57 results about "Diagnosis recommendations" patented technology

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.

The invention discloses a nervous system disease auxiliary evaluation method based on a large model and related equipment. The method comprises the following steps: constructing a structured knowledge graph based on multi-modal heterogeneous data of nervous system disease guide texts, symptom descriptions and examination results; a graph neural network module is embedded based on the structured knowledge graph in combination with a large language model, an inference network model of a dynamic graph structure is generated, and a dynamic diagnosis path tree is formed by calculating the relation weight between the medical nodes in the graph; and generating interpretable diagnosis suggestions through the reasoning network model according to the current symptom description, the examination result and the historical medical history of the patient. The problems that an existing model still has obvious bottlenecks and clinical challenges, depends on single modal data, is insufficient in quantification of dynamic signs such as a reasoning level, tremor frequency and gait abnormity, does not achieve disease development time sequence modeling and is difficult in clinical conversion can be solved.

The invention discloses a cervical lesion intercellular relation modeling and analysis system based on a graph neural network, and the system comprises a medical image collection module which is used for collecting a digital image of a cervical tissue pathological section or a cervical TCT slide; the cell detection and segmentation module is used for extracting spatial position information and morphological characteristics of cells; the cell feature extraction module is used for extracting and fusing the spatial position, morphology, texture and biological marker features of the cells; the cell relation graph construction module is used for constructing a heterogeneous cell relation graph with cells as nodes and inter-cell relations as edges; the graph neural network analysis module is used for carrying out feature learning and modeling on the heterogeneous cell relation graph; the intelligent auxiliary diagnosis module is used for generating auxiliary diagnosis suggestions; and the data management and automatic control module is used for realizing automatic control and case data management of the whole process of the data. The intelligent and automatic level of cervical lesion cell analysis can be comprehensively improved, and the accuracy and efficiency of diagnosis are improved.

The invention relates to the technical field of medical image diagnosis, and discloses a cardiovascular disease diagnosis model construction method based on image processing. The method comprises the steps that cardiac medical image data of a target patient is collected, a standardized data set is generated through preprocessing, and a morphological and hemodynamic feature set is extracted; establishing a heart state evolution characteristic spectrum according to a characteristic dynamic evolution rule, dividing a pathological state space, and calculating the characteristic distribution density of a historically diagnosed case; acquiring real-time image data of a patient to be diagnosed, and constructing a real-time diagnosis feature vector; mapping the vector to a pathological state space, and calculating a space matching degree to generate a pathological association index; and combining the association index and the two types of feature sets to construct a heart pathology probability prediction model, outputting a pathology probability prediction value and generating a hierarchical diagnosis suggestion. According to the method, through multi-dimensional feature analysis and space matching analysis, precise and graded diagnosis of the cardiovascular diseases is realized, and an efficient and feasible technical path is provided for diagnosis of the cardiovascular diseases.

The invention relates to the field of rail transit vehicle intelligent diagnosis, in particular to a train fault intelligent auxiliary diagnosis method fusing multi-source knowledge, which comprises the following steps of: constructing a multi-source knowledge document library, and performing text cleaning, segmented disassembly and vectorization processing to obtain a vector database; then obtaining a fault alarm list and capturing event information, and performing diagnosis according to the vector database to obtain a diagnosis suggestion; then time sequence operation parameters before and after the fault are extracted and analyzed, and a trend chart and characteristic indexes are generated to obtain an analysis result; constructing a multi-dimensional auxiliary analysis mechanism based on the diagnosis suggestion and the analysis result, and generating a maintenance suggestion in combination with historical maintenance work order information; and finally, generating a comprehensive auxiliary diagnosis report according to the maintenance suggestion, automatically generating a maintenance dispatching task list, determining a task execution sequence according to a preset priority scheduling algorithm, and executing the maintenance dispatching task list. The accuracy and timeliness of train fault diagnosis can be improved, and the safety, the operation and maintenance efficiency and the intelligent level of rail transit equipment are improved.

