825results about "Epidemiological alert systems" patented technology

The invention discloses an adverse drug reaction event recognition method and system based on multivariate knowledge mixed retrieval enhancement, and the method comprises the steps: extracting drug entities from a to-be-recognized clinical disease course record, segmenting the disease course record, and obtaining a drug entity set and a sentence set; for each extracted drug entity, retrieving drug concept knowledge having a hyponymy relationship with the drug entity and drug adverse reaction knowledge having an adverse reaction relationship with the drug entity in a pre-constructed multivariate knowledge base; for each sentence obtained through segmentation, searching suspected adverse drug reaction events meeting a first similarity requirement and drug field text knowledge meeting a second similarity requirement in a pre-constructed multivariate knowledge base; and finally, calling a large language model, taking all the retrieved knowledge as reference knowledge, and identifying the adverse drug reaction event from the to-be-identified clinical disease course record. According to the invention, the accuracy and reliability of adverse drug reaction event identification can be improved.

The invention discloses a medical insurance cost prediction and optimization method based on big data analysis, and belongs to the technical field of medical data processing. The method comprises the following steps: constructing a multi-source data acquisition module, and integrating data of an HIS system, a medical insurance platform and wearable equipment by using an FHIR interface and a block chain; a BERT-BiLSTM-CRF model is adopted to fuse multi-modal data, a dynamic model group containing Transform anomaly detection and LSTM-ARIMA time sequence prediction is established, and the prediction error rate is reduced to 12% (reduced by 28% compared with that of a traditional method); a multi-objective optimization system is designed, medical quality and cost control are balanced based on an improved NSGA-II algorithm, the cost of a single disease is reduced by 18%-25%, and the quality standard reaching rate exceeds 95%; a hybrid cloud and homomorphic encryption module is deployed, and the data leakage risk is reduced by 90%; a policy sandbox system is constructed, medical insurance policy simulation deduction is supported, and the response time is shortened to 72 hours. According to the invention, the problems of data islands, low prediction precision and privacy risks are solved, and the whole-process intelligent management and control of medical insurance fees is realized.

The invention relates to the technical field of animal monitoring, and provides a wild animal epidemic disease monitoring, prevention and control method and system based on artificial intelligence, and the system collects the video stream, the shell temperature, the air pathogen concentration, the sound characteristics, the VOCs spectrogram and other data of wild animals in real time through arranging multi-mode sensor nodes. Real-time reasoning is carried out through a wildness degree AI model in the edge calculation unit, the health state and epidemic disease risk of animals are evaluated, a multi-source risk knowledge graph and a graph neural network are combined, an epidemic situation occurrence probability threshold value is dynamically adjusted by the system, and accurate prevention and control instructions such as risk area division, isolation early warning and material putting schemes are generated; after the epidemic disease risk is confirmed, the system sends early warning information to a prevention and control center through various communication links, and continuously optimizes a prevention and control strategy through reinforcement learning. According to the method, intelligence and precision of epidemic disease prevention and control of wild animals are achieved, complex ecological environment changes can be coped with in real time, and prevention and control efficiency and accuracy are remarkably improved.

The invention discloses a multi-agent real-world clinical curative effect evaluation and accurate decision-making system, and relates to the technical field of medical data processing. According to the invention, a research demand analysis agent generates a research scheme after understanding the input content of a user and transmits the research scheme to a data management agent, a statistical analysis modeling agent, a result report generation agent, a data security agent and the data management agent privacy and encrypt data uploaded by the user; meanwhile, additional feature construction is carried out according to the research requirement of a user to form brand new analysis data used by a subsequent statistical analysis modeling agent, and after the statistical analysis modeling agent obtains the data, an analysis result is obtained by calling external software and is transmitted to a result report generation agent; and generating a clinical evaluation report together with the research scheme transmitted by the research demand analysis agent. According to the invention, full-process automatic, specialized and safe research support is realized.

The invention relates to the technical field of public health safety, and discloses a quantitative evaluation method and system for pathogen transmission risk, and the evaluation method comprises the following steps: step 1, obtaining virulence genes, reference virulence genes, environmental parameters, transmission factors and host risk factor parameters of pathogen samples; 2, performing variation detection on the virulence gene by adopting various variation detection algorithms to obtain all variation sites, and processing all variation sites to obtain effective variation sites; 3, constructing a virulence index model, inputting the variation distance into the virulence index model, and calculating the virulence index of the pathogenic bacteria sample; and 4, constructing a propagation risk assessment model, inputting the virulence index, the environmental factor, the propagation coefficient and the host susceptibility risk index into the propagation risk assessment model, and outputting a risk value by the propagation risk assessment model. According to the method, accurate prediction of the pathogen transmission risk can be realized.

