Metal implant segmentation in CT images based on diffusion model.

BACKGROUND: Computed tomography (CT) is widely in clinics and is affected by metal implants. Metal segmentation is crucial for metal artifact correction, and the common threshold method often fails to accurately segment metals. PURPOSE: This study aims to segment metal implants in CT images using a diffusion model and further validate it with clinical artifact images and phantom images of known size. METHODS: A retrospective study was conducted on 100 patients who received radiation therapy without metal artifacts, and simulated artifact data were generated using publicly available mask data. The study utilized 11,280 slices for training and verification, and 2,820 slices for testing. Metal mask segmentation was performed using DiffSeg, a diffusion model incorporating conditional dynamic coding and a global frequency parser (GFParser). Conditional dynamic coding fuses the current segmentation mask and prior images at multiple scales, while GFParser helps eliminate high-frequency noise in the mask. Clinical artifact images and phantom images are also used for model validation. RESULTS: Compared with the ground truth, the accuracy of DiffSeg for metal segmentation of simulated data was 97.89% and that of DSC was 95.45%. The mask shape obtained by threshold segmentation covered the ground truth and DSCs were 82.92% and 84.19% for threshold segmentation based on 2500 HU and 3000 HU. Evaluation metrics and visualization results show that DiffSeg performs better than other classical deep learning networks, especially for clinical CT, artifact data, and phantom data. CONCLUSION: DiffSeg efficiently and robustly segments metal masks in artifact data with conditional dynamic coding and GFParser. Future work will involve embedding the metal segmentation model in metal artifact reduction to improve the reduction effect.

DiffRecon: Diffusion-based CT reconstruction with cross-modal deformable fusion for DR-guided non-coplanar radiotherapy.

In non-coplanar radiotherapy, DR is commonly used for image guiding which needs to fuse intraoperative DR with preoperative CT. But this fusion task performs poorly, suffering from unaligned and dimensional differences between DR and CT. CT reconstruction estimated from DR could facilitate this challenge. Thus, We propose a unified generation and registration framework, named DiffRecon, for intraoperative CT reconstruction based on DR using the diffusion model. Specifically, we use the generation model for synthesizing intraoperative CTs to eliminate dimensional differences and the registration model for aligning synthetic CTs to improve reconstruction. To ensure clinical usability, CT is not only estimated from DR but the preoperative CT is also introduced as prior. We design a dual-encoder to learn prior knowledge and spatial deformation among pre- and intra-operative CT pairs and DR parallelly for 2D/3D feature deformable conversion. To calibrate the cross-modal fusion, we insert cross-attention modules to enhance the 2D/3D feature interaction between dual encoders. DiffRecon has been evaluated by both image quality metrics and dosimetric indicators. The high image synthesis metrics are with RMSE of 0.02±0.01, PSNR of 44.92±3.26, and SSIM of 0.994±0.003. The mean gamma passing rates between rCT and sCT for 1%/1 mm, 2%/2 mm and 3%/3 mm acceptance criteria are 95.2%, 99.4% and 99.9% respectively. The proposed DiffRecon can reconstruct CT accurately from a single DR projection with excellent image generation quality and dosimetric accuracy. These demonstrate that the method can be applied in non-coplanar adaptive radiotherapy workflows.

Uncertainty estimation- and attention-based semi-supervised models for automatically delineate clinical target volume in CBCT images of breast cancer.

