145 results about "Disease characteristic" patented technology

The invention discloses an intelligent electronic medical record generation management system and method, a terminal and a medium. The generation management system comprises a data acquisition module, a template generation module, an intelligent prompt module, an editing and saving module, a data security protection module and a permission retrieval recording module. The generation management method comprises the following steps: performing identity permission verification; integrating to form multi-modal structured data, and screening and matching to generate a template framework; a diagnosis suggestion sorting table is generated, and data is stored and backed up; transmitting and storing data, and recording operation behavior data. According to the invention, real-time automatic acquisition of basic information and examination results of patients is realized; a medical record template is automatically matched and optimized based on disease characteristics of a patient and an examination result, so that the workload of a doctor is reduced, and the medical record structure conforms to clinical specifications; transmission and storage data are encrypted in a layered mode, current account information operation behavior data are recorded based on a timestamp, and safe storage and real-time management of medical record data are effectively ensured.

The invention relates to the technical field of model construction, in particular to a construction method of a disease feature recognition and evaluation model based on the Internet of Things. The method comprises the following steps that multi-modal physiological data of a patient are collected in real time through Internet of Things medical equipment, data preprocessing is conducted on the multi-modal physiological data of the patient, standard patient physiological data are generated, and the standard patient physiological data are stored in a distributed database; extracting gene sequencing data based on the biological sample library; performing global health risk factor perception on the standard patient physiological data to generate user level health risk perception data; and carrying out feature fusion on the user level health hazard perception data and the gene sequencing data to construct a disease feature matrix containing a space-time dimension. Through multi-modal data fusion, spatio-temporal evolution modeling, deep learning architecture and personalized risk assessment, the precision and timeliness of disease feature recognition assessment model construction are improved.

The invention relates to a road dielectric constant inversion and disease discrimination method based on a neural network. The method comprises the following steps: expanding the number of jump connections based on a TransUNet baseline, introducing a gating attention mechanism, and additionally arranging a multi-scale feature fusion module, an asymmetric depth-space attention module and a deep supervision solution terminal; an AdamW optimizer and a Warmup-cosine annealing learning rate scheduling strategy are adopted to construct an adaptive region weighted loss function so as to amplify dielectric constant mutation boundary loss weight. After a model is trained through a road GPR simulation data set and parameter adjustment of the set is verified to be convergent, a to-be-detected road GPR-B-scan image is input, dielectric constant distribution is output, and according to dielectric constant characteristics of different diseases, accurate discrimination of hidden diseases such as cavities, voids, loose bodies and water-rich bodies is achieved. By adopting the method, high-resolution dielectric constant end-to-end inversion can be realized, and multi-scale disease characteristics can be accurately captured.

The invention relates to the field of biological medicine, and discloses a method and system for analyzing cough sound by using disease characteristics to diagnose respiratory diseases, and the method comprises the steps: deploying a six-microphone annular array to achieve the precise positioning and triggering of a sound source; self-adaptive spectral subtraction and Wiener filtering cascade are adopted to enhance the audio; segmenting a cough segment based on energy envelope; fusing the Mel-cepstrum, the linear prediction residual error, the harmonic energy ratio and the transient zero-crossing rate to construct a pathological feature matrix; extracting local, medium-range and global time sequence features through a three-branch parallel convolutional network; inputting a disease specific classifier to discriminate asthma, pneumonia and laryngitis respectively, and applying a feature decoupling regular term to improve interpretability. The system correspondingly realizes the modularized processing flow. According to the method, the cough sound collection quality and the disease subtype recognition accuracy in a complex environment are improved, meanwhile, the thermodynamic diagram is output to assist clinical decision making, and the diagnosis credibility and practicability are enhanced.

The invention relates to the field of forestry monitoring data processing, in particular to a forestry pest and disease damage intelligent monitoring system and method based on unmanned aerial vehicle inspection, and the system comprises a data interaction unit which is used for building a real-time communication channel between an unmanned aerial vehicle and a server through employing a WebSocket protocol; the multi-scale feature extraction unit is used for constructing a three-stage pyramid convolution structure, and respectively extracting microscopic, medium and macroscopic scale features of the blade through convolution kernels of different specifications and cavity convolution; the time sequence parameter adjusting unit is used for fusing Transform and GRU, analyzing a feature map through a multi-head self-attention mechanism, learning a tree phenological law through the GRU, performing condition normalization adjustment on convolutional layer parameters based on the tree phenological law, and distinguishing physiological and disease features; and the decision fusion unit is used for splicing and fusing the composite feature map and the time sequence feature vector, carrying out parallel processing on an SVM and a Softmax classifier, and outputting a disease and pest recognition conclusion according to a dynamic threshold decision.

