Computed Tomography Datasets Research Articles

Merlin: A Vision Language Foundation Model for 3D Computed Tomography.

Over 85 million computed tomography (CT) scans are performed annually in the US, of which approximately one quarter focus on the abdomen. Given the current shortage of both general and specialized radiologists, there is a large impetus to use artificial intelligence to alleviate the burden of interpreting these complex imaging studies while simultaneously using the images to extract novel physiological insights. Prior state-of-the-art approaches for automated medical image interpretation leverage vision language models (VLMs) that utilize both the image and the corresponding textual radiology reports. However, current medical VLMs are generally limited to 2D images and short reports. To overcome these shortcomings for abdominal CT interpretation, we introduce Merlin - a 3D VLM that leverages both structured electronic health records (EHR) and unstructured radiology reports for pretraining without requiring additional manual annotations. We train Merlin using a high-quality clinical dataset of paired CT scans (6+ million images from 15,331 CTs), EHR diagnosis codes (1.8+ million codes), and radiology reports (6+ million tokens) for training. We comprehensively evaluate Merlin on 6 task types and 752 individual tasks. The non-adapted (off-the-shelf) tasks include zero-shot findings classification (31 findings), phenotype classification (692 phenotypes), and zero-shot cross-modal retrieval (image to findings and image to impressions), while model adapted tasks include 5-year chronic disease prediction (6 diseases), radiology report generation, and 3D semantic segmentation (20 organs). We perform internal validation on a test set of 5,137 CTs, and external validation on 7,000 clinical CTs and on two public CT datasets (VerSe, TotalSegmentator). Beyond these clinically-relevant evaluations, we assess the efficacy of various network architectures and training strategies to depict that Merlin has favorable performance to existing task-specific baselines. We derive data scaling laws to empirically assess training data needs for requisite downstream task performance. Furthermore, unlike conventional VLMs that require hundreds of GPUs for training, we perform all training on a single GPU. This computationally efficient design can help democratize foundation model training, especially for health systems with compute constraints. We plan to release our trained models, code, and dataset, pending manual removal of all protected health information.

Assessing CT-based Volumetric Analysis via Transfer Learning with MRI and Manual Labels for Idiopathic Normal Pressure Hydrocephalus.

Brain computed tomography (CT) is an accessible and commonly utilized technique for assessing brain structure. In cases of idiopathic normal pressure hydrocephalus (iNPH), the presence of ventriculomegaly is often neuroradiologically evaluated by visual rating and manually measuring each image. Previously, we have developed and tested a deep-learning-model that utilizes transfer learning from magnetic resonance imaging (MRI) for CT-based intracranial tissue segmentation. Accordingly, herein we aimed to enhance the segmentation of ventricular cerebrospinal fluid (VCSF) in brain CT scans and assess the performance of automated brain CT volumetrics in iNPH patient diagnostics. The development of the model used a two-stage approach. Initially, a 2D U-Net model was trained to predict VCSF segmentations from CT scans, using paired MR-VCSF labels from healthy controls. This model was subsequently refined by incorporating manually segmented lateral CT-VCSF labels from iNPH patients, building on the features learned from the initial U-Net model. The training dataset included 734 CT datasets from healthy controls paired with T1-weighted MRI scans from the Gothenburg H70 Birth Cohort Studies and 62 CT scans from iNPH patients at Uppsala University Hospital. To validate the model's performance across diverse patient populations, external clinical images including scans of 11 iNPH patients from the Universitatsmedizin Rostock, Germany, and 30 iNPH patients from the University of Alabama at Birmingham, United States were used. Further, we obtained three CT-based volumetric measures (CTVMs) related to iNPH. Our analyses demonstrated strong volumetric correlations (ϱ=0.91, p<0.001) between automatically and manually derived CT-VCSF measurements in iNPH patients. The CTVMs exhibited high accuracy in differentiating iNPH patients from controls in external clinical datasets with an AUC of 0.97 and in the Uppsala University Hospital datasets with an AUC of 0.99. CTVMs derived through deep learning, show potential for assessing and quantifying morphological features in hydrocephalus. Critically, these measures performed comparably to gold-standard neuroradiology assessments in distinguishing iNPH from healthy controls, even in the presence of intraventricular shunt catheters. Accordingly, such an approach may serve to improve the radiological evaluation of iNPH diagnosis/monitoring (i.e., treatment responses). Since CT is much more widely available than MRI, our results have considerable clinical impact.

