Intelligent accompanying diagnosis cloud platform system and accompanying diagnosis method

Abstract

The invention belongs to the technical field of medical services, and provides an intelligent accompanying diagnosis cloud platform system and an accompanying diagnosis method, and the method comprises the steps: a pre-inquiry module collects the input information of a patient and the intention hospital tendency; the intelligent triage module analyzes the illness state by using a natural language processing technology and a medical knowledge base and gives out consulting room recommendation; the doctor-seeing determination module determines a doctor-seeing willingness according to the information; the intelligent scheduling module schedules nearby hospitals when the number sources are insufficient; the intelligent matching module screens the accompanied diagnosers according to the doctor seeing requirements and the accompanied diagnoser portraits; the real-time remote accompanying diagnosis module transmits pictures and sounds of a doctor seeing site to a remote family member terminal in real time to realize three-party real-time communication; according to the invention, the resources of the nearby hospitals are reasonably allocated, so that the medical resources are more fully utilized, the suitable hospitals, consulting rooms and accompanying doctors can be quickly matched for the patients, the overall treatment efficiency is improved, and the overall medical service efficiency and quality are improved.

Description

Technical Field

[0001] The present invention belongs to the field of medical service technology, and specifically relates to an intelligent accompanying medical consultation cloud platform system and accompanying medical consultation method. Background Art

[0002] With the accelerated aging of the population and the continuous growth of medical needs, the need for patients to be accompanied in medical services is becoming increasingly prominent. This is especially true for the elderly, patients with chronic diseases and those with mobility difficulties. Accompanying patients has become an important part of ensuring their smooth medical treatment.

[0003] However, traditional medical escort services rely heavily on manual arrangements, which can lead to information asymmetry, irrational resource allocation, and low efficiency. Patients often face difficulties registering, long wait times, and complex treatment processes, resulting in suboptimal utilization of medical resources. Furthermore, the rational allocation of medical escorts lacks scientific basis, leading to a shortage of medical escorts in some hospitals and the unused and wasted resources of some.

[0004] Furthermore, traditional accompanying medical services lack intelligent management tools, making it difficult to achieve efficient collaboration between patients, accompanying physicians, and medical institutions. Family members are often unable to attend a patient's appointment due to work, distance, and other reasons. Traditional medical service models rely solely on patients' recounting of the treatment process, preventing family members from understanding the patient's condition in real time. Patients also lack the psychological support of having family members present during their appointments, leading to a lack of security for both patients and their families and significantly reducing the transparency of medical services. Manual scheduling and triage are not only labor-intensive and inefficient, but also significantly increase labor costs and the burden on patients. Faced with limited appointment availability, existing systems often fail to effectively coordinate resources across different medical institutions, resulting in multiple referrals and repeated visits for patients, adding unnecessary time and financial resources.

[0005] To this end, those skilled in the art have proposed an intelligent accompanying cloud platform system and accompanying method, which aims to rationally allocate resources from nearby hospitals, make more effective use of medical resources, and quickly match patients with appropriate hospitals, clinics and accompanying physicians, thereby improving overall medical efficiency and enhancing the efficiency and quality of overall medical services. Summary of the Invention

[0006] In order to solve the above technical problems, the present invention provides an intelligent accompanying medical consultation cloud platform system and accompanying medical consultation method to solve the problems raised in the background technology.

[0007] According to the first aspect of the present disclosure, an intelligent medical companion cloud platform system is proposed, comprising:

[0008] The pre-diagnosis module is used to collect the patient's input information and preferred hospital preference information through the patient terminal; the input information includes symptom description, allergy history and family medical history;

[0009] An intelligent triage module is used to automatically analyze the condition based on the input information using natural language processing technology combined with a medical knowledge base to obtain clinic recommendation information;

[0010] A consultation determination module is used to determine the hospital and clinic to be consulted through the patient terminal based on the preferred hospital preference information and clinic recommendation information, and obtain consultation willingness information;