The invention relates to a voice interaction large model family health assistant dialogue method and device, equipment and a medium. The method comprises the following steps: carrying out fragmentation processing according to an original voice stream of a user to generate an audio fragment with a medical mark, and carrying out voice recognition and entity extraction on the audio fragment to generate a dynamic entity map; generating an evidence-based decision prompt based on the map, inputting the prompt into a preset medical big model for processing, and outputting a result containing an essential symptom list; and according to the symptom matching degree of the necessary symptom list and the dynamic entity map, generating a diagnosis report or a question-asking list, if the diagnosis report is output, performing medical rule chain verification operation on the diagnosis report to generate a quality control report, and based on the question-asking list or the quality control report, generating a synthetic voice stream. According to the method, through medical intention directional screening, map entity analysis, large model diagnosis, voice synthesis and the like, the voice recognition accuracy, the diagnosis suggestion reliability and the inquiry interaction efficiency of the family health assistant in the medical scene are improved.

The invention relates to the technical field of medical health intelligent assessment, in particular to a patient nutrition risk intelligent assessment system based on multi-parameter fusion. The system comprises a data acquisition module for acquiring nutrition-related indexes of a patient; the conflict detection module is used for establishing a parameter logic constraint rule base and generating a conflict parameter matrix; the evidence fusion module is used for carrying out fusion processing on the contradictory evidence by adopting an improved Dempster-Shafer evidence theory; the knowledge distillation module is used for realizing parameter complementation through a multi-hop propagation inference mechanism of a nutrition parameter association diagram; the uncertainty quantization module is used for establishing a causal inference graph to quantify the contradictory contribution degree; and the diagnosis suggestion module is used for generating clinical examination suggestions. According to the method, the problem that a traditional method cannot be used for evaluation due to nutrition parameter deficiency and metabolic index contradiction of a patient is solved, and the accuracy and practicability of nutrition risk evaluation are improved through deficiency parameter inference, contradiction evidence fusion and causal root tracing.

The invention relates to the technical field of medical information processing, and discloses an acute pulmonary embolism auxiliary diagnosis decision-making system based on a large model, and the system comprises a data collection and preprocessing module which is used for receiving multi-source clinical data and preprocessing the multi-source clinical data; the large model reasoning module is used for inputting the preprocessed multi-source clinical data into a large model fused with a multi-modal causal attention and identification adversarial learning mechanism, and outputting a reasoning result; the dynamic learning module is used for accessing desensitized case data of a hospital electronic medical record system, continuously learning new cases through federal learning, and optimizing the recognition capability of a large model pair; the diagnosis suggestion generation module is used for generating structured diagnosis suggestions according to the reasoning result output by the large model; the result output and interaction module is used for displaying diagnosis suggestions to doctors and updating reasoning results in real time by a large model when the doctors supplement data; the diagnosis efficiency is improved.

A diagnostic device (10) for guiding a doctor in performing diagnostic procedures on a patient's body comprises a main body (12) having a top surface (13) and a bottom surface (14), with one or more user output modules (22) positioned on the top surface (13) and a set of sensors comprising peripheral sensors (16) and diagnostic sensors (18) arranged on the bottom surface (14). A processor (302) executes instructions stored in a memory (306) for receiving and processing preliminary inputs about the patient through a preliminary data acquisition module (307), determining a target body region and starting point, and performing scans to determine boundaries and examine specific organs. The device employs an artificial intelligence / machine learning (AI / ML) engine (310) to generate real-time audio, visual, and haptic instructions for guiding the operator, process multiple data streams, identify abnormalities, and provide diagnostic recommendations.

The invention provides an intelligent processing system for hospital medical record informatization management, and the system comprises a data storage module which is used for carrying out the butt joint with an existing information system of a hospital through an API gateway, constructing an original database through cloud object storage, and carrying out the centralized storage of structured medical record data and unstructured medical record data; and the term standardization module is used for mapping the medical terms in the medical record to a national standard medical term set. According to the method, automatic analysis, recognition and relation extraction are performed on the text information in the medical record through the medical natural language processing engine, and the text information is filled into the structured data model, so that the medical record information is standardized, and a good basis is provided for clinical scientific research and hospital management. By integrating patient information, historical medical records and medical knowledge maps, diagnosis suggestions and safe medication early warning are provided for doctors in the diagnosis and treatment process, the scientificity of clinical decision making is enhanced, medical risks are reduced, and patient safety is improved.

