148 results about "Patient state" patented technology

The invention relates to a patient examination item recommendation method based on a medical knowledge graph. The method comprises the steps that firstly, historical data in the medical field is acquired, key information is extracted from the historical data to generate an entity relation triple, and then a medical knowledge graph is constructed; then, generating a patient state sub-graph based on patient input information, processing the medical knowledge graph by adopting an improved TransD model, and determining unique representation of an entity-relationship pair; mapping the patient state sub-map to a unique representation space, obtaining vector representation of a patient and an examination item related entity, and calculating a similarity value between vectors to obtain a similarity value matrix; and finally, performing priority ranking on the inspection items according to the matrix, and generating a final recommendation list through threshold screening and combination with graph neural network verification, optimization and adjustment of priorities. By adopting the method, examination items can be accurately recommended for patients, doctors are assisted to improve the treatment efficiency, the queuing time of the patients is shortened, and the medical experience of the patients is remarkably improved.

The invention relates to the technical field of electronic prescriptions, in particular to a dynamic electronic prescription generation method and system based on artificial intelligence. The method comprises the following steps: collecting the latest physiological state parameter flow of a patient, carrying out real-time health state evaluation, and generating a personalized patient state map; historical medical records of a patient are extracted, time sequence pathological evolution tracking is carried out, and a pathological evolution trajectory is generated; performing multi-parameter time sequence difference comparison calculation and drug curative effect quantitative evaluation based on the personalized patient state map and the pathological evolution trajectory to obtain a curative effect evaluation report; performing allergic drug identification on the patient based on the historical medical record of the patient, and performing secondary drug screening to obtain a drug candidate set; and performing intelligent matching calculation on the drug candidate set according to the curative effect evaluation report, and performing combinatorial optimization analysis to generate a final effective electronic prescription. The electronic prescription is automatically updated and adjusted based on the state change of the patient, the risk of allergic prescriptions is reduced, and the safety of the prescriptions is improved.

The invention discloses an illness state reasoning and medical record generation method and system based on large model fusion and knowledge enhancement. The method comprises the following steps: acquiring and preprocessing medical data of a patient; performing feature extraction and medical entity recognition on the preprocessed medical data based on a pre-constructed medical entity recognition model to establish a patient state sub-graph; performing topology analysis and multi-hop reasoning on the patient state sub-graph based on a graph neural network to obtain a diagnosis candidate set and a corresponding reasoning evidence chain set; performing double-track parallel illness state reasoning based on the medical data and the reasoning evidence chain set to respectively generate inner track illness state probability distribution and outer track illness state probability distribution, performing weighted fusion on the inner track illness state probability distribution and the outer track illness state probability distribution to generate optimized illness state reasoning probability distribution, and generating a structured medical record with an evidence traceability identifier according to the optimized illness state reasoning probability distribution; according to the method, the problems of logic fracture, knowledge illusion and uninterpretability in large-model medical application are solved, and the accuracy and transparency of clinical diagnosis reasoning are remarkably improved.

The invention discloses a chronic disease management AI recommendation anti-illusion method and system based on a knowledge graph, and relates to the technical field of medical health artificial intelligence, and the method comprises the steps: obtaining chronic disease medical data of a patient, extracting a feature vector, and constructing a multi-modal health trajectory vector; constructing a chronic disease tag vector and a directional causal knowledge graph; executing map path search to obtain candidate treatment paths; constructing structured cue words, and inputting the structured cue words into the medical large language model to generate candidate schemes; and calculating a comprehensive illusion score to carry out anti-illusion judgment and correction, and outputting a credible personalized treatment recommendation. By constructing a directional causal knowledge graph, the causal reasonability and semantic consistency of treatment path retrieval are enhanced, and the matching precision of candidate paths and patient states is improved; by calculating the comprehensive illusion score and introducing the four-dimensional score item for anti-illusion judgment, the credibility of the output content of the large language model is improved, and the availability of AI recommendation in clinical aid decision making is guaranteed.

