CN116153505B - Intelligent critical patient sign identification method and system based on medical pressure sensor - Google Patents
Intelligent critical patient sign identification method and system based on medical pressure sensor Download PDFInfo
- Publication number
- CN116153505B CN116153505B CN202310432157.0A CN202310432157A CN116153505B CN 116153505 B CN116153505 B CN 116153505B CN 202310432157 A CN202310432157 A CN 202310432157A CN 116153505 B CN116153505 B CN 116153505B
- Authority
- CN
- China
- Prior art keywords
- data
- sensor
- abnormal
- sample
- identification
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Pathology (AREA)
- Veterinary Medicine (AREA)
- Computational Linguistics (AREA)
- Surgery (AREA)
- Databases & Information Systems (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- Animal Behavior & Ethology (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Heart & Thoracic Surgery (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention discloses a critical patient sign intelligent identification method and system based on a medical pressure sensor, and relates to the technical field of data processing, wherein the method comprises the following steps: acquiring data of the medical pressure sensor through a data acquisition device to obtain a sensor data set; obtaining a sensor identification image set; basic information of a target critical patient is acquired, and feature extraction is carried out to obtain an influence feature data set; inputting the sensor data set and the sensor identification image set into a sensor abnormality identification model to obtain abnormality identification characteristics; based on the acquisition time, carrying out association elimination on the data abnormal characteristics to obtain the data abnormal characteristics to be identified; performing correction analysis to obtain a correction coefficient; obtaining a corrected data set after correction; and inputting the sign recognition model to obtain a sign recognition result. The invention solves the technical problems of large sign recognition error and low intelligent degree in the prior art, and achieves the technical effects of improving the data reliability and the recognition accuracy.
Description
Technical Field
The application relates to the technical field of data processing, in particular to an intelligent critical patient sign identification method and system based on a medical pressure sensor.
Background
With the rapid development of economy, the speed of updating and iterating the science and technology is gradually increased. And with the use of new materials, the stability of the pressure sensor is obviously improved, and the application range is wider and wider. With the improvement of the performance of the pressure sensor, the application of the medical pressure sensor in the medical industry is also developed.
At present, with the wide application of scientific instruments in clinic, the condition of identifying physical signs of critical patients by doctors is more and more dependent on data sent by the instruments, and the accuracy of the data has great influence on the accuracy of judgment. However, the medical pressure sensor has larger formed data errors due to the interference of the device and the influence of external factors in the use process, and the medical pressure sensor only depends on doctors to determine the data reliability, so that the time period is long and the accuracy cannot be ensured. In the prior art, the technical problems of large sign recognition error and low intelligent degree exist.
Disclosure of Invention
The application provides an intelligent critical patient sign recognition method and system based on a medical pressure sensor, which are used for solving the technical problems of large sign recognition error and low intelligent degree in the prior art.
In view of the above problems, the application provides a critical patient sign intelligent identification method and system based on a medical pressure sensor.
In a first aspect of the present application, there is provided a method for intelligently identifying critical patient symptoms based on a medical pressure sensor, wherein the method is applied to a sign identification system, the sign identification system being in communication connection with a sensor array and a data acquisition device, the method comprising:
acquiring data of the medical pressure sensors in the sensor array through a data acquisition device to obtain a sensor data set;
acquiring real-time images of the positions of the sensor arrays through image recognition equipment to obtain a sensor recognition image set;
basic information of a target critical patient is acquired, medical record information is extracted from the basic information by taking etiology as an index, and feature extraction is carried out on the medical record information according to influence feature indexes to obtain an influence feature data set;
inputting the sensor data set and the sensor identification image set into a sensor abnormality identification model to obtain abnormality identification features, wherein the abnormality identification features comprise data abnormality features and position abnormality features;
Based on the acquisition time, carrying out association elimination on the data abnormal characteristics by utilizing the position abnormal characteristics, and obtaining the data abnormal characteristics to be identified according to the elimination result;
carrying out correction analysis on the abnormal characteristics of the data to be identified and the influence characteristic data set, and obtaining correction coefficients according to correction analysis results;
performing data correction on the sensor data set by using the correction coefficient to obtain a corrected data set;
and inputting the corrected data set into a sign recognition model to obtain a sign recognition result.
