WO2021232591A1 - 基于物联网数据的用户数据分类方法、装置、计算机设备及存储介质 - Google Patents
基于物联网数据的用户数据分类方法、装置、计算机设备及存储介质 Download PDFInfo
- Publication number
- WO2021232591A1 WO2021232591A1 PCT/CN2020/105413 CN2020105413W WO2021232591A1 WO 2021232591 A1 WO2021232591 A1 WO 2021232591A1 CN 2020105413 W CN2020105413 W CN 2020105413W WO 2021232591 A1 WO2021232591 A1 WO 2021232591A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- current user
- medical
- historical
- user
- Prior art date
Links
Images
Classifications
-
- 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
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
-
- 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
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
- G16Y10/00—Economic sectors
- G16Y10/60—Healthcare; Welfare
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
- G16Y20/00—Information sensed or collected by the things
- G16Y20/40—Information sensed or collected by the things relating to personal data, e.g. biometric data, records or preferences
Definitions
- This application relates to the field of data processing and blockchain technology, and in particular to a method, device, computer equipment, and storage medium for user data classification based on Internet of Things data.
- doctor recommend follow-up treatment plans based on the user’s physical sign data measured on-site in the hospital, they are generally based on the doctor’s own experience, which requires a higher professional level of the doctor, that is, manual labor to obtain more accurate treatment plan data. higher cost.
- the embodiments of the application provide a user data classification method, device, computer equipment, and storage medium based on Internet of Things data, which are designed to solve the problem of obtaining physical sign data of patient users in the prior art and need to go to the hospital for manual measurement, and based on measurement
- the doctor’s professional level is required to be higher, which leads to a longer acquisition period of medical data for chronic diseases, and the problem of higher labor costs for obtaining more accurate treatment plan data.
- an embodiment of the present application provides a user data classification method based on Internet of Things data, which includes:
- the current user medical data includes user identification data, drug injection data, and physical sign measurement data
- the historical medical data set corresponding to the identification data is obtained; wherein, the historical medical data set includes a subset of electronic medical record data, a subset of examination report data, and medication Record data subset, historical treatment plan data subset;
- the current user data input vector is composed of the semantic vector corresponding to each field in the current user input data, and the current user data input vector is input to the pre-trained convolutional neural network model to obtain the corresponding current user data input vector Classification results;
- an embodiment of the present application provides a user data classification device based on Internet of Things data, which includes:
- the current data receiving unit is used to determine whether the current user medical data uploaded by the Internet of Things medical terminal is received; wherein the current user medical data includes user identification data, drug injection data, and physical sign measurement data;
- the historical data search unit is configured to obtain a historical medical data set corresponding to the identity recognition data if the current user medical data uploaded by the medical terminal of the Internet of Things is received; wherein, the historical medical data set includes a subset of electronic medical record data, Examination report data subset, medication record data subset, historical treatment plan data subset;
- the current user input data acquisition unit is used to call the preset first keyword set and the first extraction strategy corresponding to the first keyword set, and extract the target data in the historical medical data set corresponding to the identification data Collection to obtain current user input data by merging with the current user medical data;
- the classification result acquisition unit is used to form a current user data input vector from the semantic vector corresponding to each field in the current user input data, and input the current user data input vector into the pre-trained convolutional neural network model to obtain the current user data input vector. State the classification result corresponding to the current user data input vector;
- the data type judgment unit is used to judge whether the treatment path data type corresponding to the classification result is greater than 1;
- the standard vector obtaining unit is configured to obtain the standard user data input vector corresponding to each treatment path data type corresponding to the classification result if the treatment path data type corresponding to the classification result is greater than one;
- the optimal path data acquisition unit is used to calculate the vector similarity between the standard user data input vector corresponding to the classification result and the current user data input vector, and obtain the standard user data input vector corresponding to the maximum value of the vector similarity, and the corresponding The data of the treatment path is used as the best treatment path data.
- an embodiment of the present application provides a computer device, which includes a memory, a processor, and a computer program stored on the memory and running on the processor, and the processor executes the computer The following steps are implemented during the program:
- the current user medical data includes user identification data, drug injection data, and physical sign measurement data
- the historical medical data set corresponding to the identification data is obtained; wherein, the historical medical data set includes a subset of electronic medical record data, a subset of examination report data, and medication Record data subset, historical treatment plan data subset;
- the current user data input vector is composed of the semantic vector corresponding to each field in the current user input data, and the current user data input vector is input to the pre-trained convolutional neural network model to obtain the corresponding current user data input vector Classification results;
- the embodiments of the present application also provide a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, which when executed by a processor causes the processor to perform the following operations :
- the current user medical data includes user identification data, drug injection data, and physical sign measurement data
- the historical medical data set corresponding to the identification data is obtained; wherein, the historical medical data set includes a subset of electronic medical record data, a subset of examination report data, and medication Record data subset, historical treatment plan data subset;
- the current user data input vector is composed of the semantic vector corresponding to each field in the current user input data, and the current user data input vector is input to the pre-trained convolutional neural network model to obtain the corresponding current user data input vector Classification results;
- the embodiments of the application provide a user data classification method, device, computer equipment, and storage medium based on Internet of Things data.
- the method realizes rapid acquisition of user medical data based on Internet of Things equipment, and is fast and accurate based on historical medical data. Recommend the best treatment plan data to reduce the cost of obtaining the best treatment plan data.
- FIG. 1 is a schematic diagram of an application scenario of a user data classification method based on Internet of Things data provided by an embodiment of the application;
- FIG. 2 is a schematic flowchart of a user data classification method based on Internet of Things data provided by an embodiment of the application;
- FIG. 3 is a schematic block diagram of a user data classification device based on Internet of Things data provided by an embodiment of the application;
- Fig. 4 is a schematic block diagram of a computer device provided by an embodiment of the application.
- Figure 1 is a schematic diagram of an application scenario of a method for user data classification based on Internet of Things data provided by an embodiment of this application
- Figure 2 is a schematic diagram of a method for user data classification based on Internet of Things data provided by an embodiment of this application Schematic diagram of the process.
- the user data classification method based on Internet of Things data is applied to a cloud server, and the method is executed by application software installed in the cloud server.
- the method includes steps S110 to S170.
- S110 Determine whether the current user medical data uploaded by the IoT medical terminal is received; wherein, the current user medical data includes user identification data, medication injection data, and physical sign measurement data.
- the Internet of Things medical terminal can be a blood glucose meter, insulin syringe, thermometer and other Internet of Things devices when implemented.
- the difference between these Internet of Things devices and traditional blood glucose meters, insulin syringes, and thermometers is that these things
- the networked devices are equipped with wireless communication modules such as low-power Bluetooth modules and 4G/5G communication modules. After this transformation, some of the parameter values detected by the IoT medical terminal are: 1) Can be directly uploaded to the cloud server; 2 ) The parameter values can be uploaded to the smart terminal after being interconnected with a smart terminal such as a smart phone, and then uploaded to the server by the smart terminal.
- a patient with a chronic disease needs to use these IoT devices to measure blood sugar, body temperature and other parameters for a long time, or to inject insulin regularly and quantitatively.
- the corresponding detection is detected through the IoT medical terminal.
- the parameter value of it can be finally uploaded to the cloud server as the parameter value of chronic disease monitoring.
- the second is a smart terminal, such as a smart phone, which is used to receive the current user medical data uploaded by the IoT medical terminal bound to it, and send the current user medical data to a cloud server for storage.
- a smart terminal such as a smart phone
- the third is a cloud server, which is used to receive the current user medical data uploaded by the Internet of Things medical terminal or smart terminal, and convert the current user medical data into structured data and store the corresponding historical data of the user. Moreover, similar case data and recommended treatment path data can be matched in the database based on the current user's medical data. Furthermore, the cloud server can also generate patient portraits, patient medication data curves, medication risk predictions, etc. based on the data corresponding to each user.
- the Internet of Things medical terminal When the user uses the Internet of Things medical terminal to measure blood sugar, body temperature and other parameters, or when injecting insulin regularly and quantitatively, the current user's medical data can be accurately obtained. Since in order to identify which user uses the user medical data uploaded by the Internet of Things medical terminal device, the Internet of Things medical terminal device needs to bind the user identification data during initialization.
- the user interaction interface of the smart terminal will get the information
- the historical user medical data stored in the IoT medical terminal (generally, when the connection is initialized, the historical user medical data stored in the IoT medical terminal is an empty set, that is, the data has not been stored), and it will also prompt whether it is related to the object.
- the networked medical terminal is bound. Once the user chooses to bind the IoT medical terminal with the smart terminal, it is equivalent to establishing a mapping relationship between the device unique identification code of the IoT medical terminal and the device unique identification code of the smart terminal.
- the unique identification code of the smart terminal is in correspondence with the user identification data (for example, the user uses the WeChat ID code to bind the unique identification code of the smart terminal), so the IoT medical terminal will be current
- the corresponding current user medical data includes not only drug injection data (for example, insulin injection volume), physical sign measurement data (for example, blood glucose value, body temperature value) and other data, but also uploads user identification data.
- a historical medical data set corresponding to the identity recognition data is obtained; wherein the historical medical data set includes an electronic medical record data subset and an examination report data subset , A subset of medication record data, a subset of historical treatment plan data.
- the cloud server After the cloud server receives the current user medical data uploaded by the IoT medical terminal, it needs to know which user corresponds to the data, so as to determine to store the current user medical data in the data table corresponding to the user .
- the cloud server There are at least four data sources in the historical medical data set corresponding to the cloud server and the identification data, which are electronic medical records, inspection reports, historical medication data, and historical treatment plans stored in the hospital's medical system. Since the above-mentioned data stored in the medical system is unstructured data, the cloud server needs to perform data cleaning and structured storage after acquiring the electronic medical records, inspection reports, historical medication data, and historical treatment data from the medical system. If the cloud server does not receive the current user medical data uploaded by the IoT medical terminal, after waiting for a preset delay waiting time (such as 10s), it continues to return to step S110.
- a preset delay waiting time such as 10s
- the method before step S120, the method further includes:
- the fourth extraction strategy is to extract the initial subset of the medication record data corresponding to the historical medication data;
- the initial subset of historical treatment plan data corresponding to the historical treatment plan is extracted through the preset fifth keyword set and the fifth extraction strategy corresponding to the fifth keyword set.
- the medical system server will store a large number of electronic medical records, examination reports, historical medication data, and historical treatment plan data corresponding to a large number of patients.
- the patient's electronic medical record is the patient-related information that the doctor directly uses the computer to record when the patient is examined, at least including the home page, the disease history record, the examination and inspection result, the doctor's order, the operation record, the nursing record, etc.
- the patient’s examination report is generally a blood test report, CT examination report and other examination reports.
- the patient’s CT examination report generally includes the patient’s name, gender, age, department, hospitalization number, bed number, examination site, registration date, and examination Name, examination method, image performance, report to doctor and other information.
- the patient's historical medication data is generally written in a doctor's prescription note, and the doctor can take a photo of the doctor's prescription note and upload it to the medical system server.
- the doctor's prescription note generally include the patient's name, name, age, department, clinical diagnosis, address/telephone, prescription list, prescribing physician, dispenser, proofreader, drug price, date of prescription and other information.
- the historical treatment plan of the patient is generally registered by the doctor directly in the medical system server, that is, a treatment plan will be recorded correspondingly for each medical experience of the patient.
- the second keyword set and the second extraction strategy corresponding to the second keyword set can be preset at this time.
- the keywords included in the second set of keywords include the patient’s name, ID number, gender, age, the patient’s region, family medical history, disease history, examination results, medical advice, surgical records, nursing records, and the second
- the extraction strategy is to extract the specific value of each keyword corresponding to the second keyword set.
- the keywords included in the third keyword set set include name, ID number, gender, age, department, hospitalization number, bed number, examination site, registration date, examination name, examination method, image performance (also Understand as an examination result), report to the doctor, the third extraction strategy is to extract the specific value of each keyword corresponding to the third keyword set.
- the OCR text recognition model is used to perform text recognition on the pictures corresponding to the historical medication data to obtain the recognized text corresponding to the historical medication data.
- the recognition text is computer text that can be understood by the computer) as the recognition text corresponding to the historical medication data.
- the keywords included in the fourth keyword set include name, ID number, age, department, clinical diagnosis, address/telephone, prescription list, prescribing physician, dispenser, proofreader, drug price,
- the date of prescription the fourth extraction strategy is to extract the specific value of each keyword corresponding to the fourth keyword set.
- the keywords included in the fifth keyword set include name, ID number, gender, age, department, hospitalization number, treatment plan process, and attending doctor.
- the fifth extraction strategy is to correspond to the fifth keyword set. Extract the specific value of each keyword.
- the electronic medical record data subset, the inspection report data subset, the medication record data subset, and the historical treatment plan data subset all include the common fields of patient name, ID number, and gender
- the electronic medical record data subset, inspection A piece of data with the same name, ID number, and gender in the report data subset, medication record data subset, and historical treatment plan data subset are merged into the historical medical data set of the same patient.
- the data cleaning of the historical raw medical data corresponding to the user is realized, so as to obtain the initial set of historical medical data corresponding to each user.
- the historical medical data corresponding to each user can also be set for each initial set of historical medical data.
- the specific value of each field in a data subset is transformed into the data structure; for example, the specific value of the field of the treatment plan process in the initial subset of the historical treatment plan data corresponding to the historical treatment plan is composed of a large section of text. Converting it into structured data can extract keywords to obtain the keyword combination of the field of the treatment plan process, which is used as the structured data corresponding to the field of the treatment plan process.
- the initial subset of the electronic medical record data, the initial subset of inspection report data, the initial subset of medication record data, and the initial subset of historical treatment plan data are extracted by keywords to obtain the initial subset of the electronic medical record data.
- the electronic medical record data subset corresponding to the subset, the inspection report data subset corresponding to the initial subset of the inspection report data, the medication record data subset corresponding to the initial subset of the medication record data, and the historical treatment plan The historical treatment plan data subset corresponding to the initial subset of data.
- a large text description corresponds to the specific value of the field of the treatment program process in the initial subset of historical treatment program data.
- the keywords that do not exceed the preset ranking value in the word segmentation result are extracted as the structured data corresponding to the specific value of this field in the treatment plan process; After the specific values are extracted by keywords, the data structure conversion is realized.
- the current user medical data uploaded by the current user through the Internet of Things medical terminal generally only includes user identification data, medication injection data, and physical sign measurement data. These values are very structured field values. There is no need to perform structured data conversion on the current user medical data, and directly merge with the current user medical data to obtain the current user input data.
- the first keyword set and the first extraction strategy corresponding to the first keyword set are stored in a blockchain network.
