CN108877928A - Patient information acquisition method, device, computer equipment and storage medium - Google Patents

Patient information acquisition method, device, computer equipment and storage medium Download PDF

Info

Publication number
CN108877928A
CN108877928A CN201810546481.4A CN201810546481A CN108877928A CN 108877928 A CN108877928 A CN 108877928A CN 201810546481 A CN201810546481 A CN 201810546481A CN 108877928 A CN108877928 A CN 108877928A
Authority
CN
China
Prior art keywords
data
paraphrase
tongue picture
template
node
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201810546481.4A
Other languages
Chinese (zh)
Other versions
CN108877928B (en
Inventor
励超磨
魏海彬
苟永亮
翁志龙
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Health Cloud Co Ltd
Ping An Healthcare Technology Co Ltd
Original Assignee
Ping An Health Cloud Co Ltd
Ping An Healthcare Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ping An Health Cloud Co Ltd, Ping An Healthcare Technology Co Ltd filed Critical Ping An Health Cloud Co Ltd
Priority to CN201810546481.4A priority Critical patent/CN108877928B/en
Publication of CN108877928A publication Critical patent/CN108877928A/en
Application granted granted Critical
Publication of CN108877928B publication Critical patent/CN108877928B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • 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
    • 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
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Biomedical Technology (AREA)
  • Data Mining & Analysis (AREA)
  • Epidemiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Databases & Information Systems (AREA)
  • Pathology (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

This application involves a kind of patient information acquisition method, device, computer equipment and storage mediums.The method includes：Tongue picture attribute is extracted from patient's tongue picture image；Obtain the tongue picture paraphrase with tongue picture attributes match；Symptom characteristic is extracted from patient main suit's data and tongue picture paraphrase；It searches and collects template with the matched information of symptom characteristic, load information collects the corresponding template data of template；Node acquisition problems are generated according to the node data in template data and are exported, and are obtained patient corresponding with node acquisition problems and are replied data；Data, which are replied, according to patient generates patient's acquisition information.The accuracy of patient information record can be capable of using this method.