OBJECTIVES: Accurate segmentation of the clinical target volume (CTV) of CBCT images can observe the changes of CTV during patients' radiotherapy, and lay a foundation for the subsequent implementation of adaptive radiotherapy (ART). However, segmentation is challenging due to the poor quality of CBCT images and difficulty in obtaining target volumes. An uncertainty estimation- and attention-based semi-supervised model called residual convolutional block attention-uncertainty aware mean teacher (RCBA-UAMT) was proposed to delineate the CTV in cone-beam computed tomography (CBCT) images of breast cancer automatically. METHODS: A total of 60 patients who undergone radiotherapy after breast-conserving surgery were enrolled in this study, which involved 60 planning CTs and 380 CBCTs. RCBA-UAMT was proposed by integrating residual and attention modules in the backbone network 3D UNet. The attention module can adjust channel and spatial weights of the extracted image features. The proposed design can train the model and segment CBCT images with a small amount of labeled data (5%, 10%, and 20%) and a large amount of unlabeled data. Four types of evaluation metrics, namely, dice similarity coefficient (DSC), Jaccard, average surface distance (ASD), and 95% Hausdorff distance (95HD), are used to assess the model segmentation performance quantitatively. RESULTS: The proposed method achieved average DSC, Jaccard, 95HD, and ASD of 82%, 70%, 8.93, and 1.49 mm for CTV delineation on CBCT images of breast cancer, respectively. Compared with the three classical methods of mean teacher, uncertainty-aware mean-teacher and uncertainty rectified pyramid consistency, DSC and Jaccard increased by 7.89-9.33% and 14.75-16.67%, respectively, while 95HD and ASD decreased by 33.16-67.81% and 36.05-75.57%, respectively. The comparative experiment results of the labeled data with different proportions (5%, 10% and 20%) showed significant differences in the DSC, Jaccard, and 95HD evaluation indexes in the labeled data with 5% versus 10% and 5% versus 20%. Moreover, no significant differences were observed in the labeled data with 10% versus 20% among all evaluation indexes. Therefore, we can use only 10% labeled data to achieve the experimental objective. CONCLUSIONS: Using the proposed RCBA-UAMT, the CTV of breast cancer CBCT images can be delineated reliably with a small amount of labeled data. These delineated images can be used to observe the changes in CTV and lay the foundation for the follow-up implementation of ART.

[Research on Position Verification of Multi-Leaf Collimator (MLC) and Dose Verification Based on Electronic Portal Imaging Device].

Objective: A quality control (QC) system based on the electronic portal imaging device (EPID) system was used to realize the Multi-Leaf Collimator (MLC) position verification and dose verification functions on Primus and VenusX accelerators. Methods: The MLC positions were calculated by the maximum gradient method of gray values to evaluate the deviation. The dose of images acquired by EPID were reconstructed using the algorithm combining dose calibration and dose calculation. The dose data obtained by EPID and two-dimensional matrix (MapCheck/PTW) were compared with the dose calculated by Pinnacle/TiGRT TPS for γ passing rate analysis. Results: The position error of VenusX MLC was less than 1 mm. The position error of Primus MLC was significantly reduced after being recalibrated under the instructions of EPID. For the dose reconstructed by EPID, the average γ passing rates of Primus were 98.86% and 91.39% under the criteria of 3%/3 mm, 10% threshold and 2%/2 mm, 10% threshold, respectively. The average γ passing rates of VenusX were 98.49% and 91.11%, respectively. Conclusion: The EPID-based accelerator quality control system can improve the efficiency of accelerator quality control and reduce the workload of physicists.

Prognosis Prediction of Diffuse Large B-Cell Lymphoma in 18F-FDG PET Images Based on Multi-Deep-Learning Models.

Diffuse large B-cell lymphoma (DLBCL), a cancer of B cells, has been one of the most challenging and complicated diseases because of its considerable variation in clinical behavior, response to therapy, and prognosis. Radiomic features from medical images, such as PET images, have become one of the most valuable features for disease classification or prognosis prediction using learning-based methods. In this paper, a new flexible ensemble deep learning model is proposed for the prognosis prediction of the DLBCL in 18F-FDG PET images. This study proposes the multi-R-signature construction through selected pre-trained deep learning models for predicting progression-free survival (PFS) and overall survival (OS). The proposed method is trained and validated on two datasets from different imaging centers. Through analyzing and comparing the results, the prediction models, including Age, Ann abor stage, Bulky disease, SUVmax, TMTV, and multi-R-signature, achieve the almost best PFS prediction performance (C-index: 0.770, 95% CI: 0.705-0.834, with feature adding fusion method and C-index: 0.764, 95% CI: 0.695-0.832, with feature concatenate fusion method) and OS prediction (C-index: 0.770 (0.692-0.848) and 0.771 (0.694-0.849)) on the validation dataset. The developed multiparametric model could achieve accurate survival risk stratification of DLBCL patients. The outcomes of this study will be helpful for the early identification of high-risk DLBCL patients with refractory relapses and for guiding individualized treatment strategies.

A 4D-CBCT correction network based on contrastive learning for dose calculation in lung cancer.