The invention relates to the technical field of medical informatics, in particular to a neurosurgery patient comprehensive diagnosis and treatment system which comprises a multi-mode state monitoring module, a dynamic trend evaluation module, a consciousness function analysis module, an optimization nursing intervention module and a nursing risk early warning management module. According to the method, key features are extracted in combination with image information, health indexes and behavior changes, serialized monitoring of intracranial lesions, physiological fluctuations and behavior modes is achieved, vital signs and behavior change trends are fused, continuity and objectivity of patient state evaluation are improved, the accuracy of risk identification is enhanced, and the risk identification efficiency is improved. According to the method, a health state change rate evaluation mode is introduced, the prediction capability of function recovery potential is improved, an unstable region is dynamically identified and potential risks are pre-judged in combination with an original state trajectory and intervention response difference, continuous optimization of an intervention path and active avoidance of nursing risks are realized, and the collaboration and continuity of comprehensive management are enhanced.

The application discloses a kind of intelligent service systems of advanced cancer pain pain, including: analgesic device, massage piece, driving source and ice piece, ice piece is located inside the mounting seat side surface, ice piece is used to contact cooling massage ball;Push module is used to daily push the predetermined check quantization table to patient, check quantization table at least includes one or more of the following each item: diet condition, sleep condition, body pain condition, self cognitive ability and mental state;Evaluation module, evaluation module obtains patient state evaluation result according to the data of patient to check quantization table, and according to evaluation result, push management scheme to patient and adjust analgesic device analgesic scheme, the living condition of patient every day is obtained regularly by push module, and the individualized, centralized pain intervention scheme is formed by evaluation module, one aspect realizes the timely evaluation of patient life quality, while quantized mode is convenient for others to view specific change situation, to improve specific condition of patient in a targeted manner.

The invention discloses a self-adaptive wearing and monitoring method of an intelligent drainage monitoring system, and relates to the technical field of medical intelligent wearing, and the method comprises the steps: collecting the drainage part data of a patient through multi-modal biological characteristics, optimizing the wearing parameters through a particle swarm algorithm, and driving a shape memory polymer to adjust the equipment fitting degree, biochemical indexes, pressure and flow of drainage liquid are detected in real time through a micro-fluidic chip and various sensors, intelligent early warning of abnormity is achieved in combination with a self-adaptive neural network, parameters such as drainage negative pressure are automatically adjusted according to physiological signals of a patient, risks are predicted through a digital twinborn model, and a maintenance scheme is generated. According to the intelligent drainage monitoring system, self-adaptive wearing is achieved, and the comfort level of a patient is improved; real-time accurate monitoring and intelligent early warning are achieved, and medical safety is guaranteed; drainage parameters are automatically adjusted according to patient states, and treatment efficiency is improved; the intelligent dredging and anti-infection functions are achieved, the medical risk and cost are reduced, and the clinical application value is high.

The invention relates to the technical field of intelligent medical treatment, in particular to a basic nursing risk intelligent identification and early warning system and method. The system comprises a multi-source data perception and fusion module which is used for acquiring and fusing dominant risk data and implicit risk data of a patient in real time and outputting a structured patient state feature set; the risk intelligent identification core module simulates physiological state changes through a patient personalized digital twinborn model and is coupled with a risk conduction knowledge graph to deduce a risk evolution trajectory so as to realize intelligent identification of potential risks and intervention measure simulation; the intelligent early warning and intervention module is used for executing hierarchical early warning and active intervention linkage; and the man-machine collaborative feedback and self-evolution module continuously optimizes the system model and strategy through clinical feedback data. The problems that in the prior art, the risk perception dimension is single, early warning lags behind and intervention is disjointed are solved, and prospective and precise recognition, prevention and control of the nursing risk are achieved.

The invention provides a patient abnormal state identification monitoring early warning method and system, and relates to the technical field of patient state monitoring. The method comprises the following steps: acquiring a facial image sequence of a patient, tracking a texture direction and constructing a variation comparison sequence, sorting linkage differences to generate dynamic clues, extracting non-shielding track mark picture distribution, identifying life fluctuations in a segmented manner, generating a fluctuation list through cross pairing, establishing a mapping chain to mark centralized anomalies, and generating an early warning scheme. According to the method, the recognition precision of facial subtle changes is improved, region credibility is calibrated in combination with a shielding ratio, non-shielding continuous image regions are screened, a physiological signal change sequence is segmented through trend continuity and direction consistency, high-sensitivity fluctuation paragraphs are extracted, and the credible image regions and physiological data are subjected to cross pairing in time and space; and merging intersection signals, identifying continuous high-frequency abnormal segments, and constructing a multi-dimensional mapping chain, thereby realizing high-precision early warning and real-time monitoring of complex state changes.

