A case scoring method, device, equipment and computer readable storage medium

Abstract

The embodiment of the present application discloses a case scoring method, which includes: obtaining case information of a pending case and a doctor-patient dialogue message corresponding to the pending case; wherein the case information represents information generated after the doctor diagnoses the patient; extracting keywords from the case information and the doctor-patient dialogue message to determine target keywords; obtaining the type of the pending case, and determining the target vocabulary corresponding to the pending case based on the type of the pending case; matching the target keyword with the target vocabulary, and determining multiple scores of the pending case based on the matching results. The embodiment of the present application also discloses a case scoring device, an apparatus, and a computer-readable storage medium.

Description

Technical Field

[0001] The present application relates to case scoring technology in the field of case review, and in particular to a case scoring method, device, equipment and computer-readable storage medium. Background Art

[0002] With the development of Internet technology, it has become a common phenomenon to seek medical treatment through the Internet. Patients do not need to go to the hospital, but can register online and communicate with doctors in Internet hospitals. Doctors can provide patients with diagnosis and treatment plans based on the content of the patient's communication. Therefore, how to review the diagnosis and treatment plans provided by doctors to patients is particularly important. In related technologies, the diagnosis and treatment plans provided by doctors to patients are generally reviewed manually (other medical workers), but the accuracy and efficiency of manual review are low. Summary of the invention

[0003] In order to solve the above technical problems, the embodiments of the present application hope to provide a case scoring method, device, equipment and computer-readable storage medium, which solves the problem of low accuracy and low efficiency in manual review of pending cases.

[0004] The technical solution of this application is implemented as follows:

[0005] A case scoring method, the method comprising:

[0006] Obtaining case information of a pending case and a doctor-patient dialogue message corresponding to the pending case; wherein the case information represents information generated after the doctor diagnoses the patient;

[0007] Extract keywords from the case information and the doctor-patient conversation messages to determine target keywords;

[0008] Acquire the type of the case to be processed, and determine the target vocabulary corresponding to the case to be processed based on the type of the case to be processed;

[0009] The target keyword is matched with the target word library, and multiple scores of the case to be processed are determined based on the matching results.

[0010] In the above scheme, the step of obtaining the case information of the case to be processed and the doctor-patient dialogue message corresponding to the case to be processed includes:

[0011] Obtaining the medical questionnaire identification of the case to be processed, and obtaining the case information of the case to be processed from a case information database based on the medical questionnaire identification;

[0012] Based on the case information, determining identification information of the doctor and identification information of the patient;

[0013] Based on the identification information of the doctor and the identification information of the patient, a doctor-patient dialogue message of the case to be processed is obtained from a doctor-patient dialogue platform.

[0014] In the above solution, extracting keywords from the case information and the doctor-patient dialogue message to determine target keywords includes:

[0015] Extracting the doctor's dialogue message from the doctor-patient dialogue message based on the doctor's identification information, and extracting the patient's dialogue message from the doctor-patient dialogue message based on the patient's identification information;

[0016] Extracting keywords from the doctor's conversation message to obtain a first keyword, and extracting keywords from the patient's conversation message to obtain a second keyword;

[0017] Acquiring prescription information, recommendation information, and diagnosis information based on the case information, and performing keyword extraction on the prescription information to obtain a third keyword;

[0018] Keyword extraction is performed on the suggestion information to obtain a fourth keyword, and keyword extraction is performed on the diagnosis information to obtain a fifth keyword; wherein the target keywords include: the first keyword, the second keyword, the third keyword, the fourth keyword and the fifth keyword.

[0019] In the above scheme, the target keyword is matched with the target word library, and multiple scores of the case to be processed are determined based on the matching results, including:

[0020] Matching the first keyword with the keyword corresponding to the symptom label in the target vocabulary to obtain a first matching result, and determining the consultation score of the case to be processed based on the first matching result;

[0021] Determining a disease keyword from the target word library based on the second keyword, and matching the fifth keyword with the disease keyword to obtain a second matching result;

[0022] Determining a possible diagnosis score of the case to be processed based on the second matching result;

[0023] Determining a drug keyword from the target word library based on the second keyword, and matching the third keyword with the drug keyword to obtain a third matching result;

[0024] Determine the medication score of the case to be processed based on the third matching result;

[0025] The fourth keyword is matched with the keyword corresponding to the suggested tag in the target vocabulary to obtain a fourth matching result, and a suggested score for the case to be processed is determined based on the fourth matching result; wherein the multiple scores include at least: the consultation score, the proposed diagnosis score, the medication score and the suggested score.

[0026] In the above solution, determining the disease keyword from the target word library based on the second keyword includes:

[0027] Acquire a target symptom keyword matching the second keyword from keywords corresponding to the symptom tag in the target vocabulary;

[0028] Based on the target symptom keyword and the first association relationship, the disease keyword associated with the target symptom keyword is determined from the target vocabulary; wherein the first association relationship is an association relationship between symptoms and diseases.

[0029] In the above solution, the step of determining the drug keyword from the target word library based on the second keyword includes:

[0030] Based on the target symptom keyword and the second association relationship, determining a drug keyword associated with the target symptom keyword from the target word library; wherein the second association relationship is an association relationship between symptoms and drugs;

[0031] Accordingly, determining the medication score of the case to be processed based on the third matching result includes:

[0032] In a case where the third matching result indicates that the third keyword does not match the drug keyword, determining the medication score;

[0033] When the third matching result indicates that the third keyword matches the drug keyword and the drug type is the first type, obtaining a first matching degree between the third keyword and the drug keyword, and determining the medication score based on the first matching degree;

[0034] When the third matching result indicates that the third keyword matches the drug keyword, and the drug type is the second type, obtaining the first matching degree, and determining the drug symptomatic score based on the first matching degree;

[0035] Based on the third keyword and the third association relationship, determining a prompt keyword associated with the third keyword from the target word library, and determining a second matching degree between the first keyword and the prompt keyword; wherein the third association relationship is an association relationship between a drug and a prompt suggestion;

[0036] A drug inquiry score is determined based on the second matching degree, and the medication score is determined based on the drug inquiry score and the drug symptomatic score.

[0037] In the above solution, determining the recommended score for the case to be processed based on the fourth matching result includes:

[0038] In a case where the fourth matching result indicates that the fourth keyword does not match the keyword corresponding to the suggested tag in the target vocabulary, determining a suggested score for the case to be processed;

[0039] In the case where the fourth matching result indicates that the fourth keyword matches the keyword corresponding to the suggestion tag in the target vocabulary, based on the target symptom keyword and the fourth association relationship, determining the suggestion keyword associated with the target symptom keyword from the target vocabulary; wherein the fourth association relationship is an association relationship between symptoms and prompt suggestions;

[0040] The fourth keyword is matched with the suggested keyword to obtain a third matching degree, and a suggested score for the case to be processed is determined based on the third matching degree.

[0041] In the above scheme, the method further comprises:

[0042] Get query information;

[0043] Based on the identifier of the case to be queried carried in the query information, query whether there is a user score in the case scoring database; wherein the user score is a score given by the user to review the case to be processed; wherein the case scoring database stores multiple scores of the case to be processed;

[0044] If the user score exists in the case score database, determining the user score to be the score corresponding to the identifier of the case to be queried;

[0045] In the case that the case scoring database does not have the user score, the multiple scores of the case to be processed are determined as the scores corresponding to the identifiers of the case to be queried.

[0046] In the above solution, the consultation score, the proposed diagnosis score, the medication score and the recommendation score are stored in the case scoring database using different fields.

[0047] A case scoring device, comprising:

[0048] An acquisition unit, used to acquire case information of a case to be processed and a doctor-patient dialogue message corresponding to the case to be processed; wherein the case information represents information generated after the doctor diagnoses the patient;

[0049] A determination unit, configured to extract keywords from the case information and the doctor-patient dialogue message to determine target keywords;

[0050] The acquisition unit is further used to acquire the type of the case to be processed, and determine the target vocabulary corresponding to the case to be processed based on the type of the case to be processed;

[0051] The determination unit is further configured to match the target keyword with the target vocabulary, and determine multiple scores of the case to be processed based on the matching results.