OBJECTIVE: This study aimed to present a deep-learning network called contrastive learning-based cycle generative adversarial networks (CLCGAN) to mitigate streak artifacts and correct the CT value in four-dimensional cone beam computed tomography (4D-CBCT) for dose calculation in lung cancer patients. METHODS: 4D-CBCT and 4D computed tomography (CT) of 20 patients with locally advanced non-small cell lung cancer were used to paired train the deep-learning model. The lung tumors were located in the right upper lobe, right lower lobe, left upper lobe, and left lower lobe, or in the mediastinum. Additionally, five patients to create 4D synthetic computed tomography (sCT) for test. Using the 4D-CT as the ground truth, the quality of the 4D-sCT images was evaluated by quantitative and qualitative assessment methods. The correction of CT values was evaluated holistically and locally. To further validate the accuracy of the dose calculations, we compared the dose distributions and calculations of 4D-CBCT and 4D-sCT with those of 4D-CT. RESULTS: The structural similarity index measure (SSIM) and peak signal-to-noise ratio (PSNR) of the 4D-sCT increased from 87% and 22.31 dB to 98% and 29.15 dB, respectively. Compared with cycle consistent generative adversarial networks, CLCGAN enhanced SSIM and PSNR by 1.1% (p < 0.01) and 0.42% (p < 0.01). Furthermore, CLCGAN significantly decreased the absolute mean differences of CT value in lungs, bones, and soft tissues. The dose calculation results revealed a significant improvement in 4D-sCT compared to 4D-CBCT. CLCGAN was the most accurate in dose calculations for left lung (V5Gy), right lung (V5Gy), right lung (V20Gy), PTV (D98%), and spinal cord (D2%), with the relative dose difference were reduced by 6.84%, 3.84%, 1.46%, 0.86%, 3.32% compared to 4D-CBCT. CONCLUSIONS: Based on the satisfactory results obtained in terms of image quality, CT value measurement, it can be concluded that CLCGAN-based corrected 4D-CBCT can be utilized for dose calculation in lung cancer.

GAN-based metal artifacts region inpainting in brain MRI imaging with reflective registration.

BACKGROUND AND OBJECTIVE: Metallic magnetic resonance imaging (MRI) implants can introduce magnetic field distortions, resulting in image distortion, such as bulk shifts and signal-loss artifacts. Metal Artifacts Region Inpainting Network (MARINet), using the symmetry of brain MRI images, has been developed to generate normal MRI images in the image domain and improve image quality. METHODS: T1-weighted MRI images containing or located near the teeth of 100 patients were collected. A total of 9000 slices were obtained after data augmentation. Then, MARINet based on U-Net with a dual-path encoder was employed to inpaint the artifacts in MRI images. The input of MARINet contains the original image and the flipped registered image, with partial convolution used concurrently. Subsequently, we compared PConv with partial convolution, and GConv with gated convolution, SDEdit using a diffusion model for inpainting the artifact region of MRI images. The mean absolute error (MAE) and peak signal-to-noise ratio (PSNR) for the mask were used to compare the results of these methods. In addition, the artifact masks of clinical MRI images were inpainted by physicians. RESULTS: MARINet could directly and effectively inpaint the incomplete MRI images generated by masks in the image domain. For the test results of PConv, GConv, SDEdit, and MARINet, the masked MAEs were 0.1938, 0.1904, 0.1876, and 0.1834, respectively, and the masked PSNRs were 17.39, 17.40, 17.49, and 17.60 dB, respectively. The visualization results also suggest that the network can recover the tissue texture, alveolar shape, and tooth contour. Additionally, for clinical artifact MRI images, MARINet completed the artifact region inpainting task more effectively when compared with other models. CONCLUSIONS: By leveraging the quasi-symmetry of brain MRI images, MARINet can directly and effectively inpaint the metal artifacts in MRI images in the image domain, restoring the tooth contour and detail, thereby enhancing the image quality.

A transformer-based dual-domain network for reconstructing FOV extended cone-beam CT images from truncated sinograms in radiation therapy.

