222 results about "Drug action" patented technology

The invention discloses a drug target activation and inhibition relation prediction method based on a depth map neural network, and aims to improve the modeling precision and prediction performance of an activation or inhibition action mechanism between a drug and a target. According to the method, on the basis of a fine-grained graph interaction modeling mechanism, multi-scale structural characteristics of drug molecules and three-dimensional space structural information of protein residue levels are fused, and a heterogeneous interaction graph between drugs and proteins is constructed. The method comprises the following steps: firstly, acquiring a drug-target sample with an activation / inhibition tag through a public database, predicting a protein structure by utilizing AlphaFold2, and constructing a protein residue map and a drug molecular map; multi-scale structure semantic representation is obtained through sub-graph decomposition, atomic-scale feature extraction and graph neural network coding of drug graph features; protein graph node features are combined with context embedding generated by a pre-training language model, DSSP coding, secondary structure spectrum and atomic structure features are constructed, and edge features are designed based on the geometrical relationship between residues. Then, based on constraints such as spatial distance and biochemical similarity, a fine-grained mapping relation between drug atoms and protein residues is established, an interaction graph is constructed, and coding is carried out through a GraphSAGE network; and finally, fusing the interacted multi-source embedding, and completing the prediction of the activation / suppression relationship through a multi-layer perceptron. A cross entropy loss function, an Adam optimizer and hyper-parameter grid search are adopted in model training; in the evaluation stage, five-fold cross validation and an independent test set are adopted, and indexes such as the accuracy rate, the recall rate, the F1 score, the specificity and the Morse correlation coefficient are used for comprehensively evaluating the performance of the model. Experimental results show that compared with an existing method, the method has the advantages that the prediction accuracy and mechanism interpretability are remarkably improved, and the method has good generalization ability and application prospects and is suitable for multiple fields of drug action mechanism research, new drug discovery and the like.

The invention belongs to the technical field of biological medicines, and particularly relates to application of oxibenzophenone in preparation of a medicine for treating pathological cardiac hypertrophy and / or heart failure. Research results in myocardial cells of primary newborn rats show that Exibenzophenone can inhibit pathological hypertrophy of myocardial cells induced by phenylephrine (PE). Animal experiments show that the epoxybenzophenone can improve cardiac dysfunction induced by aortic arch constriction (TAC) and reduce cardiac hypertrophy and cardiac tissue fibrosis. The invention provides a new drug research and development approach and a drug action target for treating pathological cardiac hypertrophy and heart failure, and has very important medicinal value.

The invention discloses a wound healing state evaluation method and system based on artificial intelligence, and belongs to the field of wound healing evaluation. The method comprises the following steps: integrating medical record data and wound physiological parameters, and extracting relevance between a medication history and an infection type to form a multi-dimensional pathological factor; hemoglobin oxygenation, blood flow and metabolic heat are subjected to cross-period correlation analysis to construct dynamic physiological time-space information, pathological factors and the physiological time-space information are fused into a heterogeneous graph, nodes comprise drug intensity quantized values, infection risk levels and microcirculation fluctuation values, and edge weights are determined by the time sequence matching degree of drug doses and blood flow changes; based on the node weight and the edge association relationship, the treatment time and the physiological parameter acquisition period are aligned to generate a spatio-temporal evolution matrix, the spatial distribution of the infection risk level and the metabolic heat value characteristics are combined, the dynamic association of the drug action intensity and the microcirculation fluctuation is analyzed, and wound healing evaluation information is generated. The accuracy of wound healing state evaluation can be improved.