[0052] A case scoring device, the device comprising: a processor, a memory and a communication bus;

[0053] The communication bus is used to realize the communication connection between the processor and the memory;

[0054] The processor is used to execute the case scoring program in the memory to implement the following steps:

[0055] Obtaining case information of a pending case and a doctor-patient dialogue message corresponding to the pending case; wherein the case information represents information generated after the doctor diagnoses the patient;

[0056] Extract keywords from the case information and the doctor-patient conversation messages to determine target keywords;

[0057] Acquire the type of the case to be processed, and determine the target vocabulary corresponding to the case to be processed based on the type of the case to be processed;

[0058] The target keyword is matched with the target word library, and multiple scores of the case to be processed are determined based on the matching results.

[0059] A computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the case scoring method described above.

[0060] Because the case information of the pending case and the doctor-patient dialogue messages corresponding to the pending case can be obtained; wherein the case information represents the information generated after the doctor diagnoses the patient; then the case information and the doctor-patient dialogue messages are subjected to keyword extraction to determine the target keyword; the type of the pending case is obtained, and the target vocabulary corresponding to the pending case is determined based on the type of the pending case; the target keyword and the target vocabulary are matched, and multiple scores of the pending case are determined based on the matching results; in this way, the multiple scores of the pending case can be automatically determined based on the case information of the pending case and the doctor-patient dialogue messages corresponding to the pending case, and there is no need to manually review the pending case to obtain the multiple scores of the pending case as in the related art, which improves the efficiency of determining the multiple scores of the pending case, and avoids the subjectivity of manual review that causes the multiple scores of the pending case to be inaccurate, further improving the accuracy of the determined multiple scores of the pending case. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A schematic diagram of a process flow of a case scoring method according to an embodiment of the present application;

[0062] Figure 2 A schematic diagram of another case scoring method provided in an embodiment of the present application;

[0063] Figure 3 A schematic diagram of a process flow of a case scoring system provided in an embodiment of the present application;

[0064] Figure 4 A flowchart of another case scoring method provided in an embodiment of the present application;

[0065] Figure 5 A schematic diagram of the structure of a case scoring device provided in an embodiment of the present application;

[0066] Figure 6 A schematic diagram of the structure of a case scoring device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0067] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0068] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0069] The present application embodiment provides a case scoring method, which is applied to a case scoring device, such as Figure 1 As shown, the method comprises the following steps:

[0070] Step 101: Obtain case information of a pending case and a doctor-patient dialogue message corresponding to the pending case.

[0071] Among them, case information represents the information generated after the doctor diagnoses the patient.

[0072] In an embodiment of the present application, a pending case may be an event in which a doctor provides medical services to a patient; the case information of the pending case is not limited to: patient information, doctor information, prescription information, diagnosis information, and advice information. The case information of the pending case may be actively obtained by the case scoring device from the case system of the Internet hospital; it may also be actively sent to the case scoring device by the case system after the consultation. Doctor conversation messages refer to conversation messages between doctors and patients involved in the pending case. Among them, the types of doctor-patient conversation messages may include at least one of audio and text.

[0073] It should be noted that if the type of the doctor-patient dialogue message is audio, the doctor-patient dialogue message can be semantically recognized to obtain a text-type doctor-patient dialogue message, so that the text-type doctor-patient dialogue message can be subsequently analyzed.

[0074] Step 102: extract keywords from case information and doctor-patient conversation messages to determine target keywords.

[0075] In the embodiment of the present application, a keyword extraction model can be used to process the sentences in the case information and the doctor-patient dialogue messages, and the words output by the keyword extraction model are used as target keywords. The keyword extraction model is a model obtained by training the sample keywords using a keyword extraction algorithm. The sample keywords include at least symptom keywords, disease keywords, drug keywords, and suggestion keywords.

[0076] Step 103: Obtain the type of the case to be processed, and determine the target vocabulary corresponding to the case to be processed based on the type of the case to be processed.

[0077] The type of the pending case is the type of service provided by the doctor to the patient; the type of the pending case can be actively sent to the case scoring device by the case system, or can be obtained from the case scoring device by the case scoring device initiating an acquisition instruction. The types of pending cases include but are not limited to disease consultation, drug consultation, life health and physical examination reports.

[0078] In an embodiment of the present application, the type of the case to be processed can be matched with multiple vocabulary types, a target vocabulary type that matches the type of the case to be processed can be determined among the multiple vocabulary types, and the target vocabulary can be determined as the vocabulary corresponding to the target vocabulary type.

[0079] Step 104: Match the target keyword with the target vocabulary, and determine multiple scores for the case to be processed based on the matching results.

[0080] Among them, the target vocabulary includes a keyword set corresponding to each of the multiple tags; the tags include at least: symptom tags and suggestion tags; the keyword set corresponding to the symptom tags has multiple symptom keywords; the keyword set corresponding to the suggestion tags has multiple suggestion keywords.

[0081] In the embodiment of the present application, the target keyword can be matched with the keyword in each keyword set, and the matching result matching each keyword set can be obtained, and multiple scores of the pending cases can be determined based on the multiple matching results. The comprehensive consultation ability of the doctor can be evaluated by determining the multiple scores of the pending cases, so as to encourage the doctor to improve the quality of consultation, reduce the risk of online diagnosis and treatment, and further enhance the consultation experience of the user.

[0082] It should be noted that if the scoring options and scores of pending cases need to be adjusted and optimized subsequently, there is no need to train the personnel (doctors) who review the pending cases and then perform manual adjustment and optimization. Instead, the case scoring device automatically adjusts and optimizes the scoring options and scores of the pending cases by running the adjustment and optimization program, thereby improving the efficiency of adjusting and optimizing the scoring options and scores of pending cases.

[0083] The case scoring method provided in the embodiment of the present application obtains case information of the case to be processed and the doctor-patient dialogue message corresponding to the case to be processed; wherein the case information represents the information generated after the doctor diagnoses the patient; keyword extraction is performed on the case information and the doctor-patient dialogue message to determine the target keyword; the type of the case to be processed is obtained, and the target vocabulary corresponding to the case to be processed is determined based on the type of the case to be processed; the target keyword and the target vocabulary are matched, and multiple scores of the case to be processed are determined based on the matching results; in this way, multiple scores of the case to be processed can be automatically determined based on the case information of the case to be processed and the doctor-patient dialogue message corresponding to the case to be processed, and there is no need to manually review the case to be processed to obtain multiple scores of the case to be processed as in the related art, which improves the efficiency of determining the multiple scores of the case to be processed, and avoids the subjectivity of manual review that causes inaccurate multiple scores of the case to be processed, further improving the accuracy of the determined multiple scores of the case to be processed.

[0084] Based on the above embodiments, the present application provides a case scoring method, such as Figure 2 As shown, the method includes:

[0085] Step 201: The case scoring device obtains the medical questionnaire identifier of the case to be processed, and obtains the case information of the case to be processed from the case information database based on the medical questionnaire identifier.

[0086] Among them, the medical consultation form ID is automatically generated after the patient places an order in the Internet hospital; the medical consultation form ID can uniquely refer to a medical consultation form, which includes patient information, doctor information, and various medical condition analyses and medical diagnosis information.

[0087] In the embodiment of the present application, the medical questionnaire information can be obtained from the medical questionnaire database based on the medical questionnaire identifier, and the pending case identifier associated with the medical questionnaire information can be obtained, and the case information of the pending case can be obtained from the case information database based on the pending case identifier. The pending case is a case generated after the medical consultation is completed; the identifier of the pending case is obtained by marking the generated case, and the identifier of the pending case can uniquely refer to the pending case.

