521 results about "Drugs sensitivity" patented technology

The invention belongs to the technical field of medicines, and particularly relates to application of sesquiterpenoids and / or pharmaceutically acceptable salts thereof in preparation of medicines for treating or preventing infectious diseases caused by staphylococcus aureus. The invention relates to the field of medicines, in particular to application in preparation of medicines for treating or preventing infectious diseases caused by drug-sensitive staphylococcus aureus or drug-resistant staphylococcus aureus. The structure of the sesquiterpenoids is shown as a formula I in the specification. The sesquiterpenoids applied in the invention are derived from rattan chrysanthemum, the source is wide, and the production cost is low. The sesquiterpenoids shown in the formula I have a very good activity inhibition effect on staphylococcus aureus for the first time, and the sesquiterpenoids have a very good activity inhibition effect on staphylococcus aureus with drug sensitivity, drug medium or drug resistance.

The invention discloses a single-cell drug causal gene discovery method and system based on adversarial transfer learning, and relates to the technical field of drug sensitivity prediction.The method comprises the steps that a source domain data set and a target domain data set are obtained, an adversarial transfer model is constructed, and the adversarial transfer model is trained and optimized according to the source domain data set and the target domain data set; obtaining a single-cell drug sensitivity prediction result; modeling single-cell gene expression data, combining a k-adjacent matching algorithm and a single-cell drug sensitivity prediction result to obtain an anti-fact value, and further calculating a hybrid effect and a difference effect; and calculating the causal effect based on the hybrid effect and the difference effect, and screening out the causal gene of the single-cell drug. On one hand, the key problem of lack of drug sensitivity labeling data in single cell analysis is solved, on the other hand, the causal gene is successfully found, and an important clue is provided for deep understanding of a drug action mechanism.

The invention relates to the technical field of knowledge maps, in particular to a multi-task hepatitis B drug screening method and system based on knowledge map assistance, and the method comprises the following steps: based on hepatitis B virus genotype sequence data recorded along with time, patient drug use records and drug sensitivity. According to the method, virus genotype sequence data, patient drug use records, drug sensitivity and other multi-dimensional dynamic time sequence data are integrated, interaction events between entities are marked through timestamps, and virus variation tracks, drug curative effect changes and other dynamic characteristics are embedded into atlas node attributes; and the knowledge representation can reflect the time dependence in a real scene. A hyperedge connection multi-entity set is defined based on biological pathway annotation information and protein interaction data, the limitation that only a binary relation is supported in a traditional knowledge graph is expanded, a drug combination and multi-target synergistic effect mechanism is explicitly modeled, and misjudgment of the combination effect caused by simplification of the interaction relation is avoided.

The invention provides a drug screening and curative effect evaluation method and system based on a glioma organ-like model, and relates to the technical field of biological material analys.The method includes the steps that a three-dimensional fluorescence image of the glioma organ-like model is collected, signal intensity analysis and compensation processing are carried out, a growth feature mapping model is constructed, and the three-dimensional fluorescence image of the glioma organ-like model is obtained; and phenotypic characteristics for representing the glioma organoid heterogeneity are obtained. Inputting the phenotypic characteristics into a drug response evaluation network, and calculating a drug sensitivity score; the drug molecular feature library and the drug collaborative analysis model are combined, the synergistic interaction index of different drug combinations is calculated, the optimal drug combination is screened, the drug administration dosage and the drug administration time sequence are determined in combination with phenotypic features, and finally an individualized treatment scheme is generated. According to the invention, the curative effect of the drug on the glioma organs can be effectively evaluated, and the optimal drug combination and administration scheme can be screened out, so that the accuracy and effectiveness of glioma treatment are improved.