The invention relates to double-encapsulated lipid nanoparticles entering brain through nose and a preparation method and application thereof, the double-encapsulated lipid nanoparticles comprise lipid nanoparticles, drugs in the lipid nanoparticles and rana odorata lectin coupled to the surfaces of the lipid nanoparticles, and the drugs comprise hydrophilic drugs and hydrophobic drugs. According to the invention, the rana odorata lectin and the lipid nanoparticles are coupled, and the rana odorata lectin can be specifically combined with L-fucose residues expressed by olfactory epithelial cells, so that the adhesion of the lipid nanoparticles to olfactory epithelium during nasal administration is enhanced, and the brain entering efficiency of the lipid nanoparticles is finally improved. After entering the brain, the lipid nanoparticles respond to diseased regions with high active oxygen content so as to release drugs. In addition, double-drug encapsulation can target different pathogenesis of central nervous system diseases, and the limitation that a traditional treatment drug is single in action target is improved.

The invention provides a nanocrystalline composite oral ulcer film agent as well as a preparation method and application thereof. The nanocrystalline composite oral ulcer film agent comprises a waterproof backing layer and a drug-loading adhesion layer coated on the waterproof backing layer, the drug-loaded adhesion layer comprises a macromolecular hydrogen bond compound formed by a weakly acidic polymer and a nonionic proton receptor polymer, and a nanocrystalline drug. Compared with the prior art, the preparation method has the advantages that the polyacrylic acid polymer and the nonionic cellulose polymer interact to form IPC serving as an adhesion layer film-forming material, the nanocrystals are loaded into the adhesion layer film-forming material, the waterproof backing layer is coated with the adhesion layer film-forming material, and the oral ulcer film agent is prepared after drying and cutting. The obtained film agent has good comfort and adhesive power, shows good drug stability and permeation promotion performance, can be used as a carrier of an oral administration route, prolongs the drug action time and improves the curative effect. The preparation method of the film agent is simple and easy for industrial mass production.

The invention relates to the technical field of biological medicine, in particular to anti-acne hydrogel and a preparation method thereof. The anti-acne hydrogel is prepared from metal-polyphenol network nanoparticles and a hydrogel base material, the metal-polyphenol network nanoparticles comprise a first polyphenol compound, a second polyphenol compound and metal ions, the first polyphenol compound is a tannic acid compound, and the second polyphenol compound is a flavonoid compound; the hydrogel base material comprises a hydrophilic high-molecular polymer and a cationic polymer. The metal-polyphenol network nano-particles are creatively utilized, the anti-inflammatory and anti-oxidation activity of flavonoid compounds, the convergence and protein binding characteristic of tannin and the antibacterial activity of metal ions are integrated at the same time, and the vicious circle of acne can be intervened at the same time from multiple links of antibacterial, anti-inflammatory, anti-oxidation and keratinization adjustment; and the defect of single action target spot of the existing medicine is overcome.

The invention provides an artificial intelligence customer service management method and system for medical services, and relates to the technical field of artificial intelligence, and the method comprises the steps: processing consultation information through a medical term mapping network and a medical image unified recognition framework; fusing the standardized term information, the drug feature information and the prescription feature information to generate standardized consultation data; carrying out risk identification and grading processing based on a drug action relation network; outputting corresponding response information according to the grading result; and updating the identification framework parameters through user feedback. The intelligent medical consultation service is realized, and the risk identification accuracy and the response efficiency are improved.

The invention relates to a heart rehabilitation management method based on rehabilitation requirements. The method comprises the following steps: acquiring oxygen saturation time sequence data of a patient through wearable equipment, performing biological rhythm feature extraction on the data to construct a fluctuation model so as to capture a physiological cycle rule, and constructing a drug action factor based on a drug taking record to perform drug action calibration on original data so as to eliminate exogenous interference. Candidate time windows with consistent physiological statuses are screened through sliding window stability analysis; dynamic baseline calculation is performed on the candidate window based on long-term monitoring data to establish an individualized reference standard, when the dynamic baseline reaches the standard, the biological rhythm parameters and the drug action factors are fused to score and screen the time window, and finally, an individualized rehabilitation training scheme is generated in combination with physiological feature matching of a patient. Dynamic synergy of drug metabolism, physiological rhythm and historical baselines is achieved, the problem of physiological load imbalance caused by a static training plan is solved, and the rehabilitation efficiency is improved while the heart compensation risk is remarkably reduced.

