CN115662593B - Doctor-patient matching method, device, equipment and medium based on symptom knowledge graph - Google Patents
Doctor-patient matching method, device, equipment and medium based on symptom knowledge graph Download PDFInfo
- Publication number
- CN115662593B CN115662593B CN202211394235.4A CN202211394235A CN115662593B CN 115662593 B CN115662593 B CN 115662593B CN 202211394235 A CN202211394235 A CN 202211394235A CN 115662593 B CN115662593 B CN 115662593B
- Authority
- CN
- China
- Prior art keywords
- symptom
- current
- historical
- target
- patient
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- 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
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The method carries out symptom matching on the current symptoms of the patient through the symptom knowledge maps corresponding to all hospitals in the current position area of the patient, can carry out preliminary diagnosis on the current symptoms of the patient, and can determine the hospitals with treatment capability on the current symptoms of the patient; the treatment capacity of each hospital in the current position area for the current symptoms of the patient is determined through weight calculation of the target related parameters corresponding to the current patient, and the cure sequence is obtained, so that the hospital with the highest cure probability for the current symptoms of the patient in each hospital in the current position area is determined as the target hospital, and the cure probability of the patient is improved. Therefore, based on the symptom knowledge graph generated by the historical diagnosis data of each hospital, the rapid diagnosis and hospital matching of the current symptoms of the patient can be realized, and the accuracy of doctor-patient matching is improved.
Description
Technical Field
The application relates to the technical field of model matching, in particular to a doctor-patient matching method, device, equipment and medium based on a symptom knowledge graph.
Background
There are two forms of doctor-patient matching, including patient-initiative doctor-patient matching, such as patient-initiative doctor-seeking, and doctor-initiative doctor-patient matching, such as ambulance, to bring the patient to a designated hospital.
However, the patient does not have diagnostic capability, and the ambulance has a doctor on the ambulance, but the doctor on the ambulance is only responsible for emergency treatment tasks, but not for diagnostic tasks, so the ambulance generally adopts a nearby principle to send the patient to the nearest hospital without considering the treatment capability of the hospital. Thus, treatment of a patient's condition may be affected by delayed diagnosis or misestimation of hospital rescue capabilities.
Therefore, how to improve the accuracy of doctor-patient matching is a technical problem to be solved.
Disclosure of Invention
The application provides a doctor-patient matching method, device and equipment based on symptom knowledge graph and a storage medium, which can improve the accuracy of doctor-patient matching.
In a first aspect, the present application provides a method for matching a doctor-patient based on a symptom knowledge graph, the method for matching a doctor-patient based on a symptom knowledge graph comprising the steps of:
acquiring current symptoms and a current position area of a current patient;
based on a symptom knowledge graph, matching the current symptom to obtain at least one target related parameter corresponding to the current symptom;
based on a preset matching model and preset weights corresponding to relevant parameters of each target, calculating the cure probability of each hospital in the current position area for the current symptom, and obtaining the cure sequence of each hospital for the current symptom;
and determining a target hospital matched with the current patient in each hospital in the current position area based on the cure sequence of each hospital for the current symptom.
Further, before the step of acquiring the current symptom and the current location area of the current patient, the method further comprises:
acquiring historical medical history data and historical symptom data of historical patients of all hospitals in a current location area;
extracting target standard words from the historical medical history data and the historical symptom data based on preset standard words;
and taking the target standard word as a symptom node and constructing the symptom knowledge graph.
Further, the extracting the target standard word from the historical medical history data and the historical symptom data based on the preset standard word includes:
traversing at least one diagnostic medical record in the historical medical history data;
and matching target symptoms with the preset standard words in the historical symptom data corresponding to the diagnosis medical record and the symptom data corresponding to the preset standard words, and taking the preset standard words corresponding to the target symptoms as the target standard words.
Further, the matching the target symptom having the preset standard word in the historical symptom data corresponding to the diagnostic medical record and the symptom data corresponding to the preset standard word, and taking the preset standard word corresponding to the target symptom as the target standard word includes:
when the same diagnosis result exists in the diagnosis medical record, determining at least two similar symptoms corresponding to the same diagnosis result;
comparing the degree of deviation between the similar symptoms, and distinguishing the similar symptoms into standard words and synonyms;
and marking the synonyms as the corresponding standard words, and taking the standard words as the target standard words.
Further, after the historical medical history data and the historical symptom data of the historical patients of each hospital in the current location area are obtained, the method further comprises the following steps:
based on the historical medical history data and the historical symptom data, obtaining the occurrence frequency, the symptom grade, the treatment result and the criticality of each historical symptom as training related parameters;
calculating a predicted treatment result corresponding to each historical symptom based on the preset weight corresponding to each training related parameter;
obtaining actual treatment results corresponding to each historical symptom, comparing the actual treatment results with the predicted treatment results, and determining target training symptoms in each historical symptom;
and training a matching model based on the historical medical history corresponding to the target training symptom and the historical symptom data to obtain the preset matching model.