The invention belongs to the technical field of medical artificial intelligence, and discloses an auxiliary system based on artificial intelligence medical data analysis. A hidden trend turning point in time sequence physiological index data is analyzed and identified through a micro-period sliding window, time axis accurate alignment of multi-modal medical data is realized based on the turning point, and a time-space synchronization matrix is constructed to quantify a time coupling relationship among different modal data. And extracting a curvature distribution and tangent angle sequence by adopting geometric characteristic analysis of a feature vector change trajectory, and constructing a multi-branch path prediction model of illness state evolution. The system monitors the fit degree change rate of the patient state and the prediction path in real time, scientifically identifies the switching time of the illness state evolution branch, foresight marks the optimal intervention opportunity, and generates personalized clinical intervention suggestions. According to the invention, the change from passive monitoring to active prediction is realized, the timeliness and accuracy of clinical intervention are improved, and the medical resource configuration is optimized.

The invention relates to the technical field of medical data processing, in particular to a multi-feature recognition based postoperative anesthesia patient state quantitative evaluation method. According to the method, the anesthesia postoperative state is divided into a deep anesthesia stage, a superficial consciousness recovery stage and a cognitive recovery stage based on clinical observation. In the deep anesthesia stage, physiological feature data are collected at high frequency, and a physical sign health index is obtained by calculating the standardized deviation between the physiological feature data and a normal reference value; in the shallow consciousness recovery stage, physiological feature data and unconsciousness behavior data are collected, a weighted model is constructed in combination with the physical sign health index, and an unconsciousness behavior response index is obtained; in the cognitive recovery stage, physiological feature data and subjective feature data are collected, and short-term trend prediction is performed on a cognitive recovery index based on a grey prediction model. Furthermore, multi-source feature data of the three stages are fused to construct a dynamic trajectory prediction model, the postoperative recovery trend and state change of the patient are comprehensively analyzed, and quantitative evaluation of the postoperative delirium risk is achieved.

The invention provides a closed-loop reinforcement learning diagnosis method and system based on a high-fidelity virtual clinical environment, and the method comprises the steps: 1, carrying out the structural processing of basic information and an examination record set of a patient, and obtaining a standardized examination and diagnosis data set; 2, training to obtain a high-fidelity clinical environment system; and step 3, based on the high-fidelity clinical environment system, obtaining a diagnosis agent through reinforcement learning training. A simulation examination result highly consistent with the state of a patient can be generated based on real electronic health record data, a diagnosis agent performs multiple rounds of interaction in a virtual environment, autonomously explores a diagnosis path, and performs strategy optimization through a double-reward mechanism, so that the crossing of a diagnosis model from static prediction to dynamic decision is realized, and the diagnosis efficiency is improved. And the diagnosis accuracy and the reasonability of examination recommendation are greatly improved.

The invention relates to the technical field of medical information smart wards, in particular to a smart ward management method and system based on the Internet of Things. Environmental parameters in a smart ward and physiological parameters of a patient in the smart ward are obtained; performing comprehensive analysis according to the environmental parameters and the physiological parameters to obtain corresponding analysis results; and according to an analysis result, a control instruction used for adjusting environment equipment in the smart ward is generated and / or corresponding guidance information is sent to the management end, so that the ward environment and the patient state are combined, and integrity and collaboration are realized. Intelligent management and real-time monitoring and response of the ward are realized through the control instruction and / or guidance information, the timeliness of medical services is improved, dynamic allocation of medical resources is facilitated, the resource utilization efficiency is improved, the operation process is greatly simplified, and the use experience of medical staff and patients is improved.

The invention relates to an intelligent platform management system carrying infusion port patients, and relates to the technical field of data management, and the system comprises a monitoring unit which is used for collecting impedance data, temperature data and medicine use time sequence data; the intelligent analysis unit is used for constructing a dynamic health baseline of the patient, calling a preset machine learning model based on the dynamic health baseline and outputting a complication risk level of the patient; the closed-loop execution unit is used for triggering a three-dimensional life behavior guidance scheme matched with the patient state based on the complication risk level, calculating a corresponding infusion port maintenance period based on the real-time smoothness score, and generating a medicine supply instruction in combination with medicine use time sequence data; and the hierarchical interaction platform is used for generating an early warning view containing the risk level of the patient and the key attention information based on the complication risk level, and pushing maintenance appointment suggestions and planning contents at the patient side based on the maintenance period of the infusion port. The system can improve the management efficiency and safety of the patients in the infusion port.

