Appointment scheduling method and system for remote medical patients

Abstract

A method for scheduling appointments for telemedicine patients includes: establishing a patient selection model, and the patient selection model outputs the probability of telemedicine patients selecting alternative options in combination with the preferences and no-show behavior of telemedicine patients, wherein the preferences include: appointment time preference and preference for the selection of expert doctors; and constructing a telemedicine appointment scheduling model based on probability, and the telemedicine appointment scheduling model outputs a scheduling arrangement that maximizes the total benefit. The relationship between the preferences of telemedicine patients, utility theory and the no-show rate of telemedicine patients is comprehensively considered, and the telemedicine appointment scheduling model gives a scheduling arrangement that maximizes the total benefit, reduces the no-show rate of telemedicine patients, and the loss of medical equipment resources and human resources, thereby maximizing the total benefit of the superior hospital of the doctor expert.

Description

Technical Field

[0001] The present application relates to the research field of appointment scheduling for patients, and in particular to an appointment scheduling method and system for telemedicine patients. Background Art

[0002] In the field of medical service operation, the scheduling and optimization of medical resources has always been the focus and hotspot of research. At present, the research scenarios for appointment scheduling of medical resources mainly include outpatient clinics, operating rooms and telemedicine visits. Among them, the research on appointment scheduling of outpatient and operating room patients at home and abroad has formed a relatively mature and complete research system. However, due to the characteristics of telemedicine such as "multiple subjects, multiple demands and multiple factors", the results of traditional outpatient and operating room appointment scheduling are difficult to be directly applied to telemedicine scenarios.

[0003] In addition, different telemedicine patients have certain preferences for the products and services of different medical service agencies, that is, the degree of preference for the products or services provided by medical service agencies. In the traditional outpatient and operating room appointment scheduling research, it was found that patients' preferences are mainly reflected in their preferences for appointment times and their different degrees of loyalty to designated or specific doctors. However, since it involves medical health, the privacy of telemedicine patients must be taken into account, and it is also difficult to store targeted data on the preferences of telemedicine patients in actual operations, so that the actual data on telemedicine patients' selection preferences is limited. Of course, factors that affect the appointment scheduling of telemedicine patients also include: patient no-shows, and patient service time. However, there are few existing technologies that take into account both patient preferences and patient no-show behaviors to schedule appointments for telemedicine patients. Summary of the invention

[0004] The present application provides an appointment scheduling method and system for telemedicine patients, in order to solve or partially solve the above-mentioned problems involved in the background technology or at least one other deficiency in the prior art.

[0005] The present application proposes an appointment scheduling method for telemedicine patients, which may include: establishing a patient selection model, and the patient selection model combines the telemedicine patients' preferences and no-show behavior to output the selection probability of the telemedicine patients choosing alternative options, wherein the preferences include: appointment time preference, and preference for expert doctors; and based on the selection probability, constructing a telemedicine appointment scheduling model, and the telemedicine appointment scheduling model outputs a scheduling arrangement that maximizes the total benefit.

[0006] In some embodiments, a telemedicine patient is defined as a decision maker, and multiple alternatives corresponding to the telemedicine patient are determined. A patient selection model is constructed based on the decision maker and the multiple alternatives. The patient selection model combines the preferences and no-show behavior of the telemedicine patient to output the selection probability of the telemedicine patient selecting the alternative.

[0007] In some implementations, the probability of a telemedicine patient selecting an alternative is expressed as:

[0008]

[0009] Where i represents the serial number of the telemedicine patient, i=1,2,...,M; j represents the category serial number of the alternative plan, j=1,2,...,q,...,J; V ij Denotes the utility term determined by the i-th telemedicine patient.

[0010] In some embodiments, a telemedicine appointment scheduling model is constructed based on the selection probability, and the telemedicine appointment scheduling model outputs a scheduling arrangement that maximizes the total benefit, which may include: constructing multiple objective functions of the telemedicine appointment scheduling model from four aspects: the income of expert doctors treating telemedicine patients, the waiting cost of telemedicine patients in the appointment system, the overtime cost of expert doctors, and the sunk cost of expert doctors according to the selection probability; integrating multiple objective functions to establish a telemedicine appointment scheduling model with the goal of maximizing the total benefit; and the telemedicine appointment scheduling model outputs a scheduling arrangement that maximizes the total benefit.

[0011] In some implementations, the telemedicine appointment scheduling model may be expressed as:

[0012]

[0013] Among them, θ j represents the no-show rate of telemedicine patients who meet the jth alternative; n represents the sequence number of the time slot, n = 1, 2, ..., N; x n represents the number of telemedicine patients assigned to the nth consultation period; c w E[W(x)] represents the waiting cost of telemedicine patients; c I represents the unit idle cost of expert doctors in remote consultation; B0(x n-1 ) represents the probability that there are 0 telemedicine patients waiting for consultation at the end of the n-1th consultation period; π represents the unit penalty cost of telemedicine patients who do not respond to traditional outpatient appointments due to expert doctors participating in remote consultations; P ij represents the probability that the i-th telemedicine patient chooses the j-th alternative; k represents the number of remaining telemedicine patients; c oO(x) represents the overtime cost of expert doctors.

[0014] The present application also proposes such an appointment scheduling system for telemedicine patients, which may include: a selection probability acquisition module and a scheduling module. The selection probability acquisition module is used to establish a patient selection model, and the patient selection model combines the preferences and no-show behavior of telemedicine patients to output the selection probability of telemedicine patients choosing alternative options, wherein the preferences include: appointment time preference, and preference for expert doctors; the scheduling module is used to build a telemedicine appointment scheduling model based on the selection probability, and the telemedicine appointment scheduling model outputs a scheduling arrangement that maximizes the total benefit.

[0015] In some embodiments, the execution steps of the selection probability acquisition module may include: defining the telemedicine patient as a decision maker, and determining multiple alternatives corresponding to the telemedicine patient. According to the decision maker and the multiple alternatives, a patient selection model is constructed. The patient selection model combines the preferences and no-show behavior of the telemedicine patient to output the selection probability of the telemedicine patient selecting the alternative.

[0016] In some implementations, the probability of a telemedicine patient selecting an alternative may be expressed as:

[0017]

[0018] Where i represents the serial number of the telemedicine patient, i=1,2,...,M; j represents the category serial number of the alternative plan, j=1,2,...,q,...,J; V ij Denotes the utility term determined by the i-th telemedicine patient.

[0019] In some embodiments, the execution steps of the scheduling module may include: constructing multiple objective functions of the telemedicine appointment scheduling model from four aspects, namely, the income of expert doctors treating telemedicine patients, the waiting cost of telemedicine patients in the appointment system, the overtime cost of expert doctors, and the sunk cost of expert doctors, based on the selection probability; integrating multiple objective functions to establish a telemedicine appointment scheduling model with the goal of maximizing total benefits; and outputting a scheduling arrangement that maximizes the total benefits from the telemedicine appointment scheduling model.

[0020] In some implementations, the telemedicine appointment scheduling model may be expressed as:

[0021]

[0022] Among them, θ j represents the no-show rate of telemedicine patients who meet the jth alternative; n represents the sequence number of the time slot, n = 1, 2, ..., N; x nrepresents the number of telemedicine patients assigned to the nth consultation period; c w E[W(x)] represents the waiting cost of telemedicine patients; c I represents the unit idle cost of expert doctors in remote consultation; B0(x n-1 ) represents the probability that there are 0 telemedicine patients waiting for consultation at the end of the n-1th consultation period; π represents the unit penalty cost of telemedicine patients who do not respond to traditional outpatient appointments due to expert doctors participating in remote consultations; P ij represents the probability that the i-th telemedicine patient chooses the j-th alternative; k represents the number of remaining telemedicine patients; c o O(x) represents the overtime cost of expert doctors.

[0023] According to the technical solution of the above-mentioned implementation mode, at least one of the following beneficial effects can be obtained.

[0024] According to an appointment scheduling method and system for telemedicine patients in accordance with one embodiment of the present application, the relationship between the preferences of telemedicine patients, utility theory and the no-show rate of telemedicine patients is comprehensively considered, and the telemedicine appointment scheduling model provides a scheduling arrangement that maximizes the total benefit, thereby reducing the no-show rate of telemedicine patients and the loss of medical equipment resources and human resources, thereby maximizing the total benefit of the superior hospitals of doctors and experts. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0026] Figure 1 is a flow chart of an appointment scheduling method for a telemedicine patient according to an exemplary embodiment of the present application;

[0027] Figure 2 is a schematic diagram of a telemedicine consultation service matching process considering a telemedicine patient's preference for appointment time and preference for expert doctor selection according to an exemplary embodiment of the present application; and

[0028] Figure 3 is a schematic diagram of an appointment scheduling system for telemedicine patients according to an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0029] In order to better understand the present application, a more detailed description will be made of various aspects of the present application with reference to the accompanying drawings. It should be understood that these detailed descriptions are only descriptions of exemplary embodiments of the present application, and are not intended to limit the scope of the present application in any way. Throughout the specification, the same reference numerals refer to the same elements. The expression "and / or" includes any and all combinations of one or more of the associated listed items.

[0030] In the accompanying drawings, the size, dimensions and shapes of the elements have been slightly adjusted for ease of illustration. The drawings are for illustration only and are not strictly drawn to scale. As used herein, the terms "substantially", "approximately" and similar terms are used as terms of approximation, not as terms of degree, and are intended to illustrate the inherent deviations in measurements or calculations that will be recognized by those of ordinary skill in the art. In addition, in this application, the order in which the steps are described does not necessarily represent the order in which these processes occur in actual operation, unless otherwise specified or can be derived from the context.

[0031] It should also be understood that expressions such as "include", "including", "have", "contain" and / or "comprising" are open rather than closed expressions in this specification, which indicate the presence of the stated features, elements and / or components, but do not exclude the presence of one or more other features, elements, components and / or combinations thereof. In addition, when expressions such as "at least one of..." appear after a list of listed features, they modify the entire list of features rather than just the individual elements in the list. In addition, when describing embodiments of the present application, "may" is used to mean "one or more embodiments of the present application". And, the term "exemplary" is intended to refer to an example or illustration.

[0032] Unless otherwise specified, all words (including engineering terms and scientific and technological terms) used in this article have the same meaning as those commonly understood by ordinary technicians in the field to which this application belongs. It should also be understood that, unless clearly stated in this application, words defined in common dictionaries should be interpreted as having the same meaning as their meaning in the context of the relevant technology, and should not be interpreted in an idealized or overly formal sense.

[0033] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0034] Figure 1 is a flow chart of an appointment scheduling method for a telemedicine patient according to an exemplary embodiment of the present application.

[0035] like Figure 1As shown, the present application proposes such an appointment scheduling method for telemedicine patients, which may include: step S1, establishing a patient selection model, and the patient selection model combines the preferences and no-show behavior of telemedicine patients to output the selection probability of telemedicine patients selecting alternative options, wherein the preferences include: appointment time preference, and preference for expert doctors. Step S2, and based on the selection probability, construct a telemedicine appointment scheduling model, and the telemedicine appointment scheduling model outputs a scheduling arrangement that maximizes the total benefit.

[0036] In some embodiments, different telemedicine patients have different preferences, such as preferences for appointment times and preferences for the selection of specialist doctors, so that any telemedicine patient may make multiple alternatives. In order to determine the probability of a telemedicine patient selecting a certain alternative, the present application constructs a patient selection model.

[0037] In some implementations, MNL (Multinominal Logit Model) can be used as a patient selection model to determine the probability of a telemedicine patient choosing a certain alternative. Specifically, the Logit model is widely used in the field of user selection behavior research as a simple discrete choice model. It is mainly used to describe the decision maker's choice of alternatives. Its theoretical basis is the decision maker's utility maximization principle, that is, it is believed that the decision maker will choose the solution with the greatest utility in the problem being studied. MNL is the basic form of the Logit model. The probability of a decision maker choosing different alternatives can be obtained by calculating the determination items of the utility function.

[0038] Based on the above, we must first determine the decision maker and alternatives in this implementation. The decision maker is the research object. Obviously, in this implementation, the decision maker can be defined as a telemedicine patient. In this implementation, the alternatives are all the options that telemedicine patients must choose from according to their preferences. In order to ensure that the alternatives have modeling significance, the alternatives should have the following three characteristics: for telemedicine patients, the alternatives must meet the mutually exclusive conditions; the options in the alternatives must be exhaustive, that is, all possible options are included, and the decision maker must choose one option from them and cannot not make a choice; and the number of options in the alternatives is finite and countable.

[0039] In some embodiments, based on the telemedicine patient's preference for appointment time and preference for specialist doctor selection, the telemedicine patient's alternative options may include: the telemedicine patient has a preference for appointment time and the appointment is met, and has a preference for specialist doctors and the appointment is met; the telemedicine patient has a preference for appointment time and the appointment is met, and has a preference for specialist doctors and the appointment is not met; the telemedicine patient has a preference for appointment time and the appointment is met, and has no preference for specialist doctors; the telemedicine patient has a preference for appointment time and the appointment is not met, and has a preference for specialist doctors and the appointment is met; the telemedicine patient has a preference for appointment time and the appointment is not met, and has a preference for specialist doctors and the appointment is not met; the telemedicine patient has a preference for appointment time and the appointment is not met, and has a preference for specialist doctors and the appointment is not met; the telemedicine patient has a preference for appointment time and the appointment is not met, and has no preference for specialist doctors; the telemedicine patient has no preference for appointment time, has a preference for specialist doctors and the appointment is met; the telemedicine patient has no preference for appointment time, has a preference for specialist doctors and the appointment is not met; and the telemedicine patient has no preference for appointment time and has no preference for specialist doctors.

[0040] Furthermore, the MNL model outputs the probability of any telemedicine patient choosing a certain alternative, where the probability of selection can characterize the utility of the telemedicine patient. In the MNL model, each telemedicine patient is asked to choose whether to participate in the remote consultation arranged by the superior hospital based on the principle of utility maximization. Specifically, since telemedicine patients have different preferences for appointment times and expert doctors, and the utility of different telemedicine patients for the same appointment time with expert doctors has not only the deterministic utility that can be estimated, but also an uncertain part of the utility that cannot be observed. Therefore, the utility of telemedicine patients participating in the consultation can be expressed as:

[0041] U ij =V ij +ε ij , (1)

[0042] In formula (1), i is the number of the telemedicine thinker, i = 1, 2, ..., N; j is the category number of the alternative solution, j = 1, 2, ..., J; V ij represents the deterministic utility term, ε ij Indicates that it includes the total utility but is not included in V ij Other unobservable random influencing factors in are independent and identically distributed.

[0043] More specifically, determine the utility term V ij It can be expressed as:

[0044] V ij =α+β1T ij +β2C ij +γFij , (2)

[0045] In formula (2), α represents the initial utility of telemedicine patients participating in consultations. The initial utility is also the utility of telemedicine patients participating in consultations when both the telemedicine patients' preferences for appointment time and their preferences for expert doctors are satisfied. At this time, the utility is the largest; T ij represents the difference between the appointment time and the consultation time of telemedicine patients; β1 is T ij The weight coefficient of ij represents the degree of fit between telemedicine patients and expert doctors; β2 is C ij The weight coefficient of ij represents the registration fee of different specialist doctors; γ is F ij The weight coefficient of .

[0046] Furthermore, the utility of telemedicine patients participating in consultation can be expressed as:

[0047] U ij =V(x ij )+ε ij , (3)

[0048] In formula (3), x ij The attribute variables representing the telemedicine patient i satisfying alternative j are the attributes of the telemedicine patient’s preferences for appointment time and for the selection of expert doctors.

[0049] Furthermore, let ε i =ε i1 , ..., ε iJ The joint density of i ).

[0050] In summary, the selection probability of telemedicine patient i choosing alternative j can be expressed as:

[0051]

[0052] In formula (4), P ij For each random term ε ij to ε iq Lower than the observed value V iq To V ij probability.

[0053] Furthermore, the probability P of telemedicine patient i choosing alternative j is ij It can be expressed as:

[0054]

[0055] In some implementations, the MNL model's thinker selection model can be used to obtain the probability of a telemedicine patient choosing a certain alternative plan after considering the telemedicine patient's no-show behavior through formula (5). It should be noted that if the category number j of the alternative plan is equal to zero, it represents nominal filing, that is, the telemedicine patient does not participate in the remote appointment activity.

[0056] In some embodiments, based on the probability of a telemedicine patient choosing a certain alternative plan after considering the telemedicine patient's behavior of breaking an appointment, multiple objective functions of the telemedicine appointment scheduling model are constructed from four aspects: the income of expert doctors treating telemedicine patients, the waiting cost of telemedicine patients in the appointment system, the overtime cost of expert doctors, and the sunk cost of expert doctors.

[0057] Specifically, the income of expert doctors treating telemedicine patients can be expressed as:

[0058]

[0059] In formula (6), r represents the unit revenue brought by a single telemedicine patient; θ j represents the no-show rate of telemedicine patients who meet the jth alternative; x n represents the number of telemedicine patients assigned to the nth consultation period; It represents the total number of telemedicine patients who actually completed the teleconsultation service after taking into account the telemedicine patients' no-show behavior.

[0060] Specifically, due to the overbooking strategy, when there are multiple telemedicine patients in a consultation period, only one telemedicine patient can be treated, and the other patients need to wait, which will generate waiting costs for telemedicine patients. More specifically, the probability of m telemedicine patients waiting to be treated at the end of the nth consultation period can be expressed as: B m (x n ), (7)

[0061] Formula (7) represents the probability that m telemedicine patients are left waiting for consultation at the end of the nth consultation period, where n represents the sequence number of the time slot, n=1, 2, ..., N; and m represents the number of telemedicine patients left waiting for consultation at the end of a consultation period.

[0062] In some embodiments, when n is zero, that is, when a telemedicine patient who has missed an appointment arrives at the system on time for consultation, a remote consultation will be conducted immediately. Therefore, there are no assigned telemedicine patients at this moment, and there are no remaining telemedicine patients waiting for consultation. x0=0, and at this time, B0(x0)=1.

[0063] In some embodiments, when n=1, 2, ..., N, the value of formula (7) is: when there are no remaining telemedicine patients at the end of the nth time period, the following situations are included: first, there are no remaining telemedicine patients at the end of the n-1th time period, and there are no telemedicine patients waiting for consultation or undergoing consultation at the nth time period; second, there are no remaining telemedicine patients at the end of the n-1th time period, and there is one telemedicine patient visiting the nth time period; and third, at the end of the nth time period, there is one remaining telemedicine patient waiting for consultation, and there are no telemedicine patients waiting for consultation or undergoing consultation at the nth time period. In the first and second cases, it is possible to use In the third case, it is available Therefore, when the nth period ends, the probability that there are no remaining telemedicine patients waiting for consultation can be expressed as:

[0064]

[0065]

[0066] Based on the above, the probability of m remaining telemedicine patients waiting for consultation at the end of the nth consultation period can be expressed as:

[0067]

[0068] When the nth period ends, the probability that there are m remaining telemedicine patients waiting for consultation can be expressed as

[0069]

[0070] Combining formula (8) and formula (9), we can get:

[0071]

[0072] Among them, l(x n ) represents the maximum number of telemedicine patients waiting for consultation in the system at the end of the nth consultation period. Where 1≤n≤N.

[0073] Based on the above m (x n ), assuming that at the beginning of the consultation period, there are k remaining telemedicine patients waiting for consultation, and x telemedicine patients are assigned to the medical system at this time. Under this condition, the expected waiting time of telemedicine patients can be expressed as:

[0074]

[0075] Therefore, the expected total waiting time for telemedicine patients can be expressed as:

[0076]

[0077] W(x) in formula (12) n , m) is expressed by formula (11). In summary, the waiting cost of telemedicine patients can be expressed as c w E[W(x)], where c w is the waiting cost per unit time of a telemedicine patient; x represents the allocation scheme for allocating telemedicine patients to each consultation period, x = (x1, x2, ..., x N ).

[0078] Specifically, the sunk costs in the appointment system mainly include the cost of the expert doctors being idle during working hours due to the failure of telemedicine patients to show up or the fact that certain time periods in the appointment system are not fully booked during the remote consultation process. In other words, since the expert doctors themselves are responsible for both traditional outpatient clinics and remote consultations in the same work cycle, if the expert doctors are idle during the remote consultation, they will also be unable to respond to traditional outpatient clinics, which will result in a waste of expert resources and professional remote equipment, resulting in sunk costs.

[0079] More specifically, the sunk cost of wasting resources in higher-level hospitals and equipment due to the lack of telemedicine patients and the lack of telemedicine patients caused by expert doctors can be expressed as:

[0080]

[0081] Among them, c I Represents the unit idle cost of expert doctors in remote consultation.

[0082] More specifically, the sunk costs caused by the fact that the expert doctors are idle during the remote consultation and are unable to respond to the scheduled patients in the traditional outpatient clinic during this period can be expressed as:

[0083]

[0084] Among them, π represents the unit penalty cost of the expert doctor for not responding to the traditional outpatient appointment for the telemedicine patient due to participating in the teleconsultation; Represents the total number of telehealth patients who expect to no-show overall.

[0085] Based on the above, the sunk cost of expert doctors can be expressed as:

[0086]

[0087] The above formula can be simplified to get:

[0088]

[0089] Specifically, there are two different scenarios for expert doctors to work overtime. Scenario 1: When the number of time slots actually allocated to telemedicine patients is equal to the total number of time slots in the system (i.e., N = t), the overtime can be expressed as:

[0090]

[0091] Among them, t represents the total number of equal-length consultation sessions in a remote consultation cycle, and only one telemedicine patient is served in one consultation session.

[0092] Scenario 2 is when the number of time slots actually allocated to telemedicine patients is greater than the total number of time slots in the system (i.e., N>t). There are three possible situations for overtime: the first is that in the Nth period, all scheduled telemedicine patients fail to show up and there are no remaining patients who visited the doctor in the previous time period. Since there are still scheduled telemedicine patients in the Nth period before the consultation begins, the expert doctor will work overtime in the Nt-1 period; the second is that from the t+1th period to the Nth period, there is only one telemedicine patient in the Nth period for remote consultation, and all scheduled telemedicine patients in other periods fail to show up or when the Nth period starts, there are no remaining patients who visited the doctor in the previous time period, then the expert doctor will work overtime in the Nt period; and the last situation is that when the Nth period ends, there are still remaining telemedicine patients waiting for consultation. At this time, the overtime time includes the Nt period and the consultation time corresponding to the remaining telemedicine patients at the end of the Nth period. At this time, the overtime time can be expressed as:

[0093]

[0094] Based on the above, the overtime hours of expert doctors can be expressed as:

[0095]

[0096] In summary, the overtime cost of expert doctors can be expressed as c o O(x), where c o The overtime cost per unit time of an expert doctor.

[0097] In some embodiments, based on the above four objective functions, a telemedicine appointment scheduling model is established with the goal of maximizing the total benefits of expert doctor-level hospitals and taking into account the preferences of telemedicine patients, referred to as TPEP. The telemedicine appointment scheduling model is a nonlinear integer programming model, which can be expressed as:

[0098]

[0099] The restriction condition of formula (14) is:

[0100] And x n is an integer;

[0101]

[0102]

[0103] Finally, the Gurobi optimizer built into the telemedicine appointment scheduling model is used to solve the decision variable matrix and calculate the optimal solution of the objective function to obtain a scheduling arrangement that maximizes the total benefits of the expert doctor-level hospital and takes into account the preferences and no-show behavior of telemedicine patients. It should be noted that the optimizer can handle large-scale mathematical programming problems, and its latest version, Gurobi9.1.1, can support nonlinear integer programming solutions.

[0104] Figure 2 FIG. 1 is a schematic diagram of a telemedicine consultation service matching process according to an exemplary embodiment of the present application, which takes into account the telemedicine patient's preference for appointment time and the preference for expert doctor selection. Figure 2 As shown in the figure, it is assumed that the telemedicine patients currently waiting for consultation are patient 1, patient 2, patient 3, patient 4, patient 5 and patient 6. The preferred doctors of patients 1 to 5 are chief physicians, and the preferred doctor of patient 6 is a general physician. The preferred time period of patient 1 is time period 1, the preferred time period of patient 2 is time period 2, the preferred time period of patient 3 is time period 3, and so on, where the preferred time period of patient 6 is the overtime period. Based on the preferences of the above six patients for appointment time periods and the selection preferences for expert doctors, the telemedicine appointment scheduling model finally gives the scheduling arrangement that maximizes the total benefits of the expert doctor-level hospital and takes into account the preferences and no-show behavior of telemedicine patients: Patients 1 and 4 are arranged to see the chief physician during appointment time period 1, patient 2 is arranged to see the chief physician during appointment time period 3, patients 3 and patient 5 are arranged to see the general physician during appointment time period 2, and patient 6 is arranged to see the chief physician during the overtime period.

[0105] According to an appointment scheduling method for telemedicine patients according to one embodiment of the present application, the relationship between the preferences of telemedicine patients, utility theory and the no-show rate of telemedicine patients is comprehensively considered, and the telemedicine appointment scheduling model gives a scheduling arrangement that maximizes the total benefit, thereby reducing the no-show rate of telemedicine patients and the loss of medical equipment resources and human resources, thereby maximizing the total benefit of the superior hospital of the doctor expert.

[0106] More specifically, (1) the present invention comprehensively considers the patient's preference for appointment time, the preference for expert doctors, and the various uncertain factors of patient no-shows, establishes a telemedicine appointment scheduling model, uses the Gurobi optimizer to solve the model and analyze the influencing factors, and obtains the telemedicine patient arrival situation with the greatest total benefit. (2) The present invention proposes that the patient's preference for appointment time and the preference for expert doctors jointly affect the patient no-show rate, and comprehensively considers the relationship between the two patient preferences and the change in the no-show rate, which has an important impact on the maximum total benefit optimization result of appointment scheduling. (3) The present invention proposes that there is a critical point in the average arrival rate of telemedicine patients. When the patient arrival rate is lower than the critical point, the main factor affecting the total cost of the system is the sunk cost of the expert doctors; when the patient arrival rate is higher than the critical point, the main factor affecting the total cost of the system is the overtime cost of the expert doctors. The discovery of the critical point can help superior hospitals supervise the sunk cost losses caused by the waste of telemedicine equipment resources and human resources, and assist in the daily operation management of hospitals. (4) Liu Huajun (2006) proposed that utility theory is the basis of consumer choice behavior, and consumer choice behavior determines consumer demand. He transformed the complex consumer choice behavior into an explainable quantitative analysis model. The development relationship between the three is as follows: utility theory guides consumer choice behavior, and consumer choice behavior determines consumer demand. Qian Dongfu (2008) integrated the preference relationship into the original consumer choice and utility relationship in his study of farmers' medical choice behavior in Gansu Province, and proposed that the establishment of the utility function is based on consumer preference theory. He believes that utility theory is a way to describe preferences, and proposed a progressive development relationship as follows: preference relationship → utility function → demand function. Based on the above technology, the present invention proposes a relationship between patient preference, utility theory and telemedicine patient no-show rate, which is expressed as follows: preference relationship → utility function → no-show probability. Therefore, using scientific theoretical methods to study the problem of optimal allocation of medical resources in the field of telemedicine has certain practical significance and research value. (5) In view of the characteristics of telemedicine, the present invention takes total benefit as the objective function, where total benefit is expressed by total medical income minus total cost of superior hospital. Part of the cost included in the total cost is different from the cost division of outpatient appointment scheduling in the past, and sunk cost is taken into consideration. The study found that by considering the impact of sunk cost on total cost, it helps superior hospitals analyze and find that in the scenario where patient preferences affect patient no-shows, different no-show rates have different impacts on total cost. When the no-show rate is below a certain threshold, sunk cost has a greater impact on total cost, which also means that the waste of remote consultation resources may be more serious at this time; when the no-show rate is above the threshold, overtime cost has a greater impact on total cost, which means that the workload of expert doctors is heavier. This discovery not only expands the existing research work on cost of outpatient appointment scheduling, but also can better optimize the scheduling of telemedicine patient appointments. (6) Sunk cost is creatively added to the optimization target.Sunk costs refer to the sunk costs of expert doctors and the sunk costs of remote equipment. It is mainly reflected in the fact that expert doctors cannot respond to traditional outpatients when they are idle in remote consultations, resulting in idle waste of expert resources and professional remote equipment. This has not been considered in previous technical methods. The above is only a preferred numerical example of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0107] Figure 3 is a schematic diagram of an appointment scheduling system for telemedicine patients according to an exemplary embodiment of the present application.

[0108] like Figure 3 As shown, such an appointment scheduling system for telemedicine patients is also proposed, which may include: a selection probability acquisition module 1 and a scheduling module 2. The selection probability acquisition module 1 is used to establish a patient selection model, and the patient selection model combines the preferences and no-show behavior of telemedicine patients to output the selection probability of telemedicine patients choosing alternative plans, wherein the preferences include: appointment time preference, and preference for expert doctors; the scheduling module 2 is used to build a telemedicine appointment scheduling model based on the selection probability, and the telemedicine appointment scheduling model outputs a scheduling arrangement that maximizes the total benefit.

[0109] In some embodiments, the execution steps of the selection probability acquisition module 1 may include: defining a telemedicine patient as a decision maker, and determining a plurality of alternatives corresponding to the telemedicine patient. A patient selection model is constructed based on the decision maker and the plurality of alternatives. The patient selection model combines the preferences and no-show behavior of the telemedicine patient to output the selection probability of the telemedicine patient selecting the alternative.

[0110] In some implementations, the probability of a telemedicine patient selecting an alternative may be expressed as:

[0111]

[0112] Where i represents the serial number of the telemedicine patient, i=1,2,...,M; j represents the category serial number of the alternative plan, j=1,2,...,q,...,J; V ij Denotes the utility term determined by the i-th telemedicine patient.

[0113] In some embodiments, the execution steps of the scheduling module 2 may include: constructing multiple objective functions of the telemedicine appointment scheduling model from four aspects, namely, the income of expert doctors treating telemedicine patients, the waiting cost of telemedicine patients in the appointment system, the overtime cost of expert doctors, and the sunk cost of expert doctors, based on the selection probability; integrating multiple objective functions to establish a telemedicine appointment scheduling model with the goal of maximizing total benefits; and outputting a scheduling arrangement that maximizes the total benefits from the telemedicine appointment scheduling model.

[0114] In some implementations, the telemedicine appointment scheduling model may be expressed as:

[0115]

[0116] Among them, θ j represents the no-show rate of telemedicine patients who meet the jth alternative; n represents the sequence number of the time slot, n = 1, 2, ..., N; x n represents the number of telemedicine patients assigned to the nth consultation period; c w E[W(x)] represents the waiting cost of telemedicine patients; c I represents the unit idle cost of expert doctors in remote consultation; B0(x n-1 ) represents the probability that there are 0 telemedicine patients waiting for consultation at the end of the n-1th consultation period; π represents the unit penalty cost of telemedicine patients who do not respond to traditional outpatient appointments due to expert doctors participating in remote consultations; P ij represents the probability that the i-th telemedicine patient chooses the j-th alternative; k represents the number of remaining telemedicine patients; c o O(x) represents the overtime cost of expert doctors.

[0117] According to the technical solution of the above-mentioned implementation mode, at least one of the following beneficial effects can be obtained.

[0118] According to an appointment scheduling system for telemedicine patients in accordance with one embodiment of the present application, the relationship between the preferences of telemedicine patients, utility theory and the no-show rate of telemedicine patients is comprehensively considered, and the telemedicine appointment scheduling model provides a scheduling arrangement that maximizes the total benefit, thereby reducing the no-show rate of telemedicine patients and the loss of medical equipment resources and human resources, thereby maximizing the total benefit of the superior hospitals of doctors and experts.

Claims

1. A method for scheduling appointments for telemedicine patients, characterized in that: include: A patient selection model is established, and the patient selection model combines the preferences and no-show behavior of the telemedicine patient to output the selection probability of the telemedicine patient selecting the alternative plan, wherein the preferences include: appointment time preference, and selection preference for expert doctors, including: defining a telemedicine patient as a decision maker and identifying multiple alternatives corresponding to the telemedicine patient; constructing a patient selection model based on the decision maker and a plurality of the alternatives; and The patient selection model combines the preferences and no-show behavior of the telemedicine patient to output the probability of the telemedicine patient selecting the alternative plan; as well as Based on the selection probability, a telemedicine appointment scheduling model is constructed, and the telemedicine appointment scheduling model outputs a scheduling arrangement that maximizes the total benefit, including: According to the selection probability, multiple objective functions of the telemedicine appointment scheduling model are constructed from four aspects: the income of the expert doctor treating the telemedicine patient, the waiting cost of the telemedicine patient, the overtime cost of the expert doctor, and the sunk cost of the expert doctor; Integrate multiple objective functions to establish a telemedicine appointment scheduling model with the goal of maximizing total benefits; and The telemedicine appointment scheduling model outputs a scheduling arrangement that maximizes the total benefit; The revenue from specialist doctors treating telemedicine patients is expressed as: ； represents the unit revenue brought by a single telemedicine patient; represents the no-show rate of telemedicine patients who meet the jth alternative; represents the probability that the i-th telemedicine patient chooses the j-th alternative; represents the number of telemedicine patients assigned to the nth consultation period; It represents the total number of telemedicine patients who actually completed teleconsultation services after considering the no-show behavior of telemedicine patients; The waiting cost of telemedicine patients is expressed as ,in is the waiting cost per unit time of a telemedicine patient; x represents the allocation scheme for telemedicine patients in each consultation period, ; The overtime cost of expert doctors is expressed as ,in The overtime cost per unit time of a specialist doctor; The sunk cost of the specialist doctor is: ， Among them, the sunk cost of the waste of superior hospitals and equipment resources caused by the lack of telemedicine patients and the lack of telemedicine patients caused by expert doctors is expressed as: ， represents the unit idle cost of expert doctors in remote consultation; represents the probability that there are 0 telemedicine patients waiting for consultation at the end of the n-1th consultation period; Since the expert doctors are idle during the remote consultation and are unable to respond to the appointments of patients in the traditional outpatient clinic during this period, the sunk costs are expressed as: ， represents the unit penalty cost for telemedicine patients who are not responded to traditional clinic appointments due to the specialist physicians’ participation in teleconsultation; Indicates the total number of telehealth patients who expect to no-show overall.

2. The appointment scheduling method for remote medical patients according to claim 1, characterized in that: The probability of the telemedicine patient selecting the alternative is expressed as: ， in, Indicates the serial number of the telemedicine patient, = ; Indicates the category number of the alternative plan. = ; Indicates Telemedicine patients determine the utility item.

3. The appointment scheduling method for remote medical patients according to claim 1, characterized in that: The telemedicine appointment scheduling model is expressed as: ， in, Indicates compliance with The no-show rate of telemedicine patients for the alternative schemes; n represents the sequence number of the time slot, ; represents the number of telemedicine patients assigned to the nth visit time slot; represents the waiting cost of the telemedicine patient; represents the number of remaining telemedicine patients; Represents the overtime cost of specialist doctors.

4. An appointment scheduling system for telemedicine patients, characterized in that: include: The selection probability acquisition module is used to establish a patient selection model, and output the selection probability of the telemedicine patient selecting an alternative plan by combining the patient selection model with the telemedicine patient's preferences and no-show behavior, wherein the preferences include: appointment time preference, and preference for expert doctor selection, including: defining a telemedicine patient as a decision maker and identifying multiple alternatives corresponding to the telemedicine patient; constructing a patient selection model based on the decision maker and a plurality of the alternatives; and Outputting the selection probability of the telemedicine patient selecting the alternative plan by the patient selection model in combination with the telemedicine patient's preference and no-show behavior; and A scheduling module is used to construct a telemedicine appointment scheduling model based on the selection probability, and the telemedicine appointment scheduling model outputs a scheduling arrangement that maximizes the total benefit, including: According to the selection probability, multiple objective functions of the telemedicine appointment scheduling model are constructed from four aspects: the income of the expert doctor treating the telemedicine patient, the waiting cost of the telemedicine patient, the overtime cost of the expert doctor, and the sunk cost of the expert doctor; Integrate multiple objective functions to establish a telemedicine appointment scheduling model with the goal of maximizing total benefits; and The telemedicine appointment scheduling model outputs a scheduling arrangement that maximizes the total benefit; The revenue from specialist doctors treating telemedicine patients is expressed as: ； represents the unit revenue brought by a single telemedicine patient; represents the no-show rate of telemedicine patients who meet the jth alternative; represents the probability that the i-th telemedicine patient chooses the j-th alternative; represents the number of telemedicine patients assigned to the nth consultation period; It represents the total number of telemedicine patients who actually completed teleconsultation services after considering the no-show behavior of telemedicine patients; The waiting cost of telemedicine patients is expressed as ,in is the waiting cost per unit time of a telemedicine patient; x represents the allocation scheme for telemedicine patients in each consultation period, ; The overtime cost of expert doctors is expressed as ,in The overtime cost per unit time of a specialist doctor; The sunk cost of the specialist doctor is: ， Among them, the sunk cost of the waste of superior hospitals and equipment resources caused by the lack of telemedicine patients and the lack of telemedicine patients caused by expert doctors is expressed as: ， represents the unit idle cost of expert doctors in remote consultation; represents the probability that there are 0 telemedicine patients waiting for consultation at the end of the n-1th consultation period; Since the expert doctors are idle during the remote consultation and are unable to respond to the appointments of patients in the traditional outpatient clinic during this period, the sunk costs are expressed as: ， represents the unit penalty cost for telemedicine patients who are not responded to traditional clinic appointments by specialist doctors due to participation in teleconsultation; Indicates the total number of telehealth patients who expect to no-show overall.

5. The appointment scheduling system for remote medical patients according to claim 4, characterized in that: The probability of the telemedicine patient selecting the alternative is expressed as: ， Among them, i represents the serial number of the telemedicine patient, ; j represents the category number of the alternative plan, ; Denotes the utility term determined by the i-th telemedicine patient.

6. The appointment scheduling system for remote medical patients according to claim 4, characterized in that: The telemedicine appointment scheduling model is expressed as: ， Where n represents the sequence number of the time slot. ; represents the number of telemedicine patients assigned to the nth visit time slot; represents the unit penalty cost for telemedicine patients who are not responded to traditional clinic appointments by specialist doctors due to participation in teleconsultation; represents the probability that the i-th telemedicine patient chooses the j-th alternative; represents the number of remaining telemedicine patients; Represents the overtime cost of specialist doctors.

Images