- the second set of keywords and the second extraction strategy corresponding to the second set of keywords are stored in a blockchain network;
- the third set of keywords corresponds to the third set of keywords
- the third extraction strategy of is stored in the blockchain network;
- the fourth keyword set and the fourth extraction strategy corresponding to the fourth keyword set are stored in the blockchain network;
- the fifth keyword set And the fifth extraction strategy corresponding to the fifth keyword set is stored in the blockchain network.
- the above keyword set and extraction strategy in the cloud server can all be stored in the blockchain network.
- the cloud server can be used as one of the blockchain node devices in the blockchain network.
- Blockchain is a new application mode of computer technology such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm.
- Blockchain essentially a decentralized database, is a series of data blocks associated with cryptographic methods. Each data block contains a batch of network transaction information for verification. The validity of the information (anti-counterfeiting) and the generation of the next block.
- the blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.
- step S130 includes:
- Data combination is performed on the repeated fields in the target data set and the medical data of the current user, and duplicates are removed to obtain the input data of the current user.
- the user identification data is the current user medical data and the target data in the corresponding historical medical data set. Repeated fields of the collection, at this time, de-duplicate this repeated field and keep one of them.
- the current user medical data includes the three fields of user identification data, drug injection data, and physical sign measurement data and their specific field values
- the target data set in the historical medical data set corresponding to the identification data includes the patient's name , User identification data (specifically, ID number), the patient’s region, family medical history, patient’s symptoms, examination results, prescription list, these 7 fields and their specific field values, because the user identification data field is Repeatedly, the current user medical data is merged with the current user medical data, and the current user input data obtained includes user identification data, drug injection data, physical sign measurement data, patient name, patient area, family medical history, and patient The 10 fields of illness, examination result, prescription list and their specific field values.
- a current user data input vector is formed by the semantic vector corresponding to each field in the current user input data, and the current user data input vector is input to a pre-trained convolutional neural network model to obtain the current user data input.
- the classification result corresponding to the vector is obtained by the semantic vector corresponding to each field in the current user input data, and the current user data input vector is input to a pre-trained convolutional neural network model to obtain the current user data input.
- the classification result corresponding to the vector is obtained by the semantic vector corresponding to each field in the current user input data, and the current user data input vector is input to a pre-trained convolutional neural network model to obtain the current user data input.
- the acquired current user input data includes user identification data, drug injection data, physical sign measurement data, patient name, patient's region, family medical history, patient symptoms, examination results, prescription list, etc. Fields and their specific field values.
- the semantic vector corresponding to these 10 fields can be formed according to the word vectors corresponding to the specific values of these 10 fields (that is, based on 10 word vectors, and 10 word vectors). The weights corresponding to the vectors are obtained, and the semantic vectors corresponding to these 10 fields are obtained).
- the convolutional neural network model that has been stored and completed in the cloud server is called to input the current user data
- the vector is input to the pre-trained convolutional neural network model to obtain the classification result corresponding to the current user data input vector.
- each piece of training data in the training set includes two parts: one is the training set input data, which includes user identification Data, drug injection data, physical sign measurement data, patient name, patient area, family medical history, patient symptoms, examination results, prescription list of the specific field values of these 10 fields (generally, the specific field values of these fields have been Converted into a keyword combination, so the specific field values of these 10 fields can be conveniently converted into corresponding word vectors to form a semantic vector, so that the semantic vectors corresponding to the specific values of the 10 fields form the corresponding training set input Vector data);
- the second is the training set output data, which includes the treatment path classification results.
- each type of disease corresponds to at least one treatment path. Therefore, the classification result is generally judged based on the current user data input vector can be obtained The classification results of the disease to determine the treatment path to which the disease belongs.
- step S140 includes:
- the current user data input vector corresponding to the current user input data is used as the input of the convolutional neural network model to obtain a corresponding classification result.
- the semantic vector corresponding to each field in the current user input data is a 1*300 one-dimensional row vector.
- the semantic vector corresponding to each field in the current user input data is subjected to vector addition, thereby The current user data input vector corresponding to the current user input data is obtained.
- the current user data input vector corresponding to the current user input data is used as the input of the convolutional neural network model to perform a classification operation to obtain a corresponding classification result.
- the treatment path data generally includes the specific value of the treatment plan process field. Therefore, when the classification result corresponding to the current user input data is determined, the treatment plan process corresponding to the current user data can be obtained. If the type of treatment path data corresponding to the classification result is greater than 1, at this time, the user corresponding to the current user input data can also recommend the best treatment path data.
- the general training set when the convolutional neural network to be trained is trained through the training set, the general training set includes multiple types of training subsets, and one type of treatment path data corresponding to each training subset is used as the training set Output Data. In this way, one piece of training data is selected in each training subset as the input data of the standard training set corresponding to the type of treatment path data.
- each training subset corresponds to 1000 kinds of treatment path data training set input data
- one training set input data is selected for each training subset corresponding to each training subset Standard training set input data. Since each training subset corresponds to the standard training set input data, it also corresponds to a standard user data input vector.
- the standard user data input vectors corresponding to the three types of treatment path data are obtained at this time.
- the treatment path data corresponding to the type of treatment path data is acquired as the optimal treatment path data.
- the treatment path data corresponding to the treatment path data is selected as the optimal treatment path data.
- the optimal treatment path data can be pushed to the smart terminal corresponding to the current user input data.
- the cloud server can also generate patient portraits, patient medication data curves, medication risk predictions, etc. for the data corresponding to each user, and the generation of these data can be based on the user's historical medical data set.
- step S170 the method further includes:
- the keywords in the target portrait data set corresponding to each user are filtered to obtain a simplified set of target portrait data corresponding to each user;
- the keywords included in the set sixth keyword set include the patient's region, family medical history, patient symptoms, and examination results.
- the sixth extraction strategy is to assign each key corresponding to the sixth keyword set.
- the specific value of the word is extracted.
- the specific value of each keyword corresponding to the sixth keyword set is extracted to form a target portrait data set corresponding to each user in the historical medical data set.
- the important keywords can be filtered and the keywords can be converted into corresponding tags.
- a tag conversion strategy corresponding to each keyword is set in the cloud server (the keyword can also be converted into a strategy corresponding to the tag).
- the user portrait data corresponding to each user can be obtained after the tag conversion of the keyword.
- the method realizes the rapid acquisition of user medical data based on the Internet of Things devices, and quickly and accurately recommends the best treatment plan data based on historical medical data, reducing the cost of obtaining the best treatment plan data.
- This method belongs to the field of smart medical care, and the construction of smart cities can be promoted through this solution.
- the embodiments of the present application also provide a user data classification device based on Internet of Things data.
- the user data classification device based on Internet of Things data is used to execute any embodiment of the aforementioned method for user data classification based on Internet of Things data.
- FIG. 3 is a schematic block diagram of a user data classification device based on Internet of Things data provided by an embodiment of the present application.
- the user data classification device 100 based on Internet of Things data can be configured in a cloud server.
- the user data classification device 100 based on Internet of Things data includes: a current data receiving unit 110, a historical data searching unit 120, a current user input data obtaining unit 130, a classification result obtaining unit 140, a data type judging unit 150, The standard vector acquiring unit 160 and the optimal path data acquiring unit 170.
- the current data receiving unit 110 is used to determine whether the current user medical data uploaded by the IoT medical terminal is received; wherein the current user medical data includes user identification data, medication injection data, and physical sign measurement data.
- the IoT medical terminal device when the user uses the Internet of Things medical terminal to measure blood glucose, body temperature and other parameters, or injects insulin regularly and quantitatively, the current user's medical data can be accurately obtained.
- the IoT medical terminal device In order to identify which user uses the medical data uploaded by the IoT medical terminal device, the IoT medical terminal device needs to bind the user identification data during initialization.
- the user interaction interface of the smart terminal will get the information
- the historical user medical data stored in the IoT medical terminal (generally, when the connection is initialized, the historical user medical data stored in the IoT medical terminal is an empty set, that is, the data has not been stored), and it will also prompt whether it is related to the object.
- the networked medical terminal is bound. Once the user chooses to bind the IoT medical terminal with the smart terminal, it is equivalent to establishing a mapping relationship between the device unique identification code of the IoT medical terminal and the device unique identification code of the smart terminal.
- the unique identification code of the smart terminal is in correspondence with the user identification data (for example, the user uses the WeChat ID code to bind the unique identification code of the smart terminal), so the IoT medical terminal will be current
- the corresponding current user medical data includes not only drug injection data (for example, insulin injection volume), physical sign measurement data (for example, blood glucose value, body temperature value) and other data, but also uploads user identification data.
- the historical data search unit 120 is configured to, if the current user medical data uploaded by the IoT medical terminal is received, obtain a historical medical data set corresponding to the identification data; wherein, the historical medical data set includes a subset of electronic medical record data , Inspection report data subset, medication record data subset, historical treatment plan data subset.
- the cloud server After the cloud server receives the current user medical data uploaded by the IoT medical terminal, it needs to know which user corresponds to the data, so as to determine to store the current user medical data in the data table corresponding to the user .
- the cloud server needs to perform data cleaning and structured storage after acquiring the electronic medical records, inspection reports, historical medication data, and historical treatment data from the medical system. If the cloud server does not receive the current user medical data uploaded by the IoT medical terminal, after waiting for the preset delay waiting time (such as 10s), continue to return to the step of judging whether the current user medical data uploaded by the IoT medical terminal has been received .
- the preset delay waiting time such as 10s
- the user data classification device 100 based on Internet of Things data further includes:
- Historical data uploading and receiving unit for receiving electronic medical records, inspection reports, historical medication data and historical treatment plans uploaded by the medical system server;
- the electronic medical record data initial subset acquisition unit is configured to extract the corresponding electronic medical record data initial subset in the electronic medical record by calling a preset second keyword set and a second extraction strategy corresponding to the second keyword set set;
- the initial subset of inspection report data acquisition unit is configured to extract the corresponding initial subset of inspection report data in the inspection report by calling a preset third keyword set and a third extraction strategy corresponding to the third keyword set. set;
- the initial subset of medication record data acquisition unit is used to perform text recognition on the pictures corresponding to the historical medication data through the OCR text recognition model to obtain the recognized text corresponding to the historical medication data through a preset fourth keyword set And a fourth extraction strategy corresponding to the fourth keyword set, extracting an initial subset of medication record data corresponding to the historical medication data;
- the historical treatment plan data initial subset acquisition unit is configured to extract the initial historical treatment plan data corresponding to the historical treatment plan through a preset fifth keyword set and a fifth extraction strategy corresponding to the fifth keyword set Subset.
- the medical system server will store a large number of electronic medical records, examination reports, historical medication data, and historical treatment plan data corresponding to a large number of patients.
- the patient's electronic medical record is the patient-related information that the doctor directly uses the computer to record when the patient is examined, at least including the home page, the disease history record, the examination and inspection result, the doctor's order, the operation record, the nursing record, etc.
- the patient’s examination report is generally a blood test report, CT examination report and other examination reports.
- the patient’s CT examination report generally includes the patient’s name, gender, age, department, hospitalization number, bed number, examination site, registration date, and examination Name, examination method, image performance, report to doctor and other information.
- the patient's historical medication data is generally written in a doctor's prescription note, and the doctor can take a photo of the doctor's prescription note and upload it to the medical system server.
- the doctor's prescription note it generally includes the patient's name, name, age, department, clinical diagnosis, address/telephone, prescription list, prescribing physician, dispenser, proofreader, drug price, date of prescription and other information.
- the historical treatment plan of the patient is generally registered by the doctor directly in the medical system server, that is, a treatment plan will be recorded correspondingly for each medical experience of the patient.
- the second keyword set and the second extraction strategy corresponding to the second keyword set can be preset at this time.
- the keywords included in the second set of keywords include the patient’s name, ID number, gender, age, the patient’s region, family medical history, disease history, examination results, medical advice, surgical records, nursing records, and the second
- the extraction strategy is to extract the specific value of each keyword corresponding to the second keyword set.
- the keywords included in the third keyword set set include name, ID number, gender, age, department, hospitalization number, bed number, examination site, registration date, examination name, examination method, image performance (also Understand as an examination result), report to the doctor, the third extraction strategy is to extract the specific value of each keyword corresponding to the third keyword set.
- the OCR text recognition model is used to perform text recognition on the pictures corresponding to the historical medication data to obtain the recognized text corresponding to the historical medication data.
- the recognition text is computer text that can be understood by the computer) as the recognition text corresponding to the historical medication data.
- the keywords included in the fourth keyword set include name, ID number, age, department, clinical diagnosis, address/telephone, prescription list, prescribing physician, dispenser, proofreader, drug price,
- the date of prescription the fourth extraction strategy is to extract the specific value of each keyword corresponding to the fourth keyword set.
- the keywords included in the fifth keyword set include name, ID number, gender, age, department, hospitalization number, treatment plan process, and attending doctor.
- the fifth extraction strategy is to correspond to the fifth keyword set. Extract the specific value of each keyword.
- the electronic medical record data subset, the inspection report data subset, the medication record data subset, and the historical treatment plan data subset all include the common fields of patient name, ID number, and gender
- the electronic medical record data subset, inspection A piece of data with the same name, ID number, and gender in the report data subset, medication record data subset, and historical treatment plan data subset are merged into the historical medical data set of the same patient.
- the data cleaning of the historical raw medical data corresponding to the user is realized, so as to obtain the initial set of historical medical data corresponding to each user.
- the historical medical data corresponding to each user can also be set for each initial set of historical medical data.
- the specific value of each field in a data subset is transformed into the data structure; for example, the specific value of the field of the treatment plan process in the initial subset of the historical treatment plan data corresponding to the historical treatment plan is composed of a large section of text. Converting it into structured data can extract keywords to obtain the keyword combination of the field of the treatment plan process, which is used as the structured data corresponding to the field of the treatment plan process.
- the user data classification device 100 based on Internet of Things data further includes:
- the data subset acquisition unit is used to extract keywords from the initial subset of the electronic medical record data, the initial subset of examination report data, the initial subset of medication record data, and the initial subset of historical treatment plan data to extract keywords.
- Obtain an electronic medical record data subset corresponding to the initial subset of electronic medical record data, an inspection report data subset corresponding to the initial subset of inspection report data, and a medication record data subset corresponding to the initial subset of medication record data A set, a historical treatment plan data subset corresponding to the initial subset of the historical treatment plan data.
- a large text description corresponds to the specific value of the field of the treatment program process in the initial subset of historical treatment program data.
- the keywords that do not exceed the preset ranking value in the word segmentation result are extracted as the structured data corresponding to the specific value of this field in the treatment plan process; After the specific values are extracted by keywords, the data structure conversion is realized.