Further, the step of obtaining an actual treatment result corresponding to each historical symptom, comparing the actual treatment result with the predicted treatment result, and determining a target training symptom in each historical symptom includes:
when the predicted treatment result corresponding to the historical symptom is not matched with the actual treatment result, the historical symptom is used as a training correction symptom;
and when the historical symptoms correspond to the predicted treatment result and the actual treatment result and are matched, determining that the historical symptoms are the target training symptoms.
Further, the matching the current symptom based on the symptom knowledge graph to obtain at least one target related parameter corresponding to the current symptom includes:
obtaining at least one historical symptom hit by the current symptom in the symptom knowledge graph;
and acquiring the hit frequency of the current symptom hitting the historical symptom, and/or the target symptom grade corresponding to the current symptom, and/or acquiring the treatment result corresponding to each historical symptom, and/or the critical sequencing result of each historical symptom hit by the current symptom, as the target related parameter.
In a second aspect, the present application further provides a doctor-patient matching device based on a symptom knowledge graph, where the doctor-patient matching device based on the symptom knowledge graph includes:
the current symptom acquisition module is used for acquiring the current symptom and the current position area of the current patient;
the symptom matching module is used for matching the current symptom based on a symptom knowledge graph to obtain at least one target related parameter corresponding to the current symptom;
the cure ordering module is used for calculating the cure probability of each hospital for the current symptom in the current position area based on a preset matching model and preset weights corresponding to the relevant parameters of each target, and obtaining the cure ordering of each hospital for the current symptom;
and the target hospital determining module is used for determining a target hospital matched with the current patient in each hospital in the current position area based on the cure sequence of each hospital for the current symptom.
In a third aspect, the present application further provides a computer device, the computer device comprising a processor, a memory, and a computer program stored on the memory and executable by the processor, wherein the computer program when executed by the processor implements the steps of the above-described method for matching a doctor-patient based on a symptom knowledge graph.
In a fourth aspect, the present application further provides a computer readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the above-mentioned method for matching a doctor and patient based on a symptom knowledge graph.
The application provides a doctor-patient matching method, device, equipment and storage medium based on symptom knowledge graph, wherein the method comprises the steps of obtaining current symptoms and current position areas of a current patient; based on a symptom knowledge graph, matching the current symptom to obtain at least one target related parameter corresponding to the current symptom; based on a preset matching model and preset weights corresponding to relevant parameters of each target, calculating the cure probability of each hospital in the current position area for the current symptom, and obtaining the cure sequence of each hospital for the current symptom; and determining a target hospital matched with the current patient in each hospital in the current position area based on the cure sequence of each hospital for the current symptom. According to the method and the device, through the symptom knowledge maps corresponding to all hospitals in the current position area of the patient, the current symptoms of the patient are subjected to symptom matching, the current symptoms of the patient can be subjected to preliminary diagnosis, and the hospitals with treatment capability on the current symptoms of the patient can be determined; the treatment capacity of each hospital in the current position area for the current symptoms of the patient is determined through weight calculation of the target related parameters corresponding to the current patient, and the cure sequence is obtained, so that the hospital with the highest cure probability for the current symptoms of the patient in each hospital in the current position area is determined as the target hospital, and the cure probability of the patient is improved. Therefore, based on the symptom knowledge graph generated by the historical diagnosis data of each hospital, the rapid diagnosis and hospital matching of the current symptoms of the patient can be realized, the problem that symptoms cannot be diagnosed when no diagnosis equipment is provided is avoided, and the accuracy of doctor-patient matching is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a first embodiment of a doctor-patient matching method based on a symptom knowledge graph according to an embodiment of the present application;
FIG. 2 is a flowchart illustrating an embodiment of a method for constructing a symptom knowledge graph according to the present disclosure;
FIG. 3 is a flowchart illustrating an embodiment of a method for constructing a preset matching model according to the present application;
FIG. 4 is a schematic block diagram of a doctor-patient matching device based on a symptom knowledge-graph provided by an embodiment of the present application;
fig. 5 is a schematic block diagram of a computer device according to an embodiment of the present application.
The realization, functional characteristics and advantages of the present application will be further described with reference to the embodiments, referring to the attached drawings.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
The flow diagrams depicted in the figures are merely illustrative and not necessarily all of the elements and operations/steps are included or performed in the order described. For example, some operations/steps may be further divided, combined, or partially combined, so that the order of actual execution may be changed according to actual situations.
Some embodiments of the present application are described in detail below with reference to the accompanying drawings. The following embodiments and features of the embodiments may be combined with each other without conflict.
The embodiment of the application provides a doctor-patient matching method based on a symptom knowledge graph, a doctor-patient matching device based on the symptom knowledge graph, computer equipment and a storage medium, which are used for matching and calling target business events in a preset business event library according to page configuration data, so as to render a target page assembly, generate a target page, and improve the research and development efficiency of a front-end page.