The invention belongs to the technical field of computer biology, and relates to a lightweight drug-target interaction prediction method based on double pre-training language models. Downloading a protein information data set of required drug compounds and drug action targets from a public database; sending the protein sequence into an encoder to obtain a generated vector; the method comprises the following steps: labeling a compound sequence of a drug by using a tokenizer, pre-defining a corresponding relation between a compound sequence label and an integer, then converting into an integer value, and inputting the integer value into an encoder to generate a vector; splicing the class tokens to obtain a drug target pair, and feeding the drug target pair into an interaction head; and calculating an interaction probability between the two. According to the method, the drug pre-training language model and the protein pre-training language model are introduced at the same time, deep semantic coding is performed on the SMILES sequence and the amino acid sequence, molecular structure characteristics and protein sequence characteristics can be comprehensively captured, and the expression ability is remarkably enhanced.

The invention discloses a microfluidic multi-organ tumor chip targeted drug test AIGC system, and relates to the technical field of medical drug test information processing. The microfluidic multi-organ tumor chip targeted drug test AIGC system comprises a multi-omics data acquisition module for acquiring and preprocessing disturbance response data and dynamic expression data; the disturbance response pre-screening module is used for carrying out strength evaluation on drug action strength and reconstructing channel mapping; the characteristic space reconstruction module is used for performing action analysis on the whole medicine action state and dynamically adjusting a liquid medicine blending rule; the drug effect prediction module is used for predicting and evaluating the response degree of the organoid under the intervention of the specific drug and deducing the individualized drug effect intensity; and the feature augmentation self-closed loop module constructs a negative supervision signal and dynamically corrects a disturbance perception and feature embedding path. The problems that current multi-omics data is high in dimension, large in redundancy and large in noise, key features are difficult to extract under limited samples, and model generalization is difficult to maintain are solved.

The invention relates to the technical field of traditional Chinese medicine extract preparation, and discloses taste-masking enteric-coated traditional Chinese medicine extract particles which comprise drug-loaded pellets and enteric-coated coatings coating the drug-loaded pellets. The invention relates to a drug-loaded pellet. The coating comprises 20-45 parts of pure powder of the traditional Chinese medicine extract and 5-10 parts of dextrin; according to the taste-masking enteric-coated traditional Chinese medicine extract particle and the preparation method thereof, a stable barrier is formed in gastric juice through a carboxymethyl chitosan-lipidosome copolymer coating, and the degradation rate of effective components is reduced to 1t from 40-60% of a traditional preparation; the pH-responsive coating ensures that the rapid release rate of the intestinal tract is as high as 90%, the bioavailability is improved by 2-3 times, the bitter taste is completely covered by the coating layer, and the coating layer can be mixed with food to feed pets; the prepared sustained-release dosage form prolongs the action time of the medicine, reduces the medication frequency, reduces the medication cost of pet owners, and improves pet welfare.

This invention discloses an AfBPP-based active molecular probe for thiostreptin, its preparation, and its application, belonging to the field of chemical biology. This invention introduces a bifunctional tag integrating a photocrosslinking group (bisacrididine) and a bioorthogonal reactive group (alkynyl group) into the structure of thiostreptin. While retaining the original biological activity of the parent compound, it endows the probe with highly efficient probe function, solving the technical problem of target loss during washing and purification of traditional non-covalent probes. After target labeling, the probe can specifically connect to reporter groups such as fluorescein or biotin through click-chemical reactions, achieving efficient enrichment of drug targets. Combined with mass spectrometry analysis, it enables global identification of potential targets in cells or complex biological samples at the omics level, such as chemical proteomics, providing a powerful molecular tool for in-depth revelation of the potential targets and pharmacological mechanisms of thiostreptin.

The invention discloses a mitochondrial / lysosome dual-targeting viscosity response AIE probe as well as a preparation method and application thereof. According to the probe, tetraphenylethylene (TPE) is used as an AIE light-emitting framework, and two flexible side chains containing hydrophilic units are introduced by utilizing the excellent structural modifiability of the TPE. The tail end of each side chain is modified with a trimethylamine group, quaternary ammonium salt cations derived from the trimethylamine group have localized positive charges, and combination with water molecules can be enhanced through ion-dipole interaction, so that the water solubility and biocompatibility of the molecules are effectively improved. Based on the design, the invention provides the AIE polymer probe which is constructed by taking TPE as a core and taking an amphiphilic flexible chain as a functional unit. The probe can establish a three-dimensional linear relationship among drug action time-regional viscosity-fluorescence lifetime (t-eta-tau) by capturing fluorescence lifetime signals inside and outside cells, so as to realize visual detection and quantitative analysis of average viscosity change of mitochondria and lysosome in the mitochondrial autophagy process.

The invention discloses an application of DL-3-phenyllactic acid in treatment of cardiac mucinoma. Based on metabonomics, it is found that the level of endogenous DL-3-phenyllactic acid in serum of a patient is remarkably reduced, pharmacodynamic verification is conducted by means of a cardiac mucinoma tissue-derived cardiac mucinoma organ model of the patient, it is proved that DL-3-phenyllactic acid can inhibit proliferation of cardiac mucinoma organs in a dose-dependent mode, and the half inhibitory concentration IC50 value of DL-3-phenyllactic acid is 95.52 mu M. The DL-3-phenyllactic acid can inhibit proliferation of cardiac mucinoma organs in a dose-dependent mode. Network pharmacological analysis indicates that the drug action mechanism may involve key targets such as AKT1 and SRC. The invention discloses the therapeutic efficacy of the DL-3-phenyllactic acid on the cardiac mucinoma for the first time, and provides a solid preclinical experimental basis for developing the DL-3-phenyllactic acid into the medicine for treating the cardiac mucinoma. The invention has non-obvious and important clinical application value.

The invention discloses a drug-miRNA interaction prediction method based on a meta-path expert network, belongs to the field of biological information, and relates to a drug-miRNA interaction prediction method. Prediction of the interaction between the drug and the miRNA not only contributes to revealing the drug action mechanism, but also can provide theoretical support for drug target discovery, drug relocation and individualized treatment. A traditional method mainly verifies drug-miRNA association based on biological experiments, but has the problems of high cost and long consumed time. However, the existing calculation method does not fully excavate the relationship between the drug and the miRNA. The invention provides a drug-miRNA interaction prediction method based on a meta-path expert network, and the method comprises the steps: firstly fusing various similarities, and constructing a heterogeneous graph in combination with the known drug-miRNA interaction; a heterogeneous graph is converted into a plurality of isomorphic graphs through meta-paths, semantic information of different meta-path graphs is learned by using a graph convolutional network, and node features with rich semantics are obtained. And finally, fusing node features of different meta path diagrams by using a hybrid expert model to obtain final drug node features and miRNA node features, and performing interaction prediction through MLP.

The invention discloses a precise regulation and control injection device for extraocular muscle toxin, and belongs to the technical field of medical instruments. The precise regulation and control injection device for extraocular muscle toxin comprises a microneedle assembly, a handle, a suction head and an injection regulation and control part, wherein the injection regulation and control part is used for controlling the microneedle assembly to be connected with the suction head. When the precise regulation and control injection device for the extraocular muscle toxin is used, the micro-needle assembly can be in suction connection with the suction head arranged at the tail end of the handle through adjustment of the injection regulation and control part, so that in the process that an operator injects the extraocular muscle toxin to a patient, it is ensured that a micro-needle does not deviate or loosen in the injection process, and the injection efficiency is improved. The precise positioning of an injection part is ensured; and the injection regulation and control part can regulate the action range of the microneedle on the microneedle assembly to be more concentrated, so that the concentration of the action range of the medicine and the accurate delivery of the medicine to the target muscle layer can be ensured, the out-of-control diffusion of the medicine is avoided, and the risk of penetrating through the sclera is reduced.

