An intelligent triage assistance system for improving the work efficiency of in-hospital outpatient doctors

Abstract

The application discloses an intelligent triage assistance system for improving the work efficiency of in-hospital outpatient doctors, and relates to the technical field of medical management. The system comprises: an information acquisition module, which is used for acquiring the waiting personnel of each doctor, the disease diagnosis and treatment information, and the target disease information of a target patient; a first information analysis module, which is used for determining the triage trade-off degree of each department for the target patient based on the target disease information of the target patient and the waiting personnel of each doctor, and the triage trade-off degree is used for representing the comprehensive matching degree of the target patient suitable for going to the department for treatment; a second information analysis module, which is used for determining the diagnosis and treatment ability adaptation degree of each doctor for the target patient based on the disease diagnosis and treatment information of each doctor; and a patient triage module, which is used for triaging the target patient based on each triage trade-off degree and each diagnosis and treatment ability adaptation degree. The application effectively improves the triage accuracy and the work efficiency of the outpatient doctors, and realizes more accurate and efficient triage.

Description

Technical Field

[0001] The present invention relates to the field of medical management technology, and in particular to an intelligent triage assistance system for improving the work efficiency of outpatient doctors in a hospital. Background Art

[0002] In today's era, the wave of informatization and intelligentization is powerfully driving the healthcare industry towards greater efficiency and convenience. In the critical area of ​​outpatient care, the number of outpatients has been climbing year by year, driven by an aging population and growing healthcare needs. To effectively improve the efficiency of outpatient physicians, many hospitals are actively exploring feasible ways to leverage artificial intelligence to optimize triage processes and improve patient services.

[0003] At present, the intelligent triage assistance system has been used in some hospitals. This intelligent triage assistance system can determine the department the patient should go to based on the symptoms described by the patient.

[0004] However, existing intelligent triage assistance systems have significant limitations. They can often only identify the department corresponding to a single symptom, making it difficult to analyze the potential connections hidden behind symptom combinations. This limitation can easily lead to inaccurate triage results, preventing patients from being accurately assigned to the appropriate department. Furthermore, the intelligent triage assistance system does not take the workload of outpatient doctors into consideration during the triage process, resulting in low outpatient doctor efficiency. Summary of the Invention

[0005] The embodiment of the present invention provides an intelligent triage assistance system for improving the work efficiency of outpatient doctors in the hospital, which can effectively improve the accuracy of triage and the work efficiency of outpatient doctors, and realize more accurate and efficient triage.

[0006] A first aspect of an embodiment of the present invention provides an intelligent triage assistance system for improving the work efficiency of outpatient doctors in a hospital, comprising:

[0007] The information acquisition module is used to obtain the waiting list of each doctor, the diagnosis and treatment information of the disease, and the target disease information of the target patient;

[0008] The first information analysis module is used to determine the triage weight of each department for the target patient based on the target disease information of the target patient and the waiting list of each doctor. The triage weight is used to represent the comprehensive matching degree of the target patient to the department for treatment;

[0009] The second information analysis module is used to determine the suitability of each doctor's diagnosis and treatment ability for the target patient based on the disease diagnosis and treatment information of each doctor. The diagnosis and treatment ability suitability is used to represent the doctor's suitability for diagnosing and treating the disease of the target patient;

[0010] The patient triage module is used to triage target patients based on the trade-offs of each triage and the adaptability of each diagnosis and treatment capability.

[0011] In some possible implementations, the first information analysis module includes:

[0012] A diagnosis and treatment scope involvement determination unit is used to determine the diagnosis and treatment scope involvement of each department for the target patient based on the target disease information of the target patient;

[0013] a first correlation determination unit, configured to determine, based on target disease information of the target patient and diagnosis results of each first historical patient, a correlation between the target patient and the first diagnosis results of each first historical patient, wherein the first historical patient is a historical patient with the same disease as the current symptom of the target patient;

[0014] a second correlation determination unit, configured to determine a correlation between the target patient and the second diagnosis results of each department by using the correlation between the target patient and the first diagnosis results of each first historical patient;

[0015] The triage weight determination unit is used to determine the triage weight of each department for the target patient based on the degree of involvement of each diagnosis and treatment scope, the correlation degree of each second diagnosis result and the number of waiting personnel of each doctor.

[0016] In some possible implementations, the target condition information includes current symptoms;

[0017] The unit for determining the degree of coverage of diagnosis and treatment is used to:

[0018] Perform word segmentation on the target patient's current symptoms to obtain the target patient's first symptom description sequence;

[0019] Comparing the first symptom description sequence with the second symptom description sequence of each historical user to determine the first historical patient having the same symptom as the target patient;

[0020] Utilize the triage departments corresponding to each first historical patient to determine the extent to which each department covers the diagnosis and treatment of the target patient.

[0021] In some possible implementations, the target disease information includes current symptoms and historical medical history;

[0022] The first correlation determination unit is configured to:

[0023] Marking the diagnosis results corresponding to the first historical patient and the current symptoms of the target patient as matching diagnosis results;

[0024] Comparing the matched diagnosis result of the first historical patient with the reference diagnosis result before the matched diagnosis result to obtain a first similarity between the matched diagnosis result and each reference diagnosis result;

[0025] Comparing each historical medical record of the target patient with the target reference diagnosis result to obtain a second similarity between the target reference diagnosis result and each historical medical record, wherein the target reference diagnosis result is the reference diagnosis result corresponding to the maximum first similarity;

[0026] The maximum value of the first similarity and the maximum value of the second similarity are used to determine the correlation between the first diagnosis results of the target patient and the first historical patient.

[0027] In some possible implementations, the triage weight determination unit is configured to:

[0028] Determine the doctor's busyness by using the number of patients waiting for consultation and the predicted consultation time for each patient;

[0029] The average of the diagnosis and treatment busyness of each doctor in the department is calculated to obtain the diagnosis and treatment busyness of the department;

[0030] The degree of triage weight of each department for target patients is determined by using the busyness of each department, the coverage of each diagnosis and treatment scope, and the correlation of each second diagnosis result.

[0031] In some possible implementations, before determining the doctor's busyness by using the number of people waiting for the doctor's consultation and the predicted consultation duration corresponding to each person waiting for the consultation, the first information analysis module further includes:

[0032] A duration acquisition unit is used to acquire the first historical consultation duration of a second historical patient, where the second historical patient is a historical patient who has been treated by the doctor and has the same illness as the patient waiting for treatment;

[0033] The duration prediction unit is used to perform average processing on each first historical consultation duration to obtain the doctor's predicted consultation duration for the waiting person.

[0034] In some possible implementations, the disease diagnosis and treatment information includes the number of disease diagnosis and treatments and the second historical consultation duration corresponding to each diagnosis and treatment;

[0035] The second information analysis module includes:

[0036] a treatment proficiency determination unit, configured to determine the treatment proficiency of each doctor for each type of disease based on the number of times each doctor has diagnosed and treated each type of disease and the second historical consultation duration corresponding to each diagnosis and treatment;

[0037] A unit for determining the proficiency of each doctor in treating each type of disease is used to determine the proficiency of each doctor in treating the target patient;

[0038] A potential disease length determination unit is configured to determine a potential disease length of each doctor for the target patient based on each diagnosis result of each first historical patient, the first historical patient being a historical patient having the same disease as the current symptom of the target patient;

[0039] A diagnosis and treatment ability adaptation degree determination unit is configured to determine a diagnosis and treatment ability adaptation degree of each doctor for the target patient based on the adapted disease length and the potential disease length of each doctor for the target patient.