The invention discloses a multi-modal drug recommendation method and system based on large language model driving, and relates to the technical field of medical informatics and artificial intelligence. The invention aims to solve the limitation of an existing drug recommendation model in the aspects of patient characterization construction, drug multi-modal modeling and large language model application. Comprising the following steps: constructing patient characterization, and aligning a patient state and a potential medication space in a characterization learning stage through a cooperative prompt project driven by a large language model; multi-modal drug characterization is constructed, and clinical logic and chemical characteristics of drugs are deeply depicted through semantic expert and molecular structure expert dual-channel design; a'static anchoring dynamic 'time sequence reasoning mechanism is provided, context modulation is performed on a dynamic evolutionary clinical event sequence by utilizing a static treatment baseline formed by global medication history, and finally, an accurate and safe medication combination is generated through a label sensing prediction module. According to the method, the accuracy, safety and clinical logic self-consistency of drug recommendation are remarkably improved.

The invention discloses a medical record data association mining method and system based on reinforcement learning, and relates to the technical field of medical record data association mining, and the method comprises the following steps: obtaining medical record historical data, carrying out multi-modal feature extraction, and constructing a medical record data multi-modal state vector; constructing state input of a reinforcement learning agent based on the multi-modal state vector, constructing an action space and a reward evaluation mechanism, performing updating and iteration according to the reward evaluation mechanism, and outputting potential correlation information of the medical record; a world model for simulating a patient trajectory is introduced in the reinforcement learning process, patient state transition characteristics are learned according to medical record historical data, and a simulation sample for auxiliary training is generated for reinforcement training; performing interpretable output on the finally obtained medical record potential association information according to an enhanced training result, and constructing a medical record association network; the method effectively solves the problems that multi-modal heterogeneous data fusion is difficult, dynamic evolution of the disease course is difficult to model, and potential association interpretability is insufficient.

The invention discloses a visual asynchronous event triggering and calling method based on a medical information system, which comprises the following steps of: asynchronously acquiring multi-source data streams such as clinical events, equipment monitoring, doctor advice execution and patient states in the medical information system to form a multi-source asynchronous medical data stream set; generating a system event triggering association group based on the set and creating an asynchronous event scheduling strategy table; performing system event association degree calculation on the data stream to obtain a multi-dimensional asynchronous event association matrix; and carrying out system asynchronous time delay compensation and event chain visualization processing on the incidence matrix based on the scheduling strategy table, and finally generating a dynamic asynchronous event correlation topological graph. According to the invention, by constructing a system event detection operator set, establishing cross-flow system relevance calculation and establishing a priority-based asynchronous scheduling optimization mechanism, intelligent triggering, calling and early warning functions of medical system events are realized.

The invention discloses a big data-based digestive system department prediction analysis system. The system comprises a data acquisition module, a knowledge graph fusion module, a disease course evolution prediction module and an intelligent diagnosis and treatment decision module. The invention relates to the technical field of medical health information, in particular to a big-data-based predictive analysis system for the department of gastroenterology. According to the scheme, a physiological signal is reconstructed by using a multi-scale mask auto-encoder, and the signal and a text are aligned; mapping knowledge graph regularization is introduced, the joint code is mapped to a digestive system knowledge graph space, and medical semantic representation is obtained; constructing a time sequence attenuation dynamic graph through the dynamic memory graph, and adding graph convolution gating; an embedded differential equation module simulates inflammation and tumor marker evolution to realize accurate prediction; patient state vectors are constructed based on multi-modal coding, medicines are rapidly selected by using a scoring function, a treatment process is modeled as a Markov decision, and a strategy network, a value network and PPO optimization are combined to realize balance of curative effect and income.

Systems, methods, and computer-readable media for providing a decision support solution to medical professionals to optimize medical care through data monitoring and feedback treatment are provided herein. In another embodiment, a computer-implemented method for modeling patient outcomes resulting from treatment in a specific medical area includes receiving patient-specific data associated with a patient, determining a plurality of possible patient states under which the patient can be categorized, a current patient state under which the patient can be categorized and determining probabilities of the patient transitioning from any of the possible patient states to every other possible patient state.

The application discloses a peritoneal dialysis prescription dynamic adjustment method and system, relates to the field of data processing, and comprises the following steps: under the authorization condition of a patient, acquiring multi-dimensional data related to current peritoneal dialysis of the patient; integrating the multi-dimensional data into a structured patient state description text; calling a large language model, generating an adjusted peritoneal dialysis prescription based on the patient state description text according to a set frequency by using a peritoneal dialysis prescription prompt word template. The application can significantly improve the accuracy and timeliness of the dialysis prescription.