The invention discloses a spatio-temporal trajectory fusion-based closed-loop monitoring method and system for drug-resistant bacteria in a hospital. The method comprises the following steps: collecting and processing strain drug sensitive test results and patient clinical index data in a hospital information system, and screening out key features through an improved genetic algorithm; wherein the clinical index data of the patient comprises original spatio-temporal trajectory data and nursing operation records; analyzing spatio-temporal trajectory data of the patient according to the key features, combining with drug sensitivity spectrum consistency, and constructing a propagation network model by using a graph convolutional network to identify infection source nodes and corresponding topological influence; and generating an early warning signal based on the topological influence of the infection source node, and continuously optimizing prevention and control measures by dynamically adjusting model parameters and a real-time feedback mechanism. By implementing the method provided by the invention, accurate analysis and real-time prevention and control of a drug-resistant bacterium transmission link can be realized.

The invention discloses a method for constructing a tumor organoid by using stromal stem cells, which improves the survival rate and success rate of the tumor organoid by improving the introduction of the stromal stem cells in the construction of the human tumor organoid, especially for tumor samples which fail to be cultured by the currently disclosed method. The organoid constructed by the invention can be used for analyzing biological characteristics of human tumor organoid; and a clinical tumor drug sensitivity prediction platform is established by taking a tumor organ as a model tool, so that the effectiveness of a treatment scheme is evaluated more systematically and comprehensively, and support is provided for screening an optimal individualized precise treatment scheme for a tumor patient. In addition, on the platform, screening, verification and mining can be carried out on tumor batch chemotherapy drugs or other potential anti-cancer small molecule drugs in clinical application.

The invention discloses a circular RNA drug sensitivity correlation identification method based on integrated multi-instance learning. The circular RNA drug sensitivity correlation identification method comprises the following steps: collecting experimental verification circular RNA and drug sensitivity correlation data; establishing a feature representation model based on a heterogeneous network and vertexes; embedding a heterogeneous graph node into the model, and extracting deep feature representation of the node; designing a meta-path instance embedding projector, and generating a plurality of meta-path instances; a circular RNA and drug sensitivity association predictor is constructed and completed; constructing an integrated heterogeneous graph network deep learning model; and outputting the meta-path instance of the circular RNA and drug pair and the attention coefficient, and carrying out interpretable analysis. According to the method, integrated learning and a deep learning model are combined, so that the reliability of the model is improved; according to the invention, interpretable analysis is carried out by utilizing the meta-path and the attention coefficient, the potential action mechanism of the circular RNA associated with the drug sensitivity can be explained, and the guidance of medical research is facilitated.

The invention relates to the technical field of drug sensitivity prediction, and discloses a platinum drug sensitivity prediction model construction method and application thereof, and the method comprises the steps: constructing a training set; according to the training set, a plurality of candidate models are obtained through training through a plurality of machine learning algorithms; and screening the plurality of candidate models according to a preset model index to obtain at least two platinum drug sensitivity prediction models. Based on the cancer species data of the target group, the applied platinum drug type data, the immune cell infiltration level data of the tumor sample tissue and the PTC data, a plurality of high-precision platinum drug sensitivity prediction models are obtained through training by utilizing a machine learning algorithm, and the method can be used for accurately predicting the platinum drug sensitivity in combination with a hard voting method. The method is high in model prediction accuracy, suitable for people, more targeted, simple and clear in prediction result, short in service cycle and low in cost.

According to the device and the reagent for extracting the pathogen, the pathogen product separated from the uncultured sample extracted by using the device and the reagent can be subjected to a drug sensitivity test after simple culture and bacteria collection treatment, so that the culture process is shortened compared with pathogen diagnosis and identification of a clinical infected sample, and a drug sensitivity result can be obtained more quickly; and a more timely diagnosis and treatment reference is provided for clinicians.

