Application of deep learning in isolated tooth identification.

BACKGROUND: Teeth identification has a pivotal role in the dental curriculum and provides one of the important foundations of clinical practice. Accurately identifying teeth is a vital aspect of dental education and clinical practice, but can be challenging due to the anatomical similarities between categories. In this study, we aim to explore the possibility of using a deep learning model to classify isolated tooth by a set of photographs. METHODS: A collection of 5,100 photographs from 850 isolated human tooth specimens were assembled to serve as the dataset for this study. Each tooth was carefully labeled during the data collection phase through direct observation. We developed a deep learning model that incorporates the state-of-the-art feature extractor and attention mechanism to classify each tooth based on a set of 6 photographs captured from multiple angles. To increase the validity of model evaluation, a voting-based strategy was applied to refine the test set to generate a more reliable label, and the model was evaluated under different types of classification granularities. RESULTS: This deep learning model achieved top-3 accuracies of over 90% in all classification types, with an average AUC of 0.95. The Cohen's Kappa demonstrated good agreement between model prediction and the test set. CONCLUSIONS: This deep learning model can achieve performance comparable to that of human experts and has the potential to become a valuable tool for dental education and various applications in accurately identifying isolated tooth.

CrossZoom: Simultaneous Motion Deblurring and Event Super-Resolving.

Even though the collaboration between traditional and neuromorphic event cameras brings prosperity to frame-event based vision applications, the performance is still confined by the resolution gap crossing two modalities in both spatial and temporal domains. This paper is devoted to bridging the gap by increasing the temporal resolution for images, i.e., motion deblurring, and the spatial resolution for events, i.e., event super-resolving, respectively. To this end, we introduce CrossZoom, a novel unified neural Network (CZ-Net) to jointly recover sharp latent sequences within the exposure period of a blurry input and the corresponding High-Resolution (HR) events. Specifically, we present a multi-scale blur-event fusion architecture that leverages the scale-variant properties and effectively fuses cross-modal information to achieve cross-enhancement. Attention-based adaptive enhancement and cross-interaction prediction modules are devised to alleviate the distortions inherent in Low-Resolution (LR) events and enhance the final results through the prior blur-event complementary information. Furthermore, we propose a new dataset containing HR sharp-blurry images and the corresponding HR-LR event streams to facilitate future research. Extensive qualitative and quantitative experiments on synthetic and real-world datasets demonstrate the effectiveness and robustness of the proposed method. Codes and datasets are released at https://bestrivenzc.github.io/CZ-Net/.

A new resorcylic acid lactone from the endophytic fungus Chaetosphaeronema sp.

A new 14-membered resorcylic acid lactone (RAL14), chaetolactone A (1), along with three known ones (2-4), was obtained from the fermentation of the soil-derived fungus Chaetosphaeronema sp. SSJZ001. Their structures were established based on extensive spectroscopic data analyses (UV, IR, HRESIMS, 1D, and 2D NMR),13C NMR chemical shifts calculations coupled with the DP4+ probability method, theoretical calculations of ECD spectra, as well as X-ray diffraction analysis. All compounds were evaluated for their cytotoxic effects against A549, HO-8910, and MCF-7 cell lines.

Oleanane and 30-noroleanane triterpenoids from the roots of Paeonia lactiflora.

An investigation of EtOAc extract from the roots of Paeonia lactiflora yielded three new 30-noroleanane triterpenoids paeonenoides L-N (1-3) and one new oleanane triterpenoid paeonenoide O (4) together with 7 known compounds (5-11). Extensive spectrographic experiments were applied to identify the structures of 1-4, and their absolute configurations were unambiguously determined by theoretical calculations of ECD spectra, as well as the single-crystal X-ray diffraction analysis. Compounds 8, 9 and 10 were isolated from the Paeonia genus for the first time. Moreover, compounds 8, 9 and 11 showed inhibitory activities against LPS-induced nitric oxide (NO) production in RAW264.7 macrophages with the IC50 values of 72. 17 ± 4.74, 30.02 ± 2.03 and 28.34 ± 1.85 µM, respectively.

Dual RNA sequencing of Helicobacter pylori and host cell transcriptomes reveals ontologically distinct host-pathogen interaction.