In a second aspect of the application, there is provided a medical pressure sensor-based critical patient sign intelligent identification system, the system comprising:
the data acquisition module is used for acquiring data of the medical pressure sensors in the sensor array through the data acquisition device to obtain a sensor data set;
the identification image set acquisition module is used for acquiring real-time images of the positions of the sensor arrays through the image identification equipment to acquire a sensor identification image set;
the influence data set acquisition module is used for extracting basic information of a target critical patient by taking a cause as an index to obtain medical record information, and extracting features of the medical record information according to influence feature indexes to obtain an influence feature data set;
The abnormal feature obtaining module is used for inputting the sensor data set and the sensor identification image set into a sensor abnormal identification model to obtain abnormal identification features, wherein the abnormal identification features comprise data abnormal features and position abnormal features;
the to-be-identified feature obtaining module is used for carrying out association elimination on the data abnormal features by utilizing the position abnormal features based on acquisition time, and obtaining to-be-identified data abnormal features according to elimination results;
the correction coefficient obtaining module is used for carrying out correction analysis on the abnormal characteristics of the data to be identified and the influence characteristic data set, and obtaining a correction coefficient according to a correction analysis result;
the correction data set obtaining module is used for carrying out data correction on the sensor data set by utilizing the correction coefficient to obtain a correction data set;
and the sign recognition module is used for inputting the corrected data set into the sign recognition model to obtain a sign recognition result.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
The method comprises the steps of acquiring data of a medical pressure sensor in a sensor array through a data acquisition device, acquiring a sensor data set, acquiring a sensor identification image set through image identification equipment, further acquiring basic information of a target critical patient, extracting medical record information from the basic information by taking a cause as an index, extracting characteristics of the medical record information according to influence characteristic indexes, acquiring an influence characteristic data set, inputting the sensor data set and the sensor identification image set into a sensor abnormal identification model, acquiring abnormal identification characteristics, wherein the abnormal identification characteristics comprise data abnormal characteristics and position abnormal characteristics, then carrying out correlation elimination on the data abnormal characteristics by utilizing the position abnormal characteristics based on acquisition time, acquiring abnormal characteristics of the data to be identified according to elimination results, carrying out correction analysis on the data abnormal characteristics to be identified and the influence characteristic data set, acquiring correction coefficients according to correction analysis results, carrying out data correction on the sensor data set by utilizing the correction coefficients, and further obtaining a physical sign identification result according to the correction data set. The technical effects of improving the reliability of the data of the medical pressure sensor and improving the efficiency and accuracy of identifying critical patient signs in an intelligent identification mode are achieved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of an intelligent critical patient sign recognition method based on a medical pressure sensor provided by an embodiment of the application;
FIG. 2 is a schematic flow chart of obtaining an influence characteristic index in the critical patient sign intelligent identification method based on the medical pressure sensor provided by the embodiment of the application;
FIG. 3 is a schematic flow chart of generating a sensor abnormality recognition model in a critical patient sign intelligent recognition method based on a medical pressure sensor according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of an intelligent critical patient sign recognition system based on a medical pressure sensor according to an embodiment of the present application.
Reference numerals illustrate: the system comprises a data acquisition module 11, an identification image set acquisition module 12, an influence data set acquisition module 13, an abnormal feature acquisition module 14, a feature to be identified acquisition module 15, a correction coefficient acquisition module 16, a correction data set acquisition module 17 and a physical sign identification module 18.
Detailed Description
The application provides an intelligent critical patient sign recognition method and system based on a medical pressure sensor, which are used for solving the technical problems of large sign recognition error and low intelligent degree in the prior art.
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. It will be apparent that the described embodiments are only some, but not all, embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
It should be noted that the terms "comprises" and "comprising," along with any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or server that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus, but may include other steps or modules not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
As shown in fig. 1, the present application provides a critical patient sign intelligent identification method based on a medical pressure sensor, wherein the method is applied to a sign identification system, the sign identification system is in communication connection with a sensor array and a data acquisition device, and the method comprises:
Step S100: acquiring data of the medical pressure sensors in the sensor array through a data acquisition device to obtain a sensor data set;
specifically, the sensor array is a sensor combination formed by disposing medical pressure sensors according to a certain arrangement rule so as to comprehensively acquire data. The data acquisition device is a device which acquires data acquired by the sensor in real time after being in communication connection with the sensor in the sensor array through a port, such as an industrial intelligent gateway.
Specifically, the data acquisition device is used for acquiring the medical pressure sensor data in the sensor array in real time, and uploading the acquired data to the sign recognition system for storage, so that reliable analysis data, namely the sensor data set, is provided for the follow-up. Preferably, the pressure data of the critical patient is acquired according to the sensor array, so that corresponding data is obtained, and the obtained data is formed into the sensor data set, wherein the sensor data set comprises arterial pressure data, central venous pressure data, pulmonary artery pressure data and the like.
Step S200: acquiring real-time images of the positions of the sensor arrays through image recognition equipment to obtain a sensor recognition image set;
Specifically, the image recognition device is a device for acquiring images at distributed positions of the sensor array in real time, and includes a camera, an infrared camera, and the like. The set of sensor identification images is obtained by acquiring a placement location image of the sensor array with the image recognition device. Wherein the sensor identification image set is an image set reflecting the placement position of the sensor in real time. The technical effect of providing basic analysis data for subsequent analysis is achieved.
Step S300: basic information of a target critical patient is acquired, medical record information is extracted from the basic information by taking etiology as an index, and feature extraction is carried out on the medical record information according to influence feature indexes to obtain an influence feature data set;
further, as shown in fig. 2, step S300 of the embodiment of the present application further includes:
step S310: collecting deviation rectifying information from a database of a target department, wherein the deviation rectifying information is data adjustment information recorded in a medical pressure sensor use record and comprises adjustment items, adjustment reasons and adjustment parameters;
step S320: judging whether the adjustment items in the deviation correction information are adjusted due to the influence of the illness state of the patient or not based on the adjustment reasons, if so, setting the adjustment items as items to be analyzed;
Step S330: and extracting the item association indexes of the items to be analyzed, and obtaining the influence characteristic indexes according to the extraction results.
Specifically, the target critically ill patient is any patient for which sign identification is required. The basic information is information which can reflect the difference of target critical patients and the characteristics of other critical patients and is obtained after description, and comprises names, ages, etiologies, admission time and the like. Preferably, the medical record information is obtained by extracting data from the basic information by taking the cause as an index. The medical record information is a medical document describing the health condition of a target critical patient and comprises information such as objective description of the disease condition, objective examination results of medical staff on the patient, disease condition analysis results and the like.