The invention relates to the technical field of bridge disease diagnosis and maintenance, in particular to a bridge disease diagnosis and maintenance measure recommendation method which comprises the following steps: constructing a bridge disease database which comprises feature data and cause data of various types of bridge diseases and maintenance measure data matched with the various types of diseases; receiving field disease information of a target bridge input by a user, wherein the field disease information comprises a disease type and a disease characteristic parameter; performing matching analysis on the field disease information and data in the bridge disease database, and diagnosing a disease cause of the target bridge based on a matching result; according to the diagnosed disease causes, one or more maintenance measures corresponding to the disease causes are called from the bridge disease database and output as recommended schemes, and a complete data link from disease detection to maintenance decision is established by constructing the standardized bridge disease database and a matching mechanism.

The invention belongs to the technical field of road detection, and aims to provide a roadbed disease detection method and system, electronic equipment and a product. The method comprises the following steps: collecting multi-source detection data of a specified roadbed in real time and carrying out data fusion processing on the multi-source detection data to obtain fused detection data; performing crack detection processing on the roadbed surface image to obtain crack characteristic data of the specified roadbed, performing surface settlement analysis processing on the laser point cloud data to obtain settlement characteristic data of the specified roadbed, and performing deep disease analysis processing on the ultrasonic detection data and the electromagnetic wave reflection information to obtain the subgrade settlement characteristic data of the specified roadbed. Obtaining deep disease characteristic data of the specified roadbed; according to the crack characteristic data, the settlement characteristic data and the deep-layer disease characteristic data, constructing a disease three-dimensional model of the specified roadbed; and carrying out visualization processing on the disease three-dimensional model. According to the invention, omnibearing disease detection on the surface and the deep layer of the roadbed can be realized, and the intelligent degree is high.

The invention provides a drug-disease association prediction method and system, computer equipment and a medium, and belongs to the technical field of computers. The method comprises the following steps: constructing a drug-protein-disease heterogeneous network, and extracting a plurality of element path sub-graphs; inputting the meta-path sub-graph into a multi-scale diffusion graph convolution module, executing learnable multi-step graph diffusion on the basis of graph convolution, synchronously capturing local adjacency and high-order topological information, and generating node embedding; and performing dynamic weighted fusion by utilizing meta-path attention to obtain unified representation. In order to relieve imbalance of positive and negative samples, implementing difficult negative sampling in the embedding space, and constructing a balance training set with the positive samples; medicine-disease features are spliced, a regularization XGBoost classifier is trained, and unknown correlation accurate prediction is achieved. By adopting the method, the drug-disease association prediction precision and efficiency are improved, multi-scale topology and priori knowledge are fused, and a powerful calculation tool is provided for drug relocation.

The embodiment of the invention provides a tunnel disease detection method and system based on an unmanned aerial vehicle, and belongs to the technical field of defect optical detection. The method comprises the steps that an unmanned aerial vehicle is controlled to fly along a tunnel to collect multichannel image data of the surface of a structure, and the flight attitude is adjusted based on environment illumination information to execute image illumination compensation; performing image alignment of each target anchor point based on a structure anchor point atlas constructed based on historical acquisition images in combination with the flight pose information of the unmanned aerial vehicle and the multi-channel image data after illumination compensation; identifying a disease area in the aligned image, extracting disease features of each target anchor point at this time, and updating the disease of each target anchor point at this time into a time sequence feature data sequence corresponding to each target anchor point; and based on the updated time sequence characteristic data sequence of each target anchor point, Bayesian point change detection is adopted to analyze the disease evolution trend of the tunnel. According to the scheme, the alignment precision, the time sequence comparability and the risk judgment capability of tunnel disease detection are integrally improved.

The invention discloses a subgrade concealment disease quantitative identification method based on a ground penetrating radar, relates to the technical field of image processing, and aims to respectively establish ground penetrating radar forward simulation models for highway subgrade cavity, crack, looseness and void diseases, simulate the propagation process of radar emission electromagnetic waves in a medium, and identify the subgrade concealment disease. A B-scan radar image with clear disease types and sizes is obtained, a data set is obtained by expanding a simulation radar image, a high-quality standard data set of highway roadbed diseases is created, and quantitative recognition of the roadbed diseases is facilitated; according to the method, highway subgrade disease characteristics of different forms, different development degrees and different fillers are analyzed, the disease characteristics at least comprise cavity diseases, crack diseases, loosening diseases and void diseases, the relation between the preset disease size and the width and area in a GPR image is analyzed, and quantitative analysis of subgrade diseases is achieved. And intelligent recognition of highway subgrade diseases is realized by using a deep learning algorithm.

The invention discloses an intelligent inquiry dialogue method and device, computer equipment and a storage medium. The method comprises the steps of obtaining an inquiry dialogue of a target user through an intelligent inquiry dialogue interface; extracting disease feature information from the inquiry dialogue; querying a medical decision database based on the disease feature information to obtain a suspected disease set; calculating a suspicion degree corresponding to a suspected disease in the suspected disease set; screening a target suspected disease from the suspected disease set based on the suspicion; according to the disease feature information, determining whether the target suspected disease has an unconfirmed disease factor; if yes, generating a question generation task based on the unconfirmed disease factors, executing the question generation task through a large language model to obtain questions for the unconfirmed disease factors, and outputting the questions through an intelligent inquiry dialogue interface. According to the method, a transparent and controllable inquiry process is provided, and the risk of diagnosis errors is reduced.

The invention discloses a brain disease risk prediction method and system based on big data analysis, and belongs to the technical field of brain disease risk prediction. The method comprises the following steps: carrying out standardized preprocessing and tagged classification on brain disease related big data to generate a feature data set; mining specific disease characteristics and risk factors in the set, and carding an association rule; training a risk prediction sub-model for each disease type based on the data, and building a multi-sub-model hierarchical prediction system; and collecting to-be-predicted object data, matching a disease type, and calling the corresponding sub-model to complete risk assessment. The system comprises multiple modules for collaborative operation, and a full-process closed loop of data storage, feature processing, model management and result output is realized. According to the scheme, the pertinence, the accuracy and the efficiency of risk prediction are improved, the traceability of the whole process and the dynamic optimization of the model are realized, and reliable technical support is provided for early screening and risk early warning of brain diseases.

The invention discloses a drug-disease interaction prediction method and device, a medium and a product, and relates to the field of intelligent medical treatment. Firstly, drug data, disease data and biological data are acquired; carrying out preprocessing and feature extraction on the drug data to obtain drug molecular features; extracting target spot features based on the biological data; inputting the drug molecular features and the target features into a first full-connection neural network for predicting a drug-target interaction relationship; generating drug biological network characteristics based on the drug-target interaction relationship; performing preprocessing and feature extraction on the disease data to obtain disease features; training a second full-connection neural network based on the drug-biological network features and the disease features, and obtaining a drug-disease interaction prediction model after training is completed; the drug-disease interaction is predicted by using the drug-disease interaction prediction model, and the prediction efficiency and accuracy can be improved.

The invention relates to a road infrastructure full-automatic inspection method, device and equipment based on an unmanned aerial vehicle group, and a medium thereof. The method aims at the problem that disease characteristics collected by unmanned aerial vehicles in traffic flow and vegetation shielding scenes are broken, and the method comprises the following steps of: decoupling texture and motion characteristics of a dynamic shielding area; feature maps of the effective area and the shielding area are separated; performing cross-frame splicing and continuous repair on the fracture features by using an optical flow constraint generation network, and reconstructing the geometric integrity of a shielded region; infrared data are fused to correct thermal deformation errors, and a high-precision continuous curved surface model is generated; and finally, based on a disease quantitative feature vector matching maintenance strategy, realizing full-process automation from data acquisition to decision output. According to the method, the continuous modeling precision and the maintenance decision reliability of pavement disease detection in the shielded environment are remarkably improved.

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 a Chinese patent medicine prescription intelligent assistance method and system, and the method comprises the steps: determining a possible traditional Chinese medicine diagnosis range through disease information / patient information, carrying out the further analysis, obtaining the disease feature information / patient feature information which needs to be obtained for determining the traditional Chinese medicine diagnosis, and determining the traditional Chinese medicine diagnosis according to the related information. The Chinese patent medicine prescription intelligent auxiliary system comprises a demand acquisition module, a feature information analysis module, an information acquisition module and a traditional Chinese medicine diagnosis analysis module. According to the Chinese patent medicine prescription intelligent assistance method and system, doctors and patients can be helped to accurately select Chinese patent medicines, and damage caused by misuse of the medicines is prevented.

The invention provides a hospitalization first-time disease course record generation method and system based on a large model, and the method comprises the steps: carrying out the text segmentation, format conversion, term standardization and theme recognition through employing a historical hospitalization document record, obtaining semantic transcription data through the similarity, building a mapping relation, and carrying out the calculation of the semantic transcription data. According to the method, the medical record documents called and stored in the electronic medical record system are processed, semantic transcription data corresponding to the medical record documents are obtained, and case medical records with similarity meeting a threshold value and stored in the electronic medical record system are matched through keyword and semantic mixed retrieval according to the medical record documents and the semantic transcription data. According to the semantic transcription data and the first course record initial cue word, constructing an optimized cue word with department and disease type characteristics, and according to the optimized cue word, the semantic transcription data and the case medical record, inputting the optimized cue word, the semantic transcription data and the case medical record into a hospitalization first course record generation model trained by adopting historical data to finally obtain a hospitalization first course record. The problem that in the prior art, the accuracy of the first record content is low is solved.

The invention requests to protect a drug comprehensive evaluation system based on machine learning and an expert database, and the system comprises a multi-source data processing module which is used for processing and standardizing the molecular structure, genome characteristics and clinical data of a drug; the expert knowledge base comprises drug attributes, disease characteristics and clinical guidelines; the prediction model cluster is composed of a deep neural network, an integrated learning model and a symbol inference engine and used for predicting the IC50 value, the clinical response rate and the adverse reaction probability of the medicine; the dynamic optimization module is used for uncertainty quantification of the model, incremental learning of a training set and dynamic updating of a knowledge base; and the interpretability output module is used for interpreting the prediction result and visually outputting the prediction result.

The invention discloses a tumor patient portrait construction method and system based on multi-source heterogeneous data. The method comprises the following steps: collecting heterogeneous data related to a tumor patient; preprocessing the heterogeneous data, including format standardization, missing value completion, time sequence alignment and privacy desensitization processing, to obtain a data set which can be used for unified modeling; respectively extracting disease features, psychological features, behavior features, social features and regional features, and generating corresponding feature vectors for various features; mapping the feature vectors to a unified feature space by using a heterogeneous graph neural network based on an attention mechanism, dynamically calculating contribution degrees of different features to patient portraits, and establishing a multi-dimensional feature association graph of the tumor patients; and generating a structured tumor patient portrait including a disease dimension, a psychological dimension, a behavior dimension, a social dimension and a regional dimension. According to the method, high-precision, multi-dimensional and explainable patient feature description is realized, and a data basis is provided for precise service.

The invention discloses a drug effect prediction method for drug research and development, and relates to the technical field of drug effect analysis, and the method comprises the steps: collecting drug and target data from a plurality of data sources, and carrying out the preprocessing of the drug and target data; and carrying out feature analysis on the basis of the preprocessed drug and target data to obtain a comprehensive feature sequence, carrying out gene and pathway analysis on a disease applied by the drug, carrying out disease feature coding, and then modeling a disease background. According to the method, drug-target pairs with potential drug effects can be rapidly screened in the early stage of drug research and development through drug effect prediction, further research on a large number of invalid compounds is avoided, time and resources are saved, and then, the interaction strength between the drug and the target is predicted through the model, so that the drug effect is improved. And comprehensive analysis is carried out in combination with a disease background, so that potential drug candidate molecules can be identified more accurately, and the research and development cycle is further shortened.

The invention provides a multi-element machine learning model-based cross-species lung disease characteristic gene screening method and system, an electronic system and a storage device. The method comprises the following steps of: acquiring single cell / transcriptome data related to mouse lung diseases from a public database and preprocessing the single cell / transcriptome data; training the model by adopting six machine learning algorithms and outputting a gene importance score; calculating the weight according to the model performance and normalizing the score; and integrating the cross-species scores through a weighted fusion formula, and outputting a feature gene list and a visual report. The system comprises a data acquisition and preprocessing module, a multi-element machine learning model training module, a weight calculation and normalization module, a cross-species comprehensive scoring module and a result output module. The screening accuracy, stability and generalization ability are improved through multi-algorithm integration and cross-species fusion, and the method can be widely applied to the fields of mechanism research of lung diseases, diagnosis marker development and drug target verification.

The invention relates to the technical field of tunnel safety, and discloses a safety detection method and system for a large-span multi-arch tunnel, and the method comprises the following steps: a preventive monitoring step: carrying out the dynamic monitoring of a middle rock pillar, a double-hole junction section and a vault region of the multi-arch tunnel through a monitoring device, and collecting the deformation data and stress data of surrounding rock and a structure; a disease detection step: detecting the tunnel lining and the surrounding rock by adopting a mode of combining surface detection and hidden detection to obtain surface disease characteristics and hidden disease characteristics; and a data fusion step: carrying out fusion processing on the deformation data, the stress data, and the surface disease characteristics and the hidden disease characteristics obtained in the disease detection step. The deformation data and the stress data are obtained by dynamically monitoring the middle rock pillar, the double-hole junction section and the arch crown area, and the problem that the tunnel safety state evaluation is not comprehensive due to the fact that a single-dimension monitoring or detection mode is mostly adopted in traditional tunnel safety detection is solved.

The invention discloses a lung function evaluation method for a patient in a pneumonia rehabilitation period, and relates to the technical field of rehabilitation management. Extracting a lung function feature set based on the image data; analyzing based on multi-dimensional parameter association, and establishing a dynamic mapping relationship between the lung function feature set and associated physiological parameters; identifying abnormal associated parameters through a dynamic mapping relation between the lung function feature set and the associated physiological parameters, and when the deviation between the image feature recovery degree and the corresponding physiological parameter recovery degree exceeds a preset threshold value, judging that functional abnormal parameters exist; and matching a potential complication database based on the abnormal parameter type, analyzing the relevance between the abnormal parameter and a preset disease feature model, and outputting secondary disease risk early warning. The method has the advantages that through cross-modal fusion of the lung image features and the dynamic physiological parameters, the early recognition precision of the pneumonia convalescence complication is remarkably improved.

The invention provides a plant disease feature extraction and classification method and system based on deep learning, and relates to the technical field of plant disease identification, and the method comprises the steps: mining the interdependence rule of different disease features through constructing a disease feature symbiosis wake-up network, and setting wake-up conditions; inputting a to-be-analyzed plant disease image into the network, capturing initial disease features, triggering associated feature wakeup, and generating a disease feature symbiotic set; conducting the characteristic information according to the hierarchical relationship to generate enhanced symbiotic disease characteristics; recording morphological change details to construct a disease characteristic evolution sequence; and inputting the disease characteristic evolution sequence into a pre-trained deep learning classification model, analyzing disease essential characteristics, and generating a classification result containing a disease type and a matching basis. The method can comprehensively and accurately extract features, and improves the accuracy and reliability of plant disease classification.

The invention relates to the technical field of intelligent disease and insect pest recognition, and discloses an improved ResNet50-based agricultural disease and insect pest recognition method and system, and the method comprises the steps: recognizing the crop variety of an agricultural crop and the corresponding disease and insect pest type; an intelligent monitoring network of agricultural planting crops is set, and pest and disease damage characteristics of the agricultural planting crops are extracted; inputting the pest and disease damage characteristics and the pest and disease damage types into a pre-trained ResNet50 network model as input data, so as to identify disease crop characteristics of the agricultural planting crops through an attention mechanism in the ResNet50 network model; according to the characteristics of the diseased crops, extracting key disease areas of the agricultural planting crops by using a disease and pest behavior detection mechanism of a ResNet50 network model, and analyzing disease causes of the diseased crops; and in combination with the key disease areas and disease causes, generating disease and pest identification results of the agricultural planting crops. The accuracy and efficiency of agricultural pest identification can be improved.

The invention relates to the technical field of medical data management, and discloses a GIST special disease database intelligent management system, which comprises a multi-source data acquisition module, a standardized processing module, an intelligent analysis module and a security sharing module, the multi-source data acquisition module is in communication connection with a hospital information system H IS, a laboratory information system L IS, an image archiving and communication system PACS and a patient end application, and is used for acquiring electronic medical records, inspection reports, image data and patient active report data; a standardized data model is built in the standardized processing module, and the standardized processing module is used for performing field mapping and format conversion on the acquired multi-modal data; the intelligent analysis module comprises a machine learning algorithm engine. According to the medical data management system, the multi-source data acquisition module is in communication connection with the hospital information system and the laboratory system, so that the problem that most traditional medical data management systems can only acquire single-source or single-mode data, and consequently disease characteristics cannot be comprehensively reflected is solved.

The invention belongs to the technical field of disease detection, and discloses a rice multi-disease detection method based on improved YOLOv11n, a novel model MSAF-YOLO obtained by improving a YOLOv11n model structure is adopted, the novel model MSAF-YOLO uses a CSPC module to replace an SPPF module in a backbone network part, uses an ACM module to replace a C3k2 module in the backbone network part, and constructs a CSMS-BiFPN as a neck architecture; the rice multi-disease detection method comprises the following steps: S1, image acquisition; s2, preprocessing the image and constructing a data set; and S3, inputting the data set into the novel model MSAF-YOLO for training, and evaluating the model. According to the invention, the capturing capability of the model for disease characteristics of different scales is enhanced, the sensitivity of the model to tiny disease spots is improved, and background noise is effectively inhibited. While the high and low layer feature fusion efficiency is optimized, the number of parameters and the calculation overhead are greatly reduced.

The invention discloses a drug-disease association prediction method, system, device and medium, and relates to the technical field of drug relocalization, and the method comprises the steps: constructing a drug similarity network and a disease similarity network; obtaining a binary adjacency matrix through k-neighbor graphs of different nodes in the two similarity networks, carrying out weighted fusion on the binary adjacency matrix through an attention coefficient to obtain a soft adjacency matrix, and carrying out fine-grained graph convolution updating to obtain drug features and disease features; extracting heterogeneous node representations of drugs and diseases in the biochemical heterogeneous network, and carrying out dynamic weight distribution on different representations to obtain final embedding of drug nodes and disease nodes; and splicing the final embedding of the drug nodes and the final embedding of the disease nodes, and performing prediction according to the spliced features to obtain the drug-disease association probability. According to the method, the potential complementary relationship between the two is fully mined, and the heterogeneous feature fusion effect is improved, so that the performance and robustness of drug-disease association prediction are integrally enhanced.

The invention relates to an intelligent internet bed reservation system, which comprises a diagnosis and treatment group intelligent management module for receiving disease characteristic data of a patient, generating a diagnosis and treatment group distribution scheme and outputting an operation schedule table to a dynamic scheduling engine; the dynamic scheduling engine module is used for generating time window prediction according to the operation scheduling table and starting a grading notification process; the notification and interaction module is used for generating patient customized notification content, feeding back a confirmation state to the patient management module and updating the hospitalization preparation progress; and the patient full-period management module is used for generating an electronic hospitalization certificate and optimizing a period prediction algorithm. According to the system, method and device for reserving the bed through the intelligent internet, the processor and the computer readable storage medium, the working efficiency of medical staff is improved, the patient receiving and treating plan is flexibly adjusted, and time is saved. The queuing condition can be checked at any time, and the patient can conveniently and reasonably arrange personal affairs.

The invention provides a disease portrait generation method and device, electronic equipment and a storage medium, and belongs to the technical field of artificial intelligence, and the method comprises the steps: constructing a feature association map according to multi-source data; inputting the atlas into a hierarchical graph attention network, learning a direct association and combination relationship of feature nodes by using a hierarchical structure, and obtaining an attention weight; candidate key feature nodes are screened accordingly; performing causal effect estimation on the candidate key feature nodes to obtain a causal effect value representing real causal intensity; and finally, based on the attention weight and the causal effect value, determining an association level and generating a disease portrait. According to the method, the depth feature mining capability of the hierarchical graph attention network and the counterfeit elimination and true storage capability of causal inference are fused, so that automatic discovery of the high-order combination relationship of the disease features and quantitative verification of the true causal are realized, and a dynamic disease portrait with depth correlation information and high-credibility causal interpretation can be generated.