Original Computed Tomography Images Research Articles

Diagnostic accuracy of dual-energy computed tomography angiography in the differentiation of benign and malignant pelvic masses

PurposeThe dual-energy computed tomography(CT) angiography can accurately display subtle details of blood vessels and their surroundings. We aimed to apply dual-energy CT angiography and virtual monochromatic spectral(VMS) images to pelvic mass imaging and evaluate its value of distinguishing between benign and malignant pelvic masses. MethodsThe prospective study was approved by the Institutional Review Board and all participants signed informed consent forms. From August 2018 to July 2020, consecutive female patients with pelvic mass undergone dual-source third-generation CT angiography. The 40 keV VMS images were reconstructed to display mass morphology and corresponding feeding arteries. All images were evaluated by two radiologists blinded to diagnosis(with 9 years and 10 years of experience in CT reading).Disagreements were solved by consensus. The final diagnosis was using histopathology results as the gold standard. Interobserver variability was calculated using Cohen’s kappa and Quadratic Weighted Kappa. The differences between benign and malignant masses were compared using the chi-square test. Accuracy rate, sensitivity, specificity, and the area under the curve (AUC) were calculated as the primary indices for diagnostic accuracy. McNemar test was used to evaluate the difference in diagnostic accuracy between dual-energy CT angiography images and original CT images. A two-tailed P < 0.05 was considered statistically significant. ResultsA total of 64 patients with 28 malignant masses and 47 benign masses were included. The characteristics of malignant masses showed the branch number of the main feeding artery was ≥ 3(71.4%, 20/28), the course of feeding artery(100%, 28/28) and the mass shape(85.7%, 24/28) were both irregulars. Those characteristics all had statistical differences between benign and malignant masses(all P = 0.000). The Models using the course of feeding artery as a predictor, the accuracy, sensitivity and positive likelihood ratio were 89.3% (95 % CI: 0.801, 0.947), 100% (95 % CI: 0.850, 1) and 5.875(95 % CI: 3.125, 11.044), respectively. The diagnostic accuracy of every model by dual-energy CT angiography was significantly higher than that by original CT imaging(all P = 0.000). ConclusionsThe dual-energy CT angiography can distinguish malignant pelvic masses from benign masses by providing characteristic images of feeding arteries and mass shape.

A CT-Based Radiomics Nomogram to Predict Complete Ablation of Pulmonary Malignancy: A Multicenter Study.

ObjectiveThermal ablation is a minimally invasive procedure for the treatment of pulmonary malignancy, but the intraoperative measure of complete ablation of the tumor is mainly based on the subjective judgment of clinicians without quantitative criteria. This study aimed to develop and validate an intraoperative computed tomography (CT)-based radiomic nomogram to predict complete ablation of pulmonary malignancy.MethodsThis study enrolled 104 individual lesions from 92 patients with primary or metastatic pulmonary malignancies, which were randomly divided into training cohort (n=74) and verification cohort (n=30). Radiomics features were extracted from the original CT images when the study clinicians determined the completion of the ablation surgery. Minimum redundancy maximum relevance (mRMR) and least absolute shrinkage and selection operator (LASSO) were adopted for the dimensionality reduction of high-dimensional data and feature selection. The prediction model was developed based on the radiomics signature combined with the independent clinical predictors by multiple logistic regression analysis. The area under the curve (AUC), accuracy, sensitivity, and specificity were calculated. Receiver operating characteristic (ROC) curves and calibration curves were used to evaluate the predictive performance of the model. Decision curve analysis (DCA) was applied to estimate the clinical usefulness and net benefit of the nomogram for decision making.ResultsThirteen CT features were selected to construct radiomics prediction model, which exhibits good predictive performance for determination of complete ablation of pulmonary malignancy. The AUCs of a CT-based radiomics nomogram that integrated the radiomics signature and the clinical predictors were 0.88 (95% CI 0.80-0.96) in the training cohort and 0.87 (95% CI: 0.71–1.00) in the validation cohort, respectively. The radiomics nomogram was well calibrated in both the training and validation cohorts, and it was highly consistent with complete tumor ablation. DCA indicated that the nomogram was clinically useful.ConclusionA CT-based radiomics nomogram has good predictive value for determination of complete ablation of pulmonary malignancy intraoperatively, which can assist in decision-making.