The invention discloses a self-adaptive training parameter adjusting method for a hand rehabilitation exoskeleton, and relates to the technical field of rehabilitation medical instruments, and the method comprises the specific steps: deploying a multi-modal sensing system to synchronously collect full-dimensional original signals of a patient, and building a healthy hand motion rhythm feature library; a plurality of indexes are obtained through preprocessing and feature quantitative analysis; judging conflicts and types through intention-conflict decoding; constructing a self-adaptive parameter adjustment model, and executing a flexible solution strategy; a closed-loop adjustment period is set, all the steps are executed circularly, scene parameter adjustment is completed, and advanced rehabilitation training is implemented; according to the method, a comprehensive and accurate patient state sensing system is constructed, the intention and ability matching state is accurately judged, adaptive training parameters are generated, training risks are avoided, safety and adaptability are improved, a closed-loop adjustment mechanism is further constructed, scene-based and advanced training is combined, training is dynamically adjusted, the enthusiasm of a patient is aroused, hand function recovery is accelerated, and the patient experience is improved. The rehabilitation effect and efficiency are improved.

This invention discloses a secure data transmission access system and method for intelligent medical IoT, relating to the field of data transmission control technology. It includes: S1, real-time acquisition of medical monitoring data and transmission information, and data preprocessing; S2, assessment of the temporal consistency of medical monitoring data sequences from each transmission channel, and data rearrangement and completion processing for transmission channels with sequence distortion; S3, analysis of the correlation of data change trends between different transmission channels, assessment of the time synchronization relationship between channels, and time shift alignment for channels with stable synchronization relationships; S4, construction of a patient physiological state representation, assessment of the coupling consistency between various physiological parameters, and determination of whether to allow subsequent patient state snapshot construction. This solves the problem of inaccurate physiological state representation caused by temporal disorder and cross-channel synchronization distortion during multi-source medical monitoring data transmission in an intelligent medical IoT environment.

The invention discloses anesthesia injection equipment capable of monitoring the state of a patient, and the equipment comprises a main frame which is provided with a mounting groove; the anesthesia unit comprises a modular rack, an anesthesia injection mechanism and an anesthesia operation mechanism, the anesthesia injection mechanism and the anesthesia operation mechanism are integrated on the modular rack, a mounting cavity is formed in the modular rack, the anesthesia injection mechanism is arranged in the mounting cavity, and the anesthesia operation mechanism is arranged in the mounting cavity. The anesthesia operation mechanism is mounted on the modular rack; the anesthesia injection equipment capable of monitoring the state of the patient relates to the technical field of anesthesia equipment, the anesthesia injection equipment capable of monitoring the state of the patient is divided into the anesthesia unit and the monitoring unit, the anesthesia unit is detachably connected with the monitoring unit, and the occupied space of the equipment is saved through modularized integration of the anesthesia injection function and effective combination of anesthesia and monitoring; and the problem of large occupied area of the equipment is effectively relieved.

According to the personalized debugging method and system of the intelligent breathing training device, the breathing function evaluation result of the patient, the breathing training control information and the breathing training similarity are obtained according to at least two of the breathing function evaluation result, the breathing training control information and the breathing training similarity; determining a breathing training personalized debugging result of immersive training information provided by the to-be-processed patient in the first virtual reality lung function data, the real-time monitoring element information is used as the occurrence number of the lung function indexes; according to the method, at least three of four different forms of respiratory training information of the number of breathing times of the patient state of each piece of real-time monitoring element information and different influence element information of immersive training information provided by a patient to be processed in virtual reality lung function data are used; and whether the immersive training information provided by the to-be-processed patient is reasonable or not is judged, so that the debugging accuracy and reliability of the respiratory training device can be improved.

The invention belongs to an abdominal drainage ultrasound contrast small sample segmentation method and system based on meta-learning. The method comprises the steps of collecting multi-source data and images, analyzing target patient data to obtain patient state difference data, and preprocessing an ultrasound contrast image to obtain a standard contrast image. Performing feature extraction on the historical clinical data to obtain clinical operation difference data, and obtaining physiological influence drainage data according to the influence of the physiological state change of the target patient on the drainage area. The patient state difference data, the clinical operation difference data and the physiological influence drainage data are analyzed to obtain general segmentation features, and a meta-learning feature set is formulated in combination with a standard angiography image; constructing a meta-learning feature segmentation model, inputting a to-be-segmented ultrasound contrast image, and outputting a drainage region segmentation result; the method can adapt to physiological and pathological states and scenes of different patients, significantly reduces the segmentation boundary deviation, the false drop rate and the omission rate, and improves the accuracy of drainage region segmentation.

