Search

To establish a breast lesion risk stratification system using ultrasound images to predict breast malignancy and assess Breast Imaging Reporting and Data System (BI-RADS) categories simultaneously.This multicenter study prospectively collected a dataset of ultrasound images for 5012 patients at thirty-two hospitals from December 2018 to December 2020. A deep learning (DL) model was developed to conduct binary categorization (benign and malignant) and BI-RADS categories (2, 3, 4a, 4b, 4c, and 5) simultaneously. The training set of 4212 patients and the internal test set of 416 patients were from thirty hospitals. The remaining two hospitals with 384 patients were used as an external test set. Three experienced radiologists performed a reader study on 324 patients randomly selected from the test sets. We compared the performance of the DL model with that of three radiologists and the consensus of the three radiologists.In the external test set, the DL model achieved areas under the receiver operating characteristic curve (AUCs) of 0.980 and 0.945 for the binary categorization and six-way categorizations, respectively. In the reader study set, the DL BI-RADS categories achieved a similar AUC (0.901 vs. 0.933, p = 0.0632), sensitivity (90.98% vs. 95.90%, p = 0.1094), and accuracy (83.33% vs. 79.01%, p = 0.0541), but higher specificity (78.71% vs. 68.81%, p = 0.0012) than those of the consensus of the three radiologists.The DL model performed well in distinguishing benign from malignant breast lesions and yielded outcomes similar to experienced radiologists. This indicates the potential applicability of the DL model in clinical diagnosis.• The DL model can achieve binary categorization for benign and malignant breast lesions and six-way BI-RADS categorizations for categories 2, 3, 4a, 4b, 4c, and 5, simultaneously. • The DL model showed acceptable agreement with radiologists for the classification of breast lesions. • The DL model performed well in distinguishing benign from malignant breast lesions and had promise in helping reduce unnecessary biopsies of BI-RADS 4a lesions.

The outbreak of COVID-19 threatens the lives and property safety of countless people and brings a tremendous pressure to health care systems worldwide. The principal challenge in the fight against this disease is the lack of efficient detection methods. AI-assisted diagnosis based on deep learning can detect COVID-19 cases for chest X-ray images automatically, and also improve the accuracy and efficiency of doctors' diagnosis. However, large scale annotation of chest X-ray images is difficult because of limited resources and heavy burden on the medical system. To meet the challenge, we propose a capsule network model with multi-head attention routing algorithm, called MHA-CoroCapsule, to provide fast and accurate diagnostics for COVID-19 diseases from chest X-ray images. The MHA-CoroCapsule consists of convolutional layers, two capsule layers, and a non-iterative, parameterized multi-head attention routing algorithm is used to quantify the relationship between the two capsule layers. The experiments are performed on a combined dataset constituted by two publicly available datasets including normal, non-COVID pneumonia and COVID-19 images. The model achieves the accuracy of 97.28%, recall of 97.36%, and precision of 97.38% even with a limited number of samples. The experimental results demonstrate that, contrary to the transfer learning and deep feature extraction approaches, the proposed MHA-CoroCapsule has an encouraging performance with fewer trainable parameters and does not require pretraining and plenty of training samples.

We evaluated the role of raw-material features for machine-learning prediction of the flux crystal growth of Al2O3 in MoO3 based on 185 types of growth trials.

BACKGROUND: Vascular invasion is an important risk factor of poor prognosis in hepatocellular carcinoma (HCC) patients. The detection of circulating tumor cells (CTCs) in the blood is direct evidence of tumor presence. There are few reports on CTCs and metastasis and vascular invasion of HCC. The purpose of this study was to analyze the significance of CTCs in the portal vein regarding metastases and vascular invasion in HCC patients. METHODS: A total of 104 HCC patients diagnosed and treated in Zhengzhou University People’s Hospital were enrolled. Surgery was performed in 60 individuals. Portal vein blood samples were collected before treatment for CTCs detection. We used the isolation by size of epithelial tumor cells (ISET) and fluorescence in situ hybridization (FISH) to enrich and classify CTCs from blood samples. The patients were divided into metastasis and nonmetastasis groups according to the metastasis status before treatment. Differences in clinical indicators such as alpha-fetoprotein (AFP) levels, tumor size, CTCs count, and macrovascular tumor thrombus between the two groups were analyzed as well as the associations of CTCs count with the above indicators. For individuals with postoperative pathology, the relationship between CTCs counts and microvascular invasion (MVI) was analyzed. RESULTS: The amounts of portal vein CTCs were higher in patients with metastases compared with the nonmetastases group (20 vs. 7; z=3.795; P

For the prediction of economic expenses involved in construction industry, cost estimation has become an important aspect of construction management for the prediction of economic expenses and successful completion of the construction work. Cost analysis is crucial and require expertise for accurate and comprehensive estimation. In order to effectively improve the accuracy of construction project cost, this paper establishes an estimation model based on gray BP neural network. It combines the MATLAB toolbox for program design, and learns and tests the input and output of training samples. This article determines the application of grey system theory to optimize the estimation model of Back Propagation (BP) neural network. The viability of the method established in this article, is tested by collecting the engineering cost data in Zhengzhou city and comparing between the standard BP neural network and the gray BP neural network methods. The results show that the average error of the gray system theory optimized BP neural network model designed in this paper is 2.33%. The gray BP neural network model studied in this paper can not only quickly estimate the project cost, but also has high accuracy rate. The outcomes obtained establishes a model with scientific and reasonable construction project cost estimation.

A catalytic system for a phosphorus ligand-free Suzuki–Miyaura reaction in water at room temperature was disclosed. Ammonium 2,2′-azino-bis(3-ethylbenzthiazoline-6-sulfonate) (ABTS) was an efficient promoter and acted both as a ligand and as a surfactant for the synthesis of biaryl compounds via the Suzuki–Miyaura reaction in water. The targeted biaryl architectures were achieved under mild conditions with high efficiency and good functional group tolerance.

In recent years, compressive sensing has been one promising technique for denoising images. This paper presents a new denoising model based on blocking sparsity. First, an image is blocked. Second, the split-Bregman method is used to solve for each block image. Finally, all denoised block images are combined into one image. Compared with the latest HTV, GHNS, FastATV, CSR and BM3D models, experimental results demonstrate that the proposed method is efficient, and has better denoising capability.

In recent years, high dimensional data reconstruction based on sparse tensor has been one promising research field, especially for color images reconstruction and denoising magnetic resonance images. This paper aims at developing a new high dimensional data reconstruction model based on L2,1-norm. A fast algorithm is devised to solve the optimal problem. Additionally, error bound analysis of proposed scheme is discussed. Experiments show that the proposed algorithm is efficient, and the reconstruction results are better than two state-of-the-art methods. The proposed method has higher efficiency in terms of computational time and iterations.

Purpose This paper aims to present the design and experimental tests of an active circulating cooling system for the Experimental Advanced Superconducting Tokamak in-vessel inspection manipulator, which will help the current manipulator prototype to achieve a full-scale in-vessel high temperature environment compatibility. Design/methodology/approach The high-temperature effects and heat transfer conditions of the manipulator under in-vessel environment were analyzed. An active circulating cooling system was designed and implemented on the manipulator prototype. A simulative in-vessel inspection task in a high temperature environment of 100°C was carried out to evaluate the performance of the active circulating cooling system. Findings The proposed active circulating cooling system was proved effective in helping the manipulator prototype to achieve its basic in-vessel inspection capability in a high temperature environment. The active circulating cooling system performance can be further improved considering the cooling structure coefficient differences in different manipulator parts. Originality/value For the first time, the active circulating cooling system was implemented and tested on a full-scale of the in-vessel inspection manipulator. The experimental data of the temperature distribution inside the manipulator and the operating status of the circulating system were helpful to evaluate the current active circulating cooling system design and provided effective guidance for improving the overall system performance.

To study the effect of some parameters, such as, length and fraction of glass fiber (GF), and the fraction of maleic anhydride grafted polypropylene (PP-g-MAH), on the mechanical properties of glass fiber reinforced polypropylene (GF/PP) composites, tensile tests, bending tests and impact tests were conducted. Scanning electron microscope (SEM) was used to characterize the fracture mechanisms of the composites. The results show that, compared with 3 mm GF, 9 mm GF can significantly improve the strength of the composite better. Addition of PP-g-MAH, a kind of grafting agent, into the PP-30% LGF composite can result in a better mechanical properties because of the strengthening of the bonding interface between the matrix and the fiber. When the mass fraction of GF is 30% and the PP-g-MAH fraction is 6%, the mechanical properties of the composite are the best.

Recursive least-squares algorithms often use forgetting factors as a heuristic to adapt to non-stationary data streams. The first contribution of this paper rigorously characterizes the effect of forgetting factors for a class of online Newton algorithms. For exp-concave and strongly convex objectives, the algorithms achieve the dynamic regret of $\max\{O(\log T),O(\sqrt{TV})\}$, where $V$ is a bound on the path length of the comparison sequence. In particular, we show how classic recursive least-squares with a forgetting factor achieves this dynamic regret bound. By varying $V$, we obtain a trade-off between static and dynamic regret. In order to obtain more computationally efficient algorithms, our second contribution is a novel gradient descent step size rule for strongly convex functions. Our gradient descent rule recovers the order optimal dynamic regret bounds described above. For smooth problems, we can also obtain static regret of $O(T^{1-\beta})$ and dynamic regret of $O(T^\beta V^*)$, where $\beta \in (0,1)$ and $V^*$ is the path length of the sequence of minimizers. By varying $\beta$, we obtain a trade-off between static and dynamic regret., Comment: Accepted by AAAI 2020

A Gram-stain-negative, cocci or short rod-shaped, yellow-pigmented bacterium, designated strain S2-4-2T, was isolated from coastal sediment of Spartina alterniflora in Quanzhou Bay, PR China. Growth was observed at 15–40 °C (optimum, 30 °C) and pH 6.0–8.0 (optimum, pH 7.0). Strain S2-4-2T tolerated 0–10 % NaCl (optimum, 1 %). The 16S rRNA gene of strain S2-4-2T showed highest sequence similarity to Croceicoccus pelagius Ery9T (98.2 %), followed by Croceicoccus naphthovorans PQ-2T, Croceicoccus marinus E4A9T and Croceicoccus mobilis Ery22T (97.6%, 96.3 and 96.3 %, respectively). Phylogenetic analysis based on 16S rRNA gene sequences and phylogenomic analysis based on 92 up-to-date bacterial core gene sets indicated that strain S2-4-2T forms a distinct monophyletic branch affiliated to the genus Croceicoccus . The average nucleotide identity value between strain S2-4-2T and its close relatives were estimated to be 74.8–85.7 %. The respiratory quinone was found to be Q-10. The predominant fatty acids (>10 %) were identified as summed feature 8 (C18 : 1ω7c and/or C18 : 1ω6c) and summed feature 3 (C16 : 1ω7c and C16 : 1ω6c). The polar lipids were identified as phosphatidylglycerol, phosphatidylethanolamine, phosphatidylcholine, sphingoglycolipid and an unidentified phospholipid. The draft genome size of strain S2-4-2T was 3.5 Mb with a genomic G+C content of 63.0 mol%. Based on these results, strain S2-4-2T is concluded to represent a novel species within the genus Croceicoccus , for which the name Croceicoccus sediminis sp. nov. is proposed with the type strain S2-4-2T (=MCCC 1K03706T=KCTC 72146T).

We aimed to evaluate the safety of the ultrasound contrast agent sulfur hexafluoride microbubbles in a large group of patients referred for routine contrast-enhanced ultrasound (CEUS).A retrospective assessment was made of all patients that received sulfur hexafluoride microbubbles intravenously for CEUS at 24 centers between January 2006 and April 2019. Patient demographic details, examination type, and the dose of sulfur hexafluoride microbubbles administered were recorded with specific adverse events (AEs) documentation tools at each center. All AEs were recorded as serious or non-serious. Non-serious AEs were classified by intensity as mild, moderate, or severe according to ACR criteria. The frequencies of AEs across patient subgroups were compared using the chi-square test.A total of 463,434 examinations were evaluated. Overall, 157 AEs (153 [0.033%] non-serious; 4 [0.001%] serious) were reported after sulfur hexafluoride microbubbles administration, giving an AE frequency of 0.034% (157/463,434). Among the non-serious AEs, 66 (0.014%) were mild, 70 (0.015%) moderate, and 17 (0.004%) severe in intensity. The liver was the most common examination site, presenting an AE frequency of 0.026%. The highest AE frequency (0.092%) was for patients undergoing CEUS for vascular disease. There were no significant gender differences in either the total number or the severity of non-serious AEs (chi-square = 2.497, p = 0.287). The onset of AEs occurred within 30 min of sulfur hexafluoride microbubbles administration in 91% of cases.The frequency of AEs to sulfur hexafluoride microbubbles is very low and severe reactions are rare, confirming that sulfur hexafluoride microbubbles are appropriate for routine CEUS applications.• The frequency of AEs to sulfur hexafluoride microbubbles is very low and severe reactions are rare.

BackgroundStudies on deep learning (DL)-based models in breast ultrasound (US) remain at the early stage due to a lack of large datasets for training and independent test sets for verification. We aimed to develop a DL model for differentiating benign from malignant breast lesions on US using a large multicenter dataset and explore the model’s ability to assist the radiologists.MethodsA total of 14,043 US images from 5012 women were prospectively collected from 32 hospitals. To develop the DL model, the patients from 30 hospitals were randomly divided into a training cohort (n = 4149) and an internal test cohort (n = 466). The remaining 2 hospitals (n = 397) were used as the external test cohorts (ETC). We compared the model with the prospective Breast Imaging Reporting and Data System assessment and five radiologists. We also explored the model’s ability to assist the radiologists using two different methods.ResultsThe model demonstrated excellent diagnostic performance with the ETC, with a high area under the receiver operating characteristic curve (AUC, 0.913), sensitivity (88.84%), specificity (83.77%), and accuracy (86.40%). In the comparison set, the AUC was similar to that of the expert (p = 0.5629) and one experienced radiologist (p = 0.2112) and significantly higher than that of three inexperienced radiologists (p ConclusionsThe DL model yielded satisfactory predictions in distinguishing benign from malignant breast lesions. The model showed the potential value in improving the diagnosis of breast lesions by radiologists.

Osteosarcoma is a common primary bone tumour in children and adolescents. Circular RNAs (circRNAs) exert vital functions in human diseases, including osteosarcoma. Therefore, we explored the role of circ_0016347 in osteosarcoma.The real-time quantitative polymerase chain reaction (RT-qPCR) was used to detect the expression levels of circ_0016347, microRNA-661 (miR-661), and Interleukin-6 receptor (IL6R) in osteosarcoma tissues and cells. The proliferation of osteosarcoma cells was assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazol-3-ium bromide (MTT) and EdU experiments. The migration and invasion were determined by transwell assay. The cell cycle distribution and apoptosis were assessed by flow cytometry assay. The association relationships among circ_0016347, miR-661, and IL6R were analyzed by dual-luciferase reporter assays. The western blot assay was employed to assay the protein expression. A xenograft experiment was established to clarify the functional role of circ_0016347 inhibitionCirc_0016347 was obviously overexpressed in osteosarcoma tissues and cells compared with control groups. The suppression of circ_0016347 impeded proliferation, migration, invasion, and cell cycle and induced apoptosis in osteosarcoma cells, which was overturned by knockdown of miR-661. Consistently, circ_0016347 knockdown repressed tumour growthCirc_0016347 functioned as an oncogene in osteosarcoma at least in part by the miR-661/IL6R axis and JAK2/STAT3 signalling pathway.

Cellular senescence is one of the most significant factors involved in aging and age-related diseases. Senescence of vascular smooth muscle cells (VSMCs) adversely affects the function of the cardiovascular system and contributes to the development of atherosclerosis, hypertension, and other cardiovascular diseases. Glucagon-like peptide-1 (GLP-1) is an important incretin hormone involved in insulin release and vascular tone. GLP-1 is quickly degraded by the enzyme dipeptidyl peptidase-4 (DPP-4). Omarigliptin is a new DPP-4 inhibitor that has demonstrated anti-inflammatory and antioxidative stress properties. In the present study, we investigated the effects of the selective DPP-4 inhibitor omarigliptin (OMG) on VSMCs exposed to insult from tumor necrosis factor-α (TNF-α), one of the main inflammatory signaling molecules involved in cellular senescence. We found that OMG could suppress TNF-α-induced expression of pro-inflammatory cytokines (interleukin-1β (IL-1β), IL-6, and IL-8) and inhibit oxidative stress by reducing the production of H2O2 and protein carbonyl. OMG ameliorated the increase in senescence-associated β-galactosidase (SA-β-gal) and telomerase activity induced by TNF-α. The plasminogen activator inhibitor-1 (PAI-1)/p53/p21 pathway is a key inducer of cellular senescence. OMG ameliorated the acetylation of p53 at lysine 382 (K382) and subsequent activation of p21 via inhibition of PAI-1. Importantly, our experiments revealed that blockage of silent information-regulator 1 (SIRT1) abolished the inhibitory effects of OMG on p53 acetylation, SA-β-gal activity, and telomerase activity in VSMCs. These results suggest that OMG may have the potential to delay or prevent the progression of age-related cardiovascular diseases by modulating the activity of SIRT1.

This paper proposes a variable artificial potential field method (VAPF) that can be used to generate the gait of a quadruped robot, which combines the virtual potential field theory with the quadruped motion behavior to generate a smooth motion trajectory of the foot. According to the different stages of single-leg movement, which includes the support phase and the swing phase, the VAPF method constructs the point potential field-based model including the attraction domain and the repulsion domain, respectively. The joint torque directly output by the potential field model can realize the desired trajectory of the foot in the robot body coordinate system. An independent attitude controller is designed, and the joint torque respectively calculated by this controller and the VAPF are applied to the robot together to generate the actual gait. Finally, a simulation of trot gait generation was performed and analysis.

The virtual model control (VMC) method realizes the direct mapping from the overall control target to the joint-level control commands by establishing the virtual force corresponding to the motion control target. This control method that does not directly involve the dynamic model is easy to calculate and implement, while the values of the stiffness and damping parameters in the virtual force equation have a direct impact on the control effect. Therefore, realizing the reasonable setting of the above parameters is an important problem that needs to be solved in the application of the VMC method. This paper takes the swing motion of the leg mechanism of a quadruped robot as an example to study. In the VMC framework, an optimization method based on energy consumption and motion trajectory errors as evaluation indicators is proposed. The genetic algorithm was used to implement specific optimization calculations in detail, and the comprehensive optimization of the stiffness coefficient and damping coefficient in the virtual force model was realized. Moreover, the optimization results obtained by using multiple fitness functions are compared and analyzed. Finally, cosimulation and basic physical experiments are performed to verify the effectiveness of the method proposed in this paper.