- the current user input data acquisition unit 130 is configured to call a preset first keyword set and a first extraction strategy corresponding to the first keyword set, and extract the target in the historical medical data set corresponding to the identification data
- the data set is combined with the medical data of the current user to obtain the input data of the current user.
- the current user medical data uploaded by the current user through the Internet of Things medical terminal generally only includes user identification data, medication injection data, and physical sign measurement data. These values are very structured field values. There is no need to perform structured data conversion on the current user medical data, and directly merge with the current user medical data to obtain the current user input data.
- the first keyword set and the first extraction strategy corresponding to the first keyword set are stored in a blockchain network.
- the second set of keywords and the second extraction strategy corresponding to the second set of keywords are stored in a blockchain network;
- the third set of keywords corresponds to the third set of keywords
- the third extraction strategy of is stored in the blockchain network;
- the fourth keyword set and the fourth extraction strategy corresponding to the fourth keyword set are stored in the blockchain network;
- the fifth keyword set And the fifth extraction strategy corresponding to the fifth keyword set is stored in the blockchain network.
- the above keyword set and extraction strategy in the cloud server can all be stored in the blockchain network.
- the cloud server can be used as one of the blockchain node devices in the blockchain network.
- Blockchain is a new application mode of computer technology such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm.
- Blockchain essentially a decentralized database, is a series of data blocks associated with cryptographic methods. Each data block contains a batch of network transaction information for verification. The validity of the information (anti-counterfeiting) and the generation of the next block.
- the blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.
- the current user input data acquisition unit 130 includes:
- the target data set obtaining unit is used to call the first keyword set and the first extraction strategy corresponding to the first keyword set, and obtain the identification data corresponding to the identification data and the first extraction strategy in the historical medical data set.
- the data deduplication unit is used for data combination and deduplication of repeated fields in the target data set and the current user's medical data to obtain the current user input data.
- the user identification data is the current user medical data and the target data in the corresponding historical medical data set. Repeated fields of the collection, at this time, de-duplicate this repeated field and keep one of them.
- the current user medical data includes the three fields of user identification data, drug injection data, and physical sign measurement data and their specific field values
- the target data set in the historical medical data set corresponding to the identification data includes the patient's name , User identification data (specifically, ID number), the patient’s region, family medical history, patient’s symptoms, examination results, prescription list, these 7 fields and their specific field values, because the user identification data field is Repeatedly, the current user medical data is merged with the current user medical data, and the current user input data obtained includes user identification data, drug injection data, physical sign measurement data, patient name, patient area, family medical history, and patient The 10 fields of illness, examination result, prescription list and their specific field values.
- the classification result obtaining unit 140 is configured to form a current user data input vector from the semantic vector corresponding to each field in the current user input data, and input the current user data input vector into a pre-trained convolutional neural network model to obtain The classification result corresponding to the current user data input vector.
- the acquired current user input data includes user identification data, drug injection data, physical sign measurement data, patient name, patient's region, family medical history, patient symptoms, examination results, prescription list, etc. Fields and their specific field values.
- the semantic vector corresponding to these 10 fields can be formed according to the word vectors corresponding to the specific values of these 10 fields (that is, based on 10 word vectors, and 10 word vectors). The weights corresponding to the vectors are obtained, and the semantic vectors corresponding to these 10 fields are obtained).
- the convolutional neural network model that has been stored and completed in the cloud server is called to input the current user data
- the vector is input to the pre-trained convolutional neural network model to obtain the classification result corresponding to the current user data input vector.
- each piece of training data in the training set includes two parts: one is the input data of the training set, which includes user identification Data, drug injection data, physical sign measurement data, patient name, patient area, family medical history, patient symptoms, examination results, prescription list of the specific field values of these 10 fields (generally, the specific field values of these fields have been Converted into a keyword combination, so the specific field values of these 10 fields can be conveniently converted into corresponding word vectors to form a semantic vector, so that the semantic vectors corresponding to the specific values of the 10 fields form the corresponding training set input Vector data);
- the second is the training set output data, which includes the treatment path classification results.
- each type of disease corresponds to at least one treatment path. Therefore, the classification result is generally judged based on the current user data input vector can be obtained The classification results of the disease to determine the treatment path to which the disease belongs.
- the classification result obtaining unit 140 includes:
- the current user data input vector obtaining unit is used to obtain the semantic vector corresponding to each field in the current user input data, and sum each semantic vector to obtain the current user data input vector corresponding to the current user input data ;
- the model calling unit is used to call the pre-stored convolutional neural network model
- the vector calculation and classification unit is configured to use the current user data input vector corresponding to the current user input data as the input of the convolutional neural network model to obtain the corresponding classification result.
- the semantic vector corresponding to each field in the current user input data is a 1*300 one-dimensional row vector.
- the semantic vector corresponding to each field in the current user input data is subjected to vector addition, thereby The current user data input vector corresponding to the current user input data is obtained.
- the current user data input vector corresponding to the current user input data is used as the input of the convolutional neural network model to perform a classification operation to obtain a corresponding classification result.
- the data type judging unit 150 is used to judge whether the treatment path data type corresponding to the classification result is greater than one.
- the treatment path data generally includes the specific value of the treatment plan process field. Therefore, when the classification result corresponding to the current user input data is determined, the treatment plan process corresponding to the current user data can be obtained. If the type of treatment path data corresponding to the classification result is greater than 1, at this time, the user corresponding to the current user input data can also recommend the best treatment path data.
- the standard vector obtaining unit 160 is configured to obtain standard user data input vectors corresponding to each treatment path data type corresponding to the classification result if the treatment path data type corresponding to the classification result is greater than one.
- the general training set when the convolutional neural network to be trained is trained through the training set, the general training set includes multiple types of training subsets, and one type of treatment path data corresponding to each training subset is used as the training set Output Data. In this way, one piece of training data is selected in each training subset as the input data of the standard training set corresponding to the type of treatment path data.
- each training subset corresponds to 1000 kinds of treatment path data training set input data
- one training set input data is selected for each training subset corresponding to each training subset Standard training set input data. Since each training subset corresponds to the standard training set input data, it also corresponds to a standard user data input vector.
- the standard user data input vectors corresponding to the three types of treatment path data are obtained at this time.
- the treatment path data corresponding to the type of treatment path data is acquired as the optimal treatment path data.
- the optimal path data obtaining unit 170 is configured to calculate the vector similarity between the standard user data input vector corresponding to the classification result and the current user data input vector, and obtain the standard user data input vector corresponding to the maximum value of the vector similarity, and The corresponding treatment path data is used as the optimal treatment path data.
- the treatment path data corresponding to the treatment path data is selected as the optimal treatment path data.
- the optimal treatment path data can be pushed to the smart terminal corresponding to the current user input data.
- the cloud server can also generate patient portraits, patient medication data curves, medication risk predictions, etc. for the data corresponding to each user, and the generation of these data can be based on the user's historical medical data set.
- the user data classification device 100 based on Internet of Things data further includes:
- the target portrait data set acquiring unit is used to extract the target portrait data set corresponding to each user in the historical medical data set by calling the preset sixth keyword set and the sixth extraction strategy corresponding to the sixth keyword set ;
- the data set simplification unit is used to filter the keywords in the target portrait data set corresponding to each user according to the target portrait data set corresponding to each user and the pre-stored keyword screening strategy to obtain a simplified target portrait data set corresponding to each user;
- the user portrait data generating unit is used to call a pre-stored keyword conversion strategy to correspondingly convert the simplified set of target portrait data corresponding to each user into user portrait data.
- the keywords included in the set sixth keyword set include the patient's region, family medical history, patient symptoms, and examination results.
- the sixth extraction strategy is to assign each key corresponding to the sixth keyword set.
- the specific value of the word is extracted.
- the specific value of each keyword corresponding to the sixth keyword set is extracted to form a target portrait data set corresponding to each user in the historical medical data set.
- the important keywords can be filtered and the keywords can be converted into corresponding tags.
- a tag conversion strategy corresponding to each keyword is set in the cloud server (the keyword can also be converted into a strategy corresponding to the tag).
- the user portrait data corresponding to each user can be obtained after the tag conversion of the keyword.
- the device realizes rapid acquisition of user medical data based on IoT devices, and quickly and accurately recommends the best treatment plan data based on historical medical data, reducing the cost of obtaining the best treatment plan data.
- the above-mentioned user data classification device based on Internet of Things data can be implemented in the form of a computer program, and the computer program can be run on a computer device as shown in FIG. 4.
- FIG. 4 is a schematic block diagram of a computer device according to an embodiment of the present application.
- the computer device 500 is a server, and the server may be an independent server or a server cluster composed of multiple servers.
- the computer device 500 includes a processor 502, a memory, and a network interface 505 connected through a system bus 501, where the memory may include a non-volatile storage medium 503 and an internal memory 504.
- the non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032.
- the processor 502 can execute a user data classification method based on Internet of Things data.
- the processor 502 is used to provide calculation and control capabilities, and support the operation of the entire computer device 500.
- the internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503.
- the processor 502 can make the processor 502 execute a user data classification method based on Internet of Things data.
- the network interface 505 is used for network communication, such as providing data information transmission.
- the structure shown in FIG. 4 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device 500 to which the solution of the present application is applied.
- the specific computer device 500 may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.
- the processor 502 is configured to run a computer program 5032 stored in a memory to implement the user data classification method based on Internet of Things data disclosed in the embodiment of the present application.
- the embodiment of the computer device shown in FIG. 4 does not constitute a limitation on the specific configuration of the computer device.
- the computer device may include more or less components than those shown in the figure. Or some parts are combined, or different parts are arranged.
- the computer device may only include a memory and a processor. In such an embodiment, the structures and functions of the memory and the processor are consistent with the embodiment shown in FIG. 4, and will not be repeated here.
- the processor 502 may be a central processing unit (Central Processing Unit, CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (Digital Signal Processors, DSPs), Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc.
- the general-purpose processor may be a microprocessor or the processor may also be any conventional processor.
- a computer-readable storage medium may be a non-volatile computer-readable storage medium, or may be a volatile computer-readable storage medium.
- the computer-readable storage medium stores a computer program, where the computer program is executed by a processor to implement the user data classification method based on Internet of Things data disclosed in the embodiments of the present application.
- the disclosed equipment, device, and method may be implemented in other ways.
- the device embodiments described above are only illustrative.
- the division of the units is only a logical function division. In actual implementation, there may be other division methods, or the units with the same function may be combined into one. Units, for example, multiple units or components can be combined or integrated into another system, or some features can be omitted or not implemented.
- the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may also be electrical, mechanical or other forms of connection.
- the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments of the present application.
- the functional units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
- the above-mentioned integrated unit can be implemented in the form of hardware or software functional unit.
- the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium.
- the technical solution of this application is essentially or the part that contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium. It includes several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present application.
- the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), magnetic disk or optical disk and other media that can store program codes.
