Effect of cytochrome P450 3A4 on tissue distribution of humantenmine, koumine, and gelsemine, three alkaloids from the toxic plant Gelsemium.

Humantenmine, koumine, and gelsemine are three indole alkaloids found in the highly toxic plant Gelsemium. Humantenmine was the most toxic, followed by gelsemine and koumine. The aim of this study was to investigate and analyze the effects of these three substances on tissue distribution and toxicity in mice pretreated with the Cytochrome P450 3A4 (CYP3A4) inducer ketoconazole and the inhibitor rifampicin. The in vivo test results showed that the three alkaloids were absorbed rapidly and had the ability to penetrate the blood-brain barrier. At 5 min after intraperitoneal injection, the three alkaloids were widely distributed in various tissues and organs, the spleen and pancreas were the most distributed, and the content of all tissues decreased significantly at 20 min. Induction or inhibition of CYP3A4 in vivo can regulate the distribution and elimination effects of the three alkaloids in various tissues and organs. Additionally, induction of CYP3A4 can reduce the toxicity of humantenmine, and vice versa. Changes in CYP3A4 levels may account for the difference in toxicity of humantenmine. These findings provide a reliable and detailed dataset for drug interactions, tissue distribution, and toxicity studies of Gelsemium alkaloids.

L2NLF: a novel linear-to-nonlinear framework for multi-modal medical image registration.

In recent years, deep learning has ushered in significant development in medical image registration, and the method of non-rigid registration using deep neural networks to generate a deformation field has higher accuracy. However, unlike monomodal medical image registration, multimodal medical image registration is a more complex and challenging task. This paper proposes a new linear-to-nonlinear framework (L2NLF) for multimodal medical image registration. The first linear stage is essentially image conversion, which can reduce the difference between two images without changing the authenticity of medical images, thus transforming multimodal registration into monomodal registration. The second nonlinear stage is essentially unsupervised deformable registration based on the deep neural network. In this paper, a brand-new registration network, CrossMorph, is designed, a deep neural network similar to the U-net structure. As the backbone of the encoder, the volume CrossFormer block can better extract local and global information. Booster promotes the reduction of more deep features and shallow features. The qualitative and quantitative experimental results on T1 and T2 data of 240 patients' brains show that L2NLF can achieve excellent registration effect in the image conversion part with very low computation, and it will not change the authenticity of the converted image at all. Compared with the current state-of-the-art registration method, CrossMorph can effectively reduce average surface distance, improve dice score, and improve the deformation field's smoothness. The proposed methods have potential value in clinical application.

The Activation of GABAAR Alleviated Cerebral Ischemic Injury via the Suppression of Oxidative Stress, Autophagy, and Apoptosis Pathways.

Ischemic stroke is a devastating disease leading to neurologic impairment. Compounding the issue is the very limited array of available interventions. The activation of a γ-aminobutyric acid (GABA) type A receptor (GABAAR) has been reported to produce neuroprotective properties during cerebral ischemia, but its mechanism of action is not yet fully understood. Here, in a rat model of photochemically induced cerebral ischemia, we found that muscimol, a GABAAR agonist, modulated GABAergic signaling, ameliorated anxiety-like behaviors, and attenuated neuronal damage in rats suffering cerebral ischemia. Moreover, GABAAR activation improved brain antioxidant levels, reducing the accumulation of oxidative products, which was closely associated with the NO/NOS pathway. Notably, the inhibition of autophagy markedly relieved the neuronal insult caused by cerebral ischemia. We further established an oxygen-glucose deprivation (OGD)-induced PC12 cell injury model. Both in vivo and in vitro experiments demonstrated that GABAAR activation obviously suppressed autophagy by regulating the AMPK-mTOR pathway. Additionally, GABAAR activation inhibited apoptosis through inhibiting the Bax/Bcl-2 pathway. These data suggest that GABAAR activation exerts neuroprotective effects during cerebral ischemia through improving oxidative stress and inhibiting autophagy and apoptosis. Our findings indicate that GABAAR serves as a target for treating cerebral ischemia and highlight the GABAAR-mediated autophagy signaling pathway.

The performance of a new type accelerator uRT-linac 506c evaluated by a quality assurance automation system.