Specifically, the influence characteristic index information is an index determined according to influence generated in the process of data acquisition of the medical pressure sensor according to the illness state of a patient, and comprises a placement position influence characteristic, a behavior influence characteristic and the like. Preferably, since the medical pressure sensor cannot be placed in the standard position due to the condition of the patient in practice, a deviation exists between the measured data of the medical pressure sensor and the actual data, and the deviation is unavoidable, so that in order to ensure the accuracy of the data for sign recognition, the influence of the influence characteristic index on the sensor data needs to be considered. The placement position influencing feature is a feature for adjusting the placement position of the medical pressure sensor, and comprises an adjustment position, an adjustment distance and the like. The behavior influence features are features for describing abnormal behaviors of critical patients due to disease influence, and include influence amplitude and the like. For example, when the wrist of a critically ill patient is injured, a sensor which should be originally placed on the wrist needs to be placed on the upper side of the wrist, and due to the movement of the measurement position, the measured data cannot be analyzed according to the original analysis standard, that is, whether the current sensor data is qualified or not cannot be judged according to the original data range, and the data range needs to be adjusted, so that the analysis is not in compliance with the facts.
Specifically, the correction is used as an index to collect information from a database of a target department, and the correction information obtained is data adjustment information recorded in the medical pressure sensor usage record and comprises adjustment items, adjustment reasons and adjustment parameters. That is, the deviation correcting information is relevant information for the medical staff to adjust the sensor according to the actual situation in the actual working process. The adjustment items are information describing the content of sensor adjustment, and include adjustment positions, sensor model exchange, and the like. The adjustment reason is information describing the reason why the sensor needs to be adjusted. The adjustment parameters are parameters describing actual operation dimensions of sensor adjustment, and include adjustment of distance, model before and after replacement, and the like.
Specifically, the data in the adjustment reason is extracted by using keywords, which may be adjustment, patient, behavior disturbance, etc., by locating the relevant position in the correction information according to the keywords, so as to analyze the content at the relevant position, determine whether the adjustment item in the correction information is adjusted due to the influence of the patient's illness, if yes, indicate that the correction information is sensor-adjusted according to the factors of the patient, and set the adjustment item as the item to be analyzed. Wherein the items to be analyzed are items for which related indexes need to be further analyzed. The measurement index related to the items to be analyzed is used as an influence characteristic index. The influence characteristic index is an index influenced by the reason of the illness state after the sensor needs to be adjusted.
Specifically, after the impact characteristic index is obtained, characteristic extraction is performed on the medical record information according to the index, namely, information acquisition is performed on data related to the index in the medical record information according to the impact characteristic index, so as to obtain an impact characteristic data set of the target critical patient. Providing basis for data correction in the follow-up process.
Step S400: inputting the sensor data set and the sensor identification image set into a sensor abnormality identification model to obtain abnormality identification features, wherein the abnormality identification features comprise data abnormality features and position abnormality features;
further, as shown in fig. 3, step S400 of the embodiment of the present application further includes:
step S410: obtaining N sample sensor data sets, N sample sensor identification image sets, N sample data abnormal features and N position abnormal features as sample data sets, wherein the N sample sensor data sets are in one-to-one correspondence with the N sample sensor identification image sets, and N is an integer greater than or equal to 1;
step S420: constructing a sensor data anomaly analysis sub-model by utilizing the N sample sensor data sets and N sample data anomaly characteristics;
step S430: utilizing the N sample sensor identification image sets and N position abnormality features to construct a sensor position abnormality analysis sub-model;
Step S440: and generating the sensor abnormality recognition model according to the sensor data abnormality analysis sub-model and the sensor position abnormality analysis sub-model.
Specifically, the sensor abnormality recognition model is a model for intelligently recognizing abnormal conditions of data obtained by a sensor and a sensor placement position, and comprises a sensor data abnormality analysis sub-model and a sensor position abnormality analysis sub-model. The sensor data anomaly analysis sub-model is a sub-model which takes a sensor data set as input data and takes data anomaly characteristics as output data, and identifies the anomaly data of the sensor. The sensor position abnormality analysis sub-model is a sub-model which takes a sensor identification image set as input data and position abnormality characteristics as output data, and identifies abnormal data of the sensor. The data anomaly characteristic is a characteristic for describing the degree of exceeding the data range specified by the index, and comprises a numerical value exceeding range, a data fluctuation characteristic and the like. The position abnormality feature is a feature describing the degree of difference between the sensor and the prescribed position, including the moving distance and the like.
Specifically, by collecting sensor data of a target department in a historical time period, N sample sensor data sets, N sample sensor identification image sets, N sample data abnormal features and N position abnormal features are obtained, and the obtained data are used as sample data sets. Wherein the sample dataset is used to construct a sensor anomaly identification model.
Specifically, the sensor position abnormality analysis submodel constructed by taking the convolutional neural network as a model is supervised and trained by utilizing the N sample sensor identification image sets and N position abnormality features until the submodel converges, so as to obtain the sensor abnormality identification model meeting the requirements.
Further, step S400 of the embodiment of the present application further includes:
step S450: randomly selecting one sample sensor data set from the N sample sensor data sets without returning to serve as a first dividing node, and performing two-classification on the N sample sensor data sets according to a preset numerical threshold value to obtain a first dividing result;
step S460: randomly selecting one sample sensor data set from the N sample sensor data sets without returning as a second dividing node, and performing two-classification on the first dividing result according to a preset numerical threshold value to obtain a second dividing result;
Step S470: randomly selecting a sample sensor data set from the N sample sensor data sets without returning to serve as a P division node, and performing second classification on the P-1 division result according to a preset numerical threshold value to obtain the P division result;
step S480: and identifying the P-th division result by utilizing the N sample data abnormal characteristics, and constructing the sensor data abnormal analysis sub-model according to the identification result, the first division node, the second division node and the P-th division node.