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Physics & Mathematics (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Bioethics (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
一种基于物联网数据的用户数据分类方法、装置、计算机设备及存储介质,应用于智慧城市的智慧医疗领域,涉及数据处理及区块链,通过接收物联网医疗终端上传的当前用户医疗数据后,云服务器根据当前用户医疗数据、及在云服务器中检索得到的与当前用户对应的其他关键数据组成待分析数据集,由云服务器中已存储的卷积神经网络作为上述当前用户医疗数据和关键数据组成的输入数据的运算模型,从而得到分类结果确定最佳治疗路径。该方法实现了基于物联网设备对用户医疗数据的快速获取,且基于历史医疗数据快速且准确的推荐最佳治疗方案数据,降低获取最佳治疗方案数据的成本。
Description
本申请要求于2020年05月22日提交中国专利局、申请号为202010442300.0,发明名称为“基于物联网数据的用户数据分类方法、装置及计算机设备”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
本申请涉及数据处理及区块链技术领域,尤其涉及一种基于物联网数据的用户数据分类方法、装置、计算机设备及存储介质。
目前,当患者去医院看病(尤其是高血压、糖尿病、心脑血管疾病等慢性疾病)时,一般是用户去医院现场去测量体征数据,由患者人工读取和手动记录后告知医生,之后由医生对这些体征数据记录至医疗***服务器中,以作为该患者的历史医疗数据。但是患者频繁的去医院现场进行人工测量,不仅耗费患者时间,而且占用医生等医疗资源,导致慢性疾病的医疗数据的获取周期较长。
而且发明人发现医生根据用户在医院现场测量的体征数据进行后续治疗方案的推荐时,一般是基于医生的自身经验,对医生的专业水平要求较高,也即获取较为准确的治疗方案数据的人工成本较高。
发明内容
本申请实施例提供了一种基于物联网数据的用户数据分类方法、装置、计算机设备及存储介质,旨在解决现有技术中获取患者用户的体征数据需去医院现场进行人工测量,且基于测量的体征数据进行后续治疗方案的推荐时对医生专业水平要求较高,导致慢性疾病的医疗数据的获取周期较长,且获取较为准确的治疗方案数据的人工成本较高的问题。
第一方面,本申请实施例提供了一种基于物联网数据的用户数据分类方法,其包括:
判断是否接收到物联网医疗终端上传的当前用户医疗数据;其中,所述当前用户医疗数据包括用户身份识别数据、药物注射数据、体征测量数据;
若接收到物联网医疗终端上传的当前用户医疗数据,获取与所述身份识别数据对应的历史医疗数据集;其中,所述历史医疗数据集包括电子病历数据子集、检查报告数据子集、用药记录数据子集、历史治疗方案数据子集;
调用预先设置的第一关键词集合及与所述第一关键词集合对应的第一提取策略,提取与所述身份识别数据对应的历史医疗数据集中的目标数据集合,以与所述当前用户医疗数据进行合并得到当前用户输入数据;
由所述当前用户输入数据中各字段对应的语义向量组成当前用户数据输入向量,将所述当前用户数据输入向量输入至预先训练的卷积神经网络模型,得到与所述当前用户数据输入向量对应的分类结果;
判断所述分类结果对应的治疗路径数据种类是否大于1;
若所述分类结果对应的治疗路径数据种类大于1,获取分类结果对应的各治疗路径数据种类分别对应的标准用户数据输入向量;以及
将分类结果分别对应的标准用户数据输入向量与所述当前用户数据输入向量计算向量相似度,获取向量相似度为最大值对应的标准用户数据输入向量、及对应的治疗路径数据,以作为最佳治疗路径数据。
第二方面,本申请实施例提供了一种基于物联网数据的用户数据分类装置,其包括:
当前数据接收单元,用于判断是否接收到物联网医疗终端上传的当前用户医疗数据;其中,所述当前用户医疗数据包括用户身份识别数据、药物注射数据、体征测量数据;
历史数据搜索单元,用于若接收到物联网医疗终端上传的当前用户医疗数据,获取与所述身份识别数据对应的历史医疗数据集;其中,所述历史医疗数据集包括电子病历数据子集、检查报告数据子集、用药记录数据子集、历史治疗方案数据子集;
当前用户输入数据获取单元,用于调用预先设置的第一关键词集合及与所述第一关键词集合对应的第一提取策略,提取与所述身份识别数据对应的历史医疗数据集中的目标数据集合,以与所述当前用户医疗数据进行合并得到当前用户输入数据;
分类结果获取单元,用于由所述当前用户输入数据中各字段对应的语义向量组成当前用户数据输入向量,将所述当前用户数据输入向量输入至预先训练的卷积神经网络模型,得到与所述当前用户数据输入向量对应的分类结果;
数据种类判断单元,用于判断所述分类结果对应的治疗路径数据种类是否大于1;
标准向量获取单元,用于若所述分类结果对应的治疗路径数据种类大于1,获取分类结果对应的各治疗路径数据种类分别对应的标准用户数据输入向量;以及
最佳路径数据获取单元,用于将分类结果分别对应的标准用户数据输入向量与所述当前用户数据输入向量计算向量相似度,获取向量相似度为最大值对应的标准用户数据输入向量、及对应的治疗路径数据,以作为最佳治疗路径数据。
第三方面,本申请实施例又提供了一种计算机设备,其包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序,所述处理器执行所述计算机程序时实现以下步骤:
判断是否接收到物联网医疗终端上传的当前用户医疗数据;其中,所述当前用户医疗数据包括用户身份识别数据、药物注射数据、体征测量数据;
若接收到物联网医疗终端上传的当前用户医疗数据,获取与所述身份识别数据对应的历史医疗数据集;其中,所述历史医疗数据集包括电子病历数据子集、检查报告数据子集、用药记录数据子集、历史治疗方案数据子集;
调用预先设置的第一关键词集合及与所述第一关键词集合对应的第一提取策略,提取与所述身份识别数据对应的历史医疗数据集中的目标数据集合,以与所述当前用户医疗数据进行合并得到当前用户输入数据;
由所述当前用户输入数据中各字段对应的语义向量组成当前用户数据输入向量,将所述当前用户数据输入向量输入至预先训练的卷积神经网络模型,得到与所述当前用户数据输入向量对应的分类结果;
判断所述分类结果对应的治疗路径数据种类是否大于1;
若所述分类结果对应的治疗路径数据种类大于1,获取分类结果对应的各治疗路径数据种类分别对应的标准用户数据输入向量;以及
将分类结果分别对应的标准用户数据输入向量与所述当前用户数据输入向量计算向量相似度,获取向量相似度为最大值对应的标准用户数据输入向量、及对应的治疗路径数据,以作为最佳治疗路径数据。
第四方面,本申请实施例还提供了一种计算机可读存储介质,其中所述计算机可读存储介质存储有计算机程序,所述计算机程序当被处理器执行时使所述处理器执行以下操作:
判断是否接收到物联网医疗终端上传的当前用户医疗数据;其中,所述当前用户医疗数据包括用户身份识别数据、药物注射数据、体征测量数据;
若接收到物联网医疗终端上传的当前用户医疗数据,获取与所述身份识别数据对应的历史医疗数据集;其中,所述历史医疗数据集包括电子病历数据子集、检查报告数据子集、用药记录数据子集、历史治疗方案数据子集;
调用预先设置的第一关键词集合及与所述第一关键词集合对应的第一提取策略,提取与所述身份识别数据对应的历史医疗数据集中的目标数据集合,以与所述当前用户医疗数据进行合并得到当前用户输入数据;
由所述当前用户输入数据中各字段对应的语义向量组成当前用户数据输入向量,将所述 当前用户数据输入向量输入至预先训练的卷积神经网络模型,得到与所述当前用户数据输入向量对应的分类结果;
判断所述分类结果对应的治疗路径数据种类是否大于1;
若所述分类结果对应的治疗路径数据种类大于1,获取分类结果对应的各治疗路径数据种类分别对应的标准用户数据输入向量;以及
将分类结果分别对应的标准用户数据输入向量与所述当前用户数据输入向量计算向量相似度,获取向量相似度为最大值对应的标准用户数据输入向量、及对应的治疗路径数据,以作为最佳治疗路径数据。
本申请实施例提供了一种基于物联网数据的用户数据分类方法、装置、计算机设备及存储介质,该方法实现了基于物联网设备对用户医疗数据的快速获取,且基于历史医疗数据快速且准确的推荐最佳治疗方案数据,降低获取最佳治疗方案数据的成本。
为了更清楚地说明本申请实施例技术方案,下面将对实施例描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1为本申请实施例提供的基于物联网数据的用户数据分类方法的应用场景示意图;
图2为本申请实施例提供的基于物联网数据的用户数据分类方法的流程示意图;
图3为本申请实施例提供的基于物联网数据的用户数据分类装置的示意性框图;
图4为本申请实施例提供的计算机设备的示意性框图。
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本申请保护的范围。
应当理解,当在本说明书和所附权利要求书中使用时,术语“包括”和“包含”指示所描述特征、整体、步骤、操作、元素和/或组件的存在,但并不排除一个或多个其它特征、整体、步骤、操作、元素、组件和/或其集合的存在或添加。
还应当理解,在此本申请说明书中所使用的术语仅仅是出于描述特定实施例的目的而并不意在限制本申请。如在本申请说明书和所附权利要求书中所使用的那样,除非上下文清楚地指明其它情况,否则单数形式的“一”、“一个”及“该”意在包括复数形式。
还应当进一步理解,在本申请说明书和所附权利要求书中使用的术语“和/或”是指相关联列出的项中的一个或多个的任何组合以及所有可能组合,并且包括这些组合。
请参阅图1和图2,图1为本申请实施例提供的基于物联网数据的用户数据分类方法的应用场景示意图;图2为本申请实施例提供的基于物联网数据的用户数据分类方法的流程示意图,该基于物联网数据的用户数据分类方法应用于云服务器中,该方法通过安装于云服务器中的应用软件进行执行。
如图2所示,该方法包括步骤S110~S170。
S110、判断是否接收到物联网医疗终端上传的当前用户医疗数据;其中,所述当前用户医疗数据包括用户身份识别数据、药物注射数据、体征测量数据。
在本实施例中,为了更清楚的理解技术方案,对具体实现场景所涉及的终端进行详细介绍。
一是物联网医疗终端,物联网医疗终端具体实施时可以是血糖仪、胰岛素注射器、温度计等物联网装置,这些物联网设备与传统的血糖仪、胰岛素注射器、温度计的不同之处在于,这些物联网设备上设置有低功耗蓝牙模块、4G/5G通讯模块等无线通讯模块,通过这一改造 之后,通过物联网医疗终端检测到的一些参数值是:1)可以直接上传至云服务器;2)可以先与智能手机等智能终端互联后上传参数值至智能终端,之后再由智能终端上传至服务器。例如一慢性病患者(如高血压、糖尿病、心脑血管疾病等疾病患者)需要长期使用这些物联网设备测量血糖、体温等参数,或者是定时定量注射胰岛素,此时通过物联网医疗终端检测到对应的参数值后,即可最终联网上传至云服务器作为慢性病监测的参数值。
二是智能终端,例如智能手机等,其用于接收与之绑定的物联网医疗终端上传的当前用户医疗数据,将当前用户医疗数据发送至云服务器进行存储。
三是云服务器,用于接收物联网医疗终端或是智能终端上传的当前用户医疗数据,并将该当前用户医疗数据进行结构化数据转化后与相对应的用户的历史数据进行存储。而且能基于当前用户医疗数据在数据库中匹配出相似病例数据和推荐治疗路径数据。进一步的,在云服务器中还能针对各用户对应的数据生成患者画像、患者用药数据曲线、用药风险情况预测等。
当用户使用物联网医疗终端进行血糖、体温等参数测量,或者是定时定量注射胰岛素时,可以准确获取当前用户医疗数据。由于为了识别是哪一用户使用的物联网医疗终端设备上传的用户医疗数据,物联网医疗终端设备在初始化时需要进行用户身份识别数据的绑定。
例如,参考现有的智能手环与智能手机的蓝牙连接和用户身份识别数据的初次绑定过程,当物联网医疗终端与智能终端进行蓝牙连接后,在智能终端的用户交互界面上会获取在物联网医疗终端中已经存储的历史用户医疗数据(一般初始化连接时,物联网医疗终端中已经存储的历史用户医疗数据为空集,也就是还未存储数据),而且还会提示是否与该物联网医疗终端进行绑定。一旦用户选择将该该物联网医疗终端与智能终端进行绑定,相当于将该物联网医疗终端的设备唯一识别码与智能终端的设备唯一识别码建立了映射对应关系。而且由于该智能终端的设备唯一识别码又是与用户身份识别数据是有对应关系(例如,用户使用了微信号码绑定了该智能终端的设备唯一识别码),故此时物联网医疗终端将当前用户医疗数据上传至云服务器时,对应的当前用户医疗数据中除了包括药物注射数据(例如胰岛素注射量)、体征测量数据(例如血糖值、体温值)等数据,还上传了用户身份识别数据。
S120、若接收到物联网医疗终端上传的当前用户医疗数据,获取与所述身份识别数据对应的历史医疗数据集;其中,所述历史医疗数据集包括电子病历数据子集、检查报告数据子集、用药记录数据子集、历史治疗方案数据子集。
在本实施例中,当云服务器接收了物联网医疗终端上传的当前用户医疗数据后,需获知是哪一用户对应的数据,从而确定将该当前用户医疗数据存储至该用户对应的数据表中。
其中,云服务器与所述身份识别数据对应的历史医疗数据集中至少有四个数据来源,而分别是医院的医疗***中存储的电子病历、检查报告、历史用药数据以及历史治疗方案。由于医疗***中存储的上述数据均是非结构化数据,故云服务器获取了来自医疗***中的电子病历、检查报告、历史用药数据以及历史治疗这些数据后,需要进行数据清洗和结构化存储。若云服务器未接收到物联网医疗终端上传的当前用户医疗数据,在等待预设的延迟等待时间(如10s)后,继续返回执行步骤S110。
在一实施例中,步骤S120之前还包括:
接收医疗***服务器上传的电子病历、检查报告、历史用药数据以及历史治疗方案;
通过调用预先设置的第二关键词集合及与所述第二关键词集合对应的第二提取策略,提取所述电子病历中对应的电子病历数据初始子集;
通过调用预先设置的第三关键词集合及与所述第三关键词集合对应的第三提取策略,提取所述检查报告中对应的检查报告数据初始子集;
对所述历史用药数据对应的图片通过OCR文本识别模型进行文本识别,得到与所述历史用药数据对应的识别文本,通过预先设置的第四关键词集合及与所述第四关键词集合对应的第四提取策略,提取所述历史用药数据对应的用药记录数据初始子集;
通过预先设置的第五关键词集合及与所述第五关键词集合对应的第五提取策略,提取所 述历史治疗方案对应的历史治疗方案数据初始子集。
在本实施例中,由于医疗***服务器上会存储有大量患者分别对应的电子病历、检查报告、历史用药数据以及历史治疗方案的数据。
其中,患者的电子病历是医生对患者进行检查时直接用电脑操作记录的患者相关信息,至少包括有首页、病程记录、检查检验结果、医嘱、手术记录、护理记录等。
患者的检查报告一般是验血检查报告、CT检查报告等检查报告,例如患者的CT检查报告一般包括患者的姓名、性别、年龄、科别、住院号、床号、***位、登记日期、检查名称、检查方法、影像表现、报告医生等信息。
患者的历史用药数据一般是书写在医生处方笺中,医生可拍照医生处方笺后上传至医疗***服务器上。在医生处方笺中,一般包括患者的姓名、姓名、年龄、科别、临床诊断、住址/电话、开药清单、处方医师、调配者、校对者、药价、开药日期等信息。
患者的历史治疗方案一般是医生直接登记在医疗***服务器中,也即针对患者的每一次看病经历,均会对应的记录一次治疗方案。
由于电子病历中信息众多,为了选取其中的关键信息实现数据筛选和清洗,此时可以预先设置好第二关键词集合,和与所述第二关键词集合对应的第二提取策略。
例如设置的第二关键词集合中包括的关键词包括患者姓名、身份证号码、性别、年龄、患者所属地区区域、家族病史、病程记录、检查检验结果、医嘱、手术记录、护理记录,第二提取策略即是将第二关键词集合对应的各关键词的具体取值进行提取。
例如设置的第三关键词集合中包括的关键词包括姓名、身份证号码、性别、年龄、科别、住院号、床号、***位、登记日期、检查名称、检查方法、影像表现(也可以理解为检查结果)、报告医生,第三提取策略即是将第三关键词集合对应的各关键词的具体取值进行提取。