PURPOSE: The purpose of this study was to evaluate the performance of our quality assurance (QA) automation system and to evaluate the machine performance of a new type linear accelerator uRT-linac 506c within 6 months using this system. METHODS: This QA automation system consists of a hollow cylindrical phantom with 18 steel balls in the phantom surface and an analysis software to process electronic portal imaging device (EPID) measurement image data and report the results. The performance of the QA automation system was evaluated by the tests of repeatability, archivable precision, detectability of introduced errors, and the impact of set-up errors on QA results. The performance of this linac was evaluated by 31 items using this QA system over 6 months. RESULTS: This QA system was able to automatically deliver QA plan, EPID image acquisition, and automatic analysis. All images acquiring and analysis took approximately 4.6 min per energy. The preset error of 0.1 mm in multi-leaf collimator (MLC) leaf were detected as 0.12 ± 0.01 mm for Bank A and 0.10 ± 0.01 mm in Bank B. The 2 mm setup error was detected as -1.95 ± 0.01 mm, -2.02 ± 0.01 mm, 2.01 ± 0.01 mm for X, Y, Z directions, respectively. And data from the tests of repeatability and detectability of introduced errors showed the standard deviation were all within 0.1 mm and 0.1°. and data of the machine performance were all within the tolerance specified by AAPM TG-142. CONCLUSIONS: The QA automation system has high precision and good performance, and it can improve the QA efficiency. The performance of the new accelerator has also performed very well during the testing period.

Deep learning-based 3D brain multimodal medical image registration.

Medical image registration is a critical preprocessing step in medical image analysis. While traditional medical image registration techniques have matured, their registration speed and accuracy still fall short of clinical requirements. In this paper, we propose an improved VoxelMorph network incorporating ResNet modules and CBAM (RCV-Net), for 3D multimodal unsupervised registration. Unlike popular convolution-based U-shaped registration networks like VoxelMorph, RCV-Net incorporates the convolutional block attention module (CBAM) during the convolution process. This inclusion enhances the feature map information extraction capabilities during training and effectively prevents information loss. Additionally, we introduce a lightweight and residual network module at the network's base, which enhances learning ability without significantly increasing training parameters. To evaluate the superiority of our registration model, we utilize evaluation metrics such as structural similarity (SSIM), peak signal-to-noise ratio (PSNR), and mean square error (MSE). Experimental results demonstrate that our proposed network structure outperforms current state-of-the-art methods, yielding better performance in multimodal registration tasks. Furthermore, generalization testing on databases outside of the training set has confirmed the optimal registration effectiveness of our model.

Machine learning-based ensemble prediction model for the gamma passing rate of VMAT-SBRT plan.

PURPOSE: The purpose of this study was to accurately predict or classify the beam GPR with an ensemble model by using machine learning for SBRT(VMAT) plans. METHODS: A total of 128 SBRT VMAT plans with 330 arc beams were retrospectively selected, and 216 radiomics and 34 plan complexity features were calculated for each arc beam. Three models for GPR prediction and classification using support vector machine algorithm were as follows: (1) plan complexity feature-based model (plan model); (2) radiomics feature-based model (radiomics model); and (3) an ensemble model combining the two models (ensemble model). The prediction performance was evaluated by calculating the mean absolute error (MAE), root mean square error (RMSE), and Spearman's correlation coefficient (SC), and the classification performance was measured by calculating the area under the receiver operating characteristic curve (AUC), accuracy, specificity, and sensitivity. RESULTS: The MAE, RMSE and SC at the 2 %/2 mm gamma criterion in the test dataset were 1.4 %, 2.57 %, and 0.563, respectively, for the plan model; 1.42 %, and 2.51 %, and 0.508, respectively, for the radiomics model; and 1.33 %, 2.49 %, and 0.611, respectively, for the ensemble model. The accuracy, specificity, sensitivity, and AUC at the 2 %/2 mm gamma criterion in the test dataset were 0.807, 0.824, 0.681, and 0.854, respectively, for the plan model; 0.860, 0.893, 0.624, and 0.877, respectively, for the radiomics model; and 0.852, 0.871, 0.710, and 0.896, respectively, for the ensemble model. CONCLUSIONS: The ensemble model can improve the prediction and classification performance for the GPR of SBRT (VMAT).

A multi-view assisted registration network for MRI registration pre- and post-therapy.