Description

Patient information acquisition method, device, computer equipment and storage medium
Technical field
This application involves field of computer technology, set more particularly to a kind of patient information acquisition method, device, computer Standby and storage medium.
Background technique
Traditional Chinese medical doctor needs to carry out the four methods of diagnosis to patient when to patient's interrogation, and needs to inquire many bodies of patient The problems such as Signs shape, living habit.
Traditional Chinese medical doctor while to patient's interrogation, will also the information answered of result to the four methods of diagnosis, patient carry out it is whole Reason record.Therefore, traditional Chinese medical doctor is when recording, often only by virtue of experience to itself feel important key message into Row summary record, and due to the word speed of patient, pronunciation etc. are personal, lead to not it is accurate, record patient information comprehensively.
Summary of the invention
Based on this, it is necessary in view of the above technical problems, provide a kind of patient's letter for capableing of patient information record accuracy Cease acquisition method, device, computer equipment and storage medium.
A kind of patient information acquisition method, the method includes：
Obtain patient main suit's data and patient's tongue picture image；
Tongue picture attribute is extracted from patient's tongue picture image；
Obtain the tongue picture paraphrase with the tongue picture attributes match；
Symptom characteristic is extracted from patient main suit's data and the tongue picture paraphrase；
It searches and collects template with the matched information of the symptom characteristic, load the information and collect the corresponding template number of template According to；
Node acquisition problems are generated according to the node data in the template data and are exported, and are obtained and are acquired with the node The corresponding patient of problem replys data；
Data, which are replied, according to the patient generates patient's acquisition information.
Node acquisition problems and defeated are generated according to the node data in the template data in one of the embodiments, Out, it obtains patient corresponding with the node acquisition problems and replys data, including：
Obtain the start node and the corresponding start node data of the start node in the template data；
Starting acquisition problems are generated according to the start node data and are exported, and are obtained corresponding with the starting acquisition problems Starting patient reply data；
The corresponding connecting node of the start node is obtained from the template data, from the connecting node selection with The starting patient replys the corresponding first node of data, using the first node as present node, and continues according to The corresponding present node data of present node generate current acquisition problems and export, until collected current patents reply data Until corresponding present node is template frontier node.
The information is loaded in one of the embodiments, collects the corresponding template data of template, including：
Obtain the formwork structure type that the information collects template；
When the formwork structure type is nested template, the information is collected into the caster data in template from template Database is loaded onto local cache；
It is described to continue according to the current acquisition problems of the corresponding present node data generation of the present node and before exporting, Further include：
Judge whether the first node is that nested template jumps node；
When determining the first node is that the nested template jumps node, nesting belonging to the first node is obtained The corresponding subtemplate data of the nesting subtemplate are loaded onto local cache from database by subtemplate.
Subtemplate modification instruction and upgrading subtemplate data are received in one of the embodiments,；
The submodule panel sign in the subtemplate modification instruction is read, the corresponding son to be upgraded of the submodule panel sign is searched Template data；
The upgrading subtemplate data are compared with the subtemplate data to be upgraded and generate template change data；
Search the corresponding author's mark of the corresponding relation template of the submodule panel sign；
Data are changed according to the template and generate relation template upgrade tip, and the relation template upgrade tip is sent to The author identifies corresponding author's terminal.
The tongue picture paraphrase with the tongue picture attributes match is obtained in one of the embodiments, including：
The tongue picture attribute is inputted into default neural network classifier and obtains the first paraphrase matching probability of each tongue picture paraphrase；
The tongue picture attribute is inputted into default Bayes classifier and obtains each the second paraphrase of tongue picture paraphrase matching probability；
The paraphrase of each tongue picture paraphrase is obtained according to the first paraphrase matching probability and the second paraphrase matching probability With rate, the highest tongue picture paraphrase of paraphrase matching rate is extracted as the tongue picture paraphrase with the tongue picture attributes match.
After obtaining the tongue picture paraphrase with the tongue picture attributes match in one of the embodiments, further include：
The corresponding doctor data of online doctor is obtained, buty cycle data are extracted from the doctor data；
Idle doctor is filtered out from the online doctor according to the buty cycle data；
Patient's tongue picture image and the tongue picture paraphrase are sent to the corresponding doctor terminal of the idle doctor；
It is described to extract symptom characteristic from patient main suit's data and the tongue picture paraphrase, including：
When receiving the paraphrase acknowledgement notification corresponding with the tongue picture paraphrase that the doctor terminal returns, then from described Symptom characteristic is extracted in patient main suit's data and the tongue picture paraphrase.
Further include in one of the embodiments,：
When the paraphrase acknowledgement notification corresponding with the tongue picture paraphrase and amendment tongue picture for receiving the doctor terminal return When paraphrase, symptom characteristic is extracted from patient main suit's data and the amendment tongue picture paraphrase；
The tongue picture attribute is associated with the amendment tongue picture paraphrase and is added in tongue picture amendment sample set；
Neuron weight and the default shellfish according to the amendment sample set to the default neural network classifier The probability distribution of this classifier of leaf is adjusted.
A kind of patient information acquisition device, described device include：
Patient data obtains module, for obtaining patient main suit's data and patient's tongue picture image；
Tongue is as property extracting module, for extracting tongue picture attribute from patient's tongue picture image；
Paraphrase obtains module, for obtaining and the tongue picture paraphrase of the tongue picture attributes match；
Characteristic extracting module, for extracting symptom characteristic from patient main suit's data and the tongue picture paraphrase；
Template searching module collects template with the matched information of the symptom characteristic for searching, loads the information and receive Collect the corresponding template data of template；
Data acquisition module, for generating node acquisition problems according to the node data in the template data and exporting, It obtains patient corresponding with the node acquisition problems and replys data；
Information generating module generates patient's acquisition information for replying data according to the patient.
A kind of computer equipment, including memory and processor, the memory are stored with computer program, the processing The step of device realizes the above method when executing the computer program.
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor The step of above method is realized when row.
Above-mentioned patient information acquisition method, device, computer equipment and storage medium, when acquisition patient main suit's data and trouble After person's tongue picture image, image recognition is carried out to tongue picture image and attributes extraction, the tongue picture for the attributes match searched and extracted are released Justice, and matched information is searched according to the symptom characteristic for including in patient main suit's data and tongue picture paraphrase and collects template, Corresponding patient information is acquired according to the node data in template data, so as to realize the automatic paraphrase of tongue picture image, and The automation collection that patient information is realized according to information collection template, can obtain comprehensive and accurate patient information.
Detailed description of the invention
Fig. 1 is the application scenario diagram of patient information acquisition method in one embodiment；
Fig. 2 is the flow diagram of patient information acquisition method in one embodiment；
Fig. 3 is the flow diagram of paraphrase matching step in one embodiment；
Fig. 4 is the structural block diagram of patient information acquisition device in one embodiment；
Fig. 5 is the internal structure chart of computer equipment in one embodiment.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to paraphrase the application, not For limiting the application.
Patient information acquisition method provided by the present application, can be applied in application environment as shown in Figure 1.Wherein, eventually End is communicated by network with server.Terminal obtains patient main suit's data and patient's tongue picture image, from patient's tongue picture Tongue picture attribute is extracted in image, obtain with the tongue picture paraphrase of the tongue picture attributes match, from patient main suit's data and described Symptom characteristic is extracted in tongue picture paraphrase, searches and collects template with the matched information of the symptom characteristic, and terminal to server is sent Template data load request, server send template corresponding with information collection template to terminal according to template data load request Data, the node data in the terminal template data that server is sent based on the received, which generates, node acquisition problems and to be exported, and is obtained Patient corresponding with node acquisition problems is taken to reply data；Data, which are replied, according to patient generates patient's acquisition information.Wherein, terminal It can be, but not limited to be various personal computers, laptop, smart phone, tablet computer and portable wearable device, Server can be realized with the server cluster of the either multiple server compositions of independent server.
In one embodiment, as shown in Fig. 2, providing a kind of patient information acquisition method, it is applied to Fig. 1 in this way In terminal for be illustrated, include the following steps：
Step 210, patient main suit's data and patient's tongue picture image are obtained.
Patient main suit's data are patient to the description data of self health status, and patient main suit's data may include patient's Description data of the severity of physical condition, illnesses and extent, illness symptom and symptom etc., patient's tongue is as image For the image of patient's tongue.
Terminal can prompt user to input main suit's data, such as can prompt user " please be detailed by voice or text mode The problem of carefully describing you, including physical condition, disease and symptom etc. ", user can be inputted by voice or text to terminal Patient main suit's data, terminal obtain patient main suit's data of user's input.User can upload patient's tongue for prestoring as image, The image of patient's tongue can be directly acquired by the image capture device of terminal.
Step 220, tongue picture attribute is extracted from patient's tongue picture image.
Terminal to patient's tongue of acquisition as image carries out pretreatment operation, pretreatment operation may include to color of image into Row correction, tongue body part is extracted from image and carries out the segmentation of tongue body position, removes the operation such as shadow region of image.
Further, terminal is before carrying out pretreatment operation as image to tongue, and progress tongue is detected as picture quality, and root Judge whether tongue meets image quality requirements as image according to quality measurements.May include as image carries out quality testing to tongue Tongue is detected as image parameters data such as the resolution ratio, contrast, brightness of image, the image parameter data that will test are joined with corresponding Several preset standard parameter areas are compared, and judge whether the image parameter data detected are in preset standard parameter area It is interior, when in preset standard parameter area, the quality testing of image parameter data is passed through；Pre- bidding is not at when existing When image parameter in quasi- parameter area, fail to the quality testing of image parameter data, and it is underproof to generate picture quality Prompt, will be prompted to be shown, so that user uploads patient's tongue according to prompt as image again.
Tongue may include the attributes such as tongue color, coating colour, coating nature as attribute.Wherein, the specific attribute value of tongue color attribute can wrap It includes deep red red, livid purple, pale purple, withered white, light white etc.；The specific attribute value of coating colour attribute may include white tongue fur, yellow tongue fur, grayish fur and black tongue fur Deng；Coating nature may include thickness, moisturize, it is rough it is viscous, it is rotten it is greasy, peel off, have many aspects such as root unrooted, the specific attribute of coating nature attribute Value may include thick coating, thin tire, profit tire, greasy tire etc..Terminal is to pretreated patient's tongue as image carries out image analysis and spy Sign detection, extracts corresponding tongue as attribute according to feature testing result.
