Information processor, method for processing information, and information processing program

Abstract

To easily acquire a predicted value of a medical consultation fee based on one of disease data and a disease classification code.SOLUTION: An information processor has a first learning model generation unit for generating a first learning model to predict a medical consultation fee by using one of disease data and a disease classification code which correspond to a medical action performed in one of a medical institution and a nursing facility and facility constituting data of one of a medical institution and a nursing facility as input data and by using receipt data corresponding to one of disease data and a disease classification code and the facility constituting data as teacher data.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and an information processing program. [Background technology]

[0002] In the above technical field, Patent Document 1 discloses a technique for inputting a disease classification code and referring to facility standards and calculation requirements to calculate a theoretical value of medical fee points. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7158093 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the technology described in the above document requires reference to a large number of complex facility standards and calculation requirements, which requires many calculations to calculate theoretical medical fee points based on disease classification codes, making it difficult to predict medical fees.

[0005] An object of the present invention is to provide a technique for solving the above-mentioned problems. [Means for solving the problem]

[0006] In order to achieve the above object, the device according to the present invention comprises: a first learning model generation unit that generates a first learning model for predicting medical fees using injury / illness data or disease classification code corresponding to medical procedures performed at a medical institution or a nursing facility and facility configuration data of the medical institution or the nursing facility as input data and using receipt data corresponding to the injury / illness data or disease classification code and the facility configuration data as training data; The information processing device is provided with:

[0007] In order to achieve the above object, the device according to the present invention comprises: This is an information processing device that takes as input data illness / injury data or disease classification code corresponding to medical treatment performed at a medical institution or nursing facility, and facility configuration data of the medical institution or nursing facility, and outputs a predicted value of medical fees using a first learning model generated using prescription data corresponding to the illness / injury data or disease classification code and the facility configuration data as training data.

[0008] In order to achieve the above object, the method according to the present invention comprises: a first learning model generation step in which a first learning model generation unit uses injury / illness data or disease classification code corresponding to a medical procedure performed at a medical institution or a nursing facility and facility configuration data of the medical institution or the nursing facility as input data, and uses receipt data corresponding to the injury / illness data or disease classification code and the facility configuration data as training data to generate a first learning model for predicting medical fees; The present invention relates to an information processing method.

[0009] In order to achieve the above object, the program according to the present invention comprises: a first learning model generation step of generating a first learning model for predicting medical fees using injury / illness data or disease classification code corresponding to medical procedures performed at a medical institution or a nursing facility and facility configuration data of the medical institution or the nursing facility as input data and prescription data corresponding to the injury / illness data or disease classification code and the facility configuration data as training data; It is an information processing program that causes a computer to execute the above. Effect of the Invention

[0010] According to the present invention, a predicted value of medical fees based on injury / illness data or disease classification code can be easily obtained. [Brief description of the drawings]

[0011] [Figure 1]1 is a block diagram showing a configuration of an information processing device according to a first embodiment. [Diagram 2] FIG. 1 is a diagram explaining a receipt listing medical fees. [Diagram 3] FIG. 11 is a block diagram showing a functional configuration of an information processing device according to a second embodiment. [Figure 4] FIG. 11 is a diagram showing input data to a learning model according to the second embodiment. [Figure 5A] FIG. 11 is a diagram showing the configuration of a predicted medical fee table according to the second embodiment. [Figure 5B] FIG. 11 is a diagram showing the configuration of a medical fee comparison table according to the second embodiment. [Figure 6A] FIG. 11 is a diagram showing a display example of an information processing device according to the second embodiment. [Figure 6B] FIG. 11 is a diagram showing a display example of an information processing device according to the second embodiment. [Figure 7A] FIG. 11 is a block diagram showing a hardware configuration of an information processing device according to a second embodiment. [Figure 7B] 10 is a flowchart showing a flow of processing in an information processing device according to a second embodiment. [Figure 8] FIG. 11 is a block diagram showing a functional configuration of an information processing device according to a third embodiment. [Figure 9] FIG. 13 is a diagram showing input data to a learning model according to the third embodiment. [Figure 10A] A diagram showing the configuration of a predicted medical fee table for a changed facility in the third embodiment. [Figure 10B] FIG. 13 is a diagram showing the configuration of a medical fee comparison table according to the third embodiment. [Figure 11A] FIG. 13 is a diagram showing a display example of an information processing device according to a third embodiment. [Figure 11B] FIG. 13 is a diagram showing a display example of an information processing device according to a third embodiment. [Figure 11C] FIG. 13 is a diagram showing a display example of an information processing device according to a third embodiment. [Figure 12] 13 is a flowchart showing a flow of processing in an information processing device according to the third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. However, the components described in the following embodiment are merely examples, and are not intended to limit the technical scope of the present invention.