[0088] It should be noted that the prerequisite for obtaining the case information of the pending case is that the pending case can be generated after the consultation is completed, and whether the case can be generated can be determined by the type of the medical questionnaire and / or the doctor information on the medical questionnaire. In a feasible implementation, the case can be generated when the type of the medical questionnaire is consistent with the type of the preset medical questionnaire, and the doctor information indicates that the doctor is a full-time doctor, so as to filter out medical questionnaires that have no reference value according to the preset medical questionnaire type and doctor information, and then generate cases with reference value, and the number of generated cases can also be reduced, further reducing the workload of the case scoring device in reviewing the cases.

[0089] Step 202: The case scoring device determines the identification information of the doctor and the identification information of the patient based on the case information.

[0090] Among them, case information includes doctor information and patient information.

[0091] In an embodiment of the present application, the doctor information and the patient information may be analyzed separately to determine the doctor's identification information and the patient's identification information. The doctor's identification information can uniquely refer to the doctor, and the patient's identification information can uniquely refer to the patient.

[0092] In a feasible implementation, the identification information of the doctor may be the doctor's identity document (ID), and the identification information of the patient may be the patient's ID;

[0093] Step 203: The case scoring device obtains the doctor-patient dialogue message of the pending case from the doctor-patient dialogue platform based on the doctor's identification information and the patient's identification information.

[0094] Among them, the doctor-patient dialogue platform refers to a platform used for communication between doctors and patients, and the doctor-patient dialogue platform stores doctor-patient dialogue messages; among them, the type of doctor-patient dialogue messages is text; the doctor-patient dialogue messages of the pending cases are dialogue messages between doctors and patients when patients visit the hospital.

[0095] It should be noted that the doctor-patient dialogue message can be information directly collected by the doctor-patient dialogue platform, or it can be information collected by the doctor's corresponding doctor terminal and sent to the doctor-patient dialogue platform. If the type of the doctor-patient dialogue message collected by the doctor dialogue platform or the doctor's corresponding doctor terminal is audio, the doctor-patient dialogue message can be voiceprint recognized, and the doctor's audio part and the patient's audio part in the audio can be determined, and then the doctor's audio part and the patient's audio part can be semantically recognized to obtain the doctor's dialogue message and the patient's dialogue message, and the doctor's dialogue message and the patient's dialogue message are stored in the doctor-patient dialogue platform.

[0096] Step 204: The case scoring device extracts the doctor's conversation message from the doctor-patient conversation message based on the doctor's identification information, and extracts the patient's conversation message from the doctor-patient conversation message based on the patient's identification information.

[0097] In an embodiment of the present application, it is necessary to separate the doctor's conversation messages and the patient's conversation messages in the doctor-patient conversation messages so that the doctor's conversation messages and the patient's conversation messages can be processed separately subsequently to determine multiple scores for the case to be processed.

[0098] Step 205: The case scoring device extracts keywords from the doctor's conversation message to obtain a first keyword, and extracts keywords from the patient's conversation message to obtain a second keyword.

[0099] In an embodiment of the present application, the sentences in the doctor's conversation message can be input into the keyword extraction model, and the words output by the keyword extraction model can be used as the first keyword; and the sentences in the patient's conversation message can be input into the keyword extraction model, and the words output by the keyword extraction model can be used as the second keyword.

[0100] Step 206: The case scoring device obtains prescription information, recommendation information, and diagnosis information based on the case information, and extracts keywords from the prescription information to obtain a third keyword.

[0101] Among them, prescription information is the drug-related information prescribed by the doctor based on the patient's condition; recommendation information is the recommendation information given by the doctor based on the patient's symptoms; and diagnosis information is the disease information determined by the doctor based on the patient's symptoms.

[0102] In an embodiment of the present application, case information can be analyzed to obtain prescription information, recommendation information, and diagnosis information, and sentences in the prescription information can be processed using target keywords to obtain third keywords.

[0103] Step 207: The case scoring device extracts keywords from the suggestion information to obtain a fourth keyword, and extracts keywords from the diagnosis information to obtain a fifth keyword.

[0104] Among them, the target keywords include: the first keyword, the second keyword, the third keyword, the fourth keyword and the fifth keyword.

[0105] In an embodiment of the present application, a keyword extraction model can be used to process sentences in prescription information, and the words output by the keyword extraction model can be used as the third keyword; a keyword extraction model can be used to process sentences in suggestion information, and the words output by the keyword extraction model can be used as the fourth keyword; a keyword extraction model can be used to process sentences in diagnosis information, and the words output by the keyword extraction model can be used as the fifth keyword.

[0106] Step 208: The case scoring device obtains the type of the case to be processed, and determines a target vocabulary corresponding to the case to be processed based on the type of the case to be processed.

[0107] Step 209: The case scoring device matches the first keyword with the keyword corresponding to the symptom label in the target vocabulary to obtain a first matching result, and determines the consultation score of the case to be processed based on the first matching result.

[0108] Among them, the first keyword is the keyword extracted from the doctor's conversation message; among them, the keyword corresponding to the symptom label is the symptom keyword; and the target vocabulary includes keywords of multiple labels.

[0109] In the embodiment of the present application, if the first keyword matches the symptom keyword in the target word library, it indicates that the first keyword contains the symptom keyword, and the first consultation score of the case to be processed can be determined. If the first keyword does not match the symptom keyword in the target word library, the second consultation score of the case to be processed can be determined. The first consultation score can be set to 0 points; the second consultation score can be set to -8 points.

[0110] In a feasible implementation, during the consultation, the doctor will send a message to the patient saying "You have fever, cough and other symptoms now, please confirm whether it is consistent with your actual condition." The case scoring device can extract the keywords "fever" and "cough" from the doctor's conversation message, and match "fever" and "cough" with the keywords with symptom labels in the target vocabulary to obtain the matching degree. If the matching degree is not 0, the consultation score of the case to be processed can be determined to be 0 points. If the matching degree is 0, the consultation score of the case to be processed can be determined to be -8 points. Among them, the scores corresponding to different matching degrees can be preset.

[0111] Step 210: The case scoring device determines a disease keyword from the target vocabulary based on the second keyword, and matches the fifth keyword with the disease keyword to obtain a second matching result.

[0112] The disease keyword may be a set of disease keywords corresponding to the patient's symptoms determined from a target database based on the patient's symptoms; the number of disease keywords may be one or more.

[0113] In an embodiment of the present application, the second keyword can be matched with the keywords in the target vocabulary to determine the keywords in the target vocabulary that match the second keyword, and based on the keywords that match the second keyword, the disease keyword can be determined from the target vocabulary, and then the fifth keyword can be matched with the determined disease keyword to obtain a second matching result; wherein the second matching result represents the matching degree between the fifth keyword and the disease keyword; the fifth keyword is a keyword that represents the patient's disease diagnosed by the doctor.

[0114] Wherein, determining the disease keyword from the target word library based on the second keyword in step 210 can be implemented by the following steps:

[0115] A1. The case scoring device obtains a target symptom keyword that matches the second keyword from the keywords corresponding to the symptom label in the target vocabulary.

[0116] Among them, the keyword corresponding to the symptom label is the symptom keyword, and the number of symptom keywords can be multiple.

[0117] In the embodiment of the present application, the second keyword can be matched with multiple symptom keywords corresponding to the symptom label to determine the symptom keyword matching the second keyword to obtain the target symptom keyword. That is, the symptom keyword (target symptom keyword) corresponding to the patient's self-reported symptoms is determined by matching the second keyword with multiple symptom keywords corresponding to the symptom label.

[0118] A2. The case scoring device determines a disease keyword associated with the target symptom keyword from the target vocabulary based on the target symptom keyword and the first association relationship.

[0119] Among them, the first correlation is the correlation between symptoms and diseases.

[0120] In a feasible implementation, if the target symptom keyword is "fever", the disease corresponding to "fever" can be queried from the target vocabulary based on "fever" and the first association relationship, where the disease corresponding to fever can be febrile influenza.

[0121] Step 211: The case scoring device determines a possible diagnosis score of the case to be processed based on the second matching result.

[0122] In an embodiment of the present application, if the second matching result represents that the fifth keyword does not match any keyword in the determined disease keyword set at all, that is, the matching degree is 0, then the tentative diagnosis score of the case to be processed can be determined as the first tentative diagnosis score; if the second matching result represents that the matching degree of the fifth keyword and the determined disease keyword is greater than the preset disease matching degree, the tentative diagnosis score of the case to be processed is determined to be the second tentative diagnosis score; if the second matching result represents that the matching degree of the fifth keyword and the determined disease keyword is greater than 0 and less than or equal to the preset disease matching degree, the tentative diagnosis score of the case to be processed is determined to be the third tentative diagnosis score.

[0123] In a feasible implementation, the first tentative diagnosis score may be -8 points, the second tentative diagnosis score may be 0 points, and the third tentative diagnosis score may be -4 points.

[0124] Step 212: The case scoring device determines a drug keyword from the target vocabulary based on the second keyword, and matches the third keyword with the drug keyword to obtain a third matching result.

[0125] Among them, the drug keywords may include but are not limited to the name of the drug; the drug keywords may be the information of the drug used to treat the patient's condition determined by the case scoring device based on the second keyword; the third keyword is the information of the drug used to treat the patient's condition determined by the doctor; the number of drug keywords can be multiple; the third matching result represents the matching degree between the drug keyword and the third keyword.

[0126] In the embodiment of the present application, determining the drug keyword from the target word library based on the second keyword in step 212 can also be achieved by the following steps:

[0127] B1. The case scoring device determines the drug keyword associated with the target symptom keyword from the target vocabulary based on the target symptom keyword and the second association relationship.

[0128] Among them, the second correlation is the correlation between symptoms and drugs.

[0129] In a feasible implementation, if the target symptom keyword is "fever", the information of medicines corresponding to "fever" can be queried from the target vocabulary based on "fever" and the second association relationship, wherein the information of medicines corresponding to fever can be the name of the medicines corresponding to fever.

[0130] Step 213: The case scoring device determines the medication score of the case to be processed based on the third matching result.

[0131] In the embodiment of the present application, different medication scores for the cases to be processed can be determined based on the matching degree between the drug keyword represented by the third matching result and the third keyword, and based on different matching degrees.

[0132] Wherein, step 213 can be implemented by the following steps:

[0133] B2. The case scoring device determines a medication score when the third matching result indicates that the third keyword does not match the drug keyword.

[0134] In the embodiment of the present application, if the third keyword does not match the drug keyword, it means that the information of the drug determined by the doctor for treating the patient's condition does not correspond to the patient's symptoms, that is, the drug is not suitable for the disease, and the drug score can be directly determined to be the lowest score. In a feasible implementation, the drug score can be -8 points when the drug is not suitable for the disease.

[0135] B3. When the third matching result indicates that the third keyword matches the drug keyword and the drug type is the first type, the case scoring device obtains a first matching degree between the third keyword and the drug keyword, and determines a medication score based on the first matching degree.

[0136] Among them, the first type indicates that the drug determined by the doctor to be used to treat the patient's condition is a non-allergic drug, that is, the pre-drug inquiry score of the drug is not considered, and the drug symptomatic score determined according to the first matching degree is used as the medication score; wherein the first matching degree is greater than 0.

[0137] In an embodiment of the present application, when the third keyword matches the drug keyword and the drug type is the first type, it indicates that the information about the drug determined by the doctor for treating the patient's condition is symptomatic of the patient's symptoms. The degree of symptomatic response of the information about the drug determined by the doctor for treating the patient's condition to the patient's symptoms can be determined based on the first degree of matching, and different medication scores can be determined based on different degrees of symptomatic response.

[0138] In a feasible implementation, when the third keyword matches the drug keyword and the drug type is the first type, if the first matching degree is greater than the preset medication matching degree, the medication score is determined to be the first medication score; if the first matching degree is less than or equal to the preset medication matching degree, the medication score is determined to be the second medication score; wherein, the first medication score can be 0 points and the second medication score can be -4 points.

[0139] B4. When the third matching result indicates that the third keyword matches the drug keyword and the drug type is the second type, the case scoring device obtains a first matching degree and determines a drug symptomatic score based on the first matching degree.

[0140] B5. The case scoring device determines, based on the third keyword and the third association relationship, a prompt keyword associated with the third keyword from the target vocabulary, and determines a second matching degree between the first keyword and the prompt keyword.

[0141] Among them, the third association relationship is the association relationship between drugs and prompt suggestions;

[0142] It should be noted that if the type of drug is the second type, it means that the drug determined by the doctor to be used to treat the patient's condition is an allergic drug. Therefore, when determining the medication score, it is necessary to consider not only the drug symptomatic score but also the drug's pre-drug inquiry score; among them, the pre-drug inquiry score is used to determine whether the doctor has asked or reminded the patient when prescribing allergy drugs; the prompt keyword is the prompt information corresponding to allergy drugs.

[0143] In an embodiment of the present application, if the second matching degree is 0, it indicates that the doctor did not prompt or ask the patient before prescribing allergy drugs; if the second matching degree is not 0, it indicates that the doctor has prompted or asked the patient, and the accuracy of the prompt information given by the doctor to prompt or ask the patient can be further determined based on the second matching degree, and the pre-drug inquiry score can be determined based on the accuracy.

[0144] B6. The case scoring device determines a drug inquiry score based on the second matching degree, and determines a medication score based on the drug inquiry score and the drug symptomatic score.

[0145] In an embodiment of the present application, the pre-drug inquiry score corresponding to the second matching degree can be determined based on the second matching degree, and the pre-drug inquiry score and the drug's symptomatic score can be averaged to determine the medication score; the medication score can also be determined by taking the highest score from the pre-drug inquiry score and the drug's symptomatic score; the medication score can also be determined based on the weights of the pre-drug inquiry score and the drug's symptomatic score.

[0146] The pre-drug inquiry scores corresponding to different second matching degrees may be preset.

[0147] In a feasible implementation, if the second matching degree is 0, the drug inquiry score can be determined to be -8 points; if the second matching degree is greater than the preset drug inquiry matching degree, the drug inquiry score is determined to be 0 points; if the second matching degree is less than or equal to the preset drug inquiry matching degree, the drug inquiry score is determined to be -4 points. If the drug's symptomatic score is -4 points and the pre-drug inquiry score is -8 points, the medication score can be determined to be -6 points.

[0148] Step 214: The case scoring device matches the fourth keyword with the keyword corresponding to the suggested tag in the target vocabulary to obtain a fourth matching result, and determines a suggested score for the case to be processed based on the fourth matching result.

[0149] Among them, the multiple scores include at least: consultation score, diagnosis score, medication score and suggestion score. The keywords corresponding to the suggestion tag are suggestion keywords.

[0150] In an embodiment of the present application, the fourth keyword is matched with the keyword corresponding to the suggested tag in the target vocabulary to obtain a fourth matching result. If the fourth matching result indicates that the matching degree between the fourth keyword and the keyword corresponding to the suggested tag in the target vocabulary is 0, it indicates that the doctor has not given any advice regarding the patient's symptoms. If the fourth matching result indicates that the matching degree between the fourth keyword and the keyword corresponding to the suggested tag in the target vocabulary is not 0, it indicates that the doctor has given advice regarding the patient's symptoms. At this time, it is necessary to further determine whether the advice given by the doctor corresponds to the patient's symptoms based on the matching degree.

[0151] Determining the recommended score of the case to be processed based on the fourth matching result in step 214 can also be achieved by the following steps:

[0152] C1. The case scoring device determines a recommended score for the case to be processed when the fourth matching result indicates that the fourth keyword does not match the keyword corresponding to the recommended tag in the target vocabulary.