Referring to fig. 1, fig. 1 is a flowchart of a first embodiment of a doctor-patient matching method based on a symptom knowledge graph according to an embodiment of the present application.
As shown in fig. 1, the doctor-patient matching method based on the symptom knowledge graph includes steps S101 to S103.
Step S101, acquiring current symptoms and current position areas of a current patient;
in this embodiment, the symptoms of the current patient may be obtained according to the patient description or the preliminary diagnosis of the ambulance onboard doctor, so that the predictive model can more accurately perform the preliminary diagnosis of the symptoms of the current patient. The current position area of the current patient is acquired, and information of each hospital in the current position area can be acquired, wherein the information comprises symptom knowledge maps comprising diagnosis data and the like and position information of each hospital.
In an embodiment, for example, the current symptom of the current patient is chest pain, and the cause of chest pain is many, for example, the current patient suffers from heavy impact, then it can be judged that the chest pain is caused by external impact, further, for example, the pain is aggravated by compression, then the phenomenon of chest fracture may exist, at this time, a doctor can further diagnose or guide the patient to describe more specific symptoms according to the common symptom, so as to diagnose the current symptom of the current patient more accurately.
In an embodiment, the regional map may be divided according to the location distribution of hospitals, for example, a certain city has a plurality of hospitals, and a plurality of hospitals with more centralized distribution or closer distribution distances may be divided into the same location area, and the remote locations may be separately divided, so as to divide the city into a plurality of location areas.
In an embodiment, an area with a certain distance range of the current position may also be obtained according to the current position of the current patient, and the area is used as a current position area, for example, an area with a surrounding 5km range with the current position of the current patient as a center of a circle, and as a current position area, all hospital information in the current position area including a symptom knowledge graph and position information corresponding to each hospital is obtained.
Step S102, matching the current symptom based on a symptom knowledge graph to obtain at least one target related parameter corresponding to the current symptom;
in this embodiment, in the symptom knowledge graph, the historical symptoms corresponding to the current symptoms are matched to obtain at least one diagnosis result corresponding to the current symptoms, and at least one target related parameter is obtained according to the diagnosis result to further diagnose the current symptoms.
In an embodiment, obtaining at least one historical symptom for which the current symptom hit in the symptom knowledge graph; and acquiring the hit frequency of the current symptom hitting the historical symptom, and/or the target symptom grade corresponding to the current symptom, and/or acquiring the treatment result corresponding to each historical symptom, and/or the critical sequencing result of each historical symptom hit by the current symptom, as the target related parameter.
In an embodiment, in the symptom knowledge graph, for each symptom, the frequency of occurrence of each symptom in the symptom knowledge graph may be classified as a first related parameter, the patient associated with the symptom is classified as a second related parameter, the patient treatment result associated with the symptom is a third related parameter, all the symptoms are classified by a doctor, and the classification is performed according to the criticality, and the secondary classification is used as a fourth related parameter.
Step S103, calculating the cure probability of each hospital for the current symptom in the current position area based on a preset matching model and preset weights corresponding to the relevant parameters of each target, and obtaining the cure sequence of each hospital for the current symptom;
in this embodiment, the relevant parameters are used as input, input into a preset matching model, cure probabilities of the hospitals for the current symptoms are output according to preset weights set by doctors for the relevant parameters, and the hospitals are ranked according to the cure probabilities.
In an embodiment, a weight may be set according to the influence degree of each relevant parameter on symptom diagnosis, for example, the first relevant parameter is a symptom hit frequency, when the symptom hit frequency is more, it indicates that the symptom corresponds to more diagnosis results, and the symptom may be a common symptom, and the weight may be set smaller.
In an embodiment, the second related parameter indicates a patient rating, if the patient rating is low, it indicates that the current patient is not particularly urgent, a hospital with the highest cure probability may be selected if the patient's current condition is less weighted, conversely, if the patient's current condition is particularly urgent, a hospital with greater weight needs to be set, and more time needs to be considered for visit, a hospital that can treat the current patient most quickly needs to be selected in order to stabilize the patient's condition.
In an embodiment, the third relevant parameter represents a patient treatment outcome, which represents the treatment effect of each hospital on the current symptoms, which may be weighted more heavily, and in the matching process, hospitals where the treatment situation is high may be recommended.
In an embodiment, the fourth related parameter represents a criticality level, the part may be ranked according to expert opinion, and the corresponding weights may be set according to different criticality levels, e.g. a higher criticality level may be set to a larger weight and a lower criticality level may be set to a smaller weight.
Step S104, determining a target hospital matched with the current patient in each hospital in the current position area based on the cure sequence of each hospital for the current symptom.
In this embodiment, after comprehensively considering a plurality of relevant parameters, all hospitals which can treat the current symptoms of the current patient in the current location area are determined, and the hospitals are ranked according to the treatment conditions of the current symptoms of the hospitals, and the hospitals with high overall treatment conditions are selected as the target hospitals to be recommended preferentially.