The present invention relates to the technical field of artificial intelligence and medical decision-making systems, and discloses a method and system for constructing a reasoning agent for intelligent medical guidance. The method comprises: constructing a hierarchical modality completion network and a modality correlation knowledge graph; constructing a medical feature cross-modality mapping network based on contrastive learning to map medical data of different modalities to a shared feature space; constructing an uncertainty quantification model and automatically adjusting the diagnostic confidence threshold based on the uncertainty of the completed data; constructing a disease modality association matrix and a modality reliability assessment network for different disease types and symptom combinations; and realizing cross-validation and collaborative decision-making of multi-specialty knowledge through weighted voting, evidence aggregation and specialist authority assessment to form a final diagnostic recommendation. The present invention solves the problem of limited diagnostic capabilities caused by missing medical data in an environment with limited medical resources.

The invention provides an edge-end collaborative medical image diagnosis method based on a graph theory and deep reinforcement learning. The method is used for solving the problems that in edge-end collaborative reasoning scheduling of a medical image DAG topology DNN, the structure is complex, segmentation is difficult, and the scheduling efficiency is poor. The method comprises the following steps: firstly, converting a complex DAG structure into a chain structure through topological sorting to obtain a CDNN structure; meanwhile, for an end-side collaborative system model, the number of times of floating-point operation per second is used for representing computing resources of user side equipment and an edge server; then, constructing a scheduling unloading model based on a CDNN structure, and carrying out MDP modeling on the scheduling unloading model to minimize inference delay and energy consumption; thirdly, a DP-A3C strategy is put forward for optimization solution, and an optimal model division strategy is obtained; and finally, the model division strategy is fed back to user side equipment, and rapid and effective diagnosis suggestions are provided for doctors to process complex tasks. The method can adapt to different network topologies and task requirements, and provides stable performance.

The invention discloses an ear-nose-throat lesion intelligent auxiliary diagnosis method and system based on multi-modal fusion, and belongs to the technical field of medical auxiliary diagnosis medical record management. The method comprises the following steps: firstly, establishing an ear-nose-throat electronic medical record library containing pathological feature words and pathological images in a pathological stage, and uniformly coding; then, through a bimodal architecture processing method, early, middle and advanced lesion relation chains are constructed and matched to form a medical record sample library, and an electronic medical record chain cluster is formed through multi-part chain selection; and finally, screening an effective diagnosis case based on the lesion relevance of the pathological image, and quantifying an auxiliary diagnosis recommendation weight of the lesion relation chain. The system correspondingly comprises an electronic medical record library construction module, a bimodal architecture processing module and a diagnosis support module. According to the method, the limitation of single-part / stage diagnosis is broken through, multi-dimensional medical record data association is realized, the accuracy and efficiency of auxiliary diagnosis are improved, quantifiable diagnosis reference is provided for medical personnel, and subjective errors are reduced.

The application discloses a fusion enterprise training course intelligent diagnosis recommendation system of deep learning, including: including: course data multidimensional acquisition module, student behavior characteristic extraction module, training demand intelligent diagnosis module, deep learning course matching module, recommended result dynamic optimization module, system running state monitoring module; through multidimensional acquisition of internal and external courses and student behavior data, deep pattern and preference are mined through feature extraction, knowledge short board and potential demand are determined through hierarchical analysis of the training demand intelligent diagnosis module, a multidimensional correlation model is constructed through the deep learning course matching module to realize accurate matching, the recommended result dynamic optimization module is combined to iteratively adjust the scheme according to real-time feedback, and the system running state monitoring module guarantees stable and efficient operation of each link. The system effectively improves recommendation accuracy and dynamic adaptability, improves training resource utilization, and meets the personalized and efficient demand of enterprise talent training.

The application relates to a large model family health assistant dialogue method, device, equipment and medium for voice interaction. The method comprises the following steps: generating an audio slice with a medical mark according to user original voice stream slice processing, and performing voice recognition and entity extraction to generate a dynamic entity graph; generating an evidence-based decision prompt based on the graph, inputting the prompt into a preset medical large model for processing to output a result containing a required symptom list; generating a diagnosis report or a follow-up question list according to the matching degree of the required symptom list and the symptoms of the dynamic entity graph, performing a medical rule chain verification operation on the diagnosis report to generate a quality control report, and generating a synthesized voice stream based on the follow-up question list or the quality control report. The method improves the voice recognition accuracy, diagnosis suggestion reliability and inquiry interaction efficiency of the family health assistant in the medical scene through medical intent directional screening, graph entity analysis, large model diagnosis and voice synthesis.