The invention discloses application of tannic acid combined triazole medicines in preparation of medicines for treating diseases caused by fungi, and belongs to the technical field of antifungal. A series of drug sensitive tests show that tannic acid lower than 3 [mu] g / ml has no influence on the activity of human vaginal epithelial cells, and after the tannic acid is combined with fluconazole, the tannic acid can significantly inhibit the growth activity of drug-resistant candida (including drug-resistant candida albicans and drug-resistant candida glabrata). In view of the natural property, wide source and low toxicity of tannic acid, the invention provides a new antifungal treatment strategy, and the tannic acid has important clinical application value for treating infection (such as vaginal candidiasis) caused by drug-resistant fungi, especially under the condition that the existing antifungal drugs are poor in effect. Therefore, the tannic acid combined triazole medicine can be used for preparing the medicine for treating diseases caused by fungi.

The invention belongs to the technical field of intelligent medical treatment, and provides a drug recommendation method and system based on feature fusion and graph construction, and the method comprises the steps: carrying out the local feature extraction and global feature extraction of a diagnosis sequence, an operation sequence and a drug sequence, fusing the local feature extraction and the global feature extraction, and carrying out the global feature extraction; obtaining a diagnostic representation, a surgical representation and a drug representation; by fusing local feature extraction and global feature extraction, the health condition of the patient can be represented more comprehensively; according to the electronic health record data set, establishing an electronic health record graph, a drug interaction graph and a drug sensitivity graph, obtaining a drug representation by using the established electronic health record graph, drug interaction graph and drug sensitivity graph and a preset graph attention network, and fusing the drug representation with the drug representation obtained by feature extraction to obtain a drug representation; weight contributions of different medical events in recommendation decisions can be displayed, doctors are helped to understand logic behind recommendation, and the model has remarkable result interpretation ability.

The invention discloses a lung squamous cell carcinoma drug discovery method and system based on disturbance response scanning analysis, and belongs to the field of biological medicines.The method comprises the steps that lung squamous cell carcinoma protein expression profile data is downloaded and preprocessed, a robust network module is constructed and multi-scale analysis and evaluation are conducted, and drug genomic data are obtained based on drug genomic data. Predicting the drug sensitivity of the patient by using machine learning, and generating a DRN through difference analysis of the drug sensitivity; predicting drug-target protein binding affinity in combination with deep learning, quantifying node sensitivity through a PRS technology, calculating a DPI disturbance score and ranking drug priorities; high-ranking drugs are screened through comprehensive sensitivity analysis and literature investigation; a drug for a lung squamous carcinoma cell line is screened out through a pre-experiment and is relocated. According to the framework developed by the invention, by integrating proteomics, pharmacogenomics and dynamic network analysis, systematic analysis is provided for relocation drug identification of lung squamous cell carcinoma, and good news is brought to treatment of lung squamous cell carcinoma.

An ensemble of predictive models that elucidate how cancer mutations impact the response to common replication stress-inducing (RSi) agents. The models implement recent advances in deep learning to facilitate multi-drug prediction and mechanistic interpretation. Initial studies in tumor cells identify 41 molecular assemblies that integrate alterations in hundreds of genes for accurate drug response prediction. These cover roles in transcription, repair, cell-cycle checkpoints, and growth signaling, of which 30 are shown by loss-of-function genetic screens to regulate drug sensitivity or replication restart. The model translates to cisplatin-treated cervical cancer patients, highlighting an RTK (receptor tyrosine kinase)-JAK-STAT assembly governing resistance. This invention defines a compendium of mechanisms by which mutations affect therapeutic responses, with implications for precision medicine.

The invention provides a glycolipid metabolism related gene-based endometrial cancer prognosis prediction model construction method, which comprises the following steps of 1, acquiring data containing gene expression and clinical information, and preprocessing the data; 2, differential expression and prognosis gene screening; 3, constructing a prognosis model; 4, analyzing model gene enrichment; 5, evaluating the immunocompetence of the two GLRG related dangerous groups; and step 6, statistical analysis. According to the technical scheme, more accurate and reliable prognosis evaluation is provided for endometrial cancer by comprehensively analyzing multi-dimensional information such as gene expression, immune characteristics, mutation characteristics and drug sensitivity.