Helicobacter pylori is a highly successful pathogen that poses a substantial threat to human health. However, the dynamic interaction between H. pylori and the human gastric epithelium has not been fully investigated. In this study, using dual RNA sequencing technology, we characterized a cytotoxin-associated gene A (cagA)-modulated bacterial adaption strategy by enhancing the expression of ATP-binding cassette transporter-related genes, metQ and HP_0888, upon coculturing with human gastric epithelial cells. We observed a general repression of electron transport-associated genes by cagA, leading to the activation of oxidative phosphorylation. Temporal profiling of host mRNA signatures revealed the downregulation of multiple splicing regulators due to bacterial infection, resulting in aberrant pre-mRNA splicing of functional genes involved in the cell cycle process in response to H. pylori infection. Moreover, we demonstrated a protective effect of gastric H. pylori colonization against chronic dextran sulfate sodium (DSS)-induced colitis. Mechanistically, we identified a cluster of propionic and butyric acid-producing bacteria, Muribaculaceae, selectively enriched in the colons of H. pylori-pre-colonized mice, which may contribute to the restoration of intestinal barrier function damaged by DSS treatment. Collectively, this study presents the first dual-transcriptome analysis of H. pylori during its dynamic interaction with gastric epithelial cells and provides new insights into strategies through which H. pylori promotes infection and pathogenesis in the human gastric epithelium. IMPORTANCE: Simultaneous profiling of the dynamic interaction between Helicobacter pylori and the human gastric epithelium represents a novel strategy for identifying regulatory responses that drive pathogenesis. This study presents the first dual-transcriptome analysis of H. pylori when cocultured with gastric epithelial cells, revealing a bacterial adaptation strategy and a general repression of electron transportation-associated genes, both of which were modulated by cytotoxin-associated gene A (cagA). Temporal profiling of host mRNA signatures dissected the aberrant pre-mRNA splicing of functional genes involved in the cell cycle process in response to H. pylori infection. We demonstrated a protective effect of gastric H. pylori colonization against chronic DSS-induced colitis through both in vitro and in vivo experiments. These findings significantly enhance our understanding of how H. pylori promotes infection and pathogenesis in the human gastric epithelium and provide evidence to identify targets for antimicrobial therapies.

Detecting Line Segments in Motion-Blurred Images With Events.

Making line segment detectors more reliable under motion blurs is one of the most important challenges for practical applications, such as visual SLAM and 3D line mapping. Existing line segment detection methods face severe performance degradation for accurately detecting and locating line segments when motion blur occurs. While event data shows strong complementary characteristics to images for minimal blur and edge awareness at high-temporal resolution, potentially beneficial for reliable line segment recognition. To robustly detect line segments over motion blurs, we propose to leverage the complementary information of images and events. Specifically, we first design a general frame-event feature fusion network to extract and fuse the detailed image textures and low-latency event edges, which consists of a channel-attention-based shallow fusion module and a self-attention-based dual hourglass module. We then utilize the state-of-the-art wireframe parsing networks to detect line segments on the fused feature map. Moreover, due to the lack of line segment detection datasets with pairwise motion-blurred images and events, we contribute two datasets, i.e., synthetic FE-Wireframe and realistic FE-Blurframe, for network training and evaluation. Extensive analyses on the component configurations demonstrate the design effectiveness of our fusion network. When compared to the state-of-the-arts, the proposed approach achieves the highest detection accuracy while maintaining comparable real-time performance. In addition to being robust to motion blur, our method also exhibits superior performance for line detection under high dynamic range scenes.

[Comprehensive identification of metabolites and metabolic characteristics of luteolin and kaempferol in Simiao Yong'an Decoction in rats by UHPLC-LTQ-Orbitrap MS/MS].

Simiao Yong'an Decoction is a classic prescription for treating gangrene. Modern medical evidence has proven that Si-miao Yong'an Decoction has therapeutic effects on atherosclerosis(AS), vascular occlusion angeitides, and hypertension, while its pharmacodynamic mechanism remains unclear. The evidence of network pharmacology, molecular docking, literature review, and our previous study suggests that luteolin and kaempferol are two major flavonoids in Simiao Yong'an Decoction and can inhibit macrophage inflammation and exert anti-AS effects. However, due to lack of the metabolism studies in vivo, little is known about the metabolic characteristics of luteolin and kaempferol. This study employed ultra-performance liquid chromatography coupled with linear ion trap-Orbitrap mass spectrometry(UHPLC-LTQ-Orbitrap MS/MS) and relevant software to identify the metabolites and metabolic pathways of luteolin and kaempferol in rat plasma, urine, and feces, after oral administration of luteolin and kaempferol, respectively. After the administration of luteolin, 10, 11, and 3 metabolites of luteolin were detected in the plasma, urine, and feces, respectively. After the administration of kaempferol, 9, 3, and 1 metabolites of kaempferol were detected in the plasma, urine, and feces, respectively. The metabolic pathways mainly involved methylation, glucuronidation, and sulfation. This study enriches the knowledge about the pharmacological mechanism of luteolin and kaempferol and supplies a reference for revealing the metabolic process of other flavonoids in Simiao Yong'an Decoction, which is of great significance for elucidating the pharmacological effects and effective substances of this decoction in vivo.

Holistically-Attracted Wireframe Parsing: From Supervised to Self-Supervised Learning.

This article presents Holistically-Attracted Wireframe Parsing (HAWP), a method for geometric analysis of 2D images containing wireframes formed by line segments and junctions. HAWP utilizes a parsimonious Holistic Attraction (HAT) field representation that encodes line segments using a closed-form 4D geometric vector field. The proposed HAWP consists of three sequential components empowered by end-to-end and HAT-driven designs: 1) generating a dense set of line segments from HAT fields and endpoint proposals from heatmaps, 2) binding the dense line segments to sparse endpoint proposals to produce initial wireframes, and 3) filtering false positive proposals through a novel endpoint-decoupled line-of-interest aligning (EPD LOIAlign) module that captures the co-occurrence between endpoint proposals and HAT fields for better verification. Thanks to our novel designs, HAWPv2 shows strong performance in fully supervised learning, while HAWPv3 excels in self-supervised learning, achieving superior repeatability scores and efficient training (24 GPU hours on a single GPU). Furthermore, HAWPv3 exhibits a promising potential for wireframe parsing in out-of-distribution images without providing ground truth labels of wireframes.

NOPE-SAC: Neural One-Plane RANSAC for Sparse-View Planar 3D Reconstruction.

This article studies the challenging two-view 3D reconstruction problem in a rigorous sparse-view configuration, which is suffering from insufficient correspondences in the input image pairs for camera pose estimation. We present a novel Neural One-PlanE RANSAC framework (termed NOPE-SAC in short) that exerts excellent capability of neural networks to learn one-plane pose hypotheses from 3D plane correspondences. Building on the top of a Siamese network for plane detection, our NOPE-SAC first generates putative plane correspondences with a coarse initial pose. It then feeds the learned 3D plane correspondences into shared MLPs to estimate the one-plane camera pose hypotheses, which are subsequently reweighed in a RANSAC manner to obtain the final camera pose. Because the neural one-plane pose minimizes the number of plane correspondences for adaptive pose hypotheses generation, it enables stable pose voting and reliable pose refinement with a few of plane correspondences for the sparse-view inputs. In the experiments, we demonstrate that our NOPE-SAC significantly improves the camera pose estimation for the two-view inputs with severe viewpoint changes, setting several new state-of-the-art performances on two challenging benchmarks, i.e., MatterPort3D and ScanNet, for sparse-view 3D reconstruction.

Learning to See Through With Events.

Although synthetic aperture imaging (SAI) can achieve the seeing-through effect by blurring out off-focus foreground occlusions while recovering in-focus occluded scenes from multi-view images, its performance is often deteriorated by dense occlusions and extreme lighting conditions. To address the problem, this paper presents an Event-based SAI (E-SAI) method by relying on the asynchronous events with extremely low latency and high dynamic range acquired by an event camera. Specifically, the collected events are first refocused by a Refocus-Net module to align in-focus events while scattering out off-focus ones. Following that, a hybrid network composed of spiking neural networks (SNNs) and convolutional neural networks (CNNs) is proposed to encode the spatio-temporal information from the refocused events and reconstruct a visual image of the occluded targets. Extensive experiments demonstrate that our proposed E-SAI method can achieve remarkable performance in dealing with very dense occlusions and extreme lighting conditions and produce high-quality images from pure events. Codes and datasets are available at https://dvs-whu.cn/projects/esai/.

Bayesian Collaborative Learning for Whole-Slide Image Classification.

Whole-slide image (WSI) classification is fundamental to computational pathology, which is challenging in extra-high resolution, expensive manual annotation, data heterogeneity, etc. Multiple instance learning (MIL) provides a promising way towards WSI classification, which nevertheless suffers from the memory bottleneck issue inherently, due to the gigapixel high resolution. To avoid this issue, the overwhelming majority of existing approaches have to decouple the feature encoder and the MIL aggregator in MIL networks, which may largely degrade the performance. Towards this end, this paper presents a Bayesian Collaborative Learning (BCL) framework to address the memory bottleneck issue with WSI classification. Our basic idea is to introduce an auxiliary patch classifier to interact with the target MIL classifier to be learned, so that the feature encoder and the MIL aggregator in the MIL classifier can be learned collaboratively while preventing the memory bottleneck issue. Such a collaborative learning procedure is formulated under a unified Bayesian probabilistic framework and a principled Expectation-Maximization algorithm is developed to infer the optimal model parameters iteratively. As an implementation of the E-step, an effective quality-aware pseudo labeling strategy is also suggested. The proposed BCL is extensively evaluated on three publicly available WSI datasets, i.e., CAMELYON16, TCGA-NSCLC and TCGA-RCC, achieving an AUC of 95.6%, 96.0% and 97.5% respectively, which consistently outperforms all the methods compared. Comprehensive analysis and discussion will also be presented for in-depth understanding of the method. To promote future work, our source code is released at: https://github.com/Zero-We/BCL.

Learning to Super-Resolve Blurry Images With Events.

Super-Resolution from a single motion Blurred image (SRB) is a severely ill-posed problem due to the joint degradation of motion blurs and low spatial resolution. In this article, we employ events to alleviate the burden of SRB and propose an Event-enhanced SRB (E-SRB) algorithm, which can generate a sequence of sharp and clear images with High Resolution (HR) from a single blurry image with Low Resolution (LR). To achieve this end, we formulate an event-enhanced degeneration model to consider the low spatial resolution, motion blurs, and event noises simultaneously. We then build an event-enhanced Sparse Learning Network (eSL-Net++) upon a dual sparse learning scheme where both events and intensity frames are modeled with sparse representations. Furthermore, we propose an event shuffle-and-merge scheme to extend the single-frame SRB to the sequence-frame SRB without any additional training process. Experimental results on synthetic and real-world datasets show that the proposed eSL-Net++ outperforms state-of-the-art methods by a large margin. Datasets, codes, and more results are available at https://github.com/ShinyWang33/eSL-Net-Plusplus.

[Advances in animal models of chronic heart failure and its applications in traditional Chinese medicine].

Chronic heart failure(CHF) is a series of clinical syndromes in which various heart diseases progress to their end stage. Its morbidity and mortality are increasing year by year, which seriously threatens people's life and health. The diseases causing CHF are complex and varied, such as coronary heart disease, hypertension, diabetes, cardiomyopathy and so on. It is of great significance to establish animal models of CHF according to different etiologies to explore the pathogenesis of CHF and develop drugs to prevent and treat CHF induced by different diseases. Therefore, based on the classification of the etiology of CHF, this paper summarizes the animal models of CHF widely used in recent 10 years, and the application of these animal models in traditional Chinese medicine(TCM) research, in order to provide ideas and strategies for studying the pathogenesis and treatment of CHF, and provide ideas for TCM modernization research.

VILLIN2 regulates cotton defense against Verticillium dahliae by modulating actin cytoskeleton remodeling.

The active structural change of actin cytoskeleton is a general host response upon pathogen attack. This study characterized the function of the cotton (Gossypium hirsutum) actin-binding protein VILLIN2 (GhVLN2) in host defense against the soilborne fungus Verticillium dahliae. Biochemical analysis demonstrated that GhVLN2 possessed actin-binding, -bundling, and -severing activities. A low concentration of GhVLN2 could shift its activity from actin bundling to actin severing in the presence of Ca2+. Knockdown of GhVLN2 expression by virus-induced gene silencing reduced the extent of actin filament bundling and interfered with the growth of cotton plants, resulting in the formation of twisted organs and brittle stems with a decreased cellulose content of the cell wall. Upon V. dahliae infection, the expression of GhVLN2 was downregulated in root cells, and silencing of GhVLN2 enhanced the disease tolerance of cotton plants. The actin bundles were less abundant in root cells of GhVLN2-silenced plants than in control plants. However, upon infection by V. dahliae, the number of actin filaments and bundles in the cells of GhVLN2-silenced plants was raised to a comparable level as those in control plants, with the dynamic remodeling of the actin cytoskeleton appearing several hours in advance. GhVLN2-silenced plants exhibited a higher incidence of actin filament cleavage in the presence of Ca2+, suggesting that pathogen-responsive downregulation of GhVLN2 could activate its actin-severing activity. These data indicate that the regulated expression and functional shift of GhVLN2 contribute to modulating the dynamic remodeling of the actin cytoskeleton in host immune responses against V. dahliae.

Enabling country-scale land cover mapping with meter-resolution satellite imagery.

High-resolution satellite images can provide abundant, detailed spatial information for land cover classification, which is particularly important for studying the complicated built environment. However, due to the complex land cover patterns, the costly training sample collections, and the severe distribution shifts of satellite imageries caused by, e.g., geographical differences or acquisition conditions, few studies have applied high-resolution images to land cover mapping in detailed categories at large scale. To fill this gap, we present a large-scale land cover dataset, Five-Billion-Pixels. It contains more than 5 billion labeled pixels of 150 high-resolution Gaofen-2 (4 m) satellite images, annotated in a 24-category system covering artificial-constructed, agricultural, and natural classes. In addition, we propose a deep-learning-based unsupervised domain adaptation approach that can transfer classification models trained on labeled dataset (referred to as the source domain) to unlabeled data (referred to as the target domain) for large-scale land cover mapping. Specifically, we introduce an end-to-end Siamese network employing dynamic pseudo-label assignment and class balancing strategy to perform adaptive domain joint learning. To validate the generalizability of our dataset and the proposed approach across different sensors and different geographical regions, we carry out land cover mapping on five megacities in China and six cities in other five Asian countries severally using: PlanetScope (3 m), Gaofen-1 (8 m), and Sentinel-2 (10 m) satellite images. Over a total study area of 60,000 km2, the experiments show promising results even though the input images are entirely unlabeled. The proposed approach, trained with the Five-Billion-Pixels dataset, enables high-quality and detailed land cover mapping across the whole country of China and some other Asian countries at meter-resolution.

(±)-Yanhusuomide A, a pair of ornithine-fused benzylisoquinoline enantiomers from Corydalis yanhusuo.

(±)-Yanhusuomide A (1), a novel enantiomeric pair of ornithine-fused benzylisoquinoline, were characterized from the dried tubers of Corydalis yanhusuo, along with a biogenetically related intermediate oblongine (2). Yanhusuomide A features an unprecedented skeleton based on a benzylisoquinoline coupled with an ornithine derivative to form a rare 5,6-dihydro-4H-pyrido[3,4,5-de]quinazoline motif. Plausible biosynthetic pathway of 1 was proposed, and (±)-yanhusuomide A (1) presented potential inhibitory bioactivity against acetylcholinesterase (AChE) with IC50 = 14.07 ± 2.38 µM. The simulation of molecular docking displayed that 1 generated strong interaction with Asp-74 and Trp-86 residues of AChE through attractive charge of the quaternary nitrogen.

Learning to Extract Building Footprints From Off-Nadir Aerial Images.

Extracting building footprints from aerial images is essential for precise urban mapping with photogrammetric computer vision technologies. Existing approaches mainly assume that the roof and footprint of a building are well overlapped, which may not hold in off-nadir aerial images as there is often a big offset between them. In this paper, we propose an offset vector learning scheme, which turns the building footprint extraction problem in off-nadir images into an instance-level joint prediction problem of the building roof and its corresponding "roof to footprint" offset vector. Thus the footprint can be estimated by translating the predicted roof mask according to the predicted offset vector. We further propose a simple but effective feature-level offset augmentation module, which can significantly refine the offset vector prediction by introducing little extra cost. Moreover, a new dataset, Buildings in Off-Nadir Aerial Images (BONAI), is created and released in this paper. It contains 268,958 building instances across 3,300 aerial images with fully annotated instance-level roof, footprint, and corresponding offset vector for each building. Experiments on the BONAI dataset demonstrate that our method achieves the state-of-the-art, outperforming other competitors by 3.37 to 7.39 points in F1-score. The codes, datasets, and trained models are available at https://github.com/jwwangchn/BONAI.git.

[Pharmacological research progress of five classical prescriptions in treatment of chronic heart failure].

Chronic heart failure(CHF) is a comprehensive clinical syndrome caused by multiple factors that result in structural and/or functional abnormalities of the heart, leading to impaired ventricular contraction and/or relaxation functions. This medical condition represents the final stage of various cardiovascular diseases. In the treatment of CHF, multiple clinical studies have demonstrated the benefits of using traditional Chinese medicine(TCM) to control oxidative stress, inflammation, and apoptosis, thereby delaying ventricular remodeling and reducing myocardial fibrosis. In this study, common TCM syndromes in the diagnosis and treatment of CHF in recent years were reviewed and summarized. Five common treatment methods including benefiting Qi and activating blood circulation, enhancing Qi and nourishing Yin, warming Yang for diuresis, eliminating phlegm and dampness, rescuing from collapse by restoring Yang, and corresponding classic prescriptions in prevention and treatment of CHF were concluded under the guidance of TCM syndrome differentiation thinking. Meanwhile, research progress on the modern pharmacological effects of these classic prescriptions was systematically discussed, so as to establish a unique treatment system for CHF by classic prescriptions under the guidance of TCM syndrome differentiation theory and provide innovative diagnosis and treatment strategies for clinical CHF.

Association between plasma Irisin concentrations and changes in blood pressure among children: a cohort study / 中国学校卫生

Objective@#To investigate the longitudinal association of plasma Irisin concentrations with changes in blood pressure (BP) levels among children,and to assess the moderating effect of physical activity (PA) or sedentary behavior (SB) on the relationship between Irisin levels and BP.@*Methods@#Based on a cohort study, a cluster sampling method was used to select 3 651 school aged children from five schools in Guangzhou in 2017 at the baseline survey and follow up in 2019. Both at baseline and during follow up, PA and SB were assessed by validated questionnaires, and BP levels were measured by an electronic sphygmomanometer. A final sample of 521 children were enrolled based on the PA and SB at baseline. Plasma Irisin concentrations were measured by ELISA at baseline. Logistic regression analysis was recruited for exploring the associations of plasma Irisin concentrations with changes in BP. Moderating effects of PA and SB on the relationship between Irisin concentrations and BP were estimated using stratified analysis.@*Results@#Logistic regression analysis indicated that there was no significant association between Irisin concentrations and changes in BP levels among children ( OR =0.98, P >0.05). After stratification for SB, Irisin levels in the low SB subgroup were inversely associated with changes in diastolic blood pressure ( OR=0.87, 95%CI=0.77-0.98, P =0.02). In addition, SB level had a moderating effect on the relationship between Irisin levels and the DBP changes ( P =0.01).@*Conclusion@#Increased Irisin concentration is associated with the decrease of DBP level among low SB children. Furthermore, SB level shows moderating role in the relationship between Irisin concentrations and changes in DBP levels.

[Progress of target determination and mechanism of bioactive components of traditional Chinese medicine].

The pharmacodynamic substances of traditional Chinese medicine(TCM) are the basis for the research of TCM and the development of innovative drugs. However, the lack of clarity of targets and molecular mechanisms is the bottleneck problem that restricts the research of pharmacodynamic substances of TCM. Bioactive components are the material basis of the efficacy of TCM, which exert activity by regulating the corresponding targets. Therefore, it is very important to identify the targets of the bioactive components to elucidate the pharmacological mechanism of TCM. Proteins are the most important drug targets, and study of the interaction between the proteins and bioactive components of TCM plays a key role in the development of pharmacological mechanism of TCM. In recent years, the main techniques for detecting the interaction between the bioactive components and proteins include surface plasmon resonance, fluorescence resonance energy transfer, bio-layer interference, molecular docking, proteome chip, target fishing, target mutant, and protein crystallization techniques, etc. This review summarized the biological target detection techniques and their applications in locating the targets of the bioactive components in TCM in the last decade, and this paper will provide useful strategies to elucidate the pharmacological mechanisms of TCM.