Search

Background: Whether machine learning (ML) can assist in the diagnosis of Crohn's disease (CD) and intestinal tuberculosis (ITB) remains to be explored. Methods: We collected clinical data from 241 patients, and 51 parameters were included. Six ML methods were tested, including logistic regression, decision tree, k-nearest neighbor, multinomial NB, multilayer perceptron, and XGBoost. SHAP and LIME were subsequently introduced as interpretability methods. The ML model was tested in a real-world clinical practice and compared with a multidisciplinary team (MDT) meeting. Results: XGBoost displays the best performance among the six ML models. The diagnostic AUROC and the accuracy of XGBoost were 0.946 and 0.884, respectively. The top three clinical features affecting our ML model's result prediction were T-spot, pulmonary tuberculosis, and onset age. The ML model's accuracy, sensitivity, and specificity in clinical practice were 0.860, 0.833, and 0.871, respectively. The agreement rate and kappa coefficient of the ML and MDT methods were 90.7% and 0.780, respectively (P< 0.001). Conclusion: We developed an ML model based on XGBoost. The ML model could provide effective and efficient differential diagnoses of ITB and CD with diagnostic bases. The ML model performs well in real-world clinical practice, and the agreement between the ML model and MDT is strong. [ABSTRACT FROM AUTHOR]

Due to several complex factors such as the type, size, and shape of target object, vision-assisted robot grasping technology still faces serious challenges. In this research, a deep learning-based robot hand vision grasping algorithm was developed considering semi-structural environmental constraints. The proposed algorithm could build a deep learning network on the basis of the desired object, perform object recognition, category classification and position judgment, and complete robot hand-grasping tasks. The obtained experimental results demonstrated that the proposed algorithm effectively solved the problem of recognizing and classifying multi-category objects in a semi-structured environment, improving recognition rate and grasping rate and reducing collision rate. [ABSTRACT FROM AUTHOR]

Xylanase is a novel and safe flour additive that effectively enhances dough properties and baking quality. To explore new sources of xylanase and evaluate its application in bread baking, xylanase from Caldicellulosiruptor owensensis was heterogeneously expressed in Escherichia coli by fusion with Sumo tag. The enzymatic properties of the xylanase SumoXyn and its impact on bread quality were investigated. The xylanase from Caldicellulosiruptor owensensis was efficiently expressed in E.coli and the enzyme was obtained with high purity by simply heating the cell lysate at 75 °C for 30 minutes. Nano DSF analysis showed that the Tm value of this enzyme was 83.5 °C. The optimum pH of the enzyme was 9.0, but it still had more than 60% activity at pH 5-10. Treated in pH 4-10 buffer for 24 h, SumoXyn still had more than 80% residual activity. SumoXyn displayed the highest activity at 80 °C, but it still has 50% of the highest activity at 100 °C, and the residual activity was about 75% after heating at 75 °C for 1 h, indicating that Sumoxyn had good pH stability and thermal stability. Using bagasse xylan as substrate, the Vmax was 2 354.52 μmol/(min·mg), the Kmwas 2.26 mg/mL and the Kcat was 214.68 S-1. Using 200 g high gluten whole wheat flour to make bread, the effects of SumoXyn supplemented with 3 mg, 6 mg, 9 mg, respectively, on the quality of whole wheat bread were investigated. It was found that with the increase of SumoXyn supplemented, the specific volume and elasticity of bread increased, the hardness and chewability decreased, and the quality of whole wheat bread was significantly improved. These findings suggest that SumoXyn has great potential as a flour improver. [ABSTRACT FROM AUTHOR]

In this research, a 3D visual recognition system has been developed based on Wei deep learning algorithm using GPU. The proposed system consisted of a GPU with depth image function library, which performed image data acquisition, depth information operation, coordinate conversion, image contour search, convolutional class neural network model training, etc., and achieved object pinning by TCP/IP communication with motion control system. The obtained experimental results revealed that the recognition rate of the developed algorithm for target objects at different positions was as high as 92%. Experimental target recognition rates for different angles were relatively low, but reached 87%, and experimental accuracy rates of different luminance values also reached 89%. The errors of robot hand clamping targets also fell within 1–4 mm, which were higher than experimental expectation. [ABSTRACT FROM AUTHOR]

Based on the investigation of the welding behavior of the heterogeneous high strength steel under the coupling of multi factors of welding parameters, PLC was used to control the laser welding process, the welding process of HC550/DP780 high strength steel was optimized by orthogonal experiment. The results showed that, the laser power is the most important influence on the weld width, the width and the tensile strength of the HC550/DP780 heterogeneous high strength steel, followed by welding speed and defocusing amount; The optimum welding parameters of HC550/DP780 unequal thickness and high strength steel are A2B2C2, which the laser power is 1200 W, the welding speed is 2000 mm/min, the defocusing amount is −0.5 mm; tensile strength and elongation of HC550/DP780 welding joint under optimized welding parameters are 998.6 MPa and 11.9%, the fracture location in the HC550 base metal heat affected zone, which shows ductile fracture characteristics. Keywords: HC550/DP780, PLC control, Laser welding, Welding process

The effects of welding current and welding speed on the forming, microstructure and mechanical properties of Q890D steel/AZ31 alloy dissimilar metal welded joints were studied, and the mechanism of welding parameters on the interface zone of Q890D steel/AZ31 alloy dissimilar metal welded joints was analyzed. With the increase of welding current, the alpha-Mg grains in weld zone coarsen obviously, the thickness of interfacial layer increases, the unconnected defects decrease, and the crack sensitivity increases. With the increase of welding speed, the alpha-Mg grains in weld zone refine obviously, the thickness of interfacial layer decreases and the local unconnected defects increase. With the increase of welding current or welding speed, the tensile strength of the welded joint increases first and then decreases, when the welding current is 85 A and the welding speed is 45 cm/min, the welded joint has the highest strength and better appearance. The influence of welding current and welding speed on mechanical properties of Q890D steel/AZ31 alloy MIG welded joint is mainly related to the change of welding line energy and microstructure of interface zone. Keywords: Q890D steel/AZ31 alloy, MIG welding, Welding parameters, Microstructure, Mechanical properties

If the appearance of an eco-friendly textile fabric is problematic, the product quality will be substantially deteriorated. Defect measurement is one of the most important quality control measures for eco-friendly textile fabrics. Compared to previously employed manual measurements, the application of image processing technology for the detection of eco-friendly textile defects is characterized by high efficiency and high precision. In this study, the main objectives of textile reinforcement based on texture enhancement are as follows: (1) Summarize the description methods of texture maps in a certain space and a certain frequency and investigate the gray-scale co-occurrence matrix of textile fabrics, which aimed at the characteristics of a unique texture of textile fabrics, the texture of the background caused by noise, and the texture of the defect area. The error between them was analyzed; (2) Apply a scheme based on principal component analysis-non local means to improve the eco-friendly textile quality. The image information used in the calculation process of the neighborhood similarity in nonlocal average filtering algorithm (NLM) includes the problem of an excess amount due to noise, and the NLM method is employed to estimate the parameters. On the other hand, to remove the noise, it is also possible to display the texture image content of the textile fabric, which is more conducive to the defect detection; and (3) Apply a texture-based textile defect measurement method, that is, a class-separable characteristic between non-defective and defect textures, which increases the measurement of the gray matrix characteristics that distinguish the defect regions and improves the correctness of the detected texture. [ABSTRACT FROM AUTHOR]

In the production process of lithium battery, the quality inspection requirements of lithium battery are very high. At present, most of the work is done manually. Aiming at the problem of large manual inspection workload and large error, the robot visual inspection technology is applied to the production of lithium battery. In recent years, with the rapid development and progress of science and technology, the rapid development of visual detection hardware and algorithms, making it possible to screen defective products through visual detection algorithms. This paper takes lithium battery as the research object, and studies its vision detection algorithm. As a common commodity, the quality of lithium battery is the key for users to choose. With the increasing requirements of users for battery quality, how to produce high-quality battery is the key problem to be solved by manufacturers. However, at present, the defects of battery surface are mostly carried out manually. There are low efficiency and low detection rate in the process of manual detection. In this paper, the visual detection algorithm is studied to detect the defects such as pits, rust marks and broken skin on the surface of lithium battery, specifically to design the imaging experimental platform of lithium battery; use different lighting schemes to design different battery positioning and extraction algorithms; use Hough detection method to locate the battery surface, and design the battery defect algorithm for this, and compare the algorithm through experiments. [ABSTRACT FROM AUTHOR]

In today's globalized economy, various industries are promoting product transformation and upgrading. The textile industry is also facing a harsh international situation and fierce market competition. While constantly promoting the upgrade of automation equipment, the original quality control mode relying on manual testing has been unable to meet the modern production requirements and the market demand for product quality. This paper investigates the product inspection and quality control in the textile industry at home and abroad, and puts forward the application of machine vision technology in textile automated inspection and quality control, so As to strengthen the product quality control system and improve the product's physical quality. By studying the composition of machine vision detection technology, this paper studies the two core technologies of image acquisition and image processing in machine vision, summarizes and analyzes the common defects of textile Products, and proves that the common defects can be detected and repaired in time by machine vision detection method. [ABSTRACT FROM AUTHOR]

It is difficult to adapt to the actual situation of the production and testing of the rivets by examining the defects of low working efficiency and low accuracy. In order to solve this defect, through a series of research and analysis, it can be found that the non-contact automatic inspection method can solve the existing defects. The OSTU operation method is improved and used for the image segmentation of the rivets, which can reduce the error segmentation. In order to eliminate the uncertainty of the rivet position, the method of the least external rectangle method is obtained to determine the main shaft of the rivet. This approach reduces randomness at work. Through the study of rivet contour curvature method to analyze the characteristics of the part of the rivet contour, can better reduce in proportion of noise in the process of operation, and has a lot of improvement for accurate precision. Through experiment that this method of testing with high accuracy, errors generated during inspection situation is less, and less affected when testing can improve the efficiency of detection in industrial production, can well meet the demands of rivet real-time inspection. [ABSTRACT FROM AUTHOR]

In the diversified development of glass industry and technology products, the demand of quality performance of the glass surface increases. The surface quality detection is becoming more and more strict, especially the evaluation of the surface quality defects. The glass surface quality detection means, and ability has certain limitations, most methods for artificial detection, improve the detection accuracy and objectivity. The authors studied the glass surface quality of machine vision detection method based on the detailed discussion of computer vision and image processing technology. This study draws the conclusion that machine vision plays an important role in the quality detection of the glass surface and its high efficiency & accuracy has broad application prospects. [ABSTRACT FROM AUTHOR]

In recent years, with the aggravation of environmental pollution, relying on the traditional detection of environmental pollution has been unable to meet the current requirements for the environment, the use of advanced technology to detect environmental pollution has become crucial. In order to meet the requirements of today's era, with the rapid development of image recognition technology, it is necessary to study how to apply it to environmental pollution detection. The image recognition technology based on embedded technology has the characteristics of high discrimination, high acceptability and strong touch sense. It is of great significance for environmental monitoring to make full use of image recognition technology. Image recognition technology is to capture the same environment by digital camera, and then store the obtained image by computer. The HSI model is used to quantify the stored photos, and the variance of the gradient function is used to form photos and images with different levels of clarity, so as to analyze the environmental pollution according to the quantified results. In this paper, the ambient air quality is taken as the research object, and HSI model is used to quantify, so as to analyze the correlation of air detection results. [ABSTRACT FROM AUTHOR]