An emerging network for COVID-19 CT-scan classification using an ensemble deep transfer learning model

Over the past few years, the widespread outbreak of COVID-19 has caused the death of millions of people worldwide. Early diagnosis of the virus is essential to control its spread and provide timely treatment. Artificial intelligence methods are often used as powerful tools to reach a COVID-19 diagnosis via computed tomography (CT) samples. In this paper, artificial intelligence-based methods are introduced to diagnose COVID-19. At first, a network called CT6-CNN is designed, and then two ensemble deep transfer learning models are developed based on Xception, ResNet-101, DenseNet-169, and CT6-CNN to reach a COVID-19 diagnosis by CT samples. The publicly available SARS-CoV-2 CT dataset is utilized for our implementation, including 2481 CT scans. The dataset is separated into 2108, 248, and 125 images for training, validation, and testing, respectively. Based on experimental results, the CT6-CNN model achieved 94.66% accuracy, 94.67% precision, 94.67% sensitivity, and 94.65% F1-score rate. Moreover, the ensemble learning models reached 99.2% accuracy. Experimental results affirm the effectiveness of designed models, especially the ensemble deep learning models, to reach a diagnosis of COVID-19.

MedYOLO: A Medical Image Object Detection Framework.

Artificial intelligence-enhanced identification of organs, lesions, and other structures in medical imaging is typically done using convolutional neural networks (CNNs) designed to make voxel-accurate segmentations of the region of interest. However, the labels required to train these CNNs are time-consuming to generate and require attention from subject matter experts to ensure quality. For tasks where voxel-level precision is not required, object detection models offer a viable alternative that can reduce annotation effort. Despite this potential application, there are few options for general-purpose object detection frameworks available for 3-D medical imaging. We report on MedYOLO, a 3-D object detection framework using the one-shot detection method of the YOLO family of models and designed for use with medical imaging. We tested this model on four different datasets: BRaTS, LIDC, an abdominal organ Computed tomography (CT) dataset, and an ECG-gated heart CT dataset. We found our models achieve high performance on a diverse range of structures even without hyperparameter tuning, reaching mean average precision (mAP) at intersection over union (IoU) 0.5 of 0.861 on BRaTS, 0.715 on the abdominal CT dataset, and 0.995 on the heart CT dataset. However, the models struggle with some structures, failing to converge on LIDC resulting in a mAP@0.5 of 0.0.

Orthognathic surgery improves compromised natural head position and pharyngeal airway in patients with Skeletal Class II or III malocclusion.

Natural head position (NHP), pharyngeal airway and maxillofacial growth pattern are correlated. The author's previous studies proved that following surgical correction of Skeletal Class II malocclusion, the over-extended NHP returned upright, and the pharyngeal airway space (PAS) dimension expanded. The present study compares the post-operative change in NHP and PAS after orthognathic surgery in Skeletal Class II and III malocclusion patients. Patients receiving orthognathic procedures to correct Skeletal Class II or III malocclusions were reviewed in this retrospective study. Pre-operative and 6-week post-operative cone-beam computed tomography datasets were collected. Variables representing the craniofacial pattern, the NHP and the PAS were measured three-dimensionally. Post-operative variables were compared with their pre-operative counterparts using either repeat-measure 2-way analysis of variance or Wilcoxon matched-pairs signed rank test. Thirty cases of Skeletal Class II malocclusion and 13 cases of Skeletal Class III malocclusion were collected. Preoperatively, the inter-group differences were significant in craniofacial pattern (68.14 ± 3.552 degree vs. 79.63 ± 2.497 degree, p < .0001) and the NHP (68.77 ± 11.02 degree vs. 82.83 ± 7.738 degree, p = .0002) while not significant in PAS; after surgery, the intergroup differences in craniofacial pattern and the NHP between groups decreased, and the PAS increased in both groups. Orthognathic surgery may improve compromised NHP and increase PAS in Skeletal Class II and III malocclusion patients.

