55 results about "Disease description" patented technology

The invention discloses an electronic medical record text structuring method. The method comprises the following steps that 1, a medical knowledge database is loaded; 2, an electronic medical record text is read; 3, word segmentation is carried out on short sentences through a forwards maximum match algorithm, and words in sentences, word properties and relative position relations are obtained; 4, whether the semantics in the sentences for disease information description is positive or negative is judged; 5, disease information elements are extracted; 6, the steps 2-5 are repeated till all interesting contents in the electronic medical record are obtained; 7, different expressions of the disease information elements are combined, the same disease information is combined according to a medical synonym bank, and redundant information is removed; 8, disease description information elements are stored in a structure / class mode, the structuring process is finished, disease relevant information can be effectively extracted from medical record description text, structural expression for disease information is formed, and therefore disease occurrence regularity, definite diagnosis ways, treatment effects and the like are deeply explored.

The invention relates to a multimode intelligent analysis method and a system. The disease description is multimode data, and the multimode data comprises character data, time sequence signal data and visual data; the method can be applied to the diagnosis navigation and diagnosis decision according to the disease situation; the method includes steps of converting visual data to character description through an image-text decoder; converting a time sequence signal to the text description through a signal character decoder; respectively converting the character data, the character description of the visual data and the time sequence signal data to a semantic vector taking a medical treatment mapping knowledge domain as an axis, wherein the medical treatment mapping knowledge domain describes the topological relation among the symptom, laboratorial index, disease, drug and operation; according to the semantic vector, a first convolution neutral network CNN model is applied to perform diagnosis navigation and diagnosis decision. According to the embodiment, the best diagnosis path and the diagnosis result with high accuracy can be acquired.

The invention provides a method and device for building diagnosis inquiring channels. The method includes the steps of receiving diagnosis inquiring questions input by patients through preset communication tools, decomposing diagnosis inquiring keywords in the diagnosis inquiring questions, finding diagnosis inquiring doctors matched with the diagnosis inquiring keywords, obtaining contact information of communication tools of the diagnosis inquiring doctors, and building the diagnosis inquiring channels between the doctors and the patients through the communication tools according to the contact information. The diagnosis inquiring keywords in the diagnosis inquiring questions asked by the patients are decomposed with the data extracting technology, the diagnosis inquiring doctors matched with the diagnosis inquiring keywords are found with the data retrieval technology, the diagnosis inquiring channels between the patients and the diagnosis inquiring doctors are rapidly built in time, the patients do not need to have medical science knowledge and also do not need to wait for registration for a long time, the diagnosis inquiring doctors can be automatically distributed for the patients only through simple disease descriptions, the patients can be in communication with the diagnosis inquiring doctors through the communication tools in time, and primary diagnosis and treatment on diseases can be achieved.

The invention discloses a medical knowledge graph construction method which can make full use of a multi-network knowledge base to dynamically construct a knowledge graph and is good in integrity andhigh in accuracy. The medical knowledge graph construction method comprises the following steps: (10) medical text data acquisition: crawling medical text data including disease description, symptom description, questions and answers and comments in a plurality of data sources through a web crawler technology; (20) obtaining a medical knowledge graph: extracting medical entities and relationshipsbetween the medical entities from the medical text data through a dictionary and rule method to obtain the medical knowledge graph comprising a medical entity graph and a medical entity relationship graph; (30) knowledge graph storage: storing the medical knowledge graph in a Neo4j graph database; and (40) entity fusion: performing entity fusion by adopting an integrated entity alignment method.

The invention relates to a mobile terminal disease intelligent auxiliary diagnosis system based on deep learning. The system is provided with a mobile medical terminal and a mobile medical cloud server. The mobile medical terminal comprises a voice recognition module, a character input module and a mobile medical cloud server connection interface. The mobile medical cloud server comprises an imageprocessing module, a medical knowledge base and a knowledge reasoning module; and the knowledge reasoning module in the mobile medical cloud server receives a text chief statement, performs knowledgereasoning on the text chief statement by utilizing the medical knowledge base to obtain a disease diagnosis result and a judgment result, and feeds back the disease diagnosis result and the judgmentresult to the mobile medical terminal through the mobile medical cloud server connection interface. Diseases are input through the mobile medical terminal and transmitted to the mobile medical cloud server, a medical common sense, a clinical guideline and a medical knowledge base are established through the cloud storage technology and used for fuzzy matching, a patient is guided to supplement disease description according to the optimal judgment result, and therefore the efficiency and accuracy of disease diagnosis are improved.