The invention relates to a multi-round dialogue type intelligent diagnosis interaction system, in particular to a cervical cell panorama-oriented multi-round dialogue type intelligent diagnosis interaction system. The objective of the invention is to solve the problems that existing diagnostic information is single, insufficient in explanatory property and poor in interaction flexibility, and finer diagnostic information is difficult to obtain; the problems that multi-round interaction and context memory ability are lacked, and continuous and coherent response cannot be made in combination with historical information are solved. Comprising an image block acquisition module used for acquiring a pre-processed full slice image WSI; the image segmentation and effective area screening module is used for obtaining effective image blocks; the image-text pair construction and pathological attribute acquisition module is used for forming attribute vectors; the context-aware multi-round question and answer module is used for obtaining a final answer; and the final diagnosis suggestion generation module is used for obtaining a final diagnosis suggestion based on knowledge graph reasoning and probability model prediction. The method is applied to the field of multi-round dialogue type intelligent diagnosis interaction.

The invention relates to the technical field of veterinary intelligent diagnosis, and discloses a beef cattle disease diagnosis-oriented hierarchical multi-dimensional knowledge retrieval method, which comprises the following steps of: firstly, receiving and analyzing a diagnosis query text in a natural language form, and extracting a symptom entity set and a background information set; semantic matching is carried out in a low-layer entity set of the layered beef cattle disease knowledge graph based on the symptom entity set, and a candidate disease entity set is reversely derived through a cause relation and a differential diagnosis relation in the graph; and respectively calculating the interpretation coverage degree of the candidate diseases on symptoms and the consistency degree of epidemiological attributes and background information of the candidate diseases, sorting the candidate diseases through multi-dimensional weighted scoring, and generating a structured diagnosis suggestion containing an evidence chain in combination with a reasoning path. According to the method, the problem of misdiagnosis caused by a traditional retrieval method is solved by fusing the clinical representation and the individual environment background information, and the accuracy and interpretability of beef cattle disease diagnosis are improved.

The invention discloses an intelligent image auxiliary diagnosis method and system based on multi-modal fusion and incremental learning, and the method comprises the steps: obtaining a multi-modal medical image of a patient, and generating a time-space aligned multi-source medical image data set through the multi-modal registration and fusion technology; performing focus region segmentation and feature extraction on the multi-source medical image data set, and constructing a plurality of omics feature vectors including morphological features, textural features and functional features; inputting the multiple omics feature vectors into an intelligent diagnosis model based on incremental learning, and outputting benign and malignant probabilities of a focus area and an anatomical structure accumulation range prediction result through a dynamically updated pathology classifier; and generating personalized auxiliary diagnosis suggestions and follow-up monitoring strategies based on benign and malignant probabilities and anatomical structure affected range prediction results. According to the embodiment of the invention, deep feature extraction and dynamic knowledge evolution of the multi-source medical image can be realized, and the accuracy and comprehensiveness of auxiliary diagnosis are effectively improved.

The invention relates to the technical field of artificial intelligence, and discloses an intelligent inquiry method and device, equipment and a medium, and the method comprises the steps: obtaining disease symptom information inputted by a user, and carrying out the feature extraction of the disease symptom information, and obtaining a target feature vector; performing unified embedding representation on the language feature vector, the text feature vector and the image feature vector through a multi-modal alignment model to obtain a standardized illness state representation vector; analyzing the standardized illness state representation vector through an illness state reasoning model to obtain a diagnosis suggestion text; and analyzing the diagnosis suggestion text through an intelligent inquiry model, and outputting virtual image information and tone output content corresponding to the diagnosis suggestion text. The method can be applied to a medical care health program system, and the increasing intelligent and humanized diagnosis and treatment requirements in medical health and care scenes can be met.