[0013] In addition, the "patient data" used in this specification includes the patient's name, age, sex, address, injury name, height, weight, vital index, severity, nursing need, required nursing care level, medical history, medication history, surgery history, and hospitalization and discharge history. In addition, the "facility configuration" includes at least one of the following: the number of hospital beds, the number of medical staff, the configuration of a rehabilitation center, the presence or absence of a specialized space configuration, and the type of medical equipment. In addition, the "facility configuration" is not limited to these, but includes all configurations related to the "facility standards" that affect medical fees. In addition, the "facility data" includes the facility name, location, facility type (hospital, bed-equipped clinic, etc.), permitted number of hospital beds, bed classification (general hospital beds, convalescent hospital beds, etc.), advertised medical department, average length of hospital stay, medical and nursing needs, specialized space (functional training room, speech and hearing room, etc.), medical equipment (radiation equipment, surgery and examination equipment, etc.), number of medical personnel (doctors, paramedicals), and medical personnel expertise (clinical experience, qualifications, etc.).

[0014] [First embodiment] An information processing device 100 according to a first embodiment of the present invention will be described with reference to Fig. 1. The information processing device 100 is a device that predicts appropriate medical fees from injury / illness data or disease classification codes.

[0015] 1, the information processing device 100 includes a learning model generation unit 101. The learning model generation unit 101 receives, as input data, injury / illness data or disease classification codes corresponding to medical procedures performed at a medical institution or a nursing facility, and facility configuration data of the medical institution or the nursing facility, and receives, as training data, receipt data corresponding to the injury / illness data or disease classification codes and the facility configuration data, to generate a first learning model 111 for predicting medical fees.

[0016] According to this embodiment, a learning model can be used to easily obtain predicted medical fees based on injury / illness data or disease classification codes.

[0017] [Second embodiment] Next, an information processing device according to a second embodiment of the present invention will be described. The information processing device according to this embodiment uses injury / illness data or disease classification code corresponding to a medical procedure and facility configuration data of a medical institution or a nursing facility as input data, and uses receipt data corresponding to the injury / illness data or disease classification code and the facility configuration data as training data to generate a learning model for predicting medical fees. Then, based on the injury / illness data or disease classification code corresponding to an actually occurring medical procedure and the facility configuration data, a predicted value of medical fees predicted by the learning model (hereinafter also referred to as predicted medical fees) is displayed in comparison with the calculated medical fees (hereinafter also referred to as calculated medical fees) included in the issued receipt data, and an improvement of the receipt is proposed. In addition, the training data further includes the results of an examination by a receipt examination institution, and the learning model predicts receipt data approved by the examination by the receipt examination institution. Note that, in this embodiment, the medical fees included in the receipt data as the sales of the medical institution or nursing facility will be described as an example, but the sales are not limited to the medical fees included in the receipt data.

[0018] <Explanation regarding medical fees> Before explaining the configuration of the information processing device, the remuneration calculated for the action performed at the medical institution or the nursing facility will be explained. FIG. 2 is a diagram explaining a receipt 280 on which the medical remuneration is written. For example, FIG. 2 shows a receipt 280 in the case of "visiting a family clinic because of a fever, undergoing an influenza test and receiving a prescription", and the medical remuneration for this treatment is 731 points obtained by adding up all the accumulated points for the medical remuneration points for each item of treatment and the number of times. The medical expenses are 7,310 yen, and in the case of a 30% burden, the patient pays 2,190 yen, and if the medical remuneration examination is approved by the medical remuneration examination institution, 5,120 yen is paid to the medical institution by the insurance institution. If the medical remuneration examination is not approved by the medical remuneration examination institution, a notice of return is given and the medical remuneration is reissued.

[0019] <Functional configuration of information processing device> FIG. 3 is a block diagram showing the functional configuration of the information processing device 300. As shown in FIG.

[0020] The information processing device 300 includes a learning model generation unit 301, a medical fee prediction unit 302, a display control unit 303, and a database 304. The database 304 may be located outside the information processing device 300. For example, this includes a case where an external server installed outside Japan functions as the database 304.

[0021] The learning model generation unit 301 uses injury / illness data or disease classification code and facility configuration data of a medical institution or a nursing facility as input data, and uses receipt data corresponding to the injury / illness data or disease classification code and the facility configuration data as teacher data to generate a learning model (AI) 311. The injury / illness data may include, for example, an ICD (International Statistical Classification of Diseases and Related Health Problems) code defined by the World Health Organization (WHO) or an injury / illness name code defined by the Ministry of Health, Labor and Welfare.