Image registration of magnetic resonance imaging (MRI) pre- and post-therapy is an important part of evaluating the effect of therapy in tumor patients. The accuracy of evaluation results heavily relies on the alignment of the MRI image after registration. Although recent advancements have been made in medical image registration, applying these methods to MRI registration pre- and post-therapy remains challenging. Existing methods typically utilize single-view data for registration. However, when applied to MRI data where some slices are clear while others are blurred, these methods can be misled by erroneous spatial information in the blurred regions, leading to poor registration outcomes. To mitigate the interference caused by erroneous spatial information in single-view data, this paper proposes a multi-stream fusion-assisted registration network that incorporates different-view MRIs of the same patient at the same site. Additionally, a cross-attention guided fusion module is designed within the network to effectively utilize accurate spatial information from multi-view data. The proposed approach was evaluated on clinical data, and the experimental results demonstrated that incorporating multiple view data as auxiliary information significantly enhances the accuracy of MRI image registration before and after radiotherapy.

Dosimetric impact of CT metal artifact reduction for spinal implants in stereotactic body radiotherapy planning.

Background: Metal artifacts due to spinal implants may affect the accuracy of dose calculation for radiotherapy. However, the dosimetric impact of metal artifact reduction (MAR) for spinal implants in stereotactic body radiotherapy (SBRT) plans has not been well studied. The objective of this study was to evaluate the dosimetric impact of MAR in spinal SBRT planning with three clinically common dose calculation algorithms. Methods: Gammex phantom and 10 patients' computed tomography (CT) images were studied to investigate the effects of titanium implants. A commercial orthopedic MAR algorithm was employed to reduce artifacts. Dose calculations for SBRT were conducted on both artifact-corrected and uncorrected images using three commercial algorithms [analytical anisotropic algorithm (AAA), Acuros XB (AXB), and Monte Carlo (MC)]. Dose discrepancies between artifact-corrected and uncorrected cases were appraised using a dose-volume histogram (DVH) and 3-dimensional (3D) gamma analysis with different distance to agreement (DTA) and dose difference criteria. The gamma agreement index (GAI) was denoted as G(∆D, DTA). Statistical analysis of t-test was utilized to evaluate the dose differences of different algorithms. Results: The phantom study demonstrated that titanium metal artifacts can be effectively reduced. The patient cases study showed that dose differences between the artifact-corrected and uncorrected datasets were small evaluated by gamma index and DVH. Gamma analysis found that even the strict criterion local G(1,1) had average values ≥93.9% for the three algorithms. For all DVH metrics, average differences did not exceed 0.7% in planning target volume (PTV) and 2.1% in planning risk volume of spinal cord (PRV-SC). Statistical analysis showed that the observed dose differences of MC method were significantly larger than those of AAA (P<0.01 for D98% of PTV and P<0.001 for D0.1cc of spinal cord) and AXB methods (P<0.001 for D98% and P<0.0001 for D0.1cc). Conclusions: Dosimetric impact of artifacts caused by titanium implants is not significant in spinal SBRT planning, which indicates that dose calculation algorithms might not be very sensitive to CT number variation caused by titanium inserts.

SECP-Net: SE-Connection Pyramid Network for Segmentation of Organs at Risk with Nasopharyngeal Carcinoma.

Nasopharyngeal carcinoma (NPC) is a kind of malignant tumor. The accurate and automatic segmentation of computed tomography (CT) images of organs at risk (OAR) is clinically significant. In recent years, deep learning models represented by U-Net have been widely applied in medical image segmentation tasks, which can help to reduce doctors' workload. In the OAR segmentation of NPC, the sizes of the OAR are variable, and some of their volumes are small. Traditional deep neural networks underperform in segmentation due to the insufficient use of global and multi-size information. Therefore, a new SE-Connection Pyramid Network (SECP-Net) is proposed. For extracting global and multi-size information, the SECP-Net designs an SE-connection module and a pyramid structure for improving the segmentation performance, especially that of small organs. SECP-Net also uses an auto-context cascaded structure to further refine the segmentation results. Comparative experiments are conducted between SECP-Net and other recent methods on a private dataset with CT images of the head and neck and a public liver dataset. Five-fold cross-validation is used to evaluate the performance based on two metrics; i.e., Dice and Jaccard similarity. The experimental results show that SECP-Net can achieve SOTA performance in these two challenging tasks.