The invention relates to the technical field of biomedical materials and wound repair, and discloses a preparation method and application of 4 '-hydroxychalcone-loaded nanofiber hydrogel, the preparation method comprises the following steps: mixing 4'-hydroxychalcone and polylactic acid according to a mass ratio of 1: (6-12), and carrying out ultrasonic treatment to obtain 4 '-hydroxychalcone-loaded nanofiber hydrogel; preparing nanofibers containing 4 '-hydroxychalcone by adopting an electrostatic spinning technology; according to the present invention, the 4 '-hydroxychalcone-loaded nanofiber hydrogel has the following characteristics that the drug in the 4'-hydroxychalcone-loaded nanofiber hydrogel can be continuously released, and the cumulative release rate can achieve 70-85% within 72 h so as to maintain the stable drug action concentration on the wound surface; meanwhile, the nanofiber hydrogel has good water retention capacity, the water retention rate of the nanofiber hydrogel can reach 90-95%, a stable wet healing environment can be provided for the wound surface, and the wound surface repairing process is facilitated.

The present disclosure relates to systems and methods for analyzing drug-drug and gene-drug interactions of a user and providing suggestions therefrom. The systems and methods may utilize a database containing a list of genes associated with a plurality of drugs based on clinical relevance of the genes to the effects of the plurality of drugs. The systems and methods may include recommending alternative drugs to a user based on any identified problematic drug-drug or gene-drug interactions.

This invention discloses a method and system for evaluating the effectiveness of cardiac rehabilitation, relating to the field of medical and health technology. The method includes: acquiring physiological monitoring data, individualized parameter data, comorbidity data, multidrug therapy data, pharmacokinetic parameters, and pharmacodynamic interaction parameters of the target patient; generating baseline physiological index data without drug intervention based on the individualized parameter data and comorbidity data through a pre-set individualized cardiac response baseline model; calculating drug concentration based on multidrug therapy data and pharmacokinetic parameters; calculating comprehensive pharmacodynamic data by combining pharmacodynamic interaction parameters; generating predicted physiological data under multidrug intervention; subsequently obtaining purified physiological data through adaptive filtering and frequency domain decomposition; and finally generating a rehabilitation effect score through a pre-trained long short-term memory network model. Its beneficial effects include: distinguishing between drug interference and real cardiac function improvement, and improving the accuracy of rehabilitation assessment for patients undergoing multidrug therapy.

This invention relates to the application of mitochondrial extracts in reversing drug resistance in solid tumors. By combining mitochondrial extracts with chemotherapeutic drugs, the resistance of chemotherapeutic-resistant solid tumors to these drugs can be reversed. Compared to using mitochondrial extracts and chemotherapeutic drugs alone, this method effectively enhances drug efficacy, thereby overcoming chemotherapeutic resistance in solid tumors. This invention, through intervention with mitochondrial extracts in different chemotherapeutic-resistant solid tumor cell lines, has verified that the combined use of mitochondrial transplantation and chemotherapeutic drugs can reverse chemotherapeutic resistance, thereby improving the efficacy of chemotherapy and opening up the possibility of continued treatment for patients with advanced solid tumors who are resistant to chemotherapeutic drugs.