由于患者的历史用药数据一般是书写在医生处方笺中,且采用拉丁文或中文书写,此时通过OCR文本识别模型对历史用药数据对应的图片进行文本识别,得到与历史用药数据对应的识别文本(识别文本即是计算机可以理解的计算机文字),以作为与所述历史用药数据对应的识别文本。
之后,例如设置的第四关键词集合中包括的关键词包括姓名、身份证号码、年龄、科别、临床诊断、住址/电话、开药清单、处方医师、调配者、校对者、药价、开药日期,第四提取策略即是将第四关键词集合对应的各关键词的具体取值进行提取。
例如设置的第五关键词集合中包括的关键词包括姓名、身份证号码、性别、年龄、科别、住院号、治疗方案流程、主治医生,第五提取策略即是将第五关键词集合对应的各关键词的具体取值进行提取。
由于电子病历数据子集、检查报告数据子集、用药记录数据子集、历史治疗方案数据子集中均包括患者姓名、身份证号码、性别这些共有字段,此时可以将电子病历数据子集、检查报告数据子集、用药记录数据子集、历史治疗方案数据子集中具有相同姓名、身份证号码、性别的一条数据,合并成同一患者的历史医疗数据集。
当通过上述处理之后,即实现了对用户对应的历史原始医疗数据进行数据清洗,从而获取每一用户对应的历史医疗数据初始集,此时还可对每一用户对应的历史医疗数据初始集中每一数据子集每一字段的具体取值进行数据结构化转化;例如所述历史治疗方案对应的历史治疗方案数据初始子集中治疗方案流程这一字段的具体取值是一大段文字组成,为了将其进行结构化数据转化,可以对其进行关键词抽取,得到治疗方案流程这一字段的关键词组合,以作为治疗方案流程这一字段对应的结构化数据。
在一实施例中,所述通过预先设置的第五关键词集合及与所述第五关键词集合对应的第五提取策略,提取所述历史治疗方案对应的历史治疗方案数据初始子集之后,还包括:
将所述电子病历数据初始子集、检查报告数据初始子集、用药记录数据初始子集、历史治疗方案数据初始子集中各字段的取值进行关键词抽取,以得到与所述电子病历数据初始子 集对应的电子病历数据子集、与所述检查报告数据初始子集对应的检查报告数据子集、与所述用药记录数据初始子集对应的用药记录数据子集、与所述历史治疗方案数据初始子集对应的历史治疗方案数据子集。
在本实施例中,例如历史治疗方案数据初始子集中治疗方案流程这一字段的具体取值对应的一大段文字描述,先是将治疗方案流程这一字段的具体取值通过基于概率统计分词模型进行分词,得到与治疗方案流程这一字段的具体取值对应的分词结果;
然后通过词频-逆文本频率指数模型,抽取所述分词结果中未超出预设的排名值之前的关键词,以作为治疗方案流程这一字段的具体取值对应的结构化数据;其他各字段的具体取值均经过关键词提取后即实现了数据结构化转化。
S130、调用预先设置的第一关键词集合及与所述第一关键词集合对应的第一提取策略,提取与所述身份识别数据对应的历史医疗数据集中的目标数据集合,以与所述当前用户医疗数据进行合并得到当前用户输入数据。
在本实施例中,由于当前用户通过物联网医疗终端上传的当前用户医疗数据,一般也仅包括用户身份识别数据、药物注射数据、体征测量数据这些取值十分结构化的字段取值,此时无需对所述当前用户医疗数据进行结构化数据转化,直接与所述当前用户医疗数据进行合并得到当前用户输入数据。
在一实施例中,所述第一关键词集合及与所述第一关键词集合对应的第一提取策略存储于区块链网络中。同样的,所述第二关键词集合及与所述第二关键词集合对应的第二提取策略存储于区块链网络中;所述第三关键词集合及与所述第三关键词集合对应的第三提取策略存储于区块链网络中;所述第四关键词集合及与所述第四关键词集合对应的第四提取策略存储于区块链网络中;所述第五关键词集合及与所述第五关键词集合对应的第五提取策略存储于区块链网络中。
在本实施例中,云服务器中的上述关键词集合及提取策略均可存储于区块链网络中。云服务器可作为区块链网络中的其中一个区块链节点设备。区块链是分布式数据存储、点对点传输、共识机制、加密算法等计算机技术的新型应用模式。区块链(Blockchain),本质上是一个去中心化的数据库,是一串使用密码学方法相关联产生的数据块,每一个数据块中包含了一批次网络交易的信息,用于验证其信息的有效性(防伪)和生成下一个区块。区块链可以包括区块链底层平台、平台产品服务层以及应用服务层。
在一实施例中,步骤S130包括:
调用所述第一关键词集合及与所述第一关键词集合对应的第一提取策略,在历史医疗数据集中获取与所述身份识别数据相对应且与第一关键词集合中各字段分别对应的目标数据集合;
将所述目标数据集合中与所述当前用户医疗数据中重复字段进行数据组合并去重,得到当前用户输入数据。
在本实施例中,但是一般由于当前用户医疗数据和对应的历史医疗数据集中的目标数据集合中有重复的字段,例如用户身份识别数据是当前用户医疗数据和对应的历史医疗数据集中的目标数据集合的重复字段,此时将这一重复字段进行去重并保留其一即可。
例如,当前用户医疗数据包括用户身份识别数据、药物注射数据、体征测量数据这3个字段及其具体字段取值,而与所述身份识别数据对应的历史医疗数据集中的目标数据集合包括患者姓名、用户身份识别数据(具体就是身份证号码)、患者所属地区区域、家族病史、患者病症、检查结果、开药清单这7个字段及其具体字段取值,由于用户身份识别数据这一字段是重复的,当前用户医疗数据与所述当前用户医疗数据进行合并,所得到当前用户输入数据示包括用户身份识别数据、药物注射数据、体征测量数据、患者姓名、患者所属地区区域、家族病史、患者病症、检查结果、开药清单这10个字段及其具体字段取值。
S140、由所述当前用户输入数据中各字段对应的语义向量组成当前用户数据输入向量,将所述当前用户数据输入向量输入至预先训练的卷积神经网络模型,得到与所述当前用户数 据输入向量对应的分类结果。
在本实施例中,例如获取的当前用户输入数据中包括用户身份识别数据、药物注射数据、体征测量数据、患者姓名、患者所属地区区域、家族病史、患者病症、检查结果、开药清单这10个字段及其具体字段取值,此时可以根据这10个字段对应的具体取值相对应的词向量,组成与这10个字段对应的语义向量(即根据10个词向量,及10个词向量分别对应的权重,获取这10个字段对应的语义向量)。
当获取了与所述当前用户输入数据中各字段对应的语义向量组成当前用户数据输入向量后,再调用云服务器中已存储且完成了训练的卷积神经网络模型,将所述当前用户数据输入向量输入至预先训练的卷积神经网络模型,即可得到与所述当前用户数据输入向量对应的分类结果。
例如,先在云服务器中对待训练卷积神经网络模型进行训练时,一般是先获取训练集,其中训练集中每一条训练数据均包括两个部分:一是训练集输入数据,其包括用户身份识别数据、药物注射数据、体征测量数据、患者姓名、患者所属地区区域、家族病史、患者病症、检查结果、开药清单这10个字段的具体字段取值(一般这些字段的具体字段取值已被转化为关键字组合,故可以方便的将这10个字段的具体字段取值转化为对应的词向量以组成语义向量,从而由10个字段的具体取值对应的语义向量组成对应的训练集输入向量数据);二是训练集输出数据,其包括治疗路径分类结果,一般针对每一种类型的疾病均至少对应一种治疗路径,故该分类结果一般是判断基于当前用户数据输入向量所能得到的分类结果,从而确定该疾病所归属的治疗路径。
在一实施例中,步骤S140包括:
获取所述当前用户输入数据中各字段对应的语义向量,将每一语义向量进行求和,以得到与所述当前用户输入数据对应的当前用户数据输入向量;
调用预先存储的卷积神经网络模型;
将与所述当前用户输入数据对应的当前用户数据输入向量作为所述卷积神经网络模型的输入,得到对应的分类结果。
在本实施例中,假设所述当前用户输入数据中各字段对应的语义向量均是1*300的一维行向量,此时将当前用户输入数据中各字段对应的语义向量进行向量加法,从而得到了与所述当前用户输入数据对应的当前用户数据输入向量。之后将与所述当前用户输入数据对应的当前用户数据输入向量作为所述卷积神经网络模型的输入进行分类运算,即可得到对应的分类结果。
S150、判断所述分类结果对应的治疗路径数据种类是否大于1。
在本实施例中,由于根据所述当前用户输入数据确定了分类结果后,由于每一分类结果对应至少一种类型的治疗路径数据,治疗路径数据一般包括治疗方案流程字段的具体取值。故当确定了当前用户输入数据对应的分类结果后,即可获取该当前用户数据对应的治疗方案流程。若该分类结果对应的治疗路径数据种类是大于1,此时还可以向当前用户输入数据对应的用户推荐最佳治疗路径数据。
S160、若所述分类结果对应的治疗路径数据种类大于1,获取分类结果对应的各治疗路径数据种类分别对应的标准用户数据输入向量。
在本实施例中,当通过训练集对待训练的卷积神经网络进行训练时,一般训练集包括多个种类的训练子集,每一训练子集对应的一个种类的治疗路径数据以作为训练集输出数据。这样在每一训练子集中选定一条训练数据,以作为该种类的治疗路径数据对应的标准训练集输入数据。
例如,包括1000个训练子集,对应了1000种治疗路径数据的训练集输入数据,在1000个训练子集中每一训练子集中均选定一个训练集输入数据,以作各训练子集对应的标准训练集输入数据。由于各训练子集对应的标准训练集输入数据,也是对应一个标准用户数据输入向量。
当获取了所述分类结果为1000种治疗路径数据中其中3种治疗路径数据,此时获取这3种治疗路径数据对应的标准用户数据输入向量。
若所述分类结果对应的治疗路径数据种类等于1,则获取该治疗路径数据种类对应的治疗路径数据,以作为最佳治疗路径数据。
S170、将分类结果分别对应的标准用户数据输入向量与所述当前用户数据输入向量计算向量相似度,获取向量相似度为最大值对应的标准用户数据输入向量、及对应的治疗路径数据,以作为最佳治疗路径数据。
在本实施例中,当获取了上述3种治疗路径数据对应的标准用户数据输入向量后,若其中一种治疗路径数据对应的标准用户数据输入向量与所述当前用户数据输入向量之间的向量相似度为最大值,此时选取该治疗路径数据对应的治疗路径数据,以作为最佳治疗路径数据。
在计算向量之间的相似度时,可以采用欧几里得距离(Eucledian Distance)、余弦相似度(Cosine Similarity)、明可夫斯基距离(Minkowski Distance)、杰卡德相似系数(Jaccard Similarity)等算法。
此时即可将最佳治疗路径数据推送至所述当前用户输入数据对应的智能终端。进一步的,在云服务器中还能针对各用户对应的数据生成患者画像、患者用药数据曲线、用药风险情况预测等,这些数据的生成都可基于用户的历史医疗数据集。
在一实施例中,步骤S170之后还包括:
通过调用预先设置的第六关键词集合及与所述第六关键词集合对应的第六提取策略,提取所述历史医疗数据集中各用户对应的目标画像数据集;
根据各用户对应的目标画像数据集及预先存储的关键词筛选策略,将各用户对应的目标画像数据集中关键词进行筛选,得到各用户对应的目标画像数据简化集;
调用预先存储的关键词转化策略,将各用户对应的目标画像数据简化集对应转化为用户画像数据。
在本实施例中,例如设置的第六关键词集合中包括的关键词包括患者所属地区区域、家族病史、患者病症、检查结果,第六提取策略即是将第六关键词集合对应的各关键词的具体取值进行提取,此时将第六关键词集合对应的各关键词的具体取值进行提取即组成了所述历史医疗数据集中各用户对应的目标画像数据集。
由于上述各用户对应的目标画像数据集对应的关键词还是较多,而且有些是可能还是具体的数字参数,此时可以筛选其中重要的关键词并将该关键词转化为对应的标签。
在云服务器中设置了各关键词对应的标签转化策略(也可以将关键词转化为对应标签的策略),此时即可通过关键词的标签转化后,得到各用户对应的用户画像数据。
该方法实现了基于物联网设备对用户医疗数据的快速获取,且基于历史医疗数据快速且准确的推荐最佳治疗方案数据,降低获取最佳治疗方案数据的成本。该方法属于智慧医疗领域,通过本方案能够推动智慧城市的建设。
本申请实施例还提供一种基于物联网数据的用户数据分类装置,该基于物联网数据的用户数据分类装置用于执行前述基于物联网数据的用户数据分类方法的任一实施例。具体地,请参阅图3,图3是本申请实施例提供的基于物联网数据的用户数据分类装置的示意性框图。该基于物联网数据的用户数据分类装置100可以配置于云服务器中。
如图3所示,基于物联网数据的用户数据分类装置100包括:当前数据接收单元110、历史数据搜索单元120、当前用户输入数据获取单元130、分类结果获取单元140、数据种类判断单元150、标准向量获取单元160、最佳路径数据获取单元170。
当前数据接收单元110,用于判断是否接收到物联网医疗终端上传的当前用户医疗数据;其中,所述当前用户医疗数据包括用户身份识别数据、药物注射数据、体征测量数据。
在本实施例中,当用户使用物联网医疗终端进行血糖、体温等参数测量,或者是定时定量注射胰岛素时,可以准确获取当前用户医疗数据。由于为了识别是哪一用户使用的物联网医疗终端设备上传的用户医疗数据,物联网医疗终端设备在初始化时需要进行用户身份识别 数据的绑定。
例如,参考现有的智能手环与智能手机的蓝牙连接和用户身份识别数据的初次绑定过程,当物联网医疗终端与智能终端进行蓝牙连接后,在智能终端的用户交互界面上会获取在物联网医疗终端中已经存储的历史用户医疗数据(一般初始化连接时,物联网医疗终端中已经存储的历史用户医疗数据为空集,也就是还未存储数据),而且还会提示是否与该物联网医疗终端进行绑定。一旦用户选择将该该物联网医疗终端与智能终端进行绑定,相当于将该物联网医疗终端的设备唯一识别码与智能终端的设备唯一识别码建立了映射对应关系。而且由于该智能终端的设备唯一识别码又是与用户身份识别数据是有对应关系(例如,用户使用了微信号码绑定了该智能终端的设备唯一识别码),故此时物联网医疗终端将当前用户医疗数据上传至云服务器时,对应的当前用户医疗数据中除了包括药物注射数据(例如胰岛素注射量)、体征测量数据(例如血糖值、体温值)等数据,还上传了用户身份识别数据。
历史数据搜索单元120,用于若接收到物联网医疗终端上传的当前用户医疗数据,获取与所述身份识别数据对应的历史医疗数据集;其中,所述历史医疗数据集包括电子病历数据子集、检查报告数据子集、用药记录数据子集、历史治疗方案数据子集。
在本实施例中,当云服务器接收了物联网医疗终端上传的当前用户医疗数据后,需获知是哪一用户对应的数据,从而确定将该当前用户医疗数据存储至该用户对应的数据表中。
其中,云服务器与所述身份识别数据对应的历史医疗数据集中至少有四个数据来源,而分别是医院的医疗***中存储的电子病历、检查报告、历史用药数据以及历史治疗方案。由于医疗***中存储的上述数据均是非结构化数据,故云服务器获取了来自医疗***中的电子病历、检查报告、历史用药数据以及历史治疗这些数据后,需要进行数据清洗和结构化存储。若云服务器未接收到物联网医疗终端上传的当前用户医疗数据,在等待预设的延迟等待时间(如10s)后,继续返回执行判断是否接收到物联网医疗终端上传的当前用户医疗数据的步骤。
在一实施例中,基于物联网数据的用户数据分类装置100还包括:
历史数据上传接收单元,用于接收医疗***服务器上传的电子病历、检查报告、历史用药数据以及历史治疗方案;
电子病历数据初始子集获取单元,用于通过调用预先设置的第二关键词集合及与所述第二关键词集合对应的第二提取策略,提取所述电子病历中对应的电子病历数据初始子集;
检查报告数据初始子集获取单元,用于通过调用预先设置的第三关键词集合及与所述第三关键词集合对应的第三提取策略,提取所述检查报告中对应的检查报告数据初始子集;
用药记录数据初始子集获取单元,用于对所述历史用药数据对应的图片通过OCR文本识别模型进行文本识别,得到与所述历史用药数据对应的识别文本,通过预先设置的第四关键词集合及与所述第四关键词集合对应的第四提取策略,提取所述历史用药数据对应的用药记录数据初始子集;
历史治疗方案数据初始子集获取单元,用于通过预先设置的第五关键词集合及与所述第五关键词集合对应的第五提取策略,提取所述历史治疗方案对应的历史治疗方案数据初始子集。
在本实施例中,由于医疗***服务器上会存储有大量患者分别对应的电子病历、检查报告、历史用药数据以及历史治疗方案的数据。
其中,患者的电子病历是医生对患者进行检查时直接用电脑操作记录的患者相关信息,至少包括有首页、病程记录、检查检验结果、医嘱、手术记录、护理记录等。