CT synthesis from CBCT using a sequence-aware contrastive generative network.

Computerized tomography (CT) synthesis from cone-beam computerized tomography (CBCT) is a key step in adaptive radiotherapy. It uses a synthetic CT to calculate the dose to correct and adjust the radiotherapy plan in a timely manner. The cycle-consistent adversarial network (Cycle GAN) is commonly used in CT synthesis tasks but it has some defects: (a) the premise of the cycle consistency loss is that the conversion between domains is bijective, but the CBCT and CT conversion does not fully satisfy the bijective relationship, and (b) it does not take advantage of the complementary information between multiple sets of CBCTs for the same patient. To address these problems, we propose a novel framework named the sequence-aware contrastive generative network (SCGN) that introduces an attention sequence fusion module to improve the CBCT quality. In addition, it not only applies contrastive learning to the generative adversarial networks (GANs) to pay more attention to the anatomical structure of CBCT in feature extraction but also uses a new generator to improve the accuracy of the anatomical details. Experimental results on our datasets show that our method significantly outperforms the existing unsupervised CT synthesis methods.

Improving performance of medical image alignment through super-resolution.

Medical image alignment is an important tool for tracking patient conditions, but the quality of alignment is influenced by the effectiveness of low-dose Cone-beam CT (CBCT) imaging and patient characteristics. To address these two issues, we propose an unsupervised alignment method that incorporates a preprocessing super-resolution process. We constructed the model based on a private clinical dataset and validated the enhancement of the super-resolution on alignment using clinical and public data. Through all three experiments, we demonstrate that higher resolution data yields better results in the alignment process. To fully constrain similarity and structure, a new loss function is proposed; Pearson correlation coefficient combined with regional mutual information. In all test samples, the newly proposed loss function obtains higher results than the common loss function and improve alignment accuracy. Subsequent experiments verified that, combined with the newly proposed loss function, the super-resolution processed data boosts alignment, can reaching up to 9.58%. Moreover, this boost is not limited to a single model, but is effective in different alignment models. These experiments demonstrate that the unsupervised alignment method with super-resolution preprocessing proposed in this study effectively improved alignment and plays an important role in tracking different patient conditions over time.

Synthetic CT generation from CBCT using double-chain-CycleGAN.

PURPOSE: Cone-beam CT (CBCT) has the advantage of being less expensive, lower radiation dose, less harm to patients, and higher spatial resolution. However, noticeable noise and defects, such as bone and metal artifacts, limit its clinical application in adaptive radiotherapy. To explore the potential application value of CBCT in adaptive radiotherapy, In this study, we improve the cycle-GAN's backbone network structure to generate higher quality synthetic CT (sCT) from CBCT. METHOD: An auxiliary chain containing a Diversity Branch Block (DBB) module is added to CycleGAN's generator to obtain low-resolution supplementary semantic information. Moreover, an adaptive learning rate adjustment strategy (Alras) function is used to improve stability in training. Furthermore, Total Variation Loss (TV loss) is added to generator loss to improve image smoothness and reduce noise. RESULTS: Compared to CBCT images, the Root Mean Square Error (RMSE) dropped by 27.97 from 158.49. The Mean Absolute Error (MAE) of the sCT generated by our model improved from 43.2 to 32.05. The Peak Signal-to-Noise Ratio (PSNR) increased by 1.61 from 26.19. The Structural Similarity Index Measure (SSIM) improved from 0.948 to 0.963, and the Gradient Magnitude Similarity Deviation (GMSD) improved from 12.98 to 9.33. The generalization experiments show that our model performance is still superior to CycleGAN and respath-CycleGAN.

A deformable patch-based transformer for 3D medical image registration.