The invention provides a disease self-analysis method based on a medical ontology database. The method comprises the following steps that the medical ontology (MO) database is built, namely the MO database; according to the disease description, feature vocabularies are extracted; searching and matching are carried out in the MO database according to the feature vocabularies; according to the terms label collection obtained by matching, the probability of diseases of the MO terms collection corresponding to patients is calculated. By means of the method, the MO database based on disease features is built, the disease self-analysis method based on the medical ontology database is built, effective data and actual conditions of human bodies are combined, the symptoms in the initial stage of patients are more effectively discriminated, and an accurate and effective data basis is provided for helping patients perform self-diagnosis at the utmost.

The invention discloses a triage method, device and equipment based on a medical knowledge graph and a storage medium. The method comprises the steps of receiving disease description statements inputby a user; performing character encoding on the statement by using a pre-trained BERT model to generate a word vector, and decoding the word vector by using a BILSTM model and a CRF model to obtain adisease entity; linking the disease entities to standard disease entities in a knowledge graph by using an entity linking algorithm; vectorizing the standard symptom entity text, inputting the vectorized standard symptom entity text into a pre-trained XGBoost classification model based on a medical knowledge graph, and taking a model output result as a recommended doctor-seeing department; and returning the recommended doctor-seeing department to the user. According to the method, the main complaint of the user is directly regarded as the serialized text, the XGBoost classification model is adopted for classification, the classification result is the recommended doctor-seeing department, the trouble brought by multiple rounds of input of the user is reduced, the use experience is good, theadvantages in the aspect of spoken language recognition are obvious, and the triage result is accurate.

An acupuncture treatment instrument comprises a display unit, a memory unit, an input unit, an acupuncture unit, a central control unit and a power supply unit, human acupuncture point information and common disease information are stored in the memory unit, the acupuncture point information includes acupuncture point position image information, related meridians and collaterals and effects, the common disease information includes disease descriptions, symptoms and acupuncture point treatment prescriptions thereof, an acupuncture point-related image is displayed through the input of the input unit, and a user finds acupuncture points to be treated according to the displayed image and then uses the acupuncture unit to carry out treatment. The acupuncture treatment instrument can be conveniently used by the user, the user can find a treatment method by operating the acupuncture treatment instrument and then uses the instrument to carry out treatment as long as the user knows where the user feels uncomfortable, acupuncture treatment is no longer a professional and particularly mysterious matter, and an ordinary person can become a doctor for self.

The embodiments of the application disclose a method and a device for generating information. According to a specific embodiment, the method includes the following steps: acquiring at least two typesof disease description information describing the same disease, wherein disease description information can be a text type, a voice type or an image type; for each type of disease description information acquired, importing the type of disease description information to a pre-established feature extraction model corresponding to the type to generate feature information corresponding to the type ofdisease description information, wherein the feature extraction model is used to characterize the correspondence between disease description information and feature information; and importing the generated feature information to a pre-established feature fusion model to generate a value of probability that the disease described by the disease description information belongs to a pre-defined disease, wherein the feature fusion model is used to characterize the correspondence between feature information and probability values. The embodiment enriches the means of information generation.

The invention provides a disease automatic coding system, a method, equipment and a storage medium, wherein the method comprises the steps of acquiring a target object which is a to-be-coded disease name or a disease description; screening a code which is related with the target object from a disease code database, forming a candidate code set by the screened codes; and based on a meaning relationbetween the disease name which corresponds with each code in the candidate code set and the target object, determining the code which corresponds with the target object in the candidate code set. Thedisease automatic coding method according to the invention can automatically, accurately and efficiently determine the code which corresponds with the to-be-coded disease name or disease description.

The invention discloses an ICD automatic coding method and system for diagnosis reason visualization. The method comprises the following steps: acquiring medical record data from a medical record document library, and constructing a multi-label classification data set; preprocessing the data set, and converting the multi-label classified data set into a plurality of single-label classified data sets; completing training of a hierarchical attention neural network model based on the data sets of the plurality of single-label classifications; inputting doctor writing diagnosis data and illness state description data into the trained hierarchical attention neural network model so as to obtain predicted ICD codes and names; and according to the predicted ICD name, extracting a corresponding sentence from the disease description data as a diagnosis reason for visualization. According to the method, doctor writing diagnosis and medical record description data are used as mode features at thesame time, doctor writing diagnosis is ingeniously used as supervision information in the classification process, traditional multi-label classification is converted into a simpler single-label classification problem, and therefore the accuracy of model coding is improved.