患者的检查报告一般是验血检查报告、CT检查报告等检查报告,例如患者的CT检查报告一般包括患者的姓名、性别、年龄、科别、住院号、床号、***位、登记日期、检查名称、检查方法、影像表现、报告医生等信息。
患者的历史用药数据一般是书写在医生处方笺中,医生可拍照医生处方笺后上传至医疗***服务器上。在医生处方笺中,一般包括患者的姓名、姓名、年龄、科别、临床诊断、住 址/电话、开药清单、处方医师、调配者、校对者、药价、开药日期等信息。
患者的历史治疗方案一般是医生直接登记在医疗***服务器中,也即针对患者的每一次看病经历,均会对应的记录一次治疗方案。
由于电子病历中信息众多,为了选取其中的关键信息实现数据筛选和清洗,此时可以预先设置好第二关键词集合,和与所述第二关键词集合对应的第二提取策略。
例如设置的第二关键词集合中包括的关键词包括患者姓名、身份证号码、性别、年龄、患者所属地区区域、家族病史、病程记录、检查检验结果、医嘱、手术记录、护理记录,第二提取策略即是将第二关键词集合对应的各关键词的具体取值进行提取。
例如设置的第三关键词集合中包括的关键词包括姓名、身份证号码、性别、年龄、科别、住院号、床号、***位、登记日期、检查名称、检查方法、影像表现(也可以理解为检查结果)、报告医生,第三提取策略即是将第三关键词集合对应的各关键词的具体取值进行提取。
由于患者的历史用药数据一般是书写在医生处方笺中,且采用拉丁文或中文书写,此时通过OCR文本识别模型对历史用药数据对应的图片进行文本识别,得到与历史用药数据对应的识别文本(识别文本即是计算机可以理解的计算机文字),以作为与所述历史用药数据对应的识别文本。
之后,例如设置的第四关键词集合中包括的关键词包括姓名、身份证号码、年龄、科别、临床诊断、住址/电话、开药清单、处方医师、调配者、校对者、药价、开药日期,第四提取策略即是将第四关键词集合对应的各关键词的具体取值进行提取。
例如设置的第五关键词集合中包括的关键词包括姓名、身份证号码、性别、年龄、科别、住院号、治疗方案流程、主治医生,第五提取策略即是将第五关键词集合对应的各关键词的具体取值进行提取。
由于电子病历数据子集、检查报告数据子集、用药记录数据子集、历史治疗方案数据子集中均包括患者姓名、身份证号码、性别这些共有字段,此时可以将电子病历数据子集、检查报告数据子集、用药记录数据子集、历史治疗方案数据子集中具有相同姓名、身份证号码、性别的一条数据,合并成同一患者的历史医疗数据集。
当通过上述处理之后,即实现了对用户对应的历史原始医疗数据进行数据清洗,从而获取每一用户对应的历史医疗数据初始集,此时还可对每一用户对应的历史医疗数据初始集中每一数据子集每一字段的具体取值进行数据结构化转化;例如所述历史治疗方案对应的历史治疗方案数据初始子集中治疗方案流程这一字段的具体取值是一大段文字组成,为了将其进行结构化数据转化,可以对其进行关键词抽取,得到治疗方案流程这一字段的关键词组合,以作为治疗方案流程这一字段对应的结构化数据。
在一实施例中,基于物联网数据的用户数据分类装置100,还包括:
数据子集获取单元,用于将所述电子病历数据初始子集、检查报告数据初始子集、用药记录数据初始子集、历史治疗方案数据初始子集中各字段的取值进行关键词抽取,以得到与所述电子病历数据初始子集对应的电子病历数据子集、与所述检查报告数据初始子集对应的检查报告数据子集、与所述用药记录数据初始子集对应的用药记录数据子集、与所述历史治疗方案数据初始子集对应的历史治疗方案数据子集。
在本实施例中,例如历史治疗方案数据初始子集中治疗方案流程这一字段的具体取值对应的一大段文字描述,先是将治疗方案流程这一字段的具体取值通过基于概率统计分词模型进行分词,得到与治疗方案流程这一字段的具体取值对应的分词结果;
然后通过词频-逆文本频率指数模型,抽取所述分词结果中未超出预设的排名值之前的关键词,以作为治疗方案流程这一字段的具体取值对应的结构化数据;其他各字段的具体取值均经过关键词提取后即实现了数据结构化转化。
当前用户输入数据获取单元130,用于调用预先设置的第一关键词集合及与所述第一关键词集合对应的第一提取策略,提取与所述身份识别数据对应的历史医疗数据集中的目标数 据集合,以与所述当前用户医疗数据进行合并得到当前用户输入数据。
在本实施例中,由于当前用户通过物联网医疗终端上传的当前用户医疗数据,一般也仅包括用户身份识别数据、药物注射数据、体征测量数据这些取值十分结构化的字段取值,此时无需对所述当前用户医疗数据进行结构化数据转化,直接与所述当前用户医疗数据进行合并得到当前用户输入数据。
在一实施例中,所述第一关键词集合及与所述第一关键词集合对应的第一提取策略存储于区块链网络中。同样的,所述第二关键词集合及与所述第二关键词集合对应的第二提取策略存储于区块链网络中;所述第三关键词集合及与所述第三关键词集合对应的第三提取策略存储于区块链网络中;所述第四关键词集合及与所述第四关键词集合对应的第四提取策略存储于区块链网络中;所述第五关键词集合及与所述第五关键词集合对应的第五提取策略存储于区块链网络中。
在本实施例中,云服务器中的上述关键词集合及提取策略均可存储于区块链网络中。云服务器可作为区块链网络中的其中一个区块链节点设备。区块链是分布式数据存储、点对点传输、共识机制、加密算法等计算机技术的新型应用模式。区块链(Blockchain),本质上是一个去中心化的数据库,是一串使用密码学方法相关联产生的数据块,每一个数据块中包含了一批次网络交易的信息,用于验证其信息的有效性(防伪)和生成下一个区块。区块链可以包括区块链底层平台、平台产品服务层以及应用服务层。
在一实施例中,当前用户输入数据获取单元130包括:
目标数据集合获取单元,用于调用所述第一关键词集合及与所述第一关键词集合对应的第一提取策略,在历史医疗数据集中获取与所述身份识别数据相对应且与第一关键词集合中各字段分别对应的目标数据集合;
数据去重单元,用于将所述目标数据集合中与所述当前用户医疗数据中重复字段进行数据组合并去重,得到当前用户输入数据。
在本实施例中,但是一般由于当前用户医疗数据和对应的历史医疗数据集中的目标数据集合中有重复的字段,例如用户身份识别数据是当前用户医疗数据和对应的历史医疗数据集中的目标数据集合的重复字段,此时将这一重复字段进行去重并保留其一即可。
例如,当前用户医疗数据包括用户身份识别数据、药物注射数据、体征测量数据这3个字段及其具体字段取值,而与所述身份识别数据对应的历史医疗数据集中的目标数据集合包括患者姓名、用户身份识别数据(具体就是身份证号码)、患者所属地区区域、家族病史、患者病症、检查结果、开药清单这7个字段及其具体字段取值,由于用户身份识别数据这一字段是重复的,当前用户医疗数据与所述当前用户医疗数据进行合并,所得到当前用户输入数据示包括用户身份识别数据、药物注射数据、体征测量数据、患者姓名、患者所属地区区域、家族病史、患者病症、检查结果、开药清单这10个字段及其具体字段取值。
分类结果获取单元140,用于由所述当前用户输入数据中各字段对应的语义向量组成当前用户数据输入向量,将所述当前用户数据输入向量输入至预先训练的卷积神经网络模型,得到与所述当前用户数据输入向量对应的分类结果。
在本实施例中,例如获取的当前用户输入数据中包括用户身份识别数据、药物注射数据、体征测量数据、患者姓名、患者所属地区区域、家族病史、患者病症、检查结果、开药清单这10个字段及其具体字段取值,此时可以根据这10个字段对应的具体取值相对应的词向量,组成与这10个字段对应的语义向量(即根据10个词向量,及10个词向量分别对应的权重,获取这10个字段对应的语义向量)。
当获取了与所述当前用户输入数据中各字段对应的语义向量组成当前用户数据输入向量后,再调用云服务器中已存储且完成了训练的卷积神经网络模型,将所述当前用户数据输入向量输入至预先训练的卷积神经网络模型,即可得到与所述当前用户数据输入向量对应的分类结果。
例如,先在云服务器中对待训练卷积神经网络模型进行训练时,一般是先获取训练集, 其中训练集中每一条训练数据均包括两个部分:一是训练集输入数据,其包括用户身份识别数据、药物注射数据、体征测量数据、患者姓名、患者所属地区区域、家族病史、患者病症、检查结果、开药清单这10个字段的具体字段取值(一般这些字段的具体字段取值已被转化为关键字组合,故可以方便的将这10个字段的具体字段取值转化为对应的词向量以组成语义向量,从而由10个字段的具体取值对应的语义向量组成对应的训练集输入向量数据);二是训练集输出数据,其包括治疗路径分类结果,一般针对每一种类型的疾病均至少对应一种治疗路径,故该分类结果一般是判断基于当前用户数据输入向量所能得到的分类结果,从而确定该疾病所归属的治疗路径。
在一实施例中,分类结果获取单元140包括:
当前用户数据输入向量获取单元,用于获取所述当前用户输入数据中各字段对应的语义向量,将每一语义向量进行求和,以得到与所述当前用户输入数据对应的当前用户数据输入向量;
模型调用单元,用于调用预先存储的卷积神经网络模型;
向量计算分类单元,用于将与所述当前用户输入数据对应的当前用户数据输入向量作为所述卷积神经网络模型的输入,得到对应的分类结果。
在本实施例中,假设所述当前用户输入数据中各字段对应的语义向量均是1*300的一维行向量,此时将当前用户输入数据中各字段对应的语义向量进行向量加法,从而得到了与所述当前用户输入数据对应的当前用户数据输入向量。之后将与所述当前用户输入数据对应的当前用户数据输入向量作为所述卷积神经网络模型的输入进行分类运算,即可得到对应的分类结果。
数据种类判断单元150,用于判断所述分类结果对应的治疗路径数据种类是否大于1。
在本实施例中,由于根据所述当前用户输入数据确定了分类结果后,由于每一分类结果对应至少一种类型的治疗路径数据,治疗路径数据一般包括治疗方案流程字段的具体取值。故当确定了当前用户输入数据对应的分类结果后,即可获取该当前用户数据对应的治疗方案流程。若该分类结果对应的治疗路径数据种类是大于1,此时还可以向当前用户输入数据对应的用户推荐最佳治疗路径数据。
标准向量获取单元160,用于若所述分类结果对应的治疗路径数据种类大于1,获取分类结果对应的各治疗路径数据种类分别对应的标准用户数据输入向量。
在本实施例中,当通过训练集对待训练的卷积神经网络进行训练时,一般训练集包括多个种类的训练子集,每一训练子集对应的一个种类的治疗路径数据以作为训练集输出数据。这样在每一训练子集中选定一条训练数据,以作为该种类的治疗路径数据对应的标准训练集输入数据。
例如,包括1000个训练子集,对应了1000种治疗路径数据的训练集输入数据,在1000个训练子集中每一训练子集中均选定一个训练集输入数据,以作各训练子集对应的标准训练集输入数据。由于各训练子集对应的标准训练集输入数据,也是对应一个标准用户数据输入向量。
当获取了所述分类结果为1000种治疗路径数据中其中3种治疗路径数据,此时获取这3种治疗路径数据对应的标准用户数据输入向量。
若所述分类结果对应的治疗路径数据种类等于1,则获取该治疗路径数据种类对应的治疗路径数据,以作为最佳治疗路径数据。
最佳路径数据获取单元170,用于将分类结果分别对应的标准用户数据输入向量与所述当前用户数据输入向量计算向量相似度,获取向量相似度为最大值对应的标准用户数据输入向量、及对应的治疗路径数据,以作为最佳治疗路径数据。
在本实施例中,当获取了上述3种治疗路径数据对应的标准用户数据输入向量后,若其中一种治疗路径数据对应的标准用户数据输入向量与所述当前用户数据输入向量之间的向量相似度为最大值,此时选取该治疗路径数据对应的治疗路径数据,以作为最佳治疗路径数据。
在计算向量之间的相似度时,可以采用欧几里得距离(Eucledian Distance)、余弦相似度(Cosine Similarity)、明可夫斯基距离(Minkowski Distance)、杰卡德相似系数(Jaccard Similarity)等算法。
此时即可将最佳治疗路径数据推送至所述当前用户输入数据对应的智能终端。进一步的,在云服务器中还能针对各用户对应的数据生成患者画像、患者用药数据曲线、用药风险情况预测等,这些数据的生成都可基于用户的历史医疗数据集。
在一实施例中,基于物联网数据的用户数据分类装置100还包括:
目标画像数据集获取单元,用于通过调用预先设置的第六关键词集合及与所述第六关键词集合对应的第六提取策略,提取所述历史医疗数据集中各用户对应的目标画像数据集;
数据集简化单元,用于根据各用户对应的目标画像数据集及预先存储的关键词筛选策略,将各用户对应的目标画像数据集中关键词进行筛选,得到各用户对应的目标画像数据简化集;
用户画像数据生成单元,用于调用预先存储的关键词转化策略,将各用户对应的目标画像数据简化集对应转化为用户画像数据。
在本实施例中,例如设置的第六关键词集合中包括的关键词包括患者所属地区区域、家族病史、患者病症、检查结果,第六提取策略即是将第六关键词集合对应的各关键词的具体取值进行提取,此时将第六关键词集合对应的各关键词的具体取值进行提取即组成了所述历史医疗数据集中各用户对应的目标画像数据集。
由于上述各用户对应的目标画像数据集对应的关键词还是较多,而且有些是可能还是具体的数字参数,此时可以筛选其中重要的关键词并将该关键词转化为对应的标签。
在云服务器中设置了各关键词对应的标签转化策略(也可以将关键词转化为对应标签的策略),此时即可通过关键词的标签转化后,得到各用户对应的用户画像数据。
该装置实现了基于物联网设备对用户医疗数据的快速获取,且基于历史医疗数据快速且准确的推荐最佳治疗方案数据,降低获取最佳治疗方案数据的成本。
上述基于物联网数据的用户数据分类装置可以实现为计算机程序的形式,该计算机程序可以在如图4所示的计算机设备上运行。
请参阅图4,图4是本申请实施例提供的计算机设备的示意性框图。该计算机设备500是服务器,服务器可以是独立的服务器,也可以是多个服务器组成的服务器集群。
参阅图4,该计算机设备500包括通过***总线501连接的处理器502、存储器和网络接口505,其中,存储器可以包括非易失性存储介质503和内存储器504。
该非易失性存储介质503可存储操作***5031和计算机程序5032。该计算机程序5032被执行时,可使得处理器502执行基于物联网数据的用户数据分类方法。
该处理器502用于提供计算和控制能力,支撑整个计算机设备500的运行。
该内存储器504为非易失性存储介质503中的计算机程序5032的运行提供环境,该计算机程序5032被处理器502执行时,可使得处理器502执行基于物联网数据的用户数据分类方法。
该网络接口505用于进行网络通信,如提供数据信息的传输等。本领域技术人员可以理解,图4中示出的结构,仅仅是与本申请方案相关的部分结构的框图,并不构成对本申请方案所应用于其上的计算机设备500的限定,具体的计算机设备500可以包括比图中所示更多或更少的部件,或者组合某些部件,或者具有不同的部件布置。
其中,所述处理器502用于运行存储在存储器中的计算机程序5032,以实现本申请实施例公开的基于物联网数据的用户数据分类方法。
本领域技术人员可以理解,图4中示出的计算机设备的实施例并不构成对计算机设备具体构成的限定,在其他实施例中,计算机设备可以包括比图示更多或更少的部件,或者组合某些部件,或者不同的部件布置。例如,在一些实施例中,计算机设备可以仅包括存储器及处理器,在这样的实施例中,存储器及处理器的结构及功能与图4所示实施例一致,在此不再赘述。
应当理解,在本申请实施例中,处理器502可以是中央处理单元(Central Processing Unit,CPU),该处理器502还可以是其他通用处理器、数字信号处理器(Digital Signal Processor,DSP)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现成可编程门阵列(Field-Programmable Gate Array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。其中,通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