Specifically, a sample sensor data set is selected randomly from the N sample sensor data sets without being put back as a first dividing node, the first dividing node is based on the selected sample sensor data set, the data set with the data difference value of the selected sample sensor data set in the preset value threshold is divided into one type based on the preset value threshold, and the data set not in the preset value threshold range is divided into one type, so that two classification is performed, and a first dividing result is obtained. The first division result is a two-class set. By not replacing the randomly selected sample sensor data set, the sample sensor data set which is already used as the division basis can not be reused in the subsequent division process, so that the accuracy of division can be ensured, and invalid division caused by repeated division basis is avoided. Wherein the first division results divide the N sample sensor datasets into two categories. By dividing the data according to the range of the numerical value fluctuation, the data set can be divided more quickly and accurately. The preset value threshold is set by a worker, and is not limited herein.
Specifically, a sample sensor data set is selected from the N sample sensor data sets as a second dividing node without being replaced by the N sample sensor data sets, and similarly, the first dividing result is subjected to two-class classification according to a preset numerical threshold value to obtain a second dividing result. The second division result is to divide the first division result again, that is to say, the second division result is a three-classification set. And then, carrying out second classification on the P-1 division result by using the P division node so as to obtain the P division result. The set corresponding to the P-th division result is a p+1 classification set. Preferably, the total data set corresponding to the first division result, the second division result and the P-th division result is still the N sample sensor data sets.
And further, carrying out fusion summarization identification on the data abnormal characteristics corresponding to each partitioning result in the P-th partitioning result by utilizing the N sample data abnormal characteristics, namely fusing the same data abnormal characteristics, summarizing different data abnormal characteristics, and identifying each partitioning result according to the summarized result. And constructing the sensor data anomaly analysis sub-model according to the identification result, the first partition node, the second partition node and the P-th partition node. The technical effect of quickly obtaining data abnormal characteristics according to the data and improving the data analysis efficiency is achieved.
Step S500: based on the acquisition time, carrying out association elimination on the data abnormal characteristics by utilizing the position abnormal characteristics, and obtaining the data abnormal characteristics to be identified according to the elimination result;
further, step S500 of the embodiment of the present application further includes:
step S510: constructing a position-data anomaly mapping relation according to the acquisition time of the position anomaly characteristic and the data anomaly characteristic;
step S520: based on the mapping relation with the empty mapping relation, the mapping relation is called, and the data abnormal characteristics corresponding to the mapping relation with the empty mapping relation are added into the data abnormal characteristics to be identified;
step S530: extracting a mapping relation successfully constructed in the position-data abnormal mapping relation, judging whether the data abnormal characteristics are caused by the position abnormal characteristics, and if so, eliminating the corresponding data abnormal characteristics;
step S540: if not, adding the corresponding data abnormal characteristics into the data abnormal characteristics to be identified.
Specifically, according to the time of data acquisition, the position abnormal characteristics are utilized to correlate and reject the data abnormal characteristics, so that the data abnormal characteristics are subjected to dimension reduction processing, and the purpose of reducing interference data is achieved. The data abnormal characteristics to be identified are data abnormal characteristics which need to be further corrected and analyzed.
Specifically, a time axis is constructed according to the acquisition time of the position abnormal feature and the data abnormal feature and the corresponding time point, the position abnormal feature is arranged above the time axis, the data abnormal feature is arranged below the time axis, and the position abnormal feature and the data abnormal feature are input into the time axis to obtain the position-data abnormal mapping relation. The position-data anomaly mapping relation reflects the association degree of the position anomaly characteristic and the data anomaly characteristic in time.
Specifically, based on the mapping relation that the position-data abnormality mapping relation is empty, the mapping relation is empty, which indicates that the data abnormality feature and the position abnormality feature do not occur simultaneously, the data abnormality is not caused by the position abnormality, and the corresponding data abnormality feature is added into the data abnormality feature to be identified.
Specifically, a successful mapping relation is constructed in the position-data abnormality mapping relation, whether the data abnormality reflected by the data abnormality features is caused by the position abnormality features is analyzed according to the position abnormality conditions reflected by the position abnormality features, if so, the corresponding data abnormality features are removed, the dimension of the data is reduced, and if not, the data abnormality features are not caused by the position abnormality. The technical effect of reducing the dimension of the abnormal data features is achieved.
Step S600: carrying out correction analysis on the abnormal characteristics of the data to be identified and the influence characteristic data set, and obtaining correction coefficients according to correction analysis results;
step S700: performing data correction on the sensor data set by using the correction coefficient to obtain a corrected data set;
further, step S600 of the embodiment of the present application further includes:
step S610: performing index matching according to the influence characteristic index and the abnormal characteristic of the data to be identified to obtain an index matching result;
step S620: determining the adjustment amplitude of the influence characteristic index according to the adjustment parameters and the influence characteristic data set, and generating a correction coefficient set;
step S630: and carrying out coefficient matching from the correction coefficient set according to the index matching result to obtain the correction coefficient.
Specifically, the correction analysis is to correct abnormal data caused by patient reasons in abnormal characteristics of the data to be identified, so as to obtain data capable of reflecting actual conditions of patient signs. The correction coefficient is a coefficient for adjusting abnormal characteristics of the data to be identified. And performing index matching according to the influence characteristic index and the abnormal characteristic of the data to be identified to obtain an index matching result, wherein the index matching result is an index successfully matched after the monitoring index corresponding to the abnormal characteristic of the data to be identified is matched with the influence characteristic index. And determining an adjustment amplitude according to the numerical value difference degree between the data set corresponding to the adjustment parameter and the influence characteristic data set, and determining the adjustment amplitude as a correction coefficient set. And then acquiring the corresponding correction coefficient from the correction coefficient set according to the index in the index matching result. And correcting the data affected by the illness state of the patient in the sensor data set according to the correction coefficient, thereby obtaining a correction data set. Wherein the corrected data set is a data set from which the sensor's positional factors and patient's effects are eliminated. The technical effect of providing reliable and accurate analysis data is achieved.