A computer-aided diagnostic framework for coronavirus diagnosis using texture-based radiomics images.

The accurate and rapid detection of the novel coronavirus infection, coronavirus is very important to prevent the fast spread of such disease. Thus, reducing negative effects that influenced many industrial sectors, especially healthcare. Artificial intelligence techniques in particular deep learning could help in the fast and precise diagnosis of coronavirus from computed tomography images. Most artificial intelligence-based studies used the original computed tomography images to build their models; however, the integration of texture-based radiomics images and deep learning techniques could improve the diagnostic accuracy of the novel coronavirus diseases. This study proposes a computer-assisted diagnostic framework based on multiple deep learning and texture-based radiomics approaches. It first trains three Residual Networks (ResNets) deep learning techniques with two texture-based radiomics images including discrete wavelet transform and gray-level covariance matrix instead of the original computed tomography images. Then, it fuses the texture-based radiomics deep features sets extracted from each using discrete cosine transform. Thereafter, it further combines the fused texture-based radiomics deep features obtained from the three convolutional neural networks. Finally, three support vector machine classifiers are utilized for the classification procedure. The proposed method is validated experimentally on the benchmark severe respiratory syndrome coronavirus 2 computed tomography image dataset. The accuracies attained indicate that using texture-based radiomics (gray-level covariance matrix, discrete wavelet transform) images for training the ResNet-18 (83.22%, 74.9%), ResNet-50 (80.94%, 78.39%), and ResNet-101 (80.54%, 77.99%) is better than using the original computed tomography images (70.34%, 76.51%, and 73.42%) for ResNet-18, ResNet-50, and ResNet-101, respectively. Furthermore, the sensitivity, specificity, accuracy, precision, and F1-score achieved using the proposed computer-assisted diagnostic after the two fusion steps are 99.47%, 99.72%, 99.60%, 99.72%, and 99.60% which proves that combining texture-based radiomics deep features obtained from the three ResNets has boosted its performance. Thus, fusing multiple texture-based radiomics deep features mined from several convolutional neural networks is better than using only one type of radiomics approach and a single convolutional neural network. The performance of the proposed computer-assisted diagnostic framework allows it to be used by radiologists in attaining fast and accurate diagnosis.

Feasibility of Synthetic Computed Tomography Images Generated from Magnetic Resonance Imaging Scans Using Various Deep Learning Methods in the Planning of Radiation Therapy for Prostate Cancer.

Simple SummaryMRI-only simulation in radiation therapy (RT) planning has received attention because the CT scan can be omitted. For MRI-only simulation, synthetic CT (sCT) is necessary for the dose calculation. Various methodologies have been suggested for the generation of sCT and, recently, methods using the deep learning approaches are actively investigated. GAN and cycle-consistent GAN (CycGAN) have been mainly tested, however, very limited studies compared the qualities of sCTs generated from these methods or suggested other models for sCT generation. We have compared GAN, CycGAN, and, reference-guided GAN (RgGAN), a new model of deep learning method. We found that the performance in the HU conservation for soft tissue was poorest for GAN. All methods could generate sCTs feasible for VMAT planning with the trend that sCT generated from the RgGAN showed best performance in dosimetric conservation D98% and D95% than sCTs from other methodologies.We aimed to evaluate and compare the qualities of synthetic computed tomography (sCT) generated by various deep-learning methods in volumetric modulated arc therapy (VMAT) planning for prostate cancer. Simulation computed tomography (CT) and T2-weighted simulation magnetic resonance image from 113 patients were used in the sCT generation by three deep-learning approaches: generative adversarial network (GAN), cycle-consistent GAN (CycGAN), and reference-guided CycGAN (RgGAN), a new model which performed further adjustment of sCTs generated by CycGAN with available paired images. VMAT plans on the original simulation CT images were recalculated on the sCTs and the dosimetric differences were evaluated. For soft tissue, a significant difference in the mean Hounsfield unites (HUs) was observed between the original CT images and only sCTs from GAN (p = 0.03). The mean relative dose differences for planning target volumes or organs at risk were within 2% among the sCTs from the three deep-learning approaches. The differences in dosimetric parameters for D98% and D95% from original CT were lowest in sCT from RgGAN. In conclusion, HU conservation for soft tissue was poorest for GAN. There was the trend that sCT generated from the RgGAN showed best performance in dosimetric conservation D98% and D95% than sCTs from other methodologies.