The embodiment of the invention discloses a medical service method, device, computer equipment and storage medium. The method comprises the steps as follows: obtaining disease description of a targetuser; searching a target disease type with a mapping relationship with the disease description referring to a preset diagnostic model library according to a preset matching rule; and allocating an outpatient department and out-call personnel according to the target disease type. Before searching the target disease type, the user's prior case can be obtained to confirm whether it is old disease recurrence, so that the time taken for confirming the disease type of the user is effectively reduced. When the user is in an emergency, the user can trigger an emergency rescue function to get help fromothers. By adopting the method, the clinic time and process of the user can be effectively reduced, the waste of medical resources can be reduced and timely treat, and curing can be provided for a patient.

The invention discloses a man-machine interaction hospital guide method and system based on global attention intention recognition, and the method comprises the steps: obtaining the disease description of a patient, carrying out the natural language understanding, obtaining a standard medical term, updating a current dialogue state, inputting the current dialogue state into a preset dialogue strategy model, and obtaining a dialogue action, wherein the dialogue action comprises a system intention and a processing result, the system intention is negative, confirmation, inquiry, notification or interaction ending, and the processing result is a symptom, a disease or a department obtained by processing; and when the system intention is negative, confirmation and inquiry, selecting continued iterative execution; otherwise, executing notification or ending interaction. According to the invention, defects and problems existing in related technologies of a conventional intelligent hospital guide system can be solved, i.e., how to analyze related disease information of oral disease description of a patient and mine disease information except the disease description of the patient is solved,natural man-machine interaction is realized, and the effect of the intelligent hospital guide system in practical application is improved.

The invention provides an international disease classification and marking method and device, a computer device and a storage medium. At least one target ICD code corresponding to acquired target disease description information to be marked is looked up in a pre-established parallel corpus, the final ICD codes selected from the looked-up at least one target ICD code are used for making the targetdisease description information according to set rules and are utilized to classified marking of the target disease description information to be marked. By adopting the international disease classification and marking method, the final ICD codes can be automatically for the target disease description information from the parallel corpus, the marking efficiency is improved, selection is performedaccording to the set rules, and the marking uniformity is ensured.

The invention provides a disease early warning method and system, relates to the technical field of the Internet of Things, and regards an Internet of Things server as an execution main body. The method includes steps: receiving disease description information transmitted by an acquisition terminal; processing the disease description information, and generating disease information; assessing the disease information according to a preset rule, and generating disease early warning information; and performing format conversion on the disease early warning information and transmitting the converted information to a user terminal to enable the user terminal to display the disease early warning information. According to the method and system, the Internet of Things server is adopted so that thedisease early warning time is shortened, and the disease prevention and control range is extended.

The invention provides a method and a device for modifying disease description, wherein, the method comprises: obtaining a target medical record homepage from a local or medical institution server; Identifying the front page of the target medical record to obtain various data in the front page of the medical record; Determining disease type characteristic data corresponding to the front page of the target medical record according to the data; Determining the disease species corresponding to the front page of the target medical record according to the disease species characteristic data and thedisease species classification model established based on the classification algorithm; Determining a disease description of the front page of the target medical record registered in the medical performance appraisal form, and modifying the disease description of the front page of the target medical record in the medical performance appraisal form to the disease type if the disease description isdifferent from the disease type corresponding to the front page of the target medical record. The technical proposal of the invention can correct the nonstandard description of the disease type in the front page of the medical record and facilitate the performance analysis according to the disease type corresponding to the front page of the medical record.

The embodiment of the invention discloses an intelligent hospital guide and registration method and device, equipment and a storage medium. The method comprises the following steps: receiving diseasepart information sent by terminal equipment, obtaining disease description information according to the disease part information, and sending the disease description information to the terminal equipment; receiving target disease description information sent by the terminal device, and establishing a target case; obtaining a matching case corresponding to the target case and a hospital guide record of the matching case from a preset case library; obtaining a number source matching result corresponding to the target case from a preset number source library according to the hospital guide record, and sending the number source matching result to the terminal equipment; and receiving a number source selection result sent by the terminal device, and completing registration according to the number source selection result. According to the embodiment of the invention, the user can accurately understand departments and doctors matched with own diseases and quickly obtain proper number sourcesto complete registration, so that the medical service efficiency is improved, and the medical experience of the user is improved.

The invention discloses a disease risk value determination method, for solving the problem that in the prior art, a patient difficultly finds and treats diseases in time, and consequently the death rate of the diseases is high. The disease risk value determination method includes the steps: acquiring a plurality of disease description data of a patient, input on a disease evaluation interface; determining a weight coefficient corresponding to each piece of disease description data; determining an interactive item coefficient of the target disease according to a matching result of each diseasedescription data and standard disease description data of the target disease; and determining the disease probability of the target disease according to the interaction item coefficient and each weight coefficient. The invention also discloses a disease risk value determination device, electronic equipment, and a computer readable storage medium.

The invention provides a program control device for controlling implantable medical equipment, electronic equipment and a system. The program control device comprises a reservation unit, a video unit,a data recording unit and a control unit, wherein the reservation unit is used for transmitting time information provided by a user to the external electronic equipment and receiving feedback information of the external electronic equipment according to the time information, the video unit is used for transmitting real-time video communication requests to the external electronic equipment and performing real-time video communication with the external electronic equipment when the external electronic equipment allows the real-time video communication requests, the data recording unit is used for acquiring disease description information inputted by a user and transmitting the disease description information to the external electronic equipment, and the control unit is used for receiving implantable medical equipment control parameters transmitted by the external electronic equipment and controlling the implantable medical equipment by the aid of the control parameters.

The invention relates to an intelligent doctor-patient interaction demonstrating method and system based on 3D. The intelligent doctor-patient interaction demonstrating method and system is designed for making ambiguous medical treatment knowledge specific and visual. The intelligent doctor-patient interaction demonstrating method based on 3D comprises the steps that after user identity verification is passed, inquirers' identity information, patients' identity information and patients' medical history information are obtained; demonstrating item selection results are obtained, standard demonstrating information corresponding to selected demonstrating items, selected by users, in a standard demonstrating base are called to user display terminals for display; modification operation instructions of the users to standard demonstrating information are obtained, final demonstrating information is generated and displayed on the user display terminals, wherein the final demonstrating information is displayed on the user display terminals, synchronous screen recording and / or sound recording is performed. The intelligent doctor-patient interaction demonstrating method is conductive to comprehensive disease description and provides a visual and understanding communication based for patients and makes doctors and patients better perform disease communication.

The invention discloses a patient consulting suggestion generation device and system. The patient consulting suggestion generation device includes a file generation module used for obtaining medical data of a consulting patient and generating an electronic file of the consulting patient in advance; a semantic analysis module used for obtaining consulting description of the consulting patient, making a semantic analysis of the consulting description, and obtaining sign description in the consulting description; a standardization module used for generating standard disease description accordingto the sign description; and a suggestion generation module used for generating consulting suggestions according to the disease description and electronic file of the consulting patient. According tothe patient consulting suggestion generation device, when feedback is made for consultation of the consulting patient, the current sign description about the state of an illness of the consulting patient is used for reference, the medical history of the consulting patient is also obtained through the electronic file corresponding to the consulting patient, and thus more comprehensive medical dataof the consulting patient can be synthesized, thereby improving accuracy and safety of consulting suggestions.

The invention relates to a patient data processing method which is characterized in that the method comprises when a patient sees a doctor for a first disease for the first time, generating first treatment data and a first disease file of the first disease, and storing the first treatment data to the first disease file, wherein the first treatment data comprises a first treatment doctor identification code, and the first disease file comprises patient information, disease description information and identification information, the identification information comprising a first disease identification code and a first-time doctor identification code; and when the patient sees a doctor for the first disease thereafter, generating follow-up treatment data of the first disease and storing the follow-up treatment data to the first disease file according to the time sequence, wherein the follow-up treatment data comprises follow-up treatment doctor identification codes. The patient data processing method can effectively processing the patient data.

The invention relates to an artificial intelligence technology, discloses a drug recommendation method, device based on artificial intelligence and related equipment, and is applied to the technical field of medical informatization. The method provided by the invention comprises the following steps: acquiring current diagnosis information of a patient, and acquiring symptom description from the current diagnosis information; inputting the disease description into the neural network model, and generating disease information corresponding to the disease description; based on the disease information, searching corresponding drugs in a preset drug knowledge base to obtain a candidate drug list; acquiring historical diagnosis information of the patient, and generating a patient portrait of the patient according to the diagnosis information and the historical diagnosis information; and on the basis of the patient portrait, through a keyword extraction algorithm, generating a medication taboo rule of the patient, and on the basis of the medication taboo rule, screening medicines in the candidate medicine list to obtain a recommended medicine list. The method is used for eliminating drugs having risks for patients, and the accuracy of generating drug recommendation information is improved.