在本申请的另一实施例中提供计算机可读存储介质。该计算机可读存储介质可以为非易失性的计算机可读存储介质,也可以是易失性的计算机可读存储介质。该计算机可读存储介质存储有计算机程序,其中计算机程序被处理器执行时实现本申请实施例公开的基于物联网数据的用户数据分类方法。
所属领域的技术人员可以清楚地了解到,为了描述的方便和简洁,上述描述的设备、装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。本领域普通技术人员可以意识到,结合本文中所公开的实施例描述的各示例的单元及算法步骤,能够以电子硬件、计算机软件或者二者的结合来实现,为了清楚地说明硬件和软件的可互换性,在上述说明中已经按照功能一般性地描述了各示例的组成及步骤。这些功能究竟以硬件还是软件方式来执行取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但是这种实现不应认为超出本申请的范围。
在本申请所提供的几个实施例中,应该理解到,所揭露的设备、装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为逻辑功能划分,实际实现时可以有另外的划分方式,也可以将具有相同功能的单元集合成一个单元,例如多个单元或组件可以结合或者可以集成到另一个***,或一些特征可以忽略,或不执行。另外,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口、装置或单元的间接耦合或通信连接,也可以是电的,机械的或其它的形式连接。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本申请实施例方案的目的。
另外,在本申请各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以是两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个存储介质中。基于这样的理解,本申请的技术方案本质上或者说对现有技术做出贡献的部分,或者该技术方案的全部或部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(ROM,Read-Only Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述,仅为本申请的具体实施方式,但本申请的保护范围并不局限于此,任何熟悉本技术领域的技术人员在本申请揭露的技术范围内,可轻易想到各种等效的修改或替换,这些修改或替换都应涵盖在本申请的保护范围之内。因此,本申请的保护范围应以权利要求的保护范围为准。
Claims (20)
- 一种基于物联网数据的用户数据分类方法,其中,包括:判断是否接收到物联网医疗终端上传的当前用户医疗数据;其中,所述当前用户医疗数据包括用户身份识别数据、药物注射数据、体征测量数据;若接收到物联网医疗终端上传的当前用户医疗数据,获取与所述身份识别数据对应的历史医疗数据集;其中,所述历史医疗数据集包括电子病历数据子集、检查报告数据子集、用药记录数据子集、历史治疗方案数据子集;调用预先设置的第一关键词集合及与所述第一关键词集合对应的第一提取策略,提取与所述身份识别数据对应的历史医疗数据集中的目标数据集合,以与所述当前用户医疗数据进行合并得到当前用户输入数据;由所述当前用户输入数据中各字段对应的语义向量组成当前用户数据输入向量,将所述当前用户数据输入向量输入至预先训练的卷积神经网络模型,得到与所述当前用户数据输入向量对应的分类结果;判断所述分类结果对应的治疗路径数据种类是否大于1;若所述分类结果对应的治疗路径数据种类大于1,获取分类结果对应的各治疗路径数据种类分别对应的标准用户数据输入向量;以及将分类结果分别对应的标准用户数据输入向量与所述当前用户数据输入向量计算向量相似度,获取向量相似度为最大值对应的标准用户数据输入向量、及对应的治疗路径数据,以作为最佳治疗路径数据。
- 根据权利要求1所述的基于物联网数据的用户数据分类方法,其中,所述获取与所述身份识别数据对应的历史医疗数据集之前,还包括:接收医疗***服务器上传的电子病历、检查报告、历史用药数据以及历史治疗方案;通过调用预先设置的第二关键词集合及与所述第二关键词集合对应的第二提取策略,提取所述电子病历中对应的电子病历数据初始子集;通过调用预先设置的第三关键词集合及与所述第三关键词集合对应的第三提取策略,提取所述检查报告中对应的检查报告数据初始子集;对所述历史用药数据对应的图片通过OCR文本识别模型进行文本识别,得到与所述历史用药数据对应的识别文本,通过预先设置的第四关键词集合及与所述第四关键词集合对应的第四提取策略,提取所述历史用药数据对应的用药记录数据初始子集;通过预先设置的第五关键词集合及与所述第五关键词集合对应的第五提取策略,提取所述历史治疗方案对应的历史治疗方案数据初始子集。
- 根据权利要求2所述的基于物联网数据的用户数据分类方法,其中,所述通过预先设置的第五关键词集合及与所述第五关键词集合对应的第五提取策略,提取所述历史治疗方案对应的历史治疗方案数据初始子集之后,还包括:将所述电子病历数据初始子集、检查报告数据初始子集、用药记录数据初始子集、历史治疗方案数据初始子集中各字段的取值进行关键词抽取,以得到与所述电子病历数据初始子集对应的电子病历数据子集、与所述检查报告数据初始子集对应的检查报告数据子集、与所述用药记录数据初始子集对应的用药记录数据子集、与所述历史治疗方案数据初始子集对应的历史治疗方案数据子集。
- 根据权利要求1所述的基于物联网数据的用户数据分类方法,其中,所述由所述当前用户输入数据中各字段对应的语义向量组成当前用户数据输入向量,将所述当前用户数据输入向量输入至预先训练的卷积神经网络模型,得到与所述当前用户数据输入向量对应的分类结果,包括:获取所述当前用户输入数据中各字段对应的语义向量,将每一语义向量进行求和,以得到与所述当前用户输入数据对应的当前用户数据输入向量;调用预先存储的卷积神经网络模型;将与所述当前用户输入数据对应的当前用户数据输入向量作为所述卷积神经网络模型的输入,得到对应的分类结果。
- 根据权利要求1所述的基于物联网数据的用户数据分类方法,其中,所述调用预先设置的第一关键词集合及与所述第一关键词集合对应的第一提取策略,提取与所述身份识别数据对应的历史医疗数据集中的目标数据集合,以与所述当前用户医疗数据进行合并得到当前用户输入数据,包括:调用所述第一关键词集合及与所述第一关键词集合对应的第一提取策略,在历史医疗数据集中获取与所述身份识别数据相对应且与第一关键词集合中各字段分别对应的目标数据集合;将所述目标数据集合中与所述当前用户医疗数据中重复字段进行数据组合并去重,得到当前用户输入数据。
- 根据权利要求1所述的基于物联网数据的用户数据分类方法,其中,所述将分类结果分别对应的标准用户数据输入向量与所述当前用户数据输入向量计算向量相似度,获取向量相似度为最大值对应的标准用户数据输入向量、及对应的治疗路径数据,以作为最佳治疗路径数据之后,还包括:通过调用预先设置的第六关键词集合及与所述第六关键词集合对应的第六提取策略,提取所述历史医疗数据集中各用户对应的目标画像数据集;根据各用户对应的目标画像数据集及预先存储的关键词筛选策略,将各用户对应的目标画像数据集中关键词进行筛选,得到各用户对应的目标画像数据简化集;调用预先存储的关键词转化策略,将各用户对应的目标画像数据简化集对应转化为用户画像数据。
- 根据权利要求1所述的基于物联网数据的用户数据分类方法,其中,所述第一关键词集合及与所述第一关键词集合对应的第一提取策略存储于区块链网络中。
- 根据权利要求7所述的基于物联网数据的用户数据分类方法,其中,所述第二关键词集合及与所述第二关键词集合对应的第二提取策略存储于区块链网络中;所述第三关键词集合及与所述第三关键词集合对应的第三提取策略存储于区块链网络中;所述第四关键词集合及与所述第四关键词集合对应的第四提取策略存储于区块链网络中;所述第五关键词集合及与所述第五关键词集合对应的第五提取策略存储于区块链网络中。
- 根据权利要求1所述的基于物联网数据的用户数据分类方法,其中,所述判断所述分类结果对应的治疗路径数据种类是否大于1之后,还包括:若所述分类结果对应的治疗路径数据种类等于1,则获取该治疗路径数据种类对应的治疗路径数据,以作为最佳治疗路径数据。
- 一种基于物联网数据的用户数据分类装置,其中,包括:当前数据接收单元,用于判断是否接收到物联网医疗终端上传的当前用户医疗数据;其中,所述当前用户医疗数据包括用户身份识别数据、药物注射数据、体征测量数据;历史数据搜索单元,用于若接收到物联网医疗终端上传的当前用户医疗数据,获取与所述身份识别数据对应的历史医疗数据集;其中,所述历史医疗数据集包括电子病历数据子集、检查报告数据子集、用药记录数据子集、历史治疗方案数据子集;当前用户输入数据获取单元,用于调用预先设置的第一关键词集合及与所述第一关键词集合对应的第一提取策略,提取与所述身份识别数据对应的历史医疗数据集中的目标数据集合,以与所述当前用户医疗数据进行合并得到当前用户输入数据;分类结果获取单元,用于由所述当前用户输入数据中各字段对应的语义向量组成当前用户数据输入向量,将所述当前用户数据输入向量输入至预先训练的卷积神经网络模型,得到与所述当前用户数据输入向量对应的分类结果;数据种类判断单元,用于判断所述分类结果对应的治疗路径数据种类是否大于1;标准向量获取单元,用于若所述分类结果对应的治疗路径数据种类大于1,获取分类结果对应的各治疗路径数据种类分别对应的标准用户数据输入向量;以及最佳路径数据获取单元,用于将分类结果分别对应的标准用户数据输入向量与所述当前用户数据输入向量计算向量相似度,获取向量相似度为最大值对应的标准用户数据输入向量、及对应的治疗路径数据,以作为最佳治疗路径数据。
- 一种计算机设备,包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序,其中,所述处理器执行所述计算机程序时实现以下步骤:判断是否接收到物联网医疗终端上传的当前用户医疗数据;其中,所述当前用户医疗数据包括用户身份识别数据、药物注射数据、体征测量数据;若接收到物联网医疗终端上传的当前用户医疗数据,获取与所述身份识别数据对应的历史医疗数据集;其中,所述历史医疗数据集包括电子病历数据子集、检查报告数据子集、用药记录数据子集、历史治疗方案数据子集;调用预先设置的第一关键词集合及与所述第一关键词集合对应的第一提取策略,提取与所述身份识别数据对应的历史医疗数据集中的目标数据集合,以与所述当前用户医疗数据进行合并得到当前用户输入数据;由所述当前用户输入数据中各字段对应的语义向量组成当前用户数据输入向量,将所述当前用户数据输入向量输入至预先训练的卷积神经网络模型,得到与所述当前用户数据输入向量对应的分类结果;判断所述分类结果对应的治疗路径数据种类是否大于1;若所述分类结果对应的治疗路径数据种类大于1,获取分类结果对应的各治疗路径数据种类分别对应的标准用户数据输入向量;以及将分类结果分别对应的标准用户数据输入向量与所述当前用户数据输入向量计算向量相似度,获取向量相似度为最大值对应的标准用户数据输入向量、及对应的治疗路径数据,以作为最佳治疗路径数据。
- 根据权利要求11所述的计算机设备,其中,所述获取与所述身份识别数据对应的历史医疗数据集之前,还包括:接收医疗***服务器上传的电子病历、检查报告、历史用药数据以及历史治疗方案;通过调用预先设置的第二关键词集合及与所述第二关键词集合对应的第二提取策略,提取所述电子病历中对应的电子病历数据初始子集;通过调用预先设置的第三关键词集合及与所述第三关键词集合对应的第三提取策略,提取所述检查报告中对应的检查报告数据初始子集;对所述历史用药数据对应的图片通过OCR文本识别模型进行文本识别,得到与所述历史用药数据对应的识别文本,通过预先设置的第四关键词集合及与所述第四关键词集合对应的第四提取策略,提取所述历史用药数据对应的用药记录数据初始子集;通过预先设置的第五关键词集合及与所述第五关键词集合对应的第五提取策略,提取所述历史治疗方案对应的历史治疗方案数据初始子集。
- 根据权利要求12所述的计算机设备,其中,所述通过预先设置的第五关键词集合及与所述第五关键词集合对应的第五提取策略,提取所述历史治疗方案对应的历史治疗方案数据初始子集之后,还包括:将所述电子病历数据初始子集、检查报告数据初始子集、用药记录数据初始子集、历史治疗方案数据初始子集中各字段的取值进行关键词抽取,以得到与所述电子病历数据初始子集对应的电子病历数据子集、与所述检查报告数据初始子集对应的检查报告数据子集、与所述用药记录数据初始子集对应的用药记录数据子集、与所述历史治疗方案数据初始子集对应的历史治疗方案数据子集。
- 根据权利要求11所述的计算机设备,其中,所述由所述当前用户输入数据中各字段对应的语义向量组成当前用户数据输入向量,将所述当前用户数据输入向量输入至预先训 练的卷积神经网络模型,得到与所述当前用户数据输入向量对应的分类结果,包括:获取所述当前用户输入数据中各字段对应的语义向量,将每一语义向量进行求和,以得到与所述当前用户输入数据对应的当前用户数据输入向量;调用预先存储的卷积神经网络模型;将与所述当前用户输入数据对应的当前用户数据输入向量作为所述卷积神经网络模型的输入,得到对应的分类结果。
- 根据权利要求11所述的计算机设备,其中,所述调用预先设置的第一关键词集合及与所述第一关键词集合对应的第一提取策略,提取与所述身份识别数据对应的历史医疗数据集中的目标数据集合,以与所述当前用户医疗数据进行合并得到当前用户输入数据,包括:调用所述第一关键词集合及与所述第一关键词集合对应的第一提取策略,在历史医疗数据集中获取与所述身份识别数据相对应且与第一关键词集合中各字段分别对应的目标数据集合;将所述目标数据集合中与所述当前用户医疗数据中重复字段进行数据组合并去重,得到当前用户输入数据。
- 根据权利要求11所述的计算机设备,其中,所述将分类结果分别对应的标准用户数据输入向量与所述当前用户数据输入向量计算向量相似度,获取向量相似度为最大值对应的标准用户数据输入向量、及对应的治疗路径数据,以作为最佳治疗路径数据之后,还包括:通过调用预先设置的第六关键词集合及与所述第六关键词集合对应的第六提取策略,提取所述历史医疗数据集中各用户对应的目标画像数据集;根据各用户对应的目标画像数据集及预先存储的关键词筛选策略,将各用户对应的目标画像数据集中关键词进行筛选,得到各用户对应的目标画像数据简化集;调用预先存储的关键词转化策略,将各用户对应的目标画像数据简化集对应转化为用户画像数据。
- 根据权利要求11所述的计算机设备,其中,所述第一关键词集合及与所述第一关键词集合对应的第一提取策略存储于区块链网络中。
- 根据权利要求17所述的计算机设备,其中,所述第二关键词集合及与所述第二关键词集合对应的第二提取策略存储于区块链网络中;所述第三关键词集合及与所述第三关键词集合对应的第三提取策略存储于区块链网络中;所述第四关键词集合及与所述第四关键词集合对应的第四提取策略存储于区块链网络中;所述第五关键词集合及与所述第五关键词集合对应的第五提取策略存储于区块链网络中。
- 根据权利要求11所述的计算机设备,其中,所述判断所述分类结果对应的治疗路径数据种类是否大于1之后,还包括:若所述分类结果对应的治疗路径数据种类等于1,则获取该治疗路径数据种类对应的治疗路径数据,以作为最佳治疗路径数据。
- 一种计算机可读存储介质,其中,所述计算机可读存储介质存储有计算机程序,所述计算机程序当被处理器执行时使所述处理器执行以下操作:判断是否接收到物联网医疗终端上传的当前用户医疗数据;其中,所述当前用户医疗数据包括用户身份识别数据、药物注射数据、体征测量数据;若接收到物联网医疗终端上传的当前用户医疗数据,获取与所述身份识别数据对应的历史医疗数据集;其中,所述历史医疗数据集包括电子病历数据子集、检查报告数据子集、用药记录数据子集、历史治疗方案数据子集;调用预先设置的第一关键词集合及与所述第一关键词集合对应的第一提取策略,提取与所述身份识别数据对应的历史医疗数据集中的目标数据集合,以与所述当前用户医疗数据进行合并得到当前用户输入数据;由所述当前用户输入数据中各字段对应的语义向量组成当前用户数据输入向量,将所述当前用户数据输入向量输入至预先训练的卷积神经网络模型,得到与所述当前用户数据输入 向量对应的分类结果;判断所述分类结果对应的治疗路径数据种类是否大于1;若所述分类结果对应的治疗路径数据种类大于1,获取分类结果对应的各治疗路径数据种类分别对应的标准用户数据输入向量;以及将分类结果分别对应的标准用户数据输入向量与所述当前用户数据输入向量计算向量相似度,获取向量相似度为最大值对应的标准用户数据输入向量、及对应的治疗路径数据,以作为最佳治疗路径数据。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010442300.0 | 2020-05-22 | ||