Step S800: and inputting the corrected data set into a sign recognition model to obtain a sign recognition result.
Further, step S800 of the embodiment of the present application further includes:
step S810: based on a BP neural network, constructing the sign recognition model, wherein input data of the sign recognition model is a correction data set, and output data is a feature recognition result;
step S820: acquiring critical patient sign identification data of a target department in a past time period to obtain a plurality of sample correction data sets and a plurality of sample feature identification results, and performing data annotation on the plurality of sample feature identification results to obtain a correction sample data set;
step S830: and performing supervision training on the sign recognition model by adopting the correction sample data set, and updating network parameters of the sign recognition model according to the output error of the sign recognition model until convergence conditions are reached to obtain the sign recognition model after training is completed.
Specifically, the BP neural network is taken as a basic framework, the sign recognition model is constructed, input data of the sign recognition model is a correction data set, and output data is a feature recognition result. The sign recognition model is a functional model for determining the sign of the target critical patient by intelligently recognizing the sensor data in the correction data set.
Specifically, the plurality of sample feature recognition results are subjected to data labeling, and a plurality of sample correction data sets and labeled plurality of sample feature recognition results are used as correction sample data sets. And training the basic framework by using the corrected sample data set, monitoring by using the marked multiple sample feature recognition results, updating the network parameters of the sign recognition model according to the output errors of the model in the training process, and obtaining the trained sign recognition model by continuous training and continuous updating of the parameters until the model is converged. The intelligent identification of critical patient signs is achieved, and the technical effects of improving the identification accuracy and efficiency are achieved.
In summary, the embodiment of the application has at least the following technical effects:
according to the method, data of the medical pressure sensor are obtained and serve as objects for analysis and correction, whether the data are correctly screened from the position of the sensor or not is further judged, abnormal interference data caused by improper positions are removed, further unavoidable interference in a monitoring process is subjected to data correction by utilizing an influence characteristic data set, a correction data set is obtained, and then the correction data set is identified by utilizing an intelligent sign identification model, so that a sign identification result is obtained. The technical effects of improving the recognition efficiency and the data reliability are achieved.
Example 2
Based on the same inventive concept as the critical patient sign intelligent recognition method based on the medical pressure sensor in the foregoing embodiment, as shown in fig. 4, the present application provides a critical patient sign intelligent recognition system based on the medical pressure sensor, and the system and method embodiments in the embodiments of the present application are based on the same inventive concept. Wherein the system comprises:
the data acquisition module 11 is used for acquiring data of the medical pressure sensors in the sensor array through the data acquisition device to obtain a sensor data set;
the identification image set obtaining module 12 is used for carrying out real-time image acquisition on the position of the sensor array through image identification equipment to obtain a sensor identification image set;
the influence data set acquisition module 13 is used for acquiring basic information of a target critical patient, extracting medical record information from the basic information by taking a cause as an index, and extracting features of the medical record information according to influence feature indexes to acquire an influence feature data set;
an anomaly characteristic obtaining module 14, wherein the anomaly characteristic obtaining module 14 is configured to input the sensor data set and the sensor identification image set into a sensor anomaly identification model to obtain anomaly identification characteristics, and the anomaly identification characteristics include data anomaly characteristics and position anomaly characteristics;
The feature to be identified obtaining module 15, wherein the feature to be identified obtaining module 15 is used for carrying out association elimination on the data abnormal features by utilizing the position abnormal features based on acquisition time, and obtaining the data abnormal features to be identified according to elimination results;
the correction coefficient obtaining module 16 is configured to perform correction analysis on the to-be-identified data abnormal feature and the influencing feature data set, and obtain a correction coefficient according to a correction analysis result;
a correction data set obtaining module 17, where the correction data set obtaining module 17 is configured to perform data correction on the sensor data set using the correction coefficient to obtain a correction data set;
the sign recognition module 18 is configured to input the sign recognition module 18 into the sign recognition model according to the correction data set, and obtain a sign recognition result.
Further, the system further comprises:
the data adjustment information acquisition unit is used for acquiring deviation correction information of a database of a target department, wherein the deviation correction information is data adjustment information recorded in a medical pressure sensor use record and comprises adjustment items, adjustment reasons and adjustment parameters;
The influence judging unit is used for judging whether the adjustment items in the deviation correcting information are adjusted due to the influence of the illness state of the patient or not based on the adjustment reason, and if yes, the adjustment items are set as items to be analyzed;
and the characteristic index obtaining unit is used for extracting the item association index of the item to be analyzed and obtaining the influence characteristic index according to the extraction result.
Further, the system further comprises:
the sample data set setting unit is used for obtaining N sample sensor data sets, N sample sensor identification image sets, N sample data abnormal features and N position abnormal features as sample data sets, wherein the N sample sensor data sets are in one-to-one correspondence with the N sample sensor identification image sets, and N is an integer greater than or equal to 1;
the data submodel construction unit is used for constructing a sensor data anomaly analysis submodel by utilizing the N sample sensor data sets and N sample data anomaly characteristics;
a position sub-model construction unit for constructing a sensor position abnormality analysis sub-model by using the N sample sensor identification image sets and N position abnormality features;
And the recognition model generation unit is used for generating the sensor abnormality recognition model according to the sensor data abnormality analysis sub-model and the sensor position abnormality analysis sub-model.