Step 230, the tongue picture paraphrase with tongue picture attributes match is obtained.
A variety of tongues are previously stored in terminal as paraphrase, tongue picture is interpreted as the parsing result to tongue as image, for example, " tongue Red, while there is indentation, tongue fur thickness is greasy ".The tongue stored in terminal carries out attribute tags mark, the tongue that terminal will extract as paraphrase in advance As attribute with each tongue as the attribute tags of paraphrase are matched, find out attribute tags and tongue as the completely the same tongue picture of attribute is released Justice.
Step 240, symptom characteristic is extracted from patient main suit's data and tongue picture paraphrase.
Symptom characteristic is used to indicate the physical condition feature of patient, may include such as cough of Symptoms feature, abundant expectoration, goes out Cold sweat etc., symptom degree feature such as has an intense pain, slight flatulence, and physical trait is such as fat, thin, and tongue is as attributive character such as tongue It is red, tongue nature is light, tongue fur is thick greasy etc..
Terminal obtains symptom characteristic word list, includes curing in symptom characteristic list to multiple clinical case data and traditional Chinese and western medicine The symptom characteristic word summed up in dictionary is learned, the near synonym of each symptom characteristic word, which are also associated with, to be added in symptom characteristic word list. Terminal is from patient main suit's data and tongue as extracting the disease to match with the symptom characteristic word in symptom characteristic word list in paraphrase Shape feature.
Step 250, it searches and collects template with the matched information of symptom characteristic, load information collects the corresponding template number of template According to.
It is to be formulated for certain class disease or symptom for carrying out information collection to patient health status that information, which collects template, Template.Information collects template and carries out symptom characteristic label in advance, and symptom characteristic label is according to the disease in symptom characteristic word list Shape Feature Words are set.One information collects template and can be marked one or more symptom characteristic labels, it can to one The patient health information of kind or a variety of diseases for belonging to identical classification or symptom is acquired.The symptom characteristic that terminal will extract The symptom characteristic label that template is marked is collected with each information to be matched, and is mentioned if the symptom characteristic label being marked completely includes Then successful match when the symptom characteristic of taking-up.The information that terminal obtains successful match collects the template identification of template, template identification It is identified for uniquely collecting template to information, terminal searches and loads the corresponding template data of template identification.As terminal mentions The symptom characteristic of taking-up includes that phlegm wet, the deficiency of vital energy, obesity and tongue nature are light, information corresponding with symptom characteristic that is extracting that is being matched to Collecting template is " adult general Spleen-Qi Deficiency constitution tongue nature is light ".
Step 260, node acquisition problems are generated according to the node data in template data and exported, obtained and acquired with node The corresponding patient of problem replys data.
Information is collected in template between node data and multiple information collection nodes of the data containing multiple information collection nodes Connection relationship, each information collection node is used to be acquired the adjoint performance information of a certain health characteristics of patient.It is more It is connected with each other between a information collection node, certain information collection nodes positioned at template upstream may be located under template with multiple Trip information collection node is attached, and the collected patient information of upstream information collector node institute determines that downstream information collects section The route selection of point.It therefore, can be with structure according to the collected specific patient information of each information collection node in information collection template At a plurality of information collection path.
The node data of each information collection node can include but is not limited to nodename, acquisition problems title, acquisition The data such as problem types, acquisition problems content, node condition of contact relationship.Wherein, acquisition problems type may include individual event, it is more Item selection type problem, or the types such as open-ended question.Terminal is raw according to the acquisition problems content in present node data It is shown at current acquisition problems, and by the current acquisition problems of generation, such as can be a variety of by text, picture or voice Form is shown.It, can be by inputting problem answers to terminal after user receives current acquisition problems, terminal obtains user The answer of the problem of input generates patient according to problem answers and replys data.
Step 270, data are replied according to patient and generates patient's acquisition information.
Terminal replys data to the nodal information of each node passed through in template execution route and corresponding collected patient It carries out Data Integration, data processing and generates patient's acquisition information.
In above-mentioned patient information acquisition method, after terminal obtains patient main suit's data and patient's tongue picture image, to tongue picture Image progress image recognition and attributes extraction, the tongue picture paraphrase for the attributes match searched and extracted, and according to patient main suit's number Template is collected according to matched information is searched with the symptom characteristic for including in tongue picture paraphrase, according to the node in template data Data acquire corresponding patient information, so as to realize the automatic paraphrase of tongue picture image, and are realized according to information collection template The automation collection of patient information can obtain comprehensive and accurate patient information.
In one embodiment, generated and node acquisition problems and exported according to the node data in template data, obtain with The corresponding patient of node acquisition problems replys data：The start node and start node obtained in template data corresponds to Start node data；Starting acquisition problems are generated according to start node data and are exported, and are obtained corresponding with starting acquisition problems Starting patient reply data；The corresponding connecting node of start node is obtained from template data, from connecting node selection with It originates patient and replys the corresponding first node of data, using first node as present node, and continue corresponding according to present node Present node data generate current acquisition problems and export, until data are corresponding works as prosthomere for collected starting patient reply Until point is template frontier node.
Start node is the first information collection node in information collection template, and terminal obtains the corresponding node of start node Data, terminal extracts egress acquisition problems data from node data, and generates acquisition according to node acquisition problems data and ask Topic exports the acquisition problems of generation.
For example, the nodename of an information collection node is " whether night sweat ", it is individual event selection type problem, corresponding problem Content may include that " you are frequent night sweat to problem title？", problem option is " making " and "no".Node condition of contact relationship packet Node alternative condition decision logic is included, decision logic is for judging whether the acquisition data of upstream connecting node meet the node Selection is adjusted, and the node is executed if meeting, if being unsatisfactory for refusing to execute.If nodename is the upper of " sleep dreaminess situation " Swimming node is " whether night sweat ", and " sleep dreaminess situation " node is just met when information collected " whether night sweat " is "Yes" Alternative condition decision logic.
It, can be by inputting problem answers to terminal after user receives acquisition problems, terminal obtains asking for user's input Answer is inscribed, patient is generated according to problem answers and replys data.When problem answers are selection option, directly user can be selected Answer as patient reply data, when problem answers be user open answer when, terminal to problem answers carry out information It extracts or information conversion, specific extraction or transformation rule is configured according to the particular problem of node in advance in advance, terminal will The information of extraction replys data as patient.If the collected problem answers of " sleeping time " node are " sleep 6 hours ", then eventually Digital " 6 " therein are extracted as patient and reply data by end.
Terminal obtains the connection section positioned at start node downstream for having connection relationship with start node from template data Point judges that the patient of collected start node replys the node alternative condition whether data meet connecting node, saves from connection The first node for meeting node alternative condition is found out in point, and using first node as present node, and continue cycling through execution Current acquisition problems are generated according to the corresponding present node data of present node and are exported, and acquisition obtains and current acquisition problems pair The patient answered replys data step, until it is that template frontier node is that collected patient, which replys the corresponding present node of data, Only, template frontier node is that information collects the information collection node that downstream connecting node is not present in template.
In one embodiment, the corresponding template data of load information collection template may include：It obtains information and collects mould The formwork structure type of plate；When formwork structure type is nested template, information is collected into the caster data in template from mould Plate database is loaded onto local cache；Continue to continue to generate current acquisition problems according to the corresponding present node data of present node And can also include before exporting：Judge whether first node is that nested template jumps node；When judgement first node is nesting When template jumps node, nested subtemplate belonging to first node is obtained, by the corresponding subtemplate data of nested subtemplate from number Local cache is loaded onto according to library.
The formwork structure type that information collects template includes independent template and nested template two major classes, wherein independent template by An individual template is constituted, and nested template is connected with each other by caster with multiple nested subtemplates or nesting forms. Wherein, nested subtemplate can be considered as a node in caster, and name master mold using the template name of nested subtemplate Corresponding node in plate, and the connecting node between different templates is that nested template jumps node.
Terminal obtains the formwork structure type that the information that finds collects template, when judging formwork structure type for nesting When template, from the caster data in the server where template database in load information collection template, caster data The mark datas such as the corresponding nodename of nested subtemplate are only included, the specific template number not comprising specific nested subtemplate According to.
When terminal carries out information collection according to nested template, returned when continuing acquisition patient corresponding with the present node Before complex data, judge whether first node is that nested template jumps node, is jumped when determining first node for nested template When node, terminal obtains the first node mark of first node, the corresponding nested subtemplate of first node mark is searched, from template The corresponding subtemplate data of the nested subtemplate found are loaded into local cache in server where database.
In one embodiment, terminal is before load caster data with nested subtemplate data, first from local cache In search whether to be present in master mold panel sign or the corresponding template data of nested submodule panel sign, when it is present then without from number According to being loaded in library, loaded from template database again when it be not present.Template data is loaded in terminal timing statistics local cache Execution frequency, frequency will be executed and be removed from the cache lower than the template data of predeterminated frequency, local redundant data is removed.