The invention relates to the technical field of data processing, and particularly provides a hospital guide method, system and device based on multi-modal data and a storage medium, and the method comprises the steps: obtaining the multi-modal data which is provided by a patient and comprises texts, medical images and physiological signals; extracting each modal feature vector and fusing the modal feature vectors into a unified patient state characterization vector; on the basis, confidence distribution of the target department is calculated through a probabilistic reasoning model; if the result does not reach the termination condition, iteratively executing the following core steps: calculating the information gain of each potential action in the predefined action set, obtaining new data according to the optimal action, updating the state of the patient and recalculating the confidence coefficient; and finally, outputting a hospital guide result meeting a termination condition. According to the invention, through multi-modal fusion and dynamic interaction decision making, the accuracy and efficiency of intelligent hospital guide are effectively improved.

This application provides a multi-agent, multimodal collaborative method, device, equipment, and storage medium for diagnosing the condition of patients in the intensive care unit (ICU), belonging to the field of medical artificial intelligence and intelligent analysis technology for intensive care. The method includes: receiving multimodal clinical monitoring data in an ICU setting, including bedside medical images, continuous life time series, and critical care clinical text; identifying and adaptively routing the input data through a modality detection agent, distributing it to corresponding domain expert agents for pathological and physiological feature extraction; constructing a patient-centric graph structure representing the dynamic clinical condition of ICU patients through a knowledge graph agent based on the features output by each domain expert agent, this graph structure integrating structured medical knowledge of entity extraction and relational reasoning; further, inputting the constructed graph structure and original multimodal features into a collaborative agent, generating the final ICU patient condition diagnosis result (such as mortality prediction, ICU long-stay prediction, etc.) and a traceable reasoning path through graph traversal and collaborative reasoning. This application effectively breaks down data silos between different monitoring devices in the intensive care unit environment, dynamically captures cross-modal associations specific to critically ill patients, and significantly improves the accuracy of critical care clinical diagnosis and medical trust.

The application discloses a kind of based on brain-computer interface and digital twinning acupuncture remote diagnosis and treatment system and method, it is related to wisdom medical field.Therein, doctor side brain-computer interface module is parsed into operation instruction vector by the movement imagination decoding model with the movement intention signal of doctor;Patient side brain-computer interface module is parsed into patient state vector by the feeling coding decoding model with the feeling intention signal of patient;Digital twinning interaction unit is used to construct and update digital twinning body, operation instruction vector is mapped to digital twinning body to generate expected qi reaction, patient state vector is mapped to digital twinning body to update actual qi state, and the difference of two is calculated to generate closed-loop control instruction;Execution unit is used to receive closed-loop control instruction to drive acupuncture execution device to carry out acupuncture operation.The application can quantitatively transfer qi feeling, realize doctor-patient two-way thought interaction fusion and have closed-loop self-adapting control ability, improve the precision, safety and doctor-patient collaboration of remote acupuncture.

Systems and techniques can be used to estimate patient state during radiotherapy treatment. For example, a method can include generating, with a preliminary motion model, an extended dictionary of potential patient measurements and corresponding potential patient states. The method can include training, with a machine learning technique, a correspondence motion model that associates input patient measurements with output patient states using the dictionary. The method can include estimating, with a processor, a patient state corresponding to an input image using the correspondence motion model.

The application discloses a disease risk stratification method and system based on reinforcement learning, relates to the technical field of reinforcement learning, and comprises the following steps: acquiring multi-modal time series data of a target patient in a monitoring process; dynamically fusing the multi-modal time series data to extract a patient state representation vector; inputting the patient state representation vector into a pre-trained risk stratification model to output a risk stratification action according to a current strategy; generating and pushing a clinical monitoring prompt according to the risk stratification action; after a preset time window, acquiring response data of the patient to clinical intervention, calculating a reward signal, and updating the risk stratification model by using the reward signal. The technical problems that the existing disease risk stratification model is static and fixed, cannot dynamically optimize a decision strategy according to the effect of clinical intervention, and results in insufficient accuracy of risk stratification results are solved.