Further, the system further comprises:
the first dividing node setting unit is used for randomly selecting one sample sensor data set from the N sample sensor data sets without returning the N sample sensor data sets as a first dividing node, and performing two-classification on the N sample sensor data sets according to a preset numerical threshold value to obtain a first dividing result;
the second dividing node setting unit is used for randomly selecting one sample sensor data set from the N sample sensor data sets without returning the N sample sensor data sets as a second dividing node, and performing two-classification on the first dividing result according to a preset numerical threshold value to obtain a second dividing result;
the P-th partition node setting unit is used for randomly selecting a sample sensor data set from the N sample sensor data sets without returning the N sample sensor data sets as P-th partition nodes, and carrying out second classification on the P-1-th partition result according to a preset numerical threshold value to obtain a P-th partition result;
The data anomaly analysis sub-model building unit is used for identifying the P-th division result by utilizing the N sample data anomaly characteristics and building the sensor data anomaly analysis sub-model according to the identification result, the first division node, the second division node and the P-th division node.
Further, the system further comprises:
the abnormal mapping relation construction unit is used for constructing a position-data abnormal mapping relation according to the acquisition time of the position abnormal characteristics and the data abnormal characteristics;
the feature adding unit is used for extracting a mapping relation with an empty mapping relation based on the position-data abnormal mapping relation and adding the data abnormal feature corresponding to the mapping relation with the empty mapping relation into the data abnormal feature to be identified;
the feature eliminating unit is used for extracting a mapping relation successfully constructed in the position-data abnormal mapping relation, judging whether the data abnormal feature is caused by the position abnormal feature, and eliminating the corresponding data abnormal feature if the data abnormal feature is caused by the position abnormal feature;
and the abnormal feature adding unit is used for adding the corresponding data abnormal feature into the data abnormal feature to be identified if not.
Further, the system further comprises:
the index matching unit is used for carrying out index matching according to the influence characteristic index and the data to be identified abnormal characteristic to obtain an index matching result;
the correction coefficient set generation unit is used for determining the adjustment amplitude of the influence characteristic index according to the adjustment parameter and the influence characteristic data set and generating a correction coefficient set;
and the coefficient matching unit is used for carrying out coefficient matching from the correction coefficient set according to the index matching result to obtain the correction coefficient.
Further, the system further comprises:
the system comprises a sign recognition model construction unit, a feature recognition model generation unit and a feature recognition unit, wherein the sign recognition model construction unit is used for constructing the sign recognition model based on a BP neural network, input data of the sign recognition model is a correction data set, and output data is a feature recognition result;
the correction data set obtaining unit is used for collecting critical patient sign identification data of a target department in a past time period to obtain a plurality of sample correction data sets and a plurality of sample feature identification results, and carrying out data labeling on the plurality of sample feature identification results to obtain a correction sample data set;
And the supervision and training unit is used for performing supervision and training on the sign recognition model by adopting the correction sample data set, and updating the network parameters of the sign recognition model according to the output error of the sign recognition model until convergence conditions are reached, so as to obtain the sign recognition model after training is completed.
It should be noted that the sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
The foregoing description of the preferred embodiments of the application is not intended to limit the application to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the application are intended to be included within the scope of the application.
The specification and figures are merely exemplary illustrations of the present application and are considered to cover any and all modifications, variations, combinations, or equivalents that fall within the scope of the application. It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the scope of the application. Thus, the present application is intended to include such modifications and alterations insofar as they come within the scope of the application or the equivalents thereof.
Claims (8)
1. The utility model provides a critical patient sign intelligent identification method based on medical pressure sensor, its characterized in that, the method is applied to sign identification system, sign identification system and sensor array and data acquisition device communication connection, the method includes:
acquiring data of the medical pressure sensors in the sensor array through a data acquisition device to obtain a sensor data set;
acquiring real-time images of the positions of the sensor arrays through image recognition equipment to obtain a sensor recognition image set;
basic information of a target critical patient is acquired, medical record information is extracted from the basic information by taking etiology as an index, and feature extraction is carried out on the medical record information according to influence feature indexes to obtain an influence feature data set;
Inputting the sensor data set and the sensor identification image set into a sensor abnormality identification model to obtain abnormality identification features, wherein the abnormality identification features comprise data abnormality features and position abnormality features;
based on the acquisition time, carrying out association elimination on the data abnormal characteristics by utilizing the position abnormal characteristics, and obtaining the data abnormal characteristics to be identified according to the elimination result;
carrying out correction analysis on the abnormal characteristics of the data to be identified and the influence characteristic data set, and obtaining correction coefficients according to correction analysis results;
performing data correction on the sensor data set by using the correction coefficient to obtain a corrected data set;
and inputting the corrected data set into a sign recognition model to obtain a sign recognition result.
2. The method of claim 1, wherein the method comprises:
collecting deviation rectifying information from a database of a target department, wherein the deviation rectifying information is data adjustment information recorded in a medical pressure sensor use record and comprises adjustment items, adjustment reasons and adjustment parameters;
judging whether the adjustment items in the deviation correction information are adjusted due to the influence of the illness state of the patient or not based on the adjustment reasons, if so, setting the adjustment items as items to be analyzed;
And extracting the item association indexes of the items to be analyzed, and obtaining the influence characteristic indexes according to the extraction results.
