Automatic registration of dental CT and 3D scanned model using deep split jaw and surface curvature.

BACKGROUND AND OBJECTIVES: In the medical field, various image registration applications have been studied. In dentistry, the registration of computed tomography (CT) volume data and 3D optically scanned models is essential for various clinical applications, including orthognathic surgery, implant surgical planning, and augmented reality. Our purpose was to present a fully automatic registration method of dental CT data and 3D scanned models. METHODS: We use a 2D convolutional neural network to regress a curve splitting the maxilla (i.e., upper jaw) and mandible (i.e., lower jaw) and the points specifying the front and back ends of the crown from the CT data. Using this regressed information, we extract the point cloud and vertices corresponding to the tooth crown from the CT and scanned data, respectively. We introduce a novel metric, called curvature variance of neighbor (CVN), to discriminate between highly fluctuating and smoothly varying regions of the tooth crown. The registration based on CVN enables more accurate fine registration while reducing the effects of metal artifacts. Moreover, the proposed method does not require any preprocessing such as extracting the iso-surface for the tooth crown from the CT data, thereby significantly reducing the computation time. RESULTS: We evaluated the proposed method with the comparison to several promising registration techniques. Our experimental results using three datasets demonstrated that the proposed method exhibited higher registration accuracy (i.e., 2.85, 1.92, and 7.73 times smaller distance errors for individual datasets) and smaller computation time (i.e., 4.12 times faster registration) than one of the state-of-the-art methods. Moreover, the proposed method worked considerably well for partially scanned data, whereas other methods suffered from the unbalancing of information between the CT and scanned data. CONCLUSIONS: The proposed method was able to perform fully automatic and highly accurate registration of dental CT data and 3D scanned models, even with severe metal artifacts. In addition, it could achieve fast registration because it did not require any preprocessing for iso-surface reconstruction from the CT data.

Voxel-wise adversarial semi-supervised learning for medical image segmentation.

BACKGROUND AND OBJECTIVE: Semi-supervised learning for medical image segmentation is an important area of research for alleviating the huge cost associated with the construction of reliable large-scale annotations in the medical domain. Recent semi-supervised approaches have demonstrated promising results by employing consistency regularization, pseudo-labeling techniques, and adversarial learning. These methods primarily attempt to learn the distribution of labeled and unlabeled data by enforcing consistency in the predictions or embedding context. However, previous approaches have focused only on local discrepancy minimization or context relations across single classes. METHODS: In this paper, we introduce a novel adversarial learning-based semi-supervised segmentation method that effectively embeds both local and global features from multiple hidden layers and learns context relations between multiple classes. Our voxel-wise adversarial learning method utilizes a voxel-wise feature discriminator, which considers multilayer voxel-wise features (involving both local and global features) as an input by embedding class-specific voxel-wise feature distribution. Furthermore, our previous representation learning method is improved by overcoming information loss and learning stability problems, which enables rich representations of labeled data. RESULT: In the experiments, we used the Left Atrial Segmentation Challenge dataset and the Abdominal Multi-Organ dataset to prove the effectiveness of our method in both single class and multiclass segmentation. The experimental results demonstrate that our method outperforms current best-performing state-of-the-art semi-supervised learning approaches. Our proposed adversarial learning-based semi-supervised segmentation method successfully leveraged unlabeled data to improve the network performance by 2% in Dice score coefficient for multi-organ dataset. CONCLUSION: We compare our approach to a wide range of medical datasets, and showed our method can be adapted to embed class-specific features. Furthermore, visual interpretation of the feature space demonstrates that our proposed method enables a well-distributed and separated feature space from both labeled and unlabeled data, which improves the overall prediction results.

Cardiac segmentation on CT Images through shape-aware contour attentions.