BACKGROUND AND OBJECTIVE: Cone-beam computed tomography (CBCT) is widely used in clinical radiotherapy, but its small field of view (sFOV) limits its application potential. In this study, a transformer-based dual-domain network (dual_swin), which combined image domain restoration and sinogram domain restoration, was proposed for the reconstruction of complete CBCT images with extended FOV from truncated sinograms. METHODS: The planning CT images with large FOV (LFOV) of 330 patients who received radiation therapy were collected. The synthetic CBCT (sCBCT) images with LFOV were generated from CT images by the trained cycleGAN network, and CBCT images with sFOV were obtained through forward projection, projection truncation, and filtered back projection (FBP), comprising the training and test data. The proposed dual_swin includes sinogram domain restoration, image domain restoration, and FBP layer, and the swin transformer blocks were used as the basic feature extraction module in the network to improve the global feature extraction ability. The proposed dual_swin was compared with the image domain method, the sinogram domain method, the U-Net based dual domain network (dual_Unet), and the traditional iterative reconstruction method based on prior image and conjugate gradient least-squares (CGLS) in the test of sCBCT images and clinical CBCT images. The HU accuracy and body contour accuracy of the predicted images by each method were evaluated. RESULTS: The images generated using the CGLS method were fuzzy and obtained the lowest structural similarity (SSIM) among all methods in the test of sCBCT and clinical CBCT images. The predicted images by the image domain methods are quite different from the ground truth and have low accuracy on HU value and body contour. In comparison with image domain methods, sinogram domain methods improved the accuracy of HU value and body contour but introduced secondary artifacts and distorted bone tissue. The proposed dual_swin achieved the highest HU and contour accuracy with mean absolute error (MAE) of 23.0 HU, SSIM of 95.7%, dice similarity coefficient (DSC) of 99.6%, and Hausdorff distance (HD) of 4.1 mm in the test of sCBCT images. In the test of clinical patients, images that were predicted by dual_swin yielded MAE, SSIM, DSC, and HD of 38.2 HU, 91.7%, 99.0%, and 5.4 mm, respectively. The predicted images by the proposed dual_swin has significantly higher accuracy than the other methods (P < 0.05). CONCLUSIONS: The proposed dual_swin can accurately reconstruct FOV extended CBCT images from the truncated sinogram to improve the application potential of CBCT images in radiotherapy.

Inpainting truncated areas of CT images based on generative adversarial networks with gated convolution for radiotherapy.

This study aimed to inpaint the truncated areas of CT images by using generative adversarial networks with gated convolution (GatedConv) and apply these images to dose calculations in radiotherapy. CT images were collected from 100 patients with esophageal cancer under thermoplastic membrane placement, and 85 cases were used for training based on randomly generated circle masks. In the prediction stage, 15 cases of data were used to evaluate the accuracy of the inpainted CT in anatomy and dosimetry based on the mask with a truncated volume covering 40% of the arm volume, and they were compared with the inpainted CT synthesized by U-Net, pix2pix, and PConv with partial convolution. The results showed that GatedConv could directly and effectively inpaint incomplete CT images in the image domain. For the results of U-Net, pix2pix, PConv, and GatedConv, the mean absolute errors for the truncated tissue were 195.54, 196.20, 190.40, and 158.45 HU, respectively. The mean dose of the planning target volume, heart, and lung in the truncated CT was statistically different (p < 0.05) from those of the ground truth CT ([Formula: see text]). The differences in dose distribution between the inpainted CT obtained by the four models and [Formula: see text] were minimal. The inpainting effect of clinical truncated CT images based on GatedConv showed better stability compared with the other models. GatedConv can effectively inpaint the truncated areas with high image quality, and it is closer to [Formula: see text] in terms of image visualization and dosimetry than other inpainting models.

New technique and application of truncated CBCT processing in adaptive radiotherapy for breast cancer.

OBJECTIVE: A generative adversarial network (TCBCTNet) was proposed to generate synthetic computed tomography (sCT) from truncated low-dose cone-beam computed tomography (CBCT) and planning CT (pCT). The sCT was applied to the dose calculation of radiotherapy for patients with breast cancer. METHODS: The low-dose CBCT and pCT images of 80 female thoracic patients were used for training. The CBCT, pCT, and replanning CT (rCT) images of 20 thoracic patients and 20 patients with breast cancer were used for testing. All patients were fixed in the same posture with a vacuum pad. The CBCT images were scanned under the Fast Chest M20 protocol with a 50% reduction in projection frames compared with the standard Chest M20 protocol. Rigid registration was performed between pCT and CBCT, and deformation registration was performed between rCT and CBCT. In the training stage of the TCBCTNet, truncated CBCT images obtained from complete CBCT images by simulation were used. The input of the CBCT→CT generator was truncated CBCT and pCT, and TCBCTNet was applied to patients with breast cancer after training. The accuracy of the sCT was evaluated by anatomy and dosimetry and compared with the generative adversarial network with UNet and ResNet as the generators (named as UnetGAN, ResGAN). RESULTS: The three models could improve the image quality of CBCT and reduce the scattering artifacts while preserving the anatomical geometry of CBCT. For the chest test set, TCBCTNet achieved the best mean absolute error (MAE, 21.18±3.76 HU), better than 23.06±3.90 HU in UnetGAN and 22.47±3.57 HU in ResGAN. When applied to patients with breast cancer, TCBCTNet performance decreased, and MAE was 25.34±6.09 HU. Compared with rCT, sCT by TCBCTNet showed consistent dose distribution and subtle absolute dose differences between the target and the organ at risk. The 3D gamma pass rates were 98.98%±0.64% and 99.69%±0.22% at 2 mm/2% and 3 mm/3%, respectively. Ablation experiments confirmed that pCT and content loss played important roles in TCBCTNet. CONCLUSIONS: High-quality sCT images could be synthesized from truncated low-dose CBCT and pCT by using the proposed TCBCTNet model. In addition, sCT could be used to accurately calculate the dose distribution for patients with breast cancer.

