278 results about "Disease risk" patented technology

The invention discloses a medical intelligent decision-making method based on Deepseek and a time sequence causal knowledge graph, and the method comprises the following steps: 1, constructing an initial static medical knowledge graph, and generating a dynamic time sequence causal knowledge graph; 2, finely adjusting and training the DeepSeek model to enable the DeepSeek model to adapt to the medical field; and step 3, receiving and analyzing the text uploaded by the patient, performing intelligent triage and disease risk prediction, and realizing accurate matching of patient symptoms and target departments and intelligent prediction and early warning of potential diseases. The method aims at providing accurate triage and disease risk prediction for patients, constructing a scientific and efficient medical intelligent decision-making mechanism and optimizing medical resource allocation.

The embodiment of the invention provides a method and system for generating medical suggestions based on multi-modal data fusion, and the method comprises the steps: integrating a medical image, a physical examination report and dynamic physiological parameters of a patient through a multi-source data fusion module, generating a multi-modal data set, and synchronously inputting the multi-modal data set into a hybrid reasoning module and a dynamic knowledge graph engine. And the dynamic knowledge graph engine accurately recall a target diagnosis and treatment guide associated with the current multi-modal data set. The rule reasoning sub-module generates a first diagnosis suggestion containing a diagnosis conclusion, a treatment scheme and an evidence level based on a guide structured rule, and meanwhile, the neural network reasoning sub-module analyzes a multi-modal data set by relying on a triple topological structure and an edge weight; and generating a second diagnosis suggestion comprising the disease risk probability, the differentiated treatment suggestion and the evidence source. And finally, the interactive output module fuses the two suggestions to generate a medical suggestion report covering the diagnosis basis, the evidence level and the treatment scheme, so that the diagnosis and treatment precision of chronic disease management and health risk assessment is remarkably improved.

The invention discloses a multi-agent large-model disease diagnosis knowledge reasoning system based on data dual drive, and relates to the technical field of artificial intelligence assisted medical diagnosis. The system collects patient symptom follow-up records, laboratory test results, observation diagnosis probabilities and expert diagnosis recommendation results in a multi-source manner; time sequence evolution characteristics are extracted, a time sequence diagnosis sensitivity coefficient is calculated, and early recognition of disease risks is achieved; in combination with anti-fact simulation and statistical reasoning, a causal consistency coefficient is obtained and is used for verifying causal reasonability of observation diagnosis and contrast results; based on agent group consensus analysis, calculating a game consistency coefficient for judging the credibility of a diagnosis conclusion; positioning and multi-level verification are carried out on abnormal reasoning steps and knowledge fragments, so that the reliability and safety of a result are guaranteed; continuous optimization of the diagnosis model is realized through a log analysis and knowledge backflow mechanism; according to the invention, the accuracy, interpretability and safety of disease diagnosis can be obviously improved.

Artificial intelligence enabled disease profiling is described. An electrocardiogram analysis module is configured to derive disease vectors for a plurality of diseases using electrocardiogram training data from both disease-negative and disease-positive individuals. A standardized input is generated, via a data preprocessor of the electrocardiogram analysis module, from an electrocardiogram recorded from an individual. The standardized input is encoded, by a deep learning autoencoder of the electrocardiogram analysis module, into an embedding, the embedding being a lower-dimensional latent space representation of features extracted from the standardized input. At least one disease risk score for the individual is generated, by a statistical modeling algorithm of the electrocardiogram analysis module, for the plurality of diseases based on the embedding and the disease vectors.

The invention relates to the technical field of livestock and poultry epidemic disease management, and particularly discloses an intelligent livestock and poultry epidemic disease early warning and partitioned prevention and control management system and method, and the method comprises the steps: laying an integrated intelligent sensor network in a livestock and poultry farm, and collecting environment parameters, feeding management data and animal health data in real time through a low-power-consumption Internet of Things technology; transmitting the collected multi-dimensional data to a cloud platform by utilizing an edge computing technology; based on multi-dimensional data, a deep learning and multi-source data fusion analysis technology is adopted, a dynamic disease risk prediction model is constructed, and generalization ability of an adversarial generative network optimization model in a complex environment is introduced; by arranging the integrated intelligent sensor network, environmental parameters, feeding management data and animal health data in the livestock and poultry farm are comprehensively collected in real time, the early monitoring capability of epidemic diseases is remarkably improved, potential epidemic disease risks are found in time, and precious time is won for subsequent prevention and control work.