PURPOSE: Medical image registration is of great importance in clinical medicine. However, medical image registration algorithms are still in the development stage due to the challenges posed by the related complex physiological structures. The objective of this study was to design a 3D medical image registration algorithm that satisfies the need for high accuracy and speed of complex physiological structures. METHODS: We present a new unsupervised learning algorithm, "DIT-IVNet," for 3D medical image registration. Unlike the more popular convolution-based U-shaped registration network architectures like VoxelMorph, DIT-IVNet uses a combined convolution and transformer network architecture. To better extract image information features and reduce the heavy training parameters, we improved the 2D_Depatch module to a 3D_Depatch module, thus replacing the patch embedding in the original Vision Transformer which adaptively performs patch embedding based on 3D image structure information. We also designed inception blocks in the down-sampling part of the network to help coordinate feature learning from images to different scales. RESULTS: Dice score, Negative Jacobian determinant, Hausdorff distance, and Structural Similarity evaluation metrics were used to evaluate the registration effects. The results showed that our proposed network had the best metric results compared with some state-of-the-art methods. Moreover, our network obtained the highest Dice score in the generalization experiments which indicated better generalizability of our model. CONCLUSION: We proposed an unsupervised registration network and evaluated its performance in deformable medical image registration. The results of the evaluation metrics showed that the network structure outperformed state-of-the-art methods for the registration of brain datasets.

Plan quality and treatment efficiency assurance of two VMAT optimization for cervical cancer radiotherapy.

To investigate the difference of the fluence map optimization (FMO) and Stochastic platform optimization (SPO) algorithm in a newly-introduced treatment planning system (TPS). METHODS: 34 cervical cancer patients with definitive radiation were retrospectively analyzed. Each patient has four plans: FMO with fixed jaw plans (FMO-FJ) and no fixed jaw plans (FMO-NFJ); SPO with fixed jaw plans (SPO-FJ) and no fixed jaw plans (SPO-NFJ). Dosimetric parameters, Modulation Complexity Score (MCS), Gamma Pass Rate (GPR) and delivery time were analyzed among the four plans. RESULTS: For target coverage, SPO-FJ plans are the best ones (P ≤ 0.00). FMO plans are better than SPO-NFJ plans (P ≤ 0.00). For OARs sparing, SPO-FJ plans are better than FMO plans for mostly OARs (P ≤ 0.04). Additionally, SPO-FJ plans are better than SPO-NFJ plans (P ≤ 0.02), except for rectum V45Gy. Compared to SPO-NFJ plans, the FMO plans delivered less dose to bladder, rectum, colon V40Gy and pelvic bone V40Gy (P ≤ 0.04). Meanwhile, the SPO-NFJ plans showed superiority in MU, delivery time, MCS and GPR in all plans. In terms of delivery time and MCS, the SPO-FJ plans are better than FMO plans. FMO-FJ plans are better than FMO-NFJ plans in delivery efficiency. MCSs are strongly correlated with PCTV length, which are negatively with PCTV length (P ≤ 0.03). The delivery time and MUs of the four plans are strongly correlated (P ≤ 0.02). Comparing plans with fixed or no fixed jaw in two algorithms, no difference was found in FMO plans in target coverage and minor difference in Kidney_L Dmean, Mu and delivery time between PCTV width≤15.5 cm group and >15.5 cm group. For SPO plans, SPO-FJ plans showed more superiority in target coverage and OARs sparing than the SPO-NFJ plans in the two groups. CONCLUSIONS: SPO-FJ plans showed superiority in target coverage and OARs sparing, as well as higher delivery efficiency in the four plans.

Geometry-Consistent Adversarial Registration Model for Unsupervised Multi-Modal Medical Image Registration.

Deformable multi-modal medical image registration aligns the anatomical structures of different modalities to the same coordinate system through a spatial transformation. Due to the difficulties of collecting ground-truth registration labels, existing methods often adopt the unsupervised multi-modal image registration setting. However, it is hard to design satisfactory metrics to measure the similarity of multi-modal images, which heavily limits the multi-modal registration performance. Moreover, due to the contrast difference of the same organ in multi-modal images, it is difficult to extract and fuse the representations of different modal images. To address the above issues, we propose a novel unsupervised multi-modal adversarial registration framework that takes advantage of image-to-image translation to translate the medical image from one modality to another. In this way, we are able to use the well-defined uni-modal metrics to better train the models. Inside our framework, we propose two improvements to promote accurate registration. First, to avoid the translation network learning spatial deformation, we propose a geometry-consistent training scheme to encourage the translation network to learn the modality mapping solely. Second, we propose a novel semi-shared multi-scale registration network that extracts features of multi-modal images effectively and predicts multi-scale registration fields in an coarse-to-fine manner to accurately register the large deformation area. Extensive experiments on brain and pelvic datasets demonstrate the superiority of the proposed method over existing methods, revealing our framework has great potential in clinical application.

Enhancement of cone beam CT image registration by super-resolution pre-processing algorithm.

In order to enhance cone-beam computed tomography (CBCT) image information and improve the registration accuracy for image-guided radiation therapy, we propose a super-resolution (SR) image enhancement method. This method uses super-resolution techniques to pre-process the CBCT prior to registration. Three rigid registration methods (rigid transformation, affine transformation, and similarity transformation) and a deep learning deformed registration (DLDR) method with and without SR were compared. The five evaluation indices, the mean squared error (MSE), mutual information, Pearson correlation coefficient (PCC), structural similarity index (SSIM), and PCC + SSIM, were used to validate the results of registration with SR. Moreover, the proposed method SR-DLDR was also compared with the VoxelMorph (VM) method. In rigid registration with SR, the registration accuracy improved by up to 6% in the PCC metric. In DLDR with SR, the registration accuracy was improved by up to 5% in PCC + SSIM. When taking the MSE as the loss function, the accuracy of SR-DLDR is equivalent to that of the VM method. In addition, when taking the SSIM as the loss function, the registration accuracy of SR-DLDR is 6% higher than that of VM. SR is a feasible method to be used in medical image registration for planning CT (pCT) and CBCT. The experimental results show that the SR algorithm can improve the accuracy and efficiency of CBCT image alignment regardless of which alignment algorithm is used.

Network Pharmacology and Experimental Verification to Unveil the Mechanism of N-Methyl-D-Aspartic Acid Rescue Humantenirine-Induced Excitotoxicity.

Gelsemium is a medicinal plant that has been used to treat various diseases, but it is also well-known for its high toxicity. Complex alkaloids are considered the main poisonous components in Gelsemium. However, the toxic mechanism of Gelsemium remains ambiguous. In this work, network pharmacology and experimental verification were combined to systematically explore the specific mechanism of Gelsemium toxicity. The alkaloid compounds and candidate targets of Gelsemium, as well as related targets of excitotoxicity, were collected from public databases. The crucial targets were determined by constructing a protein-protein interaction (PPI) network. Subsequently, Gene Ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) were used to explore the bioprocesses and signaling pathways involved in the excitotoxicity corresponding to alkaloids in Gelsemium. Then, the binding affinity between the main poisonous alkaloids and key targets was verified by molecular docking. Finally, animal experiments were conducted to further evaluate the potential mechanisms of Gelsemium toxicity. A total of 85 alkaloids in Gelsemium associated with 214 excitotoxicity-related targets were predicted by network pharmacology. Functional analysis showed that the toxicity of Gelsemium was mainly related to the protein phosphorylation reaction and plasma membrane function. There were also 164 pathways involved in the toxic mechanism, such as the calcium signaling pathway and MAPK signaling pathway. Molecular docking showed that alkaloids have high affinity with core targets, including MAPK3, SRC, MAPK1, NMDAR2B and NMDAR2A. In addition, the difference of binding affinity may be the basis of toxicity differences among different alkaloids. Humantenirine showed significant sex differences, and the LD50 values of female and male mice were 0.071 mg·kg-1 and 0.149 mg·kg-1, respectively. Furthermore, we found that N-methyl-D-aspartic acid (NMDA), a specific NMDA receptor agonist, could significantly increase the survival rate of acute humantenirine-poisoned mice. The results also show that humantenirine could upregulate the phosphorylation level of MAPK3/1 and decrease ATP content and mitochondrial membrane potential in hippocampal tissue, while NMDA could rescue humantenirine-induced excitotoxicity by restoring the function of mitochondria. This study revealed the toxic components and potential toxic mechanism of Gelsemium. These findings provide a theoretical basis for further study of the toxic mechanism of Gelsemium and potential therapeutic strategies for Gelsemium poisoning.

Integrated radiomics, dose-volume histogram criteria and clinical features for early prediction of saliva amount reduction after radiotherapy in nasopharyngeal cancer patients.

PURPOSE: Previously, the evaluation of xerostomia depended on subjective grading systems, rather than the accurate saliva amount reduction. Our aim was to quantify acute xerostomia with reduced saliva amount, and apply radiomics, dose-volume histogram (DVH) criteria and clinical features to predict saliva amount reduction by machine learning techniques. MATERIAL AND METHODS: Computed tomography (CT) of parotid glands, DVH, and clinical data of 52 patients were collected to extract radiomics, DVH criteria and clinical features, respectively. Firstly, radiomics, DVH criteria and clinical features were divided into 3 groups for feature selection, in order to alleviate the masking effect of the number of features in different groups. Secondly, the top features in the 3 groups composed integrated features, and features selection was performed again for integrated features. In this study, feature selection was used as a combination of eXtreme Gradient Boosting (XGBoost) and SHapley Additive exPlanations (SHAP) to alleviate multicollinearity. Finally, 6 machine learning techniques were used for predicting saliva amount reduction. Meanwhile, top radiomics features were modeled using the same machine learning techniques for comparison. RESULT: 17 integrated features (10 radiomics, 4 clinical, 3 DVH criteria) were selected to predict saliva amount reduction, with a mean square error (MSE) of 0.6994 and a R2 score of 0.9815. Top 17 and 10 selected radiomics features predicted saliva amount reduction, with MSE of 0.7376, 0.7519, and R2 score of 0.9805, 0.9801, respectively. CONCLUSION: With the same number of features, integrated features (radiomics + DVH criteria + clinical) performed better than radiomics features alone. The important DVH criteria and clinical features mainly included, white blood cells (WBC), parotid_glands_Dmax, Age, parotid_glands_V15, hemoglobin (Hb), BMI and parotid_glands_V45.

Improving cone-beam CT quality using a cycle-residual connection with a dilated convolution-consistent generative adversarial network.

Objective.Cone-Beam CT (CBCT) often results in severe image artifacts and inaccurate HU values, meaning poor quality CBCT images cannot be directly applied to dose calculation in radiotherapy. To overcome this, we propose a cycle-residual connection with a dilated convolution-consistent generative adversarial network (Cycle-RCDC-GAN).Approach.The cycle-consistent generative adversarial network (Cycle-GAN) was modified using a dilated convolution with different expansion rates to extract richer semantic features from input images. Thirty pelvic patients were used to investigate the effect of synthetic CT (sCT) from CBCT, and 55 head and neck patients were used to explore the generalizability of the model. Three generalizability experiments were performed and compared: the pelvis trained model was applied to the head and neck; the head and neck trained model was applied to the pelvis, and the two datasets were trained together.Main results.The mean absolute error (MAE), the root mean square error (RMSE), peak signal to noise ratio (PSNR), the structural similarity index (SSIM), and spatial nonuniformity (SNU) assessed the quality of the sCT generated from CBCT. Compared with CBCT images, the MAE improved from 28.81 to 18.48, RMSE from 85.66 to 69.50, SNU from 0.34 to 0.30, and PSNR from 31.61 to 33.07, while SSIM improved from 0.981 to 0.989. The sCT objective indicators of Cycle-RCDC-GAN were better than Cycle-GAN's. The objective metrics for generalizability were also better than Cycle-GAN's.Significance.Cycle-RCDC-GAN enhances CBCT image quality and has better generalizability than Cycle-GAN, which further promotes the application of CBCT in radiotherapy.

Synthetic CT generation based on CBCT using respath-cycleGAN.

PURPOSE: Cone-beam computed tomography (CBCT) plays an important role in radiotherapy, but the presence of a large number of artifacts limits its application. The purpose of this study was to use respath-cycleGAN to synthesize CT (sCT) similar to planning CT (pCT) from CBCT for future clinical practice. METHODS: The method integrates the respath concept into the original cycleGAN, called respath-cycleGAN, to map CBCT to pCT. Thirty patients were used for training and 15 for testing. RESULTS: The mean absolute error (MAE), root mean square error (RMSE), peak signal to noise ratio (PSNR), structural similarity index (SSIM), and spatial nonuniformity (SNU) were calculated to assess the quality of sCT generated from CBCT. Compared with CBCT images, the MAE improved from 197.72 to 140.7, RMSE from 339.17 to 266.51, and PSNR from 22.07 to 24.44, while SSIM increased from 0.948 to 0.964. Both visually and quantitatively, sCT with respath is superior to sCT without respath. We also performed a generalization test of the head-and-neck (H&N) model on a pelvic data set. The results again showed that our model was superior. CONCLUSION: We developed a respath-cycleGAN method to synthesize CT with good quality from CBCT. In future clinical practice, this method may be used to develop radiotherapy plans.