A landmark-supervised registration framework for multi-phase CT images with cross-distillation

Objective. Multi-phase computed tomography (CT) has become a leading modality for identifying hepatic tumors. Nevertheless, the presence of misalignment in the images of different phases poses a challenge in accurately identifying and analyzing the patient’s anatomy. Conventional registration methods typically concentrate on either intensity-based features or landmark-based features in isolation, so imposing limitations on the accuracy of the registration process. Method. We establish a nonrigid cycle-registration network that leverages semi-supervised learning techniques, wherein a point distance term based on Euclidean distance between registered landmark points is introduced into the loss function. Additionally, a cross-distillation strategy is proposed in network training to further improve registration performance which incorporates response-based knowledge concerning the distances between feature points. Results. We conducted experiments using multi-centered liver CT datasets to evaluate the performance of the proposed method. The results demonstrate that our method outperforms baseline methods in terms of target registration error. Additionally, Dice scores of the warped tumor masks were calculated. Our method consistently achieved the highest scores among all the comparing methods. Specifically, it achieved scores of 82.9% and 82.5% in the hepatocellular carcinoma and the intrahepatic cholangiocarcinoma dataset, respectively. Significance. The superior registration performance indicates its potential to serve as an important tool in hepatic tumor identification and analysis.

Three-Dimensional Segmentation of Equine Paranasal Sinuses in Multidetector Computed Tomography Datasets: Preliminary Morphometric Assessment Assisted with Clustering Analysis.

The paranasal sinuses, a bilaterally symmetrical system of eight air-filled cavities, represent one of the most complex parts of the equine body. This study aimed to extract morphometric measures from computed tomography (CT) images of the equine head and to implement a clustering analysis for the computer-aided identification of age-related variations. Heads of 18 cadaver horses, aged 2-25 years, were CT-imaged and segmented to extract their volume, surface area, and relative density from the frontal sinus (FS), dorsal conchal sinus (DCS), ventral conchal sinus (VCS), rostral maxillary sinus (RMS), caudal maxillary sinus (CMS), sphenoid sinus (SS), palatine sinus (PS), and middle conchal sinus (MCS). Data were grouped into young, middle-aged, and old horse groups and clustered using the K-means clustering algorithm. Morphometric measurements varied according to the sinus position and age of the horses but not the body side. The volume and surface area of the VCS, RMS, and CMS increased with the age of the horses. With accuracy values of 0.72 for RMS, 0.67 for CMS, and 0.31 for VCS, the possibility of the age-related clustering of CT-based 3D images of equine paranasal sinuses was confirmed for RMS and CMS but disproved for VCS.

Evaluation of an Artificial Intelligence Model for Identification of Intracranial Hemorrhage Subtypes on Computed Tomography of the Head

Background Intracranial hemorrhage is a critical finding on computed tomography (CT) of the head. This study compared the accuracy of an artificial intelligence (AI) model (Annalise Enterprise CTB Triage Trauma) to consensus neuroradiologist interpretations in detecting 4 hemorrhage subtypes: acute subdural/epidural hematoma, acute subarachnoid hemorrhage, intra‐axial hemorrhage, and intraventricular hemorrhage. Methods A retrospective stand‐alone performance assessment was conducted on data sets of cases of noncontrast CT of the head acquired between 2016 and 2022 at 5 hospitals in the United States for each hemorrhage subtype. The cases were obtained from patients aged ≥18 years. The positive cases were selected on the basis of the original clinical reports using natural language processing and manual confirmation. The negative cases were selected by taking the next negative case acquired from the same CT scanner after positive cases. Each case was interpreted independently by up to 3 neuroradiologists to establish consensus interpretations. Each case was then interpreted by the AI model for the presence of the relevant hemorrhage subtype. The neuroradiologists were provided with the entire CT study. The AI model separately received thin (≤1.5 mm) and thick (&gt;1.5 and ≤5 mm) axial series as available. Results The 4 cohorts included 571 cases of acute subdural/epidural hematoma, 310 cases of acute subarachnoid hemorrhage, 926 cases of intra‐axial hemorrhage, and 199 cases of intraventricular hemorrhage. The AI model identified acute subdural/epidural hematoma with area under the curve of 0.973 (95% CI, 0.958–0.984) on thin series and 0.942 (95% CI, 0.921–0.959) on thick series; acute subarachnoid hemorrhage with area under the curve 0.993 (95% CI, 0.984–0.998) on thin series and 0.966 (95% CI, 0.945–0.983) on thick series; intraaxial hemorrhage with area under the curve of 0.969 (95% CI, 0.956–0.980) on thin series and 0.966 (95% CI, 0.953–0.976) on thick series; and intraventricular hemorrhage with area under the curve of 0.987 (95% CI, 0.969–0.997) on thin series and 0.983 (95% CI, 0.968–0.994) on thick series. Each finding had at least 1 operating point with sensitivity and specificity &gt;80%. Conclusion The assessed AI model accurately identified intracranial hemorrhage subtypes in this CT data set. Its use could assist the clinical workflow, especially through enabling triage of abnormal CTs.