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 an identification and analysis method for unknown pathogenic microorganisms, which comprises the following steps: filtering out genome sequences with low integrity, pollution and tag errors, and establishing a high-quality virus identification database; constructing a virus host prediction model through a machine learning algorithm; unknown pathogenic microorganisms are identified and analyzed, potential hosts or pathogenicity of the unknown microorganisms are identified, and whether the unknown microorganisms are unknown pathogenic viruses or bacteria or not is further judged. On the basis of metagenome data analysis, potential unknown pathogenic microorganisms in samples of human bodies, environments and the like can be identified more accurately.

The invention discloses a lower limb vein thrombus formation risk prediction method based on machine learning. The method comprises the following steps: step 1, constructing a triaxial thrombus evolution dynamic container space; 2, generating a pre-thrombus micro-state orbital chain; 3, determining a thrombus formation critical mutation window; 4, introducing a causal stem budget at the thrombus formation critical mutation window, and constructing a natural evolution path and a controlled path; step 5, obtaining an orbit offset difference value; 6, calculating a dynamic orbit stability index; 7, inputting the pre-thrombus micro-state orbit chain and the dynamic orbit stability index into the improved PatchTST model, and introducing a flow continuity regulation operator into a reconfiguration module to obtain a corrected orbit stability index; and 8, outputting a thrombus formation risk prediction result. According to the method, the thrombus formation risk prediction is realized by constructing a triaxial thrombus evolution dynamic container space and combining a mental structure differential equation model and an improved PatchTST model.

The invention relates to the technical field of space-time prediction, in particular to a brucellosis space-time prediction method and system based on artificial intelligence, and the method comprises the following steps: obtaining case information, calculating an incidence relation, extracting a path sequence, screening a trend direction, and matching a new case to generate prediction data. According to the method, continuous identification of a propagation path is realized by constructing a propagation incidence relation between cases and introducing a spatial directivity index, a non-trend propagation process is screened out through an angle average value and a variance, spatial consistency of path screening is enhanced, and through joint matching of spatio-temporal characteristics of newly-added cases and an existing propagation trend, the propagation path screening efficiency is improved. The sensitivity of prediction to the trend attribution of a new case is improved, the spatial directivity of a potential disease area is enhanced through reverse projection of a trend path and area positioning drop point frequency analysis, and through linkage processing of multi-level path extraction, trend judgment and area coding, the probability of occurrence of the new case is lowered. And the capturing capability of the prediction data on the propagation and evolution characteristics of the brucellosis is improved.

The invention discloses an industry and trade enterprise public health service resource optimization decision support system, and relates to the technical field of public health management, and the system comprises a data perception and access layer, a semantic fusion and modeling layer, a knowledge graph and risk calculation layer, an intelligent early warning and decision layer, and a feedback optimization closed loop. Multi-source heterogeneous data are collected through a protocol adapter, a unified data view is constructed through industry and trade public health field ontology semantic fusion, a space-time knowledge graph is dynamically generated, risks are deduced in combination with an improved SEIR space-time propagation model and a Monte Carlo method, visual early warning and interactive decision making are achieved based on digital twinborn bodies, and the method is applied to the field of industry and trade public health. According to the method, data islands can be broken, accurate advanced early warning of public health risks is realized, resource allocation is optimized, and the method is suitable for industry and trade enterprises with dense personnel such as machine manufacturing and electronic processing.

The invention relates to the technical field of infectious disease monitoring and early warning, and discloses an intelligent influenza early warning system based on community multi-modal data fusion. The community-level multi-modal data fusion architecture is constructed, medical health data, environmental data, crowd activity data and network behavior data are integrated, spatial-temporal features are dynamically extracted and fused in combination with a deep learning model, and the problem of community monitoring blind areas caused by a single data source of an existing early warning system is solved; a long short-term memory network and convolutional neural network cascade architecture is utilized to capture a localized propagation rule, and the defect that a region-level prediction model cannot adapt to community heterogeneity is overcome; the risk score is generated in real time, the grading response instruction is triggered, a'monitoring-early warning-intervention 'closed loop is established, the early warning timeliness is remarkably improved, a basic-level response chain scission gap is filled, early prevention and control of flu outbreak are finally achieved, and public health resource consumption is reduced.