Virtual magnetic resonance lumbar spine images generated from computed tomography images using conditional generative adversarial networks

IntroductionThe aim of this study was to generate virtual Magnetic resonance (MR) from computed tomography (CT) using conditional generative adversarial networks (cGAN). MethodsWe selected examinations from 22 adults who obtained their CT and MR lumbar spine examinations. Overall, 4 examinations were used as test data, and 18 examinations were used as training data. A cGAN was trained to generate virtual MR images from the CT images using the corresponding MR images as targets. After training, the generated virtual MR images from test data in epochs 1, 10, 50, 100, 500, and 1000 were compared with the original ones using the mean square error (MSE) and structural similarity index (SSIM). Additionally, two radiologists also performed qualitative assessments. ResultsThe MSE of the virtual MR images decreased as the epoch of the cGANs increased from the original CT images: 8876.7 ± 1192.9 (original CT), 1567.5 ± 433.9 (Epoch 1), 1242.4 ± 442.0 (Epoch 10), 1065.8 ± 478.1 (Epoch 50), 1276.1 ± 718.9 (Epoch 100), 1046.7 ± 488.2 (Epoch 500), and 1031.7 ± 400.0 (Epoch 1000). No considerable differences were observed in the qualitative evaluation between the virtual MR images and the original ones, except in the structure of the spinal canal. ConclusionVirtual MR lumbar spine images using cGANs could be a feasible technique to generate near-MR images from CT without MR examinations for evaluation of the vertebral body and intervertebral disc. Implications for practiceVirtual MR lumbar spine images using cGANs can offer virtual CT images with sufficient quality for attenuation correction for PET or dose planning in radiotherapy.

A deep learning- and CT image-based prognostic model for the prediction of survival in non-small cell lung cancer.

To assist clinicians in arranging personalized treatment, planning follow-up programs and extending survival times for non-small cell lung cancer (NSCLC) patients, a method of deep learning combined with computed tomography (CT) imaging for survival prediction was designed. Data were collected from 484 patients from four research centers. The data from 344 patients were utilized to build the A_CNN survival prognosis model to classify 2-year overall survival time ranges (730 days cut-off). Data from 140 patients, including independent internal and external test sets, were utilized for model testing. First, a series of preprocessing techniques were used to process the original CT images and generate training and test data sets from the axial, coronal, and sagittal planes. Second, the structure of the A_CNN model was designed based on asymmetric convolution, bottleneck blocks, the uniform cross-entropy (UC) loss function, and other advanced techniques. After that, the A_CNN model was trained, and numerous comparative experiments were designed to obtain the best prognostic survival model. Last, the model performance was evaluated, and the predicted survival curves were analyzed. The A_CNN survival prognosis model yielded a high patient-level accuracy of 88.8%, a patch-level accuracy of 82.9%, and an area under the receiver operating characteristic (ROC) curve (AUC) of 0.932. When tested on an external data set, the maximum patient-level accuracy was 80.0%. The results suggest that using a deep learning method can improve prognosis in patients with NSCLC and has important application value in establishing individualized prognostic models.

A Method of CT Image Denoising Based on Residual Encoder-Decoder Network.