This invention relates to the field of biomedicine, and more particularly to a hair follicle-targeted delivery system and a hair growth and anti-hair loss cosmetic. The hair follicle-targeted delivery system comprises: a lipid nanoparticle core loaded with bioactive ingredients having hair growth and anti-hair loss effects; a hair follicle-targeting molecule capable of specifically recognizing and binding to receptors on the surface of hair follicle cells; and a targeting and anchoring conjugate consisting of a linker covalently linked to the hair follicle-targeting molecule and an anchoring lipid. The targeting and anchoring conjugate is embedded in the lipid bilayer of the lipid nanoparticle core via the anchoring lipid, thereby exposing the hair follicle-targeting molecule on the surface of the lipid nanoparticle core. This application, through its innovative targeting and anchoring conjugate structure, can significantly improve the targeting and enrichment efficiency of the hair growth and anti-hair loss active ingredients on hair follicle tissue, ensuring precise drug action on the lesion site, thereby significantly increasing the number of hair follicles in the alopecia areata modeling area.

The invention discloses multifunctional selenium nanoparticles for treating inflammatory bowel diseases as well as a preparation method and application of the multifunctional selenium nanoparticles, and relates to the technical field of biological medicines. The multifunctional selenium nanoparticle is of a core-shell structure and comprises an inner core and an outer layer coating the inner core, the inner core is selenium nanoparticles; and the outer layer is a polydopamine shell embedded with chitosan oligosaccharide. The multifunctional selenium nanoparticles provided by the invention have good stability and dispersibility, can target intestinal inflammation parts, prolong the in-vivo time of the nanoparticles, and show strong antioxidant and anti-inflammatory activity on cell and animal levels. The whole nanoparticle is based on low toxicity and colon targeting of a natural material, is higher in safety compared with a traditional glucocorticoid / biological preparation, relieves intestinal inflammation by adjusting a redox steady state, prolonging drug action time and the like, and is expected to become a novel colitis treatment strategy with high efficiency, targeting property and safety.

A method for maintaining a gene description dataset associated with a plurality of genes with respect to a given disease, wherein the gene description dataset is a human- comprehensible text-based dataset generated by a Large Language Model (LLM); maintaining a dataset of drug mechanisms of action (MOAs) associated with a plurality of drugs; embedding the gene descriptions and the MOAs into high-dimensional vectors using a text embedding technique to generate gene description vectors and MOA vectors; calculating distances between the gene description vectors and the MOA vectors to generate distance scores; and ranking the drugs for the given disease based on an aggregated distance score between the MOA vectors and the vectors of genes associated with the given disease.

The application of CpG oligonucleotides in drugs for diabetic cardiac autonomic neuropathy: Experiments have confirmed that CpG oligonucleotide 1826 can reduce the activation of satellite glial cells in the stellate ganglion and the expression of inflammatory cytokines, inhibit the expression of purine 2Y12 receptors, and reduce lipid peroxidation damage to alleviate ferroptosis in stellate ganglion neurons, thereby improving cardiac autonomic nerve function. Examples of CpG oligonucleotide 1826 demonstrate that CPG-ODN with anti-inflammatory effects or with anti-damage protective effects on the sympathetic nervous system has preventive and therapeutic effects in diabetic cardiac autonomic neuropathy and sympathetic nerve injury-related diseases.

The application provides a preparation method and application of resveratrol-polylactic acid microspheres, and relates to the technical field of biological agents, quantitative resveratrol is dissolved in methanol as O1 phase; quantitative polylactic acid is dissolved in an oil phase solution (ethyl acetate, dichloromethane, trichloromethane) as O2 phase; ammonium bicarbonate, sodium bicarbonate, sodium chloride and hydrogen peroxide are used as a porogen; O1 phase, O2 phase and the porogen solution are subjected to ultrasonic emulsification to obtain O / O primary emulsion; the O / O primary emulsion is added to a polyvinyl alcohol aqueous solution under stirring to obtain O / O / W multiple emulsion; the obtained O / O / W multiple emulsion is subjected to mechanical stirring to remove the oil phase solution in the O / O / W multiple emulsion, and then is subjected to standing, centrifugation, washing, drying and the like to obtain resveratrol-polylactic acid microspheres with uniform size. The resveratrol-polylactic acid microspheres of the application break through the bottleneck of poor water solubility of resveratrol, can effectively prolong the drug action time, and have good antibacterial effect.

