and SCT2 for dosimetric comparison. Finally, dose distribution, dosimetric parameters and 3D gamma pass rate were analyzed. Results: Compared with rigidly registered CT (RCT), the MAE of CBCT, CBCT_cor, SCT1 and SCT2 were 346.11 ± 13.58HU, 145.95 ± 17.64HU, 105.62 ± 16.08HU and 83.51 ± 7.71HU, respectively. Moreover, the average dosimetric parameter differences for the CBCT_cor, SCT1 and SCT2 were 2.7% ± 1.4%, 1.2% ± 1.0% and 0.6% ± 0.6%, respectively. Using the dose distribution of RCT images as reference, the 3D gamma pass rate of the hybrid method was significantly better than the other methods. Conclusion: A novel hybrid approach based on HU-ED correction and CycleGAN was developed to generate sCT images for CBCT images of NPC patients. The image quality and dose accuracy of the hybrid approach were outperform the simple CycleGAN method. This finding has great significance for the clinical application of adaptive radiotherapy for NPC.

“…Finally, as the main deep learning baseline we implemented a 3D U-net for post-processing the FBP output and compared it to a two-dimensional post-processing baseline using a more recent Uformer model, 28 variants of which have lately been used for CBCT postprocessing in the literature. 33,34 We used U-net with 3…”

Section: Baseline Methodsmentioning

confidence: 99%

“…Finally, as the main deep learning baseline we implemented a 3D U‐net for post‐processing the FBP output and compared it to a two‐dimensional post‐processing baseline using a more recent Uformer model, 28 variants of which have lately been used for CBCT post‐processing in the literature 33,34 . We used U‐net with 3 downsampling layers, PReLU activations, valid convolutions and 64 base filters, similar to the original 3D U‐net, 27 but without Instance or Batch normalization layers.…”

Section: Methodsmentioning

confidence: 99%

End‐to‐end memory‐efficient reconstruction for cone beam CT

Moriakov,

Sonke,

Teuwen

2023

Medical Physics

BackgroundCone beam computed tomography (CBCT) plays an important role in many medical fields nowadays. Unfortunately, the potential of this imaging modality is hampered by lower image quality compared to the conventional CT, and producing accurate reconstructions remains challenging. A lot of recent research has been directed towards reconstruction methods relying on deep learning, which have shown great promise for various imaging modalities. However, practical application of deep learning to CBCT reconstruction is complicated by several issues, such as exceedingly high memory costs of deep learning methods when working with fully 3D data. Additionally, deep learning methods proposed in the literature are often trained and evaluated only on data from a specific region of interest, thus raising concerns about possible lack of generalization to other regions.PurposeIn this work, we aim to address these limitations and propose LIRE: a learned invertible primal‐dual iterative scheme for CBCT reconstruction.MethodsLIRE is a learned invertible primal‐dual iterative scheme for CBCT reconstruction, wherein we employ a U‐Net architecture in each primal block and a residual convolutional neural network (CNN) architecture in each dual block. Memory requirements of the network are substantially reduced while preserving its expressive power through a combination of invertible residual primal‐dual blocks and patch‐wise computations inside each of the blocks during both forward and backward pass. These techniques enable us to train on data with isotropic 2 mm voxel spacing, clinically‐relevant projection count and detector panel resolution on current hardware with 24 GB video random access memory (VRAM).ResultsTwo LIRE models for small and for large field‐of‐view (FoV) setting were trained and validated on a set of 260 + 22 thorax CT scans and tested using a set of 142 thorax CT scans plus an out‐of‐distribution dataset of 79 head and neck CT scans. For both settings, our method surpasses the classical methods and the deep learning baselines on both test sets. On the thorax CT set, our method achieves peak signal‐to‐noise ratio (PSNR) of 33.84 ± 2.28 for the small FoV setting and 35.14 ± 2.69 for the large FoV setting; U‐Net baseline achieves PSNR of 33.08 ± 1.75 and 34.29 ± 2.71 respectively. On the head and neck CT set, our method achieves PSNR of 39.35 ± 1.75 for the small FoV setting and 41.21 ± 1.41 for the large FoV setting; U‐Net baseline achieves PSNR of 33.08 ± 1.75 and 34.29 ± 2.71 respectively. Additionally, we demonstrate that LIRE can be finetuned to reconstruct high‐resolution CBCT data with the same geometry but 1 mm voxel spacing and higher detector panel resolution, where it outperforms the U‐Net baseline as well.ConclusionsLearned invertible primal‐dual schemes with additional memory optimizations can be trained to reconstruct CBCT volumes directly from the projection data with clinically‐relevant geometry and resolution. Such methods can offer better reconstruction quality and generalization compared to classical deep learning baselines.