The patient information that some diseases or symptom need to acquire is very more, and corresponding information collects the node that template includes Quantity and the data volume of node data are very huge, and some information collects template may be comprising several hundred a nodes of tens layers, several The case where ten sub- template nestings, need to occupy a large amount of storages if all template datas are loaded onto local cache Space simultaneously influences treatment effeciency, and the template number of several subtemplates may be only carried out according to the selection path that user replys data According to leading to data redundancy and waste of storage space.Therefore, in the present embodiment, only when jumping to related nested subtemplate, then The data of subtemplate are loaded into local cache from database, it is possible to reduce data redundancy improves treatment effeciency.
In one embodiment, patient information acquisition method can also include：Receive subtemplate modification instruction and upgrading Template data；The submodule panel sign in subtemplate modification instruction is read, the corresponding subtemplate number to be upgraded of submodule panel sign is searched According to；Upgrading subtemplate data are compared with subtemplate data to be upgraded and generate template change data；Search submodule panel sign The corresponding author's mark of corresponding relation template；Data are changed according to template and generate relation template upgrade tip, by relation template Upgrade tip is sent to author and identifies corresponding author's terminal.
User can be modified by the template data that terminal collects template to each information, be modified to template data It modifies including node data to each node such as nodename, node problems, node title etc., or carries out node Delete, the operation such as addition, can also pair and the nodal community such as node connection relationship of node modify, can also be to other moulds Plate data are modified.Subtemplate modification is generated when template data of the user to nested subtemplate is modified and confirmed to refer to It enables, the template identification of the nested subtemplate of modification is carried in subtemplate modification instruction.
Terminal receives subtemplate modification instruction, and obtains user's upgrading subtemplate data modified to sub- template data. Terminal reads the submodule panel sign in subtemplate modification instruction, searches subtemplate data to be upgraded corresponding with submodule panel sign, Subtemplate data to be upgraded are the template data for upgrading the corresponding initial subtemplate of subtemplate.Terminal will upgrade subtemplate data with Subtemplate data to be upgraded are compared, and find out the difference number between upgrading subtemplate data and subtemplate data to be upgraded According to according to variance data generation template change data.
Terminal searches the corresponding relation template of submodule panel sign, and relation template is in template data comprising submodule panel sign pair The information for the subtemplate to be upgraded answered collects template, i.e., subtemplate to be upgraded is nested in relation template.Terminal acquisition is found Relation template corresponding author mark, user identity of author's mark for the user to creation relation template be identified. The quantity for the corresponding relation template of submodule panel sign that terminal is found can be one or more, therefore, corresponding author's mark The quantity of knowledge may be one or more.
Terminal changes data according to template and generates relation template upgrade tip, and terminal can be in relation template upgrade tip The link of template change data is added, user can check the change conditions of subtemplate data by clickthrough.Relation template Prompt term in upgrade tip can be for " XX subtemplate is to upgrade, if needs to upgrade father's template？" etc..Terminal will The relation template upgrade tip of generation is sent to author and identifies corresponding author's terminal, so that the creation user of relation template determines Whether relation template is upgraded.
In one embodiment, as shown in figure 3, the paraphrase matching step of acquisition and the tongue picture paraphrase of tongue picture attributes match can To include：
Step 232, tongue picture attribute is inputted into default neural network classifier and obtains the first paraphrase matching of each tongue picture paraphrase Probability.
Collect the corresponding all values that may be present of each tongue picture attribute in server in advance, and by all possible value Assign a binaryzation vector.Server searches the corresponding binaryzation vector of tongue picture attribute extracted, according to all acquisitions Binaryzation vector generates combination of eigenvectors, to convert digital vectors from text data for tongue picture attribute.
As tongue picture attribute be tongue color attribute when, the red corresponding feature vector of tongue be 00000010, the corresponding feature of blue-purplish tongue Vector is 00000001, and when tongue picture attribute is coating colour attribute, the corresponding feature vector of grayish fur is 00001001, the corresponding spy of black tongue fur Levying vector is 00001010 etc..
First paraphrase matching probability is according to the default calculated feature vector of neural network classifier and all default tongue pictures The matching probability of paraphrase.Default neural network classifier is the pre- neural network classification model for first passing through great amount of samples training, mind It is used to carry out tongue picture attribute the classification of tongue picture paraphrase through network class model.
It specifically, include the tongue picture attribute pair of each interrogation case in the sample data for training neural network classification model The feature vector and the corresponding correct tongue picture paraphrase to patient's tongue picture image answered, tongue picture paraphrase can be the two of tongue picture paraphrase Value vector coding etc..Default neural network classifier is point with the corresponding combination of eigenvectors of tongue picture attribute that terminal extracts The input of class device, exports the matching probability of the tongue picture combinations of attributes for all preset tongue picture paraphrase and input, and matching probability is used In the tongue picture image of reflection patient and the matching degree of preset each tongue picture paraphrase.
In one embodiment, the generation method of default neural network classifier may include：Construct initial neural network Model acquires history diagnosis and treatment sample data, and tongue picture attribute and corresponding tongue picture paraphrase are extracted from history diagnosis and treatment sample data；It will Tongue picture attribute is set as the input data of initial neural network mould, sets initial neural network model output layer for tongue picture paraphrase Target data after, tune is trained to the initial weighting weight between each neuron in initial neural network model hidden layer It is whole, obtain optimal weighting weight；Default neural network classifier is generated according to initial neural network model and optimal weighting weight.
The initial neural network model of building has input layer, output layer and multiple hidden layers to constitute.Hidden layer is input layer The every aspect of numerous neurons and link composition between output layer, the every aspect in hidden layer may include multiple convolution Layer, pond layer, articulamentum and dropout layers etc., in hidden layer the type of the activation primitive of the number of plies and use of various levels and Initial weighting weight between each neuron can rule of thumb be set by staff, wherein be used in hidden layer Activation primitive may include sigmoid function, ReLu function or tanh function etc..
Step 234, tongue picture attribute is inputted into default Bayes classifier and obtains each the second paraphrase of tongue picture paraphrase matching probability.
Second paraphrase matching probability is released according to the default calculated feature vector of Bayes classifier with all default tongue pictures The matching probability of justice.Default Bayes classifier is the pre- Bayesian Classification Model for first passing through great amount of samples training, Bayes point Class model is used to the tongue picture image of patient carrying out tongue picture paraphrase classification.
Specifically, the tongue picture attribute in the sample data for training Bayesian Classification Model including each interrogation case is corresponding Feature vector and the corresponding correct tongue picture paraphrase to patient's tongue picture image, tongue picture paraphrase can be tongue picture paraphrase two-value Change vector coding etc..Default neural network classifier is classification with the corresponding combination of eigenvectors of tongue picture attribute that terminal extracts The input of device, exports the matching probability of the combination of eigenvectors for all preset tongue picture paraphrase and input, and matching probability is used for Reflect the tongue picture image of patient and the matching degree of preset each tongue picture paraphrase.
In one embodiment, the generation method of default Bayes classifier may include：Construct initial Bayes's classification Model acquires history diagnosis and treatment sample data, and tongue picture attribute and corresponding tongue picture paraphrase are extracted from history diagnosis and treatment sample data；From Tongue picture attribute is extracted in patient's diagnosis and treatment data and attribute feature vector combination is generated according to tongue picture attribute；It is counted according to tongue picture paraphrase The tongue picture paraphrase probability distribution of each attribute feature vector combination out；According to initial Bayesian Classification Model and tongue picture paraphrase probability point Cloth generates default Bayes classifier.In the present embodiment, initial pattra leaves can be constructed using naive Bayesian probabilistic model This disaggregated model, such as multinomial Naive Bayes Classification Model, Bernoulli Jacob's Naive Bayes Classification Model can be constructed.
Step 236, the paraphrase of each tongue picture paraphrase is obtained according to the first paraphrase matching probability and the second paraphrase matching probability With rate, the highest tongue picture paraphrase of paraphrase matching rate is extracted as the tongue picture paraphrase with tongue picture attributes match.
According to the result of the matching probability of the feature vector of the calculated each tongue picture paraphrase of different model classifiers and input Between there may be difference, terminal will be according to the first paraphrase for presetting each tongue picture paraphrase that neural network classifier be calculated With probability, comprehensive point is carried out with according to the second paraphrase matching probability for presetting each tongue picture paraphrase that Bayes classifier is calculated Analysis, obtains the comprehensive matching probability of each tongue picture paraphrase, and filter out the highest comprehensive matching probability pair of value based on the analysis results The tongue picture paraphrase answered.
In one embodiment, each tongue picture paraphrase is obtained according to the first paraphrase matching probability and the second paraphrase matching probability Paraphrase matching rate may include：Obtain corresponding first initial weight of the first paraphrase matching probability, the second paraphrase matching probability pair The second initial weight answered；According to the first paraphrase matching probability and corresponding first initial weight, the second paraphrase matching probability and Corresponding second initial weight calculates paraphrase weighted registration probability.
Terminal presets shellfish to according to the default calculated first paraphrase matching probability of neural network model, basis respectively in advance This classifier calculated of leaf go out the second paraphrase matching probability initial probability right is set, be respectively set to the first initial weight and Second initial weight.First initial weight and the second initial weight are according to default neural network classifier to historical sample data Classification results accuracy rate and the classification results classified to historical sample data of default Bayes classifier it is accurate What rate was set.Specifically, wherein the accuracy rate of classification results can different classifications device obtains according to all sample numbers According to accuracy rate statistical result, at this moment can set identical value for corresponding first initial weight of all tongue picture paraphrase, will Corresponding second initial weight of all tongue picture paraphrase is also configured as identical value, the sum of the first initial weight and the second initial weight It is 1.
The corresponding historical sample data of each tongue picture paraphrase is divided respectively it is possible to further count two classifiers The first initial weight of each paraphrase is set separately according to the accuracy rate for each paraphrase being calculated for the accuracy rate of the classification results of class Value with the second initial weight, i.e., the first initial weight of different paraphrase settings may be different, the second initial weight of setting Value may also be different, but the sum of corresponding first initial weight of same paraphrase and the second initial weight is 1.
In one embodiment, can also include after obtaining the tongue picture paraphrase with tongue picture attributes match：It obtains online The corresponding doctor data of doctor extracts buty cycle data from doctor data；It is screened from online doctor according to buty cycle data Idle doctor out；Patient's tongue picture image and tongue picture paraphrase are sent to the corresponding doctor terminal of idle doctor；From patient main suit's number According to extract symptom characteristic in tongue picture paraphrase, including：When the paraphrase corresponding with tongue picture paraphrase for receiving doctor terminal return is true When recognizing notice, then from patient main suit's data and tongue picture paraphrase extract symptom characteristic.
