14118results about How to "Increased sensitivity" patented technology

The invention discloses a deep convolutional neural network-based lung cancer preventing self-service health cloud service system. The system comprises a convolutional neural network used for deep learning and training identification, a segmentation module which segments out a lung region from a CT image based on a full convolutional neural network, a deep convolutional neural network used for lung cancer diagnosis classification, and a self-service health cloud service platform used for performing early prevention and treatment according to an identified suspected lung cancer type. According to the system, the automation and intelligentization level of mobile internet-based lung cancer screening can be effectively improved, more citizens can know and participate in self-service health detection, assessment and guidance, the sensitivity, specificity and accuracy of early lung cancer screening and clinical diagnosis are improved, the lung cancer can be early discovered, early diagnosed and early treated, and the self-health management capability is enhanced.

The invention provides a method for detecting an X-ray mammary gland lesion image based on feature pyramid network under transfer learning. The method comprises the steps: 1, establishing a source domain data set and a target domain data set; 2, establishing a deformable convolution residual network layer by a deformable convolution and extended residual network module; 3, establishing a multi-scale feature extraction sub-network based on a feature pyramid structure through a feature map up-sampling and feature fusion method in combination with a deformable convolution residual network layer;4, establishing a deformable pooling sub-network sensitive to the focus position; 5, establishing a post-processing network layer to optimize a prediction result and a loss function; and 6, migratingthe training model to a small sample molybdenum target X-ray mammary gland focus detection task so as to improve the detection precision of the network model on the focus in the small sample image. According to the method, a transfer learning strategy is combined to realize focus image processing in a small sample medical image.

Disclosed are reagents and methods for reliably detecting the presence and measuring the amount of proteins, including proteins with various post-translational modifications (phosphorylation, glycosylation, methylation, acetylation, etc.) in a sample by the use of one or more capture agents that recognize and interact with recognition sequences uniquely characteristic of a protein or a set of proteins (Proteome Epitope Tags, or PETs) in the sample. Arrays comprising these capture agents or PETs are also provided.

MRI-compatible medical instruments and appliances are made using bulk metallic glass alloys. MRI-guided methods include the use of articles that include bulk metallic glass alloys.

A robotic system that can have a body and four flippers is described. Any or all of the flippers can be rotated. The flippers can have self-cleaning tracks. The tracks can be driven or passive. The robotic system can be controlled by, and send audio and/or video to and/or from, a remote operator control module. The methods of using and making the robotic system are also described.

The invention provides a natural scene text detection method and system. The system comprises two neural network models: a text detection network based on multi-level semantic feature fusion and a detection screening network based on an attention mechanism, wherein the text detection network is an FCN-based image feature extraction fusion network, and is used for extracting multi-semantic level information of input data, carrying out full fusion of multi-scale features, and finally predicting the position and confidence degree of text information in a natural scene by carrying out convolutionoperation on the fused multi-scale information. According to the detection screening network, the trained convolutional recurrent neural network is used for discriminating and scoring an initial detection result output by the convolutional neural network of the first part, so that a background which is easy to confuse with foreground characters is filtered out, and the accuracy of natural scene text recognition is further improved.

In an exemplary embodiment, a computer-implemented method determines a set of mesh polygons or fragments of the mesh polygons visible from a navigation cell. The method includes determining a composite view frustum containing predetermined view frusta and determining mesh polygons contained in the composite view frustum. The method includes determining at least one supporting polygon between the navigation cell and the contained mesh polygons. The method further includes constructing at least one wedge from the at least one supporting polygon, the at least one wedge extending away from the navigation cell beyond at least the contained mesh polygons. The method includes determining one or more intersections of the at least one wedge with the contained mesh polygons. The method also includes determining the set of the contained mesh polygons or fragments of the contained mesh polygons visible from the navigation cell using the determined one or more intersections.

The present invention is a distribution network cable-line mixed fault circuit selection method by using HHT detection technology. The invention uses HHT singularity detection technology to calibrate fault occurring time accurately, and implements circuit selection according to a principle that, when a fault occurs transient fault zero-sequence current component of the fault circuit and the regular circuit is opposite in direction. The specific method is using HHT to proceed EMD decomposition to the sampling signal of before and after 1/4 period each circuit fault zero-sequence current that is morphological filter pretreated, obtaining highest frequency IMF component of each circuit fault zero-sequence current and proceeding a first-order backward difference treatment to the IMF component, comparing the polar mutation of the fault zero-sequence current IMF component at fault moment, and then forming a circuit selection criterion. Analysis and simulation of the principle shows that, using fault information of 1/4 period after the fault to select circuit can provide an accurate and reliable circuit selection result and avoid effects of paraphrase and CT saturation on circuit selection accrracy.

A touch switch for an electrical appliance having a capacitive flat electrode, a light source, a transparent cover and a printed circuit board connected to the electrode is disclosed. The light source is mounted on a face of the printed circuit board opposite the transparent cover and the printed circuit board presents a cut-out for allowing light transmission from the light source towards the transparent cover.

The invention relates to a method for testing high impedance faults, which is based on transient traveling waves and belongs to the electric system protecting and controlling technique field. The method comprises faults actuating and faults testing, wherein the faults actuating adopts transient traveling waves electric quantity to judge actuating, The faults testing is actuated through comparing the transient traveling waves electric quantity and a threshold value. The fault testing adopts fault components to judge, faults testing results are output through an algorithm of fault detection. Three algorithms are adopted in high-resistance fault judgments, which include a high frequency algorithm, a harmonic algorithm, and a fundamental wave algorithm. The three algorithms can be used in either a single end form or in combination to judge high-resistance actuating. The method can test thoroughly each main frequency band information which may be produced when high resistance faults occur.

A capacitive touch switch having a printed circuit board and capacitive electrode provided on a surface of the printed circuit board is disclosed. The printed circuit board is interposed between a transparent planar light guide and the electrode, the planar light guide being attached to a first face of a transparent cover whose second face is adapted to be touched by the user, a light source being connected to the printed circuit board and being able to convey light to the planar light guide.

The invention discloses a rolling bearing fault identification method under variable working conditions based on ATT-CNN, and relates to a rolling bearing fault identification technology. The problemthat the generalization ability of an existing rolling bearing fault recognition method under variable working conditions is limited to a certain extent for a complex classification problem is solved.The method comprises the following steps: firstly, mapping vibration data to a nonlinear space domain through a convolutional neural network (CNN), and adaptively extracting rolling bearing fault characteristics under variable working conditions by utilizing the characteristic that the CNN has invariance on micro displacement, scaling and other distortion forms of an input signal; Secondly, an attention mechanism (ATT) thought is put forward to be fused into a CNN structure, and the sensitivity of bearing vibration characteristics under variable working conditions is further improved; And meanwhile, more abundant and diverse training samples are obtained through a data enhancement method, so that the network can be learned more fully, and the robustness is improved. The proposed fault diagnosis model based on the attention mechanism CNN (ATT-CNN) can realize multi-state recognition and classification of the rolling bearing under variable working conditions, and compared with other methods, higher accuracy can be obtained.

The invention discloses a method for thermal deformation error compensation of a digital control gear hobbing machine, which comprises the following steps of: 1, adopting temperature and displacement sensors to detect variable values of a temperature and a thermal deformation displacement on line when the digital control gear hobbing machine is in the course of working; 2, using a fuzzy clustering analytic method to calculate linearly dependent coefficients of variables of the temperature and the displacement, performing classification and optimization on the variable of the temperature, and determining the displacement variable Xi for model building and optimized p temperature independent variables T1, T2, T3, ..., and Tp; 3, adopting a multiple linear regression-least square method to establish a mathematic model of thermal errors and temperature variables; and 4, realizing online real-time compensation of thermal deformation errors of the digital control gear hobbing machine by using a zero programming system. The method solves thermal deformation error problems of a processing gear of the digital control gear hobbing machine, and by performing the online real-time error compensation on the digital control gear hobbing machine, the method improves the gear processing quality, the accuracy and the efficiency, reduces the rejection rate, saves the cost and shortens the processing cycle.

The invention discloses a needle amperometric determination type glucose sensor for subcutaneous tissue real-time monitoring and manufacturing method thereof. The sensor comprises a needle reference electrode (2) and at least one needle working electrode (1) including a conducting layer (8) and an enzyme membrane layer (6), and is characterized in that the working electrode (1) further comprises a polymer material inner membrane layer (7) and a polymer material control diffusion layer (5), and the conducting layer (8), the polymer material inner membrane layer (7), the enzyme membrane layer (6) and the polymer material control diffusion layer (5) are coated in turn from inner to outside. According to the manufacturing method in the invention, the glucose sensor has good stability, high flexibility, wide linear range of output current and glucose concentration, short response time, continuous real-time monitoring, good fastness and stability of enzyme bonding, high flexibility, selectivity and repeatability.