BACKGROUND AND OBJECTIVE: Cardiac segmentation of atriums, ventricles, and myocardium in computed tomography (CT) images is an important first-line task for presymptomatic cardiovascular disease diagnosis. In several recent studies, deep learning models have shown significant breakthroughs in medical image segmentation tasks. Unlike other organs such as the lungs and liver, the cardiac organ consists of multiple substructures, i.e., ventricles, atriums, aortas, arteries, veins, and myocardium. These cardiac substructures are proximate to each other and have indiscernible boundaries (i.e., homogeneous intensity values), making it difficult for the segmentation network focus on the boundaries between the substructures. METHODS: In this paper, to improve the segmentation accuracy between proximate organs, we introduce a novel model to exploit shape and boundary-aware features. We primarily propose a shape-aware attention module, that exploits distance regression, which can guide the model to focus on the edges between substructures so that it can outperform the conventional contour-based attention method. RESULTS: In the experiments, we used the Multi-Modality Whole Heart Segmentation dataset that has 20 CT cardiac images for training and validation, and 40 CT cardiac images for testing. The experimental results show that the proposed network produces more accurate results than state-of-the-art networks by improving the Dice similarity coefficient score by 4.97%. CONCLUSION: Our proposed shape-aware contour attention mechanism demonstrates that distance transformation and boundary features improve the actual attention map to strengthen the responses in the boundary area. Moreover, our proposed method significantly reduces the false-positive responses of the final output, resulting in accurate segmentation.

Voxel-level Siamese Representation Learning for Abdominal Multi-Organ Segmentation.

BACKGROUND AND OBJECTIVE: Recent works in medical image segmentation have actively explored various deep learning architectures or objective functions to encode high-level features from volumetric data owing to limited image annotations. However, most existing approaches tend to ignore cross-volume global context and define context relations in the decision space. In this work, we propose a novel voxel-level Siamese representation learning method for abdominal multi-organ segmentation to improve representation space. METHODS: The proposed method enforces voxel-wise feature relations in the representation space for leveraging limited datasets more comprehensively to achieve better performance. Inspired by recent progress in contrastive learning, we suppressed voxel-wise relations from the same class to be projected to the same point without using negative samples. Moreover, we introduce a multi-resolution context aggregation method that aggregates features from multiple hidden layers, which encodes both the global and local contexts for segmentation. RESULTS: Our experiments on the multi-organ dataset outperformed the existing approaches by 2% in Dice score coefficient. The qualitative visualizations of the representation spaces demonstrate that the improvements were gained primarily by a disentangled feature space. CONCLUSION: Our new representation learning method successfully encoded high-level features in the representation space by using a limited dataset, which showed superior accuracy in the medical image segmentation task compared to other contrastive loss-based methods. Moreover, our method can be easily applied to other networks without using additional parameters in the inference.

Review of Methodological Approaches to Human Milk Small Extracellular Vesicle Proteomics.

Proteomics can map extracellular vesicles (EVs), including exosomes, across disease states between organisms and cell types. Due to the diverse origin and cargo of EVs, tailoring methodological and analytical techniques can support the reproducibility of results. Proteomics scans are sensitive to in-sample contaminants, which can be retained during EV isolation procedures. Contaminants can also arise from the biological origin of exosomes, such as the lipid-rich environment in human milk. Human milk (HM) EVs and exosomes are emerging as a research interest in health and disease, though the experimental characterization and functional assays remain varied. Past studies of HM EV proteomes have used data-dependent acquisition methods for protein detection, however, improvements in data independent acquisition could allow for previously undetected EV proteins to be identified by mass spectrometry. Depending on the research question, only a specific population of proteins can be compared and measured using isotope and other labelling techniques. In this review, we summarize published HM EV proteomics protocols and suggest a methodological workflow with the end-goal of effective and reproducible analysis of human milk EV proteomes.

Liver segmentation in abdominal CT images via auto-context neural network and self-supervised contour attention.