[0153] In the embodiment of the present application, when the fourth keyword does not match the keyword corresponding to the suggestion tag in the target word library, it indicates that the doctor has not given any suggestions for the patient's symptoms, and the first suggestion score is determined. The first suggestion score may be -8 points.

[0154] C2. When the fourth matching result indicates that the fourth keyword matches the keyword corresponding to the suggested tag in the target vocabulary, the case scoring device determines a suggested keyword associated with the target symptom keyword from the target vocabulary based on the target symptom keyword and the fourth association relationship.

[0155] Among them, the fourth association relationship is the association relationship between symptoms and prompt suggestions; the suggested keywords are words representing the doctor's suggestions to the patient, which are determined by the case scoring device from the target vocabulary based on the patient's symptoms and correspond to the patient's symptoms.

[0156] C3. The case scoring device matches the fourth keyword with the suggested keyword to obtain a third matching degree, and determines a suggested scoring of the case to be processed based on the third matching degree.

[0157] In an embodiment of the present application, whether the advice given by the doctor corresponds to the patient's symptoms is determined based on the third matching degree. If the third matching degree is greater than a preset suggestion threshold, a second suggestion score is determined, and the second suggestion score represents that the advice given by the doctor corresponds to the patient's symptoms. If the third matching degree is less than or equal to the preset suggestion threshold, a third suggestion score is determined, wherein the third suggestion score represents that the advice given by the doctor does not correspond to the patient's symptoms.

[0158] In a feasible implementation, the second suggestion score is 0 points, and the second suggestion score is -4 points.

[0159] Step 215: The case scoring device obtains query information.

[0160] The query information refers to information used to query multiple scores of the case to be processed.

[0161] In the embodiment of the present application, the query information may be information carried in a query request initiated by other doctors or reviewers through a terminal. The query information may include an identifier of the case to be queried, and the identifier of the case to be queried may be the ID of the case to be queried.

[0162] Step 216: The case scoring device queries whether there is a user score in the case scoring database based on the identifier of the case to be queried carried in the query information.

[0163] Among them, the user score is the score given by the user to the case to be processed; the case score database stores multiple scores of the case to be processed; the "user" can be other doctors or reviewers.

[0164] In the embodiment of the present application, the case scoring device can match the identification of the case to be queried with multiple case identifications in the case scoring database, determine the case identification that matches the identification of the case to be queried, and determine whether there is a user rating in the case scoring database based on the matched case identification. The user rating can be multiple ratings reviewed by the user for the case to be processed, or it can be a comprehensive score.

[0165] Step 217: If the case scoring device has the user score in the case scoring database, the user score is determined to be the score corresponding to the identifier of the case to be queried.

[0166] It should be noted that when the case scoring database contains a user score corresponding to the identifier of the case to be queried, although the case scoring data contains multiple scores of the case to be processed corresponding to the identifier of the case to be queried, the user score takes precedence over the multiple scores of the case to be processed, and the user score can be used as the score corresponding to the identifier of the case to be queried.

[0167] Step 218: If there is no user rating in the case rating database, the case rating device determines that the multiple ratings of the case to be processed are the ratings corresponding to the identifier of the case to be queried.

[0168] It should be noted that when the case scoring database does not have a user score corresponding to the identifier of the case to be queried, that is, when the case scoring database only has multiple scores of the pending cases corresponding to the identifier of the case to be queried, the multiple scores of the pending cases corresponding to the identifier of the case to be queried will be used as the score corresponding to the identifier of the case to be queried.

[0169] In other embodiments of the present application, the consultation score, the proposed diagnosis score, the medication score and the recommendation score are stored in the case scoring database using different fields.

[0170] It should be noted that the case scoring device uses different fields to store the consultation score, proposed diagnosis score, medication score, and recommendation score in the case scoring database. If other scores for the cases to be processed need to be added later, the corresponding project fields can be added to the case scoring database and the other scores can be stored through the corresponding project fields. There is no need to modify the field structure in the case scoring database, which optimizes the storage of multiple scores for the cases to be processed.

[0171] It should be noted that, for the description of the same steps and the same contents in this embodiment as those in other embodiments, reference can be made to the description in other embodiments and will not be repeated here.

[0172] The case scoring method provided in the embodiment of the present application can automatically determine multiple scores of the pending case based on the case information of the pending case and the doctor-patient conversation messages corresponding to the pending case. There is no need to manually review the pending case to obtain the multiple scores of the pending case as in the related art, which improves the efficiency of determining the multiple scores of the pending case and avoids the subjectivity of manual review that causes inaccurate multiple scores of the pending case, further improving the accuracy of the determined multiple scores of the pending case.

[0173] Based on the above embodiments, the present application also provides a case scoring system, which is applied to the case scoring method, such as Figure 3 As shown, the system includes: a doctor terminal, a case system, a case scoring device and a message middle platform; among which, the message middle platform is a doctor-patient dialogue platform.

[0174] On the doctor's terminal, when the doctor has finished providing medical services to the patient, that is, after the medical consultation is received, the end-of-consultation button on the doctor's terminal can be triggered, so that the doctor's terminal can determine whether to generate a case based on the type of the medical questionnaire and the doctor's information on the medical questionnaire. Specifically, when the type of the medical questionnaire is consistent with the type of the preset medical questionnaire, and the doctor's information indicates that the doctor is a full-time doctor, a case is generated; if the type of the medical questionnaire is inconsistent with the type of the preset medical questionnaire, or the doctor's information indicates that the doctor is not a full-time doctor, then no case is generated. Then, the next medical questionnaire can be judged to determine whether the next medical questionnaire can generate a case. In the case where the medical questionnaire generates a case, the generated case is used as a pending case. At this time, the doctor's terminal can send the pending case to the case system. If the doctor does not select the type of the pending case on the doctor's terminal, the case system defaults to the type of the pending case as a disease consultation type, and stores the pending case and the case type of the pending case. Among them, the type of the medical questionnaire is the type of medical consultation demand selected by the patient when placing an order in the Internet hospital; among them, the types of medical questionnaires include but are not limited to: graphic consultation, drugstore delivery, video consultation, and telephone consultation.

[0175] It should be noted that when a case is generated based on a medical questionnaire, an ID uniquely identifying the case can also be generated for the case, wherein the case ID is different from the medical questionnaire ID, and the medical questionnaire ID is associated with the case ID, that is, the case ID can be obtained through the medical questionnaire ID. Each of the multiple scores of pending cases stored in the case system can be initially defaulted to 0 points, wherein "0 points" is the highest score for each score.

[0176] After receiving the pending case, the case system can send the ID of the pending case and the consultation form ID of the pending case to the case scoring device, so that the case scoring device obtains the case information of the pending case according to the ID of the pending case, and obtains the doctor-patient dialogue message of the pending case from the doctor-patient dialogue platform according to the doctor's ID and the patient's ID in the case information, and uses the artificial intelligence (AI) model on the case scoring device to process the doctor-patient dialogue message and case information to determine multiple scores of the pending case. Among them, the AI ​​model is a pre-trained model for determining multiple scores of cases.

[0177] like Figure 4As shown, after the consultation is completed, if a case can be generated, the consultation form identifier can be pushed to the case scoring device in the form of a message queue (MQ), so that the case scoring device can obtain the case information of the pending case from the case information database of the case system according to the consultation form identifier, and analyze the obtained case information to obtain the identification information of the doctor and the patient involved in the pending case, and obtain the doctor-patient dialogue message according to the doctor's identification information and the patient's identification information. The AI ​​model in the case scoring device can process the case information and the doctor-patient dialogue message of the pending case, calculate each of the multiple scores of the pending case, and push the multiple scores of the pending case to the terminal of the Internet Hospital (abbreviated as Huyi) so that the Internet After receiving the multiple scores of the pending case, the terminal of the Internet hospital stores the multiple scores of the pending case in the case scoring database. When the front end calls the interface to query the multiple scores of the pending case, if the case scoring database has the multiple scores of the pending case, the multiple scores of the pending case are displayed. If the case scoring database does not have the multiple scores of the pending case, it is determined that the case scoring device has not pushed the multiple scores of the pending case to the terminal of the Internet hospital (or the push has failed). Then, a request for obtaining the multiple scores of the pending case can be initiated to the case scoring device, so that the case scoring device pushes the multiple scores of the pending case to the terminal of the Internet hospital.