The embodiment provides a doctor-patient matching method based on a symptom knowledge graph, which performs symptom matching on the current symptom of a patient through the symptom knowledge graph corresponding to each hospital in the current position area of the patient, can perform preliminary diagnosis on the current symptom of the patient, and can determine the hospitals with treatment capability on the current symptom of the patient; the treatment capacity of each hospital in the current position area for the current symptoms of the patient is determined through weight calculation of the target related parameters corresponding to the current patient, and the cure sequence is obtained, so that the hospital with the highest cure probability for the current symptoms of the patient in each hospital in the current position area is determined as the target hospital, and the cure probability of the patient is improved. Therefore, based on the symptom knowledge graph generated by the historical diagnosis data of each hospital, the rapid diagnosis and hospital matching of the current symptoms of the patient can be realized, the problem that symptoms cannot be diagnosed when no diagnosis equipment is provided is avoided, and the accuracy of doctor-patient matching is improved.
Referring to fig. 2, fig. 2 is a flow chart of an embodiment of a method for constructing a symptom knowledge graph provided in the present application.
As shown in fig. 2, the symptom knowledge graph construction method includes steps S201 to S203.
Step S201, acquiring historical medical history data and historical symptom data of historical patients of all hospitals in a current location area;
in this embodiment, an electronic medical record of a history patient in each hospital may be obtained, where the electronic medical record includes history data and history symptom data of the patient.
In an embodiment, according to the historical medical history data and the historical symptom data in the electronic medical record, various disease diagnosis results, corresponding symptoms, treatment results and the like can be obtained, for example, the diagnosis result is acute gastroenteritis, the corresponding symptoms are nausea, vomiting, abdominal pain, diarrhea, vomiting onset symptoms and the like, and the treatment results can include treatment quantity, cure proportion, recurrence proportion and the like.
Step S202, extracting target standard words from the historical medical history data and the historical symptom data based on preset standard words;
in this embodiment, different hospitals or doctors may have different names for the same disease, and a set of standard words may be set for different disease with the same or similar symptoms, so as to unify the disease, and then the diagnosis results identical or similar to the standard words are extracted from the historical medical history data and the historical symptom data, and the corresponding diagnosis results, symptoms, treatment conditions and other data are extracted.
In one embodiment, traversing at least one diagnostic medical record in the historical medical history data; and matching target symptoms with the preset standard words in the historical symptom data corresponding to the diagnosis medical record and the symptom data corresponding to the preset standard words, and taking the preset standard words corresponding to the target symptoms as the target standard words.
In one embodiment, when the diagnostic medical record has the same diagnostic result, determining at least two similar symptoms corresponding to the same diagnostic result; comparing the degree of deviation between the similar symptoms, and distinguishing the similar symptoms into standard words and synonyms; and marking the synonyms as the corresponding standard words, and taking the standard words as the target standard words.
In one embodiment, a set of standard words is established according to the description of symptoms in the teaching materials, all electronic medical records are scanned by the standard words in a character string matching mode, the deviation degree between the symptoms of the medical records with the same diagnosis is compared, then medical records with large deviation degree are extracted for manual correction, and finally a set of standard words and synonyms are obtained.
And step 203, using the target standard words as symptom nodes, and constructing the symptom knowledge graph.
In one embodiment, the standard words are used as symptom nodes, and synonyms are extracted according to the standard words to obtain a set of high-precision knowledge graph extraction method.
Referring to fig. 3, fig. 3 is a flowchart illustrating an embodiment of a preset matching model construction method provided in the present application.
As shown in fig. 3, the symptom knowledge graph construction method includes steps S301 to S304.
Step S301, based on the historical medical history data and the historical symptom data, obtaining the occurrence frequency, the symptom grade, the treatment result and the criticality of each historical symptom as training related parameters;
step S302, calculating a predicted treatment result corresponding to each historical symptom based on preset weights corresponding to each training related parameter;
step S303, obtaining an actual treatment result corresponding to each historical symptom, comparing the actual treatment result with the predicted treatment result, and determining a target training symptom in each historical symptom;
in one embodiment, the training correction symptom is used when the predicted treatment result corresponding to the historical symptom and the actual treatment result are not matched; and when the historical symptoms correspond to the predicted treatment result and the actual treatment result and are matched, determining that the historical symptoms are the target training symptoms.
And step 304, training a matching model based on the historical medical history corresponding to the target training symptom and the historical symptom data to obtain the preset matching model.
In this embodiment, after the historical medical history data and the historical symptom data are obtained, the frequency of occurrence of symptoms, the grade of symptoms, the treatment result, the degree of crisis and the like can be used as main influencing factors of symptom diagnosis, namely related parameters; according to the weight of each related parameter, each related parameter is comprehensively considered, so that a training model predicts a symptom diagnosis result, the model training effect is judged by combining with an actual diagnosis result, and the model is timely corrected until the accuracy of model prediction reaches a preset value, and then the symptom diagnosis can be performed by using the model.