Low-dose computed tomography (CT) has proved effective in lowering radiation risk for the patients, but the resultant noise and bar artifacts in CT images can be a disturbance for medical diagnosis. The difficulty of modeling statistical features in the image domain makes it impossible for the existing methods that directly process reconstructed images to maintain the detailed texture structure of images while reducing noise, which accounts for the failure in CT diagnostic images in practical application. To overcome this defect, this paper proposes a CT image-denoising method based on an improved residual encoder-decoder network. Firstly, in our approach, the notion of recursion is integrated into the original residual encoder-decoder network to lower the algorithm complexity and boost efficiency in image denoising. The original CT images and the postrecursion result graph output after recursion are used as the input for the next recursion simultaneously, and the shallow encoder-decoder network is recycled. Secondly, the root-mean-square error loss function and perceptual loss function are introduced to ensure the texture of denoised CT images. On this basis, the tissue processing technology based on clustering segmentation is optimized considering that the images after improved RED-CNN training will still have certain artifacts. Finally, the experimental results of the TCGA-COAD clinical data set show that under the same experimental conditions, our method outperforms WGAN in average postdenoising PSNR and SSIM of CT images. Moreover, with a lower algorithm complexity and shorter execution time, our method is a significant improvement on RED-CNN and is applicable for actual scenarios.

Synthetic Computed Tomography Generation from 0.35T Magnetic Resonance Images for Magnetic Resonance–Only Radiation Therapy Planning Using Perceptual Loss Models

PurposeMagnetic resonance imaging (MRI) provides excellent soft-tissue contrast, which makes it useful for delineating tumor and normal structures in radiation therapy planning, but MRI cannot readily provide electron density for dose calculation. Computed tomography (CT) is used but introduces registration uncertainty between MRI and CT. Previous studies have shown that synthetic CTs (sCTs) can be generated directly from MRI images with deep learning methods. However, mainly high-field MRI images have been validated. This study tested whether acceptable sCTs for MR-only radiation therapy planning can be synthesized using an integrated MR-guided linear accelerator at 0.35T, using MRI images and treatment plans in the liver region. Methods and MaterialsTwo models were investigated in this study: a convolutional neural network (Unet) with conventional mean square error (MSE) loss and a Unet using a secondary convolutional neural network for perceptual loss. A total of 37 cases were used in this study with 10-fold cross validation, and 37 treatment plans were generated and evaluated for target coverage and dose to organs at risk (OARs) in the MSE loss model, perceptual loss model, and original CT. ResultsThe sCTs predicted by the perceptual loss model had improved subjective visual quality compared with those predicted by the MSE loss model, but both were similar in mean absolute error (MAE), peak-signal-to-noise ratio (PSNR), and normalized cross-correlation (NCC). The MAE, PSNR, and NCC for the perceptual loss model were 35.64, 24.11, and 0.9539, respectively, and those for the MSE loss model were 35.67, 24.36, and 0.9566, respectively. No significant differences in target coverage and dose to OARs were found between the sCT predicted by the perceptual loss model or by the MSE model and the original CT image. ConclusionsThis study indicated that a Unet with both MSE loss and perceptual loss models can be used for generating sCT images from a 0.35T integrated MR linear accelerator.

Investigation of Three-dimensional Printing Materials for Printing Aorta Model Replicating Type B Aortic Dissection.