The invention discloses a children respiratory tract infection risk prediction method based on multi-source medical data, particularly relates to the field of respiratory tract infection risk prediction, and is used for solving the problems that an existing method cannot identify a cross-regional co-infection mode and cannot prevent an input transmission risk. According to the method, a high-incidence time period baseline is constructed by extracting children respiratory tract infection historical electronic medical records of all regions, a local co-infection mode feature library is established based on a pathogen positive case co-occurrence relation, and a region-time period-pathogen three-dimensional collaborative tensor is constructed to recognize a synchronous high-incidence region group. And capturing real-time pathogen data of the synchronous high-incidence area, matching the real-time pathogen data with a local co-infection mode to generate a real-time risk list, deducing an input co-infection mode in combination with interregional personnel circulation intensity, and calculating the cross-regional risk permeability. And finally, a risk heat map is generated based on the permeability, and regional linkage-oriented children respiratory tract infection risk prediction is realized.

The invention relates to the technical field of epidemic disease prediction, and discloses an intelligent epidemic disease prediction system based on deep learning. A multivariate data acquisition module of the system acquires environmental meteorological data, population flow data, medical resource distribution data and historical epidemic situation propagation data related to epidemic diseases in real time; a deep learning prediction module analyzes regional propagation dynamic characteristics through a space-time diagram neural network according to the multi-source data; a risk area identification module constructs an area risk level thermodynamic diagram according to the area propagation dynamic characteristics, and identifies a high-risk area of which the risk level exceeds a preset threshold value; and the resource matching analysis module calculates the resource coverage range of each medical institution in combination with the medical resource distribution data and the spatial position of the high-risk region. The system can integrate multi-source data, accurately analyze the epidemic propagation trend, intelligently identify high-risk areas, realize reasonable matching of medical resources, and provide effective technical support for epidemic prevention and control.

The invention provides a disease burden prediction and prevention and control decision-making method and system based on machine learning, and relates to the technical field of disease prediction and public health decision-making, and the method comprises the steps: obtaining epidemiological data of tuberculosis and related diseases from an authoritative database, and carrying out the preprocessing of the epidemiological data; respectively training an XGBoost model, an RF (Radio Frequency) model and a Prophet model; training a random forest meta-model by adopting a Stacking fusion strategy, and constructing a hybrid prediction model; calculating RMSE, MAE, MAPE and Rindex evaluation model performance; based on the obtained hybrid prediction model; based on the variable importance analysis result and the prediction result, the influence of the key independent variable on the tuberculosis burden dependent variable is quantified, the effect of the independent variable change on the dependent variable is simulated, and a tuberculosis prevention and control intervention strategy suggestion is generated. By fusing multi-source data and a hybrid modeling technology, tuberculosis prediction precision is remarkably improved, confidence interval quantization and prevention and control strategies are linked for the first time, and data-driven decision support is provided for global tuberculosis prevention and control.

The invention relates to the technical field of intelligent regulation and control and energy conservation of hospital environment air quality, in particular to a big data management and intelligent evaluation system for the hospital environment air quality. The system comprises a multi-modal data acquisition module, a data processing module, a dynamic transmission modeling module, a dynamic Bayesian network risk modeling module and a prospective risk hedging and energy consumption optimization module. The system calculates a dynamic air transmission coefficient by collecting environment and people flow data, and constructs a dynamic Bayesian network model to calculate a cross-region propagation risk probability; the method is characterized in that when the risk probability exceeds a threshold value, a system actively solves a multi-objective optimization problem with minimization of energy consumption as an objective, and an optimal HVAC control instruction is generated; according to the method, the conversion from lagging evaluation to prospective risk hedging is realized, and the risk can be actively identified and regulated before the pollution exceeds the standard.

The invention discloses a double-layer high-order network structure reconstruction method and system based on maximum likelihood estimation, and relates to the technical field of network reconstruction. The method comprises the following steps: constructing a double-layer network model based on an information transmission model and a disease transmission model; a likelihood function is constructed based on the node state time sequence, and items associated with an information diffusion layer and items associated with a disease transmission layer are decomposed; according to the method, likelihood function approximate solution is carried out through a state average field approximation method, a second-order Taylor expansion method and a two-stage reconstruction strategy; according to the first-order Taylor expansion method, likelihood function approximate solution is carried out through a state average field approximation method and the first-order Taylor expansion method; and network topology reconstruction is carried out on the basis of a solving result. A dynamic coupling process of two layers is established through an interlayer coupling mechanism; layered optimization is realized by decomposing a likelihood function; and finally, the likelihood function is converted into a linear equation or an equation set for solving, so that the calculation complexity is reduced, and the reconstruction precision is ensured.

The invention relates to an animal epidemic disease monitoring data statistical analysis system and method. The method comprises the following steps: a multi-source data fusion module performs standardized fusion processing based on input case data, environment data and population data to obtain a spatio-temporal joint data matrix; a risk probability prediction module performs area scanning shape prediction on geographic feature data and population distribution data in the matrix, outputs a dynamic scanning area set, and performs environment and small sample correction to obtain risk probability distribution; and the hierarchical prevention and control visualization module performs spatial aggregation test based on the distribution, outputs a significant epidemic disease aggregation area, and generates a hierarchical prevention and control thermodynamic diagram in combination with an environment correction factor. According to the system, through multi-source data fusion, geographically adaptive dynamic window prediction and spatial aggregation test, the animal epidemic disease risk probability can be accurately analyzed, an epidemic disease aggregation area is output, a hierarchical prevention and control thermodynamic diagram is generated, and the animal epidemic disease monitoring level and the prevention and control visualization capability are further improved.

The invention provides a PINN-based natural epidemic disease transmission mechanism modeling method and system, which are used for mechanism deduction and space-time prediction of a natural epidemic disease transmission process. The method comprises the following steps: firstly, constructing a host population model and a host-population coupling propagation dynamics model; secondly, the dynamic equation set serves as a physical constraint to be embedded into the physical information neural network; and finally, through joint loss function optimization network learning, bidirectional driving of case data prediction and parameter inversion is realized. The method solves the problems that in the prior art, a host-crowd coupling mechanism is lacked, environmental factor modeling is insufficient, and data driving and mechanism modeling are separated, two-way driving of mechanism modeling and data learning is achieved, and space-time simulation precision, interpretability and model generalization ability are remarkably improved.

The invention relates to the technical field of biomedical engineering, and discloses a method and system for early warning and analyzing high-risk groups with high altitude polycythemia, and the method comprises the steps: screening a subject data set from an extremely high altitude area; performing structured processing on the subject data set to obtain a structured feature matrix, and extracting a core prediction factor from the structured feature matrix; candidate early warning models of the core predictive factors are generated, and the candidate early warning models comprise a logistic regression model, an XGBoost model and a random forest model; screening an optimal early-warning model from the candidate early-warning models, and establishing a high altitude polycythemia early-warning system of the subject data set through the optimal early-warning model; and effect verification is carried out on the high altitude polycythemia early warning system so as to realize high altitude polycythemia high-risk group early warning analysis processing of the subject data set. According to the method, the core problem that the early warning result is one-sided and unreliable due to three defects of data dimension missing, static evaluation limitation and extensive privacy mechanism can be solved.

The invention discloses a pathogen transmission rapid early warning and traceability analysis method based on multi-source data fusion. The method comprises the steps of S1, collecting multi-source data such as drug sales, network behaviors, social media, outpatient diagnosis and traffic time and space; s2, through cleaning and standardized preprocessing, social text disease features are extracted by adopting BERT; s3, mining a comprehensive weak signal through single-source anomaly detection and multi-source correlation analysis; s4, fusing the features by using an Attention-LSTM model, and outputting a regional risk index; s5, training an early warning model based on historical data, and setting a three-level threshold to trigger early warning; and S6, positioning a propagation starting point in combination with the spatio-temporal data, and constructing a propagation chain through a graph neural network. The early warning is advanced by 3-7 days, the traceability precision reaches the community level, and the prevention and control precision is improved.

The invention discloses an infection propagation path sensing method and system based on a time-space relationship. The method comprises the following steps: acquiring and cleaning hospitalization, nursing and inspection data of patients, and constructing a time sequence file of each patient; defining the contact type of the same ward and the medical care medium based on the time sequence file, setting an infection period and an observation window to identify a potential infection transmission path, and forming a contact event table; sorting the time information of the positive patients and the contactors thereof based on the contact event table, and dynamically generating a directed propagation chain; performing cyclic path elimination and credibility grading on the directed propagation chain to obtain a corrected propagation network; and screening out cases without upstream infection sources according to the corrected propagation network, and determining infection sources and properties thereof in combination with hospitalization time and historical hospitalization records. Through the implementation of the method, the adaptive limitation of the prior art to a complex propagation scene can be solved, rapid reconstruction and accurate attribution of a hospital infection propagation chain are realized, and intelligent decision support is provided for reducing the incidence rate of hospital infection.

The application discloses a disease transmission link prediction method and system based on a dynamic hypergraph neural network, aiming to comprehensively depict virus transmission paths and risk levels from multiple dimensions. First, data is collected from a gene database, mobile device logs and clinical texts, virus variation characteristics, population contact relationships and symptom semantic information are extracted respectively, and a dynamic hypergraph that integrates virus strain nodes, individual nodes and regional nodes is constructed. The hyperedge is defined according to molecular similarity, space-time contact and semantic overlap, and the weight is dynamically adjusted according to the transmission intensity of the infectious disease. A multi-modal hypergraph neural network with variation perception is introduced to jointly model and adaptively fuse multiple types of hyperedges and multi-scale features, output node propagation embedding vectors and predict potential transmission chains, perform prevention and control responses, and realize real-time deployment and feedback loop in an edge-cloud collaborative manner, ensuring efficient and accurate infectious disease transmission modeling and intervention control while ensuring data privacy.

The invention relates to the technical field of molecular biology, and discloses a hepatitis E virus rapid detection method and system based on an animal detection technology, and the method comprises the following steps: carrying out nucleic acid extraction on a hepatitis E virus sample to obtain a nucleic acid extraction sample; establishing a PCR system of the nucleic acid extraction sample, and generating uniform microdroplets of the nucleic acid extraction sample and the PCR system; calculating a rupture coefficient of the uniform microdroplet, performing PCR amplification on the uniform microdroplet to obtain an amplified microdroplet, and performing signal detection on the amplified microdroplet by using a preset fluorescent dual-channel to obtain a fluorescent detection signal; generating a two-dimensional scatter diagram of the amplified droplets to determine FAM positive droplets in the amplified droplets; and analyzing the HEV genotype and variation site of the hepatitis E virus sample to establish an HEV strain evolutionary tree and risk-inducing factors of the hepatitis E virus sample, and generating an HEV geographical distribution heat map of the to-be-detected region by combining the HEV strain evolutionary tree and the risk-inducing factors. According to the invention, the accuracy of rapid detection of hepatitis E virus can be improved.

The invention discloses a cage rearing broiler chicken multi-level epidemic situation early warning and prevention and control optimization method, which comprises the following steps: S1, constructing a data support system; S2, constructing a cross-level epidemic situation propagation simulation system, and realizing multi-scale digital mapping from a coop to a region; s3, according to the epidemic situation simulation result in the S2, early warning grades are divided; and S4, according to the early warning level in the S3, an association rule base addition and deletion measure is called, a candidate prevention and control measure combination is generated, and dual-target optimal measure combination screening with the maximum control effect and the minimum total cost is realized. According to the method, multi-scale accurate simulation of epidemic situation propagation and quantitative optimization of prevention and control measures can be realized, the control effect and cost are balanced, the scientificity and operability of prevention and control decisions are improved, and the method is adaptive to different early warning levels and breeding scenes and is high in robustness and practicability.

The invention relates to the technical field of computers, and discloses a measure generation method and device based on a strategy model and terminal equipment. The method comprises the following steps: acquiring a plurality of time sequences, wherein each time sequence comprises intervention measure data, discrete sign data and continuous sign data corresponding to a plurality of historical moments before a target moment; aligning and splicing time sequences including the intervention measure data, the discrete sign data and the continuous sign data according to time steps to obtain a combined time sequence; carrying out vectorization processing and position coding on the combined time sequence to obtain a sequence tensor; processing the sequence tensor through a preset encoder to obtain joint conditional probability distribution corresponding to the next moment of the target moment; and processing the joint conditional probability distribution through an intervention strategy model, and determining an intervention measure corresponding to the next moment of the target moment. By adopting the method, the optimal or reasonable intervention measure at the next moment can be intelligently determined.

Techniques for geospatial-temporal pathogen tracing in zero knowledge include: generating, by a first user device, a first proximity token for contact tracing; receiving, by the first user device, a second proximity token from a second user device; generating, by the first user device, a hash based on the first proximity token and the second proximity token; generating, by the first user device using a prover function of a preprocessing zero knowledge succinct non-interactive argument of knowledge (pp-zk-SNARK), a cryptographic proof attesting that an individual associated with the first user device tested positive for a pathogen; transmitting, by the first user device, first publicly verifiable exposure data including at least the cryptographic proof and the hash to a public registry; and applying at least the first publicly verifiable exposure data and second publicly verifiable exposure data to a machine learning model, to obtain actionable intelligence associated with the pathogen.

The invention belongs to the field of infectious disease community transmission multi-subject simulation, and relates to an infectious disease community transmission multi-subject simulation method and system containing asymptomatic infectors. The method comprises the following steps: establishing a virtual community group which comprises a plurality of communities and a plurality of destinations; distinguishing subjects in the virtual community group by using the health state and the space state, wherein the subjects comprise a susceptible subject S, an exposed subject E, an asymptomatic infection subject A, an infection subject I and a rehabilitation subject R; a motion sequence of all subjects in the virtual community group is established, the infectious disease transmission process is simulated according to the motion sequence, and the motion sequence comprises movement, infection and state updating, infection detection and isolation and community sealing control; and obtaining a simulation result, and evaluating the effectiveness of the infectious disease prevention and control policy. According to the method, rapid evaluation of different prevention and control strategies on infectious disease transmission conditions, social and economic influences caused by infectious diseases and the like can be realized, and beneficial reference is provided for a macroscopic regulation and control policy for restraining infectious diseases.

The invention discloses an early warning method and system for outbreak of drug-resistant bacteria in a hospital based on dynamic aggregation degree evaluation. The method comprises the following steps: acquiring hospitalization information of a patient, microbiological inspection data and spatial layout information of an inpatient area to obtain initial data; judging whether a new case of drug-resistant bacteria is detected on the current day; if not, gradually reducing the historical aggregation degree by using an exponential decay formula until the historical aggregation degree returns to zero; if yes, calculating a newly added detection rate, a space aggregation degree and a time aggregation degree, and calculating a total aggregation degree of the day; dividing an inpatient area; dynamically generating a threshold value for the high-frequency inpatient area by adopting a quartile method to carry out abnormal value judgment, and identifying an abnormal aggregation degree for the low-frequency inpatient area by using a fixed absolute value threshold mechanism to obtain an early warning level; and early warning information of different levels is pushed according to the early warning level, and a prevention and control suggestion plan is attached. By implementing the method provided by the invention, the accuracy and efficiency of early warning can be remarkably improved, so that patients and medical personnel are better protected from the threat of drug-resistant bacterium infection.

The invention discloses a large language model liver cancer prediction method and system based on an improved sampling strategy, and relates to the technical field of artificial intelligence medical diagnosis, and the method comprises the steps: obtaining an electronic medical record of a patient, carrying out the sequential reconstruction, and constructing a structured cue word; a plurality of reasoning paths are generated in parallel by using a random decoding strategy according to a general large language model with frozen input parameters; constructing a target distribution model based on gamma distribution, and calculating the normalized weight of each path by adopting an importance sampling algorithm to suppress low-quality paths and amplify the weight conforming to medical logic paths; and finally, performing weighted aggregation on the diagnosis conclusion based on the weight to obtain a prediction result, and outputting the reasoning process with the highest weight as an interpretability report. The accuracy of liver cancer prediction and the clinical decision transparency can be remarkably improved without fine adjustment of the model.

The invention discloses an animal epidemic disease regional transmission risk prediction model construction method and prediction method based on big data analysis, and relates to the technical field of animal epidemic disease prevention and control, and the method comprises the steps: collecting multi-dimensional dynamic data related to animal epidemic disease regional transmission through a variety of devices or platforms; after data layering preprocessing, propagation association features are extracted in combination with an epidemic disease propagation kinetic model, and a three-dimensional feature vector is generated through fusion; a prediction model is constructed based on DeepSTN, spatial-temporal features of CNN and bidirectional LSTM are fused by adopting a dynamic weight distribution mechanism, and online incremental learning is realized through transfer learning; and establishing a multi-dimensional evaluation system, carrying out model parameter iterative optimization through reinforcement learning, generating a simulation data expansion training set in combination with the GAN network, and forming a closed-loop optimization mechanism to obtain an optimal prediction model and carry out propagation risk prediction. According to the method, the accuracy and timeliness of animal epidemic disease regional transmission risk prediction are improved, and a scientific basis is provided for accurate prevention and control.