[0040] In some possible implementation manners, the processing length determination unit is configured to:

[0041] divide the number of disease diagnoses and treatments of the target doctor for the target disease type by an average value of the number of disease diagnoses and treatments of each doctor in the target department for the target disease type, to obtain a diagnosis and treatment priority of the target doctor for the target disease type, the target doctor being any one doctor, the target disease type being any one disease type, and the target department being a department where the target doctor is located;

[0042] perform mean calculation on each second historical inquiry duration of the target doctor for the target disease type, to obtain a first diagnosis and treatment average duration of the target doctor for the target disease type;

[0043] determine a diagnosis and treatment time length priority of the target doctor for the target disease type based on the first diagnosis and treatment average duration and a second diagnosis and treatment average duration of each doctor in the target department for the target disease type;

[0044] determine a processing length of the target doctor for the target disease type by using the diagnosis and treatment priority and the diagnosis and treatment time length priority.

[0045] In some possible implementation manners, the potential disease length determination unit is configured to:

[0046] count each diagnosis result of each first historical patient to obtain a number of occurrences of each disease type;

[0047] determine a disease type as a potential disease type of the target patient if the number of occurrences meets a preset number condition and the disease type does not belong to the current symptom of the target patient;

[0048] determine a potential disease length of each doctor for the target patient based on the potential disease type of the target patient.

[0049] In some possible implementation manners, the patient triage module is configured to:

[0050] determine an optimal triage degree of each doctor for the target patient based on each triage trade-off degree and each diagnosis and treatment ability adaptation degree;

[0051] triage the target patient based on a size relationship between each optimal triage degree.

[0052] The present invention has the following beneficial effects:

[0053] The intelligent triage assistance system, provided by embodiments of the present invention, improves the work efficiency of in-hospital outpatient physicians by obtaining target symptom information from a target patient and, based on this information, determining the triage weights for each department. This process comprehensively considers information about multiple symptoms of the target patient, not just a single symptom. This allows for in-depth analysis of the potential connections hidden behind symptom combinations, allowing for more accurate determination of which department is appropriate for the target patient, significantly improving the accuracy of triage results. Furthermore, triage not only considers the target patient's target symptom information but also the doctor's waiting list and symptom diagnosis and treatment information, triaging the target patient based on a combination of triage weights and diagnostic and treatment capacity compatibility. This means that the triage process fully considers the physician's workload and diagnostic and treatment capacity, enabling the appropriate allocation of patients to physicians with a relatively reasonable workload and appropriate diagnostic and treatment capacity. This avoids the problem of inefficiency caused by excessive workload or a mismatch between the physician's diagnostic and treatment capacity and the patient's condition, effectively improving the work efficiency of outpatient physicians. In summary, the present invention effectively improves triage accuracy and the work efficiency of outpatient doctors by comprehensively considering the doctor's waiting list, disease diagnosis and treatment information, and target disease information of target patients, thereby achieving more accurate and efficient triage. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0055] Figure 1 A schematic diagram of the structure of an intelligent triage assistance system for improving the work efficiency of outpatient doctors in a hospital, provided by one embodiment of the present invention;

[0056] Figure 2 A schematic structural diagram of a first information analysis module provided in one embodiment of the present invention;

[0057] Figure 3 A schematic structural diagram of a second information analysis module provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0058] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features and effects of an intelligent triage assistance system for improving the work efficiency of outpatient doctors in the hospital proposed by the present invention. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics of one or more embodiments may be combined in any suitable form.

[0059] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0060] It should be noted that the acquisition, storage, use, and processing of data in the technical solution of the present invention comply with the relevant provisions of laws and regulations.

[0061] It should be noted that in the embodiments of the present invention, certain software, components, models and other existing solutions in the industry may be mentioned. They should be regarded as exemplary and their purpose is only to illustrate the feasibility of implementing the technical solution of the present invention, but it does not mean that the applicant has or will necessarily use the solution.

[0062] In today's era, the booming development of information technology and intelligent technology is accelerating the healthcare industry towards greater efficiency and convenience at an unprecedented rate. Outpatient care, as a key component of healthcare services, faces numerous challenges. With the aging population and growing demand for healthcare, the number of outpatients is steadily increasing. To effectively address this situation and improve the efficiency of outpatient physicians, many hospitals are actively exploring feasible solutions to optimize triage processes and improve patient services through the use of artificial intelligence (AI). Currently, intelligent triage assistance systems are already in practical use in some hospitals. These systems can preliminarily determine the department a patient should visit based on the symptoms they describe.

[0063] However, existing intelligent triage assistance systems have obvious limitations. They can usually only identify the department corresponding to a single symptom and lack the ability to deeply analyze the potential associations hidden behind the combination of symptoms. This limitation can easily lead to inaccurate triage results, making it impossible for patients to be accurately assigned to the appropriate department, which not only affects the patient's medical experience, but may also delay the diagnosis and treatment of the disease. In addition, the existing intelligent triage assistance system does not take the workload of outpatient doctors into consideration during the triage process. This means that some doctors may face too many patients and excessive work pressure, while other doctors may be relatively idle, resulting in an uneven distribution of medical resources and, in turn, low overall work efficiency of outpatient doctors.

[0064] The present invention aims to provide an intelligent triage assistance system for improving the work efficiency of in-hospital outpatient physicians. The intelligent triage assistance system for improving the work efficiency of in-hospital outpatient physicians, provided in embodiments of the present invention, obtains target symptom information from a target patient and, based on this information, determines the triage weights for each department. This process comprehensively considers information about multiple symptoms of the target patient, not just a single symptom. This allows for in-depth analysis of the potential connections hidden behind symptom combinations, allowing for more accurate determination of which department is appropriate for the target patient, significantly improving the accuracy of triage results. Furthermore, triage not only considers the target patient's target symptom information but also the doctor's waiting list and symptom diagnosis and treatment information, triaging the target patient based on a comprehensive assessment of triage weights and diagnostic and treatment capacity. This means that the triage process fully considers the doctor's workload and diagnostic and treatment capacity, enabling the appropriate allocation of patients to doctors with a reasonable workload and appropriate diagnostic and treatment capacity. This avoids the problem of inefficiency caused by excessive workload or a mismatch between the doctor's diagnostic and treatment capacity and the patient's condition, effectively improving the work efficiency of outpatient physicians. In summary, the present invention effectively improves triage accuracy and the work efficiency of outpatient doctors by comprehensively considering the doctor's waiting list, disease diagnosis and treatment information, and target disease information of target patients, thereby achieving more accurate and efficient triage.

[0065] The following describes a specific embodiment of an intelligent triage assistance system for improving the work efficiency of outpatient doctors in a hospital, provided by an embodiment of the present invention.

[0066] like Figure 1 FIG2 is a block diagram of an intelligent triage assistance system for improving the work efficiency of in-hospital outpatient doctors. The intelligent triage assistance system 100 for improving the work efficiency of in-hospital outpatient doctors includes an information acquisition module 110, a first information analysis module 120, a second information analysis module 130, and a patient triage module 140.

[0067] The information acquisition module 110 is used to obtain the waiting list of each doctor, disease diagnosis and treatment information, and target disease information of the target patient.

[0068] In this embodiment, the waiting list of each doctor is used to represent the information of patients currently waiting for the doctor's diagnosis and treatment, which may include the number of patients, personal information of patients, and other related information, reflecting the doctor's current workload status.

[0069] Disease diagnosis and treatment information may include data such as the types of diseases that the doctor is good at diagnosing and treating, past experience in diagnosing and treating related diseases, success rate, etc., which is used to evaluate the doctor's ability to diagnose and treat various diseases.

[0070] The target patient is the patient who currently needs triage. The target disease information refers to the specific description of the patient's disease, such as symptoms, possible causes, etc., which is an important basis for triage.

[0071] As an example, the information acquisition module 110 obtains the doctor's waiting list information by connecting with the hospital's information system (such as the HIS system), including a list of patients waiting, each patient's registration time, etc., so as to calculate the number of patients waiting for each doctor.

[0072] At the same time, disease diagnosis and treatment information is extracted from data sources such as the doctor's personal files and historical medical records, such as the doctor's professional field, the types of diseases he is good at treating, the success rate of treatment, and patient evaluations.

[0073] When a new patient (i.e., target patient) comes to see the doctor, the target patient's target disease information, such as symptom description, pain location, duration, etc., is collected through patient self-input and doctor's consultation records.

[0074] The first information analysis module 120 is used to determine the triage weight of each department for the target patient based on the target disease information of the target patient and the waiting list of each doctor. The triage weight is used to represent the comprehensive matching degree of the target patient to go to the department for treatment.

[0075] In this embodiment, the triage weight is used to measure the comprehensive matching degree of the target patient to a certain department for treatment, taking into account factors such as the target patient's condition and the current waiting conditions of each department.

[0076] As an example, the first information analysis module 120 classifies and extracts features of target symptom information of a target patient, for example, classifying symptoms into different types such as fever, cough, pain, etc., and analyzing their severity and relevance.

[0077] Then, considering the number of doctors waiting for their appointments, we consider the degree of compatibility between the department's expertise and the target patient's condition. For example, for patients with a fever, we prioritize departments within the Internal Medicine Department that specialize in infectious diseases. We also analyze the current number of patients waiting for their appointments and the estimated wait time in each department to avoid long waits.

[0078] Taking all of these factors into account, a pre-set algorithmic model (such as a weighted scoring model) is used to calculate the triage weighting of each department for the target patient. For example, factors such as symptom matching and waiting time are assigned different weights, and then a total score is calculated. A higher score indicates a higher triage weighting.

[0079] The second information analysis module 130 is configured to determine a diagnosis and treatment capability adaptation degree of each doctor to the target patient based on the diagnosis and treatment information of each doctor, and the diagnosis and treatment capability adaptation degree is used to represent the adaptation degree of the doctor to the diagnosis and treatment of the target patient.

[0080] In this embodiment, the diagnosis and treatment capability adaptation degree is used to represent the adaptation degree of the doctor to the diagnosis and treatment of the target patient, and is mainly evaluated based on the diagnosis and treatment information of the doctor, and reflects the capability and appropriateness of the doctor to the diagnosis and treatment of the target patient.

[0081] As an example, the second information analysis module 130 matches and filters the diagnosis and treatment information of the doctor based on the target diagnosis information of the target patient. For example, if the patient has a heart disease, the doctor who is good at the diagnosis and treatment of the heart disease is filtered out.

[0082] The diagnosis and treatment capability adaptation degree of the doctor to the target patient is evaluated according to the experience, success rate, professional skill level and other indicators of the doctor in the diagnosis and treatment of similar diseases. The fuzzy comprehensive evaluation method can be used to quantitatively score each indicator, and then the diagnosis and treatment capability adaptation degree is calculated.

[0083] The patient triage module 140 is configured to triage the target patient based on the triage trade-off degrees and the diagnosis and treatment capability adaptation degrees.

[0084] In this embodiment, the patient triage module 140 receives the triage trade-off degrees calculated by the first information analysis module 120 and the diagnosis and treatment capability adaptation degrees calculated by the second information analysis module 130.

[0085] The triage trade-off degree and the diagnosis and treatment capability adaptation degree are comprehensively considered, for example, a multiplication model or an addition model can be used to combine the two to calculate a comprehensive score.

[0086] According to the comprehensive score, the doctor corresponding to the department with the highest score is selected for the triage of the target patient, and the triage result is fed back to the information system of the hospital to guide the patient to the corresponding department for treatment.

[0087] As an optional embodiment, as shown in Figure 2 The first information analysis module 120 specifically can include:

[0088] The diagnosis and treatment range involvement determination unit 121 is configured to determine a diagnosis and treatment range involvement degree of each department to the target patient based on the target diagnosis information of the target patient.

[0089] A first correlation determination unit 122 is configured to determine a correlation between the target patient and the first diagnosis results of each first historical patient based on the target disease information of the target patient and the diagnosis results of each first historical patient, where the first historical patient is a historical patient with the same disease as the current symptom of the target patient;

[0090] The second correlation determination unit 123 is configured to determine the correlation between the target patient and the second diagnosis results of each department by using the correlation between the target patient and the first diagnosis results of each first historical patient;

[0091] The triage weight determination unit 124 is configured to determine the triage weight of each department for the target patient based on the degree of involvement of each diagnosis and treatment scope, the correlation degree of each second diagnosis result, and the number of waiting staff of each doctor.

[0092] In this embodiment, the diagnosis and treatment coverage is used to measure the degree of diagnosis and treatment coverage of the target patient's symptoms by each department, reflecting the relevance and capability of the department in diagnosing and treating the patient's symptoms.

[0093] The first historical patient is used to represent patients in the hospital's historical diagnosis and treatment records who have the same symptoms as the target patient. Their diagnosis results and treatment experience can provide a reference for the triage of the target patient.

[0094] The first diagnosis result correlation is used to indicate the similarity between the target patient and the first historical patient in the diagnosis results, reflecting the correlation between the target patient and these historical patients in terms of disease characteristics and treatment possibilities.

[0095] The second diagnostic result correlation reflects the degree of correlation between the target patient and the department's diagnostic results, taking into account the department's diagnostic capabilities and experience in similar diseases.

[0096] As an example, the diagnosis and treatment scope determination unit 121 extracts and classifies the target patient's target symptom information, for example, classifying symptoms into respiratory symptoms, digestive symptoms, etc. The target patient's symptom characteristics are then matched with the diagnosis and treatment scopes of each department, and the degree of coverage of each department's diagnosis and treatment scope for the target patient is calculated based on the degree of matching. A similarity algorithm, such as cosine similarity, can be used to calculate the degree of similarity between the symptom characteristics and the diagnosis and treatment scope, which serves as a quantitative indicator of the diagnosis and treatment scope coverage.

[0097] Then, the first correlation degree determination unit 122 screens out the first historical patients with the same disease as the current symptoms of the target patient from the historical diagnosis database of the hospital. Then, the diagnosis result information of the first historical patients is extracted, including the disease name diagnosed, the disease stage, the relevant examination results, etc. Then, the data mining and machine learning algorithms, such as decision tree, support vector machine, etc., are used to analyze and compare the target disease information of the target patient and the diagnosis results of the first historical patients, and the first diagnosis result correlation degree of the target patient and each first historical patient is calculated.

[0098] Then, the second correlation degree determination unit 123 calculates the first diagnosis result correlation degree of the target patient and each first historical patient according to the first diagnosis result correlation degree calculated by the first correlation degree determination unit 122, and takes the average of the first diagnosis result correlation degrees of all the first historical patients in each department as the second diagnosis result correlation degree of each department.

[0099] Finally, the triage trade-off determination unit 124 considers the waiting personnel situation of each doctor, calculates the waiting pressure index of each department, such as the ratio of the number of waiting people to the number of doctors in the department. Then, the diagnosis and treatment range involvement degree, the second diagnosis result correlation degree and the waiting pressure index are comprehensively calculated through a preset algorithm model (such as analytic hierarchy process, fuzzy comprehensive evaluation method, etc.) to calculate the triage trade-off degree of each department for the target patient. For example, the diagnosis and treatment range involvement degree, the second diagnosis result correlation degree and the waiting pressure index can be given different weights, and then the weighted average value is calculated as the triage trade-off degree.

[0100] Through the embodiment, the diagnosis and treatment range involvement degree is calculated, which can clearly determine the diagnosis and treatment correlation of each department for the target patient's disease, and avoids assigning the patient to an irrelevant department. At the same time, the first diagnosis result correlation degree and the second diagnosis result correlation degree are used to refer to the diagnosis experience of the historical patients, so that the triage is more in line with the actual condition of the target patient. In this way, the accuracy of triage can be improved.

[0101] As an optional embodiment, the target disease information includes current symptoms;

[0102] The diagnosis and treatment range involvement degree determination unit 121 is configured to:

[0103] perform word segmentation processing on the current symptoms of the target patient to obtain a first symptom description sequence of the target patient;

[0104] compare the first symptom description sequence with the second symptom description sequence of each historical user to determine the first historical patients with the same disease as the target patient;

[0105] determine the diagnosis and treatment range involvement degree of each department for the target patient by using the triage department corresponding to each first historical patient.

[0106] In this embodiment, the current symptoms are used to characterize the physical discomfort manifestations described by the target patient during the consultation, such as headache, cough, fever, etc., and are an important basis for triage.

[0107] Word segmentation is used to segment the target patient's current symptom description text into individual words or phrases according to specific rules for subsequent comparison and analysis. For example, using Jieba word segmentation technology, "persistent headache accompanied by nausea" is segmented into "persistent," "headache," "and," "accompanied by," and "nausea."

[0108] The first symptom description sequence is used to represent the ordered arrangement of the target patient's symptom words obtained after word segmentation processing, and is used to accurately represent the symptom characteristics of the target patient.

[0109] The second symptom description sequence is used to represent the ordered arrangement of historical user symptom words after word segmentation processing, and is used to compare with the symptoms of the target patient.

[0110] As an example, the diagnosis and treatment scope determination unit 121 adopts a word segmentation algorithm in natural language processing technology, such as a dictionary-based word segmentation method, a statistical-based word segmentation method, etc. Taking the dictionary-based word segmentation method as an example, the system will pre-build a dictionary containing a large number of medical terms and common words. When the target patient enters the current symptom description text, it will start from the beginning of the text and match the words in the dictionary in sequence. The successfully matched words will be segmented to form independent words, and finally the first symptom description sequence will be obtained. For example, for "the patient feels general fatigue and loss of appetite", after word segmentation, the sequence of "patient", "feeling", "general", "fatigue", "and", and "loss of appetite" may be obtained.

[0111] Then, using a string matching algorithm or similarity calculation algorithm, such as the edit distance algorithm or the cosine similarity algorithm, the target patient's first symptom description sequence is compared one by one with the historical user's second symptom description sequence. For example, if the first symptom description sequence A is "abdominal pain, dizziness, and arm numbness," the second symptom description sequence B is "abdominal pain, dizziness, and arm numbness," and the second symptom description sequence C is "abdominal pain and headache," then the second symptom description sequence B is completely consistent with the first symptom description sequence A. The historical patient corresponding to the second symptom description sequence B is determined to be the first historical patient.

[0112] Finally, the triage department information for each primary patient is counted. For each department, the ratio of the number of primary patients associated with that department to the total number of primary patients associated with the target patient is calculated. For example, if there are 10 primary patients with the same condition as the target patient, and 3 of them are triaged to the internal medicine department, then the internal medicine department's coverage of the target patient is 3 / 10 = 0.3.

[0113] Through the word segmentation processing and the symptom sequence comparison, the historical patients with the same disease as the target patient can be found more accurately, and the diagnosis and treatment range of each department for the target patient can be determined by referring to the department information of the historical patients. Thus, the triage can be more in line with the actual condition of the target patient, and the accuracy of the triage is improved.

[0114] As an optional embodiment, the target disease information includes current symptoms and historical medical records;

[0115] The first correlation degree determination unit 122 is configured to:

[0116] The diagnosis result of the first historical patient corresponding to the current symptoms of the target patient is marked as a matching diagnosis result;

[0117] The matching diagnosis result of the first historical patient is compared with the reference diagnosis results before the matching diagnosis result to obtain a first similarity between the matching diagnosis result and each reference diagnosis result;

[0118] Each historical medical record of the target patient is compared with a target reference diagnosis result to obtain a second similarity between the target reference diagnosis result and each historical medical record, and the target reference diagnosis result is a reference diagnosis result corresponding to the maximum first similarity;

[0119] The maximum value of the first similarity and the maximum value of the second similarity are used to determine the first diagnosis result correlation degree of the target patient and the first historical patient.

[0120] In this embodiment, the matching diagnosis result is used to represent the diagnosis result of the first historical patient corresponding to the current symptoms of the target patient. It reflects the disease condition of the first historical patient under similar symptoms.

[0121] The reference diagnosis result is used to represent the diagnosis result appearing before the matching diagnosis result in the diagnosis and treatment process of the first historical patient. These results can be used as a basis for comparison to evaluate the similarity and accuracy of the matching diagnosis result.

[0122] The first similarity is used to represent a similarity quantization index between the matching diagnosis result of the first historical patient and the reference diagnosis result. By calculating the similarity, the consistency of the matching diagnosis result and the previous diagnosis result can be understood.

[0123] The target reference diagnosis result is used to represent the reference diagnosis result with the maximum first similarity to the matching diagnosis result among all reference diagnosis results. It represents the most relevant historical diagnosis to the current matching situation.

[0124] Historical medical records are used to represent the medical records left by the target patient during his past visits to the hospital. They contain information such as symptoms, diagnosis, and treatment, and can be used to compare with the target reference diagnosis results.

[0125] The second similarity is used to quantitatively represent the similarity between the target reference diagnosis and the target patient's historical medical records. By calculating this similarity, the degree of correlation between the target patient's past medical records and the current reference diagnosis can be assessed.

[0126] As an example, the first correlation determination unit 122 compares the symptom description of the first historical patient with the current symptoms of the target patient one by one to identify matching symptoms. For matching symptoms, the corresponding diagnosis result of the first historical patient is marked as a matching diagnosis result. For example, if the target patient currently has symptoms of coughing and fever, and the first historical patient also has similar symptoms and is diagnosed with pneumonia, then the pneumonia diagnosis result of the first historical patient is marked as a matching diagnosis result.

[0127] Next, all reference diagnostic results for the first patient before the matching diagnostic result are collected. A similarity calculation algorithm, such as cosine similarity or Jaccard similarity, is used to compare the matching diagnostic result with each reference diagnostic result. Taking cosine similarity as an example, the diagnostic result is converted into a vector representation, and the cosine of the angle between the two vectors is calculated as the first similarity. The closer the similarity value is to 1, the more similar the two diagnostic results are.

[0128] Then, within the first similarity calculated in the previous step, the reference diagnosis corresponding to the maximum value is found and used as the target reference diagnosis. Using the same similarity calculation algorithm, the target reference diagnosis is compared with the target patient's historical medical records to calculate a second similarity. For example, feature extraction can be performed between the target reference diagnosis and the historical medical records, such as symptoms and diagnoses, and then the similarity between the vectors can be calculated.

[0129] Finally, the maximum value of the first similarity and the maximum value of the second similarity are normalized so that their values ​​range from 0 to 1. Using a weighted average method, different weights are assigned to the first and second similarities according to their importance, and then the weighted average is calculated as the correlation of the first diagnosis result. For example, if the first similarity is considered more important, it can be assigned a weight of 0.7 and the second similarity can be assigned a weight of 0.3. The calculation formula is: First diagnosis result correlation = 0.7 × first similarity maximum value + 0.3 × second similarity maximum value. Alternatively, the maximum value of the first similarity can be directly multiplied by the maximum value of the second similarity to obtain the result as the correlation of the first diagnosis result between the target patient and the first historical patient.

[0130] Through this embodiment, by comprehensively considering the similarity of the diagnosis results of the first historical patient and the similarity between the historical medical records of the target patient and the reference diagnosis results, the correlation degree of the first diagnosis result can be accurately calculated, thereby more accurately evaluating the degree of correlation between the diagnosis results of the target patient and the first historical patient.

[0131] As an optional embodiment, the triage weight determination unit 124 is configured to:

[0132] Determine the doctor's busyness by using the number of patients waiting for consultation and the predicted consultation time for each patient;

[0133] The average of the diagnosis and treatment busyness of each doctor in the department is calculated to obtain the diagnosis and treatment busyness of the department;

[0134] The degree of triage weight of each department for target patients is determined by using the busyness of each department, the coverage of each diagnosis and treatment scope, and the correlation of each second diagnosis result.

[0135] In this embodiment, the predicted consultation duration is used to represent the estimated time required for the doctor to conduct a consultation with the patient based on factors such as the patient's symptom description and the complexity of the disease.

[0136] The doctor's treatment busyness is an indicator used to measure the doctor's current workload, taking into account the number of patients waiting for the doctor and the predicted consultation time for each patient.

[0137] The department's diagnosis and treatment busyness is used to represent the indicator obtained by calculating the average diagnosis and treatment busyness of all doctors in the department, reflecting the busyness of the entire department in the current period.

[0138] As an example, the triage weight determination unit 124 first obtains information about the doctor's waiting list, including the number of patients and each patient's basic information. It then uses a prediction model (such as a regression model in machine learning) to estimate the consultation duration for each patient based on factors such as the patient's symptom description, the complexity of their condition, and their historical medical history. For example, the predicted consultation duration may be shorter for patients with simple symptoms and a clear condition, while it may be longer for patients with complex symptoms requiring further examination. The predicted consultation durations for each patient are then accumulated to determine the doctor's workload.

[0139] Then, the diagnosis and treatment busyness of all doctors in the department is collected, and the diagnosis and treatment busyness of all doctors in the department is accumulated and divided by the number of doctors in the department to obtain the diagnosis and treatment busyness of the department.

[0140] Finally, the department's triage weight for the target patient is determined by the following formula 1:

[0141] Formula 1

[0142] In formula 1, It is used to represent the triage weight of the target patient by the i-th department. The normalized value used to represent the busyness of the diagnosis and treatment of the i-th department, It is used to represent the extent to which the diagnosis and treatment scope of the target patient is covered by the i-th department. It is used to characterize the correlation between the second diagnosis results of the i-th department and the target patient, that is, the average of the correlation between the first diagnosis results of all the first historical patients treated in the i-th department and the target patient.

[0143] Among them, the busier the department's diagnosis and treatment, the more attention is paid to the extent to which the target patient's current symptoms are covered by the department's diagnostic scope; and the smaller the department's diagnosis and treatment, the more attention is paid to the correlation between the diagnostic results of the target patient and the first historical patients who have visited the department in the past, so as to comprehensively realize the comprehensive analysis of characteristics such as efficient allocation of diagnostic resources and coverage of diagnostic scope.

[0144] This embodiment comprehensively considers each department's workload, scope of care, and relevance of secondary diagnosis results, making triage decisions more scientific and comprehensive. This takes into account not only the department's diagnostic and treatment capabilities, but also its current workload and the relevance of the patient's diagnosis. This allows patients to be assigned to the most appropriate department based on the triage trade-off, improving triage accuracy and reducing patient wait times and referrals.

[0145] As an optional embodiment, before determining the doctor's busyness by using the number of people waiting for the doctor's consultation and the predicted consultation duration corresponding to each person waiting for the consultation, the first information analysis module 120 further includes:

[0146] A duration acquisition unit is used to acquire the first historical consultation duration of a second historical patient, where the second historical patient is a historical patient who has been treated by the doctor and has the same illness as the patient waiting for treatment;

[0147] The duration prediction unit is used to perform average processing on each first historical consultation duration to obtain the doctor's predicted consultation duration for the waiting person.

[0148] In this embodiment, the second historical patients are used to represent patients that the doctor has treated before, and these patients have the same symptoms as the current patient. The diagnosis and treatment data of these historical patients can provide a reference for predicting the consultation time of the current patient.

[0149] The first-history consultation duration refers to the actual time the doctor spent consulting with the second-history patient. This time record reflects the consultation time the doctor spent when treating patients with similar symptoms.

[0150] The predicted consultation time is obtained by averaging the first historical consultation time of the second historical patient, and is used to estimate the time required for the current patient to be consulted by the doctor.

[0151] For example, the duration acquisition unit filters out secondary historical patients with the same condition as a designated patient from the hospital's electronic medical record system or medical information database. Matching can be performed based on information such as the condition name, symptom description, and diagnosis code. The primary historical consultation duration of these secondary historical patients is then extracted. This data is typically recorded in the patient's medical record, including the consultation start and end times. The primary historical consultation duration is then calculated by calculating the time difference between the two.

[0152] Then, the duration prediction unit averages the first historical consultation durations of all qualified second historical patients to obtain the doctor's predicted consultation duration for the waiting patient.

[0153] This embodiment predicts the duration of consultations for patients waiting for treatment, allowing doctors to plan their consultation time in advance and rationally arrange the order and time of each patient's consultation, thereby improving work efficiency. This avoids wasted time or disrupted work rhythm caused by inaccurate consultation duration estimates, allowing doctors to treat patients more calmly.

[0154] As an optional embodiment, the disease diagnosis and treatment information includes the number of disease diagnosis and treatment and the second historical consultation duration corresponding to each diagnosis and treatment;

[0155] like Figure 3 As shown, a schematic diagram of the structure of a second information analysis module is provided. The second information analysis module 130 includes:

[0156] The treatment proficiency determination unit 131 is configured to determine the treatment proficiency of each doctor for each disease type based on the number of times each doctor has diagnosed and treated each disease type and the second historical consultation duration corresponding to each diagnosis and treatment;

[0157] The proficiency determination unit 132 for matching disease with target patients is used to determine the proficiency of each doctor in matching disease with target patients based on the proficiency of each doctor in treating each disease type;

[0158] A potential disease proficiency determination unit 133 is configured to determine the proficiency of each doctor for the target patient's potential disease based on each diagnosis result of each first historical patient, where the first historical patient is a historical patient with the same disease as the current symptoms of the target patient;

[0159] The diagnosis and treatment capability suitability determination unit 134 is configured to determine the diagnosis and treatment capability suitability of each doctor for the target patient based on the doctor's proficiency in the adapted disease and the proficiency in the potential disease of the target patient.

[0160] In this embodiment, treatment proficiency refers to the doctor's proficiency and ability level in diagnosing and treating a specific type of disease, and is measured by comprehensively considering factors such as the number of diagnoses and treatments and the time spent on each diagnosis and treatment.

[0161] The expertise in matching diseases refers to the doctor's expertise in treating the matching diseases (i.e., diseases that match the symptoms of the target patient) suffered by the target patient, based on the doctor's expertise in treating each type of disease and the specific situation of the target patient.

[0162] Proficiency in potential diseases refers to evaluating the doctor's proficiency in identifying potential diseases that may exist in the target patient by analyzing the diagnosis results of historical patients (first historical patients) who have the same disease as the target patient's current symptoms.

[0163] As an example, the treatment proficiency determination unit 131 first collects the number of times each doctor has diagnosed and treated each symptom type, as well as the second historical consultation duration corresponding to each diagnosis and treatment. Specifically, this data can be obtained from the hospital's medical record system. Different weights are then assigned to the number of diagnosis and treatment times, and the treatment proficiency is calculated using a weighted average method. The shorter the consultation time and the greater the number of diagnosis and treatment times, the more skilled the doctor is in treating the symptom, and the higher the corresponding treatment proficiency.

[0164] Then, the adaptation disease proficiency determination unit 132 extracts the treatment proficiency of each doctor for all disease types corresponding to the target patient from the results obtained by the treatment proficiency determination unit based on all disease types contained in the symptom description of the target patient and accumulates them as the adaptation disease proficiency of each doctor for the target patient.

[0165] Next, the potential condition proficiency determination unit 133 collects statistics on the diagnostic results for each of the first historical patients and analyzes the doctors' diagnostic accuracy and handling capabilities for these patients' potential conditions. For example, this unit can calculate the percentage of doctors who correctly diagnosed a particular potential condition, or assess the doctors' handling capabilities for the potential condition based on the treatment outcomes of the patients after diagnosis. Based on this analysis of the diagnostic results for the first historical patients, the unit then determines the proficiency of each doctor in the target patient's potential condition.

[0166] Finally, the diagnosis and treatment capability suitability determination unit 134 determines the doctor's diagnosis and treatment capability suitability for the target patient using the following formula 2:

[0167] Formula 2

[0168] In formula 2, It is used to characterize the suitability of the nth doctor’s diagnosis and treatment ability for the target patient. The normalized value used to represent the nth doctor's busyness in diagnosis and treatment. It is used to represent the degree of the nth doctor’s proficiency in matching the target patient with the disease. Used to represent the proficiency of the nth doctor in the target patient's potential disease.

[0169] Specifically, the expertise in different diseases is dynamically weighed and analyzed based on the busyness of the doctors. That is, when each doctor's diagnosis and treatment business is higher, more attention is paid to the expertise in diseases that are compatible with the target patients; and when each doctor's diagnosis and treatment business is lower, more attention is paid to the expertise in potential diseases of the target patients. In this way, the adaptability of each doctor's diagnosis and treatment capabilities to the target patients is comprehensively quantified.

[0170] Through this embodiment, by comprehensively considering the doctor's proficiency in treating various types of symptoms, his proficiency in adapting symptoms to the target patient, and his proficiency in potential symptoms, the doctor's adaptability to the target patient's diagnosis and treatment capabilities can be evaluated more comprehensively and accurately, thereby accurately matching the patient with the most suitable doctor and improving the accuracy and effectiveness of diagnosis and treatment.

[0171] As an optional embodiment, the processing proficiency determination unit 131 is configured to:

[0172] The target doctor's diagnosis and treatment frequency for the target disease type is divided by the average diagnosis and treatment frequency of the target disease type by each doctor in the target department to obtain the target doctor's diagnosis and treatment frequency priority for the target disease type. The target doctor is any doctor, the target disease type is any disease type, and the target department is the department where the target doctor is located.

[0173] Calculate the average of the second historical consultation time of the target doctor for the target disease type to obtain the average first consultation time of the target doctor for the target disease type;

[0174] Determine the priority of the target doctor's diagnosis and treatment time for the target disease type based on the average first diagnosis and treatment time and the average second diagnosis and treatment time of each doctor in the target department for the target disease type;

[0175] The target doctor's expertise in treating the target disease type is determined by using the priority of the number of consultations and the priority of the consultation time.

[0176] In this embodiment, the target doctor refers to any doctor selected for evaluation in a specific department; the target disease type refers to any disease category that requires evaluation of the doctor's expertise in treating it.

[0177] The priority of the number of diagnosis and treatment times is used to measure the degree of advantage of the target doctor in the number of diagnosis and treatment times of a specific type of disease compared with other doctors in the department.

[0178] The average first diagnosis and treatment time refers to the average of the second historical consultation times of the target doctor for the target disease type, and the average second diagnosis and treatment time refers to the average of the second historical consultation times of the target disease type by each doctor in the target department.

[0179] The priority of diagnosis and treatment time refers to the degree of advantage of the target doctor in the diagnosis and treatment time of a specific type of disease based on the average first diagnosis and treatment time of the target doctor and the average second diagnosis and treatment time of other doctors in the department.

[0180] As an example, the proficiency determination unit 131 obtains the number of times each doctor in the target department has diagnosed and treated a target symptom type from the hospital's medical record system. The target doctor's number of times he or she has diagnosed and treated the target symptom type is divided by the average number of times each doctor in the target department has diagnosed and treated the target symptom type to obtain the priority level of the target doctor's number of times he or she has diagnosed and treated the target symptom type.

[0181] The second historical consultation durations for the target condition type for each doctor in the target department are then collected and averaged to obtain the second average consultation duration. Simultaneously, the average of each second historical consultation duration for the target condition type is calculated to obtain the first average consultation duration for the target doctor. Based on the first and second average consultation durations, the priority of the target doctor's consultation duration for the target condition type is evaluated.

[0182] Finally, the priority of the number of consultations and the priority of the duration of consultations are used to determine the target doctor's proficiency in treating the target disease type.

[0183] Specifically, the processing proficiency can be determined by the following formula 3:

[0184] Formula 3

[0185] In formula 3, It is used to represent the proficiency of the nth doctor (i.e., target doctor) in the target department in treating the jth disease type (i.e., target disease type). It is used to represent the number of times the nth doctor in the target department diagnoses and treats the jth disease type. It is used to represent the average number of times each doctor in the target department diagnoses and treats the jth disease type, where is a non-zero constant. It is used to represent the average length of time it takes for the nth doctor in the target department to make the first diagnosis and treatment of the jth type of disease. It is used to represent the average duration of the second diagnosis and treatment of the jth disease type by each doctor in the target department, where is a non-zero constant, norm is used to represent the normalization process.

[0186] in, It is used to represent the priority of the number of times the nth doctor in the target department diagnoses and treats the jth type of disease. It is used to represent the priority of the diagnosis and treatment time of the nth doctor in the target department for the jth type of disease. When the nth doctor in the target department treats the jth type of disease more times or treats it faster than other doctors in the same department, the corresponding treatment expertise is greater.

[0187] Through this embodiment, by quantifying the two key indicators of the number of diagnosis and treatment and the duration of diagnosis and treatment, and calculating the priority of the number of diagnosis and treatment and the priority of the duration of diagnosis and treatment, it is possible to more objectively and comprehensively evaluate the doctor's ability to handle specific types of symptoms, avoiding the deviation of subjective evaluation.

[0188] As an optional embodiment, the potential disease proficiency determination unit 133 is configured to:

[0189] Counting the diagnosis results of each first historical patient to obtain the number of occurrences of each symptom type;

[0190] Determine the disease type that meets the preset number of occurrences and does not belong to the current symptoms of the target patient as the potential disease type of the target patient;

[0191] Based on the target patient's potential disease type, determine each doctor's expertise in the target patient's potential disease.

[0192] In this embodiment, the preset frequency condition represents a pre-defined criterion for the number of occurrences of a symptom type, used to screen for symptom types with a certain degree of prevalence or importance. For example, the condition can be set so that only symptom types meeting the criteria must have a frequency of occurrence exceeding a certain percentage (e.g., 10%) or a certain absolute number (e.g., 50).

[0193] Potential disease types are used to characterize disease types that meet the preset number conditions and do not belong to the current symptoms of the target patient based on the diagnosis results of the first historical patient. These disease types may be diseases that the target patient may develop in the future.

[0194] As an example, the potential disease proficiency determination unit 133 classifies the diagnosis results of each first historical patient by disease type and counts the number of occurrences of each disease type. Data classification and counting can be implemented using a database query language (e.g., SQL) or a data analysis tool (e.g., Python's Pandas library).

[0195] Then, from the statistically calculated number of occurrences of each symptom type, the symptom types that meet the preset number of occurrences are screened. The symptom types corresponding to the target patient's current symptoms are then excluded, and the remaining symptom types are identified as potential symptom types for the target patient. This screening process can be implemented by writing logic code in a programming language such as Python.

[0196] Finally, the proficiency of each doctor in treating all potential disease types corresponding to the target patient is extracted and accumulated as the proficiency of each doctor in treating the potential disease of the target patient.

[0197] This embodiment analyzes the diagnosis results of the first historical patient and determines the target patient's potential condition type, helping to identify potential health risks in advance. Furthermore, by determining each doctor's expertise in the target patient's potential condition based on the potential condition type, the patient can be more accurately matched with the most appropriate doctor, improving the doctor's ability to diagnose and treat the patient's potential condition.

[0198] As an optional embodiment, the patient triage module 140 is configured to:

[0199] Based on the trade-offs of each triage and the adaptability of each diagnosis and treatment capability, determine the optimal triage degree of each doctor for the target patient;

[0200] The target patients are triaged based on the size relationship between the preferred triage degrees.

[0201] In this embodiment, the preferred triage degree is an indicator calculated based on the triage weight and the diagnostic and treatment capability compatibility, used to determine the priority of each doctor in triaging the target patient. The higher the preferred triage degree, the more suitable the doctor is for diagnosing and treating the target patient.

[0202] As an example, the patient triage module 140 can set reasonable weights for triage weight and diagnostic and treatment capability adaptation based on actual conditions and expert experience. Assume that the weight of the triage weight is α, the weight of the diagnostic and treatment capability adaptation is β, and that α + β = 1. Then, the doctor's preferred triage degree for the target patient can be obtained by multiplying the doctor's office's triage weight by the corresponding weight and the doctor's corresponding diagnostic and treatment capability adaptation by the corresponding weight.

[0203] Finally, all doctors' preferred triage scores are sorted from highest to lowest. Based on the sorting results, the doctor with the highest preferred triage score is selected as the target patient's triage candidate. If multiple doctors have the same highest preferred triage score, other factors, such as the patient's personal preferences, can be considered to make a comprehensive decision.

[0204] This embodiment comprehensively considers the overall strengths of each department and the individual physician's diagnostic and treatment capabilities, making triage more scientific and reasonable. Patients can be assigned to the most appropriate physician, improving diagnostic accuracy and treatment effectiveness. This avoids triage bias caused by considering only departmental or physician factors, thereby improving triage accuracy.

[0205] It should be understood that the present invention is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, a detailed description of known methods is omitted. In the above embodiments, several specific steps are described and illustrated as examples. However, the method of the present invention is not limited to the specific steps described and illustrated. Those skilled in the art may make various changes, modifications, and additions, or change the order of the steps after understanding the spirit of the present invention.

[0206] It should also be noted that the exemplary embodiments described herein describe methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the steps described above. In other words, the steps may be performed in the order described in the embodiments, or in a different order, or several steps may be performed simultaneously.

[0207] The above description is only a specific embodiment of the present invention. Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited to this. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention.

Claims

1. An intelligent triage assistance system for improving the work efficiency of outpatient doctors in hospitals, characterized by: The system comprises: The information acquisition module is used to obtain the waiting list of each doctor, the diagnosis and treatment information of the disease, and the target disease information of the target patient; a first information analysis module, configured to determine, based on the target disease information of the target patient and the number of patients waiting for consultation at each doctor, a triage weight of each department for the target patient, wherein the triage weight is used to represent a comprehensive degree of suitability of the target patient for consultation at the department; A second information analysis module is configured to determine the suitability of each doctor's diagnosis and treatment ability for the target patient based on the disease diagnosis and treatment information of each doctor, wherein the suitability of the diagnosis and treatment ability is used to represent the degree of suitability of the doctor in diagnosing and treating the disease of the target patient; A patient triage module, configured to triage the target patient based on the triage weights and the diagnostic and treatment capability compatibility; The target disease information includes current symptoms and historical medical history; The first information analysis module includes: a first correlation determination unit, configured to determine a correlation between the target patient and the first diagnosis results of each of the first historical patients based on the target disease information of the target patient and the diagnosis results of each of the first historical patients, wherein the first historical patients are historical patients who have the same disease as the current symptoms of the target patient; The method for obtaining the correlation degree of the first diagnosis result specifically includes: Marking the diagnosis results corresponding to the current symptoms of the first historical patient and the target patient as matching diagnosis results; Comparing the matching diagnosis result of the first historical patient with reference diagnosis results preceding the matching diagnosis result to obtain a first similarity between the matching diagnosis result and each of the reference diagnosis results; wherein the reference diagnosis result is used to characterize the diagnosis result that occurred before the matching diagnosis result during the diagnosis and treatment process of the first historical patient; Comparing each of the historical medical records of the target patient with a target reference diagnosis result to obtain a second similarity between the target reference diagnosis result and each of the historical medical records, wherein the target reference diagnosis result is the reference diagnosis result corresponding to the largest first similarity; The maximum value of the first similarity and the maximum value of the second similarity are used to determine the correlation between the first diagnosis results of the target patient and the first historical patient.

2. The intelligent triage assistance system for improving the work efficiency of outpatient doctors in hospitals according to claim 1 is characterized in that: The first information analysis module includes: a diagnosis and treatment scope involvement determination unit, configured to determine the diagnosis and treatment scope involvement of each department for the target patient based on the target disease information of the target patient; a first correlation determination unit, configured to determine a correlation between the target patient and the first diagnosis results of each of the first historical patients based on the target disease information of the target patient and the diagnosis results of each of the first historical patients, wherein the first historical patients are historical patients who have the same disease as the current symptoms of the target patient; a second correlation determination unit, configured to determine a correlation between the target patient and the second diagnosis results of each of the departments by using the correlation between the target patient and the first diagnosis results of each of the first historical patients; The triage weight determination unit is used to determine the triage weight of each department for the target patient based on the coverage of each diagnosis and treatment scope, the correlation of each second diagnosis result, and the number of waiting staff of each doctor.

3. The intelligent triage assistance system for improving the work efficiency of outpatient doctors in hospitals according to claim 2 is characterized in that: The diagnosis and treatment scope involvement determination unit is used to: Performing word segmentation processing on the current symptoms of the target patient to obtain a first symptom description sequence of the target patient; Comparing the first symptom description sequence with the second symptom description sequence of each historical user to determine the first historical patient having the same symptom as the target patient; The triage departments corresponding to the first historical patients are used to determine the extent to which each department covers the diagnosis and treatment scope of the target patient.

4. The intelligent triage assistance system for improving the work efficiency of outpatient doctors in hospitals according to claim 2 is characterized in that: The triage weight determination unit is configured to: Determine the doctor's busyness by using the number of people waiting for the doctor and the predicted consultation duration corresponding to each of the waiting people; Calculate the average of the diagnosis and treatment busyness of each doctor in the department to obtain the diagnosis and treatment busyness of the department; The triage weight of each department for the target patient is determined by utilizing the busyness of each department, the coverage of each diagnosis and treatment scope, and the correlation of each second diagnosis result.

5. The intelligent triage assistance system for improving the work efficiency of outpatient doctors in hospitals according to claim 4 is characterized in that: Before determining the doctor's busyness by using the number of people waiting for the doctor's consultation and the predicted consultation duration corresponding to each of the people waiting for the consultation, the first information analysis module further includes: A duration acquisition unit is used to acquire the first historical consultation duration of a second historical patient, where the second historical patient is a historical patient who has been treated by the doctor and has the same disease as the patient waiting for treatment; The duration prediction unit is used to perform mean processing on each of the first historical consultation durations to obtain the doctor's predicted consultation duration for the waiting person.

6. The intelligent triage assistance system for improving the work efficiency of outpatient doctors according to claim 5 is characterized in that: The disease diagnosis and treatment information includes the number of disease diagnosis and treatment and the second historical consultation duration corresponding to each diagnosis and treatment; The second information analysis module includes: a treatment proficiency determination unit, configured to determine the treatment proficiency of each doctor for each type of disease based on the number of times each doctor diagnosed and treated each type of disease and the second historical consultation duration corresponding to each diagnosis and treatment; a proficiency determination unit for matching disease with the target patient, configured to determine the proficiency of each doctor in matching disease with the target patient by using the proficiency of each doctor in treating each disease type; a potential disease proficiency determination unit, configured to determine the proficiency of each of the doctors in treating the target patient's potential disease based on each diagnosis result of each first historical patient, wherein the first historical patient is a historical patient who suffers from the same disease as the current symptom of the target patient; The diagnosis and treatment capability suitability determination unit is used to determine the diagnosis and treatment capability suitability of each doctor for the target patient based on the doctor's proficiency in the adapted disease and the potential disease of the target patient.

7. The intelligent triage assistance system for improving the work efficiency of outpatient doctors in the hospital according to claim 6 is characterized in that: The processing proficiency determination unit is configured to: The target doctor's diagnosis and treatment frequency for the target symptom type is divided by the average diagnosis and treatment frequency of the target symptom type by each doctor in the target department, to obtain the priority of the target doctor's diagnosis and treatment frequency for the target symptom type, where the target doctor is any one of the doctors, the target symptom type is any one of the symptom types, and the target department is the department where the target doctor is located; Calculate the average of the second historical consultation durations of the target doctor for the target disease type to obtain the average first diagnosis and treatment duration of the target doctor for the target disease type; Determining a priority of the target doctor's diagnosis and treatment time for the target disease type based on the first average diagnosis and treatment time and the second average diagnosis and treatment time of each doctor in the target department for the target disease type; The treatment expertise of the target doctor for the target disease type is determined by using the priority of the number of diagnosis and treatment times and the priority of the diagnosis and treatment time.

8. The intelligent triage assistance system for improving the work efficiency of outpatient doctors according to claim 6 is characterized in that: The potential disease proficiency determination unit is used to: Counting the diagnosis results of each of the first historical patients to obtain the number of occurrences of each type of disease; Determine the disease type whose occurrence times meet the preset condition and does not belong to the current symptoms of the target patient as the potential disease type of the target patient; Based on the potential disease type of the target patient, the expertise of each doctor in treating the potential disease of the target patient is determined.

9. The intelligent triage assistance system for improving the work efficiency of outpatient doctors in hospitals according to any one of claims 1 to 8, characterized in that: The patient triage module is used to: Determining the preferred triage degree of each doctor for the target patient based on each triage weight and each diagnostic and treatment capability compatibility degree; The target patient is triaged based on the size relationship between the preferred triage degrees.

Images