Terminal obtains corresponding doctor's number of online doctor at present after obtaining the tongue picture paraphrase with tongue picture attributes match According to.Online doctor is the doctor that can be provided in line remote interrogation at present, and doctor data may include the doctor of the affiliated hospital of doctor Raw information, the affiliated department's information of doctor, doctor hold the data such as doctor's posterior infromation, current buty cycle information.
Terminal extracts buty cycle data from the doctor data of acquisition, and buty cycle data are to be able to reflect to work as examining doctor The data of preceding interrogation busy degree can specifically include the number of patient of being lined up at present, doctor to average interrogation time of patient Etc. data, terminal can according to examine doctor it is corresponding before be lined up average interrogation time of patient numbers and doctor and assess and examine The estimated waiting time of doctor, and according to respectively corresponding buty cycle is calculated in the estimated waiting time for examining doctor, generally, in advance It is higher to count waiting time longer corresponding buty cycle, it is contemplated that waiting time, lower corresponding buty cycle was lower.Buty cycle is high Indicate that doctor is relatively busy, the low expression doctor of buty cycle is notr busy.
Terminal is examining doctor's screening as idle doctor for buty cycle is minimum, and obtains the corresponding doctor terminal of idle doctor Mark, doctor terminal mark for examining terminal used in doctor carry out unique identification.Terminal is by tongue picture image and tongue picture Paraphrase is sent to the corresponding doctor terminal of idle doctor together.Doctor terminal it is corresponding examine doctor can be to receiving tongue picture figure As checking analysis, the tongue picture image for judging that whether tongue picture paraphrase uploads with user is consistent, when doctor judges tongue picture paraphrase and tongue picture When image is consistent, doctor can carry out tongue picture paraphrase confirmation by doctor terminal, and doctor terminal detects doctor to tongue picture paraphrase Confirmation operation when, generate paraphrase acknowledgement notification simultaneously return to terminal.The paraphrase confirmation that terminal receives doctor terminal return is logical After knowing, it is further continued for executing the step of symptom characteristic is extracted from patient main suit's data and tongue picture paraphrase.
In one embodiment, terminal can set physician feedback time threshold, when according to current time and physician feedback Between threshold value can calculate the estimated feedback time of doctor, sent when reaching the estimated feedback time of doctor and not receiving doctor terminal also Paraphrase feedback when, production paraphrase confirmation, which is reminded, is simultaneously sent to doctor terminal, to prompt doctor preferentially to carry out paraphrase confirmation task Processing.
In the present embodiment, it after terminal finds matched tongue picture paraphrase, is sent to is examining doctor's progress really first Recognize, executes following intelligent interrogation operation after doctor's confirmation again, avoid paraphrase matching error from causing to acquire invalid information, cause Patient wastes time.
In one embodiment, patient information acquisition method can also include：When receive doctor terminal return and tongue When as the corresponding paraphrase acknowledgement notification of paraphrase and amendment tongue picture paraphrase, disease is extracted from patient main suit's data and amendment tongue picture paraphrase Shape feature；Tongue picture attribute is associated with amendment tongue picture paraphrase and is added in tongue picture amendment sample set；According to amendment sample set to pre- If the neuron weight of neural network classifier and the probability distribution of default Bayes classifier are adjusted.
When doctor judges that tongue picture paraphrase is not consistent with tongue picture image, doctor can be by doctor terminal to received tongue picture Paraphrase is modified operation, is after tongue picture paraphrase confirmation after doctor is to the modification of input, doctor terminal detects doctor's After modification operation, obtains the modified amendment tongue picture paraphrase of doctor and generate paraphrase acknowledgement notification, by paraphrase acknowledgement notification and repair Positive tongue picture paraphrase returns to terminal.
After terminal receives paraphrase acknowledgement notification and amendment tongue picture paraphrase, from patient main suit's data and amendment tongue picture paraphrase Symptom characteristic is extracted, and continues to execute the step of lookup collects template with the matched information of the symptom characteristic.
Terminal will be corrected after tongue picture paraphrase and tongue picture attribute be associated, and amendment tongue picture paraphrase is added to tongue picture amendment sample This concentration.Terminal periodically extracts tongue picture from tongue picture amendment sample set and corrects sample data, mentions from tongue picture amendment sample data Revised tongue picture paraphrase after taking preset quantity tongue picture attribute and corresponding to, using the tongue picture attribute extracted as default nerve net The input data of network classifier, target output of the revised tongue picture paraphrase as default neural network classifier are constantly right The neuron weight of neuron in the hidden layer of default neural network classifier is adjusted and optimizes.Terminal is repaired according to tongue picture Positive sample data and historical sample data count of the combination of eigenvectors of each preset tongue picture paraphrase and tongue picture attribute again It is adjusted with probability, and to the probability distribution of default Bayes classifier.
It should be understood that although each step in the flow chart of Fig. 2-3 is successively shown according to the instruction of arrow, These steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly stating otherwise herein, these steps Execution there is no stringent sequences to limit, these steps can execute in other order.Moreover, at least one in Fig. 2-3 Part steps may include that perhaps these sub-steps of multiple stages or stage are not necessarily in synchronization to multiple sub-steps Completion is executed, but can be executed at different times, the execution sequence in these sub-steps or stage is also not necessarily successively It carries out, but can be at least part of the sub-step or stage of other steps or other steps in turn or alternately It executes.
In one embodiment, as shown in figure 4, providing a kind of patient information acquisition device, including：Patient data obtains Module 410, tongue as property extracting module 420, paraphrase obtain module 430, characteristic extracting module 440, template searching module 450, Data acquisition module 460 and information generating module 470, wherein：
Patient data obtains module 410, for obtaining patient main suit's data and patient's tongue picture image.
Tongue is as property extracting module 420, for extracting tongue picture attribute from patient's tongue picture image.
Paraphrase obtains module 430, for obtaining and the tongue picture paraphrase of tongue picture attributes match.
Characteristic extracting module 440, for extracting symptom characteristic from patient main suit's data and tongue picture paraphrase.
Template searching module 450 collects template with the matched information of symptom characteristic for searching, and load information collects template Corresponding template data；.
Data acquisition module 460 is obtained for generating node acquisition problems according to the node data in template data and exporting Patient corresponding with node acquisition problems is taken to reply data.
Information generating module 470 generates patient's acquisition information for replying data according to patient.
In one embodiment, data acquisition module 460 may include：
Present node extraction module, it is corresponding for obtaining start node in the template data and the start node Start node data.
Data acquisition module is replied, for generating starting acquisition problems according to the start node data and exporting, is obtained Starting patient corresponding with the starting acquisition problems replys data.
Template execution module, for obtaining the corresponding connecting node of the start node from the template data, from institute Selection first node corresponding with starting patient reply data is stated in connecting node, using the first node as working as prosthomere Point, and continue to generate current acquisition problems according to the corresponding present node data of the present node and export, until collecting Current patents reply until the corresponding present node of data is template frontier node.
In one embodiment, template searching module 450 may include：
Structure type obtains module, and the formwork structure type of template is collected for obtaining information.
Caster loading module, for when formwork structure type is nested template, information to be collected the master mold in template Plate data are loaded onto local cache from template database.
Patient information acquisition device can also include：
Node judgment module is jumped, for judging whether first node is that nested template jumps node.
Subtemplate loading module, for obtaining first node institute when determining that first node jumps node for nested template The corresponding subtemplate data of nested subtemplate are loaded onto local cache from database by the nested subtemplate of category.
In one embodiment, patient information acquisition device can also include：
Command reception module is modified, for receiving subtemplate modification instruction and upgrading subtemplate data.
Subtemplate data search module searches subtemplate mark for reading the submodule panel sign in subtemplate modification instruction Know corresponding subtemplate data to be upgraded.
Data generation module is changed, is compared generation mould with subtemplate data to be upgraded for subtemplate data will to be upgraded Plate changes data.
Author's identifier lookup module, for searching the corresponding author's mark of the corresponding relation template of submodule panel sign.
Upgrade tip generation module generates relation template upgrade tip for changing data according to template, by relation template Upgrade tip is sent to author and identifies corresponding author's terminal.
In one embodiment, paraphrase acquisition module 430 may include：
First probability obtains module, obtains each tongue picture paraphrase for tongue picture attribute to be inputted default neural network classifier First paraphrase matching probability.
Second probability obtains module, is released with the default Bayes classifier of tongue picture attribute input is obtained each tongue picture paraphrase second Adopted matching probability.
Paraphrase extraction module, for obtaining each tongue picture paraphrase according to the first paraphrase matching probability and the second paraphrase matching probability Paraphrase matching rate, extract the highest tongue picture paraphrase of paraphrase matching rate as the tongue picture paraphrase with tongue picture attributes match.
In one embodiment, patient information acquisition device can also include：
Doctor data obtains module and extracts busy from doctor data for obtaining the corresponding doctor data of online doctor Degree evidence.
Idle doctor's screening module, for filtering out idle doctor from online doctor according to buty cycle data.
Tongue is as data transmission blocks, for patient's tongue picture image and tongue picture paraphrase to be sent to the corresponding doctor of idle doctor Terminal.
Characteristic extracting module 440 is also used to when paraphrase corresponding with the tongue picture paraphrase confirmation for receiving doctor terminal return is logical When knowing, then from patient main suit's data and tongue picture paraphrase extract symptom characteristic.
In one embodiment, patient information acquisition device can also include：
Paraphrase receiving module is corrected, for leading to when paraphrase corresponding with the tongue picture paraphrase confirmation for receiving doctor terminal return When knowing and correcting tongue picture paraphrase, symptom characteristic is extracted from patient main suit's data and amendment tongue picture paraphrase.
Tongue is added to tongue picture amendment sample set as data adding module, for tongue picture attribute to be associated with amendment tongue picture paraphrase In.
Classifier optimization module, for according to amendment sample set to the neuron weight of default neural network classifier and The probability distribution of default Bayes classifier is adjusted.
Specific about patient information acquisition device limits the limit that may refer to above for patient information acquisition method Fixed, details are not described herein.Modules in above-mentioned patient information acquisition device can fully or partially through software, hardware and its Combination is to realize.Above-mentioned each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also be with It is stored in the memory in computer equipment in a software form, in order to which processor calls the above modules of execution corresponding Operation.