The invention provides a small cell lung cancer organoid culture medium and an application method thereof. The small cell lung cancer organoid culture medium is prepared from the following components: 10 to 500 [mu] g / mL of R-spondin-1, 10 to 500 [mu] g / mL of Noggin, 0.01 to 0.2 mg / mL of EGF (Epidermal Growth Factor), 0.05 to 0.2 [mu] g / mL of FGFb (Fibroblast Growth Factor b), 0.1 to 10 [mu] g / mL of FGF10, 0.02 to 0.5 mM of A83-01, 0.5 to 20 mM of Y-27632, 1 to 500 mM of SB202190, 0.1 to 200 mM of Nicotinamide, 0.1 to 500 mM of Insulin, 0.1 to 50 nM of protodioscin, 1 to 100 [mu] All the components are dissolved in a DMEM / F12 culture solution. The small cell lung cancer organoid culture medium provided by the invention is suitable for small cell lung cancer tissue organoid culture after component concentration optimization design, good in stability, high in growth speed and remarkably improved in passage frequency, and can still maintain the same tissue structure and genome characteristics of primary tissues after multiple passages; and an ideal model can be provided for research on the occurrence and development mechanism of small cell lung cancer, detection and research on drug sensitivity and the like.

The invention relates to the technical field of precision medical treatment, in particular to an interpretation system and method for tumor precision medication and a storage medium, and the system comprises a data collection module which is used for collecting gene data, clinical data and drug response data of tumor patients from a plurality of data sources; the data standardization processing module is used for carrying out formatting and standardization processing on the data, so that the data formats are consistent, and effective processing and analysis can be carried out in subsequent modules; the gene data analysis module is used for performing variation analysis according to the gene data of the patient, identifying a gene mutation site and generating a drug sensitivity analysis result; and the drug screening module is used for screening drugs matched with the gene variation sites of the patient from a drug database according to the gene data analysis result, and providing drug dosage recommendation. By adopting the technical scheme of combining a quantum optimization technology and gene data analysis, the technical effect of remarkably improving the drug screening and dose recommendation precision is achieved.

The invention discloses a dual-channel fusion cancer drug response prediction method, and belongs to the technical field of biological information. The method aims at solving the problems that an existing prediction method is only limited to intra-modal feature extraction, and a complex nonlinear cooperative relation between a drug and cancer cells is difficult to capture. Multi-omics data, drug molecular structure information and cancer cell line drug reaction data are integrated from databases such as CCLE, GDSC and PubChem, and unified input features are formed through standardization and feature construction. Introducing a hierarchical double-attention conversion network into the first channel to carry out characterization learning on the multi-modal features of the drug and the cell line, and constructing a high-order attention neighbor interaction graph convolutional network in the second channel to capture graph structure information of cancer drug response. And carrying out adaptive weighting on output results of the two channels by using a PPO-based fusion module so as to realize a dynamic optimal decision. And finally, generating a drug sensitivity prediction result of the cancer cell line through a classification predictor.

The invention discloses full-automatic drug sensitivity equipment, and relates to the field of drug sensitivity equipment, the full-automatic drug sensitivity equipment comprises a shell, an incubation device, a sample adding device and a reading device, the incubation device, the sample adding device and the reading device are all arranged in the shell, the reading device is arranged on the first side of the incubation device, and the sample adding device is arranged on the second side of the incubation device. The sample adding device is arranged on the second side of the incubation device, and the first side is adjacent to the second side; the incubation device is used for placing a drug sensitive plate and can rotate along a rotating shaft of the incubation device, and bar code information is arranged on the drug sensitive plate; the reading device is used for reading bar code information of the drug sensitive plate and reciprocating along the height direction of the incubation device and a first direction, and the first direction is intersected with the height direction of the incubation device; the sample adding device is used for reciprocating along a second direction, the second direction is intersected with the first direction, and the second direction is intersected with the height direction of the incubation device. The device has the advantages that the occupied space is reduced, and the operation difficulty is reduced.