In one embodiment, the relevant parameters are divided into positive relevant parameters and negative relevant parameters, the weighted values are set by an expert, the patient outcome in the clinical data is trained as very good clinical data, the clinical data are arranged according to time, the data with early historic date are used for training, the data with end historic date are used as input, whether the output conclusion is consistent with the fact or not is judged, the non-conforming event is detected manually and used for correcting a training data set, and a set of doctor-patient matching models is obtained.
Referring to fig. 4, fig. 4 is a schematic block diagram of a doctor-patient matching device based on a symptom knowledge graph according to an embodiment of the present application, where the doctor-patient matching device based on the symptom knowledge graph is used for executing the doctor-patient matching method based on the symptom knowledge graph. The doctor-patient matching device based on the symptom knowledge graph can be configured in the terminal.
As shown in fig. 4, the doctor-patient matching device 100 based on symptom knowledge graph includes: a current symptom acquisition module 101, a symptom matching module 102, a cure ordering module 103, and a target hospital determination module 104.
A current symptom acquisition module 101 for acquiring a current symptom and a current location area of a current patient;
the symptom matching module 102 is configured to match the current symptom based on a symptom knowledge graph, and obtain at least one target related parameter corresponding to the current symptom;
the cure ranking module 103 is configured to calculate cure probabilities of all hospitals in the current location area for the current symptom based on preset matching models and preset weights corresponding to all target related parameters, and obtain cure rankings of all hospitals for the current symptom;
a target hospital determination module 104, configured to determine a target hospital matched by the current patient in each hospital in the current location area based on the cure ranks of each hospital for the current symptom.
In an embodiment, the doctor-patient matching device 100 based on the symptom knowledge graph further includes a symptom knowledge graph construction module, configured to obtain historical medical history data and historical symptom data of historical patients of each hospital in the current location area; extracting target standard words from the historical medical history data and the historical symptom data based on preset standard words; and taking the target standard word as a symptom node and constructing the symptom knowledge graph.
In an embodiment, the symptom knowledge graph construction module is further configured to implement traversing at least one diagnostic medical record in the historical medical history data; and matching target symptoms with the preset standard words in the historical symptom data corresponding to the diagnosis medical record and the symptom data corresponding to the preset standard words, and taking the preset standard words corresponding to the target symptoms as the target standard words.
In an embodiment, the symptom knowledge graph construction module is further configured to determine at least two similar symptoms corresponding to the same diagnosis result when the same diagnosis result exists in the diagnosis medical record; comparing the degree of deviation between the similar symptoms, and distinguishing the similar symptoms into standard words and synonyms; and marking the synonyms as the corresponding standard words, and taking the standard words as the target standard words.
In an embodiment, the doctor-patient matching device 100 based on the symptom knowledge graph further includes a preset matching model construction module, configured to obtain, based on the historical medical history data and the historical symptom data, occurrence frequency, symptom level, treatment result and criticality of each historical symptom as training related parameters; calculating a predicted treatment result corresponding to each historical symptom based on the preset weight corresponding to each training related parameter; obtaining actual treatment results corresponding to each historical symptom, comparing the actual treatment results with the predicted treatment results, and determining target training symptoms in each historical symptom; and training a matching model based on the historical medical history corresponding to the target training symptom and the historical symptom data to obtain the preset matching model.
In an embodiment, the preset matching model building module is further configured to implement the training correction symptom when the predicted treatment result corresponding to the historical symptom and the actual treatment result are not matched; and when the historical symptoms correspond to the predicted treatment result and the actual treatment result and are matched, determining that the historical symptoms are the target training symptoms.
In an embodiment, the symptom matching module 102 is further configured to obtain at least one historical symptom hit by the current symptom in the symptom knowledge graph; and acquiring the hit frequency of the current symptom hitting the historical symptom, and/or the target symptom grade corresponding to the current symptom, and/or acquiring the treatment result corresponding to each historical symptom, and/or the critical sequencing result of each historical symptom hit by the current symptom, as the target related parameter.
It should be noted that, for convenience and brevity of description, specific working processes of the above-described apparatus and each module may refer to corresponding processes in the foregoing embodiments of the doctor-patient matching method based on the symptom knowledge graph, which are not described herein again.
The apparatus provided by the above embodiments may be implemented in the form of a computer program which may be run on a computer device as shown in fig. 5.
Referring to fig. 5, fig. 5 is a schematic block diagram of a computer device according to an embodiment of the present application. The computer device may be a terminal.
With reference to FIG. 5, the computer device includes a processor, memory, and a network interface connected by a system bus, where the memory may include a non-volatile storage medium and an internal memory.