WaveletDFDS-Net: A Dual Forward Denoising Stream Network for Low-Dose CT Noise Reduction

The challenge of denoising low-dose computed tomography (CT) has garnered significant research interest due to the detrimental impact of noise on CT image quality, impeding diagnostic accuracy and image-guided therapies. This paper introduces an innovative approach termed the Wavelet Domain Dual Forward Denoising Stream Network (WaveletDFDS-Net) to address this challenge. This method ingeniously combines convolutional neural networks and Transformers to leverage their complementary capabilities in feature extraction. Additionally, it employs a wavelet transform for efficient image downsampling, thereby preserving critical information while reducing computational requirements. Moreover, we have formulated a distinctive dual-domain compound loss function that significantly enhances the restoration of intricate details. The performance of WaveletDFDS-Net is assessed by comparative experiments conducted on public CT datasets, and results demonstrate its enhanced denoising effect with an SSIM of 0.9269, PSNR of 38.1343 and RMSE of 0.0130, superior to existing methods.

Noninvasive Atherosclerotic Phenotyping: The Next Frontier into Understanding the Pathobiology of Coronary Artery Disease.

Despite recent advances, coronary artery disease remains one of the leading causes of mortality worldwide. Noninvasive imaging allows atherosclerotic phenotyping by measurement of plaque burden, morphology, activity and inflammation, which has the potential to refine patient risk stratification and guide personalized therapy. This review describes the current and emerging roles of advanced noninvasive cardiovascular imaging methods for the assessment of coronary artery disease. Cardiac computed tomography enables comprehensive, noninvasive imaging of the coronary vasculature, and is used to assess luminal stenoses, coronary calcifications, and distinct adverse plaque characteristics, helping to identify patients prone to future events. Novel software tools, implementing artificial intelligence solutions, can automatically quantify and characterize atherosclerotic plaque from standard computed tomography datasets. These quantitative imaging biomarkers have been shown to improve patient risk stratification beyond clinical risk scores and current clinical interpretation of cardiac computed tomography. In addition, noninvasive molecular imaging in higher risk patients can be used to assess plaque activity and plaque thrombosis. Noninvasive imaging allows unique insight into the burden, morphology and activity of atherosclerotic coronary plaques. Such phenotyping of atherosclerosis can potentially improve individual patient risk prediction, and in the near future has the potential for clinical implementation.

Influence of learned landmark correspondences on lung CT registration.

Disease or injury may cause a change in the biomechanical properties of the lungs, which can alter lung function. Image registration can be used to measure lung ventilation and quantify volume change, which can be a useful diagnostic aid. However, lung registration is a challenging problem because of the variation in deformation along the lungs, sliding motion of the lungs along the ribs, and change indensity. Landmark correspondences have been used to make deformable image registration robust to largedisplacements. To tackle the challenging task of intra-patient lung computed tomography (CT) registration, we extend the landmark correspondence prediction model deep convolutional neural network-Match by introducing a soft mask loss term to encourage landmark correspondences in specific regions and avoid the use of a mask during inference. To produce realistic deformations to train the landmark correspondence model, we use data-driven synthetic transformations. We study the influence of these learned landmark correspondences on lung CT registration by integrating them into intensity-based registration as a distance-basedpenalty. Our results on the public thoracic CT dataset COPDgene show that using learned landmark correspondences as a soft constraint can reduce median registration error from approximately 5.46 to 4.08mm compared to standard intensity-based registration, in the absence of lungmasks. We show that using landmark correspondences results in minor improvements in local alignment, while significantly improving globalalignment.

Refining computer tomography data with super-resolution networks to increase the accuracy of respiratory flow simulations

Accurately computing the flow in the nasal cavity with computational fluid dynamics (CFD) simulations requires highly resolved computational meshes based on anatomically realistic geometries. Such geometries can only be obtained from computer tomography (CT) data with high spatial resolution, i.e., featuring a ≤1mm slice thickness. In practice, CT images are, however, recorded at a lower resolution to not expose patients to high radiation and to reduce the overall costs. To overcome this problem and to provide patients with a detailed physics-based diagnosis, e.g., for surgery planning, the potential of super-resolution networks (SRNs) to increase the CT resolution is analyzed. Therefore, an SRN is developed and trained on CT data. Its predictive performance is improved by an automated hyperparameter optimization technique. The training time is further reduced without predictive accuracy degradation by oversampling images with challenging regions. The performance of the SRN is assessed by an analysis of the reconstructed 3D surfaces of the human upper airway and by comparing results of CFD simulations. That is, surfaces and simulation results based on SRN-generated CT data at 1mm resolution are compared to those obtained from unmodified CT data-sets at low (3mm) and high (1mm) resolution, as well as from CT data interpolated to a 1mm resolution from coarse data. The findings reveal the SRN-based approach to have the lowest deviations in the surfaces and CFD results when compared to those based on the original high-resolution data. The pressure loss between the inflow (nostrils) and outflow (pharynx) regions averaged for three test patients differs by only 1.3%, compared to 8.7% and 8.8% in the coarse and interpolated cases. It is concluded that the SRN-based method is a promising tool to enhance underresolved CT data to yield reliable numerical results of respiratory flows.

Treatment duration by morphology and location of impacted maxillary canines: A cone-beam computed tomography investigation

This study aimed to identify whether patients with impacted maxillary canines take longer to treat than orthodontic patients without an impacted canine. We also sought to identify factors that are predictive of increased treatment duration in patients with impacted maxillary canines and treated by surgical exposure. A retrospective investigation of 37 patients with an impacted maxillary canine, treated by surgical exposure and fixed appliance therapy, was undertaken. In addition, an age- and sex-matched control group of 39 patients (without impacted canines) was also collected. Patient age, sex, and total treatment duration were recorded. For patients with an impacted canine, patient records and pretreatment cone-beam computed tomography datasets were assessed. Point coordinates identifying relevant landmarks were recorded, and a geometric method was used to calculate variables describing canine location, orientation, and apical morphology. Controlling for age and sex, linear regression identified a statistically significant increase in treatment duration of 46.7 and 41.5 weeks for palatal and labial/midalveolar impacted canines, respectively, vs controls (P<0.002). Age and sex of patients with impacted canines collectively affected treatment duration (P= 0.04), with females of increased age being treated faster than younger males. Rotation of the impacted canine crown had a highly significant effect on treatment duration, with every degree of rotation increasing treatment duration by 0.32 weeks (P<0.001). There was a significant degree of multicollinearity between the other radiographic variables. Collectively, radiographic variables describing canine displacement significantly prolonged treatment duration (P<0.001) and explained 29.8% of the variability in total treatment time. The apical morphology of impacted maxillary canines was significantly associated with increased treatment duration (P= 0.01) and explained 11.3% of the overall treatment variability (P= 0.01). Increased total treatment duration of surgically exposed impacted maxillary canines is associated with increasing mesiopalatal canine crown rotation, worsening displacement, and hooked apical morphology.

MIB-Net: Balance the mutual information flow in deep learning network for multi-dimensional segmentation of COVID-19 CT images

Computed tomography (CT) based lung screening followed by computer-vision-based automated segmentation has great potential in early diagnosis of COVID-19, but low sample size and high patient-to-patient variability of infection characteristics are two of the major challenges. We propose a novel Mutual Information Balanced Net (MIB-Net) with three distinctive features. First, it captures mutual information across feature maps, CT scans, and predicted segmentation labels to optimize the contribution of different feature maps for better segmentation accuracy. Second, it incorporates a three-dimensional structural prior of CT scans based on predicted segmentation of neighboring CT slices to improve accuracy. Third, a novel Weight Optimization Filter (WOF) is used to adjust the proportion of 3D structural prior to be incorporated, which further improves the segmentation accuracy. Experimental results show that the proposed approach outperforms state-of-the-art biomedical image segmentation models by 7.87%, 3.85%, and 5.01% in Dice Score, respectively for three popular COVID-19 CT datasets. The better performance of the proposed approach suggests that it may serve as an attractive alternative to Lung CT-based screening for COVID-19.

Prediction of recurrence-free survival in lung adenocarcinoma based on self-supervised pre-training and multi-task learning

Computed tomography (CT) imaging is a vital tool for the diagnosis and assessment of lung adenocarcinoma, and using CT images to predict the recurrence-free survival (RFS) of lung adenocarcinoma patients post-surgery is of paramount importance in tailoring postoperative treatment plans. Addressing the challenging task of accurate RFS prediction using CT images, this paper introduces an innovative approach based on self-supervised pre-training and multi-task learning. We employed a self-supervised learning strategy known as "image transformation to image restoration" to pretrain a 3D-UNet network on publicly available lung CT datasets to extract generic visual features from lung images. Subsequently, we enhanced the network's feature extraction capability through multi-task learning involving segmentation and classification tasks, guiding the network to extract image features relevant to RFS. Additionally, we designed a multi-scale feature aggregation module to comprehensively amalgamate multi-scale image features, and ultimately predicted the RFS risk score for lung adenocarcinoma with the aid of a feed-forward neural network. The predictive performance of the proposed method was assessed by ten-fold cross-validation. The results showed that the consistency index (C-index) of the proposed method for predicting RFS and the area under curve (AUC) for predicting whether recurrence occurs within three years reached 0.691 ± 0.076 and 0.707 ± 0.082, respectively, and the predictive performance was superior to that of existing methods. This study confirms that the proposed method has the potential of RFS prediction in lung adenocarcinoma patients, which is expected to provide a reliable basis for the development of individualized treatment plans.

High throughput automated characterization of enamel microstructure using synchrotron tomography and optical flow imaging

The remarkable damage-tolerance of enamel has been attributed to its hierarchical microstructure and the organized bands of decussated rods. A thorough characterization of the microscale rod evolution within the enamel is needed to elucidate this complex structure. While prior efforts in this area have made use of single particle tracking to track a single rod evolution to various degrees of success, such a process can be both computationally and labor intensive, limited to the evolution path of a single rod, and is therefore prone to error from potentially tracking outliers. Particle image velocimetry (PIV) is a well-established algorithm to derive field information from image sequences for processes that are time-dependent, such as fluid flows and structural deformation. In this work, we demonstrate the use of PIV in extracting the full-field microstructural distribution of rods within the enamel. Enamel samples from a wild African lion were analyzed using high-energy synchrotron X-ray micro-tomography. Results from the PIV analysis provide sufficient full-field information to reconstruct the growth of individual rods that can potentially enable rapid analysis of complex microstructures from high resolution synchrotron datasets. Such information can serve as a template for designing damage-tolerant bioinspired structures for advanced manufacturing. Statement of significanceThorough characterization and analysis of biological microstructures (viz. dental enamel) allows us to understand the basis of their excellent mechanical properties. Prior efforts have successfully replicated these microstructures via single particle tracking, but the process is computationally and labor intensive. In this work, optical flow imaging algorithms were used to extract full-field microstructural distribution of enamel rods from synchrotron X-ray computed tomography datasets, and a field method was used to reconstruct the growth of individual rods. Such high throughput information allows for the rapid production/prototyping and advanced manufacturing of damage-tolerant bioinspired structures for specific engineering applications. Furthermore, the algorithms used herein are freely available and open source to broaden the availability of the proposed workflow to the general scientific community.

Wasserstein-GAN-based noise reduction for preserving anatomical structures in low-dose CT

The purpose of low-dose computed tomography (LDCT) is to reduce patient's radiation dose. However, LDCT imaging often suffers from low image quality due to computed tomography (CT) noise patterns. While conventional methods such as iterative reconstruction and image processing have been studied to address this issue, their clinical application remains limited due to the degradation of critical anatomical structures. Recently, deep-learning-based noise reduction approaches have shown promising results in improving the image quality of LDCT. Particularly, generative adversarial network (GAN)-based noise reduction methods have produced outstanding results, approaching the quality of normal-dose CT (NDCT) images. In this study, we propose a hybrid framework based on Wasserstein-GAN with the non-subsampled contourlet transform for enhanced noise reduction in LDCT. The aim is to meticulously address CT noise patterns and generate CT images similar to NDCT. We implemented the proposed algorithm and conducted an experiment using clinical abdominal CT datasets to demonstrate its feasibility. According to our results, the proposed method effectively reduced noise components in LDCT images while preserving the anatomical structures of NDCT. This was confirmed through both qualitative and quantitative evaluations. The implications of our findings for clinical practice are significant, as the proposed method can improve the accuracy of LDCT and reduce radiation dose to patients, ultimately enhancing patient safety.

Exploration of postural effects on the external jugular and diploic venous system using upright computed tomography scanning.

Few studies have investigated the influence of posture on the external jugular and diploic venous systems in the head and cranial region. In this study, we aimed to investigate the effects of posture on these systems using upright computed tomography (CT) scanning. This study retrospectively analysed an upright CT dataset from a previous prospective study. In each patient, the diameters of the vessels in three external jugular tributaries and four diploic veins were measured using CT digital subtraction venography in both supine and sitting positions. Amongst the 20 cases in the original dataset, we eventually investigated 19 cases due to motion artifacts in 1 case. Compared with the supine position, most of the external jugular tributaries collapsed, and the average size significantly decreased in the sitting position (decreased by 22-49% on average). In contrast, most of the diploic veins, except the occipital diploic veins, tended to increase or remain unchanged (increased by 12-101% on average) in size in the sitting position compared with the supine position. However, the changes in the veins associated with this positional shift were not uniform; in approximately 5-30% of the cases, depending on each vein, an opposite trend was observed. Compared to the supine position, the contribution of external jugular tributaries to head venous drainage decreased in the sitting position, whilst most diploic veins maintained their contribution. These results could enhance our understanding of the physiology and pathophysiology of the head region in upright and sitting positions.

Recurrent feature propagation and edge skip-connections for automatic abdominal organ segmentation

Automatic segmentation of abdominal organs in computed tomography (CT) images can support radiation therapy and image-guided surgery workflows. Development of such automatic solutions remains challenging mainly owing to complex organ interactions and blurry boundaries in CT images. To address these issues, we focus on effective spatial context modeling and explicit edge segmentation priors. Accordingly, we propose a 3D network with four main components trained end-to-end including shared encoder, edge detector, decoder with edge skip-connections (ESCs) and recurrent feature propagation head (RFP-Head). To capture wide-range spatial dependencies, the RFP-Head propagates and harvests local features through directed acyclic graphs (DAGs) formulated with recurrent connections in an efficient slice-wise manner, with regard to spatial arrangement of image units. To leverage edge information, the edge detector learns edge prior knowledge specifically tuned for semantic segmentation by exploiting intermediate features from the encoder with the edge supervision. The ESCs then aggregate the edge knowledge with multi-level decoder features to learn a hierarchy of discriminative features explicitly modeling complementarity between organs’ interiors and edges for segmentation. We conduct extensive experiments on challenging abdominal CT datasets with eight annotated organs. Experimental results show that the proposed network outperforms several state-of-the-art models, especially for the segmentation of small and complicated structures (gallbladder, esophagus, stomach, pancreas and duodenum).

Noise aware content-noise complementary GAN with local and global discrimination for low-dose CT denoising

In response to rising concerns over radiation exposure in computed tomography (CT) imaging, effective denoising methods for low-dose CT (LDCT) images are crucial. In recent years, the use of deep learning techniques especially generative adversarial networks (GANs) significantly enhanced the efficiency of LDCT denoising methods, surpassing traditional methods. However, GAN-based denoising methods often face challenges in preserving structural consistency and fine details. This study introduces a novel GAN framework with three accretions to enhance the effectiveness of LDCT denoising. Firstly, our generator is designed to leverage a complementary learning scheme between image noise and image content via two distinct paths. One path focuses on exploring the anatomical information of the image, while the second path is dedicated to learning the noise pattern. This complementary learning scheme provides stable noise cancellation while preserving the maximum structural information of the image. Subsequently, we propose a novel noise-conscious mean absolute error loss to address the challenge posed by the non-stationary characteristic of CT noise. In contrast to conventional MAE loss, this loss attentively prioritizes the different parts of the image based on the local distribution of noise in that region. We also incorporate a gradient-domain loss into the loss function, which inspires the generator to preserve precise image details through explicit guidance. Finally, this study adopted a U-Net-based design for the discriminator to better regularize the model by discriminating between the clean image and the denoised image at both global and local levels. The merit of this discriminator is that it can better adapt to the non-stationary environment of GAN training and guide the generator to produce denoised images that are locally and globally consistent. Our thorough experiments using abdominal CT and lung CT datasets demonstrate the superior performance of our approach compared to state-of-the-art methods.