Classification for thyroid nodule using ViT with contrastive learning in ultrasound images.

The lack of representative features between benign nodules, especially level 3 of Thyroid Imaging Reporting and Data System (TI-RADS), and malignant nodules limits diagnostic accuracy, leading to inconsistent interpretation, overdiagnosis, and unnecessary biopsies. We propose a Vision-Transformer-based (ViT) thyroid nodule classification model using contrast learning, called TC-ViT, to improve accuracy of diagnosis and specificity of biopsy recommendations. ViT can explore the global features of thyroid nodules well. Nodule images are used as ROI to enhance the local features of the ViT. Contrast learning can minimize the representation distance between nodules of the same category, enhance the representation consistency of global and local features, and achieve accurate diagnosis of TI-RADS 3 or malignant nodules. The test results achieve an accuracy of 86.9%. The evaluation metrics show that the network outperforms other classical deep learning-based networks in terms of classification performance. TC-ViT can achieve automatic classification of TI-RADS 3 and malignant nodules on ultrasound images. It can also be used as a key step in computer-aided diagnosis for comprehensive analysis and accurate diagnosis. The code will be available at https://github.com/Jiawei217/TC-ViT.

Streaking artifact reduction for CBCT-based synthetic CT generation in adaptive radiotherapy.

BACKGROUND: Cone-beam computed tomography (CBCT) is widely used for daily image guidance in radiation therapy, enhancing the reproducibility of patient setup. However, its application in adaptive radiotherapy (ART) is limited by many imaging artifacts and inaccurate Hounsfield units (HUs). The correction of CBCT image is necessary and of great value for CBCT-based ART. PURPOSE: To explore the synthetic CT (sCT) generation from CBCT images of thorax and abdomen patients, which usually surfer from serious artifacts duo to organ state changes. In this study, a streaking artifact reduction network (SARN) is proposed to reduce artifacts and combine with cycleGAN to generate high-quality sCT images from CBCT and achieve an accurate dose calculation. METHODS: The proposed SARN was trained in a self-supervised manner. Artifact-CT images were generated from planning CT by random deformation and projection replacement, and SARN was trained based on paired artifact-CT and CT images. The planning CT and CBCT images of 260 patients with cancer, including 120 thoracic and 140 abdominal CT scans, were used to train and evaluate neural networks. The CBCT images of another 12 patients in late treatment fractions, which contained large anatomy changes, were also tested by trained models. The trained models include commonly used U-Net, cycleGAN, attention-gated cycleGAN (cycAT), and cascade models combined SARN with cycleGAN or cycAT. The generated sCT images were compared in terms of image quality and dose calculation accuracy. RESULTS: The sCT images generated by SARN combined with cycleGAN and cycAT showed the best image quality, removed the most artifacts, and retained the normal anatomical structure. The SARN+cycleGAN performed best in streaking artifacts removal with the maximum percent integrity uniformity (PIUm ) of 91.0% and minimum standard deviation (SD) of 35.4 HU for delineated artifact regions among all models. The mean absolute error (MAE) of CBCT images in the thorax and abdomen were 71.6 and 55.2 HU, respectively, using planning CT images after deformable registration as ground truth. Compared with CBCT, the thoracic and abdominal sCT images generated by each model had significantly improved image quality with smaller MAE (p < 0.05). The SARN+cycAT obtained the minimum MAEs of 42.5 HU in the thorax while SARN+cycleGAN got the minimum MAEs of 32.0 HU in the abdomen. The sCT generated by U-Net had a remarkably lower anatomical structure accuracy compared with the other models. The thoracic and abdominal sCT images generated by SARN+cycleGAN showed optimal dose calculation accuracy with gamma passing rates (2 mm/2%) of 98.2% and 96.9%, respectively. CONCLUSIONS: The proposed SARN can reduce serious streaking artifacts in CBCT images. The SARN combined with cycleGAN can generate high-quality sCT images with fewer artifacts, high-accuracy HU values, and accurate anatomical structures, thus providing reliable dose calculation in ART.

Inpainting the metal artifact region in MRI images by using generative adversarial networks with gated convolution.