In one embodiment, a kind of computer equipment is provided, which can be terminal, internal structure Figure can be as shown in Figure 5.The computer equipment includes processor, the memory, network interface, display connected by system bus Screen and input unit.Wherein, the processor of the computer equipment is for providing calculating and control ability.The computer equipment is deposited Reservoir includes non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system and computer journey Sequence.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating The network interface of machine equipment is used to communicate with external terminal by network connection.When the computer program is executed by processor with Realize a kind of patient information acquisition method.The display screen of the computer equipment can be liquid crystal display or electric ink is shown Screen, the input unit of the computer equipment can be the touch layer covered on display screen, be also possible on computer equipment shell Key, trace ball or the Trackpad of setting can also be external keyboard, Trackpad or mouse etc..
It will be understood by those skilled in the art that structure shown in Fig. 5, only part relevant to application scheme is tied The block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment It may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.
In one embodiment, a kind of computer equipment, including memory and processor are provided, which is stored with Computer program, the processor realize following steps when executing computer program：Obtain patient main suit's data and patient's tongue picture figure Picture；Tongue picture attribute is extracted from patient's tongue picture image；Obtain the tongue picture paraphrase with tongue picture attributes match；From patient main suit's data and Symptom characteristic is extracted in tongue picture paraphrase；It searches and collects template with the matched information of symptom characteristic, it is corresponding that load information collects template Template data；Node acquisition problems are generated according to the node data in template data and are exported, and are obtained and node acquisition problems Corresponding patient replys data；Data, which are replied, according to patient generates patient's acquisition information.
In one embodiment, processor executes computer program and realizes according to the node data generation section in template data Point acquisition problems simultaneously export, and obtain the step of patient corresponding with node acquisition problems replys data and are also used to：Obtain template number The corresponding start node data of start node and start node in generate node according to the node data in template data and adopt Collection problem simultaneously exports, and obtains patient's reply data corresponding with node acquisition problems and may include：Obtain rising in template data Beginning node and the corresponding start node data of start node；Starting acquisition problems are generated according to start node data and are exported, and are obtained Starting patient corresponding with starting acquisition problems is taken to reply data；The corresponding connection section of start node is obtained from template data Point, the selection first node corresponding with starting patient's reply data from connecting node, using first node as present node, and Continue to generate current acquisition problems according to the corresponding present node data of present node and export, until collected starting patient Until the corresponding present node of data is replied as template frontier node.
In one embodiment, processor executes computer program and realizes that load information collects the corresponding template data of template Step when be also used to：Obtain the formwork structure type that information collects template；When formwork structure type is nested template, will believe The caster data that breath is collected in template are loaded onto local cache from template database；Also realize following steps：Judge first segment Whether point is that nested template jumps node；When determining that first node jumps node for nested template, obtain belonging to first node Nested subtemplate, the corresponding subtemplate data of nested subtemplate are loaded onto local cache from database.
In one embodiment, following steps are also realized when processor executes computer program：Subtemplate modification is received to refer to Enable and upgrade subtemplate data；The submodule panel sign in subtemplate modification instruction is read, it is corresponding wait rise to search submodule panel sign Grade subtemplate data；Upgrading subtemplate data are compared with subtemplate data to be upgraded and generate template change data；It searches The corresponding author's mark of the corresponding relation template of submodule panel sign；Data, which are changed, according to template generates relation template upgrade tip, Relation template upgrade tip is sent to author and identifies corresponding author's terminal.
In one embodiment, processor executes computer program and realizes acquisition and the tongue picture paraphrase of tongue picture attributes match It is also used to when step：Tongue picture attribute is inputted into default neural network classifier and obtains the first paraphrase matching of each tongue picture paraphrase generally Rate；Tongue picture attribute is inputted into default Bayes classifier and obtains each the second paraphrase of tongue picture paraphrase matching probability；According to the first paraphrase Matching probability and the second paraphrase matching probability obtain the paraphrase matching rate of each tongue picture paraphrase, extract the highest tongue of paraphrase matching rate As paraphrase is as the tongue picture paraphrase with tongue picture attributes match.
In one embodiment, following steps are also realized when processor executes computer program：It is corresponding to obtain online doctor Doctor data, from doctor data extract buty cycle data；Idle doctor is filtered out from online doctor according to buty cycle data It is raw；Patient's tongue picture image and tongue picture paraphrase are sent to the corresponding doctor terminal of idle doctor；Execute from patient main suit's data and It is also used to realize when extracting the step of symptom characteristic in tongue picture paraphrase to work as and receives the corresponding with tongue picture paraphrase of doctor terminal return Paraphrase acknowledgement notification when, then extract symptom characteristic from patient main suit's data and tongue picture paraphrase.
In one embodiment, following steps are also realized when processor executes computer program：When receiving doctor terminal When the paraphrase acknowledgement notification corresponding with tongue picture paraphrase and amendment tongue picture paraphrase that return, released from patient main suit's data and amendment tongue picture Symptom characteristic is extracted in justice；Tongue picture attribute is associated with amendment tongue picture paraphrase and is added in tongue picture amendment sample set；According to amendment Sample set is adjusted the neuron weight of default neural network classifier and the probability distribution of default Bayes classifier.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated Machine program realizes following steps when being executed by processor：Obtain patient main suit's data and patient's tongue picture image；From patient's tongue picture figure Tongue picture attribute is extracted as in；Obtain the tongue picture paraphrase with tongue picture attributes match；It is extracted from patient main suit's data and tongue picture paraphrase Symptom characteristic；It searches and collects template with the matched information of symptom characteristic, load information collects the corresponding template data of template；According to Node data in template data generates node acquisition problems and exports, and obtains patient corresponding with node acquisition problems and replys number According to；Data, which are replied, according to patient generates patient's acquisition information.
In one embodiment, computer program is executed by processor the node data generation realized according in template data Node acquisition problems simultaneously export, and obtain the step of patient corresponding with node acquisition problems replys data and are also used to：Obtain template The corresponding start node data of start node and start node in data generate node according to the node data in template data Acquisition problems simultaneously export, and obtain patient's reply data corresponding with node acquisition problems and may include：It obtains in template data Start node and the corresponding start node data of start node；Starting acquisition problems are generated according to start node data and are exported, It obtains starting patient corresponding with starting acquisition problems and replys data；The corresponding connection section of start node is obtained from template data Point, the selection first node corresponding with starting patient's reply data from connecting node, using first node as present node, and Continue to generate current acquisition problems according to the corresponding present node data of present node and export, until collected starting patient Until the corresponding present node of data is replied as template frontier node.
In one embodiment, computer program, which is executed by processor, realizes that load information collects the corresponding template number of template According to step when be also used to：Obtain the formwork structure type that information collects template；It, will when formwork structure type is nested template The caster data that information is collected in template are loaded onto local cache from template database；Also realize following steps：Judge first Whether node is that nested template jumps node；When determining that first node jumps node for nested template, first node institute is obtained The corresponding subtemplate data of nested subtemplate are loaded onto local cache from database by the nested subtemplate of category.
In one embodiment, following steps are also realized when computer program is executed by processor：Receive subtemplate modification Instruction and upgrading subtemplate data；Read subtemplate modification instruction in submodule panel sign, search submodule panel sign it is corresponding to Upgrade subtemplate data；Upgrading subtemplate data are compared with subtemplate data to be upgraded and generate template change data；It looks into Look for the corresponding author's mark of the corresponding relation template of submodule panel sign；Data generation relation template upgrading is changed according to template to mention Show, relation template upgrade tip is sent to author and identifies corresponding author's terminal.
In one embodiment, computer program is executed by processor the tongue picture paraphrase realized and obtained with tongue picture attributes match Step when be also used to：Tongue picture attribute is inputted into default neural network classifier and obtains the first paraphrase matching of each tongue picture paraphrase generally Rate；Tongue picture attribute is inputted into default Bayes classifier and obtains each the second paraphrase of tongue picture paraphrase matching probability；According to the first paraphrase Matching probability and the second paraphrase matching probability obtain the paraphrase matching rate of each tongue picture paraphrase, extract the highest tongue of paraphrase matching rate As paraphrase is as the tongue picture paraphrase with tongue picture attributes match.
In one embodiment, following steps are also realized when computer program is executed by processor：Obtain online doctor couple The doctor data answered extracts buty cycle data from doctor data；The free time is filtered out from online doctor according to buty cycle data Doctor；Patient's tongue picture image and tongue picture paraphrase are sent to the corresponding doctor terminal of idle doctor；It executes from patient main suit's data With in tongue picture paraphrase extract symptom characteristic step when be also used to realize when receive doctor terminal return with tongue picture paraphrase pair When the paraphrase acknowledgement notification answered, then from patient main suit's data and tongue picture paraphrase extract symptom characteristic.
In one embodiment, following steps are also realized when computer program is executed by processor：It is whole when receiving doctor When holding the paraphrase acknowledgement notification corresponding with tongue picture paraphrase returned and amendment tongue picture paraphrase, from patient main suit's data and amendment tongue picture Symptom characteristic is extracted in paraphrase；Tongue picture attribute is associated with amendment tongue picture paraphrase and is added in tongue picture amendment sample set；According to repairing Positive sample collection adjusts the neuron weight of default neural network classifier and the probability distribution of default Bayes classifier It is whole.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, To any reference of memory, storage, database or other media used in each embodiment provided herein, Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM (PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms, Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
Each technical characteristic of above embodiments can be combined arbitrarily, for simplicity of description, not to above-described embodiment In each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lance Shield all should be considered as described in this specification.
The several embodiments of the application above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art It says, without departing from the concept of this application, various modifications and improvements can be made, these belong to the protection of the application Range.Therefore, the scope of protection shall be subject to the appended claims for the application patent.