The invention discloses a multi-agent-based chronic disease risk management and hierarchical diagnosis decision-making method, and particularly relates to the technical field of artificial intelligence. A multi-agent collaborative network is constructed, and multi-source data acquisition and preliminary risk assessment are performed on chronic disease patients; the multi-agent collaborative network comprises a patient terminal agent, a community screening agent and a hospital specialized agent, a private type space, an action space and a utility function of the agents are defined, a total medical resource constraint is set, a Bayesian game model of collaborative diagnosis is established, and when a risk assessment result reaches a first-level early warning threshold value, a first-level early warning model is established. And starting an event-driven and consensus-oriented multi-round iterative negotiation mechanism, searching game equilibrium by utilizing multi-round iterative negotiation and strategy optimization so as to generate an optimal hierarchical diagnosis suggestion, carrying out quantitative evaluation on behaviors of each agent in a Bayesian game collaborative diagnosis decision process, and generating a behavior evaluation report.

The application discloses a kind of based on multi-modal fusion's diabetic retinopathy auxiliary diagnosis method, comprising the following steps: S1, the fundus image of patient is preprocessed operation;S2, the text information of patient is preprocessed operation;S3, the feature extraction operation is carried out to the fundus image after preprocessing, and image feature vector is obtained;S4, the feature extraction operation is carried out to the text information after preprocessing, and text semantic vector is obtained;S5, image feature vector, text semantic vector are combined with medical mask gate and carry out high-order sparse tensor fusion, and obtain the feature vector after fusion and diagnostic category prediction probability;S6, the diagnosis result is combined with bidirectional back-propagation mechanism and carries out graph-text joint tracing, and outputs diagnosis heat map and its corresponding dynamic linkage explanation.The application can provide fast diagnosis suggestion for doctors, reduce artificial reading film pressure, reduce misdiagnosis false negative rate, accelerate the retinal screening efficiency of diabetic patient.

The invention provides a method for automatically generating an oral image report, which comprises the following steps of: receiving and processing oral medical image data; classifying the processed oral medical image data, and performing matching based on the classification result and pre-stored oral illness state text data; wherein the oral disease condition text data comprises a disease description text and a diagnosis suggestion text; correlating corresponding image information and text information based on the matching result, and generating a standardized diagnosis report corresponding to the oral medical image; detecting and correcting potential errors in the standardized diagnosis report according to expert knowledge base data, and generating a target report; wherein the expert knowledge base data comprises morphological characteristics of an anatomical structure, disease characteristics and diagnosis suggestions. According to the invention, the standardization level of oral medical diagnosis is improved, the report is more intuitive, and the workload of doctors is reduced.

The invention discloses a medical resource recommendation method and device, equipment, a medium and a program product, and relates to the technical field of recommendation. The medical resource recommendation method comprises the following steps: on a user interface of a first application, in response to a first input operation of a user, obtaining a medical preference of the user; on the user interface, in response to a second input operation of the user, obtaining a diagnosis suggestion; one or more predicted causes are included in the diagnostic suggestions; in response to a first confirmation operation of the user, displaying a medical institution recommendation list on the user interface; wherein the medical institution recommendation list is a medical institution and a doctor of the institution screened out from the first recommendation list according to the medical seeking preference, and the first recommendation list is determined according to each predicted cause of disease. In this way, the recommended medical institution can give consideration to scientificity and user experience.

The invention relates to the technical field of internal fistula safety detection, and discloses an internal fistula safety detection system and method.Image data of an internal fistula area are obtained through an image obtaining module, and physiological data of a patient are monitored in real time; the image processing module is used for processing image data, extracting internal fistula features and judging the existence and severity of internal fistula, so that the diagnosis accuracy and efficiency are improved; the recognition and diagnosis module automatically recognizes internal fistula features and predicts changes of internal fistula, valuable reference information is provided for doctors, and making of a more accurate treatment scheme is helped; the interactive display module visually displays the existence, severity and change of the internal fistula, and generates a diagnosis result and a diagnosis suggestion; and the feedback adjustment module dynamically adjusts an internal fistula judgment strategy and a prediction strategy according to an actual diagnosis result, so that the diagnosis and prediction accuracy is improved.