OBJECTIVE: Accurate image segmentation of the liver is a challenging problem owing to its large shape variability and unclear boundaries. Although the applications of fully convolutional neural networks (CNNs) have shown groundbreaking results, limited studies have focused on the performance of generalization. In this study, we introduce a CNN for liver segmentation on abdominal computed tomography (CT) images that focus on the performance of generalization and accuracy. METHODS: To improve the generalization performance, we initially propose an auto-context algorithm in a single CNN. The proposed auto-context neural network exploits an effective high-level residual estimation to obtain the shape prior. Identical dual paths are effectively trained to represent mutual complementary features for an accurate posterior analysis of a liver. Further, we extend our network by employing a self-supervised contour scheme. We trained sparse contour features by penalizing the ground-truth contour to focus more contour attentions on the failures. RESULTS: We used 180 abdominal CT images for training and validation. Two-fold cross-validation is presented for a comparison with the state-of-the-art neural networks. The experimental results show that the proposed network results in better accuracy when compared to the state-of-the-art networks by reducing 10.31% of the Hausdorff distance. Novel multiple N-fold cross-validations are conducted to show the best performance of generalization of the proposed network. CONCLUSION AND SIGNIFICANCE: The proposed method minimized the error between training and test images more than any other modern neural networks. Moreover, the contour scheme was successfully employed in the network by introducing a self-supervising metric.

Individual tooth detection and identification from dental panoramic X-ray images via point-wise localization and distance regularization.

Dental panoramic X-ray imaging is a popular diagnostic method owing to its very small dose of radiation. For an automated computer-aided diagnosis system in dental clinics, automatic detection and identification of individual teeth from panoramic X-ray images are critical prerequisites. In this study, we propose a point-wise tooth localization neural network by introducing a spatial distance regularization loss. The proposed network initially performs center point regression for all the anatomical teeth (i.e., 32 points), which automatically identifies each tooth. A novel distance regularization penalty is employed on the 32 points by considering L2 regularization loss of Laplacian on spatial distances. Subsequently, teeth boxes are individually localized using a multitask neural network on a patch basis. A multitask offset training is employed on the final output to improve the localization accuracy. Our method successfully localizes not only the existing teeth but also missing teeth; consequently, highly accurate detection and identification are achieved. The experimental results demonstrate that the proposed algorithm outperforms state-of-the-art approaches by increasing the average precision of teeth detection by 15.71 % compared to the best performing method. The accuracy of identification achieved a precision of 0.997 and recall value of 0.972. Moreover, the proposed network does not require any additional identification algorithm owing to the preceding regression of the fixed 32 points regardless of the existence of the teeth.

Automatic Registration Between Dental Cone-Beam CT and Scanned Surface via Deep Pose Regression Neural Networks and Clustered Similarities.

Computerized registration between maxillofacial cone-beam computed tomography (CT) images and a scanned dental model is an essential prerequisite for surgical planning for dental implants or orthognathic surgery. We propose a novel method that performs fully automatic registration between a cone-beam CT image and an optically scanned model. To build a robust and automatic initial registration method, deep pose regression neural networks are applied in a reduced domain (i.e., two-dimensional image). Subsequently, fine registration is performed using optimal clusters. A majority voting system achieves globally optimal transformations while each cluster attempts to optimize local transformation parameters. The coherency of clusters determines their candidacy for the optimal cluster set. The outlying regions in the iso-surface are effectively removed based on the consensus among the optimal clusters. The accuracy of registration is evaluated based on the Euclidean distance of 10 landmarks on a scanned model, which have been annotated by experts in the field. The experiments show that the registration accuracy of the proposed method, measured based on the landmark distance, outperforms the best performing existing method by 33.09%. In addition to achieving high accuracy, our proposed method neither requires human interactions nor priors (e.g., iso-surface extraction). The primary significance of our study is twofold: 1) the employment of lightweight neural networks, which indicates the applicability of neural networks in extracting pose cues that can be easily obtained and 2) the introduction of an optimal cluster-based registration method that can avoid metal artifacts during the matching procedures.

Pose-aware instance segmentation framework from cone beam CT images for tooth segmentation.

Individual tooth segmentation from cone beam computed tomography (CBCT) images is an essential prerequisite for an anatomical understanding of orthodontic structures in several applications, such as tooth reformation planning and implant guide simulations. However, the presence of severe metal artifacts in CBCT images hinders the accurate segmentation of each individual tooth. In this study, we propose a neural network for pixel-wise labeling to exploit an instance segmentation framework that is robust to metal artifacts. Our method comprises of three steps: 1) image cropping and realignment by pose regressions, 2) metal-robust individual tooth detection, and 3) segmentation. We first extract the alignment information of the patient by pose regression neural networks to attain a volume-of-interest (VOI) region and realign the input image, which reduces the inter-overlapping area between tooth bounding boxes. Then, individual tooth regions are localized within a VOI realigned image using a convolutional detector. We improved the accuracy of the detector by employing non-maximum suppression and multiclass classification metrics in the region proposal network. Finally, we apply a convolutional neural network (CNN) to perform individual tooth segmentation by converting the pixel-wise labeling task to a distance regression task. Metal-intensive image augmentation is also employed for a robust segmentation of metal artifacts. The result shows that our proposed method outperforms other state-of-the-art methods, especially for teeth with metal artifacts. Our method demonstrated 5.68% and 30.30% better accuracy in the F1 score and aggregated Jaccard index, respectively, when compared to the best performing state-of-the-art algorithms. The major implication of the proposed method is two-fold: 1) an introduction of pose-aware VOI realignment followed by a robust tooth detection and 2) a metal-robust CNN framework for accurate tooth segmentation.

Deeply self-supervised contour embedded neural network applied to liver segmentation.

OBJECTIVE: Herein, a neural network-based liver segmentation algorithm is proposed, and its performance was evaluated using abdominal computed tomography (CT) images. METHODS: A fully convolutional network was developed to overcome the volumetric image segmentation problem. To guide a neural network to accurately delineate a target liver object, the network was deeply supervised by applying the adaptive self-supervision scheme to derive the essential contour, which acted as a complement with the global shape. The discriminative contour, shape, and deep features were internally merged for the segmentation results. RESULTS AND CONCLUSION: 160 abdominal CT images were used for training and validation. The quantitative evaluation of the proposed network was performed through an eight-fold cross-validation. The result showed that the method, which uses the contour feature, segmented the liver more accurately than the state-of-the-art with a 2.13% improvement in the dice score. SIGNIFICANCE: In this study, a new framework was introduced to guide a neural network and learn complementary contour features. The proposed neural network demonstrates that the guided contour features can significantly improve the performance of the segmentation task.

Accurate liver vessel segmentation via active contour model with dense vessel candidates.

BACKGROUND AND OBJECTIVE: The purpose of this paper is to propose a fully automated liver vessel segmentation algorithm including portal vein and hepatic vein on contrast enhanced CTA images. METHODS: First, points of a vessel candidate region are extracted from 3-dimensional (3D) CTA image. To generate accurate points, we reduce 3D segmentation problem to 2D problem by generating multiple maximum intensity (MI) images. After the segmentation of MI images, we back-project pixels to the original 3D domain. We call these voxels as vessel candidates (VCs). A large set of MI images can produce very dense and accurate VCs. Finally, for the accurate segmentation of a vessel region, we propose a newly designed active contour model (ACM) that uses the original image, vessel probability map from dense VCs, and the good prior of an initial contour. RESULTS: We used 55 abdominal CTAs for a parameter study and a quantitative evaluation. We evaluated the performance of the proposed method comparing with other state-of-the-art ACMs for vascular images applied directly to the original data. The result showed that our method successfully segmented vascular structure 25%-122% more accurately than other methods without any extra false positive detection. CONCLUSION: Our model can generate a smooth and accurate boundary of the vessel object and easily extract thin and weak peripheral branch vessels. The proposed approach can automatically segment a liver vessel without any manual interaction. The detailed result can aid further anatomical studies.