The non-volatile storage medium may store an operating system and a computer program. The computer program comprises program instructions which, when executed, cause the processor to perform any one of a number of methods for matching a doctor to a patient based on a symptom knowledge graph.
The processor is used to provide computing and control capabilities to support the operation of the entire computer device.
The internal memory provides an environment for the execution of a computer program in the non-volatile storage medium, which when executed by the processor, causes the processor to perform any one of a number of methods for matching a doctor with a patient based on a symptom knowledge graph.
The network interface is used for network communication such as transmitting assigned tasks and the like. It will be appreciated by those skilled in the art that the structure shown in fig. 5 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
It should be appreciated that the processor may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field-programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. Wherein the general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Wherein in one embodiment the processor is configured to run a computer program stored in the memory to implement the steps of:
acquiring current symptoms and a current position area of a current patient;
based on a symptom knowledge graph, matching the current symptom to obtain at least one target related parameter corresponding to the current symptom;
based on a preset matching model and preset weights corresponding to relevant parameters of each target, calculating the cure probability of each hospital in the current position area for the current symptom, and obtaining the cure sequence of each hospital for the current symptom;
and determining a target hospital matched with the current patient in each hospital in the current position area based on the cure sequence of each hospital for the current symptom.
In an embodiment, the processor, prior to implementing the acquiring the current symptom and the current location area of the current patient, is further configured to implement:
acquiring historical medical history data and historical symptom data of historical patients of all hospitals in a current location area;
extracting target standard words from the historical medical history data and the historical symptom data based on preset standard words;
and taking the target standard word as a symptom node and constructing the symptom knowledge graph.
In an embodiment, the processor is configured to, when implementing the extracting the target standard word from the historical medical history data and the historical symptom data based on the preset standard word, implement:
traversing at least one diagnostic medical record in the historical medical history data;
and matching target symptoms with the preset standard words in the historical symptom data corresponding to the diagnosis medical record and the symptom data corresponding to the preset standard words, and taking the preset standard words corresponding to the target symptoms as the target standard words.
In an embodiment, the processor is configured to, when implementing the matching of the historical symptom data corresponding to the diagnostic medical record and the symptom data corresponding to the preset standard word with the target symptom having the preset standard word, take the preset standard word corresponding to the target symptom as the target standard word, implement:
when the same diagnosis result exists in the diagnosis medical record, determining at least two similar symptoms corresponding to the same diagnosis result;
comparing the degree of deviation between the similar symptoms, and distinguishing the similar symptoms into standard words and synonyms;
and marking the synonyms as the corresponding standard words, and taking the standard words as the target standard words.
In one embodiment, the processor is further configured to, after implementing the acquiring the historical medical history data and the historical symptom data of the historical patients of each hospital in the current location area, implement:
based on the historical medical history data and the historical symptom data, obtaining the occurrence frequency, the symptom grade, the treatment result and the criticality of each historical symptom as training related parameters;
calculating a predicted treatment result corresponding to each historical symptom based on the preset weight corresponding to each training related parameter;
obtaining actual treatment results corresponding to each historical symptom, comparing the actual treatment results with the predicted treatment results, and determining target training symptoms in each historical symptom;
and training a matching model based on the historical medical history corresponding to the target training symptom and the historical symptom data to obtain the preset matching model.
In an embodiment, the processor is configured to, when implementing the obtaining the actual treatment result corresponding to each historical symptom, compare the actual treatment result with the predicted treatment result, determine a target training symptom from each historical symptom, and implement:
when the predicted treatment result corresponding to the historical symptom is not matched with the actual treatment result, the historical symptom is used as a training correction symptom;
and when the historical symptoms correspond to the predicted treatment result and the actual treatment result and are matched, determining that the historical symptoms are the target training symptoms.
In an embodiment, when implementing the symptom-based knowledge graph, the processor is configured to match the current symptom to obtain at least one target-related parameter corresponding to the current symptom, where the target-related parameter is implemented by:
obtaining at least one historical symptom hit by the current symptom in the symptom knowledge graph;
and acquiring the hit frequency of the current symptom hitting the historical symptom, and/or the target symptom grade corresponding to the current symptom, and/or acquiring the treatment result corresponding to each historical symptom, and/or the critical sequencing result of each historical symptom hit by the current symptom, as the target related parameter.
The computer readable storage medium may be an internal storage unit of the computer device according to the foregoing embodiment, for example, a hard disk or a memory of the computer device. The computer readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), or the like, which are provided on the computer device.
While the invention has been described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions of equivalents may be made and equivalents will be apparent to those skilled in the art without departing from the scope of the invention. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A doctor-patient matching method based on symptom knowledge graph, which is characterized by comprising the following steps:
acquiring current symptoms and a current position area of a current patient;
based on a symptom knowledge graph, matching the current symptom to obtain at least one target related parameter corresponding to the current symptom;
the target related parameters comprise hit frequencies of each historical symptom hit by the current symptom in a symptom knowledge graph, and/or target symptom grades corresponding to the current symptom, and/or treatment results corresponding to each historical symptom hit by the current symptom in the symptom knowledge graph, and/or critical sequencing results of each historical symptom hit by the current symptom in the symptom knowledge graph;
based on a preset matching model and preset weights corresponding to relevant parameters of each target, calculating the cure probability of each hospital in the current position area for the current symptom, and obtaining the cure sequence of each hospital for the current symptom;
the preset matching model is used for calculating the cure probability of each hospital for the current symptom according to each target related parameter and each preset weight corresponding to each target related parameter, and sequencing each hospital according to the cure probability;
and determining a target hospital matched with the current patient in each hospital in the current position area based on the cure sequence of each hospital for the current symptom.
2. The method for matching a doctor to an patient based on a symptom knowledge graph according to claim 1, wherein before the step of obtaining the current symptom and the current location area of the current patient, further comprises:
acquiring historical medical history data and historical symptom data of historical patients of all hospitals in a current location area;
extracting target standard words from the historical medical history data and the historical symptom data based on preset standard words;
and taking the target standard word as a symptom node and constructing the symptom knowledge graph.
3. The method for matching doctors and patients based on the symptom knowledge graph according to claim 2, wherein the extracting target standard words from the historical medical history data and the historical symptom data based on the preset standard words comprises:
traversing at least one diagnostic medical record in the historical medical history data;
and matching target symptoms with the preset standard words in the historical symptom data corresponding to the diagnosis medical record and the symptom data corresponding to the preset standard words, and taking the preset standard words corresponding to the target symptoms as the target standard words.
4. The method for matching a doctor-patient based on a symptom knowledge graph according to claim 3, wherein the matching a target symptom having the preset standard word in the history symptom data corresponding to the diagnosis medical record and the symptom data corresponding to the preset standard word, using the preset standard word corresponding to the target symptom as the target standard word, includes:
when the same diagnosis result exists in the diagnosis medical record, determining at least two similar symptoms corresponding to the same diagnosis result;
comparing the degree of deviation between the similar symptoms, and distinguishing the similar symptoms into standard words and synonyms;
and marking the synonyms as the corresponding standard words, and taking the standard words as the target standard words.
5. The method for matching doctors and patients based on the symptom knowledge graph according to claim 2, wherein after obtaining the historical medical history data and the historical symptom data of the historical patients of each hospital in the current location area, the method further comprises:
based on the historical medical history data and the historical symptom data, obtaining the occurrence frequency, the symptom grade, the treatment result and the criticality of each historical symptom as training related parameters;
calculating a predicted treatment result corresponding to each historical symptom based on the preset weight corresponding to each training related parameter;
obtaining actual treatment results corresponding to each historical symptom, comparing the actual treatment results with the predicted treatment results, and determining target training symptoms in each historical symptom;
and training a matching model based on the historical medical history data and the historical symptom data corresponding to the target training symptom to obtain the preset matching model.
6. The method for matching a doctor to a patient based on a knowledge graph of symptoms according to claim 5, wherein the obtaining the actual treatment results corresponding to each of the historical symptoms, comparing the actual treatment results with the predicted treatment results, and determining the target training symptom among the historical symptoms comprises:
when the predicted treatment result corresponding to the historical symptom is not matched with the actual treatment result, the historical symptom is used as a training correction symptom;
and when the historical symptoms correspond to the predicted treatment result and the actual treatment result and are matched, determining that the historical symptoms are the target training symptoms.
7. The method for matching a doctor with a patient based on a symptom knowledge graph according to claim 1, wherein the matching the current symptom based on the symptom knowledge graph to obtain at least one target-related parameter corresponding to the current symptom comprises:
obtaining at least one historical symptom hit by the current symptom in the symptom knowledge graph;
and acquiring the hit frequency of the current symptom hitting the historical symptom, and/or the target symptom grade corresponding to the current symptom, and/or acquiring the treatment result corresponding to each historical symptom, and/or the critical sequencing result of each historical symptom hit by the current symptom, as the target related parameter.
8. A doctor-patient matching device based on a symptom knowledge graph, characterized in that the doctor-patient matching device based on the symptom knowledge graph comprises:
the current symptom acquisition module is used for acquiring the current symptom and the current position area of the current patient;
the symptom matching module is used for matching the current symptom based on a symptom knowledge graph to obtain at least one target related parameter corresponding to the current symptom;
the target related parameters comprise hit frequencies of each historical symptom hit by the current symptom in a symptom knowledge graph, and/or target symptom grades corresponding to the current symptom, and/or treatment results corresponding to each historical symptom hit by the current symptom in the symptom knowledge graph, and/or critical sequencing results of each historical symptom hit by the current symptom in the symptom knowledge graph;
the cure ordering module is used for calculating the cure probability of each hospital for the current symptom in the current position area based on a preset matching model and preset weights corresponding to the relevant parameters of each target, and obtaining the cure ordering of each hospital for the current symptom;
the preset matching model is used for calculating the cure probability of each hospital for the current symptom according to each target related parameter and each preset weight corresponding to each target related parameter, and sequencing each hospital according to the cure probability;
and the target hospital determining module is used for determining a target hospital matched with the current patient in each hospital in the current position area based on the cure sequence of each hospital for the current symptom.
9. A computer device comprising a processor, a memory, and a computer program stored on the memory and executable by the processor, wherein the computer program when executed by the processor performs the steps of the symptomatic knowledge-graph based doctor-patient matching method of any one of claims 1 to 7.
10. A computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, wherein the computer program, when executed by a processor, implements the steps of the symptomatic knowledge-graph based doctor-patient matching method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211394235.4A CN115662593B (en) | 2022-11-08 | 2022-11-08 | Doctor-patient matching method, device, equipment and medium based on symptom knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211394235.4A CN115662593B (en) | 2022-11-08 | 2022-11-08 | Doctor-patient matching method, device, equipment and medium based on symptom knowledge graph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115662593A CN115662593A (en) | 2023-01-31 |
| CN115662593B true CN115662593B (en) | 2023-06-23 |
Family
ID=85016766
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211394235.4A Active CN115662593B (en) | 2022-11-08 | 2022-11-08 | Doctor-patient matching method, device, equipment and medium based on symptom knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115662593B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110085307A (en) * | 2019-04-04 | 2019-08-02 | 华东理工大学 | A kind of intelligent hospital guide's method and system based on the fusion of multi-source knowledge mapping |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110993078A (en) * | 2019-11-27 | 2020-04-10 | 华中科技大学同济医学院附属协和医院 | Medical triage method, device and storage medium |
| CN111785367A (en) * | 2020-06-30 | 2020-10-16 | 平安科技（深圳）有限公司 | Triage method and device based on neural network model and computer equipment |
| CN113656601A (en) * | 2021-08-30 | 2021-11-16 | 平安医疗健康管理股份有限公司 | Doctor-patient matching method, device, equipment and storage medium |
| CN113658712A (en) * | 2021-08-31 | 2021-11-16 | 平安医疗健康管理股份有限公司 | Doctor-patient matching method, device, equipment and storage medium |
-
2022
- 2022-11-08 CN CN202211394235.4A patent/CN115662593B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110085307A (en) * | 2019-04-04 | 2019-08-02 | 华东理工大学 | A kind of intelligent hospital guide's method and system based on the fusion of multi-source knowledge mapping |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115662593A (en) | 2023-01-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107743630B (en) | Predicting likelihood of satisfying a condition using a recurrent neural network | |
| US11488718B2 (en) | Computer aided medical method and medical system for medical prediction | |
| CN107851462B (en) | Analyzing health events using a recurrent neural network | |
| Gartner et al. | Machine learning approaches for early DRG classification and resource allocation | |
| US20220044809A1 (en) | Systems and methods for using deep learning to generate acuity scores for critically ill or injured patients | |
| US20210398674A1 (en) | Method for providing diagnostic system using semi-supervised learning, and diagnostic system using same | |
| WO2021179630A1 (en) | Complications risk prediction system, method, apparatus, and device, and medium | |
| CN110111885B (en) | Attribute prediction method, attribute prediction device, computer equipment and computer readable storage medium | |
| CN114298234B (en) | Brain medical image classification method and device, computer equipment and storage medium | |
| US20210174968A1 (en) | Visualization of Social Determinants of Health | |
| US20220172805A1 (en) | System and method for automatically determining serious adverse events | |
| US20200286599A1 (en) | Experience Engine-Method and Apparatus of Learning from Similar Patients | |
| CN116312934A (en) | Medical service recommendation scheme generation method, device, equipment and readable storage medium | |
| CN115662593B (en) | Doctor-patient matching method, device, equipment and medium based on symptom knowledge graph | |
| CN115719640B (en) | Chinese medicine primary and secondary symptom recognition system, device, electronic equipment and storage medium thereof | |
| CN109102845B (en) | Medical document auditing method, device, computer equipment and storage medium | |
| CN116825359A (en) | VTE risk early warning method, system, electronic equipment and computer readable medium | |
| WO2023050668A1 (en) | Clustering model construction method based on causal inference and medical data processing method | |
| CN114203306A (en) | Medical event prediction model training method, medical event prediction method and device | |
| CN115631823A (en) | Similar case recommendation method and system | |
| US10335092B1 (en) | Building predicted future medical profiles | |
| CN117497158B (en) | Method for calculating recommended triage grade based on dimension combination and weight | |
| EP3961648A1 (en) | Computing device for estimating the probability of myocardial infarction | |
| CN117153378B (en) | Diagnosis guiding method and device, electronic equipment and storage medium | |
| US20230020543A1 (en) | Learning system, learning device, learning method, and storage medium |
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 |