PURPOSE: Magnetic resonance imaging (MRI) plays an important role in clinical diagnosis, but it is susceptible to metal artifacts. The generative adversarial network GatedConv with gated convolution (GC) and contextual attention (CA) was used to inpaint the metal artifact region in MRI images. METHODS: MRI images containing or near the teeth of 70 patients were collected, and the scanning sequence was a T1-weighted high-resolution isotropic volume examination sequence. A total of 10 000 slices were obtained after data enhancement, of which 8000 slices were used for training. MRI images were normalized to [-1,1]. Based on the randomly generated mask, U-Net, pix2pix, PConv with partial convolution, and GatedConv were used to inpaint the artifact region of MRI images. The mean absolute error (MAE) and peak signal-to-noise ratio (PSNR) for the mask were used to compare the results of these methods. The inpainting effect on the test dataset using dental masks was also evaluated. Besides, the artifact area of clinical MRI images was inpainted based on the mask sketched by physicians. Finally, the earring artifacts and artifacts caused by abnormal signal foci were inpainted to verify the generalization of the models. RESULTS: GatedConv could directly and effectively inpaint the incomplete MRI images generated by masks in the image domain. For the results of U-Net, pix2pix, PConv, and GatedConv, the masked MAEs were 0.1638, 0.1812, 0.1688, and 0.1596, respectively, and the masked PSNRs were 18.2136, 17.5692, 18.2258, and 18.3035 dB, respectively. Using dental masks, the results of U-Net, pix2pix, and PConv differed more from the real images in terms of alveolar shape and surrounding tissue compared with GatedConv. GatedConv could inpaint the metal artifact region in clinical MRI images more effectively than the other models, but the increase in the mask area could reduce the inpainting effect. Inpainted MRI images by GatedConv and CT images with metal artifact reduction coincided with alveolar and tissue structure, and GatedConv could successfully inpaint artifacts caused by abnormal signal foci, whereas the other models failed. The ablation study demonstrated that GC and CA increased the reliability of the inpainting performance of GatedConv. CONCLUSION: MRI images are affected by metal, and signal void areas appear near metal. GatedConv can inpaint the MRI metal artifact region in the image domain directly and effectively and improve image quality. Medical image inpainting by GatedConv has potential value for tasks, such as positron emission tomography (PET) attenuation correction in PET/MRI and adaptive radiotherapy of synthetic CT based on MRI.

TNSNet: Thyroid nodule segmentation in ultrasound imaging using soft shape supervision.

BACKGROUND AND OBJECTIVES: Thyroid nodules are a common disorder of the endocrine system. Segmentation of thyroid nodules on ultrasound images is an important step in the evaluation and diagnosis of nodules and an initial step in computer-aided diagnostic systems. The accuracy and consistency of segmentation remain a challenge due to the low contrast, speckle noise, and low resolution of ultrasound images. Therefore, the study of deep learning-based algorithms for thyroid nodule segmentation is important. This study utilizes soft shape supervision to improve the performance of detection and segmentation of boundaries of nodules. Soft shape supervision can emphasize the boundary features and assist the network in segmenting nodules accurately. METHODS: We propose a dual-path convolution neural network, including region and shape paths, which use DeepLabV3+ as the backbone. Soft shape supervision blocks are inserted between the two paths to implement cross-path attention mechanisms. The blocks enhance the representation of shape features and add them to the region path as auxiliary information. Thus, the network can accurately detect and segment thyroid nodules. RESULTS: We collect 3786 ultrasound images of thyroid nodules to train and test our network. Compared with the ground truth, the test results achieve an accuracy of 95.81% and a DSC of 85.33. The visualization results also suggest that the network has learned clear and accurate boundaries of the nodules. The evaluation metrics and visualization results demonstrate the superior segmentation performance of the network to other classical deep learning-based networks. CONCLUSIONS: The proposed dual-path network can accurately realize automatic segmentation of thyroid nodules on ultrasound images. It can also be used as an initial step in computer-aided diagnosis. It shows superior performance to other classical methods and demonstrates the potential for accurate segmentation of nodules in clinical applications.

Generating synthetic CT from low-dose cone-beam CT by using generative adversarial networks for adaptive radiotherapy.