The present disclosure is directed to systems and methods for analyzing drug-drug and gene-drug interactions for a user and providing recommendations accordingly. The systems and methods can utilize a database comprising a list of genes are associated with a plurality of drugs based on clinical relevance of the gene to the actions of the plurality of drugs. The systems and methods can include recommending alternative drugs to users based on any identified problematic drug-drug or gene-drug interactions.

The invention provides a marker for predicting caries occurrence based on saliva metabolites, which is characterized by comprising 4-acetaminobutyric acid, adipic acid, pantetheine and palmitoyl ethanolamide, and the marker is used for preparing a kit for detecting the caries occurrence risk of a subject. The invention provides four markers in saliva metabolites, and early warning of future caries can be realized in a non-invasive and low-cost manner; according to the basis of detection numerical values of the markers, the morbidity state of the constant dentition caries can be reflected, so that accurate prevention and control of the caries can be realized, and finally the purposes of reducing the morbidity of the caries, helping disease pathological typing and researching drug action targets, accurate medication and pathogenesis can be achieved.

The invention discloses an AI intervention analysis method for diabetic renal interstitial fibrosis, and relates to the technical field of biologication.The method comprises the following specific steps of sample collection and multi-omics data acquisition, specifically, kidney tissue samples of diabetic renal interstitial fibrosis patients before and after the diabetic renal interstitial fibrosis patients use glucose kidney health intervention and kidney tissue samples of healthy contrasts are collected; respectively acquiring epigenetic data and single-cell gene expression data, preprocessing, and integrating to construct a comprehensive data set; by combining the AI technology and the high-throughput epigenetic detection technology, the epigenetic modification change on a TGF-beta1 / Smads signal path in the process of intervening diabetic renal interstitial fibrosis by the Sushenkang can be deeply analyzed; a brand-new perspective is provided for understanding the occurrence mechanism of the diabetic renal interstitial fibrosis disease and the drug action mechanism of the TGF-beta1 / Smads pathway, epigenetic data is deeply mined through AI, and the specific mechanism of the TGF-beta1 / Smads pathway related gene expression affected by the TGF-beta1 / Smads pathway through epigenetic regulation is shown.

This invention provides a drug interaction prediction method based on a bidirectional cross-view attention network, belonging to the field of drug interaction prediction technology. The method first constructs a Morgan fingerprint similarity view of the drug, an original DDI view, and a multi-scale diffusion view based on personalized PageRank. Then, a graph convolutional network with shared weights is used to co-encode the multiple views, generating a unified drug embedding representation. Finally, a bidirectional cross-view attention mechanism is used to achieve fine-grained interaction and alignment between the structural and attribute views, and interaction prediction is completed via a multilayer perceptron. This invention effectively alleviates the sparsity problem of the DDI network through multi-scale topology enhancement and achieves complementary enhancement between views using bidirectional cross-view attention, significantly improving the accuracy and generalization ability of drug interaction prediction. It can provide reliable technical support for drug development screening and clinical combined drug safety assessment.

The application discloses a medicine patch with gradient effect on human body, which comprises, from outside to inside, a back layer, a heating layer and a solid medicine layer, wherein the solid medicine layer is provided with solid medicine A; and further comprises a medicine liquid cavity for sealing liquid medicine B; when used, the medicine patch is attached to the human body for a set time, the medicine liquid cavity can be opened and the liquid medicine B can flow into the solid medicine layer; the solid medicine A can preliminarily treat related diseases under the action of heating, and the liquid medicine B is added in a progressive manner after the set time; the medicine patch has the gradient treatment requirement and the treatment effect, and can further drive the residual effect of the solid medicine B to continuously act on the affected area, improve the utilization rate of the medicine, and make the treatment medicine more targeted; thus, the traditional inertial thinking of the medicine patch is changed, interval medicine for treatment is made possible, the medicine effect is improved, and the foundation for the final treatment effect is laid.