The invention relates to a doctor-patient data processing method. The method includes the steps that when a patient sees a doctor for a first disease, doctor-seeing data of the first disease are generated and stored in a first disease file; the doctor-seeing data comprise a doctor identification code of a doctor to which the patient sees at that time, the first disease file comprises patient information, disease description information and recognition information, and the recognition information comprises a first disease identification code and the first-time doctor identification code; a doctor-patient database is formed by the doctor-seeing data corresponding to the doctor identification code; the doctor-patient database further comprises doctor information and identification information, and the identification information comprises a disease identification code. The doctor-patient data can be effectively processed through the doctor-patient data processing method.

The embodiment of the invention provides a health plan management method and device based on data analysis, a server and a medium. The method comprises the following steps: receiving a health plan formulation request sent by a terminal; determining a health plan for a target user according to medical record data of the target user corresponding to the terminal in the health plan formulation request,; sending the health plan for the target user to the terminal; determining a disease level of the target user according to the disease type and the disease description information included in the disease information of the medical record data of the target user; determining a health plan reminder strategy corresponding to the disease level of the target user according to a preset corresponding relationship between disease levels and health plan reminder strategies; and sending prompt information to the terminal according to the reminding frequency indicated by the determined health plan reminder strategy. The method and the device can automatically and intelligently formulate a health plan for a user, and can supervise the user to complete the health plan efficiently.

The invention discloses a department guidance method, device and equipment. Design foundation of the invention are that 1) a patient can autonomously perform department guidance query in a self-service manners, so a large amount of medical care workers are not needed; and 2) during autonomous self-service query, two-dimensional problems need to be solved together; i.e., confusion of patients aboutprofessional medical information needs to be eliminated, and the capability of a self-service system in identification and analysis of fuzzy query information input by a non-professional patient needs to be improved. Therefore, a main scheme provided by the invention comprises the following steps of: performing preliminary department identification and text validity judgment on an input inquiry text; directly processing department recommendation or illegal inquiry input in advance; and then in virtue of a double-strategy combination mode of text classification and key information retrieval, deeply analyzing effective disease description in the case that departments cannot be directly recommended is adopted so as to accurately and reliably recommend doctor-seeing departments and guide patients to see doctors smoothly. Therefore, the pressure of guidance consultation in hospitals is relieved.

The invention discloses a migratable medical inquiry dialogue system and method for a low-resource scene, and the system comprises: a simulation patient construction module which is used for buildinga simulation patient which is provided with a plurality of dialogue samples, randomly selects one dialogue sample for each training, and gives an illness state description report for the simulation patient; an intelligent medical inquiry system construction module which is used for establishing an intelligent medical inquiry system, extracting context information of different hierarchies from dialogue history through the layered context encoder module according to disease description, encoding to obtain feature vectors of single-round hierarchies and multi-round hierarchies of dialogues, and establishing an intelligent medical inquiry system by evolving an external medical knowledge graph, relationship reasoning between disease symptoms is performed according to the feature vectors to obtain graph node feature vectors to generate replies of doctors to patients by utilizing a replication network under the guidance of graph node information; and a training module which is used for carrying out end-to-end training on the system and training the system on dialogue data of existing diseases by utilizing a graph evolution meta-learning algorithm.

The invention provides an intelligent diagnosis method and system for common diseases and electronic equipment. The method comprises the steps: extracting a clinical symptom semantic net range corresponding to each common disease based on an obtained descriptive text of the common diseases of a patient; fusing clinical symptom semantic network ranges corresponding to common disease diagnosis standards in a standard library to obtain a corresponding clinical symptom range and a clinical symptom term set; based on the obtained clinical symptom range and clinical symptom term set corresponding to each common disease, constructing a clinical symptom semantic network model; and inputting clinical symptom feature information of a to-be-diagnosed disease into the constructed clinical symptom semantic network model, and judging a disease type and a corresponding solution of the to-be-diagnosed disease through the clinical symptom semantic network model. The clinical symptom semantic network model is constructed according to the clinical symptom information of the existing common diseases, and the common diseases to be diagnosed can be intelligently diagnosed through the constructed clinical symptom semantic network model.

The invention discloses a disease prediction method and device, a terminal and a storage medium. The method comprises steps that the disease description information of a to-be-detected user in a userinterface is obtained, and a target disease vector is generated according to the disease description information; a matched disease vector is screened from a disease vector set, the matched disease vector refers to a disease vector which is included in the disease vector set and matches the target disease vector; the disease information of the to-be-detected user is predicted according to the matched disease vector; the disease information of the to-be-detected user can be better predicted.