This study aims to determine a printing material that has both elastic property and radiology equivalence close to the real aorta for simulation of endovascular stent-graft repair of aortic dissection. With the rapid development of Three-Dimensional (3D) printing technology, a patient- specific 3D printed model is able to help surgeons to make a better treatment plan for Type B aortic dissection patients. However, the radiological properties of most 3D printing materials have not been well characterized. This study aims to investigate the appropriate materials for printing human aorta with mechanical and radiological properties similar to the real aortic Computed Tomography (CT) attenuation. Quantitative assessment of CT attenuation of different materials used in 3D printed models of aortic dissection for developing patient-specific 3D printed aorta models to simulate type B aortic dissection. A 25-mm length of aorta model was segmented from a patient's image dataset with a diagnosis of type B aortic dissection. Four different elastic commercial 3D printing materials, namely Agilus A40 and A50, Visijet CE-NT A30 and A70 were selected and printed with different hardness. Totally four models were printed out and CT scanned twice on a 192-slice CT scanner using the standard aortic CT angiography protocol, with and without contrast inside the lumen. Five reference points with the Region Of Interest (ROI) of 1.77 mm2 were selected at the aortic wall, and intimal flap and their Hounsfield units (HU) were measured and compared with the CT attenuation of original CT images. The comparison between the patient's aorta and models was performed through a paired-sample t-test to determine if there is any significant difference. The mean CT attenuation of the aortic wall of the original CT images was 80.7 HU. Analysis of images without using contrast medium showed that the material of Agilus A50 produced the mean CT attenuation of 82.6 HU, which is similar to that of original CT images. The CT attenuation measured at images acquired with the other three materials was significantly lower than that of the original images (p<0.05). After adding contrast medium, Visijet CE-NT A30 had an average CT attenuation of 90.6 HU, which is close to that of the original images without a statistically significant difference (p>0.05). In contrast, the CT attenuation measured at images acquired with other three materials (Agilus A40, A50 and Visiject CE-NT A70) was 129 HU, 135 HU and 129.6 HU, respectively, which is significantly higher than that of original CT images (p<0.05). Both Visijet CE-NT and Agilus have tensile strength and elongation close to actual patient's tissue properties producing similar CT attenuation. Visijet CE-NT A30 is considered the appropriate material for printing aorta to simulate contrast-enhanced CT imaging of type B aortic dissection. Due to the lack of body phantoms in the experiments, further research with the simulation of realistic anatomical body environment should be conducted.

Accuracy of digital model generated from CT data with metal artifact reduction algorithm

This study investigated whether metal artifact reduction (MAR) applied computed tomography (CT) scans could be used to generate precise digital models and explored possible correlations between the amount of metal artifact and model accuracy. Thirty maxillofacial CT scans were randomly selected and a MAR algorithm was applied. By subtracting the original and MAR-applied CT images, the amount of metal artifact was quantified. Digital models were generated from the original and the MAR-applied CT data. Paired digital models were superimposed and shape deviation in planar surface was measured at 10 points in 4 planes. Statistical analyses were performed to compare deviations and to assess correlations between the amount of artifact and deviation. The MAR algorithm reduced metal artifact in all cases. The overall mean deviation of the MAR-applied models was 0.0868 mm, with no significant difference according to the reference plane. The amount of artifact did not significantly influence the accuracy of the digital models. MAR-applied CT is a convenient source for digital modeling with clinically acceptable accuracy. The MAR algorithm can be used regardless of the amount of metal artifact, which are generated by dental prostheses, for the quick and convenient manipulation of dental digital models.

Quantitative Assessment of 3D Printed Model Accuracy in Delineating Congenital Heart Disease.

Background: Three-dimensional (3D) printing is promising in medical applications, especially presurgical planning and the simulation of congenital heart disease (CHD). Thus, it is clinically important to generate highly accurate 3D-printed models in replicating cardiac anatomy and defects. The present study aimed to investigate the accuracy of the 3D-printed CHD model by comparing them with computed tomography (CT) images and standard tessellation language (STL) files. Methods: Three models were printed, comprising different CHD pathologies, including the tetralogy of Fallot (ToF), ventricular septal defect (VSD) and double-outlet right-ventricle (DORV). The ten anatomical locations were measured in each comparison. Pearson’s correlation coefficient, Bland–Altman analysis and intra-class correlation coefficient (ICC) determined the model accuracy. Results: All measurements with three printed models showed a strong correlation (r = 0.99) and excellent reliability (ICC = 0.97) when compared to original CT images, CT images of the 3D-printed models, STL files and 3D-printed CHD models. Conclusion: This study demonstrated the high accuracy of 3D-printed heart models with excellent correlation and reliability when compared to multiple source data. Further investigation into 3D printing in CHD should focus on the clinical value and the benefits to patients.

Pancreas segmentation using a dual-input v-mesh network.

