38results about How to "Improve diagnostic capabilities" patented technology

The invention discloses a two-stage multi-mode bearing fault diagnosis method based on a pre-training large model, and belongs to the technical field of bearing fault diagnosis. The method aims at solving the problems that a traditional method is poor in generalization and poor in robustness under multiple working conditions and small sample conditions. The method comprises the following steps: firstly, constructing a learnable multi-modal Tokens which comprises a multi-scale patch Token, a feature Token and a fault Token, and realizing efficient extraction and fusion of multi-modal features; a time-frequency semantic fusion module is introduced, and comprehensive time-frequency features are output through adaptive frequency coding, time coding and multi-modal fusion; and inputting the multi-modal feature sequence into a pre-training BERT model, and adopting a two-stage training strategy, in the first stage, performing self-supervised pre-training by taking mask signal reconstruction as a target, and in the second stage, performing parameter fine tuning by taking fault classification as a target. According to the method, the diagnosis accuracy and the cross-working-condition generalization ability under the small sample condition can be remarkably improved.

The application provides a wind tunnel pressure regulating valve opening set fault diagnosis method based on a composite criterion, relates to the technical field of equipment fault diagnosis, and comprises the following steps: training a fault diagnosis model by using seen fault sample, extracting sample features by using an encoder, and obtaining classification information and reconstruction information of the sample by using a classifier and a decoder; then, according to the output result of the model on training data, a composite criterion is constructed in a feature space, a probability space and a sample space, and is used for identifying unknown faults; finally, when a new fault sample is input, the composite criterion is used to judge the sample, accurate diagnosis of known faults and effective identification of unknown faults are realized. The composite criterion method has a clear structure, a simple implementation process, can be extended on the basis of an existing fault diagnosis model, and has good engineering application value.

The present application relates to the technical field of diagnostic markers, in particular to a marker combination and its application in diagnosing active tuberculosis and distinguishing between latent tuberculosis infection and active tuberculosis. The lectin combination provided by the present application can be used as a marker for ATB diagnosis and distinguishing between LTBI and ATB, and has high specificity and sensitivity. The lectin combination provided by the present application in combination with detection of specific antibodies of mycobacterium tuberculosis antigens can further improve the diagnostic effect. The marker and its detection products provided by the present application have good application potential in ATB diagnosis and distinguishing between LTBI and ATB, and are expected to overcome the limitations of existing diagnostic techniques, improve the diagnostic accuracy and sensitivity of tuberculosis, and provide new ideas and technical means for early detection, precise treatment and effective prevention and control of tuberculosis.

The application discloses a strategy for extracting fault features of electronic circuit systems, named as SSEST strategy, which is used for perceiving global information and paying attention to notable local information, and mining important local information means realizing expression of intermittent fault features of electronic circuit systems, specifically, first, S transformation is performed on a circuit output time sequence signal to acquire time-frequency domain features, then a squeeze and excitation network attention module is used to distribute channel weights, subsequently, input into a Swin Transformer framework, and pay attention to local information related to faults from global signals, and deep mining is performed on fault features, and two electronic circuits are taken as experimental circuits, the proposed diagnostic strategy realizes rapid and high-precision diagnosis, and shows that the proposed multiple attention mechanism is efficient for feature mining of intermittent faults of electronic circuit systems.

The application discloses a resource self-adaptive federated learning multi-modal knee osteoarthritis auxiliary diagnosis method, which comprises the following steps: a server initializes a multi-modal model for knee osteoarthritis diagnosis, and pre-deploys a low-rank adaptive plug-in in a trainable layer; after a client reports local hardware resources, the server allocates a resource echelon and issues a two-dimensional clipping ratio; the client clips the model in depth and width based on the clipping ratio, activates a LoRA module of a specified layer, loads local knee osteoarthritis X-ray images and clinical text data for multi-modal training, and establishes semantic mapping of image features and clinical descriptions; the client only uploads an activated LoRA parameter update amount, the server aggregates and optimizes a global model through weighted average aggregation and key layer compensation; and the learning ability of the model to key discriminative features of knee osteoarthritis is strengthened in combination with real-time resource monitoring and knowledge distillation. The application covers heterogeneous computing power devices through a resource self-adaptive mechanism, and improves the participation efficiency of primary medical institutions.

The application discloses a polypeptide ligand for brain glioma diagnosis and preparation and application of a radioactive molecular probe. The polypeptide targeting a TREM2 protein has the amino acid sequence as shown in the following: HLRKLRKR. The targeted polypeptide designed by the application has the characteristics of small molecular weight, strong blood brain barrier penetration and strong TREM2 protein binding capacity. The radioactive nuclide labeled molecular probe prepared by the application has the characteristics of strong targeting and easy observation as a brain glioma imaging drug, and is convenient for clinical application. The application improves the diagnosis effect of brain glioma and provides a new idea for early discovery and early treatment of brain glioma.

The invention discloses an aero-engine fault diagnosis method based on a conditional physical perception diffusion model, and the method is characterized in that the method comprises the steps: obtaining a working condition monitoring parameter and a performance monitoring parameter in the operation process of an aero-engine; based on the thermodynamic system characteristics of the aero-engine, constructing a physical constraint system; a conditional physical perception diffusion model is constructed, noise addition and physical violation term injection are carried out in a physical perception forward process, and noise removal and physical violation term correction are respectively carried out through a noise prediction branch and a physical correction branch of a double-branch network by taking a fault type label as a condition in a reverse process. The double-branch network completes pre-training based on a physical constraint system; training the conditional physical perception diffusion model; based on the working condition monitoring parameters and the performance monitoring parameters, physical perception features are extracted through middle layer representation of the condition physical perception diffusion model, and fault diagnosis of the aero-engine is achieved through a fault diagnosis classifier.

The invention relates to the technical field of neural image analysis, and particularly provides a brain function connection intelligent screening method and system for autism spectrum disorders, and the method comprises the steps: firstly obtaining resting state functional magnetic resonance imaging data and phenotype information thereof, and extracting a blood oxygen level dependence value time sequence of each brain region after preprocessing; then constructing a multi-scale brain network comprising a low-order function connection matrix and at least one high-order function connection matrix; the matrix is converted into a brain function connection graph containing sub-graphs of different scales through threshold sparsification; meanwhile, phenotype embedding features are extracted from phenotype information; the graph data and the phenotypic features are input into a multi-channel neural network model for parallel processing and fusion, and joint feature representation is obtained; and finally, outputting an auxiliary diagnosis result of the autism spectrum disorder through the classifier. According to the method, by fusing the multi-scale brain function connection information and the individual phenotype features, the accuracy of autism classification diagnosis and the generalization ability of the model are effectively improved.

The present application relates to the field of agricultural big data and fertilizer supply chain management, and particularly to a selenium-rich bio-organic fertilizer supply chain collaborative management system based on big data, comprising: a source evaluation step: a management center calls material attributes, a raw material evaluation unit analyzes source quality fluctuations to distinguish batches; an efficiency drift analysis step: in response to abnormal control signals, a conversion analysis unit performs efficiency drift evaluation feedback analysis; a supply-demand collaborative matching step: in response to regular control signals, a collaborative demand unit obtains a collaborative demand coefficient, and a matching division unit performs quantitative matching analysis of control decision; a control execution step: compare control evaluation indexes to generate control signals; the present application constructs a source fluctuation perception and supply-demand collaborative matching closed loop, solves the spatiotemporal mismatch contradiction of raw material fluctuations and soil adsorption specificity, and improves the overall efficiency of the supply chain.

The invention discloses an evaporator control method and a control system thereof, and relates to the technical field of evaporation process control. According to the method, a material characteristic coefficient, a thermal efficiency coefficient and a heat transfer performance coefficient are obtained, and the thermal working condition adaptation degree is obtained by combining the pressure of an evaporation chamber and the vapor phase temperature; based on the current feeding flow, the adaptation degree and the heat transfer performance coefficient, the optimized feeding flow is obtained through a flow optimization model. According to the method, through multi-model collaborative analysis, dynamic optimization adjustment of the feeding flow of the evaporator is achieved, the operation stability, the thermal working condition adaptability and the overall heat transfer efficiency of the evaporation process are effectively improved, and the self-adaptive control capacity of an evaporation system is remarkably improved.

The application discloses a method and a kit for detecting urinary system cancer based on urine DNA methylation, and belongs to the field of biotechnology and medical detection. The method only needs to take 10 mL of urine, centrifuges the supernatant, takes urine sediment to perform DNA extraction and bisulfite conversion. The converted DNA is used to detect the methylation state of four methylation target genes H4C6, SIX6, SHOX2 and SEPTIN9 by real-time fluorescent quantitative PCR. The application establishes a methylation marker combination taking H4C6, SIX6, SHOX2 and SEPTIN9 as the core and a matching detection method and kit, and realizes non-invasive, stable and efficient detection of various malignant tumors (bladder tumor, ureter tumor, renal pelvis tumor, kidney tumor and prostate tumor) of the urinary system.

According to the multi-label automatic diagnosis and evaluation system and method for ankylosing spondylitis, pelvic X-ray film images and clinical data are combined, automatic diagnosis of ankylosing spondylitis is achieved through a multi-label classification model, and the severity degree of injury of sacroiliac joints and hip joints is evaluated. The system adopts a prior attention mechanism, can automatically focus a key area in an image, and further improves the diagnosis efficiency. Experimental results show that the system still has relatively high diagnosis accuracy in an environment with limited resources, and can be comparable with diagnosis results of experienced doctors. The multi-label classification model with the priori attention mechanism provided by the invention provides a promising and cost-effective tool for diagnosis and evaluation of ankylosing spondylitis, and has important practical application value.

This invention relates to a method for NAION prediction and acute phase differentiation based on deformable convolution and multi-site OCTA, belonging to the field of ophthalmic disease diagnosis and medical image processing technology. This invention acquires OCTA images of ocular samples and extracts data from key sites. Using the OCTA deformable convolution feature extraction module as the core feature extraction backbone network, its optimal performance is verified through single-site experiments, and a multi-site joint diagnostic framework is constructed. Radiomics features are introduced, and deep fusion of deformable convolution features and radiomics features is achieved through a bidirectional cyclic feature interaction module. Finally, a hybrid expert module is added to the three-site joint model to decouple NAION prediction from the acute phase differentiation task of NAION / ON, improving diagnostic accuracy. This invention effectively solves the problems of difficult accurate differentiation between the acute phases of NAION and ON and insufficient early prediction of NAION, providing reliable technical support for ophthalmic clinical diagnosis.

The application provides a precast bending member forming process deformation intelligent monitoring method and system, relates to the technical field of deformation intelligent monitoring, and the method comprises the following steps: obtaining an interference fringe image of a precast bending member surface, performing phase synthesis and integral processing to generate a displacement gradient field, then performing phase unwrapping and denoising processing to obtain a reconstructed displacement field; synchronously collecting strain data and temperature data, and interpolating and reconstructing into a multi-physical quantity field; then, based on the multi-physical quantity field, the displacement signal in the reconstructed displacement field is decomposed into an intrinsic modal function component by using a variational modal decomposition algorithm, then recombination is performed to obtain a sub-deformation field, and the sub-deformation field is superimposed with a strain characteristic field and a temperature characteristic field to form a multi-modal characteristic tensor; finally, a complex-valued convolutional neural network is used to extract complex domain features, and the complex domain features are matched with a deformation database to identify the deformation category in the forming process. The application can accurately monitor the slight deformation of the precast member under the action of multiple fields.

This invention discloses a specific plasma metabolic biomarker combination for early-onset type 2 diabetes and its application. The specific plasma metabolic biomarker combination contains 14 peripheral blood amino acid and fatty acid biomarkers. The amino acid biomarkers include leucine, valine, isoleucine, γ-aminobutyric acid (GABA), dimethylglycine, glutamic acid, and GABA; the fatty acid biomarkers include C18:1, C18:2, C18:3a, C18:4, C18:3r, C22:5, and C22:6. This invention significantly improves the ability to identify early-onset type 2 diabetes by screening plasma free amino acid and fatty acid biomarkers and combining them with a stratified analysis strategy based on the age of onset. The 14 candidate metabolic biomarkers screened show significantly better diagnostic efficacy in early-onset type 2 diabetes than in late-onset type 2 diabetes, indicating that the metabolic biomarkers screened in this invention have higher sensitivity and specificity for early-onset type 2 diabetes and can effectively compensate for the shortcomings of existing technologies in identifying early-onset individuals.

The application discloses a feature automatic extraction method around a tooth boundary and a lesion recognition method, relates to the technical field of oral medicine, and comprises the following steps: acquiring a digital image of an oral X-ray apical film to be processed; performing automatic segmentation and contour extraction on a target tooth in the digital image of the oral X-ray apical film to obtain a tooth contour line of the target tooth; performing normalization processing on the tooth contour line; and traversing each boundary pixel point on the tooth contour line to generate an edge band feature map of the target tooth. The application can effectively extract local features highly related to lesions, especially features of a tooth boundary region, from the oral X-ray apical film, so that the recognition capability for early micro-lesions is improved.

The invention is applied to the technical field of mechanical equipment state monitoring and intelligent fault diagnosis, and discloses a multi-modal feature fusion domain self-adaptive diagnosis method and system and a storage medium, through a feature correction module (FCM), a cross-modal guide mechanism is utilized to carry out bidirectional self-adaptive calibration on depth features of vibration and current signals, and the depth features of the vibration and current signals are corrected. Feature distribution is aligned, and modal specific noise is suppressed; then, the feature fusion module uses a cross attention mechanism to realize deep interaction and selective information enhancement based on the corrected features; finally, the generated high-quality fusion features are sent into a conditional domain adversarial network, domain invariant features are learned through class-aware adversarial training, and accurate diagnosis is achieved. According to the method, heterogeneity and noise interference among modals are effectively overcome, deep interaction and complementary information enhancement of multi-modal features are realized, and the fault diagnosis accuracy and robustness of the model on a cross-working-condition migration task with relatively large inter-domain difference are remarkably improved.

This invention provides a magnetic resonance imaging (MRI) apparatus and its control method. In contrast-enhanced MRI, it improves the ability to depict tissues reached by the contrast agent and shortens the overall imaging time. The imaging unit of the MRI apparatus is equipped with a gradient echo pulse sequence for acquiring T1-emphasis images containing fat-suppressing pulses. The control unit performs the following control: the imaging unit repeats the pulse sequence for a given time from the time the contrast agent is administered to the subject, generating images of multiple phases at different arrival locations of the contrast agent. At this time, starting from the phase about to reach the target tissue, a preparatory pulse is added before the pulse sequence to suppress signals from the contrast agent present outside the target tissue (cells).

A kind of permanent magnet synchronous motor bearing fault diagnosis method based on motor current analysis method, first acquisition permanent magnet synchronous motor stator U, V phase current signal, calculate W phase current;Three-phase current is obtained by space vector dimension reduction two-phase current under rectangular coordinate system, and two-phase current is carried out vector normalization processing;Based on the mechanism of action of bearing fault to current signal and the characteristics of current signal, the signal after space vector dimension reduction is carried out variation modal decomposition, the approximate entropy of modal component is calculated and the feature matrix is formed;The improved standard artificial bee colony algorithm makes the bee colony type mutually transform according to the optimal solution, and changes the bee colony initialization population generation rule, the optimal classification model is obtained by using the optimized adaptive variable type algorithm to optimize machine learning model parameter;Using optimal model, the test set in sample set is classified according to fault type, and compared with label to obtain test set accuracy rate;The present application has the advantages of high accuracy, high diagnostic efficiency and the like.

This application discloses an incremental training method and a computer-readable storage medium for a dental caries diagnosis model. The method includes: acquiring at least one initial dental caries diagnosis image output by the dental caries diagnosis model, wherein the initial dental caries diagnosis image is an image labeled with dental caries diagnosis annotations obtained by the dental caries diagnosis model from an existing dental caries image; responding to a user's correction operation on the initial annotations in the initial dental caries diagnosis image, collecting the user's correction behavior information, and generating a first dental caries diagnosis image labeled with at least one correction annotation; determining a quality score for each correction annotation based on the correction behavior information; constructing an incremental training dataset based on the first dental caries diagnosis image labeled with target correction annotations of high quality scores, and its corresponding initial dental caries diagnosis image; and incrementally training the dental caries diagnosis model based on the incremental training dataset and the original training dataset of the dental caries diagnosis model.

The application discloses a kind of control moment gyro high-speed bearing sliding state monitoring method based on transfer learning, belong to rolling bearing fault diagnosis technical field, comprising the following steps: S1, builds control moment gyro high-speed bearing sliding test bench, obtains motor current experimental data and bearing sliding data;S2, obtains motor current simulation data;S3, extracts current feature;S4, current feature is classified;S5, based on the current feature after classification, bearing sliding prediction model is constructed and trained, and motor current experimental data and motor current simulation data are input to the bearing sliding prediction model after training, and the classification result of bearing skidding is generated.The application has the advantages of high prediction accuracy and shorter running time in bearing sliding state prediction task based on simulation data training.

This invention relates to a group of extracellular vesicle-derived biomarkers for gastrointestinal malignancies and their applications, belonging to the field of molecular biology. The biomarkers of this invention are derived from extracellular vesicles and include proteins CALR, SPP1, OLFM4, and OIT3. When used for the diagnosis of gastrointestinal cancers, these biomarkers have shown good diagnostic efficacy, superior to existing serum biomarkers. Furthermore, the biomarkers of this invention show significant elevations in the early stages of gastrointestinal cancer, which is beneficial for early screening and diagnosis of gastrointestinal cancers.

The application provides a model continuous learning optimization method, an electronic device and a storage medium. By inputting an initial sample data set into a diagnosis model, a core gradient space of the initial sample data set when training the diagnosis model is obtained. When subsequently optimizing the model by using a new sample data set, the diagnosis effect on historical data can be maintained, the diagnosis capability on new data can be improved, the catastrophic forgetting problem of the diagnosis model can be avoided, and the training cost and workload of the model are not increased. Therefore, the compatibility of the model on different data can be improved while meeting clinical requirements. A new sample data set is obtained according to the initial sample data set and the new sample data set. A gradient transfer value is obtained according to the new sample data set, the core gradient space and the new sample data set. The target gradient of the diagnosis model is determined according to the gradient transfer value, the direction of the optimization of the diagnosis model can be distinguished, and the overall effect of the optimization of the diagnosis model can be improved.

The invention belongs to the technical field of power equipment fault early warning diagnosis, and provides a power equipment fault early warning self-learning method and system, and the technical scheme is that the method comprises the steps: obtaining the historical multi-source sensing data of power equipment, and carrying out the preprocessing of the data, and constructing normal training data and abnormal training data; training the fault early warning model based on the normal training data to obtain a trained fault early warning model; training the fault diagnosis model based on the abnormal training data to obtain a trained fault diagnosis model; inputting the collected real-time operation data of the power equipment into the trained fault early warning model for state judgment to obtain early warning information; determining whether the fault early warning model has misjudgment according to the early warning information, and judging the type of the misjudgment if the fault early warning model has misjudgment; and carrying out self-learning on the fault early warning model by adopting a corresponding self-learning updating mechanism according to different misjudgment types to obtain an updated fault early warning model.

The application belongs to the technical field of tunnel engineering, and particularly relates to a kind of concrete surface automatic inspection robot and method in narrow space tunnel, the scheme completely abandons the traditional method of "uniform scanning, full collection", adopts the two-stage collaborative mechanism of "preliminary identification" and "key diagnosis" driven by intelligence, realizes structural reduction at data source, and theoretically can release more than 90% of scanning resources and data processing bandwidth from "invalid area", and redistribute them to high-value areas; at the same time, through dynamic calculation of interest threshold, the system can adapt to the differences of inspection environment and the slight differences of disease characteristics, not only solve the fundamental problem of data redundancy and storage and transmission pressure, but also ensure the attention to subtle and early diseases, fundamentally improve the detection probability of hidden dangers and the overall efficiency of the perception subsystem.

This invention discloses a method for constructing a benign / malignant prediction model based on CT images of pancreatic tumors. The method includes: acquiring preoperative pancreatic CT images through an image storage and transmission system; identifying and preprocessing a region of interest (ROI) based on the pancreatic cancer tumor region in the CT images; extracting radiomics features from the processed ROI; filtering the extracted features using a maximum correlation minimum redundancy algorithm to select features related to the benign / malignant nature of the pancreatic tumor; constructing a benign / malignant prediction model based on the selected features using a support vector machine model; and outputting the final prediction result. The method described in this invention enables rapid and accurate classification of benign / malignant pancreatic tumors, providing auxiliary information for clinical diagnosis and assisting physicians in developing personalized treatment plans.

The application discloses a group of antibodies specifically recognizing IgA1 with galactose deficiency and application thereof, and belongs to the technical field of antibodies. The main problem to be solved by the application is how to obtain anti-Gd-IgA1 antibodies with high specificity. In order to solve the above technical problem, the application provides antibodies or antigen binding fragments thereof, and the antibodies can be antibody B0020, antibody B0035 or antibody B0043. Compared with the existing antibody KM55, the obtained antibodies have the advantage of high affinity with antigens, and are expected to be more effectively used for diagnosis of IgA nephropathy, monitoring of treatment efficacy, judgment of prognosis and the like.

The application provides a photovoltaic IV data processing method and device, electronic equipment and storage medium. The method comprises the following steps: acquiring IV data obtained by scanning a target photovoltaic module in a target period; performing normalization processing on the IV data according to one of at least one preset normalization mode; the at least one preset normalization mode comprises a normalization mode based on a selected standard IV curve of the target photovoltaic module; wherein the selected standard IV curve is a standard IV curve of the target photovoltaic module corresponding to the irradiance value of the target period corrected by a preset correction mode, and the preset correction mode is a correction mode for correcting the irradiance value collected by an irradiance meter according to the angle parameter of the target photovoltaic module. The application can improve the diagnosis effect of the photovoltaic module.

The invention discloses a steam turbine defect composite voiceprint diagnosis device, and particularly relates to the technical field of steam turbine diagnosis, the steam turbine defect composite voiceprint diagnosis device comprises a track body, a diagnosis assembly, a mounting and dismounting mechanism and a telescopic assembly; the diagnosis assembly is arranged on the track body; the diagnosis assembly comprises a moving plate, a base, a rotating arm A, a connecting arm, a rotating arm B, a fixed plate, a voiceprint detector, a vibration sensor and a camera body; the moving plate is slidably arranged on the track body through the driving assembly. The base is connected with the moving plate through the mounting and dismounting mechanism; the rotating arm A is rotationally connected with the base through a rotating motor A; the connecting arm is arranged above the rotating arm A; the rotating arm B is rotationally connected with the connecting arm through a rotating motor B; the fixed plate is arranged on one side of the rotating arm B; the voiceprint detector, the vibration sensor and the camera body are all arranged on the fixing plate; the device is reasonable in structural design, capable of achieving efficient, continuous and automatic monitoring, comprehensive in coverage and low in use cost, and the diagnosis accuracy is greatly improved.

The invention belongs to the technical field of biology, and particularly relates to a group of markers for predicting the immunotherapy effect of EGFR gene mutation NSCLC patients and application of the markers. The biomarker disclosed by the invention is simple and convenient to sample and quick in result reporting. The biomarker disclosed by the invention is high in accuracy: the AUC of CCL4 is equal to 0.771, the sensitivity is 0.64, the specificity is 0.67 ([95% CI: 0.62-0.93]); plt; 0.05) of the substrate (1); the AUC of the PD-L1 is equal to 0.720, the sensitivity is 0.60, the specificity is 0.82 ([95% CI: 0.55-0.89]); plt; 0.05) of the method. In a screening queue, a combined diagnosis ROC curve of CCL4 and PD-L1 is as follows: AUC is equal to 0.907, the sensitivity is 0.667, and the specificity is 1.000 ([95% CI: 0.696-1.000]); plt; 0.05) of the substrate (1); in the verification queue, the AUC of the combined diagnosis of CCL4 and PD-L1 is equal to 0.833, the sensitivity is 0.69, the specificity is 0.91, and the AUC is less than [95% CI: 0.70-0.96]; plt; 0.05) of the substrate (1); therefore, the efficiency of combined diagnosis of the CCL4 and the PD-L1 is higher than that of diagnosis by independently using the CCL4 or the PD-L1; the marker is high in diagnosis efficiency, the immunotherapy curative effect of the NSCLC patient with EGFR gene mutation is predicted, and clinical guidance is provided.