OBJECTIVE: To develop high-quality synthetic CT (sCT) generation method from low-dose cone-beam CT (CBCT) images by using attention-guided generative adversarial networks (AGGAN) and apply these images to dose calculations in radiotherapy. METHODS: The CBCT/planning CT images of 170 patients undergoing thoracic radiotherapy were used for training and testing. The CBCT images were scanned under a fast protocol with 50% less clinical projection frames compared with standard chest M20 protocol. Training with aligned paired images was performed using conditional adversarial networks (so-called pix2pix), and training with unpaired images was carried out with cycle-consistent adversarial networks (cycleGAN) and AGGAN, through which sCT images were generated. The image quality and Hounsfield unit (HU) value of the sCT images generated by the three neural networks were compared. The treatment plan was designed on CT and copied to sCT images to calculated dose distribution. RESULTS: The image quality of sCT images by all the three methods are significantly improved compared with original CBCT images. The AGGAN achieves the best image quality in the testing patients with the smallest mean absolute error (MAE, 43.5 ± 6.69), largest structural similarity (SSIM, 93.7 ± 3.88) and peak signal-to-noise ratio (PSNR, 29.5 ± 2.36). The sCT images generated by all the three methods showed superior dose calculation accuracy with higher gamma passing rates compared with original CBCT image. The AGGAN offered the highest gamma passing rates (91.4 ± 3.26) under the strictest criteria of 1 mm/1% compared with other methods. In the phantom study, the sCT images generated by AGGAN demonstrated the best image quality and the highest dose calculation accuracy. CONCLUSIONS: High-quality sCT images were generated from low-dose thoracic CBCT images by using the proposed AGGAN through unpaired CBCT and CT images. The dose distribution could be calculated accurately based on sCT images in radiotherapy.

CrossModalNet: exploiting quality preoperative images for multimodal image registration.

A long-standing problem in image-guided radiotherapy is that inferior intraoperative images present a difficult problem for automatic registration algorithms. Particularly for digital radiography (DR) and digitally reconstructed radiograph (DRR), the blurred, low-contrast, and noisy DR makes the multimodal registration of DR-DRR challenging. Therefore, we propose a novel CNN-based method called CrossModalNet to exploit the quality preoperative modality (DRR) for handling the limitations of intraoperative images (DR), thereby improving the registration accuracy. The method consists of two parts: DR-DRR contour predictions and contour-based rigid registration. We have designed the CrossModal Attention Module and CrossModal Refine Module to fully exploit the multiscale crossmodal features and implement the crossmodal interactions during the feature encoding and decoding stages. Then, the predicted anatomical contours of DR-DRR are registered by the classic mutual information method. We collected 2486 patient scans to train CrossModalNet and 170 scans to test its performance. The results show that it outperforms the classic and state-of-the-art methods with 95th percentile Hausdorff distance of 5.82 pixels and registration accuracy of 81.2%. The code is available at https://github.com/lc82111/crossModalNet.

A study on the correlation between radiation field size and gamma index passing rate for MatriXX.

This study aimed to analyze the influence of the radiation field size on the passing rate of the treatment planning system using MatriXX if the field irradiated the circuit.Two sets of static fields which were 10 cm and 30 cm in the left-right direction (X), and was 31 cm to 40 cm in gun-target direction (Y) were designed. In these fields, the gantry was 0 and the monitor units were 200 MU. Two plans from an esophagus carcinoma patient with a planning target volume of 86.4 cm and a cervical carcinoma patient with a planning target volume (PTV) of 2094.1 cm were chosen. The passing rates of these plans were gained without and with protecting the circuit area from lead alloys. The gamma analysis was used and the standard was set to 3%/3 mm.The verification passing rate decreased from 95.0% to 69.2% when X was 10 cm while Y increased from 31 cm to 40 cm. With the protection from low melting point lead alloys, the passing rate was from 96.2% to 89.6%. The results of the second set of plans without lead alloys were similar but the passing rate decreased more sharply. The passing rates of the 2 patients were 99.5% and 57.1%. With the protection of the lead alloys, their passing rates were 99.8% and 72.1%, respectively.The results showed that with the increase of the radiation field size in the Y direction, more areas were irradiated in the circuit, and the passing rate gradually decreases and dropped sharply at a certain threshold. After putting lead alloys above the circuit, the passing rate was much better in the static field but was still less than 90% in the second patient volumetric modulated arc therapy (VMAT) because the circuit was irradiate in other directions. In daily QA, we should pay attention to these patients with long size tumor.

A comparative study of identical VMAT about two adjacent targets with and without fixed-jaw technique.