The invention discloses an auxiliary diagnosis device and method for abnormal heart blood flow and a storage medium. The auxiliary diagnosis device for the abnormal heart blood flow comprises a monitoring control module used for monitoring whether a patient to be detected executes a Valsalva action or not in real time, and if yes, a pulse signal is sent; the image acquisition module is used for acquiring an image sequence of the to-be-detected patient in a right atrium pressure rise period based on the pulse signal, and performing target ROI extraction on the image sequence to obtain a target image sequence for displaying the target ROI; the light spot recognition module is used for detecting a microbubble signal in the target ROI of each frame of target image in the target image sequence in real time and counting the number of microbubbles in the target ROI of each frame of target image; the decision module is used for generating a corresponding diagnosis suggestion based on the number of microbubbles in the target ROI of each frame of target image; the full-process automation is realized, and the accuracy of heart blood flow abnormality diagnosis is improved; the diagnosis time is shortened, and the diagnosis efficiency is improved.

This invention provides a management system for WSI pathology reports based on OFD-H. It includes a presentation layer, a business layer, a data storage layer, and a standard interface for interoperability with third-party medical systems. The presentation layer includes an editor for editing OFD reports and a WSI image reader capable of coordinate backtracking of WSI images. The business layer includes: module one for automatically generating pathology image heatmaps and textual diagnostic suggestions; module two for OFD format container encapsulation; module three for three-level streaming image backtracking; and module four for adding digital signatures and watermarks. The data storage layer is used for hierarchical storage of WSI files, ROI images, and OFD documents to meet the retrieval needs of massive multimodal data. This invention can achieve "intelligent assisted generation" of reports, "two-way interoperability" between text and images, "lightweight encapsulation" of data, and "compliant traceability" of documents.

The invention relates to the field of case management and artificial intelligence application in automobile testing, and provides an automobile bus testing AI diagnosis system and a closed-loop feedback and self-learning method thereof. The method comprises the following steps: S1, generating a fault diagnosis suggestion by a system; s2, collecting maintenance feedback data; s3, causal features are extracted, and two causal feature vectors are formed; s4, generating a causal feature differentiation code; s5, performing incremental training and fine tuning on the system model; and S6, performing fault diagnosis by using the fine-tuned system model. The system comprises a fault diagnosis module, a maintenance feedback module, a causal feature extraction module, a differential coding module and an incremental training module. According to the method, the diagnosis accuracy is improved along with the use times through closed-loop feedback and a self-learning mechanism; according to evaluation, the diagnosis accuracy can be improved by 35% within 3 months; the sustainable evolution capability enables the system to continuously adapt to the specific test environment and fault mode of an enterprise, and provides more and more accurate diagnosis suggestions.

The application discloses an aquatic pet cloud primary screening and on-site treatment whole-cycle management system and method based on artificial intelligence, which comprises an offline service terminal used for collecting multi-modal data of aquatic pets and performing on-site service operations; a cloud hub system is in communication connection with the offline service terminal through a mobile communication network, is used for receiving offline service terminal data and performing data processing, decision-making and resource scheduling processing, and comprises an AI computing cluster, which is used for analyzing the multi-modal data to generate diagnosis suggestion information; an intelligent service scheduling module is used for acquiring the diagnosis suggestion information and on-site resource states fed back by the offline service terminal, generating and issuing a service execution scheme to the offline service terminal; and the offline service terminal receives the service execution scheme and performs corresponding on-site service operations. The application constructs an aquatic pet multi-modal AI primary screening diagnosis module based on data resources accumulated by a national aquatic animal pathogen library, and significantly improves disease identification accuracy and diagnosis efficiency.

The embodiment of the invention discloses a data analysis method based on multi-modal data fusion and related equipment, and relates to the technical field of machine learning and data analysis, and the method comprises the steps: obtaining the chief complaint data of a target person through a natural language processing technology according to the voice information of the target person; analyzing the medical image information of the target person through a preset deep learning model to obtain a radiology report of the target person; inputting the chief complaint data and the radiology report into a primary diagnosis model based on a single-layer multi-head attention mechanism to obtain a primary diagnosis result; acquiring historical doctor seeing data of the target person, and constructing a historical case knowledge graph based on the historical doctor seeing data; performing embedded learning on the preliminary diagnosis result and the historical case knowledge graph by using a diagnosis model to obtain a diagnosis suggestion; and inputting the diagnosis suggestions into a domain expert knowledge base so as to verify the diagnosis suggestions.