The invention relates to a kit for detecting the sensitivity of kidney cancer to a targeted drug, which comprises an RNA (Ribonucleic Acid) nucleic acid probe, and the sequence of the RNA nucleic acid probe is shown as SEQ ID NO: 4. The invention also provides a construction method of the specific annular RNA probe. The construction method comprises the following steps: S1, designing and constructing a recombinant plasmid containing a target gene; s2, constructing a tricRNA (Ribonucleic Acid) vector containing a reverse complementary sequence crossing a backspace junction site and a Broccoli sequence, wherein the reverse complementary sequence crossing the backspace junction site and the Broccoli sequence are subjected to reverse complementary sequence crossing the backspace junction site; s3, carrying out vector transfection on the HEK293T cell to express a tricRNA probe, and extracting the tricRNA probe; and S4, constructing a tricAGAP1 Probe-Broccoli probe. The method comprises the following steps of: constructing a tricAGAP1 probe; the invention provides application of a probe in preparation of a cell imaging probe and in preparation of a kit for detecting sensitivity of kidney cancer to a targeted drug. The invention not only provides a new method and technology for guiding the medication of sunitinib for patients with advanced kidney cancer, but also provides a technical example for the development of other circRNA detection technologies.

The invention provides a single-cell drug sensitivity prediction algorithm based on multi-modal reinforcement learning, and aims to solve the problem that the accuracy of a traditional drug sensitivity prediction method is limited. According to the system, multi-modal bioinformatics data including cell line gene expression data, proteome data, metabolome data, drug reaction data and single-cell RNA sequencing data are integrated, and feature extraction, fusion and selection are performed by using a deep reinforcement learning technology, so that high-precision prediction of single-cell drug sensitivity is realized. The algorithm mainly comprises five core modules: a multi-modal data integration and preprocessing module, a deep reinforcement learning feature extraction and fusion module, a dynamic feature selection and importance evaluation module, a cross-domain migration and adaptive learning module and a deep reinforcement learning prediction model and performance optimization module. The modules work cooperatively, potential features can be learned automatically from multi-modal data, a feature fusion strategy can be optimized, a key feature subset can be selected dynamically, and a model structure and parameters can be adjusted adaptively according to features of target domain data, so that prediction accuracy and stability are improved remarkably. Experimental results show that the algorithm shows significant advantages in multiple evaluation indexes, and compared with a traditional prediction method, the algorithm has higher prediction accuracy and wider applicability. Therefore, the invention provides a novel and efficient drug sensitivity prediction tool for the fields of personalized medical treatment and drug research and development, and has important practical application value.

The invention discloses a high-precision identification method for a culture medium drug sensitive paper sheet based on a mixed model, and the method comprises the steps: S1, image collection and preprocessing: carrying out the standardization processing of a collected culture medium image, including noise removal, illumination correction and contrast enhancement; s2, semantic segmentation model processing: carrying out pixel-level classification on the preprocessed image by adopting a deep learning network architecture, accurately positioning the positions and boundaries of all drug sensitive paper sheets in a culture medium, and generating a segmentation mask; and S3, classification model processing: extracting a paper sheet area based on a result of the step S2, and identifying a drug code and concentration on each paper sheet by using a CNN classification model. And S4, hybrid model fusion: intelligently fusing the data in the steps S2 and S3, and outputting a final drug sensitive paper identification result through confidence weighting and result verification. According to the method, semantic segmentation and classification models are fused, the difficulties of low recognition precision, poor environmental adaptability and the like of a traditional drug sensitive test are overcome, and high-precision automatic recognition of the drug sensitive paper sheets is realized.