Creation of a functional graded nanobiomembrane using a new electrospinning system for drug release control and an in vitro validation of drug release behavior of the coating membrane.

Functional graded nanobiomembranes (FGMs) with multiple layers were created by a single process using a novel electrospinning system equipped with a generator and a PCI type motion board as a controller in order to control the drug release rate. By varying physical apparatus-related parameters such as nozzle-to-collector distance via a robot and the collector moving velocity the FGMs were formed. For the membrane base layer, poly-(ε-caprolactone) (PCL) with paclitaxel (PTX) was dissolved in a solvent (dichloromethane, N,N-dimethylformamide) and electrospun. For the top layers, the PCL solution was electrospun according to the distance and FGM system parameters, which can move the collector location at a constant ratio. It was observed that pore size, porosity, and permeability were higher when the membrane was spun at the far distance. The top surface of FGM is more porous, rougher, more permeable, and more hydrophilic so as to be active to the surrounding tissue cells. Meanwhile, the porous inside membrane was as low as the membrane spun at a close distance. Thus it induced a slow drug release due to the internal structure of FGM, which is considered to be very effective for slow drug release as well as bioactivity and bioconductivity.

Agenesis of the dorsal pancreas: a case report and review of the literature.

Partial or complete agenesis of the dorsal pancreas is a rare congenital anomaly that results from the embryological failure of the dorsal pancreatic bud to form the body and tail of the pancreas. To date, four cases have been reported in Korea. We report an additional case; a 25-year-old woman presented with diabetes mellitus and abdominal pain. Abdominal computed tomography (CT) revealed a normal-appearing pancreatic head, but the body and tail were not visualized. Endoscopic cholangiopancreatogram (ERCP) revealed a short pancreatic duct in the uncinate process and the head and the duct of Santorini draining into the minor papilla. Abdominal magnetic resonance imaging (MRI) findings were similar to the CT and ERCP results. The patient was diagnosed with partial agenesis of the dorsal pancreas by CT, ERCP and MRI.

The effects of thiazolidinedione treatment on the regulations of aquaglyceroporins and glycerol kinase in OLETF rats.

Aquaporins (AQPs) that transport glycerol in addition to water are classified as aquaglyceroporins (AQP3, 7, 9). AQP7 in the adipose tissue and AQP9 in the liver may coordinately contribute to the increase in hepatic gluconeogenesis in states of insulin resistance. Thiazolidinedione (TZD) has been shown to increase adipose AQP7 and induce glycerol kinase (GlyK) which is nearly absent in adipocytes. In the present study, we analyzed both GlyK and AQP gene expression in adipose and hepatic tissues, and AQP3 in kidneys from Long-Evans Tokushima Otsuka (LETO), Otsuka Long-Evans Tokushima Fatty (OLETF), and rosiglitazone (RSG)-treated OLETF (RSG-OLETF) rats. We also evaluated AQP9 protein expression in cultured human hepatoma cells treated with oleic acid, Wy14643, or RSG. A 2-week RSG treatment increased AQP7 mRNA levels in the mesenteric fat, but not in the epididymal fat of OLETF rats. Rosiglitazone treatment markedly increased GlyK expression in both fat depots, with a greater increase in the mesenteric fat. The magnitudes of GlyK induction by RSG were greater than that of AQP7 in both adipose tissues (P < .05, each). AQP9 and GlyK levels in the liver were not affected by RSG treatment in OLETF rats. Oleic acid and Wy14643 upregulated AQP9 protein expression in cultured human hepatoma cells in a dose-dependent manner. AQP3 mRNA levels tended to increase in the outer medulla of the RSG-OLETF rats. These results indicate that in the adipose tissue TZD has an important role in the glycerol metabolic pathway through the regulation of AQP and GlyK, especially by GlyK induction. Free fatty acids may directly enhance glycerol availability in the liver via the upregulation of AQP9 levels. Renal AQP3 may be related to the fluid retention caused by TZD.