[0178] Based on the above embodiments, the present invention provides a case scoring device, which can be applied to Figure 1-2 In the case scoring method provided in the corresponding embodiment, refer to Figure 5 As shown, the case scoring device 3 comprises:

[0179] The acquisition unit 31 is used to acquire case information of the case to be processed and the doctor-patient dialogue message corresponding to the case to be processed; wherein the case information represents the information generated after the doctor diagnoses the patient;

[0180] A determination unit 32, for extracting keywords from case information and doctor-patient conversation messages to determine target keywords;

[0181] The acquisition unit 31 is further used to acquire the type of the case to be processed, and determine the target word library corresponding to the case to be processed based on the type of the case to be processed;

[0182] The determination unit 32 is further configured to match the target keyword with the target vocabulary, and determine multiple scores of the case to be processed based on the matching results.

[0183] In the embodiment of the present application, the acquisition unit 31 is further configured to perform the following steps:

[0184] Obtaining the medical questionnaire identification of the case to be processed, and obtaining the case information of the case to be processed from the case information database based on the medical questionnaire identification;

[0185] Based on the case information, determine the identification information of the doctor and the identification information of the patient;

[0186] Based on the doctor's identification information and the patient's identification information, the doctor-patient dialogue message of the pending case is obtained from the doctor-patient dialogue platform.

[0187] In the embodiment of the present application, the determining unit 32 is further configured to perform the following steps:

[0188] Extracting the doctor's dialogue message from the doctor-patient dialogue message based on the doctor's identification information, and extracting the patient's dialogue message from the doctor-patient dialogue message based on the patient's identification information;

[0189] Extracting keywords from the doctor's conversation message to obtain a first keyword, and extracting keywords from the patient's conversation message to obtain a second keyword;

[0190] Obtain prescription information, recommendation information and diagnosis information based on case information, and extract keywords from the prescription information to obtain a third keyword;

[0191] Keyword extraction is performed on the suggestion information to obtain a fourth keyword, and keyword extraction is performed on the diagnosis information to obtain a fifth keyword; wherein the target keywords include: a first keyword, a second keyword, a third keyword, a fourth keyword and a fifth keyword.

[0192] In the embodiment of the present application, the determining unit 32 is further configured to perform the following steps:

[0193] Matching the first keyword with the keyword corresponding to the symptom label in the target vocabulary to obtain a first matching result, and determining the consultation score of the case to be processed based on the first matching result;

[0194] Determining a disease keyword from the target word library based on the second keyword, and matching the fifth keyword with the disease keyword to obtain a second matching result;

[0195] Determining a possible diagnosis score of the case to be processed based on the second matching result;

[0196] Determining a drug keyword from the target word library based on the second keyword, and matching the third keyword with the drug keyword to obtain a third matching result;

[0197] Determining a medication score for the case to be processed based on the third matching result;

[0198] The fourth keyword is matched with the keyword corresponding to the suggested tag in the target vocabulary to obtain a fourth matching result, and a suggested score for the case to be processed is determined based on the fourth matching result; wherein the multiple scores include at least: a consultation score, a proposed diagnosis score, a medication score and a suggested score.

[0199] In the embodiment of the present application, the determining unit 32 is further configured to perform the following steps:

[0200] Obtaining a target symptom keyword matching the second keyword from keywords corresponding to the symptom tag in the target vocabulary;

[0201] Based on the target symptom keyword and the first association relationship, a disease keyword associated with the target symptom keyword is determined from the target vocabulary; wherein the first association relationship is an association relationship between symptoms and diseases.

[0202] In the embodiment of the present application, the determining unit 32 is further configured to perform the following steps:

[0203] Based on the target symptom keyword and the second association relationship, determine the drug keyword associated with the target symptom keyword from the target word library; wherein the second association relationship is the association relationship between the symptom and the drug;

[0204] In the case where the third matching result indicates that the third keyword does not match the drug keyword, determining a medication score;

[0205] When the third matching result indicates that the third keyword matches the drug keyword and the drug type is the first type, obtaining a first matching degree between the third keyword and the drug keyword, and determining a medication score based on the first matching degree;

[0206] When the third matching result indicates that the third keyword matches the drug keyword and the drug type is the second type, obtaining a first matching degree, and determining a drug symptomatic score based on the first matching degree;

[0207] Based on the third keyword and the third association relationship, determining a prompt keyword associated with the third keyword from the target word library, and determining a second matching degree between the first keyword and the prompt keyword; wherein the third association relationship is an association relationship between the drug and the prompt suggestion;

[0208] A drug inquiry score is determined based on the second matching degree, and a medication score is determined based on the drug inquiry score and the drug symptomatic score.

[0209] In the embodiment of the present application, the determining unit 32 is further configured to perform the following steps:

[0210] When the fourth matching result indicates that the fourth keyword does not match the keyword corresponding to the suggested tag in the target vocabulary, determining a suggested score for the case to be processed;

[0211] In the case where the fourth matching result indicates that the fourth keyword matches the keyword corresponding to the suggestion tag in the target vocabulary, based on the target symptom keyword and the fourth association relationship, determining the suggestion keyword associated with the target symptom keyword from the target vocabulary; wherein the fourth association relationship is an association relationship between the symptom and the prompt suggestion;

[0212] The fourth keyword is matched with the suggested keyword to obtain a third matching degree, and a suggested score for the case to be processed is determined based on the third matching degree.

[0213] In the embodiment of the present application, the determining unit 32 is further configured to perform the following steps:

[0214] Get query information;

[0215] Based on the identifier of the case to be queried carried in the query information, query whether there is a user score in the case scoring database; wherein the user score is a score that the user reviews the case to be processed; wherein the case scoring database stores multiple scores of the case to be processed;

[0216] If there is a user score in the case score database, determine the user score as the score corresponding to the identifier of the case to be queried;

[0217] When there is no user rating in the case rating database, multiple ratings of the case to be processed are determined to be the ratings corresponding to the identifier of the case to be queried.

[0218] In the embodiment of the present application, the consultation score, the proposed diagnosis score, the medication score and the recommendation score are stored in the case scoring database using different fields.

[0219] It should be noted that the interaction process between the various units in the embodiment of the present application can refer to Figure 1-2 The implementation process of the case scoring method provided in the corresponding embodiment will not be repeated here.

[0220] The case scoring device provided in the embodiment of the present application can automatically determine multiple scores of the case to be processed based on the case information of the case to be processed and the doctor-patient dialogue messages corresponding to the case to be processed. There is no need to manually review the case to be processed to obtain the multiple scores of the case to be processed as in the related art, which improves the efficiency of determining the multiple scores of the case to be processed, and avoids the subjectivity of manual review that causes the multiple scores of the case to be processed to be inaccurate, further improving the accuracy of the determined multiple scores of the case to be processed.

[0221] Based on the above embodiments, the embodiments of the present application provide a case scoring device, which can be applied to Figure 1-2 In the case scoring method provided in the corresponding embodiment, refer to Figure 6As shown, the case scoring device 4 may include: a processor 41, a memory 42 and a communication bus 33, wherein:

[0222] The communication bus 33 is used to realize the communication connection between the processor 41 and the memory 42;

[0223] The processor 41 is used to execute the case scoring program in the memory 42 to implement the following steps:

[0224] Obtaining case information of the pending case and doctor-patient dialogue messages corresponding to the pending case; wherein the case information represents information generated after the doctor diagnoses the patient;

[0225] Extract keywords from case information and doctor-patient conversation messages to determine target keywords;

[0226] Obtaining the type of the case to be processed, and determining a target vocabulary corresponding to the case to be processed based on the type of the case to be processed;

[0227] The target keywords are matched with the target vocabulary, and multiple scores of the case to be processed are determined based on the matching results.