Accurate segmentation of the pancreas from abdomen scans is crucial for the diagnosis and treatment of pancreatic diseases. However, the pancreas is a small, soft and elastic abdominal organ with high anatomical variability and has a low tissue contrast in computed tomography (CT) scans, which makes segmentation tasks challenging. To address this challenge, we propose a dual-input v-mesh fully convolutional network (FCN) to segment the pancreas in abdominal CT images. Specifically, dual inputs, i.e., original CT scans and images processed by a contrast-specific graph-based visual saliency (GBVS) algorithm, are simultaneously sent to the network to improve the contrast of the pancreas and other soft tissues. To further enhance the ability to learn context information and extract distinct features, a v-mesh FCN with an attention mechanism is initially utilized. In addition, we propose a spatial transformation and fusion (SF) module to better capture the geometric information of the pancreas and facilitate feature map fusion. We compare the performance of our method with several baseline and state-of-the-art methods on the publicly available NIH dataset. The comparison results show that our proposed dual-input v-mesh FCN model outperforms previous methods in terms of the Dice similarity coefficient (DSC), positive predictive value (PPV), sensitivity (SEN), average surface distance (ASD) and Hausdorff distance (HD). Moreover, ablation studies show that our proposed modules/structures are critical for effective pancreas segmentation.

Generation of Brain Dual-Energy CT from Single-Energy CT Using Deep Learning.

Deep learning (DL) has shown great potential in conversions between various imaging modalities. Similarly, DL can be applied to synthesize a high-kV computed tomography (CT) image from its corresponding low-kV CT image. This indicates the feasibility of obtaining dual-energy CT (DECT) images without purchasing a DECT scanner. In this study, we investigated whether a low-to-high kV mapping was better than a high-to-low kV mapping. We used a U-Net model to perform conversions between different kV CT images. Moreover, we proposed a double U-Net model to improve the quality of original single-energy CT images. Ninety-eight patients who underwent brain DECT scans were used to train, validate, and test the proposed DL-based model. The results showed that the low-to-high kV conversion was better than the high-to-low kV conversion. In addition, the DL-based DECT images had better signal-to-noise ratios (SNRs) than the true (original) DECT images, but at the expense of a slight loss in spatial resolution. The mean CT number differences between the true and DL-based DECT images were within [Formula: see text]1 HU. No statistically significant difference in CT number measurements was found between the true and DL-based DECT images (p>0.05). The DL-based DECT images with improved SNR could produce low-noise virtual monoenergetic images. Our preliminary results indicate that DL has the potential to generate brain DECT images using single-energy brain CT images.

From Raw Pixels to Recurrence Image for Deep Learning of Benign and Malignant Mediastinal Lymph Nodes on Computed Tomography

Lung cancer causes the most cancer deaths worldwide and has one of the lowest five-year survival rates of all cancer types. It is reported that more than half of patients with lung cancer die within one year of being diagnosed. Because mediastinal lymph node status is the most important factor for the treatment and prognosis of lung cancer, the aim of this study is to improve the predictive value in assessing the computed tomography (CT) of mediastinal lymph-node malignancy in patients with primary lung cancer. This paper introduces a new method for creating pseudo-labeled images of CT regions of mediastinal lymph nodes by using the concept of recurrence analysis in nonlinear dynamics for the transfer learning. Pseudo-labeled images of original CT images are used as input into deep-learning models. Three popular pretrained convolutional neural networks (AlexNet, SqueezeNet, and DenseNet-201) were used for the implementation of the proposed concept for the classification of benign and malignant mediastinal lymph nodes using a public CT database. In comparison with the use of the original CT data, the results show the high performance of the transformed images for the task of classification. Three pretrained convolutional neural networks that are AlexNet, SqueezeNet, and DenseNet201 were trained and tested with the transformed images. Classification accuracies and areas under the receiver operating characteristic curve obtained from the ten-fold cross-validation are 93% and 0.97, 96% and 0.99, and 100% and 1 for the SqueezeNet, AlexNet, and DenseNet201, respectively. The proposed method has the potential for differentiating benign from malignant mediastinal lymph nodes on CT, and may provide a new way for studying lung cancer using radiology imaging.

Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method.

Intertrochanteric (IT) femur fractures are the most common fractures in elderly people, and they lead to significant morbidity, mortality, and reduced quality of life. The different types of fractures require a careful definition to ensure accurate surgical planning and reduce the operation time, healing time, and number of surgical failures. In this study, a deep learning-based automatic multi-class IT fracture detection model was developed using computed tomography (CT) images and based on the AO/OTA classification method. The original CT image was resized and rearranged according to the fracture location and an unsharp masking filter was applied. A multi-class classification of nine different types of IT fractures and no fracture was performed using the faster regional-convolutional neural network (R-CNN). Bayesian optimization was also implemented to determine the optimal hyperparameter values for the faster R-CNN algorithm. In our proposed model, IT fractures classified into two classes showed an average accuracy of 0.97 ± 0.02, which was 0.90 ± 0.02 when classified into ten classes. Additionally, the detected region of interest from our proposed model showed minimum root mean square error and intersection over union values of 16.34 ± 47.01 pixels and 0.87 ± 0.12, respectively. In the future, our proposed automatic multi-class IT femur fracture detection model could allow clinicians to identify the fracture region and diagnose different types of femur fractures faster and more accurately. This will increase the probability of correct surgical treatment and minimize postoperative complications.

A voxel image-based pulmonary airflow simulation method with an automatic detection algorithm for airway outlets.

Investigations of pulmonary airflows in respiratory systems are important for the diagnostics and treatment of pulmonary diseases. For accurate prediction of the flow field in an airway, a numerical simulation must be conducted using the true geometry from computed tomography (CT) data. Flow simulation is still a difficult task because of the mesh generation process and preprocessing setup procedures. In this study, we developed a voxel image-based simulation method using an automatic detection algorithm for airway outlets to simplify the simulation process and improve its applicability in the medical field. Our approach is based on the lattice Boltzmann method with a topology analysis algorithm, which can preserve all raw information from the original CT images and give an accurate flow field inside the airways. Our method can reproduce the essential flow features inside airways, is highly efficient, and decreases the overall processing time. Thus, it has a great potential for future real-time airflow analyses to provide airflow information to medical experts. HIGHLIGHTS: This paper proposed a voxel image-based simulation method with a novel automatic outlet-selecting algorithm to calculate the velocity and pressure of physiological flows in multi-generation-branched airways. Our approach simplifies the simulation process by automatically applying the boundary conditions to large numbers of outlets and minimizes the time-consuming mesh generation process. Our proposed method has considerable potential for real-time simulations improving the applicability to patient-specific medical diagnostics and treatments.

Adversarial Uni- and Multi-modal Stream Networks for Multimodal Image Registration.

Deformable image registration between Computed Tomography (CT) images and Magnetic Resonance (MR) imaging is essential for many image-guided therapies. In this paper, we propose a novel translation-based unsupervised deformable image registration method. Distinct from other translation-based methods that attempt to convert the multimodal problem (e.g., CT-to-MR) into a unimodal problem (e.g., MR-to-MR) via image-to-image translation, our method leverages the deformation fields estimated from both: (i) the translated MR image and (ii) the original CT image in a dual-stream fashion, and automatically learns how to fuse them to achieve better registration performance. The multimodal registration network can be effectively trained by computationally efficient similarity metrics without any ground-truth deformation. Our method has been evaluated on two clinical datasets and demonstrates promising results compared to state-of-the-art traditional and learning-based methods.