Claims (10)

1. a kind of patient information acquisition method, the method includes：
Obtain patient main suit's data and patient's tongue picture image；
Tongue picture attribute is extracted from patient's tongue picture image；
Obtain the tongue picture paraphrase with the tongue picture attributes match；
Symptom characteristic is extracted from patient main suit's data and the tongue picture paraphrase；
It searches and collects template with the matched information of the symptom characteristic, load the information and collect the corresponding template data of template；
Node acquisition problems are generated according to the node data in the template data and are exported, and are obtained and the node acquisition problems Corresponding patient replys data；
Data, which are replied, according to the patient generates patient's acquisition information.
2. the method according to claim 1, wherein the node data according in the template data generates Node acquisition problems simultaneously export, and obtain patient corresponding with the node acquisition problems and reply data, including：
Obtain the start node and the corresponding start node data of the start node in the template data；
Starting acquisition problems are generated according to the start node data and are exported, and are obtained and the starting acquisition problems corresponding Beginning patient replys data；
The corresponding connecting node of the start node is obtained from the template data, from the connecting node selection with it is described It originates patient and replys the corresponding first node of data, using the first node as present node, and continue according to described current The corresponding present node data of node generate current acquisition problems and export, and correspond to until collected current patents reply data Present node be template frontier node until.
3. according to the method described in claim 2, it is characterized in that, the load information collects the corresponding template number of template According to, including：
Obtain the formwork structure type that the information collects template；
When the formwork structure type is nested template, the information is collected into the caster data in template from template data Library is loaded onto local cache；
It is described to continue also to wrap according to the current acquisition problems of the corresponding present node data generation of the present node and before exporting It includes：
Judge whether the first node is that nested template jumps node；
When determining the first node is that the nested template jumps node, nested submodule belonging to the first node is obtained The corresponding subtemplate data of the nesting subtemplate are loaded onto local cache from database by plate.
4. according to the method described in claim 3, it is characterized in that, the method also includes：
Receive subtemplate modification instruction and upgrading subtemplate data；
The submodule panel sign in the subtemplate modification instruction is read, the corresponding subtemplate to be upgraded of the submodule panel sign is searched Data；
The upgrading subtemplate data are compared with the subtemplate data to be upgraded and generate template change data；
Search the corresponding author's mark of the corresponding relation template of the submodule panel sign；
Data are changed according to the template and generate relation template upgrade tip, the relation template upgrade tip are sent to described Author identifies corresponding author's terminal.
5. the method according to claim 1, wherein the acquisition and the tongue picture of the tongue picture attributes match are released Justice, including：
The tongue picture attribute is inputted into default neural network classifier and obtains the first paraphrase matching probability of each tongue picture paraphrase；
The tongue picture attribute is inputted into default Bayes classifier and obtains each the second paraphrase of tongue picture paraphrase matching probability；
The paraphrase matching rate of each tongue picture paraphrase is obtained according to the first paraphrase matching probability and the second paraphrase matching probability, The highest tongue picture paraphrase of paraphrase matching rate is extracted as the tongue picture paraphrase with the tongue picture attributes match.
6. according to the method described in claim 5, it is characterized in that, the tongue picture paraphrase of the acquisition and the tongue picture attributes match Later, further include：
The corresponding doctor data of online doctor is obtained, buty cycle data are extracted from the doctor data；
Idle doctor is filtered out from the online doctor according to the buty cycle data；
Patient's tongue picture image and the tongue picture paraphrase are sent to the corresponding doctor terminal of the idle doctor；
It is described to extract symptom characteristic from patient main suit's data and the tongue picture paraphrase, including：
When receiving the paraphrase acknowledgement notification corresponding with the tongue picture paraphrase that the doctor terminal returns, then from the patient Symptom characteristic is extracted in main suit's data and the tongue picture paraphrase.
7. according to the method described in claim 6, it is characterized in that, the method also includes：
When the paraphrase acknowledgement notification corresponding with the tongue picture paraphrase and amendment tongue picture paraphrase for receiving the doctor terminal return When, symptom characteristic is extracted from patient main suit's data and the amendment tongue picture paraphrase；
The tongue picture attribute is associated with the amendment tongue picture paraphrase and is added in tongue picture amendment sample set；
Neuron weight and the default Bayes according to the amendment sample set to the default neural network classifier The probability distribution of classifier is adjusted.
8. a kind of patient information acquisition device, which is characterized in that described device includes：
Patient data obtains module, for obtaining patient main suit's data and patient's tongue picture image；
Tongue is as property extracting module, for extracting tongue picture attribute from patient's tongue picture image；
Paraphrase obtains module, for obtaining and the tongue picture paraphrase of the tongue picture attributes match；
Characteristic extracting module, for extracting symptom characteristic from patient main suit's data and the tongue picture paraphrase；
Template searching module collects template with the matched information of the symptom characteristic for searching, loads the information and collect mould The corresponding template data of plate；
Data acquisition module is obtained for generating node acquisition problems according to the node data in the template data and exporting Patient corresponding with the node acquisition problems replys data；
Information generating module generates patient's acquisition information for replying data according to the patient.
9. a kind of computer equipment, including memory and processor, the memory are stored with computer program, feature exists In the step of processor realizes any one of claims 1 to 7 the method when executing the computer program.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program The step of method described in any one of claims 1 to 7 is realized when being executed by processor.
CN201810546481.4A 2018-05-31 2018-05-31 Patient information acquisition method and device, computer equipment and storage medium Active CN108877928B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810546481.4A CN108877928B (en) 2018-05-31 2018-05-31 Patient information acquisition method and device, computer equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810546481.4A CN108877928B (en) 2018-05-31 2018-05-31 Patient information acquisition method and device, computer equipment and storage medium