The invention discloses a chronic disease risk prediction method and system fusing a knowledge graph and a large language model, and the method comprises the following steps: obtaining a natural language problem related to a chronic disease, and carrying out the semantic analysis; according to the analysis content, hypothetical questions and answers related to chronic diseases are generated through a large language model, and key entities are extracted; mapping the key entities to corresponding nodes in a medical knowledge graph, exploring a semantic path and a causal relationship between the key entities, and constructing an inference chain pointing to potential disease risks from acquired information; introducing a fragment granularity sensing mechanism, performing fine granularity analysis on each fragment in the reasoning chain, and rearranging and optimizing a link sequence; and based on the optimized inference chain, converting the question and answer result into a structured diagnosis result for visual display. According to the method, the whole process from question asking to answer generation of the patient is optimized, the efficiency and accuracy of chronic disease risk prediction are effectively improved, and meanwhile, personalized health management service is provided for the patient.

The invention relates to the technical field of medical and multi-modal large models, and discloses a disease prediction and auxiliary diagnosis system construction method and system based on a multi-modal large model. The method comprises the steps of report format conversion, data cleaning and table image-to-structured text conversion. Constructing a retrieval knowledge base to enhance the retrieval capability; precise cue word design and reasoning optimization are carried out; small sample learning and model fine tuning; the invention discloses a multi-modal large model integration and visualization system. According to the system, the accuracy problem of a traditional disease risk prediction method is effectively solved, and a more reliable auxiliary diagnosis tool is provided.

The invention relates to the technical field of newborn medical big data risk assessment, and discloses a newborn disease risk assessment method based on big data analysis. The method comprises the following steps: acquiring an initial multi-modal health record of a newborn from a multi-heterogeneous medical data source; carrying out fusion analysis on the data, and constructing a risk assessment map driven by an event; starting a continuous risk tracking process, performing hierarchical risk scanning by using the map, and identifying a stable risk cluster, an evolution risk cluster and an isolated risk signal; deducing a hierarchical risk management and control plan based on the spatial distribution and time evolution mode of the risk cluster; at the end of each period, integrating latest clinical treatment feedback and monitoring readings, and carrying out reverse calibration on entity node states and relation edge strength in the atlas; and dynamically recombining a measure execution sequence and resource allocation in the plan according to the calibrated atlas. According to the invention, dynamic and continuous evaluation and self-adaptive intervention planning of the neonatal health risk are realized.

The invention discloses an earth and rockfill dam illness feature mining method and system based on historical text data, and the method comprises the steps: collecting earth and rockfill dam historical illness data, constructing an earth and rockfill dam illness diagnosis text corpus set, carrying out the structured preprocessing of the corpus set, and obtaining an illness feature and danger removal measure text corpus set; generating a structured word sequence subset through a word segmentation tool; processing the structured word sequence subset by adopting an LDA topic model, determining an optimal topic number through a confusion degree curve, and outputting a final topic and a corresponding topic word; constructing a visual network graph based on the subject term co-occurrence frequency; and carrying out centrality analysis on nodes in the visual network diagram, identifying key nodes in the visual network diagram, quantitatively analyzing association rules of the danger characteristics and danger removing measures, and completing feature mining of the earth and rockfill dam danger. The problems that traditional manual diagnosis is high in subjectivity and low in utilization rate of historical engineering data are solved, and intelligent auxiliary decision making of the earth and rockfill dam danger is achieved.

The invention discloses a medical record data mining and potential risk prediction method and system based on deep learning, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining a first semantic vector set containing the current symptom multi-role chief complaint information of a target child, multi-modal physiological data, basic treatment information and initial cognitive level model parameters, a conflict vector set is generated through conflict recognition, initial cognitive level model parameters are adjusted in combination with multi-modal physiological data, individual cognitive level model parameters are obtained, and then a questioning text set is generated according to the conflict vector set and the individual cognitive level model parameters; after feedback is received, conflict resolution is carried out on the first semantic vector set to obtain a second semantic vector set, the second semantic vector set is fused with basic doctor seeing information to generate medical record combination feature data, and a potential disease risk prediction result of the target child is generated according to the medical record combination feature data; the method has the beneficial effects that the multi-role chief complaint information conflict in the child medical record can be effectively processed, and the potential disease risk prediction accuracy is improved.

The invention discloses an intelligent animal epidemic disease monitoring management and early warning workstation, which comprises a multi-modal data acquisition module, an epidemic disease risk assessment module, a dynamic early warning threshold generation and judgment module, a visual human-computer interaction interface and a block chain storage module, the multi-modal data acquisition module is used for acquiring animal data of each monitoring point in a preset monitoring area and environment data of animals; the epidemic disease risk assessment module is used for comprehensively analyzing the animal data and the data of the environment where the animals are located to obtain epidemic disease risk indexes of all monitoring points in the current monitoring area; the dynamic early warning threshold generation module is used for dynamically generating an early warning threshold and judging an epidemic disease risk index to obtain an epidemic disease risk level and an epidemic disease risk area; the visual human-computer interaction interface is used for displaying the epidemic disease risk condition in the monitoring area; and the block chain storage module is used for data storage. According to the invention, the accuracy and timeliness of early warning of epidemic diseases are improved.

The invention relates to the technical field of medical data processing, and discloses a disease risk assessment method based on a disease risk knowledge graph, the disease risk knowledge graph comprises a disease risk rule and a disease knowledge entity, and the disease knowledge entity comprises a focus position entity; performing organ positioning according to the physical examination data of the target object, and determining a target organ; performing organ matching and index data matching in a disease risk rule based on the target organ and the physical examination data to obtain a risk matching rule corresponding to the target organ; performing disease risk level analysis on the target organ according to the risk matching rule to obtain a disease risk level of the target object about the target organ; generating a physical examination report of the target object according to the physical examination data, the disease risk level and the target focus position; wherein the physical examination report comprises a three-dimensional model used for displaying the target focus position. The method has the beneficial effects that the pertinence and efficiency of disease risk assessment are improved, and the target object is helped to understand the health state of the target object.

The invention provides a teenager disease risk dynamic prediction method and system, and the method comprises the steps: obtaining an electronic health record and subcutaneous metabolite dynamic concentration data of a teenager, and the electronic health record comprises chronic disease related information, a genetic risk value and historical diagnosis data; performing multi-modal fusion on the metabolite concentration data, the genetic risk value and the historical diagnosis data according to a time axis to form a mixed time sequence information set; adaptive optimization is carried out on the mixed time sequence information set, and the collaborative change trend of metabolite fluctuation characteristics and genetic characteristics is identified from the optimized mixed time sequence information set; generating a dynamic graph by adopting a graph neural network based on the collaborative change trend; and tracking an abnormal conduction chain of a metabolic pathway in the dynamic map in real time, and calculating a risk probability matched with the growth stage of the teenagers according to the abnormal conduction chain and chronic disease related information. The dynamic prediction precision and timeliness of the chronic disease risk of teenagers are improved.

The invention discloses a disease risk assessment method and screening device based on a multi-group student physical collaborative digital network, and relates to the field of intelligent medical detection. In order to solve the defect that multi-omics-level system collaborative analysis and robust risk assessment are difficult to realize in the prior art, the technical scheme provided by the invention is as follows: acquiring a plasma sample, acquiring a spectral signal by adopting an attenuated total reflection Fourier transform infrared spectrum, and establishing a plasma spectrum digital information space; the method comprises the following steps: constructing a biological collaborative digital network containing four nodes of protein, lipid, saccharides and nucleic acid based on pathophysiology priori knowledge, and defining node strength, edge weight and network collaborative efficiency; a health baseline configuration file is established by using a health sample, a standardized deviation score of a to-be-tested sample is calculated, a comprehensive risk score is obtained, a disease screening result is output in combination with a machine learning model, and digital evaluation of multi-omics collaborative characteristics is realized. The method is suitable for non-invasive rapid screening and risk assessment work of neurodegenerative diseases and mental diseases.

The invention relates to the technical field of geological survey, in particular to a rapid stratum disease risk detection method based on a ground penetrating radar, and solves the technical problem of depth positioning misalignment caused by propagation parameter fluctuation due to heterogeneity of an underground medium in stratum disease detection of the ground penetrating radar in the prior art. The method comprises the following steps: acquiring a reflected wave signal at each acquisition moment through a ground penetrating radar; performing signal decomposition processing on the reflected wave signal, separating an intrinsic mode component representing dielectric noise, and performing feature extraction on the intrinsic mode component to determine a dielectric noise intensity index; constructing a dielectric noise intensity sequence according to the dielectric noise intensity index, and performing spatial-temporal characteristic analysis to determine the confidence coefficient of stratum diseases; and according to the dielectric noise intensity index and the stratum disease confidence coefficient, in combination with a pre-calibrated stratum average dielectric constant, constructing a depth correction weight, carrying out adaptive correction on the original stratum disease depth, and outputting the corrected stratum disease depth.

The invention relates to the technical field of infectious disease risk prediction, in particular to an infectious disease risk prediction analysis method and system based on electronic cases. Comprising the following steps: S1, preprocessing a multi-modal electronic case, cleaning structured data, and performing structured conversion on unstructured data; S2, constructing dynamic risk characteristics, quantifying basic disease risks, evaluating a key symptom change rate, and calculating a dynamic risk comprehensive index; S3, predicting infectious disease risks, and evaluating dynamic relevance between a patient and a group infectious disease condition. S4, optimizing a risk judgment threshold value, calibrating the threshold value through a Kappa coefficient, iteratively optimizing a model and features in combination with clinical feedback, forming a standardized scheme, improving data quality through multi-modal data integration, achieving accurate risk assessment through two-dimensional layering, dynamically calibrating the threshold value, optimizing and enhancing model adaptability through clinical feedback, and improving the reliability of the model. A standardized scheme guarantees application consistency, and accuracy and clinical practicability of infectious disease risk prediction are effectively improved.

The invention relates to the technical field of osteoarticular diseases, in particular to an artificial intelligence auxiliary diagnosis and treatment system for osteoarticular diseases based on an internet hospital. Comprising an information management module; an intelligent diagnosis auxiliary module; a remote monitoring and rehabilitation guidance module; a disease risk prediction module; according to the system, a comprehensive system covering functions of information acquisition, data integration, image analysis, rehabilitation tracking and the like is constructed, so that whole-process information of a patient from a doctor seeing period to a rehabilitation period is systematically managed, and the continuity and personalized service level of diagnosis and treatment are effectively improved; and artificial intelligence technologies such as deep learning, natural language processing and knowledge graph are applied to automatically complete bone joint image recognition, symptom analysis and rehabilitation plan making, so that the diagnosis efficiency and accuracy of doctors are greatly improved, and meanwhile, dynamic adjustment and refined guidance of the rehabilitation plan are realized.

The invention relates to the technical field of disease monitoring, in particular to a multi-device collaborative disease risk prediction system. The system comprises a non-contact type signal receiving and transmitting device, a gait collecting device, an intelligent bracelet and a central processing unit, the non-contact type signal receiving and transmitting device captures macroscopic movement data of a user through a millimeter wave signal transmitting and receiving module, and the macroscopic movement data comprises periodic characteristics of gaits, movement tracks of four limbs and posture changes; the gait acquisition device acquires plantar pressure distribution data of a user through a pressure sensor array; the smart bracelet collects motion data of a user through an accelerometer and a gyroscope; and the central processing unit carries out space-time alignment on data of the non-contact signal transceiving device, the gait acquisition device and the intelligent bracelet through a timestamp synchronization algorithm, so that the disease risk is predicted according to the analysis data. The system supports all-weather and multi-scene monitoring, is suitable for disease risk assessment, and can early warn cardiovascular events and other health problems.

The invention provides a chronic respiratory system disease risk early warning method and device based on artificial intelligence and a medium. The method comprises the following steps: firstly, acquiring physiological monitoring information of wearable equipment of a patient; inputting the physiological monitoring information into the risk prediction model to obtain a risk early warning result; wherein the risk prediction model is established according to patient medical record information, follow-up visit information and historical physiological monitoring information; the risk prediction model is a Transform-LSTM (Long Short Term Memory) double-branch integrated model. According to the method, physiological monitoring information collected by wearable equipment in real time is input into a Transform-LSTM double-branch integrated model constructed on the basis of patient medical record, follow-up visit and historical physiological monitoring multi-source information, so that a long-distance dependency relationship of multi-modal data is captured by means of a Transform branch to identify a potential risk trend in a stable period; and the time sequence dynamic characteristics of physiological monitoring information are captured through an LSTM branch to perceive short-term signal mutation in an acute exacerbation period, so that accurate early warning of the whole course risk of the chronic respiratory system disease is realized.

The invention discloses an intelligent glasses system and method for health risk monitoring of old people, and belongs to the technical field of intelligent health monitoring. The system is integrated on a glasses body and comprises a multi-mode sensing module, an edge calculation module and an interaction module. The method comprises the following steps: synchronously acquiring physiological parameters and behavior posture data of a user, extracting head posture change characteristics for representing body instability on a local edge side, performing time sequence fusion on the head posture change characteristics and the physiological parameter data, and inputting the fusion result into a pre-trained risk identification model to identify a falling risk or a sudden disease risk. And early warning is triggered when the risk is identified. According to the invention, non-inductive, continuous and accurate monitoring and early warning of high-risk scenes of old people are realized, the defects of single data, high false alarm rate and large response delay of existing equipment are overcome, and the real-time performance and reliability of health monitoring are remarkably improved.

The embodiment of the invention provides a nurse occupational health management method and system, and the system comprises the steps: a flexible fabric pressure sensor array module is embedded in a preset position of a nurse uniform, and is used for collecting the posture biomechanical data of a nurse in real time; the microenvironment monitoring patch is used for detecting microenvironment data corresponding to the position information based on the position information of the nurse; the multi-modal data fusion processor is used for performing multi-modal data fusion on the microenvironment data to obtain fused data; the digital twin engine is used for constructing a nurse health prediction model marked with an occupational disease risk hot spot area based on the fused data and the attitude biomechanical data; and the self-adaptive intervention terminal is used for generating and outputting prompt information based on the occupational health prediction result output by the nurse occupational health prediction model. According to the system, the occurrence probability of occupational diseases of nurses is reduced, and the monitoring accuracy is improved.

The invention provides a non-serious disease risk prediction method based on multi-dimensional health data analysis, and the method can collect the multi-dimensional health data of a user, builds a high-precision serious disease risk prediction model through multi-source data fusion and intelligent analysis in combination with a machine learning algorithm and a statistical model, and improves the risk prediction accuracy. The health state of the user is dynamically evaluated based on the deep learning model, the potential disease risk of the user is predicted by analyzing the relevance between historical data and future health risks, the health trend, risk factors and improvement suggestions are covered, the user is helped to make a scientific health management plan, the limitation of a traditional prediction method is broken through, and the user experience is improved. And real-time data updating and model optimization are supported, and a more accurate and personalized prediction result is provided.

The invention relates to an aquaculture disease risk prediction method and device, equipment and a storage medium. The method comprises the steps of collecting an original data set, performing multi-scale decomposition on water quality parameter time series data, generating an intrinsic mode component set and a residual term, and screening components associated with historical disease tags from the intrinsic mode component set to obtain a sensitive mode subset; generating an environmental state prediction value based on the subset, generating an environmental baseline value based on the residual term, and calculating a dynamic weight based on the environmental state prediction value and the environmental baseline value; and performing feature extraction on the pathogen activity data, generating pathogen spatio-temporal features, generating joint features according to fusion of the dynamic weight and the environmental state predicted value, and generating a dynamic risk probability according to the joint features. According to the method, the accuracy and timeliness of aquaculture disease risk prediction are improved by fusing the sensitive mode subset after multi-scale decomposition and the spatio-temporal characteristics of the pathogens and dynamically adjusting the weight according to the environmental state prediction value and the baseline value.

The invention discloses an evolutionary intelligent agent system and method for automatically constructing a disease risk prediction model, and the system comprises a data processing module which is used for obtaining original medical data, calling a large language model to carry out standardization processing on the obtained original medical data; the feature extraction module is used for calling a large language model to extract features related to the target disease in the standardized medical data, and associating the extracted features with corresponding target disease tags to obtain a structured data set; the model construction and evaluation module is used for inputting the data set into a multi-objective Cartesian genetic programming MOCGP algorithm to construct a plurality of disease risk prediction models, and optimizing the plurality of disease risk prediction models through two optimization objectives of model interpretability and prediction performance, and at least one candidate disease risk prediction model with optimal prediction performance and high interpretability is obtained.

The invention discloses a co-disease collaborative identification and risk early warning method based on multi-modal data and an adaptive large model. The method comprises the following steps: collecting multi-modal medical data; carrying out preprocessing and feature extraction on the multi-modal medical data; based on the fine-tuned adaptive large model, deep interactive fusion is carried out on the extracted multi-modal features, potential association among different data sources and mutual influence and synergistic effect among various diseases are learned, and combined judgment and future onset risk prediction of the current common disease state of the patient are output; determining a specific co-disease combination and a co-disease risk level based on the combined judgment of the current co-disease state of the patient and a future onset risk prediction result; and when the determined co-disease risk level is equal to or higher than an early warning threshold, generating a collaborative early warning and intervention suggestion. According to the invention, a cross-modal and cross-disease feature fusion technology can be utilized to realize conjoint analysis and comprehensive risk early warning of multiple disease states of the patient, and the co-disease management efficiency and the clinical auxiliary decision-making level are significantly improved.

The invention relates to the technical field of intelligent peanut disease monitoring, in particular to an intelligent peanut disease monitoring method based on community diversity analysis. According to the method, soil environment parameters are collected in real time through a multi-source sensor array, and rhizosphere microorganism DNA is extracted based on an improved CTAB-PEG method for high-throughput sequencing; constructing a dynamic baseline model fusing a Shannon-Wiener index, a phylogenetic diversity index and an environment correction factor; synchronously analyzing spatial distribution and time evolution characteristics of an OTU abundance matrix by adopting a deep space-time convolutional network, and decoding a microbial anomaly succession signal through a gating circulation unit; and when the disease risk index exceeds a threshold value, activating an early warning terminal and generating a resistant variety adaptation scheme. According to the method, the problems of incomplete microorganism capture, misjudgment of environmental interference, early warning lag and the like of a traditional method are solved, and collaborative optimization of precise early warning and prevention and control decision of peanut diseases is realized.

The invention discloses a multi-modal disease prediction algorithm based on a dynamic health record, and the algorithm comprises the following steps: multi-modal health data collection: synchronously obtaining multi-source data comprising a time sequence physiological signal, image data, a clinical diagnosis result and a user behavior log through a non-contact sensing device and a clinical data interface; dynamic health archive construction: establishing an individual health feature network based on a graph database, and realizing association storage of real-time data and historical baselines through a time sequence alignment mechanism to form a dynamically updated health portrait; multi-modal feature fusion: carrying out cross-modal feature extraction on the acquired data; the method has the advantages that by integrating multi-modal data such as RGB / near-infrared images, rPPG physiological signals, near-infrared spectrum biochemical indexes, clinical data and behavior logs, cross-dimensional health assessment of physiology-biochemistry-behavior is achieved; the defect that traditional single-mode data cannot describe the health state completely is overcome, and the accuracy of disease risk prediction is remarkably improved.

The embodiment of the invention provides a disease risk prediction method and device, equipment, a storage medium and a program product. The method comprises the following steps: acquiring historical physical examination information of a target object; inputting the historical examination information of the target object into a disease risk prediction model to obtain a disease risk prediction result corresponding to the target object; wherein the disease risk prediction model is obtained based on training of a time-dependent loss function, and the time-dependent loss function is determined based on a weight adjustment factor related to a time step length and a smoothness regular term. The accuracy of disease risk prediction can be improved.

The invention relates to the technical field of disease risk prediction and precision medical treatment, in particular to an establishment method and application of a late-onset psoriasis risk prediction model, and the method comprises the following steps: (1) collecting lifestyle data, serum metabolite data, clinical characteristics and polygene risk scores of a subject; (2) constructing a healthy lifestyle score according to the lifestyle data; (3) screening metabolites significantly related to the healthy lifestyle by using a multiple linear regression model; (4) screening metabolites significantly related to the risk of late psoriasis by using a Cox regression model; (5) screening a key metabolite set by adopting an elastic network regression model; and (6) inputting the sample features into a machine learning model to obtain a late-onset psoriasis risk prediction model. According to the method, the prediction accuracy is high, the AUC can reach 0.86 by combining the lifestyle, serum metabolites and genetic risks, and the method is obviously superior to a prediction method only depending on clinical characteristics or genetic information.

The application provides a thymus disease risk prediction method and system based on cross-modal feature interaction, which comprises the following steps: respectively preprocessing CT image data, MRI image data and image reports; sequentially performing word segmentation, vector conversion, feature extraction and pooling on the third data, inputting the labeled semantic feature vector into an image generation network to generate a PET-CT image; inputting the first data, the second data and the PET-CT image into the same medical mamba feature extraction network respectively to obtain the first modal feature corresponding to the first data, the second modal feature corresponding to the second data and the third modal feature corresponding to the PET-CT image; performing feature fusion interaction on the first modal feature, the second modal feature and the third modal feature to obtain a fusion feature, and obtaining the thymus disease risk according to the fusion feature. The application can greatly improve the thymus disease prediction accuracy.