[0228] In other embodiments of the present application, the processor 41 is used to execute the case scoring program in the memory 42 to obtain the case information of the case to be processed and the doctor-patient dialogue message corresponding to the case to be processed, so as to implement the following steps:

[0229] Obtaining the medical questionnaire identification of the case to be processed, and obtaining the case information of the case to be processed from the case information database based on the medical questionnaire identification;

[0230] Based on the case information, determine the identification information of the doctor and the identification information of the patient;

[0231] Based on the doctor's identification information and the patient's identification information, the doctor-patient dialogue message of the pending case is obtained from the doctor-patient dialogue platform.

[0232] In other embodiments of the present application, the processor 41 is used to execute the case scoring program in the memory 42 to extract keywords from the case information and the doctor-patient dialogue message, and determine the target keywords to implement the following steps:

[0233] Extracting the doctor's dialogue message from the doctor-patient dialogue message based on the doctor's identification information, and extracting the patient's dialogue message from the doctor-patient dialogue message based on the patient's identification information;

[0234] Extracting keywords from the doctor's conversation message to obtain a first keyword, and extracting keywords from the patient's conversation message to obtain a second keyword;

[0235] Obtain prescription information, recommendation information and diagnosis information based on case information, and extract keywords from the prescription information to obtain a third keyword;

[0236] Extracting keywords from the suggestion information to obtain a fourth keyword, and extracting keywords from the diagnosis information to obtain a fifth keyword; wherein the target keywords include: the first keyword, the second keyword, the third keyword, the fourth keyword, and the fifth keyword

[0237] In other embodiments of the present application, the processor 41 is used to execute the case scoring program in the memory 42 to match the target keyword with the target vocabulary, and determine multiple scores of the case to be processed based on the matching results to implement the following steps:

[0238] Matching the first keyword with the keyword corresponding to the symptom label in the target vocabulary to obtain a first matching result, and determining the consultation score of the case to be processed based on the first matching result;

[0239] Determining a disease keyword from the target word library based on the second keyword, and matching the fifth keyword with the disease keyword to obtain a second matching result;

[0240] Determining a possible diagnosis score of the case to be processed based on the second matching result;

[0241] Determining a drug keyword from the target word library based on the second keyword, and matching the third keyword with the drug keyword to obtain a third matching result;

[0242] Determining a medication score for the case to be processed based on the third matching result;

[0243] The fourth keyword is matched with the keyword corresponding to the suggested tag in the target vocabulary to obtain a fourth matching result, and a suggested score for the case to be processed is determined based on the fourth matching result; wherein the multiple scores include at least: a consultation score, a proposed diagnosis score, a medication score and a suggested score.

[0244] In other embodiments of the present application, the processor 41 is used to execute the case scoring program in the memory 42 to determine the disease keyword from the target word library based on the second keyword to implement the following steps:

[0245] Obtaining a target symptom keyword matching the second keyword from keywords corresponding to the symptom tag in the target vocabulary;

[0246] Based on the target symptom keyword and the first association relationship, a disease keyword associated with the target symptom keyword is determined from the target vocabulary; wherein the first association relationship is an association relationship between symptoms and diseases.

[0247] In other embodiments of the present application, the processor 41 is used to execute the case scoring program in the memory 42 to determine the drug keyword from the target word library based on the second keyword to implement the following steps:

[0248] Based on the target symptom keyword and the second association relationship, determine the drug keyword associated with the target symptom keyword from the target word library; wherein the second association relationship is the association relationship between the symptom and the drug;

[0249] In the case where the third matching result indicates that the third keyword does not match the drug keyword, determining a medication score;

[0250] When the third matching result indicates that the third keyword matches the drug keyword and the drug type is the first type, obtaining a first matching degree between the third keyword and the drug keyword, and determining a medication score based on the first matching degree;

[0251] When the third matching result indicates that the third keyword matches the drug keyword, and the drug type is the second type, obtaining a first matching degree, and determining a drug symptomatic score based on the first matching degree;

[0252] Based on the third keyword and the third association relationship, determining a prompt keyword associated with the third keyword from the target word library, and determining a second matching degree between the first keyword and the prompt keyword; wherein the third association relationship is an association relationship between the drug and the prompt suggestion;

[0253] A drug inquiry score is determined based on the second matching degree, and a medication score is determined based on the drug inquiry score and the drug symptomatic score.

[0254] In other embodiments of the present application, the processor 41 is used to execute the case scoring program in the memory 42 to determine the recommended score of the case to be processed based on the fourth matching result, so as to implement the following steps:

[0255] When the fourth matching result indicates that the fourth keyword does not match the keyword corresponding to the suggested tag in the target vocabulary, determining a suggested score for the case to be processed;

[0256] In the case where the fourth matching result indicates that the fourth keyword matches the keyword corresponding to the suggestion tag in the target vocabulary, based on the target symptom keyword and the fourth association relationship, determining the suggestion keyword associated with the target symptom keyword from the target vocabulary; wherein the fourth association relationship is an association relationship between the symptom and the prompt suggestion;

[0257] The fourth keyword is matched with the suggested keyword to obtain a third matching degree, and a suggested score for the case to be processed is determined based on the third matching degree.

[0258] In other embodiments of the present application, the processor 41 is used to execute the case scoring program in the memory 42 and is also used to implement the following steps:

[0259] Get query information;

[0260] Based on the identifier of the case to be queried carried in the query information, query whether there is a user score in the case scoring database; wherein the user score is a score that the user reviews the case to be processed; wherein the case scoring database stores multiple scores of the case to be processed;

[0261] If there is a user score in the case score database, determine the user score as the score corresponding to the identifier of the case to be queried;

[0262] When there is no user rating in the case rating database, multiple ratings of the case to be processed are determined to be the ratings corresponding to the identifier of the case to be queried.

[0263] In other embodiments of the present application, the consultation score, the proposed diagnosis score, the medication score and the recommendation score are stored in the case scoring database using different fields.

[0264] The case scoring device provided in the embodiment of the present application can automatically determine multiple scores of the pending case based on the case information of the pending case and the doctor-patient conversation messages corresponding to the pending case. There is no need to manually review the pending case to obtain the multiple scores of the pending case as in the related art, which improves the efficiency of determining the multiple scores of the pending case and avoids the subjectivity of manual review that causes inaccurate multiple scores of the pending case, further improving the accuracy of the determined multiple scores of the pending case.

[0265] Based on the foregoing embodiments, the embodiments of the present application provide a computer-readable storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement Figure 1-2 The corresponding embodiments provide steps of the case scoring method.

[0266] It should be noted that the above-mentioned computer-readable storage medium can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory (Flash Memory), a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM) and other memories; it can also be various electronic devices including one or any combination of the above-mentioned memories, such as mobile phones, computers, tablet devices, personal digital assistants, etc.

[0267] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0268] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0269] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course, by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0270] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0271] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0272] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0273] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A case scoring method, characterized in that: The method comprises: Acquire case information of a pending case and a doctor-patient dialogue message corresponding to the pending case; wherein the case information represents information generated after the doctor diagnoses the patient; Keyword extraction is performed on the case information and the doctor-patient dialogue message to determine target keywords; wherein the target keywords are obtained by processing the case information and the sentences in the doctor-patient dialogue message respectively using a keyword extraction model, and include keywords representing drugs used to treat the patient, keywords representing suggestions for the patient's symptoms, and keywords representing the patient's disease; the keyword extraction model is obtained by training sample keywords including symptom keywords, disease keywords, drug keywords, and suggestion keywords using a keyword extraction algorithm; Acquire the type of the case to be processed, and determine the target vocabulary corresponding to the case to be processed based on the type of the case to be processed; Matching the target keyword with the target word library, and determining multiple scores of the case to be processed based on the matching results; The step of matching the target keyword with the target word library and determining multiple scores of the case to be processed based on the matching results includes: The target keyword is matched with keywords in a keyword set corresponding to each tag in a plurality of tags in the target vocabulary to obtain a matching result between the target keyword and each keyword set, and the multiple scores are determined based on the plurality of matching results; wherein the plurality of tags include at least a symptom tag and a suggestion tag.

2. The method according to claim 1, characterized in that The step of obtaining case information of a pending case and a doctor-patient dialogue message corresponding to the pending case includes: Obtaining the medical questionnaire identification of the case to be processed, and obtaining the case information of the case to be processed from a case information database based on the medical questionnaire identification; Based on the case information, determining identification information of the doctor and identification information of the patient; Based on the identification information of the doctor and the identification information of the patient, a doctor-patient dialogue message of the case to be processed is obtained from a doctor-patient dialogue platform.

3. The method according to claim 2, characterized in that The step of extracting keywords from the case information and the doctor-patient dialogue message to determine target keywords includes: Extracting the doctor's dialogue message from the doctor-patient dialogue message based on the doctor's identification information, and extracting the patient's dialogue message from the doctor-patient dialogue message based on the patient's identification information; Extracting keywords from the doctor's conversation message to obtain a first keyword, and extracting keywords from the patient's conversation message to obtain a second keyword; Acquiring prescription information, recommendation information, and diagnosis information based on the case information, and performing keyword extraction on the prescription information to obtain a third keyword; Keyword extraction is performed on the suggestion information to obtain a fourth keyword, and keyword extraction is performed on the diagnosis information to obtain a fifth keyword; wherein the target keywords include: the first keyword, the second keyword, the third keyword, the fourth keyword and the fifth keyword.

4. The method according to claim 3, characterized in that The target keyword is matched with the target word library, and multiple scores of the case to be processed are determined based on the matching results, including: Matching the first keyword with the keyword corresponding to the symptom label in the target vocabulary to obtain a first matching result, and determining the consultation score of the case to be processed based on the first matching result; Determining a disease keyword from the target word library based on the second keyword, and matching the fifth keyword with the disease keyword to obtain a second matching result; Determining a possible diagnosis score of the case to be processed based on the second matching result; Determining a drug keyword from the target word library based on the second keyword, and matching the third keyword with the drug keyword to obtain a third matching result; Determine the medication score of the case to be processed based on the third matching result; The fourth keyword is matched with the keyword corresponding to the suggested tag in the target vocabulary to obtain a fourth matching result, and a suggested score for the case to be processed is determined based on the fourth matching result; wherein the multiple scores include at least: the consultation score, the proposed diagnosis score, the medication score and the suggested score.

5. The method according to claim 4, characterized in that The determining of the disease keyword from the target word library based on the second keyword includes: Acquire a target symptom keyword matching the second keyword from keywords corresponding to the symptom tag in the target vocabulary; Based on the target symptom keyword and the first association relationship, the disease keyword associated with the target symptom keyword is determined from the target vocabulary; wherein the first association relationship is an association relationship between symptoms and diseases.

6. The method according to claim 5, characterized in that The determining of the drug keyword from the target word library based on the second keyword includes: Based on the target symptom keyword and the second association relationship, determining a drug keyword associated with the target symptom keyword from the target word library; wherein the second association relationship is an association relationship between symptoms and drugs; Accordingly, determining the medication score of the case to be processed based on the third matching result includes: In a case where the third matching result indicates that the third keyword does not match the drug keyword, determining the medication score; When the third matching result indicates that the third keyword matches the drug keyword and the drug type is the first type, obtaining a first matching degree between the third keyword and the drug keyword, and determining the medication score based on the first matching degree; When the third matching result indicates that the third keyword matches the drug keyword and the drug type is the second type, obtaining the first matching degree, and determining the drug symptomatic score based on the first matching degree; Based on the third keyword and the third association relationship, determining a prompt keyword associated with the third keyword from the target word library, and determining a second matching degree between the first keyword and the prompt keyword; wherein the third association relationship is an association relationship between a drug and a prompt suggestion; A drug inquiry score is determined based on the second matching degree, and the medication score is determined based on the drug inquiry score and the drug symptomatic score.

7. The method according to claim 5, characterized in that Determining the recommended score for the case to be processed based on the fourth matching result includes: In a case where the fourth matching result indicates that the fourth keyword does not match the keyword corresponding to the suggested tag in the target vocabulary, determining a suggested score for the case to be processed; In the case where the fourth matching result indicates that the fourth keyword matches the keyword corresponding to the suggestion tag in the target vocabulary, based on the target symptom keyword and the fourth association relationship, determining the suggestion keyword associated with the target symptom keyword from the target vocabulary; wherein the fourth association relationship is an association relationship between symptoms and prompt suggestions; The fourth keyword is matched with the suggested keyword to obtain a third matching degree, and a suggested score for the case to be processed is determined based on the third matching degree.

8. The method according to claim 1, characterized in that The method further comprises: Get query information; Based on the identifier of the case to be queried carried in the query information, query whether there is a user score in the case scoring database; wherein the user score is a score given by the user to review the case to be processed; wherein the case scoring database stores multiple scores of the case to be processed; If the user score exists in the case score database, determining the user score to be the score corresponding to the identifier of the case to be queried; In the case that the case scoring database does not have the user score, the multiple scores of the case to be processed are determined as the scores corresponding to the identifiers of the case to be queried.

9. The method according to claim 4, characterized in that The consultation score, the proposed diagnosis score, the medication score and the recommendation score are stored in a case scoring database using different fields.

10. A case scoring device, characterized in that: The device comprises: An acquisition unit, used to acquire case information of a case to be processed and a doctor-patient dialogue message corresponding to the case to be processed; wherein the case information represents information generated after the doctor diagnoses the patient; A determination unit is used to extract keywords from the case information and the doctor-patient dialogue message to determine target keywords; wherein the target keywords are obtained by processing the case information and the sentences in the doctor-patient dialogue message respectively using a keyword extraction model, and include keywords representing drugs used to treat the patient, keywords representing suggestions for the patient's symptoms, and keywords representing the patient's disease; the keyword extraction model is obtained by training sample keywords including symptom keywords, disease keywords, drug keywords, and suggestion keywords using a keyword extraction algorithm; The acquisition unit is further used to acquire the type of the case to be processed, and determine the target vocabulary corresponding to the case to be processed based on the type of the case to be processed; The determination unit is further used to match the target keyword with the keyword in the keyword set corresponding to each tag in the multiple tags in the target vocabulary, and determine multiple scores for the case to be processed based on multiple matching results; wherein the multiple tags include at least symptom tags and suggestion tags.

11. A case scoring device, characterized in that: The device comprises: a processor, a memory and a communication bus; The communication bus is used to realize the communication connection between the processor and the memory; The processor is used to execute the case scoring program in the memory to implement the following steps: Acquire case information of a pending case and a doctor-patient dialogue message corresponding to the pending case; wherein the case information represents information generated after the doctor diagnoses the patient; Keyword extraction is performed on the case information and the doctor-patient dialogue message to determine target keywords; wherein the target keywords are obtained by processing the case information and the sentences in the doctor-patient dialogue message respectively using a keyword extraction model, and include keywords representing drugs used to treat the patient, keywords representing suggestions for the patient's symptoms, and keywords representing the patient's disease; the keyword extraction model is obtained by training sample keywords including symptom keywords, disease keywords, drug keywords, and suggestion keywords using a keyword extraction algorithm; Acquire the type of the case to be processed, and determine the target vocabulary corresponding to the case to be processed based on the type of the case to be processed; The target keyword is matched with the keyword in the keyword set corresponding to each tag in the multiple tags in the target vocabulary, and multiple scores of the case to be processed are determined based on multiple matching results; wherein the multiple tags include at least symptom tags and suggestion tags.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the case scoring method according to any one of claims 1 to 9.

Images