Relationship between CT Numbers and Artifacts Obtained Using CT-based Attenuation Correction of PET/CT

The aim of this study was to clarify the artifacts that occurred in the non-activity signal with computed tomography (CT)-based attenuation correction (CTAC) error due to image misregistration. We used a cylindrical phantom containing a test tube with a diameter of 15 mm as the non-activity signal part. Positron emission tomography (PET) images were acquired for 30 minutes using the phantom with water in the non-activity signal part and 18F-fluoro-2-deoxy-d-glucose (18F-FDG) (5.3 kBq/ml) in the background area. CT scanning was performed by replacing the water with contrast agents at different dilutions to obtain arbitrary CT numbers (-1000 to 1000). The PET images were attenuation-corrected individually by the CT images in which the CT number of the non-activity signal part had changed. The relationship between the CT numbers and the CTAC artifact was determined by measuring the PET value in the non-activity signal part of the PET images and comparing Ci. As the CT number of the CT images increased, Ci of the artifact increased. The CT number and Ci had a correlation of y=1.48x+2.86×103 (R2 =0.99) when CTAC was performed in units of CT numbers above 0 for PET data of water (0 HU) and a correlation of y=3.15x+6.26×103 (R2 =0.97) when CTAC was performed in units of CT numbers below 0 for PET data of air (-1000 HU). Although the original CT image was air, the artifacts due to CTAC errors with different Hounsfield units showed larger changes. In particular, positive artifacts were recognized in the PET images after CTAC depending on the Hounsfield units. When the CT number was different from the original in CTAC, the PET value was different. CTAC should be performed with caution as there may be image misregistration.

Transbronchial biopsy of peripheral lung lesions using fluoroscopic guidance combined with an enhanced ray-summation display.

The aim of this study was to investigate the effectiveness of guidance assistance during transbronchial biopsy (TBB) to achieve an appropriate pathway to small and peripheral pulmonary lesions (PPLs) using a combination of fluoroscopy and specialized ray-summation (Ray-sumTBB) images, which were processed from preprocedural lung computed tomography (CT) images. To improve the visibility of the correct pathway to the PPLs, three-dimensional spatial resolution enhancement and CT number conversion processes were applied to the original CT images. The Ray-sumTBB images reconstructed from the processed CT images were used as additional guides. We compared the rates of successful tumor localization and biopsy (arrival rate) between the trial (with Ray-sumTBB) and control (without Ray-sumTBB) groups. The fluoroscopy and examination times were also compared. The arrival rate of the trial group (73.1%) was significantly better than that of the control group (42.3%) (p = 0.048). The fluoroscopy and examination times did not differ significantly between the trial and control groups. No complications were identified in the trial group. Our results suggest that Ray-sumTBB improves the diagnostic accuracy of TBB.

Deep learning-based image quality improvement for low-dose computed tomography simulation in radiation therapy.

Low-dose computed tomography (CT) is desirable for treatment planning and simulation in radiation therapy. Multiple rescanning and replanning during the treatment course with a smaller amount of dose than a single conventional full-dose CT simulation is a crucial step in adaptive radiation therapy. We developed a machine learning-based method to improve image quality of low-dose CT for radiation therapy treatment simulation. We used a residual block concept and a self-attention strategy with a cycle-consistent adversarial network framework. A fully convolution neural network with residual blocks and attention gates (AGs) was used in the generator to enable end-to-end transformation. We have collected CT images from 30 patients treated with frameless brain stereotactic radiosurgery (SRS) for this study. These full-dose images were used to generate projection data, which were then added with noise to simulate the low-mAs scanning scenario. Low-dose CT images were reconstructed from this noise-contaminated projection data and were fed into our network along with the original full-dose CT images for training. The performance of our network was evaluated by quantitatively comparing the high-quality CT images generated by our method with the original full-dose images. When mAs is reduced to 0.5% of the original CT scan, the mean square error of the CT images obtained by our method is , with respect to the original full-dose images. The proposed method successfully improved the noise, contract-to-noise ratio, and nonuniformity level to be close to those of full-dose CT images and outperforms a state-of-the-art iterative reconstruction method. Dosimetric studies show that the average differences of dose-volume histogram metrics are ( ). These quantitative results strongly indicate that the denoised low-dose CT images using our method maintains image accuracy and quality and are accurate enough for dose calculation in current CT simulation of brain SRS treatment. We also demonstrate the great potential for low-dose CT in the process of simulation and treatment planning.