[0022] The input data is not limited to injury / illness data or disease classification code and facility configuration data. It may include other data related to the issuance of medical receipts. In addition, virtual injury / illness data or disease classification code may be used as input data, and the theoretical value of medical fees calculated based on the virtual injury / illness data or disease classification code may be used as training data. The learning model (AI) 311 generated by the final data can output an approximation to the optimal theoretical value of medical fees.

[0023] Furthermore, as an option for the training data, the results of examination by a medical claims examination organization may be included. In this case, a learning model 311 that outputs medical claims approved by the examination by the medical claims examination organization can be generated.

[0024] These input data or training data are extracted from files collected from medical institutions, including, for example, the following files: UKE file: disease name, date of birth, gender, medical department, medicine, medical procedure, point data H file: Date of birth, severity, and nursing need data EF file: Date of birth, gender, ward, department, disease name, medicine, medical procedure data · FF1 file: Patient basics, ward, medical department, life history, admission and discharge route, medical condition, level of care, outcome data · Hospital bed function report: Hospital bed function, medical department, number of patients, route, staff, main number of actions, etc. of medical institutions nationwide Other public data: Prefectural medical information (unified standard), patient surveys, demographics, list of national facility standards, medical facility surveys, hospital reports, diagnosis procedure combination (DPC) data, surveys of nursing care facilities and businesses, etc. Other hospital-specific data: electronic medical record data, medical data, number of staff, hospital structure, financial statements, labor management data, departmental statistics, etc. -Other files include FF3 files, FF4 files, D files, K files, etc.

[0025] The medical fee prediction unit 302 has a predicted medical fee table 321, receives the injury / illness data or disease classification code of the prediction target for predicting the medical fee and facility configuration data from a terminal of the medical institution or nursing facility, inputs them to the learning model 311, and acquires and outputs the predicted medical fee. Note that the injury / illness data or disease classification code of the prediction target and the facility configuration data received from the terminal of the medical institution or nursing facility may be stored in the database 304 and then read out by the medical fee prediction unit 302.

[0026] The display control unit 303 has a medical fee comparison table 331, and receives issued receipt data (calculated medical fee) issued based on the injury / illness data or disease classification code of the prediction target from a terminal of a medical institution or nursing facility, and compares it with the predicted medical fee predicted by the learning model 311 based on the same injury / illness data or disease classification code of the prediction target. Then, in response to a viewing request from the terminal of the medical institution or nursing facility, it controls the output of a display screen comparing the medical fee of the issued receipt data with the predicted medical fee to the terminal of the medical institution or nursing facility.

[0027] The database 304 has an input data storage unit 342, a teacher data storage unit 343, a learning model generation algorithm 344, a facility configuration data storage unit 345, and a predicted medical fee storage unit 346.

[0028] The input data storage unit 342 stores data collected as input data for generating the learning model 311. The teacher data storage unit 343 stores data collected as teacher data for generating the learning model 311. The input data and the teacher data are linked by the same medical treatment. The learning model generation algorithm 344 is an algorithm for generating the learning model 311. In this embodiment, a supervised learning model 311 is used, but an unsupervised learning model 311 using deep learning, for example, a neural network, may also be used. The facility configuration data storage unit 345 stores facility configuration data including information on facility equipment and personnel corresponding to an identifier (ID) for identifying a medical institution or a nursing facility. The predicted medical fee storage unit 346 stores a predicted medical fee predicted by the learning model 311, and this predicted medical fee may be used for learning the learning model 311.

[0029] In FIG. 3, the learning model generation unit 301 and the medical fee prediction unit 302 are treated as an information processing unit 310, and detailed illustrations thereof are omitted in the following FIG. 8 and FIG. 13A.

[0030] (Input data for the learning model) FIG. 4 is a diagram showing input data 400 to the learning model 311.

[0031] The input data 400 includes a large number of receipt data issued in a large number of facilities and their medical fees. The input data 400 stores facility configuration data 402 corresponding to a facility ID 401. The input data 400 stores injury / illness data or disease classification codes 403 of a plurality of medical treatments issued corresponding to the facility ID 401, correspondingly issued receipt data 404, and an optional receipt examination result 405. The facility configuration data 402 stores information on personnel and equipment required for determining facility standards and calculation requirements, such as the number of hospital beds, the number of doctors, the number of nurses, the types and numbers of medical equipment, and working conditions, linked to the facility ID 401. The input data 400 may include injury / illness data or disease classification codes of virtual medical treatments in a virtual facility and medical fees of theoretically issued receipt data.