CN202010442300.0A CN111696675B (zh) | 2020-05-22 | 2020-05-22 | 基于物联网数据的用户数据分类方法、装置及计算机设备 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2021232591A1 true WO2021232591A1 (zh) | 2021-11-25 |
Family
ID=72477182
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2020/105413 WO2021232591A1 (zh) | 2020-05-22 | 2020-07-29 | 基于物联网数据的用户数据分类方法、装置、计算机设备及存储介质 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN111696675B (zh) |
WO (1) | WO2021232591A1 (zh) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114822859A (zh) * | 2022-03-31 | 2022-07-29 | 数魔方(北京)医药科技有限公司 | 治疗线程挖掘和检索方法及装置 |
CN114974564A (zh) * | 2022-05-17 | 2022-08-30 | 青岛百洋智能科技股份有限公司 | 一种规则驱动的肿瘤医疗辅助决策***的数据智能存储方法 |
CN115361424A (zh) * | 2022-10-21 | 2022-11-18 | 新亿成科技(江苏)有限公司 | 一种基于物联网数据分析的医疗设备监控方法和*** |
CN116386869A (zh) * | 2023-04-11 | 2023-07-04 | 中国科学技术大学 | 一种基于多变量的病情危重程度评估方法 |
CN116738216A (zh) * | 2023-08-15 | 2023-09-12 | 湖南警察学院 | 预警大数据的关联处理方法和装置 |
Families Citing this family (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112784720A (zh) * | 2021-01-13 | 2021-05-11 | 浙江诺诺网络科技有限公司 | 基于银行回单的关键信息提取方法、装置、设备及介质 |
CN113380358B (zh) * | 2021-06-01 | 2022-12-16 | 上海德衡数据科技有限公司 | 基于物联网的医疗信息交互的方法、装置及设备 |
CN113626429B (zh) * | 2021-07-26 | 2024-04-12 | 上海齐网网络科技有限公司 | 基于元数据的智能化范围急救医疗知识库构建方法及*** |
CN113643813B (zh) * | 2021-08-30 | 2024-07-09 | 平安医疗健康管理股份有限公司 | 基于人工智能的慢病随访监管方法、装置及计算机设备 |
CN113643818B (zh) * | 2021-09-16 | 2023-11-24 | 上海德衡数据科技有限公司 | 基于区域数据的医疗数据集成数据的方法及*** |
CN114257592A (zh) * | 2021-12-20 | 2022-03-29 | 东莞盟大集团有限公司 | 基于物联网的终端应用集成方法、装置、服务器及介质 |
CN114820225B (zh) | 2022-06-28 | 2022-09-13 | 成都秦川物联网科技股份有限公司 | 基于关键词识别和处理制造问题的工业物联网及控制方法 |
WO2024028631A1 (zh) * | 2022-08-01 | 2024-02-08 | Evyd科技有限公司 | 医疗服务套餐的推荐方法、装置、电子设备及计算机介质 |
CN115691789A (zh) * | 2022-11-15 | 2023-02-03 | 湖南春播万象科技有限公司 | 中医贴敷的数字化开方方法及*** |
CN116246749B (zh) * | 2023-05-11 | 2023-07-21 | 西南医科大学附属医院 | 集成电子病历的内分泌病人个性化健康管理*** |
CN117558460B (zh) * | 2024-01-11 | 2024-04-05 | 卓世未来(天津)科技有限公司 | 基于小样本学习和大语言模型的慢性病管理方法及*** |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109785928A (zh) * | 2018-12-25 | 2019-05-21 | 平安科技(深圳)有限公司 | 诊疗方案推荐方法、装置及存储介质 |
CN109817297A (zh) * | 2018-12-19 | 2019-05-28 | 平安科技(深圳)有限公司 | 医疗报告的生成方法、装置、计算机设备及计算机存储介质 |
US20190287684A1 (en) * | 2018-03-16 | 2019-09-19 | Vvc Holding Corporation | Medical system interface apparatus and methods to classify and provide medical data using artificial intelligence |
CN110880361A (zh) * | 2019-10-16 | 2020-03-13 | 平安科技(深圳)有限公司 | 一种个性化精准用药推荐方法及装置 |
WO2020086967A1 (en) * | 2018-10-25 | 2020-04-30 | Quadrant Biosciences Inc. | Methods and machine learning for disease diagnosis |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR102028664B1 (ko) * | 2017-12-26 | 2019-10-07 | (주)인더스웰 | 사물인터넷 기반 헬스케어 정보 전송장치 |
GB201802221D0 (en) * | 2018-02-12 | 2018-03-28 | Iota Medtech Pte Ltd | Integrative medical technology artificial intelligence platform |
CN110459313A (zh) * | 2018-05-08 | 2019-11-15 | 智森鸟(北京)科技有限责任公司 | 结合数据挖掘和智能学习算法的智能医疗云端*** |
CN109830303A (zh) * | 2019-02-01 | 2019-05-31 | 上海众恒信息产业股份有限公司 | 基于互联网一体化医疗平台的临床数据挖掘分析与辅助决策方法 |
-
2020
- 2020-05-22 CN CN202010442300.0A patent/CN111696675B/zh active Active
- 2020-07-29 WO PCT/CN2020/105413 patent/WO2021232591A1/zh active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190287684A1 (en) * | 2018-03-16 | 2019-09-19 | Vvc Holding Corporation | Medical system interface apparatus and methods to classify and provide medical data using artificial intelligence |
WO2020086967A1 (en) * | 2018-10-25 | 2020-04-30 | Quadrant Biosciences Inc. | Methods and machine learning for disease diagnosis |
CN109817297A (zh) * | 2018-12-19 | 2019-05-28 | 平安科技(深圳)有限公司 | 医疗报告的生成方法、装置、计算机设备及计算机存储介质 |
CN109785928A (zh) * | 2018-12-25 | 2019-05-21 | 平安科技(深圳)有限公司 | 诊疗方案推荐方法、装置及存储介质 |
CN110880361A (zh) * | 2019-10-16 | 2020-03-13 | 平安科技(深圳)有限公司 | 一种个性化精准用药推荐方法及装置 |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114822859A (zh) * | 2022-03-31 | 2022-07-29 | 数魔方(北京)医药科技有限公司 | 治疗线程挖掘和检索方法及装置 |
CN114822859B (zh) * | 2022-03-31 | 2023-11-03 | 数魔方(北京)医药科技有限公司 | 治疗线程挖掘和检索方法及装置 |
CN114974564A (zh) * | 2022-05-17 | 2022-08-30 | 青岛百洋智能科技股份有限公司 | 一种规则驱动的肿瘤医疗辅助决策***的数据智能存储方法 |
CN115361424A (zh) * | 2022-10-21 | 2022-11-18 | 新亿成科技(江苏)有限公司 | 一种基于物联网数据分析的医疗设备监控方法和*** |
CN116386869A (zh) * | 2023-04-11 | 2023-07-04 | 中国科学技术大学 | 一种基于多变量的病情危重程度评估方法 |
CN116386869B (zh) * | 2023-04-11 | 2024-01-19 | 中国科学技术大学 | 一种基于多变量的病情危重程度评估方法 |
CN116738216A (zh) * | 2023-08-15 | 2023-09-12 | 湖南警察学院 | 预警大数据的关联处理方法和装置 |
CN116738216B (zh) * | 2023-08-15 | 2023-12-26 | 湖南警察学院 | 预警大数据的关联处理方法和装置 |
Also Published As
Publication number | Publication date |
---|---|
CN111696675A (zh) | 2020-09-22 |
CN111696675B (zh) | 2023-09-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021232591A1 (zh) | 基于物联网数据的用户数据分类方法、装置、计算机设备及存储介质 | |
US11669791B2 (en) | Accession number correction system and methods for use therewith | |
CN109543863B (zh) | 一种就医任务管理方法、服务器及存储介质 | |
US20200388385A1 (en) | Efficient diagnosis confirmation of a suspect condition for certification and/or re-certification by a clinician | |
JP6907831B2 (ja) | コンテキストベースの患者類似性の方法及び装置 | |
CN108766561B (zh) | 病症信息处理方法、装置、计算机设备和存储介质 | |
US10319466B2 (en) | Intelligent filtering of health-related information | |
CN111710420A (zh) | 一种基于电子病历大数据的并发症发病风险预测方法、***、终端以及存储介质 | |
US11715569B2 (en) | Intent-based clustering of medical information | |
KR20110032637A (ko) | 혈당 관리 정보를 제공하는 방법 및 장치 | |
KR20120076564A (ko) | 환자 건강정보 기반의 환자대 환자관계 추천방법 | |
CN109801690A (zh) | 区域医疗电子病历共享整合查询***及方法 | |
US20230024631A1 (en) | Expression of clinical logic with positive and negative explainability | |
Zhang et al. | Computable eligibility criteria through ontology-driven data access: a case study of hepatitis C virus trials | |
CN112447270A (zh) | 一种用药推荐方法、装置、设备及存储介质 | |
US20140046694A1 (en) | Systems and methods for synoptic element structured reporting | |
CN111742372B (zh) | 用于检测医疗保健数据中的指示符的学习过滤器 | |
Lim | Future of health diagnostics | |
CN113436738A (zh) | 管理风险用户的方法、装置、设备及存储介质 | |
CN113436746B (zh) | 基于排序算法的用药推荐方法、装置、设备及存储介质 | |
CN111279424A (zh) | 用于优化图像采集工作流的设备、***和方法 | |
KR20070106942A (ko) | 의료 예상 진단 서비스 방법 | |
Tang et al. | Integrating medical code descriptions and building text classification models for diagnostic decision support | |
US20230395204A1 (en) | Survey and suggestion system | |
Taloyan et al. | Remote digital monitoring for selected chronic diseases in primary health care |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20936641 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 10/03/2023) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20936641 Country of ref document: EP Kind code of ref document: A1 |