3. The method of claim 1, wherein the method comprises:
obtaining N sample sensor data sets, N sample sensor identification image sets, N sample data abnormal features and N position abnormal features as sample data sets, wherein the N sample sensor data sets are in one-to-one correspondence with the N sample sensor identification image sets, and N is an integer greater than or equal to 1;
constructing a sensor data anomaly analysis sub-model by utilizing the N sample sensor data sets and N sample data anomaly characteristics;
utilizing the N sample sensor identification image sets and N position abnormality features to construct a sensor position abnormality analysis sub-model;
and generating the sensor abnormality recognition model according to the sensor data abnormality analysis sub-model and the sensor position abnormality analysis sub-model.
4. A method according to claim 3, wherein the method comprises:
randomly selecting one sample sensor data set from the N sample sensor data sets without returning to serve as a first dividing node, and performing two-classification on the N sample sensor data sets according to a preset numerical threshold value to obtain a first dividing result;
Randomly selecting one sample sensor data set from the N sample sensor data sets without returning as a second dividing node, and performing two-classification on the first dividing result according to a preset numerical threshold value to obtain a second dividing result;
randomly selecting a sample sensor data set from the N sample sensor data sets without returning to serve as a P division node, and performing second classification on the P-1 division result according to a preset numerical threshold value to obtain the P division result;
and identifying the P-th division result by utilizing the N sample data abnormal characteristics, and constructing the sensor data abnormal analysis sub-model according to the identification result, the first division node, the second division node and the P-th division node.
5. The method of claim 1, wherein the method comprises:
constructing a position-data anomaly mapping relation according to the acquisition time of the position anomaly characteristic and the data anomaly characteristic;
based on the mapping relation with the empty mapping relation, the mapping relation is called, and the data abnormal characteristics corresponding to the mapping relation with the empty mapping relation are added into the data abnormal characteristics to be identified;
extracting a mapping relation successfully constructed in the position-data abnormal mapping relation, judging whether the data abnormal characteristics are caused by the position abnormal characteristics, and if so, eliminating the corresponding data abnormal characteristics;
If not, adding the corresponding data abnormal characteristics into the data abnormal characteristics to be identified.
6. The method according to claim 2, wherein the method comprises:
performing index matching according to the influence characteristic index and the abnormal characteristic of the data to be identified to obtain an index matching result;
determining the adjustment amplitude of the influence characteristic index according to the adjustment parameters and the influence characteristic data set, and generating a correction coefficient set;
and carrying out coefficient matching from the correction coefficient set according to the index matching result to obtain the correction coefficient.
7. The method of claim 1, wherein the method comprises:
based on a BP neural network, constructing the sign recognition model, wherein input data of the sign recognition model is a correction data set, and output data is a feature recognition result;
acquiring critical patient sign identification data of a target department in a past time period to obtain a plurality of sample correction data sets and a plurality of sample feature identification results, and performing data annotation on the plurality of sample feature identification results to obtain a correction sample data set;
and performing supervision training on the sign recognition model by adopting the correction sample data set, and updating network parameters of the sign recognition model according to the output error of the sign recognition model until convergence conditions are reached to obtain the sign recognition model after training is completed.
8. Intelligent critical patient sign recognition system based on medical pressure sensor, its characterized in that, the system includes:
the data acquisition module is used for acquiring data of the medical pressure sensors in the sensor array through the data acquisition device to obtain a sensor data set;
the identification image set acquisition module is used for acquiring real-time images of the positions of the sensor arrays through the image identification equipment to acquire a sensor identification image set;
the influence data set acquisition module is used for extracting basic information of a target critical patient by taking a cause as an index to obtain medical record information, and extracting features of the medical record information according to influence feature indexes to obtain an influence feature data set;
the abnormal feature obtaining module is used for inputting the sensor data set and the sensor identification image set into a sensor abnormal identification model to obtain abnormal identification features, wherein the abnormal identification features comprise data abnormal features and position abnormal features;
the to-be-identified feature obtaining module is used for carrying out association elimination on the data abnormal features by utilizing the position abnormal features based on acquisition time, and obtaining to-be-identified data abnormal features according to elimination results;
The correction coefficient obtaining module is used for carrying out correction analysis on the abnormal characteristics of the data to be identified and the influence characteristic data set, and obtaining a correction coefficient according to a correction analysis result;
the correction data set obtaining module is used for carrying out data correction on the sensor data set by utilizing the correction coefficient to obtain a correction data set;
and the sign recognition module is used for inputting the corrected data set into the sign recognition model to obtain a sign recognition result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310432157.0A CN116153505B (en) | 2023-04-21 | 2023-04-21 | Intelligent critical patient sign identification method and system based on medical pressure sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310432157.0A CN116153505B (en) | 2023-04-21 | 2023-04-21 | Intelligent critical patient sign identification method and system based on medical pressure sensor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116153505A CN116153505A (en) | 2023-05-23 |
