27 results about "Lung tuberculosis" patented technology

The invention discloses a pulmonary tuberculosis image classification and recognition system based on feature aggregation and focus area perception, and the system comprises the steps: carrying out the local and global feature extraction and fusion of a pulmonary tuberculosis image through a feature extraction subsystem; the method comprises the following steps: constructing a graph structure of fusion features through a pulmonary tuberculosis classification subsystem based on focus area perception, dynamically aggregating node information, realizing three classifications of health, tuberculosis and disease but non-tuberculosis in a first stage, and carrying out active tuberculosis and old tuberculosis subdivision in a second stage on a pulmonary tuberculosis image in combination with focus structure features; through an unbalanced data set processing subsystem based on characteristic aggregation and dynamic data amplification, under-sampling is carried out on the characteristic that active tuberculosis data samples are large by adopting a characteristic aggregation technology, and resampling is carried out on old tuberculosis with few data samples by implementing a dynamic data amplification strategy; and key information learned from active tuberculosis categories is reserved through a knowledge distillation mechanism, so that more accurate focus recognition and classification are realized.

The invention discloses an active / inactive pulmonary tuberculosis recognition system based on an improved 3D network, and the system comprises an image feature obtaining module which is used for obtaining pulmonary tuberculosis CT chest image features; the data preprocessing module is used for carrying out data preprocessing on the obtained pulmonary tuberculosis CT chest image features; the image segmentation module is used for segmenting the preprocessed pulmonary tuberculosis CT chest image features to obtain pulmonary parenchyma image features, and an adopted segmentation model is a 3D ResUNet segmentation model; the image classification and recognition module is used for performing prediction and classification on the lung parenchyma image features after image segmentation; wherein the adopted image classification model is a 3D ResUNet50 classification model, and the 3D ResUNet50 classification model is combined with a multi-scale attention module to obtain image output features; then, image output features are processed by adopting grouping convolution and channel shuffling, the types of active and inactive pulmonary tuberculosis are predicted, and confidence scores of the model are generated; the CT image classification accuracy of the model can be improved.

The invention discloses a pulmonary tuberculosis focus map segmentation system based on symmetric similarity and dynamic sample amplification, which comprises a sternum suppression network module, an entropy-driven sample number dynamic amplification module, a pulmonary tuberculosis focus detection module and a sternum suppression network module, and is used for removing a sternum image in a chest X-ray map; the entropy-driven sample dynamic amplification module is used for carrying out dynamic amplification on the number of sample images based on the complexity of the image and the existing training state of the model for a single image; specifically, for each image, an information entropy is calculated for probability distribution output by a model of each training round; in the early stage of training, the number of training image samples is amplified slightly, and the number of amplified samples is increased along with model training; and the pulmonary tuberculosis focus detection module is used for designing a pulmonary tuberculosis focus detection network model by using the target detector and the symmetric similarity of the pulmonary tuberculosis images, and segmenting a tuberculosis focus map from the X-ray map.

The invention provides a data processing device for pulmonary tuberculosis disease risk assessment and application thereof. The data processing device provided by the invention comprises a memory, a processor and a computer program stored on the memory, and the processor executes the computer program to realize the following steps: receiving gene expression quantity data in a blood sample of a subject, performing pulmonary tuberculosis disease risk grouping on the subjects according to the gene expression quantity data; the gene at least comprises an aspartate beta-hydroxylase structural domain 2 gene (ASPHD2). The data processing device provided by the invention can be used for evaluating or assisting in evaluating the pulmonary tuberculosis disease risk of the subject, improves the accuracy of risk evaluation, and has a relatively high clinical application value.

The application discloses a kind of non-tuberculous mycobacterial lung disease and pulmonary tuberculosis CT image classification method, mainly including the following steps: image feature map is extracted by encoder and image feature map is divided into multiple examples, each example represents the feature information in certain area of image;The correlation weight of example and classification task is calculated, and the example is aggregated into bag feature representation according to the weight to express the characteristics of CT image;Bag feature representation is predicted as CT image category by classifier;According to the correlation weight, important examples are selected, and are evaluated by classifier, to express the correlation degree of the learning content and expected content of network;Loss function is constructed, including bag classification loss and correlation loss, respectively representing the difference between CT predicted value and true value and the difference between model learning content and expected content.The method can realize the binary classification of non-tuberculous mycobacterial lung disease and pulmonary tuberculosis CT image, and is superior to the classification accuracy of existing method.

The invention discloses a pulmonary tuberculosis detection method and system based on deep learning, and relates to the field of medical image analysis. The system comprises an image acquisition module for acquiring a lung CT image; the image preprocessing module is used for preprocessing the CT image; the shared feature extraction module is used for extracting feature vectors of the CT slices; the first tissue information extraction module is used for extracting tissue information and updating feature vectors of the CT slices; the spatial information extraction module is used for acquiring spatial information and updating feature vectors of the CT slices; the second tissue information extraction module is used for acquiring a feature vector of the CT slice containing fusion information; the splicing module is used for acquiring three groups of splicing feature vectors; the attention pooling module is used for acquiring three groups of packet feature vectors of a packet where the CT slices are located; the feature aggregation module is used for acquiring a final packet feature vector of the CT packet; and the classification prediction module is used for obtaining a dichotomy prediction result. According to the method, the focus recognition rate of the pulmonary tuberculosis diagnosis system is improved, only classification labels need to be provided, and dependence on label segmentation is reduced.

This invention discloses miRNA biomarkers for the auxiliary diagnosis of tuberculosis and their applications. This invention claims protection for the use of substances for detecting miRNAs in serum exosomes in the preparation of products; the purpose of said products is for the diagnosis or auxiliary diagnosis of tuberculosis patients or for the diagnosis or auxiliary diagnosis of active pulmonary tuberculosis patients. The miRNAs are any one or any combination of the following five miRNAs: miR-28-3p, miR-193b-5p, miR-370-3p, miR-1246, and miR-2110. This invention also claims protection for the use of substances for detecting serum exosome miRNAs in the preparation of products; the purpose of said products is for the diagnosis or auxiliary diagnosis of latent tuberculosis infection. The miRNAs are any one or any combination of the following three miRNAs: let-7e-5p, let-7d-5p, and miR-140-5p. This invention lays the foundation for further development of rapid early diagnosis techniques and methods for tuberculosis.

The invention discloses a biomarker for early risk early warning of lung cancer combined with pulmonary tuberculosis, an early risk early warning model and application of the biomarker and the early risk early warning model. In particular discloses application of a biomarker (ANK2 gene, CA1 gene, EPB42 gene, HBB gene, HBD gene and / or MYL4 gene) and / or a substance for detecting the biomarker in preparation of a product for early risk early warning of lung cancer combined with pulmonary tuberculosis or in construction of an early risk early warning model of lung cancer combined with pulmonary tuberculosis. The biomarker and the model constructed by using the biomarker are accurate and reliable through ROC curve verification, have good sensitivity and specificity, and are suitable for wide popularization. By performing early risk early warning of lung cancer and pulmonary tuberculosis on the to-be-detected subject, decision support can be provided for clinicians, early dry treatment and the like can be performed on high-risk patients, and optimal disease management is realized.

The present invention relates to a combination of (i) a flagellin polypeptide and (ii) an antibiotic, for the simultaneous or sequential use in the treatment of bacterial infections due to Pseudomonas aeruginosa antibiotic-resistant strain. The present invention also provides a flagellin polypeptide, for use in a method for enhancing sensitivity to an antibiotic of a patient suffering from bacterial infection, due to Pseudomonas aeruginosa antibiotic-resistant strain. Indeed, using an in vivo mouse model, inventors show that while standalone prophylactic flagellin attenuates infection caused by a MDR strain of PA (PAMDR), its combination with an antibiotic such as gentamicin (GNT) leads to a strong decrease in bacterial load in the lung and a significant reduction in cell infiltration and inflammatory cytokines. Moreover, mice receiving flagellin in combination with GNT showed a 100% survival rate, a result that GNT alone was unable to provide against PAMDR.

The application discloses a lung nodule diagnosis marker and application, and belongs to the technical field of biotechnology detection. The application provides the use of one or more of PLA2G6, PLA2G7, AKR1C1, AKR1C3, LTA4H and PTGER4 genes as biomarker molecules for diagnosing pulmonary sarcoidosis. The use of the biomarkers can assist biopsy and help diagnose related lung diseases. The genes for identifying sarcoidosis provided by the application have an accuracy rate (area under the ROC curve) of 96.5%, 90.9%, 89.1%, 87.5%, 87.2% and 85.8% for distinguishing healthy people, lung adenocarcinoma patients and pulmonary tuberculosis patients in sequence, and the comprehensive accuracy rate is 90.9%. The application collects a certain amount of clinical samples for research and verification, and the research results are in line with clinical practice, and have a good clinical application prospect.

The invention belongs to the field of detection, and particularly relates to a cavity pulmonary tuberculosis marker based on serum DSG4 protein and amyl salicylate and a detection kit. The marker is DSG4 protein and / or amyl salicylate. The invention provides a biomarker for distinguishing a subject with cavernous tuberculosis cPTB or non-cavernous tuberculosis PTB by systematically analyzing proteomics and metabonomics characteristics of PTB and cPTB patients. According to the technical scheme, new insights and clues are provided for early diagnosis and personalized treatment of cPTB. And new scientific evidences are provided for early diagnosis and personalized treatment strategies.

The invention discloses a construction method and application of a pulmonary tuberculosis detection model, and belongs to the technical field of image processing, and the method specifically comprises the steps: obtaining a medical image with a label category from a database, carrying out the preprocessing, positioning a lung region, inputting the medical image into a convolutional neural network, obtaining a rough segmentation result of the lung image, and obtaining a final segmentation result of the pulmonary image; in combination with the uncertain segmentation edge image, identifying the segmentation edge of the lung, obtaining a lung region image, and carrying out noise reduction on the lung region image; performing feature extraction on the lung region image after noise reduction; obtaining a normal data value and an abnormal data value according to the key features; constructing a training sample set according to the normal data values and the abnormal data values; and performing model training through the training sample set to obtain a prediction model for predicting pulmonary tuberculosis. According to the invention, rapid determination of pulmonary tuberculosis is realized.

The application discloses an aptamer-nano gold test strip for screening human pulmonary tuberculosis and an application thereof, and belongs to the technical field of biological detection. The application successfully constructs an aptamer-nano gold test strip for screening human pulmonary tuberculosis, realizes rapid detection of a pulmonary tuberculosis marker ESAT-6 antigen, adopts mercapto-modified aptamer to modify nano gold to form stable gold label probes, adopts a mixed solution composed of a biotin-modified aptamer solution and a streptavidin solution, and constructs an ESAT-6 antigen test strip by using an ESAT-6 antigen solution. The constructed test strip has good sensitivity in ESAT-6 antigen detection, has good detection effect in the antigen concentration range of 100 pg / mL-500 pg / mL, and has good specificity, stability, repeatability and accuracy.

The invention discloses an application of an extracellular vesicle B and T lymphocyte weakening factor in active tuberculosis diagnosis, relates to the technical field of biomedicine, and particularly discloses an application of the extracellular vesicle B and T lymphocyte weakening factor as a diagnostic marker of active tuberculosis in preparation of an active tuberculosis diagnostic product. The extracellular vesicle B and T lymphocyte weakening factor can effectively distinguish active tuberculosis patients from healthy people, tuberculosis latent infection patients and lung cancer patients, can be used for diagnosing active tuberculosis, and has high sensitivity and specificity.

The invention discloses an intelligent pulmonary tuberculosis detection method based on images and clinical data, and relates to the technical field of medical image detection.The method comprises the steps that chest X-ray medical images and patient clinical data are collected, cross-modal data enhancement is conducted through a conditional generative adversarial network, and an enhanced multi-modal data set is generated; feature reconstruction based on an attention mechanism is carried out on the pulmonary tuberculosis multi-modal feature sequence, association weights among different modal features in the pulmonary tuberculosis multi-modal feature sequence are calculated, and a pulmonary tuberculosis discrimination feature tensor is generated; and performing quantitative analysis on the pulmonary tuberculosis discrimination feature tensor, evaluating the global feature intensity of the pulmonary tuberculosis, performing pulmonary tuberculosis positive risk level determination according to the global feature intensity of the pulmonary tuberculosis, and generating a pulmonary tuberculosis risk determination conclusion. According to the method, cross-modal data enhancement is carried out by adopting the conditional generative adversarial network, so that the diversity and authenticity of medical image detection training data are remarkably improved, and the generalization ability of image detection is enhanced.

The invention discloses a T cell receptor for specifically recognizing mycobacterium tuberculosis antigen peptide as well as a preparation method and application of the T cell receptor. According to the invention, HLA-A * 0201 positive pulmonary tuberculosis patient PBMC is respectively stimulated by tuberculosis antigens CFP10 and Ag85B, CD8 + T cells are subjected to flow sorting, TCR alpha / beta chains are paired, corresponding HLA-A * 0201 restrictive TCR protein is successfully obtained, and eukaryotic expression and purification are carried out. The TCR protein can significantly inhibit proliferation of mycobacteria in THP-1 macrophages, shows that CFU is reduced and high content fluorescence intensity is weakened, and significantly improves expression of inflammatory factors such as IL-1beta, TNF-alpha, IL-6, IL-12 and the like. The tuberculosis antigen-specific TCR protein is proved to effectively inhibit intracellular mycobacterium tuberculosis replication by regulating the polarization state of macrophages and an inflammatory signal network for the first time, and a new strategy is provided for TCR-based tuberculosis immunotherapy.

The application discloses an activity / inactivity tuberculosis recognition system based on an improved 3D network, which comprises an image feature acquisition module, a data preprocessing module, an image segmentation module and an image classification recognition module. The image feature acquisition module is used for acquiring lung tuberculosis CT chest image features. The data preprocessing module is used for performing data preprocessing on the acquired lung tuberculosis CT chest image features. The image segmentation module is used for segmenting the preprocessed lung tuberculosis CT chest image features to acquire lung parenchyma image features, and a segmentation model adopted is a 3D ResUNet segmentation model. The image classification recognition module is used for performing prediction classification on the lung parenchyma image features after image segmentation. An image classification model adopted is a 3D ResUNet50 classification model, and the 3D ResUNet50 classification model is combined with a multi-scale attention module to obtain image output features. Then, grouping convolution and channel shuffling are used to process the image output features, the categories of active and inactive tuberculosis are predicted, and the confidence score of the model is generated. The application can improve the CT image classification accuracy of the model.

The invention discloses an AI-based medical image X-ray pulmonary tuberculosis medical information management system. The system comprises an image data analysis module, a digital model construction module, a particle dynamic simulation module, a feature focusing extraction module and an auxiliary evaluation output module. The system obtains an anatomical structure, a breathing phase state and an active focus area serving as a dissemination source by analyzing a target medical image; a personalized lung model is constructed, and model parameters including a simulated airway and airflow are generated; further simulating diffusion and deposition of biological particles in the bidirectional airflow by taking the dispersion source as a starting point, and generating dispersion risk probability distribution data; taking the data as a spatial attention weight, carrying out feature extraction on the image, focusing and detecting tiny anomalies in a high-risk area, and generating a tiny focus detection result; and finally, fusing the probability distribution data and a detection result, and outputting a visual evaluation report. Detection is guided by simulating a disease dissemination mechanism, and the recognition capacity of tiny focuses is remarkably improved.

The invention discloses an intelligent pulmonary tuberculosis identification method based on a CNN-ViT-Meta fusion model. The method comprises the following steps: carrying out data preprocessing on a disclosed pulmonary tuberculosis chest X-ray CT image data set; constructing a CNN-based metadata inversion model, performing metadata inversion and supplementation on the chest X-ray CT image, deducing the age and gender of the patient, and generating a metadata label; constructing a CNN-ViT-Meta fusion model, extracting low-level local features of the lung image by using CNN, capturing global feature information by ViT, and generating an image feature sequence; and meanwhile, inputting the metadata into a full connection layer to generate a metadata feature sequence, integrating the image feature sequence output by CNN-ViT and the metadata feature sequence through a feature fusion module, and outputting a pulmonary tuberculosis recognition result. Performance indexes such as accuracy are evaluated through the prediction result of the test set and the real label, and the recognition performance and generalization ability of the model are verified.

The present invention discloses a bimetallic synergistic promotion system and its application in the asymmetric synthesis of (R, S)-TBAJ-876. Specifically, a metal lithium, sodium or potassium salt forms an amino metal compound with a chiral or achiral amine, and the amino metal compound and another metal form a synergistic promotion system. Under the action of a suitable ligand, the addition reaction of 6-bromo-3-(2,3,6-trimethoxypyridine)-2-methoxyquinoline and 3-N,N-dimethylamino-1-(2,6-dimethoxypyridine)-1-acetone is synergistically promoted under low temperature conditions. This bimetallic synergistic promotion system has achieved the asymmetric synthesis of the anti-tuberculosis candidate drug (R, S)-TBAJ-876 with high yield and high selectivity for the first time.

The invention provides an intelligent auxiliary diagnosis method and device for pulmonary tuberculosis and a readable storage medium, and the method comprises the steps: obtaining a lung medical image of an object, the lung medical image comprising a normal position chest radiograph and a lateral position chest radiograph; performing focus recognition on the lung medical image through a pre-trained recognition model to obtain a normal focus recognition result and a lateral focus recognition result; performing three-dimensional reconstruction through an anatomical feature matching algorithm according to the orthotopic chest radiograph and the lateral chest radiograph to obtain a chest three-dimensional anatomical structure; obtaining a focus positioning result in the chest three-dimensional anatomical structure based on an anterior focus recognition result and a lateral focus recognition result; and according to the focus positioning result in the chest three-dimensional anatomical structure and the clinical data of the pulmonary tuberculosis patient, obtaining a pulmonary tuberculosis diagnosis result of the object. The method can improve the focus positioning precision, enhance the model generalization ability, and improve the accuracy of the pulmonary tuberculosis diagnosis result.

The invention discloses a pulmonary tuberculosis patient respiration assisting device based on the motion state, and belongs to the field of medical instruments, the pulmonary tuberculosis patient respiration assisting device comprises a respiration assisting device, a binding assembly is arranged on one side of the respiration assisting device, the binding assembly is used for being bound to the waist of a user, and a self-displacement assembly is arranged on a mounting frame included in the binding assembly; a sliding frame included in the self-displacement assembly is connected with a breathing assist device through a connecting assembly. The whole device is stably bound to the waist of a user through the binding assembly, a fixed installation foundation is provided for the breathing assist device, it is ensured that the device is not prone to falling off or shifting in the moving process of the user, then the self-displacement assembly plays a role to drive the breathing assist device installed on the sliding frame to move, and therefore the breathing assist device can be stably fixed. Finally, the breathing assist device is accurately adjusted to the position of the rear waist side of the user, the position can better adapt to the breathing rhythm of the user, and interference to daily activities is reduced.

The invention discloses a traditional Chinese medicine for treating cold, cough and lung diseases and a preparation method thereof, and belongs to the technical field of traditional Chinese medicines. The traditional Chinese medicine is prepared by compatibility of various common traditional Chinese medicinal materials such as honeysuckle, wild chrysanthemum flower, mulberry leaf, fritillary bulb and dogwood according to specific parts by weight, and the core treatment principle is clearing away heat and toxic materials, ventilating lung and relieving cough, resolving phlegm and relieving asthma, reinforcing qi and nourishing yin, tonifying spleen and kidney, promoting blood circulation to remove blood stasis and taking both symptoms and root causes into consideration. The preparation method comprises the following five steps: screening and treating raw materials, mixing ingredients, decocting and extracting, filtering and extracting juice, and combining and sub-packaging. The preparation method is simple and convenient in process, easy to operate, free of complex equipment, capable of effectively retaining effective components of the medicinal materials and low in cost. The traditional Chinese medicine is suitable for preventing and treating various lung diseases such as cold, cough, asthma, bronchitis, pulmonary nodule, emphysema, pneumonia and pulmonary tuberculosis, is definite in curative effect and small in side effect, and has remarkable clinical application value and popularization prospect.

A novel strain, Lactobacillus paracasei ATG-E1 (Accession No. KCTC 14245BP), and a composition containing the same for preventing or treating respiratory diseases which are caused by fine particulate matter. The Lactobacillus paracasei ATG-E1 strain reduces the number of immune cells bronchoalveolar and lung tissues and inhibits the expression of inflammatory cytokines such as interleukin-17A (IL-17A), tumor necrosis factor-α (TNF-α), macrophage inflammatory protein 2 (MIP2), C-X-C motif chemokine ligand 1 (CXCL-1), macrophage inflammatory protein-α (MIP-1α) or interleukin-6 (IL-6). Therefore, the Lactobacillus paracasei ATG-E1 strain can be used for a treatment or health functional food for various respiratory diseases such as acute and chronic bronchitis, catarrhal bronchitis, bronchitis obliterans, inflammatory bronchitis, bronchial asthma, atopic asthma, non-atopic asthma, atopic IgE-mediated asthma, allergic asthma, non-allergic asthma, chronic bronchoconstriction, acute bronchoconstriction, chronic obstructive pulmonary disease, bronchial adenoma, pulmonary tuberculosis, emphysema, lung abscess, pulmonary fibrosis, lung cancer, airway cancer, bronchoalveolar cancer, and bronchial cancer.

The application belongs to the technical field of biology and particularly relates to screening and application of a urine metabolic marker of tuberculosis. The urine metabolic marker comprises 19 kinds of metabolites. After verification, it is proved that the level change of the 19 kinds of metabolites is indeed related to pulmonary tuberculosis. When the AUC value of one or more of the markers is greater than 0.8, it indicates that the patient is a pulmonary tuberculosis patient. The application further provides a screening method of a pulmonary tuberculosis metabolic marker, a detection kit for diagnosing pulmonary tuberculosis and a model for diagnosing pulmonary tuberculosis. The 19 kinds of urine metabolic markers screened by the application have high accuracy in pulmonary tuberculosis detection and are of great significance for the diagnosis of pulmonary tuberculosis.

The invention discloses a shared feature map-based pulmonary tuberculosis focus detection method and system, and a storage medium. The method comprises the steps of obtaining a CT image sample; improvement is carried out based on a YOLO neural network model, and a PA module is adopted in a backbone network for feature extraction of tiny pulmonary tuberculosis in a CT image; a shared feature map CAB and a C2f-RFCBAMCConv module are added into the neck network, and a plurality of pieces of feature information are obtained by using the shared feature map, so that final pulmonary tuberculosis focus features are obtained; and a mode of combining NWD and WIou is added to improve a weighting mechanism so as to adjust the influence degree of different samples. The trained optimization network model is used for carrying out focus detection on a CT image sample, and in order to meet the high requirement for detecting tiny tuberculosis and multi-feature information targets, the detection sensitivity and positioning precision of the tuberculosis focus are remarkably improved, the misdiagnosis rate is effectively reduced, and effective support is provided for clinical diagnosis.

The application discloses a kind of drug combination and its application.The application provides a kind of drug combination or pharmaceutical composition, the active component of the drug combination or pharmaceutical composition includes material A, material B, material C and material D;Material A is compound I or its pharmaceutically acceptable salt;Material B is isoniazid (H) or its pharmaceutically acceptable salt;Material C is rifampicin (R) or its pharmaceutically acceptable salt;Material D is pyrazinamide (Z) or its pharmaceutically acceptable salt.The drug combination or pharmaceutical composition of the application can reduce alveolar inflammation, pulmonary tuberculosis inflammatory lung injury and pulmonary tuberculosis secondary fibrosis, while increasing the penetration of anti-tuberculosis drugs in the lesion site, to accelerate the treatment response, reduce the risk of recurrence.