[0011] An intelligent scheduling module is used to combine the patient consultation intention information with the hospital terminal number source data, and when the number source is not satisfied, to schedule and recommend the same clinic in other nearby hospitals to obtain scheduling information;

[0012] An intelligent matching module is used to intelligently screen accompanying doctors based on the medical needs obtained from the medical intention information and scheduling information, combined with the multi-dimensional profile of the accompanying doctors, to obtain matching recommendation information for accompanying doctors; the multi-dimensional profile includes personal information, professional expertise, language ability and service area preference;

[0013] The real-time remote medical accompanying module is used to transmit the images and sounds of the medical consultation site to the remote family terminal in real time through smart devices, so as to achieve real-time communication among the patient, doctor and family members.

[0014] Preferably, the intelligent triage module is further used for:

[0015] Preprocessing the input information to obtain preprocessed text information;

[0016] Natural language processing technology is used to extract feature vector x from the pre-processed text information, and the similarity is calculated based on the medical knowledge base:

[0017]

[0018] Among them, y i is the characteristic vector of the i-th disease in the medical knowledge base, x·y i For vectors x and y i The dot product of ||x|| and ||y i || are vectors x and y respectively i The model;

[0019] Select the disease with the highest similarity and make a clinic recommendation:

[0020] j = argmax i Sim(x,y i )

[0021] Among them, j is the index of the disease, and the final clinic recommendation information is obtained based on the recommended clinic corresponding to disease j in the medical knowledge base.

[0022] Preferably, the medical consultation determination module is further configured to:

[0023] The vector representation of the preferred hospital tendency information is: where h i represents the patient's propensity score for the i-th hospital, and n is the total number of hospitals;

[0024] The vector representation of the clinic recommendation information is: where c j represents the matching score between the jth clinic and the patient's condition, and m is the total number of recommended clinics;

[0025] Based on the preferred hospital preference information and clinic recommendation information, a comprehensive score is calculated:

[0026] S ij =ω1×h i +ω2×c j

[0027] Among them, S ij is the comprehensive score of the jth clinic in the i-th hospital, ω1 is the hospital propensity weight, ω2 is the clinic matching weight, and ω1+ω2=1;

[0028] According to the comprehensive score, the combination (i * ,j * ):

[0029] (i * ,j * )=argmax i,j S ij

[0030] Among them, the value range of i is 1 to n, the value range of j is 1 to m; the final hospital visited is i * Hospital, the visiting clinic is No. j * clinics to obtain confirmed information on the patient’s willingness to seek medical treatment.

[0031] Preferably, the intelligent scheduling module is further used to:

[0032] According to the treatment willingness information (i * ,j * ), obtain the i-th * The jth hospital * The number source data of the clinic, determine the jth * Whether the number of available appointments in each clinic meets the patient's needs:

[0033] when No scheduling is required;

[0034] when Scheduling is required;

[0035] in, For the i * The jth hospital * The number of available appointments in each clinic, z is the number of appointments required by the patient;

[0036] When scheduling, use the following formula to calculate the i-th * The distance between the first hospital and the kth hospital

[0037]

[0038] Among them, the i * The geographical coordinates of the hospital are The geographical coordinates of the kth hospital are (x k ,y k ) ; filter out The hospitals of the i-th hospital constitute the nearby hospital set K, where r is the distance threshold, indicating that * The hospitals in the circular area with radius r as the center are the hospitals;

[0039] For the nearby hospital set K, select the ones that meet Hospital k, among which The jth * The number of available appointments in each clinic constitutes the set K′ of nearby hospitals with available appointments;

[0040] According to the hospitals in the nearby hospital set K′, calculate the comprehensive score s of the kth hospital k :

[0041]

[0042] Among them, α and β are weight coefficients, and α+β=1, R k Score the reputation of the kth hospital;

[0043] According to the comprehensive score s of the kth hospital k Sort from largest to smallest, select s k The largest hospital * As the hospital recommended by the scheduling, the scheduling information (k * ,j * ).