[0032] (Predicted medical fee table) 5A is a diagram showing the configuration of the predicted medical fee table 321. The predicted medical fee table 321 is used by the medical fee prediction unit 302 to output the predicted medical fee predicted by the learning model 311 in correspondence with the injury / illness data or disease classification code of a certain medical treatment at a certain facility and the facility configuration data.

[0033] The predicted medical fee table 321 stores facility configuration data 512 corresponding to the facility ID 511 of the prediction target. It also stores injury / illness data or disease classification code 514 for all medical treatments in a target year / month 513, corresponding predicted medical fees 515, and predicted medical fees 516 in total by month.

[0034] (Medical fee comparison table) 5B is a diagram showing the configuration of the medical fee comparison table 331. The medical fee comparison table 331 is used by the display control unit 303 to generate a display screen that compares calculated medical fees and predicted medical fees for issued medical receipt data based on the same injury / illness data or disease classification code, and to suggest improvements to the medical receipt data.

[0035] The medical fee comparison table 331 stores, corresponding to the target year and month 522 of the facility ID 521 to be compared, issued receipt data 523 for the monthly total and each receipt, predicted medical fees 524 for the monthly total and each receipt, a comparison 525 of medical fees for the monthly total and each receipt, and improvement suggestions 526 for the monthly total and each receipt.

[0036] (Example) 6A is a diagram showing a display example 610 by the display control unit 303 of the information processing device 300. The display example 610 is generated based on the medical fee comparison table 331, and displays a comparison between the medical fee billed for past medical procedures performed at a medical institution or a nursing facility and the total predicted value.

[0037] The display example 610 displays the hospital's billed amount and the predicted value for each medical procedure. For example, a specific disease medical care management fee of 1,027,500 yen is billed, but the predicted value by the learning model 311 is 1,335,000 yen, and the billing rate is displayed as 77.7%. This actual billed amount of 1,027,500 yen is high compared to the national average of 850,500 yen, but compared to the predicted value of 1,335,000 yen, it indicates that there is room for improvement of 22%. Similarly, the actual billed amount and the predicted value are displayed for cerebrovascular rehabilitation, musculoskeletal rehabilitation, etc.

[0038] Furthermore, in the "condition" column, it is possible to suggest measures that can be taken to improve the billing rate by automatically calculating and displaying the percentage of diseases, the number of patients undergoing rehabilitation, etc. The display control unit 303 may display additional items corresponding to additional points that were not included in the medical fee claimed for past medical treatment performed at a medical institution or nursing facility.

[0039] 6B is a diagram showing another display example 620 by the display control unit 303 of the information processing device 300. In the display example 620, past medical results and improvable medical treatments are displayed in comparison for a medical fee item corresponding to a certain medical treatment based on the provisions of the calculation requirements.

[0040] In this example, it is assumed that the number of calculation days 621 and the calculation number 6222 are specified for each medical fee item as the calculation requirements. Here, the calculation number is set as the upper limit of the number of calculations per day (here, 1 time) and the upper limit of the calculation days (here, 150 days from onset). In addition, the calculation number is set as an upper limit of 13 times per month when it exceeds 150 days. It is not limited to the medical fee standards established by the Ministry of Health, Labor and Welfare, and original standards (upper limit 3 times per week, number of days excluded from calculation after onset: 14) may be set.

[0041] As an example, assume that medical fee points of 623, 624, and 625 were obtained. In other words, the injury was a femur fracture, and the onset date was June 15th. The actual billed points were 555 points for June and 3700 points for July. In contrast, the total value of 2960 points (626) for the 16 days from onset to the end of June, which is possible based on the calculation requirements for musculoskeletal rehabilitation fees (I), and 5735 points (627) for the 31 days of July are displayed in comparison.

[0042] <Hardware configuration of information processing device> Fig. 7A is a block diagram showing a hardware configuration of information processing device 300. In Fig. 7A, the same components as those in Fig. 3A are given the same reference numerals, and duplicated explanations will be omitted.

[0043] In Fig. 7A, a CPU (Central Processing Unit) 710 is a processor for arithmetic control, and executes a program to realize the components of Fig. 3A. A ROM (Read Only Memory) 720 stores fixed data and programs, such as initial data and programs. A network interface 730 controls communication with communication terminals at each facility via a network. Although the learning model 311 is illustrated as an independent component in Fig. 7A, the CPU 710 may configure the learning model 311 in the RAM 740.

[0044] The RAM (Random Access Memory) 740 is a random access memory used by the CPU 710 as a work area for temporary storage. The RAM 740 has an area reserved for storing data necessary for realizing this embodiment. The facility configuration data (for learning) 741 is facility configuration data of each facility collected for learning the learning model 311. The injury / illness data / disease classification code (for learning) 742 is injury / illness data or disease classification code collected for learning the learning model 311. The receipt data (for learning) 743 is receipt data issued corresponding to the injury / illness data / disease classification code (for learning) 742 collected for learning the learning model 311. The receipt review result (for learning) 744 is the receipt review result of whether the receipt data (for learning) 743 has been approved, assessed, or returned by the review by the receipt review organization. The facility configuration data (for prediction) 745 is facility configuration data of the facility for which the medical fee is predicted by the learning model 311. The injury / illness data / disease classification code (for prediction) 746 is injury / illness data or disease classification code for which medical fees are predicted by the learning model 311. The predicted medical fees 747 are medical fees predicted by the learning model 311 based on the facility configuration data (for prediction) 745 and the injury / illness data / disease classification code (for prediction) 746. The facility configuration data (for prediction) 745, the injury / illness data / disease classification code (for prediction) 746, and the predicted medical fees 747 correspond to the predicted medical fees table 321. The medical fees comparison table 331 is a table used by the display control unit 303 described in FIG. 3C. The viewed screen data 748 is display screen data generated by the display control unit 303 in response to a viewing request from a terminal of the facility. The transmitted / received data 749 is data transmitted to and received from a terminal of the facility via a network.

[0045] The storage 750 stores a database, various parameters, or the following data or programs required to realize this embodiment. The database 304 is a database including the input data storage unit 342, the teacher data storage unit 343, the learning model generation algorithm 344, the facility configuration data storage unit 345, and the predicted medical fee storage unit 346 described in FIG. 3. The storage 750 stores the following programs. The information processing program 751 is a program that controls the entire information processing device 300. The input data collection module 752 is a module for collecting input data and teacher data for generating the learning model 311. The learning model generation module 753 is a module for generating the learning model 311 based on the input data and teacher data. The predicted medical fee acquisition module 754 is a module for inputting the facility configuration data (for prediction) 745 and the injury / disease data / disease classification code (for prediction) 746 into the generated learning model 311 and acquiring the predicted medical fee. The medical fee comparison module 756 is a module that compares the calculated medical fee of the issued issued receipt data with the predicted medical fee based on the same facility configuration data (for prediction) 745 and injury / disease data / disease classification code (for prediction) 746. The viewing screen generation module 757 is a module that generates a viewing screen in response to a viewing request from a terminal of the facility.

[0046] It should be noted that the RAM 740 and storage 750 in FIG. 7A do not show programs and data related to the general-purpose functions of the information processing device 300 or other feasible functions.

[0047] <Processing flow of information processing device> 7B is a flowchart showing the flow of processing by the information processing device 300. This flowchart is executed by the CPU 710 in FIG. 7A using the generated learning model 311 and the RAM 740, and realizes the components in FIG.

[0048] In step S701, the information processing device 300 collects pairs of input data (injury / illness data or disease classification code, facility configuration data) and training data (corresponding medical receipt data, optional medical receipt review results). In step S703, the information processing device 300 inputs the input data and training data to the learning model 311 and causes it to learn. In step S705, the information processing device 300 determines whether sufficient learning has been completed. If not, the process returns to step S701 and the learning is repeated. When learning is sufficiently completed, in step S707, the information processing device 300 acquires pairs of the injury / illness data or disease classification code to be predicted and the facility configuration data, and inputs them to the learning model 311. Then, in step S709, the information processing device 300 acquires the predicted medical fees predicted by the learning model 311.

[0049] In step S711, the information processing device 300 waits for a request for viewing from the facility, and if there is a request, in step S713, generates a screen for comparing calculated medical fees and predicted medical fees for issued receipt data based on the same combination of injury / illness data or disease classification code and facility configuration data. Then, in step S715, the information processing device 300 displays the comparison screen and receipt improvement proposals (see FIG. 6A) on the terminal that made the request.

[0050] According to this embodiment, it is possible to easily obtain predicted medical fees based on injury / disease data or disease classification codes using a learning model and compare them with medical fees based on issued receipt data. It is also possible to propose more appropriate receipt generation based on the current facility configuration and number of medical treatments in a medical institution or nursing facility.

[0051] [Third embodiment] Next, an information processing device according to a third embodiment of the present invention will be described. The information processing device according to this embodiment is different from the second embodiment in that it proposes an improvement plan for facility configuration for more appropriate receipt generation, rather than the current facility configuration or the number of medical treatments. In this embodiment, a predetermined period or a predetermined number of multiple injury / disease data or disease classification codes and changed facility configuration data are input, and a plurality of medical fees predicted using a learning model are output. Then, when the increment from the medical fee of the current facility corresponding to the same injury / disease data or disease classification codes generated in a predetermined period or a predetermined number to the medical fee of the changed facility is greater than a predetermined value, a change in facility configuration is presented. Alternatively, the change in facility configuration, the medical fee included in the medical fee predicted based on the change in facility configuration, and the medical fee included in the receipt data issued by the medical institution or care facility are displayed in comparison. Since other configurations and operations are the same as those of the second embodiment, the same configurations and operations are assigned the same symbols and detailed descriptions thereof are omitted.

[0052] <Functional configuration of information processing device> Fig. 8 is a block diagram showing the functional configuration of an information processing device 800. In Fig. 8, the same components as those in Fig. 3A are given the same reference numerals, and duplicated explanations will be omitted.

[0053] In FIG. 8, an information processing device 800 includes a display control unit 803, a changed facility medical fee prediction unit 804, and a facility configuration change unit 805.

[0054] The facility configuration change unit 805 changes the facility configuration data of the current facility to be predicted, and inputs the changed facility to the medical fee prediction unit 804. Note that it is preferable that the changes in the facility configuration change unit 805 are not unlimited, but are limited to a predetermined range of changes from the facility configuration data of the current facility.

[0055] The medical fee prediction unit 804 for the changed facility has a predicted medical fee table (not shown), and inputs the disease data or disease classification code for which the medical fee is to be predicted and the facility configuration data changed by the facility configuration change unit 805 to the learning model 311, and acquires and outputs the predicted medical fee for the changed facility. Note that if the change in the facility configuration change unit 805 results in an increase in the number of medical treatments, this may be reflected in the number of disease data or disease classification codes to be input.

[0056] The display control unit 803 has a medical fee comparison table 831, and receives the medical fee of the issued receipt data issued based on the injury / illness data or disease classification code of the prediction target from the terminal of the medical institution or the nursing facility. Then, it compares the predicted medical fee of the current facility predicted by the learning model 311 based on the injury / illness data or disease classification code of the same prediction target and the facility configuration data of the current facility with the predicted medical fee of the changed facility predicted by the learning model 311 based on the injury / illness data or disease classification code of the same prediction target and the facility configuration data of the changed facility. Then, in response to a viewing request from the terminal of the medical institution or the nursing facility, it controls the output of a display screen that compares the calculated medical fee of the issued receipt data, the predicted medical fee of the current facility, and the predicted medical fee of the changed facility to the terminal of the medical institution or the nursing facility. Even if the display screen compares the medical fee of the issued receipt data with the predicted medical fee of the changed facility, it can present the improvement in medical fee (increase in sales) due to the change of the facility.

[0057] (Input data for the learning model) FIG. 9 is a diagram showing input data to the learning model 311.

[0058] Input data 910 in Fig. 9 shows an example of facility configuration data for a current facility, and input data 920 in Fig. 9 shows an example of facility configuration data for a changed facility. The facility configuration data includes, for example, the number of hospital beds, the number of doctors, the number of nurses, medical equipment, and the like.

[0059] (Predicted medical fee table for changed facilities) Fig. 10A is a diagram showing the configuration of a predicted medical fee table 1010 for a changed facility. The predicted medical fee table 1010 for a changed facility (not shown in Fig. 8) is used by the changed facility's medical fee prediction unit 804 to output the predicted medical fee for the changed facility predicted by the learning model 311 in correspondence with injury / illness data or disease classification code of a certain medical treatment at a certain facility and facility configuration data of the changed facility.

[0060] The predicted medical fee table 1010 of the changed facility stores facility configuration data 1012 of the current facility and the changed facility corresponding to the prediction target facility ID 1011. Each facility configuration data 1012 stores injury / illness data or disease classification code 1013 of all medical treatments in the target year / month, and corresponding predicted medical fee 1014 of the changed facility predicted by the learning model 311.

[0061] (Medical fee comparison table) 10B is a diagram showing the configuration of the medical fee comparison table 831. The medical fee comparison table 831 is used by the display control unit 803 to generate a display screen that compares the medical fee of issued receipt data, the predicted medical fee of the current facility, and the predicted medical fee of a changed facility, using the same injury / illness data or disease classification code, and to suggest improvements to the facility configuration.

[0062] The medical fee comparison table 831 stores illness / disease classification code 1022 for medical treatment in association with the target year and month of the facility ID 1021 to be compared. It also stores issued receipt data 1023, and predicted medical fees for the current facility and changed facility for the monthly total and each receipt 1024 in association with each illness / disease data / disease classification code 1022. It also stores changed facility configuration 1025 for the monthly total and maximum medical fees for each receipt, and proposed facility improvement measures 1026.

[0063] (Example) 11A to 11C are examples of display screens displayed on the display unit by the display control unit 803. For example, the display screen presents changes to facility configuration. Alternatively, the actual value of medical fees for the current facility configuration and a theoretical change value in which the increment is greater than a predetermined value are displayed. Alternatively, the profit for the current facility configuration and the profit for a changed facility configuration are displayed. Furthermore, the change in facility configuration, the actual value of medical fees included in the receipt data issued by the medical institution or care facility, and the predicted value of medical fees that will increase due to the change in facility configuration may be displayed in comparison.

[0064] Fig. 11A is a diagram showing a display example 1110 of the information processing device 800. Fig. 11A is an example of a display screen displaying the actual value of medical fees in the current facility configuration data, improvement measures for the predicted value of medical fees in the current facility configuration data, and a proposed change to the facility configuration for the predicted value of medical fees in the facility configuration data of the changed facility. Note that in Fig. 11A, explanations of display items similar to those in Fig. 6A are omitted.

[0065] As shown in the display example 1110 of FIG. 11A, the billing amount at the hospital, the predicted value for the current facility configuration, and the predicted value for the changed facility configuration are displayed for each medical procedure. That is, the "Facility Change Prediction" column displays the predicted medical fee by changing the facility configuration to have the highest improvement effect, and the "Action Plan" column displays the changes to the facility configuration to have the highest improvement effect. For example, for "cerebrovascular rehabilitation," the plan states, "Assuming one unit per patient per day, 100% billing is possible with an increase of one PT and two OTs. In addition, a speech-language-hearing room is required to raise the standard." For "feeding function therapy," the plan states, "VE / VF evaluation function needs to be improved. To do this, STs and radiographers and radiologists who can take radiographs are required. The number of billings can also be increased with the intervention of ward nurses."

[0066] Fig. 11B is a diagram showing another display example 1120 of the information processing device 800. The display example 1120 is, for example, a display screen showing the details of an increase in medical fees due to a change in medical function. As shown in Fig. 11B, when a long-term care bed is changed to a community-based comprehensive care bed, the number of nurses needs to be increased compared to when it was a long-term care bed (other measures include reducing the length of hospital stay and admitting patients with specific diseases), but the unit price also increases, which ultimately leads to increased revenue.

[0067] Figure 11B shows the before and after changes to the facility configuration, and displays the proposed changes to the facility configuration: "We need to add three nurses and three assistants. Two more patients will be admitted and discharged per month. In addition, adding one assistant will make it possible to obtain an additional surcharge for the XX department."

[0068] 11C is a diagram showing yet another display example 1130 of the information processing device 800. The display example 1130 is a display screen showing details of an increase in medical fees when, for example, a psychiatric treatment ward is changed to a medical treatment ward.

[0069] Figure 11C shows the before and after images of the facility configuration change, and displays comments on the improvement effects of the change in facility configuration, such as "Information such as occupancy rate will increase by xx%, and medical treatment cost will increase by xx yen."

[0070] <Processing flow of information processing device> Fig. 12 is a flowchart showing the flow of processing of the information processing device 800. This flowchart is executed by the CPU 710 in Fig. 7A using the generated learning model 311 and the RAM 740, and realizes the components in Fig. 8. In Fig. 12, the same steps as those in Fig. 7 are given the same step numbers, and duplicated explanations are omitted.

[0071] In step S1210, the information processing device 800 generates changed facility configuration data and inputs it to the learning model 311. Then, in step S1211, the information processing device 800 obtains the predicted medical fees predicted by the learning model 311 based on the changed facility configuration data. In step S1212, the information processing device 800 determines whether there are other facility changes, and if there are, returns to step S1210 and repeats obtaining the predicted medical fees from the learning model 311.

[0072] In step S1213, the information processing device 800 creates a viewing screen showing an appropriate change in facility configuration (e.g., the change that will result in the largest increase in medical fees). Note that the change in facility configuration may also increase the burden of expenses, such as one-time expenses for medical facilities and medical equipment, and ongoing expenses such as labor costs, and a change in facility configuration that will maximize the actual profit obtained by subtracting the increase in expenses from the increase in sales due to the increase in medical fees may be presented. In addition, the burden of expenses for medical facilities and medical equipment may be considered in terms of cost depreciation. The display on the viewing screen includes medical fees corresponding to the issued receipt data issued in the current facility configuration and predicted medical fees predicted by the learning model 311 when the facility configuration is changed. Alternatively, it may include medical fees corresponding to the optimal theoretical value of predicted medical fees predicted by the learning model 311 in the current facility configuration.

[0073] According to this embodiment, a learning model can be used to easily obtain predicted medical fees based on injury / disease data or disease classification codes and predicted medical fees with facility configurations changed, and the two medical fees can be compared. Then, it is possible to propose improvements to facility configurations that will enable more appropriate receipt generation in medical institutions or nursing facilities. That is, according to this embodiment, a medical fee database / platform can be constructed, and improvements to facility configurations that will increase the hospital's income can be proposed. Also, improvements to facility configurations that take into account the cost burden of changing the facility configuration can be proposed.

[0074] [Other embodiments] Although the present invention has been described above with reference to the embodiments, the present invention is not limited to the above-mentioned embodiments. Various modifications can be made to the configuration and details of the present invention that can be understood by a person skilled in the art within the technical scope of the present invention. In addition, systems or devices that combine separate features included in each embodiment in any way are also included in the technical scope of the present invention. For example, an embodiment in which both the facility configuration change unit 805 of the third embodiment and the location environment change unit 1305 of the fourth embodiment are provided is also possible, and a configuration in which the facility configuration change unit 805 and the location environment change unit 1305 are associated with each other can also be realized.

[0075] The present invention may be applied to a system consisting of multiple devices, or to a single device. Furthermore, the present invention is also applicable to a case where an information processing program for implementing the functions of the embodiments is supplied to a system or device and executed by a built-in processor. In order to implement the functions of the present invention with a computer, a program installed on a computer, a medium storing the program, a server for downloading the program, and a processor for executing the program are also included in the technical scope of the present invention. In particular, at least a non-transitory computer readable medium storing a program for causing a computer to execute the processing steps included in the above-mentioned embodiments is included in the technical scope of the present invention.

Claims

1. A first learning model generation unit that generates a first learning model for predicting medical fees using injury / illness data or disease classification code corresponding to medical treatment performed at a medical institution or a nursing facility and facility configuration data of the medical institution or the nursing facility as input data and prescription data corresponding to the injury / illness data or disease classification code and the facility configuration data as training data. An information processing device comprising:

2. The information processing device according to claim 1 , further comprising a medical fee prediction unit that inputs injury / illness data or disease classification code and facility configuration data, and outputs a predicted value of medical fee predicted using the first learning model.

3. The first learning model generation unit further includes an examination result by a prescription examination institution as the teacher data. The information processing device according to claim 1 .

4. The information processing device according to claim 1 , further comprising a first display control unit that displays the predicted value of the medical fee in comparison with the medical fee included in the prescription data issued by the medical institution or nursing facility.

5. the medical fee prediction unit inputs facility configuration data before and after a change to a predetermined facility configuration, and outputs a first predicted value and a second predicted value of the medical fee; The information processing apparatus according to claim 2 , further comprising a second display control unit that presents the predetermined change in the facility configuration when an increment from the first predicted value to the second predicted value is greater than a predetermined value.

6. the medical fee prediction unit receives facility configuration data after a predetermined change to the facility configuration, and outputs a second predicted value of the medical fee; The information processing device according to claim 2, further comprising a third display control unit that displays the specified facility configuration changes, the medical fees included in the prescription data issued by the medical institution or nursing facility, and the second predicted value in comparison.

7. An information processing device that receives as input data illness / injury data or disease classification code corresponding to medical treatment performed at a medical institution or nursing facility, and facility configuration data of the medical institution or nursing facility, and outputs a predicted value of medical fees using a first learning model generated using prescription data corresponding to the illness / injury data or disease classification code and the facility configuration data as training data.

8. An information processing method in which a first learning model generation unit uses injury / illness data or disease classification code corresponding to medical procedures performed at a medical institution or nursing facility and facility configuration data of the medical institution or nursing facility as input data, and prescription data corresponding to the injury / illness data or disease classification code and the facility configuration data as training data to generate a first learning model for predicting medical fees.

9. An information processing program that causes a computer to execute a first learning model generation step of generating a first learning model for predicting medical fees using injury / illness data or disease classification code corresponding to medical treatment performed at a medical institution or nursing facility and facility configuration data of the medical institution or nursing facility as input data, and prescription data corresponding to the injury / illness data or disease classification code and the facility configuration data as training data.

Images