BACKGROUND: The radiation transmission through the multileaf collimators is undesired in modern techniques such as volumetric modulated arc therapy (VMAT). According to identical plans, in this study, we aim to investigate the dosimetric impact of jaw tracking on the VMAT plans on two adjacent targets. METHODS: Two treatment plans were designed for eight pelvic (cervical) patients with two targets using the same optimization parameters. The original plan (O-plan) used automatically selected jaw positions. In the new plan (F-plan), the jaws were fixed to block two targets in two beams. The dosimetric parameters of the two plans were compared to evaluate the improvement of dose sparing for the body volume between two targets (named interOAR) in F-VMAT. RESULTS: The mean dose of interOAR reduced significantly from 654.96 ± 113.38 cGy for O-VMAT, to 490.84 ± 80.26 cGy for F-VMAT (p = 0.018). The monitor units (MUs) in the F-plans were 1.49-fold higher than that in the O-plan. The F and O-plan performed similarly in target dose homogeneity. The differences in Dmax of spinal cord, Dmax of spinal cord planning organ at risk volume, and V20, V30, and V40 of the intestine were insignificant. CONCLUSIONS: VMAT plans with the fixed-jaw method can reduce the volume between two targets effectively. However, despite the plan quality, the method can only be used when the regular methods cannot reach the clinical requirements for critical organs because of additional MUs.

Evaluation of the Effect of a Tracheal Stent on Radiation Dose Distribution via Micro-CT Imaging.

PURPOSE: To study the effect of a metal tracheal stent on radiation dose distribution. METHOD: A metal tube bracket is placed in a self-made foam tube sleeve, and micro-computed tomography scanning is performed directly. The foam sleeve containing the metal bracket is placed in a nonuniform phantom for a routine computed tomography scan. The stents in conventional computed tomography images are replaced by the stents in micro-computed tomography images. Subsequently, 2 sets of computed tomography images are obtained and then imported to a radiotherapy treatment planning system. A single photon beam at 0° is designed in a field size of 10 cm × 10 cm, a photon beam of 6 MV, and a monitor unit of 200 MU. Monte Carlo algorithm is used to calculate the dose distribution and obtain the dose curve of the central axis of the field. The dose is verified with thermoluminescence dose tablets. RESULTS: The micro-computed tomography images of the tracheal stent are clearer and less false-like than its conventional computed tomography images. The planned dose curves of the 2 groups are similar. In comparison with the images without any stents in place, the doses at the incident surface of the stent in the conventional computed tomography images and at the stent exit surface in the rear of the stent increase by 1.86% and 2.76%, respectively. In the micro-computed tomography images, the doses at the incident surface of the stent and at the exit surface behind the stent increase by 1.32% and 1.19%, respectively. Conventional computed tomography reveals a large deviation between the measured and calculated values. CONCLUSION: Tracheal stent based on micro-computed tomography imaging has a less effect on radiotherapy calculation than that based on conventional computed tomography imaging.

Imaging study of pseudo-CT images of superposed ultrasound deformation fields acquired in radiotherapy based on step-by-step local registration.

The purpose of this study is to create a new pseudo-computed tomography (CT) imaging approach under superposed ultrasound (US) deformation fields based on step-by-step local registration. Scanned CT and US 3D image datasets of three patients with postoperative cervical carcinoma were selected, including CT (CTsim) and US images (USsim) acquired during simulated positioning process and cone beam CT (CBCT) and US images for positioning verification (USpv) acquired after treatment for 10 times. Regions of interest such as urinary bladders were segmented out and accepted local registration to obtain different deformation fields. These deformation fields were successively performed according to their order and then applied to localized CT images to obtain pseudo-CT (CTps). After filtering, we obtained the final correct pseudo-CT (CTpsf). The pseudo-CT based on the mask of the whole imaging region of US images (WCTps) were acquired as control. Then, we compared CTpsf, CTps, WCTps, and CBCT in terms of their similarity in anatomical structure and differences in pseudo-CT and CTsim in terms of dosimetry. Structural similarity degree between CTpsf and CBCT was larger compared with that between CTps and WCTps. Target regions and dosages of endangered organs between CTpsf and CTsim were different under the same calculation conditions based on the Monte Carlo algorithm. Compared with the VMAT plan of CTsim, the pass rate of CTpsf in γ analysis under the standards of 2% dosage difference and 2-mm distance difference was 91.8%. The imaging quality of CTpsf was better compared with WCTps and CTps. It exhibited high similarity with CBCT in anatomical structure and had favorable application prospect in adaptive radiotherapy. Graphical abstract The local deformation registration is performed between the ultrasound images based on different regions of interest, and then stepwise applied to localized CT images to obtain pseudo-CT. After filtering, the corrected pseudo CT image is obtained.