Publications (2)

Publication Number Publication Date
CN108877928A true CN108877928A (en) 2018-11-23
CN108877928B CN108877928B (en) 2023-04-07

Family

ID=64336186

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810546481.4A Active CN108877928B (en) 2018-05-31 2018-05-31 Patient information acquisition method and device, computer equipment and storage medium

Country Status (1)

Country Link
CN (1) CN108877928B (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109770875A (en) * 2019-03-27 2019-05-21 上海铀米机器人科技有限公司 A kind of human body constitution discrimination method and system based on neural network classifier
CN112216383A (en) * 2020-10-26 2021-01-12 山东众阳健康科技集团有限公司 Traditional Chinese medicine intelligent inquiry tongue diagnosis comprehensive system based on syndrome element and deep learning
CN113257423A (en) * 2021-06-28 2021-08-13 天津慧医谷科技有限公司 Health detection system and method and electronic equipment
CN113837986A (en) * 2020-12-15 2021-12-24 京东科技控股股份有限公司 Method, apparatus, electronic device, and medium for recognizing tongue picture
TWI818203B (en) * 2020-10-23 2023-10-11 國立臺灣大學醫學院附設醫院 Classification model establishment method based on disease conditions

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101543401A (en) * 2009-04-17 2009-09-30 张庆龙 Intelligent electronic quick tongue manifestation disease diagnoser

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101543401A (en) * 2009-04-17 2009-09-30 张庆龙 Intelligent electronic quick tongue manifestation disease diagnoser

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
朱庆文 等: "基于脉诊、舌诊信息提取与识别的诊断集成技术研究策略" *
许家佗 等: "舌象量级诊断方法对521例舌象特征的临床观察" *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109770875A (en) * 2019-03-27 2019-05-21 上海铀米机器人科技有限公司 A kind of human body constitution discrimination method and system based on neural network classifier
TWI818203B (en) * 2020-10-23 2023-10-11 國立臺灣大學醫學院附設醫院 Classification model establishment method based on disease conditions
CN112216383A (en) * 2020-10-26 2021-01-12 山东众阳健康科技集团有限公司 Traditional Chinese medicine intelligent inquiry tongue diagnosis comprehensive system based on syndrome element and deep learning
CN112216383B (en) * 2020-10-26 2023-02-21 山东众阳健康科技集团有限公司 Traditional Chinese medicine intelligent inquiry tongue diagnosis comprehensive system based on syndrome element and deep learning
CN113837986A (en) * 2020-12-15 2021-12-24 京东科技控股股份有限公司 Method, apparatus, electronic device, and medium for recognizing tongue picture
CN113257423A (en) * 2021-06-28 2021-08-13 天津慧医谷科技有限公司 Health detection system and method and electronic equipment

Also Published As

Publication number Publication date
CN108877928B (en) 2023-04-07

Similar Documents

Publication Publication Date Title
CN108877928A (en) Patient information acquisition method, device, computer equipment and storage medium
Dieber et al. Why model why? Assessing the strengths and limitations of LIME
US20200364504A1 (en) System and method for interpretable sequence and time-series data modeling
CN109978022B (en) Medical text information processing method and device and storage medium
CN107516110B (en) Medical question-answer semantic clustering method based on integrated convolutional coding
CN107239529B (en) Public opinion hotspot category classification method based on deep learning
CN110021439A (en) Medical data classification method, device and computer equipment based on machine learning
CN111967495B (en) Classification recognition model construction method
CN110335653A (en) Non-standard case history analytic method based on openEHR case history format
CN106203490A (en) Based on attribute study and the image ONLINE RECOGNITION of interaction feedback, search method under a kind of Android platform
CN110277167A (en) The Chronic Non-Communicable Diseases Risk Forecast System of knowledge based map
CN112037179B (en) Method, system and equipment for generating brain disease diagnosis model
CN110705490B (en) Visual emotion recognition method
CN112101162A (en) Image recognition model generation method and device, storage medium and electronic equipment
CN117198468B (en) Intervention scheme intelligent management system based on behavior recognition and data analysis
CN115147661A (en) Chromosome classification method, device, equipment and readable storage medium
CN114201656A (en) Combining data-driven models for classifying data
Liu et al. Hipal: A deep framework for physician burnout prediction using activity logs in electronic health records
CN113779295A (en) Retrieval method, device, equipment and medium for abnormal cell image features
CN117078007A (en) Multi-scale wind control system integrating scale labels and method thereof
CN116719840A (en) Medical information pushing method based on post-medical-record structured processing
CN116311380A (en) Skin typing method for autonomous learning of small sample data
CN114139531B (en) Medical entity prediction method and system based on deep learning
CN114582449A (en) Electronic medical record named entity standardization method and system based on XLNet-BiGRU-CRF model
Zhang et al. TIM‐Net: A multi‐label classification network for TCM tongue images fusing global‐local features

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant