Medical accompanying system

Abstract

The invention relates to the technical field of medical accompanying diagnosis, and discloses a medical accompanying system, which comprises an image acquisition module for acquiring real-time images, a storage module for storing user and accompanying information, an operation database for managing operation functions, a statistical module for recording accompanying states and times, and an identity confirmation module for performing authority judgment and comparing historical images. And the processing module generates an accompanying recommendation table according to the authority and the image comparison result in combination with the statistics or matching value. According to the invention, intelligence and individuation of accompanying recommendation are realized, and the accuracy and efficiency of accompanying service are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical escort, and more particularly, to a medical escort system. Background Art

[0002] The modern medical system is cumbersome and highly intelligent for the elderly, making it difficult for them to quickly master the intelligent medical treatment process. The elderly may have communication barriers and have difficulty understanding the doctor's professional terms or treatment plans. The accompanying personnel can help coordinate the medical treatment time, optimize the medical treatment process, and alleviate the communication barriers between the elderly and the doctor. The accompanying personnel can provide a sense of security for the elderly, help them better cooperate with the treatment, and can also provide convenience for the children who are inconvenient to accompany.

[0003] With the increasing demand for medical services, the escort service, as an important auxiliary link during the patient's hospitalization, is gradually developing from a spontaneous family-based model to a platform-based and professional model. Currently, some medical institutions or third-party platforms have established a database of escort personnel to support patients in online booking of escort personnel. However, the existing technology still has the following problems: The recommendation mechanism for escort personnel is usually relatively simple, often only sorting based on ratings or service times, lacking a comprehensive consideration of key factors such as service stability and historical matching relationships, resulting in low accuracy of the recommendation results and affecting the patient experience. Summary of the Invention

[0004] To solve the above technical problems, the present invention proposes a medical escort system, including: An image acquisition module configured to collect real-time images of the operator; A storage module configured to store user information, escort personnel information, and reservation information and store the real-time images as historical images; An operation database configured to store operation functions, and the types of the operation functions include registration, login, reservation, information deletion, information addition, and information modification; A statistics module configured to collect the number of escort personnel, escort status, and escort times within the current medical institution; the escort status includes idle and non-idle, and the escort times are the number of times within a preset fixed cycle duration; An identity confirmation module configured to perform a first authority judgment or a second authority judgment according to the operation function retrieved by the operator to verify the authority of the operator, and the authorities include patient level and non-patient level; The processing module is configured to, when it is determined that the permission is at the patient level, obtain the real-time image of the latest current operator, compare the real-time image with the stored historical image, and determine whether the current operator exists in the historical image; if it is determined that the current operator does not exist, store the real-time image in the storage module, and sort each caregiver based on the data of the statistics module to form a caregiver recommendation list; If it is determined that the current operator exists, calculate the matching value between the current operator and each caregiver according to the appointment information and the caregiver information, and sort the caregivers based on the matching value to form a caregiver recommendation list.

[0005] Further, the operators include patients, users, and caregivers; The user information includes: user ID, user name, user password, age, gender, hospital where the user is located, department, and contact phone number; The caregiver information includes: caregiver ID, caregiver name, appointment time, and caregiver score; The appointment information includes: appointment ID, user ID, appointment time, and user name.

[0006] Further, the first permission determination includes: when the operator selects to execute the appointment, determining that the permission of the operator is at the patient level; The second permission determination includes: when the operator selects the login, determining the permission according to the input user ID, user password, or caregiver ID.

[0007] Further, when determining whether the current operator exists in the historical image, compare the real-time image with each historical image according to face recognition, and set a similarity threshold between the real-time image and the historical image; when the similarity between the real-time image and any historical image is greater than or equal to the similarity threshold, the determination result is output as existing; when the similarity is less than the similarity threshold, the determination result is output as not existing.

[0008] Further, sorting each caregiver based on the data of the statistics module to form a caregiver recommendation list includes: Obtain the caregiving status of all the caregivers, exclude the caregivers whose caregiving status is not idle, and generate an initial caregiver recommendation list based on the remaining caregivers; Calculate the priority value according to the caregiving score and the number of caregiving times, sort the remaining caregivers from largest to smallest according to the priority value, and form a revised caregiver recommendation list.

[0009] Further, the priority value satisfies the following relationship: ; Among them, is the priority value, is the weighted scoring trend term, is the load compensation term, is the penalty term, is the fairness compensation term, , , , are the influence coefficients respectively.

[0010] Furthermore, the weighted scoring trend term, the load compensation term, the penalty term and the fairness compensation term are calculated through the following relationships respectively: ; ; ; ; Among them, is the i-th caregiver score, is the time decay coefficient, , is the number of days since the i-th caregiver score, is the number of caregiver times within a fixed period, is the maximum number of caregiver times among all caregivers within a fixed period, is the standard deviation of the number of caregiver times of the current caregiver within a fixed period, is the maximum standard deviation among all caregivers within a fixed period, is the cancellation rate, and the cancellation rate is the number of refused caregiver times divided by the number of requested caregiver times, is the number of refused caregiver times, is the number of days since the last completed caregiving.

[0011] Furthermore, the influence coefficient is determined by the entropy weight method, including: Construct an evaluation matrix, where each row represents a caregiver and each column is an index, and the indexes include the weighted scoring trend term, the load compensation term, the penalty term and the fairness compensation term; perform positive normalization on all data in the evaluation matrix, calculate the entropy value and the difference coefficient of each column, and finally normalize the weights to obtain the influence coefficient.

[0012] Furthermore, if it is judged that there is, the matching value satisfies the following relationship: ; Among them, is the matching value, is the number of times this caregiver has been matched with the patient in the historical record, is the maximum number of times all the accompanying personnel have been matched with the patient in the historical records, is the priority value.

[0013] Furthermore, the accompanying personnel are sorted according to the descending order of the matching values to generate an accompanying recommendation form.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: Through the face recognition comparison mechanism and the historical image library, combined with the permission judgment, the patient's identity can be effectively identified, ensuring the pertinence and accuracy of the accompanying recommendation. The system no longer relies solely on scoring or the number of services, but integrates multiple factors such as the idle state, scoring trend, service frequency, penalty items, and fairness. The entropy weight method is used to assign reasonable weights to improve the scientificity and fairness of the ranking. The statistical module collects the accompanying status and number of times in real time, providing a decision-making basis for the scheduling of accompanying personnel and resource allocation, and improving the scheduling efficiency of accompanying services. The permission judgment module divides the operators into patient level and non-patient level, and performs permission judgments at different levels according to the operation content, preventing unauthorized operations and data leakage. By combining the historical matching records and the current priority value to calculate the accompanying matching value, intelligent recommendation based on individual preferences is realized, improving the patient satisfaction and service coherence. The system can assist elderly patients to complete the medical treatment process more conveniently, relieve the communication barrier between them and the doctor, provide emotional support and operation assistance, and reflect the humanized care. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered as limiting the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings: Figure 1 is the functional framework diagram of the medical accompanying system provided by the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The exemplary embodiments disclosed in the present application will be described in more detail below with reference to the drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art. It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the drawings and in conjunction with the embodiments.

[0017] Refer to Figure 1 As shown, the embodiment of the present invention provides a medical accompanying system, including: An image acquisition module, configured to collect real-time images of the operator; A storage module, configured to store user information, caregiver information, and appointment information, and store the real-time image as a historical image; An operation database, configured to store operation functions, and the types of operation functions include registration, login, appointment, information deletion, information addition, and information modification; A statistics module, configured to collect the number of caregivers, caregiving status, and caregiving times within the current medical institution; the caregiving status includes idle and non-idle, and the caregiving times are the number of times within a preset fixed cycle duration; An identity confirmation module, configured to perform a first authority judgment or a second authority judgment according to the operation function retrieved by the operator to verify the operator's authority, and the authorities include patient level and non-patient level; A processing module, configured to, when determining that the authority is at the patient level, obtain the latest real-time image of the current operator, compare the real-time image with the stored historical image, and determine whether the current operator exists in the historical image; if it is determined that the operator does not exist, store the real-time image in the storage module, and sort each caregiver based on the data of the statistics module to form a caregiver recommendation list; If it is determined that the operator exists, calculate the matching value between the current operator and each caregiver according to the appointment information and caregiver information, and sort the caregivers based on the matching value to form a caregiver recommendation list.

[0018] It should be noted that by collecting real-time images through the image acquisition module and combining face recognition technology to compare the operator with the historical images, the accuracy of identity recognition is improved, and the problem of misusing others' information for operation is effectively prevented, ensuring the safety and compliance of system operations.

[0019] The system realizes the judgment of the operator's authority level (patient level or non-patient level) through the identity confirmation module, and combines the operation function types for matching, so that all operations of the system are executed within the scope of authority, avoiding unauthorized access or over-authorization operations.

[0020] After the processing module determines the identity, it triggers different recommendation mechanisms according to different identity paths. For patients identified for the first time, a recommendation list is constructed through objective indicators such as caregiving status and historical caregiving times; for patients with successful historical identification, the matching value calculation logic is further introduced, and combined with the historical matching relationship and priority sorting factors, to achieve a more personalized caregiver recommendation.

[0021] The statistics module collects the caregiving status and service data of caregivers within the medical institution in real time, providing dynamic data support for the system, which not only helps the system accurately allocate resources, but also improves the organizational efficiency and response ability of the caregiving service within the medical institution.

[0022] This system particularly takes into account the problems such as inconvenient operation and insufficient information mastery when the elderly seek medical treatment. Through an automated mechanism for image acquisition and recommendation, it simplifies the operation process of patients, reduces manual input steps, and significantly improves the service experience and satisfaction of the elderly group.

[0023] In some embodiments of the present application, the operators include patients, users, and accompanying persons; User information includes: user ID, user name, user password, age, gender, hospital where the user is located, department, and contact phone number; Accompanying person information includes: accompanying person ID, accompanying person name, appointment time, and accompanying score; Reservation information includes: reservation ID, user ID, reservation time, and user name.

[0024] In some embodiments of the present application, the first permission judgment includes: when the operator selects to execute a reservation, it is judged that the operator's permission is at the patient level; The second permission judgment includes: when the operator selects to log in, the permission is determined according to the input user ID, user password, or accompanying person ID.

[0025] It should be noted that by distinguishing the judgment logic triggered by reservation operations and login operations, the system can clearly distinguish the identity categories of operators, effectively demarcate the permission boundaries between patients and accompanying persons, avoid non-patient identities from mistakenly performing patient-exclusive operations, and improve the compliance and controllability of system operation.

[0026] For patient users, the permission judgment is directly triggered when initiating a reservation behavior, without an additional login authentication process, reducing the operation complexity; while for the login operation, the permission is determined through account information (ID and password), enabling both accompanying persons and ordinary users to verify their identities through a unified entry, with clear operation logic and smooth process.

[0027] The dual judgment mechanism effectively prevents malicious users from bypassing permission control for illegal operations. Especially in sensitive processes such as patient privacy and accompanying service allocation, it ensures that each operation has a clear and credible permission basis, providing a solid security guarantee for the medical escort system.

[0028] The clear layering of the permission judgment mechanism facilitates subsequent system expansion. For example, when introducing roles such as medical staff permissions and administrator permissions, only the judgment conditions need to be expanded to seamlessly connect, with good scalability and maintainability.

[0029] In some embodiments of the present application, when determining whether the current operator exists in the historical images, the real-time image is compared with each historical image according to face recognition to set a similarity threshold between the real-time image and the historical image; when the similarity between the real-time image and any historical image is greater than or equal to the similarity threshold, the judgment result is output as existent; when the similarity is less than the similarity threshold, the judgment result is output as non-existent.

[0030] It should be noted that through the image comparison technology based on face recognition, the system can automatically identify whether the user is a historical user without disturbing the user operation process, realizing non-contact and password-free identity confirmation, improving the usage convenience and user experience, and is especially suitable for groups such as the elderly and those with inconvenient mobility who are not good at operating traditional systems.

[0031] Setting the similarity threshold as the judgment basis can, to a certain extent, suppress the recognition errors caused by factors such as shooting angles and lighting changes, effectively balance the sensitivity and accuracy of recognition, and ensure that the system neither over-recognizes nor easily misses historical users.

[0032] By matching and confirming each real-time image with the historical images, the misappropriation or repeated allocation of escort resources can be avoided, ensuring the stable and reliable identity of the system service objects; at the same time, when the user enters the system for the first time, the image can be automatically saved for file establishment to realize the subsequent traceability and anti-counterfeiting of the identity. Once identified as a user in the historical images, the redundant identity verification process can be skipped, and personalized recommendations and service matching can be directly carried out, effectively improving the service efficiency and enhancing the intelligent level of the system. The comparison mechanism can be applied to a variety of mainstream face recognition algorithms and modules, and the similarity threshold can be dynamically adjusted according to actual needs, so as to adapt to the different requirements for recognition accuracy and speed in different medical institutions or usage scenarios.

[0033] In some embodiments of the present application, each escort is sorted based on the data of the statistical module to form an escort recommendation list, including: Obtain the escort status of all escorts, eliminate the escorts with non-idle escort status, and generate an initial escort recommendation list according to the remaining escorts; Calculate the priority value according to the escort score and the number of escort times, sort the remaining escorts from largest to smallest according to the priority value, and form a revised escort recommendation list.

[0034] It should be noted that each escort is sorted based on the data of the statistical module to form an escort recommendation list, including: Obtain the escort status of all escorts; eliminate the escorts with non-idle escort status; generate an initial escort recommendation list according to the remaining escorts; calculate the priority value according to the escort score and the number of escort times; sort the remaining escorts from largest to smallest according to the priority value to form a revised escort recommendation list.

[0035] The above solution has the following beneficial effects: The system first eliminates the accompanying personnel who are currently in the "non-idle" state, ensuring that the accompanying personnel in the recommended list are all in a serviceable state, avoiding the recommendation of ineffective candidates, thereby improving the recommendation efficiency and accuracy, and reducing the user's waiting time or the experience gap caused by failed recommendations. The priority value of the accompanying personnel is jointly determined by comprehensive factors such as "accompanying score" and "accompanying times". It not only considers their historical service quality (reflected by the score), but also avoids over-reliance on a small number of high-scoring personnel (adjusted by the number of times), ensuring the fairness and sustainability of the recommendation system. Through the real-time analysis of the multi-dimensional status data of the accompanying personnel by the statistical module, the system can dynamically adjust the recommendation order before each recommendation, construct a "corrected accompanying recommendation form", which is more intelligent, more real-time and more in line with the actual service scenario than the static scoring and sorting method. The recommendation form based on sorting can achieve rapid and accurate matching of users with different needs, helping to improve the utilization rate of accompanying personnel, reducing the situation of idle rotation or uneven distribution of accompanying resources, and thus improving the overall operation efficiency of the medical accompanying system. The recommendation form structure formed by this sorting mechanism is clear and highly adjustable, and can be flexibly combined with the matching module, face recognition module, etc. to implement higher-order customized recommendation logic, providing a good foundation for the expansion and intelligent evolution of the system.

[0036] In some embodiments of the present application, the priority value satisfies the following relationship: ; Wherein, is the priority value, is the weighted scoring trend term, is the load compensation term, is the penalty term, is the fairness compensation term, , , , are the influence coefficients respectively.

[0037] It should be noted that a multi-dimensional comprehensive evaluation mechanism is constructed to improve the scientificity of recommendation Through multiple dimensions such as weighted scoring trend, load adjustment, service penalty and fairness compensation, the priority value calculation model breaks through the traditional single method of "sorting by score", comprehensively reflects the service history and current status of the accompanying personnel, and makes the recommendation more accurate and comprehensive.

[0038] The weighted scoring trend item Se can reflect the scoring performance of the escort staff during the long-term service process. The system assigns a higher priority value to those with a high scoring trend, thereby guiding the escort staff to improve the service quality and promoting healthy competition. The load compensation item Fc can effectively avoid the overuse of high-scoring personnel, contribute to achieving the dynamic balance of service frequencies, extend the working cycle of the escort staff, and improve the overall system stability. The penalty item Pp can deduct points from those with a high cancellation rate or poor service records, making the recommendation more inclined to stable and reliable escort resources, and enhancing the patient experience and service success rate. The fairness compensation item Eb gives appropriate weights to the escort staff who have not been scheduled for a long time, effectively reducing the problems of resource idling or "cold resources", and enhancing the operation efficiency and fairness of the entire escort manpower system. Each influence coefficient (α, β, γ, δ) can be flexibly adjusted according to the operation requirements of medical institutions or specific business scenarios, realizing the priority ranking scheme of different strategies, and having strong scalability and practicability.

[0039] In some embodiments of the present application, the weighted scoring trend item, the load compensation item, the penalty item, and the fairness compensation item are calculated through the following relationships respectively: ; ; ; ; Among them, is the i-th escort score, is the time decay coefficient, , is the number of days since the i-th escort score, is the number of escort times within a fixed period, is the maximum number of escort times among all escort staff within a fixed period, is the standard deviation of the number of escort times of the current escort staff within a fixed period, is the maximum standard deviation among all escort staff within a fixed period, is the cancellation rate, and the cancellation rate is the number of rejected escort times divided by the number of requested escort times, is the number of rejected escort times, is the number of days since the last completed escort.

[0040] It should be noted that The calculation method of is as follows: Assume that the fixed period is set to 5 days, and the number of escort times of a certain escort staff within the current fixed period is 1, 0, 2, 3, 1 in sequence every day. Then the average escort per day is (1 + 0 + 2 + 3 + 1) / 5 = 1.4 times. Taking 1.4 as the average value and substituting it into the standard deviation calculation formula, the standard deviation is 1.0198.

[0041] The priority value is used to measure the recommended priority of the escort at the current moment; the weighted scoring trend item represents the weighted trend of the escort's historical scores; the load compensation item represents the adaptability of the escort's current workload; the penalty item is used to quantify the impact of service dishonesty caused by refusing to provide escort; the fairness compensation item is used to increase the recommendation probability of the escort who has not been selected for a long time.

[0042] The calculation method of the weighted scoring trend item is: sum all scores after multiplying them by their time decay weights, and normalize the total weight. The time decay weight is jointly determined by the number of days and the time decay coefficient. The newer the score, the higher its weight.

[0043] This item can ensure that the system pays more attention to the escorts with good recent service performance, while weakening the influence of old scores, so as to reflect the real and effective current service ability.

[0044] The calculation method of the load compensation item is: obtained by multiplying two factors. The first factor is "1 minus the ratio of the current number of escort times to the maximum number of escort times", indicating that the higher the usage frequency of the escort, the smaller its value; the second factor is "1 minus the ratio of the standard deviation of the current escort's number of escort times to the maximum standard deviation", indicating that the more fluctuating the escort frequency, the smaller its value.

[0045] This item is used to prevent highly rated escorts from being called frequently, resulting in fatigue or a decline in service quality, and helps to achieve a reasonable and balanced distribution of tasks in human resources.

[0046] The calculation method of the penalty item is: the penalty item is equal to the refusal rate multiplied by "1 plus the natural logarithm of the number of times of refusing to provide escort". This formula can dynamically increase the penalty intensity, and the penalty value increases with the number of refusals.

[0047] This item encourages escorts to accept tasks, reduces the behavior of breaking appointments, and improves the schedulability of the overall service and the stability of the system.

[0048] The calculation method of the fairness compensation item is: the fairness compensation item is equal to "the square of the number of days since the last escort task was completed".

[0049] This item is used to increase the recommendation probability of the escorts who have not been called for a long time, prevent resource idleness, and improve the fairness and human resource utilization rate of the system.

[0050] Through the construction of the priority value formula and the introduction of multiple factors, the escort recommendation mechanism provided in this embodiment has the following beneficial effects: considering multiple key indicators such as scoring, usage frequency, appointment-breaking behavior, and fairness simultaneously, realizing multi-dimensional regulation of the recommendation results; improving the accuracy and rationality of scheduling recommendations, meeting the dual requirements of service quality and resource balance in the actual operation of medical institutions; providing an adjustable parameter space, facilitating the flexible adjustment of the recommendation logic according to the management strategies of different institutions; enhancing the dynamic adaptability and actual operation stability of the system, and improving patient satisfaction and the enthusiasm of escort personnel.

[0051] In some embodiments of the present application, the influence coefficient is determined by the entropy weight method, including: Construct an evaluation matrix, where each row represents an escort personnel and each column is an indicator, and the indicators include the weighted score trend item, the load compensation item, the penalty item, and the fairness compensation item; perform positive normalization on all data in the evaluation matrix, calculate the entropy value and the difference coefficient of each column, and finally normalize the weight to obtain the influence coefficient.

[0052] It should be noted that, first, construct the original evaluation matrix. Each row in this evaluation matrix corresponds to an escort personnel, and each column corresponds to an evaluation indicator. The evaluation indicators used include four aspects: weighted score trend, load compensation, penalty degree, and fairness compensation. Through this matrix, the performance of each escort personnel in each dimension can be comprehensively reflected.

[0053] Secondly, perform dimensionless processing on the numerical values of each indicator in the original evaluation matrix. Since different indicators have different dimensions and dimension scales, in order to ensure the comparability between indicators, unified processing is required. The specific processing method is to uniformly convert all numerical values into positive standardized values between zero and one, that is, convert the original data into a form where the larger the numerical value, the better the performance.

[0054] Then, on the basis of normalization, calculate the information entropy of each evaluation indicator. Information entropy is used to measure the degree of information difference provided by a certain indicator among all escort personnel. If the entropy value of an indicator is higher, it means that the numerical distribution of this indicator among all escort personnel is more average, and the discrimination ability is weaker; if the entropy value of an indicator is lower, it means that this indicator has a stronger discrimination ability and a greater impact on sorting.

[0055] Next, according to the information entropy of each indicator, calculate its corresponding difference coefficient, that is, the information utility value. The size of the difference coefficient represents the degree of influence of this indicator on the final decision. The larger the difference coefficient, the greater the role of this indicator in distinguishing the priority of escort personnel.

[0056] Finally, the coefficient of variation of each index is normalized to obtain the final weight value of each index. The sum of the weight values of each index is one, which are respectively used as the influence coefficients of the four indexes of weighted scoring trend, load compensation, penalty term, and fairness compensation in the calculation of the priority value.

[0057] The influence coefficients determined in the above manner have the following beneficial effects: It avoids the subjective deviation caused by artificial weighting, making the entire recommendation mechanism more objective and transparent; it can dynamically adjust the importance of each index according to actual data, improving the adaptive ability and decision-making accuracy of the system; it ensures the scientificity and rationality of the recommendation algorithm, effectively enhancing the credibility and fairness of the sorting results of the accompanying staff; it realizes intelligent decision-making driven by data, which helps the medical escort system to more reasonably match escort resources and optimize service efficiency and user experience.

[0058] In some embodiments of the present application, when it is judged that there is a match, the match value satisfies the following relationship: ; Wherein, is the match value, is the number of times this escort has been matched with the patient in the historical record, is the maximum number of times all escorts have been matched with the patient in the historical record, is the priority value.

[0059] It should be noted that, is not the maximum number of times within a fixed period, but the maximum cumulative number of times an escort has accompanied a patient in the historical record.

[0060] In some embodiments of the present application, the escorts are sorted from largest to smallest according to the match value to generate an escort recommendation form.

[0061] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: the specific implementation manners of the present invention can still be modified or equivalently replaced, and any modification or equivalent replacement without departing from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.

Claims

1. A medical accompanying system, characterized in that: include: An image acquisition module configured to acquire a real-time image of an operator; A storage module, configured to store user information, accompanying personnel information and appointment information and store the real-time image as a historical image; An operation database is configured to store operation functions, wherein the types of the operation functions include registration, login, reservation, information deletion, information addition and information modification; A statistical module is configured to collect the number of accompanying personnel, accompanying status and accompanying times in the current medical institution; the accompanying status includes idle and non-idle, and the accompanying times are the times within a preset fixed period; An identity confirmation module is configured to perform a first authority judgment or a second authority judgment according to the operation function called by the operator to verify the authority of the operator, wherein the authority includes a patient level and a non-patient level; The processing module is configured to, when it is determined that the authority is at the patient level, obtain the latest real-time image of the current operator, compare the real-time image with the stored historical image, and determine whether the current operator exists in the historical image; if it is determined that the current operator does not exist, store the real-time image in the storage module, and sort each accompanying person based on the data of the statistical module to form an accompanying recommendation table; Forming the accompanying recommendation table specifically includes: obtaining the accompanying status of all the accompanying personnel, eliminating the accompanying personnel whose accompanying status is not idle, and generating an initial accompanying recommendation table according to the remaining accompanying personnel; calculating the priority value according to the accompanying score and the number of accompanying times, sorting the remaining accompanying personnel from large to small according to the priority value, and forming a revised accompanying recommendation table; the priority value is calculated and obtained according to the weighted score trend item, load compensation item, penalty item and fairness compensation item of the accompanying personnel; If it is determined to exist, the matching value between the current operator and each accompanying person is calculated according to the reservation information and the accompanying person information, and the accompanying persons are sorted based on the matching value to form an accompanying recommendation table.

2. The medical accompanying system according to claim 1, characterized in that: The operators include patients, users and accompanying personnel; The user information includes: user ID, user name, user password, age, gender, hospital, department and contact number; The accompanying personnel information includes: accompanying personnel ID, accompanying personnel name, appointment time and accompanying personnel score; The reservation information includes: reservation ID, user ID, reservation time and user name.

3. The medical accompanying system according to claim 2, characterized in that: The first authority determination includes: when the operator selects to execute the appointment, determining that the operator's authority is at the patient level; The second authority determination includes: when the operator selects the login, determining the authority according to the input user ID, user password or accompanying personnel ID.

4. The medical accompanying system according to claim 3, characterized in that: When judging whether the current operator exists in the historical image, the real-time image is compared with each historical image according to face recognition, and a similarity threshold between the real-time image and the historical image is set; when the similarity between the real-time image and any historical image is greater than or equal to the similarity threshold, the output judgment result is existence; when the similarity is less than the similarity threshold, the output judgment result is non-existence.

5. The medical accompanying system according to claim 4, characterized in that: The priority values ​​satisfy the following relationship: ； in, is the priority value, is the weighted scoring trend item, is the load compensation term, is the penalty item, For fair compensation, , , , are the influence coefficients respectively.

6. The medical accompanying system according to claim 5, characterized in that: The weighted scoring trend item, load compensation item, penalty item and fairness compensation item are calculated respectively by the following relationships: ； ； ； ； in, Score the i-th caregiver, is the time attenuation coefficient, , is the number of days since the i-th care rating, is the number of accompanying times within a fixed period, is the maximum number of accompanying times among all accompanying personnel within a fixed period, is the standard deviation of the number of accompanying times of the current accompanying personnel in a fixed period, is the maximum standard deviation among all caregivers in a fixed period, is the no-show rate, which is the number of times of refusing to accompany divided by the number of times of requesting to accompany, To refuse the number of escorts, The number of days since the last care was completed.

7. The medical accompanying system according to claim 6, characterized in that: The influence coefficient is determined by the entropy weight method, including: An evaluation matrix is ​​constructed, in which each row represents a caregiver and each column is an indicator, including a weighted scoring trend item, a load compensation item, a penalty item and a fairness compensation item; all data in the evaluation matrix are forward normalized, the entropy value and the difference coefficient of each column are calculated, and finally the weight is normalized to obtain the influence coefficient.

8. The medical accompanying system according to claim 7, characterized in that: If it is judged to exist, the matching value satisfies the following relationship: ； in, is the matching value, is the number of times the accompanying person has been matched with the patient in the historical records. is the maximum number of matches between all accompanying personnel and patients in the historical records, is the priority value.

9. The medical accompanying system according to claim 8, characterized in that: The accompanying persons are sorted from large to small according to the matching values ​​to generate an accompanying recommendation table.

Images