132 results about "Medical model" patented technology

The invention relates to the technical field of medical information processing, in particular to a health data processing method and system based on a distributed ledger library, and the method comprises the steps: a node firstly links a health data abstract to obtain a differential privacy budget token and a reversible tensor hash key; then, carrying out encryption training on local model parameters by using the key, injecting Gaussian noise according to budget to generate a differential protection gradient, and carrying out uplink together with zero-knowledge proof; the account book end decrypts the threshold value, uses a structural equation model to deduce a causal correction matrix according to the reputation weight aggregation gradient, and automatically adds a budget and turns the key when the concept drifts and the budget is insufficient; the operation mechanism splices the aggregation parameters and local parameters, generates a gating vector in combination with a causal correction matrix and a reputation weight, outputs disease risk prediction, and only uploads prediction hash and error information; chain-level audible privacy protection, dynamic budget management and hybrid deviation suppression are realized, and the safety and accuracy of a cross-institution medical model are improved.

The invention discloses a retrieval enhancement generation method and device based on a medical knowledge fine-tuning language model. The method comprises the steps of constructing a medical knowledge base, performing fine-tuning on the medical language model, inputting medical questions by a user, retrieving the knowledge base, screening knowledge fragments, generating answers and the like. And the medical knowledge base is constructed through semantic partitioning, so that the integrity and continuity of related knowledge are ensured. And a medical knowledge fine tuning model is used as a screener to further screen high-quality knowledge fragments most related to user questions, so that the professionality and reliability of generating answers are improved. The method is suitable for the medical field, and can provide more accurate and more professional medical consultation services for patients and doctors. The fine-adjusted medical model is innovatively used as a filter, knowledge fragments are evaluated from multiple dimensions, it is ensured that the knowledge fragments are highly related to questions, powerful support is provided for generating professional and accurate medical answers, and the precise medical consultation requirement of a user is better met.

The invention relates to the technical field of medical image intelligent analysis and artificial intelligence auxiliary diagnosis, in particular to a medical image disease course prediction system based on an industrial neural network, and the system comprises a reference generation module which is used for obtaining static image data; processing the static image data by using a physical perception neural network to generate a pure ideal state reference; a perturbation simulation module; the industrial kinetic parameters are used as perturbation terms to be superposed to a pure ideal state reference, and a theoretical damaged state is generated; the projection verification module is used for acquiring real multi-modal observation data; generating a real residual error; generating a theoretical residual error based on the theoretical damaged state and the pure ideal state reference; calculating a manifold coupling confidence coefficient; the closed-loop correction module is used for performing inversion optimization on the industrial kinetic parameters; outputting a disease course prediction result according to the manifold coupling confidence coefficient; according to the method, the problem that a traditional medical model lacks physical consistency explanation is solved, and the credibility of artificial intelligence auxiliary diagnosis is remarkably improved.

The invention relates to the technical field of intelligent medical treatment, and discloses a medical large model driven cross-department collaborative prescription generation method and system. According to the method, electronic medical records of patients and prescription data of multiple departments are acquired, key information is structurally extracted to generate directional vectors, and transfer learning and fine adjustment are performed by using historical conflict cases and a drug knowledge graph based on a medical basic large model and an AI chip, so that a collaborative prescription model is constructed. The model can identify drug incompatibility and dosage risks among departments and generate a prescription suggestion set. A doctor can dynamically correct a prescription based on feedback, a collaborative report containing a medication time sequence, a monitoring index and an emergency scheme is generated after multiple rounds of collaborative optimization, and the safety and effectiveness of multi-department combined medication of complex diseases are remarkably improved.

The invention discloses a medical model training system and method for protecting medical data privacy. The system comprises a central server and clients of all participants participating in federated learning. The client is used for training a local model by adopting local data; in the training process, a local difference privacy technology and target noise intensity are adopted to carry out noise disturbance on the gradient of each time of training; under the condition that training is finished, gradient updating of the local model is determined, and gradient data are obtained; encrypting the gradient data by adopting a homomorphic encryption technology, and sending the encrypted gradient data to a central server; the central server is used for globally aggregating the received encrypted gradient data in a pre-created trusted execution environment; in the global aggregation process, a global differential privacy technology is adopted to add noise to aggregated model parameters, model global parameters are generated, and the model global parameters are broadcasted to each client. By adopting the embodiment of the invention, the risk of medical data leakage can be reduced.

The invention discloses a full-cycle path chronic disease management system and method based on artificial intelligence, and belongs to the technical field of intelligent medical treatment, and the method comprises the steps: generating a model result based on an artificial intelligence chronic disease risk prediction model and a prescription through an active diagnosis and treatment module, and carrying out the active recognition, intervention and management of a target group; the personalized diagnosis and treatment module is used for providing refined follow-up visit, prescription and screening services according to health states and risk characteristics of different individuals; and semantic search, index generality identification and multi-source data integration capabilities of clinical data are provided through an intelligent data governance and decision support module. According to the invention, an intelligent medical new mode of active, personalized and intelligent chronic disease management is constructed, clinical decision is assisted, the management efficiency is improved, the chronic disease management level is improved, reasonable flow of medical resources is promoted, and chronic disease prevention, management and referral are promoted to develop towards the full-life-cycle management direction.

Systems and methods for computer modeling in medicine. A sort of period table of medical models is described for personalized diagnostics, prognostics and therapeutics, including at least 80 major categories of medical models. Generative artificial intelligence and geometric deep learning techniques, and algorithms including 2D and 3D graph machine learning and GenAI algorithms, are described, tailored and applied to diagnostic disease description, prognostic prediction and therapeutic development and management, including generation of novel synthetic drugs. The AI and machine learning techniques and algorithms are applied to understand each individual's genetic, RNA and protein anomalies that represent the source of many unique patient diseases. AI-enabled software agents assist physicians and researchers in building patient medical models. Several personalized medicine applications of individualized medical modeling include cardiovascular disease, cancer, neurological disorders, immune system disorders and genetic diseases.

The embodiment of the invention discloses an evidence tracing method and device for reply content of a medical model, a medium and a product. The method comprises the following steps: matching question data with a knowledge text in a knowledge base to obtain a first confidence coefficient which is a standard similarity between the question data and the knowledge text; based on the first confidence coefficient and the first optimal parameter, screening out an actual candidate text from the knowledge base, and obtaining actual reply content in combination with the medical model; semantically segmenting the actual reply content into a plurality of actual fine-grained fragments, and matching the actual fine-grained fragments with the knowledge text to obtain a second confidence coefficient which is a standard similarity between the actual fine-grained fragments and the knowledge text; and based on the second confidence coefficient and the second optimal parameter, the tracing evidence of the actual fine-grained fragment is screened out from the knowledge base, so that the tracing result of the actual reply content can be displayed. According to the method, through RAG retrieval and traceability retrieval, the original source of the reply content can be reversely verified, high-precision traceability is achieved, and interpretability is enhanced.

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 provides a method and device for generating pre-training data of a large medical model and electronic equipment, and the method comprises the steps: obtaining initial medical data, and carrying out the format conversion of the initial medical data, and obtaining structured medical data; identifying a noise text in the structured medical data, and clearing the noise text to obtain de-noised medical data; in response to a repeated text in the de-noised medical data, determining a repetition type of the repeated text, and performing correction processing on the repeated text to obtain initial target medical data; filtering the initial target medical data based on a preset medical text to obtain target medical data; randomly extracting a preset number of target medical data from the target medical data as to-be-detected medical data, performing quality detection on the to-be-detected medical data, determining that a detection result is qualified, performing format conversion on the target medical data to obtain training medical data, and taking the training medical data as pre-training data of a medical large model; and the quality of the pre-training data is improved.

The invention discloses a multi-modal medical image data fusion analysis system based on deep learning, and relates to the field of medical images, and the system comprises a medical image module which is used for obtaining medical image data of different modals and multiple angles; the image recognition module is used for recognizing part contours in the medical image data of all the modalities, and the image recognition module aligns all the medical image data based on the part contours in all the medical image data so as to obtain fusion data of all the modalities at different angles; and the mapping module is used for extracting part contours of the medical image data of each modal from the image recognition module and generating a medical model of a corresponding dimension based on each part contour. Through efficient fusion of multi-modal image data, limitation of a single-modal image is effectively overcome, advantages of different modals are combined through introduction of multi-modal data fusion, and more comprehensive information support is provided for accurate identification of a focus area.

The invention provides a medical model training method and device and a computer readable storage medium, and the method comprises the steps: carrying out the low-rank decomposition of the gradient of a local model, and obtaining a local first low-rank matrix and a local second low-rank matrix; training the local first low-rank matrix and the local second low-rank matrix through the received updated global first low-rank matrix, the updated global second low-rank matrix and the local data until the training of the local first low-rank matrix and the training of the local second low-rank matrix reach a set iteration condition; and generating a local medical model according to the finally obtained global first low-rank matrix, the finally obtained global second low-rank matrix, the finally obtained local first low-rank matrix, the finally obtained local second low-rank matrix and the local model. According to the method, the training efficiency is improved and the communication consumption caused by model transmission is reduced by reducing the number of the training weights and the size of the transmitted model.

The invention belongs to the technical field of artificial intelligence, and more specifically relates to a medical large model migration training method based on multi-matrix aggregation multi-dimensional features. The method comprises the following steps: collecting professional medical data, and constructing a professional medical data set; federal migration training is performed on the pre-trained large model, and specific professional medical knowledge is injected into the pre-trained large model, specifically, for the pre-trained Baichuan large model, a low-rank matrix is initialized, and the low-rank matrix and the weight are quantized; the quantized low-rank matrix is adapted to a linear layer and an attention layer of a Baichuan large model; extracting three different dimensions of knowledge including original knowledge, low-rank knowledge and global complex knowledge from the professional medical data set through multi-round efficient migration training; and fusing the extracted knowledge of three different dimensions and carrying out weight aggregation. The problems that an existing training method needs a large amount of high-quality data and computing resources and is poor in performance in specific medical scenes with stricter requirements are solved.

The invention relates to the technical field of education management, in particular to a health education content management system for gynecological nursing, which is constructed by the following steps: S1, analyzing original medical content to generate a professional expression structure, and identifying a core medical entity and a logic relation chain between entities; s2, on the basis of the professional expression structure and the user cognition level label, outputting a cognition adaptation expression through a hierarchical conversion engine; and S3, binding the cognitive adaptation expression with a professional expression structure, and constructing a bidirectional version topological relation in a content storage library, so that a reversible mapping relation is formed between the professional expression and the adaptation expression. According to the method, not only is the insufficient adaptation of a general medical model in terms of gynecology ambiguity and multi-stage dependence overcome, but also the subsequent conversion process is ensured to keep consistent in terms, relationships and contexts, and an interpretable and traceable basic expression framework is provided for cognitive adaptation.

The application discloses an electronic medical record and a cataloging and classifying method, system and device thereof. The method comprises the following steps: obtaining standardized data by preprocessing multi-format medical record original data through a medical OCR model, a VAE anomaly detection algorithm, a medical knowledge graph word segmentation tool and a Transformer term standardization model; checking semantic rationality, extracting multi-dimensional features, and obtaining a comprehensive feature vector after strengthening and fusing; constructing a transfer learning classification model and training the model, inputting the feature vector to generate cataloging information; and finally, evaluating the classification result through an active learning mechanism, combining artificial labeling data and a causal forest dynamic updating framework, and incrementally learning and optimizing the model performance. The application solves the problems of low multi-format data extraction accuracy, insufficient term standardization and poor model generalization in the prior art.

The present application relates to the technical field of computers, and provides a three-dimensional model usage method and apparatus based on a medical model content library, an electronic device, and a storage medium. The method comprises: when performing medical simulation training, invoking a medical virtual content library, wherein the medical virtual content library is obtained by storing three-dimensional data files; and obtaining demand data, searching the medical virtual content library for a corresponding target three-dimensional data file on the basis of the demand data, and invoking a corresponding target three-dimensional model on the basis of a target three-dimensional medical file. The present application solves the problem in the prior art of how to quickly provide medical three-dimensional scenes and matched modes under various scenes.

The invention discloses a multi-dimensional quality cross validation evaluation method and system based on medical model training, and belongs to the technical field of medical file evaluation. Constructing a pathological feature sample library and a medical record archive library based on the pathological feature description words and the medical record archives, and classifying and recording the medical record archives through the pathological feature description words; according to the incidence process of the pathological characteristics, forming an incidence state transition relation chain and configuring incidence state verification nodes; in combination with a clinical pathology state time range of the pathology state verification node, screening associated cross-pathology features to form a cross-pathology label; and updating according to a verification node sequence to obtain a disease occurrence state cross relation chain, performing multi-dimensional quality evaluation on the medical record file, and outputting an evaluation value. According to the method, the accuracy and reliability of medical file quality evaluation are improved through multi-dimensional cross validation, high-quality medical record data support is provided for medical model training, and the method is suitable for medical record quality auditing, medical data treatment and medical model training data screening scenes of medical institutions at all levels.

The invention provides a method and system for improving reliability of medical decision and application. The method comprises the following steps: constructing a data set covering any disease according to medical data; training the medical vertical large model according to the data set to obtain a medical decision set; performing priority ranking on any decision in the medical decision set by constructing a multi-dimensional evaluation system to obtain a primary medical decision ranking result; and dynamically adjusting a weight parameter and a decision rule of any medical decision in the primary medical decision sorting result by taking the feedback suggestions and the actual diagnosis and treatment effects as reward signals to obtain a final medical decision sorting. According to the method, the problems of data fragmentation and illusion faced by the current large medical model and the industrial pain points of insufficient application suitability are effectively solved, and the clinical diagnosis and treatment efficiency and quality are remarkably improved.

This invention discloses a secure aggregation method and system for multi-party medical model parameters based on blockchain. Combining homomorphic pseudo-random number generator technology and one-time pad encryption technology, this invention designs a blockchain-based method for covertly distributing pseudo-random number generator seeds. This achieves privacy protection for local model parameters in a single round, random replacement of local model masks in multiple rounds, and efficient aggregation of global model parameters. It prevents the risk of privacy leaks due to unauthorized entities stealing medical model parameters. Furthermore, the invention designs a blockchain-based hierarchical summary record for model parameters, significantly reducing the communication and computational overhead of model training nodes and the recovery of erroneous model parameters while ensuring the consistency of the medical model.

The present invention provides a method and system for constructing a large multimodal medical model. The method comprises: acquiring multiple sample images; performing noise reduction on the sample images and generating images of each modality based on the noise reduction results; defining a reference coordinate system, mapping the myocardial contour and thickness in a second MRI image to the reference coordinate system, and registering the metabolically active region in a second PET image to the reference coordinate system to obtain a fused coordinate system; generating multimodal features based on vascular topology information, myocardial contour and thickness, and metabolically active regions in the fused coordinate system; annotating the multimodal features corresponding to each sample image as a cardiovascular disease risk level, and inputting the annotated multimodal features into a cardiovascular disease prediction model for training. The present invention can improve the accuracy and reliability of cardiovascular risk prediction.

The invention discloses an intelligent triage system, method and device based on a medical large model and a medium, and relates to the technical field of medical diagnosis, and the system comprises a first data obtaining interface which is used for obtaining multi-modal medical record data; the data fusion tool is used for analyzing the multi-modal medical record data and fusing the multi-modal medical record data according to an analysis result to obtain fused medical record data; the second data acquisition interface is used for acquiring real-time influenza data, regional climate information and target operation data; the data input module is used for inputting the fused medical record data, the real-time influenza data, the regional climate information and the target operation data into a target medical big model, so that the target medical big model generates target triage information corresponding to each target patient; and the information acquisition module is used for acquiring target triage information. The patient is triage by using the influenza data, the regional climate information and the operation data of the hospital, so that the reliability of the triage result is ensured.

This invention discloses a method and system for developing a medical dataset based on a large-scale model. The method includes: acquiring multiple medical data sets to be processed; performing privacy data identification and encryption on each medical data set based on a de-identification algorithm to obtain multiple de-identified medical data sets; labeling each de-identified medical data set with medical information according to multiple preset student models to obtain corresponding labeled data; sampling and verifying all labeled data according to the teacher model corresponding to the student model, and organizing the labeled data into a medical dataset based on the verification results; the medical dataset is a pre-training dataset, instruction fine-tuning dataset, and / or test dataset for large-scale model training. Therefore, this invention can effectively ensure privacy and security during large-scale medical information processing, providing high-quality and standardized data asset support for the robust training of subsequent large-scale medical models.

The invention discloses a medical consultation reply generation method, device and equipment based on multi-modal fusion, and belongs to the field of text generation, and the method comprises the steps: selecting a plurality of first agent experts and a plurality of second agent experts according to multi-modal data input by a user; extracting medical data features and numeric symbol features from the multi-modal data through a first agent expert, and performing joint coding on the medical data features and the numeric symbol features to obtain a multi-modal feature vector; inputting the output of the first agent expert and the multi-modal feature vector to a corresponding second agent expert, and calling the second agent expert according to a preset calling sequence; the second agent experts are connected according to a preset calling sequence; and if the consistency verification of the medical knowledge output by each agent expert and the mathematical logic passes, aggregating the output of each agent expert, and generating a medical consultation reply. The problem of low quality of medical consultation reply generated by a large medical model can be solved.

The invention discloses a curative effect prediction method based on multi-organ metastasis genome data, and belongs to the technical field of medical models, and the method specifically comprises the following steps: collecting clinical pathological characteristics, multi-organ metastasis genome data and a treatment scheme of a breast cancer patient, and recording a metastasis part and a load state; dimensionality reduction is conducted on high-dimensional genome data through a regularization algorithm, feature importance is evaluated in combination with a nonlinear model, and clinical, treatment and genome features related to treatment response are screened out; inputting the screened features into a machine learning and deep learning framework, randomly dividing a training set and a test set in a layered manner, optimizing hyper-parameters through cross validation, and constructing a classic machine learning set model and a deep learning model based on an attention mechanism; disturbing test queue treatment scheme data, evaluating the consistency of model recommendation and an actual scheme, and verifying the prediction capability and clinical practicability of the model; according to the method, multi-dimensional data are integrated, and the curative effect prediction accuracy of the metastatic breast cancer is improved.

This application provides a method, apparatus, device, and storage medium for self-correction of a medical visual language model based on a dynamic experience base (DEKB), relating to the field of medical visual language processing technology. The method includes: when the initial diagnostic result of the medical visual language model for a current clinical case does not match the fact label, constructing the current case as a structured experience unit and storing it in a dynamic experience knowledge base; upon receiving a query, retrieving historical experience cases related to the new query from the dynamic experience knowledge base; using the historical experience cases as contextual prompts to guide the medical visual language model in chain-like thinking, generating a corrected reasoning result. By constructing an endogenous dynamic experience knowledge base, designing a deep attribution analysis mechanism, and employing a dual-threshold retrieval algorithm, this application enables DEKB to transform the model's historical errors into structured knowledge that can be used for future reference, significantly improving the robustness and generalization ability of medical model image diagnosis reasoning.

The invention discloses an intelligent monitoring and early warning system for medical adverse events, and relates to the technical field of medical treatment, and the system comprises an ETL module which is used for extracting medical text data from a hospital clinical information system; the medical adverse event prediction module is used for predicting the medical text data and obtaining the medical adverse event category and grade corresponding to the medical text data; the medical adverse event prediction module comprises a medical adverse event model obtained through training; the medical adverse event model is obtained by training a medical model according to historical medical adverse event samples, and the medical adverse event model can output corresponding medical adverse event types and levels according to input medical text data; according to the medical adverse event reporting method and device, the medical adverse event reporting efficiency can be improved.

