AI-based health care service recommendation system

Abstract

The present invention discloses an AI-based health care service recommendation system, which relates to the field of Internet technology. The present invention includes a request unit, which is used for users to select a medical department and input symptom keywords; a preliminary generation unit, which generates a preliminary medical list according to the information transmitted by the request unit; a pointing division unit, which recommends a medical hospital to the user through a client according to the preliminary medical list; a matching unit: it combines additional requests uploaded by the user to match the intended medical hospital from the recommended medical hospitals for the user. The present invention recommends medical hospitals to users through the analysis of medical information input by users and medical big data, saving doctors' time for light consultations, and directly recommending eligible medical hospitals online, saving time for users, while recommending users with different needs to different hospitals to achieve medical diversion.

Description

Technical Field

[0001] The present invention belongs to the field of Internet technology, and in particular to an AI-based health care service recommendation system. Background Art

[0002] With the development of electronic information technology, medical information technology has also developed accordingly. Different hospitals also have different focuses, which has led to the emergence of specialized hospitals and specialty departments. Due to the large amount and complexity of medical information, if users search and understand the relevant information of various hospitals by themselves, and then manually select hospitals that suit their own preferences and needs, the workload is extremely large.

[0003] For example, Chinese patent CN109065185A discloses a method and system for fast and accurate medical services for patients, which includes the steps of establishing a medical database, receiving requests, waking up medical personnel services, confirming appointments, preliminary diagnosis, and feedback on diagnostic results. It can solve the problems of cumbersome existing medical procedures and difficult registration, and can effectively improve the efficiency and portability of medical services, and reduce the waiting time for medical personnel. For example, Chinese patent CN112711639A discloses a blockchain-based AI doctor medical method and system, which determines the treatment plan given by the AI ​​doctor. After the treatment plan is completed, the patient's medical choice, medical process and medical results are formed into a medical record.

[0004] In addition, there are a large number of intelligent service systems for medical treatment, such as Chinese patents CN109584093A, CN110428890A, etc., which help users screen medical institutions; however, there is a lack of technical means to provide users with a service system that integrates their own conditions and hospital conditions to solve the problem of information islands. Summary of the invention

[0005] The purpose of the present invention is to provide an AI-based health care service recommendation system, which provides users with medical treatment by combining user requests, analysis of medical big data, etc., solving the problems of existing medical information islands, complicated medical procedures, and difficult selection.

[0006] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:

[0007] The present invention is an AI-based health care service recommendation system, comprising: a request unit, which is used by users to select a treatment department and input symptom keywords; a preliminary generation unit, which generates a preliminary treatment form according to the information transmitted by the request unit; a direction division unit, which recommends a treatment hospital to the user through a client according to the preliminary treatment form; a matching unit: which combines additional requests uploaded by the user to match the user with the intended treatment hospital from the recommended treatment hospitals.

[0008] As an embodiment provided by the present invention, preferably, the medical departments are the set of departments in all hospitals, and each medical department corresponds to a number of symptom keywords.

[0009] As an embodiment provided by the present invention, preferably, when the preliminary generation unit generates a preliminary medical form according to the information transmitted by the request unit, it first pre-screens the information transmitted by the request unit in combination with the database; the following algorithm is executed during pre-screening:

[0010] Obtain the medical department selected by the user;

[0011] Retrieve all the symptom keywords input by the user and mark them as the target keyword set B, B = {b1, b2, b3, …, bn};

[0012] Retrieve all the symptom keywords corresponding to the medical department selected by the user from the database and mark them as the reference keyword set C, C = {c1, c2, c3, …, cm};

[0013] Compare the symptom keywords in the target keyword set and the reference keyword set:

[0014] If B ∩ C = B, then use the symptom keywords input by the user as the suspected diagnosis keywords;

[0015] If B ∩ C ≠ B and then obtain the number of symptom keywords in the difference set of the target keyword set B relative to the reference keyword set C and mark it as the difference score F;

[0016] When the difference score F ≥ F1, screen out F symptom keywords from the target keyword set B to make B ∩ C = B, and transmit a signal for re-entering symptom keywords to the request unit through the processor until the corresponding difference score F < F1, and then use the symptom keywords in the corresponding target keyword set as the suspected diagnosis keywords;

[0017] If then transmit a signal for re-entering symptom keywords and medical departments to the request unit through the processor;

[0018] Wherein, m and n are positive integers, and F1 is a preset value.

[0019] As an embodiment provided by the present invention, preferably, after the preliminary generation unit pre-screens the information transmitted by the request unit in combination with the database, it generates a preliminary medical form, and the preliminary medical form includes "suspected diagnosis keywords + medical department".

[0020] As an embodiment provided by the present invention, preferably, when the pointing and dividing unit recommends a medical hospital according to the preliminary medical form, the following algorithm is executed:

[0021] Obtain the emergency weight D of the visiting department corresponding to the preliminary visiting order from the database;

[0022] If D1<emergency weight D≤D2, then obtain the distance H and cure rate Y between all hospitals with the treatment department and the user, and calculate the hospital recommendation value TP based on the distance H and cure rate Y, and classify the M hospitals with the largest hospital recommendation value TP as the directed hospitals;

[0023] If the emergency weight D>D2, the distance H, the fastest consultation time, and the cure rate Y between all hospitals with the treatment department and the user are obtained respectively, and the hospital recommendation value TP1 is calculated based on the distance H, the fastest consultation time, and the cure rate Y. The M hospitals with the largest hospital recommendation value TP1 are classified as the directed hospitals;

[0024] If the emergency weight D≤D1, the distance H and the fastest consultation time between the user and all hospitals with the treatment department are obtained respectively, and the hospital recommendation value TP2 is calculated based on the distance H and the fastest consultation time, and the M hospitals with the largest hospital recommendation value TP2 are classified as the directed hospitals;

[0025] Wherein, M is a preset value.

[0026] As an embodiment provided by the present invention, preferably, the method in which the direction division unit calculates the hospital recommendation value TP in combination with the distance H and the cure rate Y is:

[0027] Obtain all the patients who visited the department within one year, and retrieve the medication status of the patients from the database;

[0028] Eliminate the users who did not pick up the medicine, and mark the total number of remaining users as the total number of actual visits;

[0029] Get the number of cured users among the users who have received the medicine, marked as the total number of cured users, which can be obtained through the user's selection after the consultation. The user can select "cured" or "uncured" through the client;

[0030] Cure rate Y = (total number of cured cases / total number of actual diagnosis cases)*β, where β is a preset value and 0<β<1;

[0031] The recommended value TP=α*H+λ*Y, where α+λ=1 and α≥λ, and α and λ are both preset values.

[0032] As an embodiment provided by the present invention, preferably, the method for calculating the hospital recommendation value TP1 by combining the distance H, the fastest consultation time, and the cure rate Y is:

[0033] The hospital recommended value TP is calculated based on the combined distance H and cure rate Y;

[0034] Arrange the hospital recommendation values ​​TP in descending order, and mark the top 5 hospitals as the proposed hospitals;

[0035] Get the fastest consultation time of the corresponding department of the proposed hospital E: Fastest consultation time Among them, PR is the current number of people in the queue, ST is the average consultation time per person, and dT is the expected arrival time;

[0036] The hospital recommended value TP1=TP+E.

[0037] As an embodiment provided by the present invention, preferably, the method for calculating the hospital recommendation value TP2 in combination with the distance H and the fastest consultation time E is:

[0038] The hospital recommended value TP2 = α*H+λ*E, where α+λ = 1 and α≥λ, and α and λ are both preset values.

[0039] As an embodiment provided by the present invention, preferably, the method in which the matching unit matches the intended hospital for the user from the recommended hospitals in combination with the additional request uploaded by the user is:

[0040] Obtaining an additional request from the user, where the additional request is a proposed inspection item input by the user;

[0041] If the recommended hospital includes the items to be examined, the recommended hospital will be directly matched to the user;

[0042] Otherwise, the hospital with the highest recommendation value and that has the intended examination items is selected from the non-recommended hospitals and matched to the user.

[0043] The present invention has the following beneficial effects:

[0044] The present invention selects a visiting department and inputs symptom keywords through a request unit; a preliminary generation unit generates a preliminary visiting list according to the information transmitted by the request unit; a pointing division unit recommends a visiting hospital to the user through a client according to the preliminary visiting list; a matching unit matches the intended visiting hospital from the recommended hospitals in combination with the additional request uploaded by the user, thereby saving the doctor's time for light consultation and directly recommending a suitable visiting hospital online, thereby saving time for the user and recommending users with different needs to different hospitals to achieve diversion of visits.

[0045] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0047] Figure 1 It is a structural schematic diagram of the Internet-based health care service management system of the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] See also Figure 1 As shown, the present invention is an AI-based health care service recommendation system, including: a request unit, wherein the request unit is used for the user to select a treatment department and input symptom keywords, such as fever, headache, dizziness, nausea, runny nose, chest tightness, etc. Preferably, the symptom keywords are preset by the system for the user to select; a preliminary generation unit, which generates a preliminary medical form according to the information transmitted by the request unit; a pointing division unit, which recommends a hospital to the user through the client according to the preliminary medical form; a matching unit: which combines the additional requests uploaded by the user to match the intended hospital for the user from the recommended hospitals, and through matching, the online stage plays a role in the medical institution.

[0050] As an embodiment provided by the present invention, preferably, the treatment department is a collection of departments of all hospitals, and each treatment department corresponds to a number of symptom keywords.

[0051] As an embodiment provided by the present invention, preferably, when the preliminary generating unit generates the preliminary medical record according to the information transmitted by the requesting unit, the information transmitted by the requesting unit is pre-screened in combination with the database; the following algorithm is executed during the pre-screening:

[0052] Get the medical department selected by the user;

[0053] Retrieve all symptom keywords input by the user and mark them as a target keyword set B, where B = {b1, b2, b3, L, bn};

[0054] Retrieve all symptom keywords corresponding to the selected department of medical treatment from the database, and mark them as the reference keyword set C, C = {c1, c2, c3, …, cm};

[0055] Compare the symptom keywords in the target keyword set and the reference keyword set:

[0056] If B ∩ C = B, then regard the symptom keywords input by the user as the keywords for preliminary diagnosis;

[0057] If B ∩ C ≠ B and then obtain the number of symptom keywords in the difference set of the target keyword set B relative to the reference keyword set C, and mark it as the difference score F;

[0058] When the difference score F ≥ F1, screen out F symptom keywords from the target keyword set B to make B ∩ C = B, and transmit a signal for re - inputting symptom keywords to the request unit through the processor until the corresponding difference score F < F1, and then regard the symptom keywords in the corresponding target keyword set as the keywords for preliminary diagnosis;

[0059] If then transmit a signal for re - inputting symptom keywords and the department of medical treatment to the request unit through the processor;

[0060] Wherein, m and n are positive integers, and F1 is a preset value.

[0061] As an embodiment provided by the present invention, preferably, after the preliminary generation unit pre - screens the information transmitted by the request unit in combination with the database, a preliminary medical treatment form is generated, and the preliminary medical treatment form includes "keywords for preliminary diagnosis + department of medical treatment".

[0062] As an embodiment provided by the present invention, preferably, when the pointing and dividing unit recommends a medical treatment hospital according to the preliminary medical treatment form, the following algorithm is executed:

[0063] Retrieve the emergency weight D of the department of medical treatment corresponding to the preliminary medical treatment form from the database;

[0064] If D1 < emergency weight D ≤ D2, respectively obtain the distance H, cure rate Y between all hospitals with this department of medical treatment and the user, and calculate the hospital recommendation value TP by combining the distance H and the cure rate Y, and divide the M hospitals with the largest hospital recommendation value TP into the pointing hospitals;

[0065] If the emergency weight D > D2, respectively obtain the distance H, the fastest medical treatment time, and the cure rate Y between all hospitals with this department of medical treatment and the user, and calculate the hospital recommendation value TP1 by combining the distance H, the fastest medical treatment time, and the cure rate Y, and divide the M hospitals with the largest hospital recommendation value TP1 into the pointing hospitals;

[0066] If the emergency weight D≤D1, the distance H and the fastest consultation time between the user and all hospitals with the treatment department are obtained respectively, and the hospital recommendation value TP2 is calculated based on the distance H and the fastest consultation time, and the M hospitals with the largest hospital recommendation value TP2 are classified as the directed hospitals;

[0067] Wherein, M is a preset value.

[0068] As an embodiment provided by the present invention, preferably, the method in which the direction division unit calculates the hospital recommendation value TP in combination with the distance H and the cure rate Y is:

[0069] Obtain all the patients who visited the department within one year, and retrieve the medication status of the patients from the database;

[0070] Eliminate the users who did not pick up the medicine, and mark the total number of remaining users as the total number of actual visits;

[0071] Get the number of cured users among the users who have received the medicine, marked as the total number of cured users, which can be obtained through the user's selection after the consultation. The user can select "cured" or "uncured" through the client;

[0072] Cure rate Y = (total number of cured cases / total number of actual diagnosis cases)*β, where β is a preset value and 0<β<1;

[0073] The recommended value TP=α*H+λ*Y, where α+λ=1 and α≥λ, and α and λ are both preset values.

[0074] As an embodiment provided by the present invention, preferably, the method for calculating the hospital recommendation value TP1 by combining the distance H, the fastest consultation time, and the cure rate Y is:

[0075] The hospital recommended value TP is calculated based on the combined distance H and cure rate Y;

[0076] Arrange the hospital recommendation values ​​TP in descending order, and mark the top 5 hospitals as the recommended hospitals; obtain the fastest consultation time E of the corresponding consultation department of the recommended hospital: the fastest consultation time Among them, PR is the current number of people in the queue, ST is the average consultation time per person, and dT is the estimated arrival time. The estimated arrival time is pushed in real time by the map navigation based on the user's path to the hospital; the average consultation time per person is calculated based on the consultation data within half a year;

[0077] The hospital recommended value TP1=TP+E.

[0078] As an embodiment provided by the present invention, preferably, the method for calculating the hospital recommendation value TP2 in combination with the distance H and the fastest consultation time E is:

[0079] The hospital recommended value TP2 = α*H+λ*E, where α+λ = 1 and α≥λ, and α and λ are both preset values.

[0080] As an embodiment provided by the present invention, preferably, the method in which the matching unit matches the intended hospital for the user from the recommended hospitals in combination with the additional request uploaded by the user is:

[0081] Obtaining an additional request from the user, where the additional request is a proposed inspection item input by the user;

[0082] If the recommended hospital includes the items to be examined, the recommended hospital will be directly matched to the user;

[0083] Otherwise, the hospital with the highest recommendation value and that has the intended examination items is selected from the non-recommended hospitals and matched to the user.

[0084] The AI-based health care service recommendation system selects the treatment department and enters symptom keywords through the request unit; the preliminary generation unit generates a preliminary treatment order based on the information transmitted by the request unit; the pointing division unit recommends the hospital to the user through the client based on the preliminary treatment order; the matching unit combines the additional requests uploaded by the user to match the user with the intended hospital from the recommended hospitals, saving the doctor's time for light consultation, and directly recommending suitable hospitals online, saving time for users while recommending users with different needs to different hospitals to achieve diversion of treatment.

[0085] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0086] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. AI-based health care service recommendation system, characterized by: Including: A request unit, which is used for a user to select a department for seeing a doctor and input symptom keywords; A preliminary generation unit, which generates a preliminary medical visit form according to the information transmitted by the request unit; A pointing and division unit, which recommends medical visit hospitals to the user through a client according to the preliminary medical visit form; A matching unit: which combines the additional requests uploaded by the user to match the intended medical visit hospitals for the user from the recommended medical visit hospitals; The department for seeing a doctor is a set of departments of all hospitals, and each department for seeing a doctor corresponds to a number of symptom keywords; When the preliminary generation unit generates a preliminary medical visit form according to the information transmitted by the request unit, it first pre-screens the information transmitted by the request unit in combination with a database; the following algorithm is executed during pre-screening: Obtain the department for seeing a doctor selected by the user; Retrieve all the symptom keywords input by the user and mark them as the target keyword set B, B = {b1, b2, b3,..., bn}; Retrieve all the symptom keywords corresponding to the department for seeing a doctor selected by the user from the database and mark them as the reference keyword set C, C = {c1, c2, c3,..., cm}; Compare the symptom keywords in the target keyword set and the reference keyword set: If B ∩ C = B, then use the symptom keywords input by the user as the suspected diagnosis keywords; If B∩C≠B and Then obtain the number of symptom keywords in the difference set between the target keyword set B and the reference keyword set C, and mark it as the difference number F; When the difference score F ≥ F1, screen out F symptom keywords from the target keyword set B so that B ∩ C = B, and transmit a signal for re-entering symptom keywords to the request unit through a processor until the corresponding difference score F < F1, and then use the symptom keywords in the corresponding target keyword set as the suspected diagnosis keywords; like Then, the processor transmits a signal of re-entering the symptom keyword and the visiting department to the request unit; Wherein, m and n are positive integers, and F1 is a preset value; After the preliminary generation unit pre-screens the information transmitted by the request unit in combination with the database, it generates a preliminary medical visit form, and the preliminary medical visit form includes "suspected diagnosis keywords + department for seeing a doctor".

2. The AI-based health care service recommendation system according to claim 1, characterized in that: When the pointing and division unit recommends medical visit hospitals according to the preliminary medical visit form, the following algorithm is executed: Obtain the emergency weight D of the department for seeing a doctor corresponding to the preliminary medical visit form from the database; If D1 < emergency weight D ≤ D2, then respectively obtain the distance H, cure rate Y between all the hospitals with this department for seeing a doctor and the user, and calculate the hospital recommendation value TP in combination with the distance H and cure rate Y, and divide the M hospitals with the largest hospital recommendation value TP into the pointed hospitals; If the emergency weight D > D2, then respectively obtain the distance H, the fastest medical visit time, and the cure rate Y between all the hospitals with this department for seeing a doctor and the user, and calculate the hospital recommendation value TP1 in combination with the distance H, the fastest medical visit time, and the cure rate Y, and divide the M hospitals with the largest hospital recommendation value TP1 into the pointed hospitals; If the emergency weight D ≤ D1, then respectively obtain the distance H, the fastest medical visit time between all the hospitals with this department for seeing a doctor and the user, and calculate the hospital recommendation value TP2 in combination with the distance H and the fastest medical visit time, and divide the M hospitals with the largest hospital recommendation value TP2 into the pointed hospitals; Wherein, M is a preset value.

3. The AI-based health care service recommendation system according to claim 2, characterized in that: The method for the pointing and division unit to calculate the hospital recommendation value TP in combination with the distance H and cure rate Y is: Obtain all the medical visit users of the department for seeing a doctor within one year, and retrieve the medicine-taking status of the medical visit users from the database; Eliminate the users who did not pick up the medicine, and mark the total number of remaining users as the total number of actual visits; Get the number of cured users among the users who have received the medicine, marked as the total number of cured users, which can be obtained through the user's selection after the consultation. The user can select "cured" or "uncured" through the client; Cure rate Y = (total number of cured cases / total number of actual diagnosis cases)*β, where β is a preset value and 0<β<1; The recommended value TP=α*H+λ*Y, where α+λ=1 and α≥λ, and α and λ are both preset values.

4. The AI-based health care service recommendation system according to claim 3, characterized in that: The method for calculating the hospital recommended value TP1 by combining distance H, fastest consultation time, and cure rate Y is: The hospital recommended value TP is calculated based on the combined distance H and cure rate Y; Arrange the hospital recommendation values ​​TP in descending order, and mark the top 5 hospitals as the proposed hospitals; Get the fastest consultation time of the corresponding department of the proposed hospital E: Fastest consultation time Among them, PR is the current number of people in the queue, ST is the average consultation time per person, and dT is the expected arrival time; The hospital recommended value TP1=TP+E.

5. The AI-based health care service recommendation system according to claim 4, characterized in that: The method for calculating the hospital recommended value TP2 by combining the distance H and the fastest consultation time E is: The hospital recommended value TP2 = α*H+λ*E, where α+λ = 1 and α≥λ, and α and λ are both preset values.

6. The AI-based health care service recommendation system according to claim 1, characterized in that: The method in which the matching unit matches the intended hospital for the user from the recommended hospitals in combination with the additional request uploaded by the user is: Obtaining an additional request from the user, where the additional request is a proposed inspection item input by the user; If the recommended hospital includes the items to be examined, the recommended hospital will be directly matched to the user; Otherwise, the hospital with the highest recommendation value and that has the intended examination items is selected from the non-recommended hospitals and matched to the user.

Images