Methods for facilitating drug discovery in various models and proposing single cell-type-specific drug candidates, and predicting cellular drug sensitivities. Predictions comprise obtaining bulk RNA sequence data, wherein bulk RNA seq data is pan-cancer cell line transcriptomic data from one or more sources; obtaining single cell RNA sequence data (scRNA-seq) from one or more sources; integrating the bulk RNA sequence data and scRNA-seq data; capturing similarities between the bulk data and the scRNA-seq data based on a shared drug response relative gene and using canonical correlation analysis; extracting one or more drug-response relevant features and selecting a pharmacogenomic subspace for accurately predicting a single-cell drug response.

The invention discloses application of a nitroimidazole derivative and a combined drug in preparation of a drug for resisting drug-sensitive tuberculosis, and belongs to the technical field of biological medicine. The invention provides application of a nitroimidazole derivative and a combined drug in preparation of a drug for resisting drug-sensitive tuberculosis in order to solve the clinical research blank of the nitroimidazole derivative in prevention and treatment of drug-sensitive tuberculosis. Animal experiments, I-stage clinic, II-stage clinic and medication dosage and mode optimization prove that the nitroimidazole derivative has no obvious drug accumulation, no obvious reproductive toxicity, no obvious cardiotoxicity and no QT interval extension, and compared with existing drugs such as deramanib, the nitroimidazole derivative is shorter in treatment time, more remarkable in curative effect, higher in safety and suitable for clinical application. Especially in lung iconography, voids of a patient are obviously reduced or focuses are obviously absorbed, so that the compound is expected to be used as a new medicine for preventing and treating drug-sensitive pulmonary tuberculosis.

The invention provides a method and system for analyzing lung cancer brain metastasis organoid cell subpopulation and functional characteristics, and the method comprises a series of steps: data quality control, batch effect removal, cell grouping, copy number variation inference, quasi-time sequence analysis, function enrichment analysis, and drug sensitivity analysis before and after radiotherapy. Single cell transcriptome sequencing data obtained after organoid culture is subjected to personalized analysis, heterogeneity of tumor cells and sensitivity of the tumor cells to drugs after radiotherapy are described, and the characteristics of all cell subgroups in the lung cancer brain metastasis organoid and the effect of the cell subgroups in the tumor microenvironment can be accurately analyzed; powerful support is provided for mechanism research and precise medicine of lung cancer brain metastasis.

The invention, which relates to the technical field of cloud data diagnosis, discloses a cloud data sharing and diagnosis platform for microbial inspection, comprising a data preprocessing module, a federal modeling module, a multi-modal fusion module, a diagnosis scoring module, a block chain evidence storage module, a trusted sharing module and a dynamic optimization module. The data preprocessing module is used for performing format unification and feature extraction on original microbiological inspection data from a plurality of medical institutions by adopting a standardized coding method, the original microbiological inspection data comprises microscopic images, mass spectrum peak maps, drug sensitivity results and clinical information, and a structured inspection feature set is output; and the federated modeling module is used for carrying out cooperative training on locally deployed initial diagnosis models of all mechanisms by adopting a federated learning method based on differential privacy protection, locally calculating gradient information by utilizing the structured test feature set, and uploading gradient parameters subjected to noise disturbance to the central server for aggregation updating.

The invention provides an organoid drug sensitivity automatic detection method and system for multi-sample parallel processing, and the method comprises the steps: obtaining the optical path length data of each hole site of a cell culture perforated plate, and determining the initial focal plane position of each hole site; matching the deformed feature point coordinates with the initial three-dimensional structure data of the similar organs, and calculating the focal plane offset caused by the deformation of the similar organs through least square fitting; adjusting the focal length of the optical imaging system according to the corrected focal plane position by adopting a fast focusing algorithm, and re-collecting an organ-like image; using a layer-by-layer stacked three-dimensional reconstruction algorithm to perform registration splicing on the recollected image sequence to obtain complete three-dimensional structure data after the organ-like body is deformed; and according to the complete three-dimensional structure data, extracting the volume, surface area and cell distribution density characteristics of the organoid, and generating a morphological characteristic report of the drug sensitivity experiment.

The invention discloses an intelligent turbidimetric instrument, belongs to the field of medical instruments, and particularly relates to a bacterium liquid preparation process of a drug sensitivity experiment. The method is mainly used for bacterial turbidity detection, drug sensitivity experiments and other application scenes in the field of microbiological detection. According to the present invention, the phenomena of repeated sampling liquid preparation and concentration determination by using a turbidimetric instrument exist in the manual preparation of the bacterial liquid, the step is tedious in the drug sensitivity experiment, and in order to reduce the manual operation of the step, two specific technical implementation schemes are introduced, and the instrument selection can be performed according to the specific conditions; according to the technical scheme, the intelligent turbidimeter is used for detecting the concentration of a sample, whether a sample bacterial solution meets the required concentration in a drug sensitivity experiment or not is automatically judged, if not, the volume of a required normal saline diluent is calculated by extracting a micro control unit through a peristaltic pump, and the volume of the normal saline diluent is added into a sample test tube; after the sample is diluted, starting the vibration motor to uniformly shake the sample, and re-measuring the concentration by the turbidimetric instrument to obtain the bacterial liquid for the subsequent drug sensitivity experiment. According to the other technical scheme, on the basis of automatically judging the concentration of the bacterial liquid, the bacterial liquid for the drug sensitivity experiment is prepared in a high-flux liquid preparation mode.

The invention discloses a high-throughput bacterial drug sensitivity test platform based on computer vision. The platform integrates the functions of automatic separation and identification of bacteria, drug sensitivity test and result analysis. Separation and identification of bacteria are realized through a microfluidic technology, and growth conditions of the bacteria under the action of different antibiotics are monitored in real time in combination with a computer vision technology, so that the antibiotic sensitivity of the bacteria is rapidly judged. The platform has the characteristics of high throughput, automation and intellectualization, can complete detection of a large number of samples in a short time, automatically generates a drug sensitivity test report, and provides powerful support for clinical diagnosis and treatment. By implementing the method, the speed and the accuracy of the bacterial drug sensitivity test are remarkably improved, and the method has a wide application prospect.

The invention discloses a gastrointestinal stromal tumor organ culture solution as well as a use method and application thereof, and belongs to the technical field of gastrointestinal stromal tumor organ culture. The gastrointestinal stromal tumor organs obtained by adopting the culture solution and the culture method provided by the invention can retain histological characteristics and genetic heterogeneity of original tumors of patients, can be used as a medicine sieve model and a disease research model for gastrointestinal stromal tumor treatment, and can promote progressive development of tumor precision medicine; the cultured gastrointestinal stromal tumor organ is high in proliferation speed and cell viability, long-term passage and amplification can be realized, and the gastrointestinal stromal tumor organ with the diameter of more than 100 microns and good activity can be obtained in 8-10 days; the method is also suitable for culturing wild gastrointestinal stromal tumors such as SDH defect type gastrointestinal stromal tumors; the culture medium does not contain components with indefinite components such as serum, is not influenced by different batches, and can maintain relatively high homogenization in the culture process; the cell culture medium can be used for culturing gastrointestinal stromal tumor organs derived from tumor tissues or malignant cell lines of patients, and is suitable for drug sensitivity detection.

The invention relates to a model based on multi-modal cross attention and uncertainty integral gradient and application, and relates to the technical field of bioinformatics. The model is obtained based on multi-modal cross attention and uncertainty integral gradient construction, the problems that in an existing drug sensitivity prediction method, multi-modal data fusion is difficult, the model generalization ability is weak, and interpretability is lacked can be solved, the model captures correlation between modals through a cross attention mechanism, and the prediction accuracy of the drug sensitivity is improved. According to the method, uncertainty estimation is realized by combining Monte Carlo Dropout (MC Dropout), feature importance is analyzed by using an integral gradient algorithm, and finally, unification of high-precision prediction and biological interpretability is realized.