The invention discloses a clinical prediction model construction method and system for treating oligometastatic non-small cell lung cancer through radioactive particle implantation, and relates to the technical field of medical model construction, and the method comprises the steps: obtaining patient data, taking survival time as a main endpoint event, employing a Cox regression model to carry out single-factor analysis on factors affecting patient prognosis, and obtaining a single-factor analysis result; determining the relationship between each clinical feature and survival time, and incorporating the important factors influencing the prognosis into multi-factor Cox regression analysis to determine independent risk factors influencing the survival prognosis of the patient; generating a first clinical prediction model according to each independent risk factor nano-construction; cD3 + AC, CD4 + AC, CD8 + AC, combined treatment, complete treatment and the number of metastatic lesions are taken as variables together to be included in construction of a clinical prediction model, and a second clinical prediction model is manufactured and generated. Independent risk factors influencing the prognosis of a radioactive particle implantation treatment few-metastatic NSCLC patient are confirmed, the clinical prediction model is established, the treatment prognosis can be accurately predicted, and the treatment cost is reduced. And screening of dominant benefit people is facilitated.

The invention provides an artificial intelligence diagnosis and treatment analog simulation system and method, and the system comprises a knowledge graph construction module which is used for constructing the knowledge association relation among the medical information, the disease type and the treatment scheme of a patient, generating a medical knowledge graph, and carrying out the iteration; the medical model construction module is used for constructing a large medical model for disease recognition and treatment scheme generation; and the diagnosis and treatment aid decision-making module is used for performing personalized analysis on the medical information of the patient in combination with the knowledge graph construction module and the medical model construction module to generate diagnosis aid information and a treatment aid decision-making scheme. Therefore, by constructing the medical knowledge spectrogram and fusing the large-scale pre-training model ability, the medical information of the patient is automatically analyzed, the auxiliary information of the personalized diagnosis and treatment scheme is generated, and the auxiliary decision making function of the AI doctor role is achieved.

The invention relates to a method for constructing a chronic kidney disease risk prediction model after acute kidney injury, and belongs to the field of medical models. The method comprises the following steps: constructing a clinical complete prediction model by using an XGBoost algorithm, obtaining SHAP values of different factors, carrying out feature importance sorting by using the SHAP values, and identifying key prediction factors through sorting; performing a forward variable selection strategy, and expanding the number of candidate variables to a preset value by adopting a step-by-step forward selection method based on an SHAP value sorting result; cross validation and model optimization are carried out, the optimization objective is to maximize AUC of a training set, and generalization performance of different factor combinations is evaluated; constructing a simplified model, selecting a core prediction factor of which the prediction effect AUC drop range is smaller than a set value, and constructing a core prediction factor model; and carrying out simplified model verification. According to the method, a model can be constructed to accurately identify high-risk people suffering from chronic kidney diseases after acute kidney injury, a reliable basis is provided for clinical intervention, and the clinical application value of the model is finally improved.

Embodiments of the present disclosure disclose a medical model reply content evidence tracing method, device, medium and product. The method comprises: matching the question data with the knowledge text in the knowledge base to obtain a first confidence degree, which is the standard similarity between the question data and the knowledge text; based on the first confidence degree and the first best parameter, filtering out the actual candidate text from the knowledge base, combining the medical model to obtain the actual reply content; after the actual reply content is semantically segmented into multiple actual fine-grained segments, matching the actual fine-grained segments with the knowledge text to obtain a second confidence degree, which is the standard similarity between the actual fine-grained segments and the knowledge text; based on the second confidence degree and the second best parameter, filtering out the tracing evidence of the actual fine-grained segments from the knowledge base, and then the actual reply content tracing result can be displayed. Through RAG retrieval and tracing retrieval, the original source of the reply content can be checked reversely, high-precision tracing is realized, and the explainability is enhanced.