[0044] Preferably, the intelligent matching module is further used for:

[0045] Quantifying the personal information, professional expertise, language ability, and service area preference data in the multi-dimensional portrait to obtain a personal information score, a professional expertise score, a language ability score, and a service area preference score;

[0046] The multi-dimensional portrait vector of the accompanying doctor p is obtained as They correspond to personal information scores, professional expertise scores, language ability scores, and service area preference scores respectively;

[0047] According to the medical treatment demand, the medical treatment demand vector is obtained. Among them, q1 is set according to the patient's expectation of the personal characteristics of the accompanying doctor, and q2 is set according to j * Clinic needs setting, q3 is set according to the patient's language needs, q4 is set according to i * and k * The area where the hospital is located;

[0048] Use the following formula to calculate the matching score S between the accompanying physician p and the treatment needs. p :

[0049]

[0050] Among them, λ i is the corresponding weight vector, and When the set of accompanying doctors is P, select v accompanying doctors with the highest matching scores to form the recommended accompanying doctor set P′, which can be expressed as follows using the sorting function:

[0051] P′={p∈sort(P,S p )[0,v]}

[0052] Among them, sort(P,S p ) indicates that the matching score S p Sort the set of accompanying physicians P in descending order, [0,v] means taking the first v elements of the sorted list, and obtaining P′ as the accompanying physician matching recommendation information.

[0053] According to a second aspect of the present disclosure, a method for accompanying a patient on an intelligent accompanying cloud platform is proposed. The cloud platform includes a patient terminal, a hospital terminal, and a family terminal. The method includes the following steps:

[0054] S1. The patient enters personal information and preferred hospital preference information through the terminal, which is then collected and integrated by the pre-diagnosis module;

[0055] S2. Preprocess the entered personal information and preferred hospital preference information, apply natural language processing technology to extract key features from the preprocessed information, match the extracted key features with disease features in the medical knowledge base, obtain matching results, and obtain clinic recommendation information based on the matching results;

[0056] S3. Quantify the preferred hospital preference information and clinic recommendation information to obtain a hospital preference score and a clinic matching score. Using a comprehensive consideration model, assign weights to the hospital preference score and clinic matching score. Calculate and compare the comprehensive scores of each clinic combination to determine the clinic preference information.

[0057] S4. Based on the patient's willingness to seek medical treatment, obtain the corresponding hospital and clinic appointment data from the hospital terminal, determine whether the number of available appointments meets the patient's needs, and if not, determine the current hospital's geographical location, filter out other hospitals within a set range from the current hospital from the hospital database, and check the appointment data of the same clinic in the filtered out hospitals to obtain nearby hospitals with available appointments. These nearby hospitals with available appointments are sorted to obtain scheduling information.

[0058] S5. quantify the multi-dimensional portrait of the accompanying physician, assign weights to the dimensions of the multi-dimensional portrait, calculate the matching score between each accompanying physician and the treatment needs, sort the accompanying physicians in descending order according to the matching score, and obtain matching recommendation information for accompanying physicians;

[0059] S6. Connect patients and their families to the real-time remote medical consultation module through smart devices, and use smart devices to transmit the images and sounds of the medical consultation site to the remote family terminal in real time, enabling real-time three-party communication between patients, doctors and their families.

[0060] Compared with the prior art, the present invention has the following beneficial effects:

[0061] 1. The present invention quickly matches patients with appropriate hospitals, clinics and accompanying physicians through intelligent triage, diagnosis confirmation and intelligent scheduling, reducing patients' waiting time and unnecessary running around during the diagnosis process, and improving overall diagnosis efficiency; and through the intelligent scheduling module, it can reasonably allocate resources from nearby hospitals when appointment numbers are insufficient, so that medical resources can be more fully utilized and resource waste can be avoided.

[0062] 2. The present invention can also reasonably allocate the resources of accompanying doctors through the intelligent matching module, thereby improving the efficiency of accompanying services. The real-time remote accompanying module allows family members to participate in the medical process remotely, enhancing the sense of security of patients and their families, improving the transparency of medical services and patient experience, and increasing patients' psychological support; the collaborative work of various modules provides patients with a one-stop, personalized accompanying service, improving patients' medical experience.

[0063] 3. The present invention reduces the labor costs of manual triage, scheduling and other links, while avoiding the additional expenses incurred by patients due to blind medical treatment or multiple referrals, which helps to reduce overall medical costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is a system block diagram of the intelligent accompanying medical consultation cloud platform of the present invention;

[0065] Figure 2 This is a flow chart of the accompanying medical treatment method of the intelligent accompanying medical treatment cloud platform of the present invention. DETAILED DESCRIPTION

[0066] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention but are not intended to limit the scope of the present invention.

[0067] As attached Figure 1 As shown:

[0068] Embodiment 1: The present invention provides an intelligent accompanying medical consultation cloud platform system, comprising:

[0069] The pre-diagnosis module is used to collect the patient's input information and preferred hospital preference information through the patient terminal; the input information includes symptom description, allergy history and family medical history; through the pre-diagnosis module, it can provide comprehensive and accurate basic data for subsequent disease analysis and medical treatment arrangements, helping doctors and platforms to understand the patient's condition more comprehensively so as to make more appropriate diagnoses and recommendations.

[0070] The intelligent triage module is used to automatically analyze the condition based on the input information using natural language processing technology combined with the medical knowledge base to obtain clinic recommendation information;

[0071] By preprocessing the input information, preprocessed text information is obtained;

[0072] Natural language processing technology is used to extract feature vector x from the preprocessed text information, and combined with the medical knowledge base, similarity is calculated:

[0073]

[0074] Among them, y i is the characteristic vector of the i-th disease in the medical knowledge base, x·y i For vectors x and y i The dot product of ||x|| and ||y i || are vectors x and y respectively i The model;

[0075] Select the disease with the highest similarity and make a clinic recommendation:

[0076] j = argmax i Sim(x,y i )

[0077] Among them, j is the index of the disease, and the final clinic recommendation information is obtained based on the recommended clinic corresponding to disease j in the medical knowledge base.

[0078] By preprocessing the input information, extracting key features, and matching them with disease characteristics in the medical knowledge base, we can obtain clinic recommendation information. By using natural language processing technology and the medical knowledge base, we can automatically analyze the condition and recommend suitable clinics, thereby improving the accuracy and efficiency of triage and reducing the errors and time costs of manual triage.

[0079] The consultation determination module is used to determine the hospital and clinic to be consulted through the patient terminal based on the preferred hospital preference information and clinic recommendation information, and obtain the consultation intention information;

[0080] The vector representation of the preferred hospital preference information is: where h i represents the patient's propensity score for the i-th hospital, and n is the total number of hospitals;

[0081] The vector representation of clinic recommendation information is where c j represents the matching score between the jth clinic and the patient's condition, and m is the total number of recommended clinics;

[0082] Calculate the comprehensive score based on the preferred hospital preference information and clinic recommendation information:

[0083] S ij =ω1×h i +ω2×c j

[0084] Among them, S ij is the comprehensive score of the jth clinic in the i-th hospital, ω1 is the hospital propensity weight, ω2 is the clinic matching weight, and ω1+ω2=1;

[0085] According to the comprehensive score, the combination with the maximum comprehensive score (i * ,j * ):

[0086] (i * ,j * )=argmax i,j S ij

[0087] Among them, the value range of i is 1 to n, the value range of j is 1 to m; the final hospital visited is i * Hospital, the visiting clinic is No. j* clinics to obtain confirmed information on the patient’s willingness to seek medical treatment.

[0088] By quantifying the information on preferred hospitals and clinic recommendations, and calculating and comparing the comprehensive scores of the treatment combinations through a comprehensive consideration model, the information on treatment willingness is determined, thereby comprehensively considering the patient's wishes and the matching degree of the disease condition, and determining the most appropriate hospital and clinic for the patient, balancing patient preferences and the reasonable allocation of medical resources, and improving patient satisfaction.

[0089] The intelligent scheduling module is used to combine the patient willingness information with the hospital terminal number source data. When the number source is not met, it will recommend the same clinic in other nearby hospitals to obtain scheduling information;

[0090] According to the information on willingness to seek medical treatment (i * ,j * ), obtain the i-th * The jth hospital * The number source data of the clinic, determine the jth * Whether the number of available appointments in each clinic meets the patient's needs:

[0091] when No scheduling is required;

[0092] when Scheduling is required;

[0093] in, For the i * The jth hospital * The number of available appointments in each clinic, z is the number of appointments required by the patient;

[0094] When scheduling, use the following formula to calculate the i-th * The distance between the first hospital and the kth hospital

[0095]

[0096] Among them, the i * The geographical coordinates of the hospital are The geographical coordinates of the kth hospital are (x k ,y k ) ; filter out The hospitals of the i-th hospital constitute the nearby hospital set K, where r is the distance threshold, indicating that * The hospitals in the circular area with radius r as the center are the hospitals;

[0097] For the set of nearby hospitals K, filter out those that meet Hospital k, among which The jth* The number of available appointments in each clinic constitutes the set K′ of nearby hospitals with available appointments;

[0098] Calculate the comprehensive score s of the kth hospital based on the hospitals in the nearby hospital set K′ k :

[0099]

[0100] Among them, α and β are weight coefficients, and α+β=1, R k Score the reputation of the kth hospital;

[0101] According to the comprehensive score s of the kth hospital k Sort from largest to smallest, select s k The largest hospital * As the hospital recommended by the scheduling, the scheduling information (k * ,j * ).

[0102] According to the information on the willingness to see a doctor, the number of appointments data is obtained. When the number of appointments is insufficient, the nearby hospitals are screened and checked. The nearby hospitals with available appointments are ranked to obtain scheduling information. This effectively solves the problem of insufficient number of appointments, expands the range of patients' medical options, increases the possibility of patients receiving timely treatment, optimizes the allocation of medical resources, and effectively realizes patient diversion. In actual medical scenarios, patients usually prefer to choose other hospitals that are closer to the current preferred hospital. By setting a distance threshold, the time and energy cost of traveling can be effectively reduced.

[0103] The intelligent matching module is used to intelligently screen accompanying doctors based on the treatment needs obtained from the treatment intention information and scheduling information, combined with the multi-dimensional profile of the accompanying doctors, to obtain matching recommendation information for accompanying doctors; the multi-dimensional profile includes personal information, professional expertise, language ability and service area preference;

[0104] Quantify the personal information, professional expertise, language ability and service area preference data in the multi-dimensional portrait to obtain personal information scores, professional expertise scores, language ability scores and service area preference scores;

[0105] The multi-dimensional portrait vector of the accompanying doctor p is obtained as They correspond to personal information scores, professional expertise scores, language ability scores, and service area preference scores respectively;

[0106] According to the medical treatment demand, the medical treatment demand vector is obtained Among them, q1 is set according to the patient's expectation of the personal characteristics of the accompanying doctor, and q2 is set according to j * Clinic needs setting, q3 is set according to the patient's language needs, q4 is set according to i *and k * The area where the hospital is located;

[0107] Use the following formula to calculate the matching score S between the accompanying physician p and the treatment needs. p :

[0108]

[0109] Among them, λ i is the corresponding weight vector, and When the set of accompanying doctors is P, select v accompanying doctors with the highest matching scores to form the recommended accompanying doctor set P′, which can be expressed as follows using the sorting function:

[0110] P′={p∈sort(P,S p )[0,v]}

[0111] Among them, sort(P,S p ) indicates that the matching score S p Sort the set of accompanying physicians P in descending order, [0,v] means taking the first v elements of the sorted list, and obtaining P′ as the accompanying physician matching recommendation information.

[0112] By quantifying the multi-dimensional portraits of accompanying doctors, calculating the matching scores with medical needs, and sorting them to obtain the recommended information for accompanying doctors, we can accurately match accompanying doctors according to the specific needs of patients and the characteristics of accompanying doctors, improve the quality and professionalism of accompanying services, and meet the diverse needs of patients.

[0113] The real-time remote medical accompanying module is used to transmit the images and sounds of the medical treatment site to the remote family terminal in real time through smart devices, so as to conduct real-time communication among the patient, doctor and family members. It enables family members to understand the patient's medical condition in a timely manner even if they are not at the scene, and participate in the medical process, thereby enhancing the sense of security of patients and families, and improving the transparency of medical services and patient experience.

[0114] As attached Figure 2 As shown:

[0115] Example 2: The present invention also provides a method for accompanying a patient on an intelligent accompanying cloud platform, wherein the cloud platform includes a patient terminal, a hospital terminal, and a family terminal. The method includes the following steps:

[0116] S1. The patient enters personal information and preferred hospital preference information through the terminal, which is then collected and integrated by the pre-diagnosis module;

[0117] S2. Preprocess the entered personal information and preferred hospital preference information, apply natural language processing technology to extract key features from the preprocessed information, match the extracted key features with disease features in the medical knowledge base, obtain matching results, and obtain clinic recommendation information based on the matching results;

[0118] S3. Quantify the preferred hospital preference information and clinic recommendation information to obtain hospital preference scores and clinic matching scores. Using a comprehensive consideration model, assign weights to the hospital preference scores and clinic matching scores, calculate and compare the comprehensive scores of each clinic combination, and determine the clinic preference information.

[0119] S4. Based on the patient's willingness to see a doctor, the hospital obtains the corresponding hospital and clinic's appointment data from the hospital terminal to determine whether the number of available appointments meets the patient's needs. If the number of available appointments does not meet the patient's needs, the hospital's geographical location is determined. Other hospitals within the set range of the current hospital are screened from the hospital database. The appointment data of the same clinic in these other screened hospitals is checked to obtain nearby hospitals with available appointments. These nearby hospitals with available appointments are sorted to obtain scheduling information.

[0120] S5. Quantify the multi-dimensional portrait of the accompanying physician, assign weights to each dimension of the multi-dimensional portrait, calculate the matching score between each accompanying physician and the treatment needs, sort the accompanying physicians in descending order according to the matching score, and obtain matching recommendation information for accompanying physicians;

[0121] S6. Connect patients and their families to the real-time remote medical consultation module through smart devices, and use smart devices to transmit the images and sounds of the medical consultation site to the remote family terminal in real time, enabling real-time three-party communication between patients, doctors and their families.

[0122] As can be seen from the above, not only the patient's condition and hospital information are taken into consideration, but also the multi-dimensional portrait of the accompanying physician is used for precise matching to provide patients with personalized accompanying physician recommendations; existing accompanying physician services mostly focus on on-site accompanying physicians, while this platform realizes real-time communication among patients, doctors and family members through a real-time remote accompanying physician module, breaking the spatial limitations and providing an effective solution for situations where family members cannot be present; and intelligent scheduling can automatically dispatch resources from nearby hospitals according to the number of appointments. This intelligent resource optimization configuration method is innovative in improving the utilization rate of medical resources, and is more efficient and accurate than traditional manual scheduling or simple information query methods; integrating multiple functions on the cloud platform realizes information sharing and collaborative work, and provides a full-process accompanying physician service.

[0123] It is important to note that the construction and arrangement of the present application shown in a plurality of different exemplary embodiments are merely illustrative. Although only a few embodiments are described in detail in this disclosure, it will be readily understood by those who consult this disclosure that many modifications are possible without departing substantially from the novel teachings and advantages of the subject matter described in this application. Other replacements, modifications, changes, and omissions may be made in the design, operating conditions, and arrangement of the exemplary embodiments without departing from the scope of the present invention. Therefore, the present invention is not limited to specific embodiments, but extends to a variety of modifications still falling within the scope of the appended claims.

[0124] Additionally, in order to provide a concise description of exemplary embodiments, all features of an actual embodiment (i.e., those features that are not relevant to the best mode presently contemplated for carrying out the invention or those that are not relevant to implementing the invention) may not be described.

[0125] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent accompanying medical consultation cloud platform system, characterized in that: include: The pre-diagnosis module is used to collect the patient's input information and preferred hospital preference information through the patient terminal; the input information includes symptom description, allergy history and family medical history; An intelligent triage module is used to automatically analyze the condition based on the input information using natural language processing technology combined with a medical knowledge base to obtain clinic recommendation information; A consultation determination module is used to determine the hospital and clinic to be consulted through the patient terminal based on the preferred hospital preference information and clinic recommendation information, and obtain consultation willingness information; An intelligent scheduling module is used to combine the patient consultation intention information with the hospital terminal number source data, and when the number source is not satisfied, to schedule and recommend the same clinic in other nearby hospitals to obtain scheduling information; An intelligent matching module is used to intelligently screen accompanying doctors based on the medical needs obtained from the medical intention information and scheduling information, combined with the multi-dimensional profile of the accompanying doctors, to obtain matching recommendation information for accompanying doctors; the multi-dimensional profile includes personal information, professional expertise, language ability and service area preference; The real-time remote medical accompanying module is used to transmit the images and sounds of the medical consultation site to the remote family terminal in real time through smart devices, which is used for real-time communication among patients, doctors and their families.

2. The intelligent medical companion cloud platform system according to claim 1, characterized in that: The intelligent triage module is also used for: Preprocessing the input information to obtain preprocessed text information; Natural language processing technology is used to extract feature vector x from the pre-processed text information, and the similarity is calculated based on the medical knowledge base: Among them, y i is the characteristic vector of the i-th disease in the medical knowledge base, x·y i For vectors x and y i The dot product of ||x|| and ||y i || are vectors x and y respectively i The model; Select the disease with the highest similarity and make a clinic recommendation: j=argmax i Sim(x,y i ) Among them, j is the index of the disease, and the final clinic recommendation information is obtained based on the recommended clinic corresponding to disease j in the medical knowledge base.

3. The intelligent medical companion cloud platform system according to claim 1, characterized in that: The medical consultation determination module is further configured to: Based on the patient's preferred hospital tendency information collected by the patient terminal, the preferred hospital tendency information is represented by a vector, and the preferred hospital tendency information vector is obtained as follows: where h i represents the patient's propensity score for the i-th hospital, and n is the total number of hospitals; Based on the clinic recommendation information obtained by the intelligent triage module, the clinic recommendation information is represented by a vector, and the vector of the clinic recommendation information is obtained as follows: where c j represents the matching score between the jth clinic and the patient's condition, and m is the total number of recommended clinics; Based on the preferred hospital preference information and clinic recommendation information, a comprehensive score is calculated: S ij =ω1×h i +ω2×c j Among them, S ij is the comprehensive score of the jth clinic in the i-th hospital, ω1 is the hospital propensity weight, ω2 is the clinic matching weight, and ω1+ω2=1; According to the comprehensive score, the combination (i * ,j * ): (i * ,j * )=argmax i,j S ij Among them, the value range of i is 1 to n, the value range of j is 1 to m; the final hospital visited is i * Hospital, the visiting clinic is No. j * clinics to obtain confirmed information on the patient’s willingness to seek medical treatment.

4. The intelligent medical companion cloud platform system according to claim 1, characterized in that: The intelligent scheduling module is also used for: According to the treatment willingness information (i * ,j * ), obtain the i-th * The jth hospital * The number source data of the clinic, determine the jth * Whether the number of available appointments in each clinic meets the patient's needs: when No scheduling is required; when Scheduling is required; in, For the i * The jth hospital * The number of available appointments in each clinic, z is the number of appointments required by the patient; When scheduling, use the following formula to calculate the i-th * The distance between the first hospital and the kth hospital Among them, the i * The geographical coordinates of the hospital are The geographical coordinates of the kth hospital are (x k ,y k ) ; filter out The hospitals of the i-th hospital constitute the nearby hospital set K, where r is the distance threshold, indicating that * The hospitals in the circular area with radius r as the center are the hospitals; For the nearby hospital set K, select the ones that meet Hospital k, among which The jth * The number of available appointments in each clinic constitutes the set K′ of nearby hospitals with available appointments; According to the hospitals in the nearby hospital set K′, calculate the comprehensive score s of the kth hospital k : Among them, α and β are weight coefficients, and α+β=1, R k Score the reputation of the kth hospital; According to the comprehensive score s of the kth hospital k Sort from largest to smallest, select s k The largest hospital * As the hospital recommended by the scheduling, the scheduling information (k * ,j * ).

5. The intelligent medical companion cloud platform system according to claim 1, characterized in that: The intelligent matching module is also used for: Quantifying the personal information, professional expertise, language ability, and service area preference data in the multi-dimensional portrait to obtain a personal information score, a professional expertise score, a language ability score, and a service area preference score; According to the personal information score, professional expertise score, language ability score and service area preference score, the multi-dimensional portrait vector of the accompanying physician p is obtained as follows: in, According to the medical treatment demand, the medical treatment demand vector is obtained. Among them, q1 is set according to the patient's expectation of the personal characteristics of the accompanying doctor, and q2 is set according to j * Clinic needs setting, q3 is set according to the patient's language needs, q4 is set according to i * and k * The area where the hospital is located; Use the following formula to calculate the matching score S between the accompanying physician p and the treatment needs. p : Among them, λ i is the corresponding weight vector, and When the set of accompanying doctors is P, select v accompanying doctors with the highest matching scores to form the recommended accompanying doctor set P′, which can be expressed as follows using the sorting function: P′={p∈sort(P,S p )[0,v]} Among them, sort(P,S p ) indicates that the matching score S p Sort the set of accompanying physicians P in descending order, [0,v] means taking the first v elements of the sorted list, and obtaining P′ as the accompanying physician matching recommendation information.

6. A method for accompanying a patient on an intelligent accompanying cloud platform, characterized in that: The cloud platform includes a patient terminal, a hospital terminal, and a family terminal, and the method includes the following steps: S1. The patient enters personal information and preferred hospital preference information through the terminal, which is then collected and integrated by the pre-diagnosis module; S2. Preprocess the entered personal information and preferred hospital preference information, apply natural language processing technology to extract key features from the preprocessed information, match the extracted key features with disease features in the medical knowledge base, obtain matching results, and obtain clinic recommendation information based on the matching results; S3. Quantify the preferred hospital preference information and clinic recommendation information to obtain a hospital preference score and a clinic matching score. Using a comprehensive consideration model, assign weights to the hospital preference score and clinic matching score. Calculate and compare the comprehensive scores of each clinic combination to determine the clinic preference information. S4. Based on the patient's willingness to seek medical treatment, obtain the corresponding hospital and clinic appointment data from the hospital terminal, determine whether the number of available appointments meets the patient's needs, and if not, determine the current hospital's geographical location, filter out other hospitals within a set range from the current hospital from the hospital database, and check the appointment data of the same clinic in the filtered out hospitals to obtain nearby hospitals with available appointments. These nearby hospitals with available appointments are sorted to obtain scheduling information. S5. quantify the multi-dimensional portrait of the accompanying physician, assign weights to the dimensions of the multi-dimensional portrait, calculate the matching score between each accompanying physician and the treatment needs, sort the accompanying physicians in descending order according to the matching score, and obtain matching recommendation information for accompanying physicians; S6. Connect patients and their families to the real-time remote medical consultation module through smart devices, and use smart devices to transmit the images and sounds of the medical consultation site to the remote family terminal in real time, enabling real-time three-party communication between patients, doctors and their families.

Images