| CN116153505B true CN116153505B (en) | 2023-08-18 |
Family
ID=86341104
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310432157.0A Active CN116153505B (en) | 2023-04-21 | 2023-04-21 | Intelligent critical patient sign identification method and system based on medical pressure sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116153505B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116933380A (en) * | 2023-09-12 | 2023-10-24 | 北京装库创意科技有限公司 | House pattern disassembly analysis method and system based on deep learning |
| CN117119646B (en) * | 2023-10-20 | 2024-08-16 | 南通医疗器械有限公司 | Shadowless lamp regulating and controlling method and system |
| CN117213532B (en) * | 2023-11-07 | 2024-01-23 | 东腾盛达科技（天津）有限公司 | Multifunctional microfluidic flexible sensor |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105249925A (en) * | 2015-09-25 | 2016-01-20 | 刘垚 | Traditional Chinese medicine pulse manifestation acquisition apparatus, noise reduction system and noise reduction method |
| CN107714023A (en) * | 2017-11-27 | 2018-02-23 | 乐普（北京）医疗器械股份有限公司 | Static ecg analysis method and apparatus based on artificial intelligence self study |
| CN108309262A (en) * | 2018-02-24 | 2018-07-24 | 乐普（北京）医疗器械股份有限公司 | Multi-parameter monitoring data analysing method and multi-parameter monitor |
| CN108742587A (en) * | 2018-06-20 | 2018-11-06 | 博动医学影像科技（上海）有限公司 | The method and device of flow characteristic value is obtained based on history information |
| CN113990500A (en) * | 2021-10-22 | 2022-01-28 | 深圳市艾美迪电子科技有限公司 | Vital sign parameter monitoring method and device and storage medium |
| CN114617535A (en) * | 2022-04-28 | 2022-06-14 | 深圳中康华盛生物医疗科技有限公司 | Ward patient vital sign wireless monitoring system |
| CN115482931A (en) * | 2022-09-16 | 2022-12-16 | 北京慧养道健康科技有限公司 | Life early warning system based on sensor acquisition |
| CN115530783A (en) * | 2022-09-21 | 2022-12-30 | 广东乐心医疗电子股份有限公司 | Blood pressure measuring method and device and electronic sphygmomanometer |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10750966B2 (en) * | 2016-03-31 | 2020-08-25 | Zoll Medical Corporation | Systems and methods of patient data compression |
| US20210398667A1 (en) * | 2020-06-18 | 2021-12-23 | Robb Takeshi Fujioka | Systems, methods, and apparatuses for secure diagnosis and treatments and authenticating diagnosed users |
| CA3193680A1 (en) * | 2020-09-25 | 2022-03-31 | Ian Shadforth | Method and system to assess disease using multi-sensor signals |
-
2023
- 2023-04-21 CN CN202310432157.0A patent/CN116153505B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105249925A (en) * | 2015-09-25 | 2016-01-20 | 刘垚 | Traditional Chinese medicine pulse manifestation acquisition apparatus, noise reduction system and noise reduction method |
| CN107714023A (en) * | 2017-11-27 | 2018-02-23 | 乐普（北京）医疗器械股份有限公司 | Static ecg analysis method and apparatus based on artificial intelligence self study |
| CN108309262A (en) * | 2018-02-24 | 2018-07-24 | 乐普（北京）医疗器械股份有限公司 | Multi-parameter monitoring data analysing method and multi-parameter monitor |
| CN108742587A (en) * | 2018-06-20 | 2018-11-06 | 博动医学影像科技（上海）有限公司 | The method and device of flow characteristic value is obtained based on history information |
| CN113990500A (en) * | 2021-10-22 | 2022-01-28 | 深圳市艾美迪电子科技有限公司 | Vital sign parameter monitoring method and device and storage medium |
| CN114617535A (en) * | 2022-04-28 | 2022-06-14 | 深圳中康华盛生物医疗科技有限公司 | Ward patient vital sign wireless monitoring system |
| CN115482931A (en) * | 2022-09-16 | 2022-12-16 | 北京慧养道健康科技有限公司 | Life early warning system based on sensor acquisition |
| CN115530783A (en) * | 2022-09-21 | 2022-12-30 | 广东乐心医疗电子股份有限公司 | Blood pressure measuring method and device and electronic sphygmomanometer |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116153505A (en) | 2023-05-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116153505B (en) | Intelligent critical patient sign identification method and system based on medical pressure sensor | |
| CN112508105B (en) | Fault detection and retrieval method for oil extraction machine | |
| US20040199482A1 (en) | Systems and methods for automatic and incremental learning of patient states from biomedical signals | |
| CN117542474A (en) | Remote nursing monitoring system and method based on big data | |
| CN113822366A (en) | Service index abnormality detection method and device, electronic equipment and storage medium | |
| CN118280562B (en) | Knowledge graph driven medical large model diagnosis method | |
| CN107358019B (en) | Recommendation method for concept-shifted medical solutions | |
| US20220246302A1 (en) | Data analysis apparatus, data analysis method, and data analysis program | |
| CN115691722A (en) | Quality control method, apparatus, device, medium and program product for medical data detection | |
| CN112702408A (en) | Internet of things system and method based on multi-sensing function | |
| CN110336860A (en) | Key node data guard method based on multidimensional data processing in industrial Internet of Things | |
| CN118280579B (en) | Sepsis patient condition assessment method and system based on multi-mode data fusion | |
| US20220208383A1 (en) | Method and system for mental index prediction | |
| CN118211882B (en) | Product quality management system and method based on big data | |
| US20240169726A1 (en) | Computer vision-based surgical workflow recognition system using natural language processing techniques | |
| CN116842330A (en) | Health care information processing method and device capable of comparing histories | |
| CN118475919A (en) | Model verification based on sub-model performance | |
| EP2889724B1 (en) | System and method for selecting features for identifying human activities in a human-computer interacting environment | |
| JP6797295B2 (en) | Diagnostic equipment, diagnostic methods and programs | |
| CN116861175B (en) | Operation track correction method based on neural network | |
| CN118629596B (en) | Psychological state analysis system and method for patients with multiple myeloma heart amyloidosis | |
| CN118236041B (en) | Cognitive function evaluation method, system, equipment and readable storage medium | |
| CN118213058B (en) | Medical instrument fault detection management system and method based on data analysis | |
| CN118262931B (en) | Medical data characteristic completion method and system in emergency rescue scene | |
| CN118504847B (en) | Intelligent beam field management method and system based on digital twin technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |