Intelligent accompanying equipment management scheduling system and method for intelligent medical treatment

Abstract

The invention discloses an intelligent accompanying equipment management scheduling system and method for intelligent medical treatment, relates to the technical field of medical accompanying equipment, and solves the technical problem that the intelligent accompanying equipment is unreasonably distributed due to different demands of the intelligent accompanying equipment in different departments in a hospital. According to the invention, the ARIMA model is trained based on the hospitalization basic information of each department to obtain the hospitalization prediction model, and the hospitalization prediction model is utilized to predict the number of hospitalization people of each department, so that the demand of accompanying equipment of each department is determined; scheduling according to a difference value between the demanded quantity of the accompanying equipment of each department and the number of the accompanying equipment so as to generate a scheduling scheme; the scheduling cost and the scheduling complexity are considered in the scheduling process, so that the scheduling cost and the scheduling complexity of the intelligent accompanying equipment between the two departments are lowest as far as possible, the intelligent accompanying equipment is efficiently scheduled, and the problem that the intelligent accompanying equipment is unreasonably distributed in each department is solved.

Description

Technical Field

[0001] This application belongs to the field of medical escort equipment, and relates to the management and scheduling technology of intelligent escort equipment. Specifically, it is an intelligent escort equipment management and scheduling system and method for smart healthcare. Background Art

[0002] Intelligent escort equipment mainly includes intelligent escort beds, intelligent escort chairs, etc., which are mainly used to solve the rest problem of escort personnel during the escort process. At present, intelligent escort equipment has facilitated the use of escort personnel and the management of equipment through technologies such as the Internet of Things, mobile payment, and intelligent locks, not only improving the escort experience but also optimizing the hospital's operation efficiency.

[0003] When not in use, intelligent escort equipment is mainly stored in the ward lockers and can be used by scanning the code with a smartphone when needed. However, due to the differences in the demand for intelligent escort equipment in different departments or floors of the hospital, the distribution of intelligent escort equipment may be unreasonable during a certain period. Moreover, since there are not many restrictions on the access of current intelligent protection equipment, some intelligent escort equipment is unreasonably occupied, thus affecting the use of other escort personnel.

[0004] This application provides an intelligent escort equipment management and scheduling system and method for smart healthcare to solve the above technical problems. Summary of the Invention

[0005] This application aims to solve at least one of the technical problems existing in the prior art; for this purpose, this application proposes an intelligent escort equipment management and scheduling system and method for smart healthcare, which is used to solve the technical problem that the distribution of intelligent escort equipment is unreasonable due to the different demands for intelligent escort equipment in different departments of the hospital.

[0006] To achieve the above object, the first aspect of this application provides an intelligent escort equipment management and scheduling system for smart healthcare, including: an intelligent scheduling module, and a data acquisition module connected thereto;

[0007] Data acquisition module: used to obtain the basic hospitalization information and the number of escort equipment in each department; preprocess the basic hospitalization information to obtain target data; where the basic hospitalization information includes the number of inpatients, date, and season, and the preprocessing includes data cleaning and data conversion;

[0008] Intelligent scheduling module: used to train an ARIMA model based on the target data to obtain a hospitalization prediction model; predict the number of inpatients on the current date based on the hospitalization prediction model; and,

[0009] Used to determine the demand for escort equipment based on the number of inpatients, calculate the difference one between the demand for escort equipment and the number of escort equipment; generate a scheduling plan based on the difference one of each department.

[0010] Preferably, training an ARIMA model based on the basic inpatient information includes:

[0011] Extracting feature data from the basic inpatient information; among them, the feature data includes time features, seasonal features, and the number of inpatients in each department;

[0012] Performing stationarity verification and processing on the feature data, initially determining the model parameters using ACF and PACF, and optimizing the model parameters through the network search method and information criteria;

[0013] Combining the optimized model parameters to fit the ARIMA model, and obtaining the inpatient prediction model after residual diagnosis and optimization.

[0014] Preferably, generating a scheduling plan according to the difference one of each department includes:

[0015] Integrating the departments with the difference one greater than 0 into data sequence one, and integrating the departments with the difference one less than 0 into data sequence two; among them, the integration process is sorted in descending order according to the absolute value of the difference one;

[0016] Sequentially select the difference one in data sequence one as the target data, and match the surplus of intelligent escort equipment in each department in data sequence two to supplement the missing amount corresponding to the target data.

[0017] Preferably, matching the surplus of intelligent escort equipment in each department in data sequence two to supplement the missing amount corresponding to the target data includes:

[0018] Taking the department corresponding to the target data as the target department, and marking the departments in data sequence two as alternative departments; calculating the difference between the target data and the difference one of the alternative departments, and marking it as difference two;

[0019] Calculating the scheduling cost between the target department and the alternative departments; calculating the scheduling evaluation coefficient based on the scheduling cost and difference two; determining the scheduling plan based on the scheduling evaluation coefficient.

[0020] Preferably, calculating the scheduling evaluation coefficient based on the scheduling cost and difference two includes:

[0021] Marking the difference two as CZ and the scheduling cost as DC; among them, the scheduling cost is proportional to the scheduling distance and the scheduling quantity of intelligent escort equipment;

[0022] Calculating the scheduling evaluation coefficient DPX through the formula DPX = α / DC + β / exp(CZ); where α and β are weight coefficients, and exp(*) is the exponential function with the natural number e as the base.

[0023] Preferably, determining a scheduling plan based on a scheduling evaluation coefficient includes:

[0024] Select the alternative department corresponding to the maximum value of the scheduling evaluation coefficient as the scheduling department, and provide intelligent escort devices for the target department;

[0025] Determine whether there is a remaining quantity in the scheduling department after the scheduling is completed; if so, re - sort the scheduling department in the second data sequence according to the remaining quantity; if not, delete the scheduling department from the second data sequence;

[0026] Determine whether the number of escort devices in the target department reaches the demand for escort devices after the scheduling is completed; if so, delete the target department from the first data sequence; if not, update the difference value one of the target department;

[0027] After completing the matching for all target departments, generate a scheduling plan.

[0028] Preferably, updating the difference value one of the target department includes:

[0029] Recalculate the difference value one of the target department;

[0030] Update the difference value one of the target department in the first data sequence according to the recalculated difference value one; or, re - sort the first data sequence according to the recalculated difference value one.

[0031] Preferably, when each department handles the hospitalization registration for each in - patient, the information of the escort personnel is registered simultaneously;

[0032] Associate the information of the escort personnel with the bed of the in - patient they accompany, and each bed is associated with an intelligent escort device; among them, the intelligent escort device associated with each bed is determined according to the scheduling plan.

[0033] Preferably, arrange staff to complete the scheduling of intelligent escort devices based on the scheduling plan during a set time period; where the staff includes medical staff or cleaning staff.

[0034] The second aspect of the present application provides an intelligent escort device management and scheduling method for intelligent medical care, including:

[0035] Obtain the basic hospitalization information and the number of escort devices of each department; pre - process the basic hospitalization information to obtain target data; where the basic hospitalization information includes the number of in - patients, date, and season, and the pre - processing includes data cleaning and data conversion;

[0036] Train an ARIMA model based on the target data to obtain a hospitalization prediction model; predict the number of inpatients on the current date based on the hospitalization prediction model; determine the demand for escort equipment according to the number of inpatients, and calculate the difference one between the demand for escort equipment and the number of escort equipment; generate a scheduling plan according to the difference one of each department.

[0037] Compared with the prior art, the beneficial effects of the present application are as follows:

[0038] 1. The present application trains an ARIMA model based on the hospitalization basic information of each department to obtain a hospitalization prediction model, and uses the hospitalization prediction model to predict the number of inpatients in each department, and then determines the demand for escort equipment in each department; performs scheduling according to the difference between the demand for escort equipment in each department and the number of escort equipment to generate a scheduling plan; the present application considers the scheduling cost and scheduling complexity during the scheduling process, and tries to make the scheduling cost and scheduling complexity of the intelligent escort equipment between two departments the lowest, and efficiently schedules the intelligent escort equipment to solve the problem that the distribution of intelligent escort equipment in each department is unreasonable.

[0039] 2. After calculating the scheduling evaluation coefficient between the target department and several alternative departments, the present application selects the alternative department corresponding to the maximum value of the scheduling evaluation coefficient as the scheduling department to provide intelligent escort equipment for the target department; then judges whether there is still a shortage in the target department after the scheduling is completed, and updates the data sequence one and the data sequence two according to the judgment result; the present application judges the situation of each department through simulated scheduling to update the data sequence, simplifies the complex and gradually completes the scheduling matching for all departments, effectively improving the generation efficiency of the scheduling plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0041] Figure 1 It is a flowchart of the intelligent escort equipment management and scheduling method according to Embodiment 1 of the present application;

[0042] Figure 2 It is a system schematic diagram of the intelligent escort equipment management and scheduling system according to Embodiment 1 of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] The technical solutions of the present application will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0044] At present, the management of intelligent escort equipment in hospitals is relatively immature. There are often situations where the intelligent escort equipment in some departments is significantly more than the inpatients, while in other departments, the intelligent escort equipment is significantly insufficient. One of the main reasons is the inability to dynamically predict the number of inpatients, lacking a data basis for management and scheduling. Of course, the difficulty of managing and scheduling intelligent escort equipment is also an important reason for the unreasonable distribution of intelligent escort equipment in each department. The present application provides an intelligent escort equipment management and scheduling system and method for intelligent healthcare to solve the problem of unreasonable distribution of intelligent escort equipment in the field of intelligent healthcare.

[0045] Embodiment 1:

[0046] Please refer to Figure 1 - Figure 2 , the first aspect embodiment of the present application provides an intelligent escort equipment management and scheduling system for intelligent healthcare, including: an intelligent scheduling module and a data acquisition module connected thereto;

[0047] Data acquisition module: used to obtain the inpatient basic information and the number of escort equipment in each department; preprocess the inpatient basic information to obtain target data;

[0048] Intelligent scheduling module: used to train an ARIMA model based on the target data to obtain an inpatient prediction model; predict the number of inpatients on the current date based on the inpatient prediction model; and, used to determine the demand for escort equipment according to the number of inpatients, calculate the difference one between the demand for escort equipment and the number of escort equipment; generate a scheduling plan according to the difference one of each department.

[0049] Accurately predicting the number of inpatients (inpatients) in each department is an important basic data for reasonably scheduling intelligent escort equipment. In this embodiment, an ARIMA model is trained based on the obtained inpatient basic information of each department, and the number of inpatients in the future every day is predicted according to the obtained inpatient prediction model. Then, the escort personnel are determined according to the number of inpatients, and finally, the demand for intelligent escort equipment in each department can be determined, and the intelligent escort equipment in each department can be coordinated and scheduled according to this demand.

[0050] The data acquisition module in this example is used to collect the basic inpatient information of each department and the number of accompanying devices in real time. The basic inpatient information can be extracted from the hospital information system (HIS), mainly extracting the relevant information of inpatients, specifically including time (seasons can be identified according to time), the number of inpatients, etc. To improve the accuracy of data prediction, data related to the number of inpatients, such as epidemic trends, can also be extracted.

[0051] Data interaction occurs between the intelligent scheduling module and the data acquisition module, which can be connected wirelessly. The intelligent scheduling module is mainly responsible for data processing and scheduling execution, specifically including predicting the number of inpatients, generating a coordination plan based on the number of inpatients in each department and the existing number of intelligent accompanying devices, and finally completing the scheduling of intelligent accompanying devices. It should be noted that intelligent accompanying devices mainly include intelligent accompanying beds, intelligent accompanying chairs, intelligent accompanying cabinets, etc., which can all be unlocked and used through Internet of Things technology and are generally used by the accompanying personnel of inpatients.

[0052] The basis for implementing the scheduling of intelligent accompanying devices in this embodiment is the prediction of the number of inpatients in each department, which is mainly achieved based on the ARIMA model, and specifically includes the following steps:

[0053] Extract feature data from the basic inpatient information; among them, the feature data includes time features, seasonal features, and the number of inpatients in each department; perform stationarity verification and processing on the feature data, initially determine the model parameters using ACF and PACF, and optimize the model parameters through the network search method and information criteria; fit the ARIMA model in combination with the optimized model parameters, and obtain the inpatient prediction model after residual diagnosis and optimization.

[0054] Exemplarily, the following example briefly illustrates the construction of the ARIMA model and the prediction of the number of inpatients.

[0055] 1. Data collection: Collect the number of inpatients in each department through the hospital information system, including the number of inpatients daily, weekly, or monthly; and focus on collecting date information and seasonal information, with a time period of at least 3 years as much as possible to discover the relationship between date, season, and the number of inpatients. The above information can also be extracted from electronic health records, and all the extracted data has been de-identified or authorized by the rights holders.

[0056] 2. Data preprocessing:

[0057] 1) Data cleaning: Check and process missing values, and check, correct, and delete outliers;

[0058] 2) Feature extraction: Extract the time features and seasonal features (such as seasonal dummy variables) of the basic inpatient information;

[0059] 3) Stationarity check: Use the ADF test to check the stationarity of the time series. If the time series is non-stationary, differencing is required to make the time series stationary.

[0060] 3. Model construction:

[0061] Use the autocorrelation function (ACF) and partial autocorrelation function (PACF) plots to preliminarily determine the model parameters, and use the network search method and information criteria (such as AIC or BIC) to optimize the parameters. The above model parameters are mainly p, d, and q of the ARIMA model; p is the autoregressive term, indicating the number of lag terms used in the model; d is the differencing order, indicating how many times differencing is required to achieve stationarity; q is the moving average term, indicating the number of error terms used in the model. According to the optimized model parameters, fit the ARIMA model. Finally, judge whether the residuals are white noise through the ACF and PACF plots of the residuals. If they are not white noise, the model parameters need to be readjusted. After passing the diagnosis, the hospitalization prediction model is obtained. There are many reference solutions for the construction and fitting of the ARIMA model in the existing materials, so it will not be elaborated here.

[0062] After predicting the number of hospitalizations in each department, it is necessary to determine the demand for escort equipment based on the number of hospitalizations, that is, how many intelligent escort devices are needed for the predicted number of hospitalizations. It should be noted that generally, the number of hospitalizations is equal to the demand for escort equipment, that is, one hospitalized patient is allowed one escort; for other special cases, the ratio of the number of hospitalizations to the demand for escort equipment can be adjusted appropriately. Moreover, the number of hospitalizations is generally predicted once a day, and of course, it can also be predicted once a week to reduce the scheduling cost.

[0063] When the demand for escort equipment in each department is not equal to the number of escort equipment, either there is a surplus of intelligent escort equipment or there is a shortage of intelligent escort equipment. Therefore, it is necessary to generate a scheduling plan based on the difference between the demand for escort equipment and the number of escort equipment in each department, that is, difference one. The specific steps can be referred to as follows:

[0064] Integrate the departments with difference one greater than 0 into data sequence one, and integrate the departments with difference one less than 0 into data sequence two; sequentially select the difference one in data sequence one as the target data, and match the surplus of intelligent escort equipment in each department in data sequence two to supplement the missing amount corresponding to the target data; the scheduling plan can be generated according to the supplementation process of data sequence two to data sequence one.

[0065] If the first difference is greater than 0, it indicates that the demand for escort equipment in the corresponding department is greater than the quantity of escort equipment, that is, the supply is insufficient, and some intelligent escort equipment needs to be dispatched from other departments. Integrate the departments with the first difference greater than 0 into the first data sequence. This first data sequence includes several departments and their corresponding first differences. Then, each department in the first data sequence needs to dispatch intelligent escort equipment from other departments. Similarly, each department in the second data sequence needs to provide redundant intelligent escort equipment for other departments.

[0066] It should be noted that during the process of integrating and generating the first data sequence and the second data sequence, they are both sorted in descending order of the absolute value of the first difference. That is, the department with the largest shortage of intelligent escort equipment ranks first in the first data sequence, and the department with the largest remaining quantity of intelligent escort equipment ranks first in the second data sequence.

[0067] To achieve the automated evaluation and scheduling of intelligent escort equipment, the following solutions are provided for how to supplement the missing quantity in the first data sequence with the redundant intelligent escort equipment in the second data sequence:

[0068] Take the department corresponding to the target data as the target department, and mark the departments in the second data sequence as alternative departments; calculate the difference between the target data and the first difference of the alternative departments, and mark it as the second difference;

[0069] Calculate the scheduling cost between the target department and the alternative departments; calculate the scheduling evaluation coefficient based on the scheduling cost and the second difference; determine the scheduling plan based on the scheduling evaluation coefficient.

[0070] Select the departments in the first data sequence as the target departments in turn. In fact, this target department is the department with the most serious shortage of intelligent escort equipment. Then calculate the difference between the first difference of the target department and the first differences of each alternative department in the second data sequence as the second difference.

[0071] Exemplarily, assume that the first difference of the target department is 10 and the first difference of an alternative department is 11. Then the second difference (taking the absolute value) is 1. It can be understood that the quantity of intelligent escort equipment to be dispatched from the alternative department to the target department is 10, not 1. The above scheduling cost is related to the scheduling quantity and, of course, also related to the scheduling distance between the target department and the alternative departments.

[0072] Exemplarily, a solution for calculating the scheduling cost is given, including: obtaining the average cost of dispatching intelligent escort equipment between departments according to historical data, and multiplying the average cost by the scheduling quantity to obtain the scheduling cost. The average cost is averaged based on historical experience and, of course, can also be set manually. It should be noted that the average cost can be a monetary cost, a time cost, or a combination of multiple costs.

[0073] The purpose of the scheduling evaluation coefficient is to select an alternative department from Data Sequence 2 to ensure the lowest scheduling cost and scheduling complexity between the alternative department and the target department. The lowest scheduling complexity here is manifested as the size of Difference 2. The smaller Difference 2 is, the lower the scheduling complexity. When Difference 2 is the smallest, either the shortage of the target department is replenished and there is not much remaining intelligent escort equipment in the alternative department; or the surplus of the alternative department is all scheduled to the target department, but the demand gap of the target department is small.

[0074] After calculating the scheduling evaluation coefficients of the target department and each alternative department, the scheduling evaluation coefficients can be used to determine which or which alternative departments will schedule the redundant intelligent escort equipment to the target department, mark them as the target department in turn, and calculate the scheduling evaluation coefficients of each target department and the remaining alternative departments, so as to schedule sufficient intelligent escort equipment for each target department, and thus determine the scheduling plan of the intelligent escort equipment between departments. It should be noted that the scheduling plan is the basis for the scheduling of intelligent escort equipment, but the actual scheduling process needs to be completed within a specific time period to avoid affecting the use of existing escort personnel.

[0075] Next, a calculation scheme for the scheduling evaluation coefficient is provided, including the following steps:

[0076] Mark Difference 2 as CZ and the scheduling cost as DC; calculate the scheduling evaluation coefficient DPX through the formula DPX = α / DC + β / exp(CZ).

[0077] In the above calculation scheme, α and β are weight coefficients, and exp(*) is an exponential function with the natural number e as the base. In this embodiment, the scheduling evaluation coefficient is inversely proportional to the scheduling cost and Difference 2, that is, the higher the scheduling cost and the larger Difference 2 are, the larger the scheduling evaluation coefficient is. The weight coefficients α and β are set according to the importance of the scheduling cost and scheduling complexity; if you want to complete the scheduling of the intelligent escort equipment for each department with fewer scheduling times, you can increase β; if you want to complete the scheduling of the intelligent escort equipment for each department in a short time, you can increase α. Of course, other existing schemes can also be used to calculate the scheduling evaluation coefficient, or other evaluation parameters.

[0078] Based on the above calculation scheme of the scheduling evaluation coefficient, to determine the scheduling plan based on the obtained scheduling evaluation coefficient, the following steps can be referred to:

[0079] Select the alternative department corresponding to the maximum value of the scheduling evaluation coefficient as the scheduling department to provide intelligent escort equipment for the target department;

[0080] Judge whether there is any remaining quantity in the scheduling department after the scheduling is completed; if yes, re - sort the scheduling department in Data Sequence 2 according to the remaining quantity; if no, delete the scheduling department from Data Sequence 2;

[0081] Determine whether the number of accompanying devices in the target department after the scheduling is completed reaches the demand for accompanying devices; if so, delete the target department from Data Sequence 1; if not, update the difference value 1 of the target department;

[0082] After matching is completed for all target departments, generate a scheduling plan.

[0083] The above "after determining the completion of scheduling" does not really complete the scheduling, but is a hypothesis, that is, if the scheduling department schedules intelligent accompanying devices for the target department. If there is a remaining quantity in the scheduling department, it means that the demand for accompanying devices in the target department after scheduling has been met, and there are still some extra intelligent accompanying devices left in the scheduling department. At this time, delete the target department from Data Sequence 1, re-determine the target department, and re-sort the scheduling department in Data Sequence 2. Conversely, if the demand for accompanying devices in the target department has not been supplemented enough, the target department still needs to be supplemented, then update its difference value 1; at the same time, there are no more extra intelligent accompanying devices in the corresponding scheduling department, and delete the scheduling department from Data Sequence 2. It can be understood that at least one of the target department and the scheduling department needs to be deleted from its respective data sequence after the scheduling is completed.

[0084] This embodiment provides two ways to update the difference value 1 of the target department, which are as follows:

[0085] First, after the scheduling is completed, recalculate the difference value 1 of the target department;

[0086] The sorting of the target department in the original Data Sequence 1 can be retained, and only its difference value 1 is updated; recalculate the scheduling evaluation coefficient with this difference value 1 until the intelligent accompanying devices required by the target department are supplemented enough, and then delete it from Data Sequence 1;

[0087] Or after the difference value 1 of the target department is updated, it is re-incorporated into Data Sequence 1 and sorted; determine the target department from the sorted Data Sequence 1 in turn, and then repeat the above process to supplement the intelligent accompanying devices for the target department.

[0088] After the matching and supplement of intelligent accompanying devices for all departments in Data Sequence 1 are completed, generate a scheduling plan according to the matching logic of each department. The staff can complete the scheduling of intelligent accompanying devices within the set time period according to this scheduling plan. It should be noted that it may not be possible to fully supplement all departments in Data Sequence 1. If the shortage is large, an alarm can be sent to the staff.

[0089] Embodiment 2: On the basis of Embodiment 1, associate intelligent accompanying devices with each bed to prevent the intelligent accompanying devices from being unreasonably occupied. Specifically:

[0090] When each department handles the hospitalization registration for each in-patient, the information of the accompanying personnel is registered at the same time; the information of the accompanying personnel is associated with the bed of the in-patient they accompany, and each bed is associated with an intelligent accompanying device; among them, the intelligent accompanying devices associated with each bed are determined according to the scheduling plan.

[0091] When handling the hospitalization procedures for in-patients, the accompanying personnel of the in-patient should be registered, such as name, mobile phone number, etc. The information of the accompanying personnel is associated with the bed of the in-patient they accompany, and each bed is associated with at least one intelligent accompanying device. If the accompanying personnel need to use the intelligent accompanying device, they can unlock and use it by scanning the code with a smart phone to verify the bed and the information of the accompanying personnel.

[0092] It should be noted that after the scheduling plan is generated, it is determined through the background data which intelligent accompanying devices need to be scheduled to the required departments, that is, the unique identifiers of these intelligent accompanying devices are assigned to the corresponding departments, and the corresponding departments can match intelligent accompanying devices for each bed, and the intelligent accompanying devices can be associated after the information of the accompanying personnel is registered. Since the intelligent accompanying devices in the hospital are generally used at night, the association does not mean that they have been transported to the corresponding beds, and they may still not be available at the current time.

[0093] Embodiment 3: On the basis of Embodiment 1, this embodiment provides a solution for how to execute the scheduling plan, specifically: arranging staff to complete the scheduling of intelligent accompanying devices based on the scheduling plan during the set time period; among them, the staff includes medical staff or cleaning staff.

[0094] Intelligent accompanying devices are generally used within a specific time period, such as at night, so the above set time period can be from 7:00 pm to 8:00 pm. During this time period, the background statistics the information of the intelligent accompanying devices that need to be scheduled to other departments, and the cleaning staff or the medical staff on shift transport them to the designated location; similarly, the departments that need intelligent accompanying devices will also receive the statistical information, and the cleaning staff or medical staff in this department will go to the designated location to transport the required intelligent accompanying devices according to the statistical information. It should be noted that the designated location is not fixed and depends on the relative positions of the target department and the scheduling department.

[0095] The second aspect of the embodiments of the present application provides an intelligent accompanying device management and scheduling method for intelligent medical care, including:

[0096] Obtaining the basic hospitalization information and the number of accompanying devices of each department; preprocessing the basic hospitalization information to obtain target data; among them, the basic hospitalization information includes the number of in-patients, date and season, and the preprocessing includes data cleaning and data conversion;

[0097] Train an ARIMA model based on the target data to obtain a hospitalization prediction model; predict the number of inpatients on the current date based on the hospitalization prediction model; determine the demand for escort equipment according to the number of inpatients, and calculate the difference one between the demand for escort equipment and the number of escort equipment; generate a scheduling plan according to the difference one of each department.

[0098] The above embodiments are only used to illustrate the technical solutions of the present application rather than to limit them. Although the present application has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. An intelligent accompanying equipment management and scheduling system for smart medical care, characterized in that: include: Intelligent scheduling module, and the data acquisition module connected thereto; Data collection module: used to obtain basic inpatient information and the number of accompanying equipment in each department; Preprocessing the basic hospitalization information to obtain target data; wherein the basic hospitalization information includes the number of hospitalized patients, date and season, and the preprocessing includes data cleaning and data conversion; Intelligent scheduling module: used to train the ARIMA model based on the target data to obtain a hospitalization prediction model; predict the number of hospitalized people on the current date based on the hospitalization prediction model; and, It is used to determine the demand for accompanying equipment according to the number of inpatients, calculate the difference between the demand for accompanying equipment and the number of accompanying equipment; and generate a scheduling plan according to the difference of each department.

2. According to claim 1, a smart accompanying equipment management and scheduling system for smart medical care is characterized in that: The ARIMA model is trained based on the basic hospitalization information, including: Extracting characteristic data from the basic hospitalization information; wherein the characteristic data includes time characteristics, seasonal characteristics and the number of hospitalized persons; Performing stationarity check and processing on the characteristic data, using ACF and PACF to preliminarily determine the model parameters, and optimizing the model parameters through network search method and information criterion; The ARIMA model was fitted with the optimized model parameters, and the hospitalization prediction model was obtained after residual diagnosis and optimization.

3. According to claim 1, the intelligent accompanying equipment management and scheduling system for smart medical care is characterized in that: Generate a scheduling plan based on the difference of each department, including: The departments whose difference value 1 is greater than 0 are integrated into data sequence 1, and the departments whose difference value 1 is less than 0 are integrated into data sequence 2; wherein the integration process is sorted from large to small according to the absolute value of difference value 1; The difference value 1 in the data sequence 1 is selected in turn as the target data, and the surplus amount of the intelligent accompanying equipment in each department in the data sequence 2 is matched to supplement the missing amount corresponding to the target data.

4. The intelligent accompanying equipment management and dispatching system for smart medical care according to claim 3 is characterized in that: Matching the surplus of intelligent accompanying equipment in each department in the data sequence 2 to supplement the missing quantity corresponding to the target data includes: The department corresponding to the target data is taken as the target department, and the department in the second data sequence is marked as the candidate department; the difference between the target data and the difference value 1 of the candidate department is calculated, and marked as the difference value 2; Calculate the scheduling cost between the target department and the alternative department; calculate the scheduling evaluation coefficient based on the scheduling cost and the difference; determine the scheduling plan based on the scheduling evaluation coefficient.

5. The intelligent accompanying equipment management and dispatching system for smart medical care according to claim 4 is characterized in that: Calculating a scheduling evaluation coefficient based on the scheduling cost and the second difference includes: The difference is marked as CZ, and the dispatch cost is marked as DC; wherein the dispatch cost is proportional to the dispatch distance and the dispatch quantity of the intelligent accompanying device; The scheduling evaluation coefficient DPX is calculated by the formula DPX=α / DC+β / exp(CZ); wherein α and β are weight coefficients, and exp(*) is an exponential function with the natural number e as the base.

6. The intelligent accompanying equipment management and dispatching system for smart medical care according to claim 4 is characterized in that: Determining a scheduling solution based on the scheduling evaluation coefficient includes: Select the candidate department corresponding to the maximum value of the scheduling evaluation coefficient as the scheduling department, and provide the target department with intelligent accompanying equipment; Determine whether there is a remaining amount in the dispatching department after the dispatching is completed; if yes, re-order the dispatching department in the data sequence 2 according to the remaining amount; if no, delete the dispatching department from the data sequence 2; Determine whether the number of accompanying equipment in the target department reaches the required number of accompanying equipment after the scheduling is completed; if yes, delete the target department from the data sequence one; if no, update the difference value one of the target department; After matching is completed for all target departments, a scheduling plan is generated.

7. The intelligent accompanying equipment management and dispatching system for smart medical care according to claim 6 is characterized in that: The updating of the difference value of the target department comprises: Recalculate the difference value of the target department -1; The difference value 1 of the target department in the data sequence 1 is updated according to the recalculated difference value 1; or, the data sequence 1 is reordered according to the recalculated difference value 1.

8. The intelligent accompanying equipment management and scheduling system for smart medical care according to claim 1 is characterized in that: When each department registers each inpatient for hospitalization, it also registers the information of the accompanying personnel; The accompanying personnel information is associated with the bed of the inpatient they are accompanying, and each bed is associated with an intelligent accompanying device; wherein the intelligent accompanying device associated with each bed is determined according to the scheduling plan.

9. The intelligent accompanying equipment management and scheduling system for smart medical care according to claim 1 is characterized in that: Arrange staff to complete the dispatch of the intelligent accompanying equipment based on the dispatch plan during the set time period; wherein the staff includes medical staff or cleaning staff.

10. A method for managing and scheduling intelligent accompanying equipment for smart medical treatment, based on the operation of an intelligent accompanying equipment management and scheduling system for smart medical treatment according to any one of claims 1 to 9, characterized in that: include: Obtain basic inpatient information and the number of accompanying equipment in each department; Preprocessing the basic hospitalization information to obtain target data; wherein the basic hospitalization information includes the number of hospitalized patients, date and season, and the preprocessing includes data cleaning and data conversion; Based on the target data, an ARIMA model is trained to obtain a hospitalization prediction model; the number of hospitalized patients on the current date is predicted based on the hospitalization prediction model; the demand for accompanying equipment is determined based on the number of hospitalized patients, and the difference between the demand for accompanying equipment and the number of accompanying equipment is calculated; and a scheduling plan is generated based on the difference for each department.

Images