A support system for determining the composition of herbal medicine preparations.

Abstract

To provide a determination support device, a determination support method, and a determination support program that can propose Chinese herbal and crude drug preparations suitable for a patient's condition by taking into account a composition of the Chinese herbal and crude drug preparations.SOLUTION: A determination support device for Chinese herbal and crude drug preparations of the present invention comprises a calculation unit that has input information including constraints based on an evidence of Chinese medicine and indication as input, and output information including Chinese herbal and crude drug preparations suitable for the input information as output. The calculation unit has a relationship model that, on the basis of a learning machine learned to have herbal and crude drugs forming the Chinese herbal and crude drug preparations as model input, and the information including constraints based on the evidence of Chinese medicine and suitable for the model input and the indication as model output, defines a relationship between the model input and the model output. The calculation unit, on the basis of the relationship model, from the model output highly related to the input information that has been input, outputs the model input corresponding to the model output, and has the Chinese herbal and crude drug preparations formed of the herbal and crude drugs included in the model input that has been output as output information.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] The present invention relates to a decision support device, a decision support method, and a decision support program for Kampo medicine preparations.

Background Art

[0002] In recent years, information processing devices for supporting the prescription of Kampo medicine have been proposed. For example, the information processing device described in Patent Document 1 uses a learned model created based on information on the symptoms of past patients and diagnostic result information for past patients by Kampo specialists. That is, by inputting information on the symptoms of a new patient into this learned model, possible diseases, treatment methods, and Kampo medicine preparations to be prescribed are output. Therefore, it is possible to provide an optimal diagnostic result regardless of the experience and proficiency of the doctor.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the learning model implemented in the above information processing device is based on the experience of past doctors and does not consider the relationship with symptoms after analyzing the Kampo medicine preparation itself in detail. Therefore, there was room for improvement in proposing Kampo medicine preparations suitable for symptoms. The present invention has been made to solve this problem, and an object thereof is to provide a decision support system, a decision support method, and a decision support program for Kampo medicine preparations that can propose a Kampo medicine preparation suitable for a patient's symptoms in consideration of the composition of the Kampo medicine preparation.

Means for Solving the Problems

[0005] The Kampo herbal medicine preparation determination support device of the present invention according to the first aspect includes a calculation unit that takes input information including restrictive conditions and indications based on the diagnosis of Kampo medicine as input, and outputs output information including Kampo herbal medicine preparations that conform to the input information, and has a relational model that defines the relationship between the model input and the model output based on a learning device that has been trained to take the crude drugs constituting the Kampo prescription preparation as model input and the information including restrictive conditions and indications based on the diagnosis of Kampo medicine that conform to the model input as model output, and is configured to output the Kampo herbal medicine preparation composed of the crude drugs included in the output model input as output information.

[0006] The decision support device for Kampo herbal medicine preparations according to the second aspect of the present invention is such that, in the decision support device according to the first aspect, the composition of the crude drugs in the Kampo herbal medicine preparation, the limiting conditions, and the indications are extracted from the package insert information for medical Kampo preparations and the manufacturing and marketing approval standards for general-use Kampo preparations.

[0007] The Kampo herbal medicine preparation determination support device of the present invention according to the third aspect is a determination support device according to the first or second aspect described above, wherein the limiting conditions of the input information include physical strength conditions relating to physical strength, the physical strength conditions related to the Kampo prescription preparation are defined, and the calculation unit is configured to exclude the Kampo herbal medicine preparation from the output information when the physical strength conditions included in the input information do not match the physical strength conditions related to the Kampo herbal medicine preparation included in the output information.

[0008] The method for supporting the determination of a Kampo herbal medicine preparation according to the fourth aspect of the present invention comprises a step of inputting the input information to a calculation unit that takes input information including limiting conditions based on the diagnosis of Kampo medicine and indications as input, and outputs output information including a Kampo herbal medicine preparation that conforms to the input information, thereby outputting the output information. The calculation unit has a relational model that defines the relationship between the model input and the model output based on a learning device that has learned to take the crude drugs constituting the Kampo prescription preparation as model input and the information including limiting conditions based on the diagnosis of Kampo medicine and indications that conforms to the model input as model output. Based on the relational model, the calculation unit outputs the corresponding model input from the model output that is highly related to the input input information, and the Kampo herbal medicine preparation composed of the crude drugs included in the output model input is configured to be output information.

[0009] The fifth aspect of the present invention relates to a Kampo herbal medicine preparation determination support program, which causes a computer to take input information including limiting conditions based on Kampo medicine and indications as input, and output information including Kampo herbal medicine preparations that conform to the input information as output, thereby executing a step of outputting the output information, wherein the calculation unit has a relational model that defines the relationship between the model input and the model output, based on a learning device that has learned to take the crude drugs constituting the Kampo prescription preparation as model input and the information including limiting conditions based on Kampo medicine and indications that conform to the model input as model output, and based on the relational model, outputs the corresponding model input from the model output that is highly related to the input input, and outputs the Kampo herbal medicine preparation composed of the crude drugs included in the output model input as output information.

[0010] The determination support device for Kampo herbal medicine preparations according to the sixth aspect of the present invention is as follows: The system includes a calculation unit that takes input information including restrictive conditions and indications based on the diagnosis of traditional Chinese medicine as input, and outputs output information including traditional Chinese herbal medicine preparations that match the input information. A device for determining Kampo herbal medicine preparations, wherein the composition of the crude drugs in the Kampo herbal medicine preparation, the limiting conditions, and the indications are extracted from the package insert information for medical Kampo medicine preparations and the manufacturing and marketing approval standards for Kampo medicine preparations for general use.

[0011] The method for supporting the determination of Kampo herbal medicine preparations according to the seventh aspect of the present invention is as follows. The system includes a step of inputting the input information to a calculation unit that takes input information including restrictive conditions and indications based on the diagnosis of traditional Chinese medicine as input, and outputs output information including traditional Chinese herbal medicine preparations that match the input information, thereby outputting the output information. A method for supporting the determination of a Kampo herbal medicine preparation, wherein the composition of the crude drugs in the Kampo herbal medicine preparation, the limiting conditions, and the indications are extracted from the package insert information for medical Kampo medicine preparations and the manufacturing and marketing approval standards for Kampo medicine preparations for general use.

[0012] The decision-making support program for Kampo herbal medicine preparations of the present invention, relating to the eighth aspect, is as follows: A calculation unit takes input information including restrictive conditions and indications based on the diagnosis of traditional Chinese medicine as input, and outputs output information including herbal medicine preparations that match the input information. By inputting the input information, the calculation unit is made to execute the step of outputting the output information. A program to support the determination of Kampo herbal preparations, wherein the composition of the crude drugs in the aforementioned Kampo herbal preparation, the limiting conditions, and the indications are extracted from the package insert information for medical Kampo preparations and the manufacturing and marketing approval standards for over-the-counter Kampo preparations.

[0013] The device for supporting the determination of Kampo herbal medicine preparations according to the ninth aspect of the present invention is as follows: The system includes a calculation unit that takes input information including restrictive conditions and indications based on the diagnosis of traditional Chinese medicine as input, and outputs output information including traditional Chinese herbal medicine preparations that match the input information. The aforementioned input information restriction conditions include physical fitness conditions, The aforementioned physical condition related to the aforementioned herbal medicine formulation is defined, A device for determining herbal medicines, wherein the calculation unit is configured to exclude herbal medicines from the output information when the physical condition included in the input information does not match the physical condition related to the herbal medicine preparation included in the output information.

[0014] The method for supporting the determination of Kampo herbal medicine preparations according to the tenth aspect of the present invention is as follows. The system includes a step of inputting the input information to a calculation unit that takes input information including restrictive conditions and indications based on the diagnosis of traditional Chinese medicine as input, and outputs output information including traditional Chinese herbal medicine preparations that match the input information, thereby outputting the output information. The aforementioned input information restriction conditions include physical fitness conditions, The aforementioned physical condition related to the aforementioned herbal medicine formulation is defined, A method for supporting the determination of a herbal medicine preparation, wherein the calculation unit is configured to exclude the herbal medicine preparation from the output information when the physical condition included in the input information does not match the physical condition related to the herbal medicine preparation included in the output information.

[0015] The decision-making support program for Kampo herbal medicine preparations of the present invention, relating to the eleventh aspect, is as follows: A calculation unit takes input information including restrictive conditions and indications based on the diagnosis of traditional Chinese medicine as input, and outputs output information including herbal medicine preparations that match the input information. By inputting the input information, the calculation unit is made to execute the step of outputting the output information. The aforementioned input information restriction conditions include physical fitness conditions, The aforementioned physical condition related to the aforementioned herbal medicine formulation is defined, A program for determining herbal medicines, wherein the calculation unit is configured to exclude herbal medicines from the output information when the physical condition included in the input information does not match the physical condition related to the herbal medicine preparation included in the output information. [Effects of the Invention]

[0016] According to the present invention, it is possible to propose a Kampo crude drug preparation suitable for the symptoms of a patient in consideration of the composition of the Kampo crude drug preparation.

Brief Description of the Drawings

[0017] [Figure 1] It is a schematic diagram of a system including a decision support device for a Kampo crude drug preparation according to an embodiment of the present invention. [Figure 2] It is a block diagram showing the hardware configuration of the decision support device used in FIG. 1. [Figure 3] It is a block diagram showing the software configuration of the decision support device used in FIG. 1. [Figure 4] It is an example of the medical Kampo preparation package insert information of the Kampo preparation. [Figure 5] It is an example of the general Kampo preparation manufacturing and sales approval standards of the Kampo preparation. [Figure 6] It is an example of the medical Kampo preparation package insert information of the crude drug preparation. [Figure 7] It is an example of the learning device included in the calculation unit. [Figure 8] It is a diagram for explaining the generation of the calculation unit using the learning device of FIG. 7. [Figure 9] It is a diagram showing the relationship model generated using the learning device of FIG. 7. [Figure 10] It is a flowchart showing the output process of the Kampo crude drug preparation. [Figure 11] It is a diagram showing an example of the output from the decision support device. [Figure 12] It is a block diagram showing another example of the decision support device.

Modes for Carrying Out the Invention

[0019] In this embodiment, the term "Kampo herbal medicine preparation" encompasses both Kampo preparations and herbal medicine preparations. Kampo preparations are preparations made using one or more types of extracts (extracts) or chopped or powdered herbs based on Kampo theory, and may contain additives. On the other hand, herbal medicine preparations are preparations made using one or more types of extracts (extracts, tinctures, fluid extracts, etc.) or chopped or powdered herbs, and may contain chemical drugs or additives. The decision support device 1 will now be described in detail.

[0020] <1. Hardware Configuration> First, the hardware configuration of the decision support device will be described. Figure 2 shows an example of the hardware configuration of the decision support device according to this embodiment.

[0021] This decision support device 1 is a computer in which a control unit 11, a memory unit 12, an external interface 13, and a communication interface 14 are electrically connected. For example, it can be made up of a general-purpose personal computer, a dedicated computer, or a tablet computer or smartphone. In Figure 2, the external interface 13 and the communication interface 14 are referred to as "external I / F" and "communication I / F".

[0022] The control unit 11 includes a CPU, RAM, ROM, etc., and is configured to perform various information processing based on programs and data. The storage unit 12 is composed of an auxiliary storage device such as an HDD or SSD, and stores a decision support program 121, herbal medicine preparation data 122, history data 123, and various data for driving the information processing device. The decision support program 121 is a program for outputting recommended herbal medicine preparations based on the above-mentioned inputs. The control unit 11 is configured to perform the processing of each step described later by interpreting and executing this decision support program 121.

[0023] The external interface 13 is an interface for connecting to an external device and is configured appropriately depending on the external device to be connected. In this embodiment, the external interface 13 is connected to the display device 4 and the input device 5. The display device 4 is, for example, a display and is used to display the inputs, outputs, etc., described above. The display is not particularly limited, and a known liquid crystal display or the like can be used. The input device 5 is a keyboard, mouse, etc., and is used to perform the inputs described above. In addition, various external devices can be connected to the external interface 13 as appropriate. For example, a touch panel display that serves as both an input device and an output device can be used.

[0024] The communication interface 14 is, for example, a wired LAN (Local Area Network) module, a wireless LAN module, etc., and is an interface for wired or wireless communication. In other words, the communication interface 14 is an example of a communication unit configured to communicate with other devices. For example, it can connect to a server via a network.

[0025] The specific hardware configuration of the decision support device 1 can be appropriately modified by omitting, substituting, and adding components depending on the embodiment. For example, the control unit 11 may include multiple processors. The control unit 11 may also be configured using an FPGA. The storage unit 12 may be configured using RAM and ROM included in the control unit 11.

[0026] <2. Software Configuration> Next, the software configuration of the decision support device 1 will be described. Figure 3 is a block diagram showing the software configuration of the decision support device. As shown in Figure 3, the control unit 11 of the decision support device 1 loads the information processing program 121 stored in the memory unit 12 into RAM, interprets and executes the decision support program 121 using the CPU, and functions as a computer equipped with a calculation unit 111.

[0027] As described above, the input to the calculation unit 111 is the indications and limiting conditions based on the diagnosis of Kampo medicine, obtained from the patient's symptoms, and the output is a Kampo herbal medicine preparation. These are extracted from the package insert information for medical Kampo preparations and the manufacturing and marketing approval standards for over-the-counter Kampo preparations. The information from the package insert information for medical Kampo preparations is not particularly limited, but can be obtained from the search site for package inserts of medical drugs on the website of the Pharmaceuticals and Medical Devices Agency (PMDA). There are 148 Kampo herbal medicine preparations specified for medical use, and there are 294 Kampo herbal medicine preparations specified for over-the-counter use, although these are not particularly limited. In this embodiment, the system is configured so that one of a total of 442 Kampo herbal medicine preparations is output from the second calculation unit 112, which will be described later. In this disclosure, the manufacturing and marketing approval standards for over-the-counter Kampo preparations refer to the manufacturing and marketing approval standards for over-the-counter Kampo preparations (Pharmaceutical and Food Safety Bureau, Ministry of Health, Labour and Welfare, April 1, 2017).

[0028] Figure 4 shows the package insert information for the herbal medicine preparation "Anchusan Extract Granules". As shown in the figure, this herbal medicine preparation contains the following crude drugs: Japanese Pharmacopoeia Cinnamon Bark, Japanese Pharmacopoeia Corydalis, Japanese Pharmacopoeia Oyster Shell, Japanese Pharmacopoeia Fennel, Japanese Pharmacopoeia Licorice, Japanese Pharmacopoeia Cardamom, and Japanese Pharmacopoeia Licorice (7 in total).

[0029] Furthermore, the package insert information for the herbal medicine preparation shown in Figure 4 includes the following description under the "Indications or Effects" section: "For the following conditions in a thin person with a tendency for abdominal muscle relaxation, experiencing stomach pain or abdominal pain, sometimes accompanied by heartburn, belching, loss of appetite, nausea, etc.: nervous gastritis, chronic gastritis, gastric atony." In this embodiment, nervous gastritis, chronic gastritis, and gastric atony are extracted from this description (a total of 3), and these become the indications included in the input information.

[0030] Traditional Chinese medicine has a concept called "sho," which refers to a patient's individual condition (constitution, physical strength, resistance, symptoms, etc.). It employs the idea of ​​treating illness by prescribing herbal medicines tailored to the individual's "sho." Therefore, in this embodiment, the "sho" is extracted from the "efficacy or effect" section. Specifically, the description "thin build with a tendency for abdominal muscle relaxation, stomach pain or abdominal pain, sometimes heartburn, belching, loss of appetite, and nausea" is broken down into its elements, and the "sho" extracted is thinness, relaxation (abdominal muscles), stomach pain, abdominal pain, heartburn, belching, loss of appetite, and nausea (a total of 8 items). These become the constraints to be included in the input information.

[0031] In this way, 532 indications and 284 limiting conditions are defined from the package insert information for 442 herbal medicine preparations for medical use and the manufacturing and marketing approval standards for over-the-counter herbal medicine preparations. These 442 herbal medicine preparations, 532 indications, 284 limiting conditions, and 173 herbs are stored in the memory unit 12 as herbal medicine preparation data 122.

[0032] Figure 5 shows a portion of the approval standards for the manufacture and sale of the over-the-counter herbal medicine "Naishitol G". Figure 6 shows a portion of the package insert information for the medical herbal medicine "Kyohyosuisan". In these cases as well, the indications, restrictions, and crude drugs have been extracted as described above.

[0033] Next, the configuration of the calculation unit 111 will be described. This calculation unit 111 is configured based on the learner described below. As shown in Figure 7, this learner 6 is a neural network comprising an input layer 61 having 173 nodes representing crude drugs, an output layer 62 having 816 nodes representing indications and limiting conditions, and an intermediate layer 63 having multiple layers. This neural network is then trained using the backpropagation method with the 442 Kampo crude drug preparations mentioned above as training data. For example, in the Kampo crude drug preparation shown in Figure 4, the seven crude drugs mentioned above are input to the corresponding nodes of the input layer 61, and the three indications and eight limiting conditions mentioned above are input to the corresponding nodes of the output layer 62 to perform the training. In other words, a multi-label classification model is trained. In this way, training is performed on the 442 Kampo crude drug preparations, and a trained learner 6 is generated.

[0034] Next, this learning unit 6 is used to generate a relationship model between each herbal medicine preparation and its indications and constraints. First, as shown in Figure 8, the herbs contained in each herbal medicine preparation are sequentially input to the corresponding nodes in the input layer 61, and scores for the indications and constraints are output. In other words, 816 scores for indications and constraints are output for 442 input pairs. The scores for each indication and constraint output at this time can be, for example, real values ​​normalized to between 0 and 1. In the example in Figure 8, for example, since a certain herbal medicine preparation contains herbs 001 and 002, these are input to the nodes in the input layer 61. For example, 1 is input to the node corresponding to herbs 001 and 002, and 0 is input to the other nodes, and the scores for each indication and constraint are output. Hereafter, the herbs constituting each herbal medicine preparation, which are the inputs, will be referred to as model inputs, and the scores for each indication and constraint, which are the outputs, will be referred to as model outputs. Thus, as shown in Figure 9, a relationship between the model input and the model output, i.e., a relational model 7, is generated and included in the calculation unit 111. This relational model 7 can be composed of, for example, a table as shown in Figure 9 (the blank spaces in the table will contain scores, although this is not shown in the illustration). The columns of this table consist of 816 indications and constraints, and the rows of the table consist of 173 crude drugs. Since Kampo herbal preparations are generated by combinations of crude drugs, the rows can also consist of 816 Kampo herbal preparations resulting from these combinations.

[0035] <3. Output of herbal medicine preparations> Next, the method for outputting Kampo herbal medicine preparations using the decision support device described above will be explained with reference to Figure 10. Figure 10 is a flowchart showing the output process for Kampo herbal medicine preparations.

[0036] First, the patient is interviewed, and the indications and limiting conditions are identified based on their symptoms (Step S1). The indications and limiting conditions identified at this time are selected from the 816 indications and limiting conditions described above. Next, the identified indications and limiting conditions are input into the decision support device 1 via the input device 5 (Step S2). The input indications and limiting conditions are then compared with the model output of the relational model 7 described above within the decision support device 1 (Step S3). Subsequently, based on the relational model 7, multiple herbal medicine preparations containing herbs corresponding to the selected indications and limiting conditions are selected. At this time, the comparison between the input indications and limiting conditions and the model output is performed using a known method such as cosine similarity. As a result, for example, as shown in Figure 11, the top four model outputs with high cosine similarity and the corresponding herbal medicine preparations are output along with their recommendation ranking and displayed on the display device 4 (Step S4). At this time, the higher the cosine similarity, the higher the recommendation ranking. Note that if the cosine similarity is close to 1, they are similar, and if it is close to 0, they are not similar. Furthermore, it is possible to exclude indications and restriction conditions included in the model output that have scores below a predetermined value (for example, less than 0.1). This prevents the recommendation of herbal medicines (traditional Chinese medicine preparations) corresponding to indications and restriction conditions with scores close to 0.

[0037] In the example above, four herbal medicine preparations were recommended based on their cosine similarity, but this number can be changed as needed. From these outputted herbal medicine preparations, the one to be prescribed to the patient is selected. Furthermore, the input information and the information regarding the outputted herbal medicine preparations can be stored in the memory unit 12 as history data 123.

[0038] <4. Features> As described above, the following effects can be obtained according to this embodiment. (1) As mentioned above, there are many herbal medicine preparations, but for example, a doctor who is not familiar with herbal medicine preparations can only use a very small number of them. In contrast, by using the decision support device 1 of this embodiment, even a doctor who is not familiar with herbal medicine preparations can suggest an appropriate herbal medicine preparation from the 442 herbal medicine preparations shown in this example, according to the patient's symptoms. Therefore, not only doctors who are not familiar with herbal medicine preparations, but also researchers of herbal medicine, pharmacists, etc. can use this decision support device 1 to suggest an herbal medicine preparation suitable for the patient.

[0039] (2) In this decision support device 1, the learning device 6 does not take Kampo herbal preparations as input, but rather takes Kampo prescription preparations and decomposes them into their constituent herbs as input. In other words, it learns the relationship between indications and constraints and herbs. This is because Kampo herbal preparations are composed of at least one herb, and each herb forms the core of the efficacy of the Kampo herbal preparation. Therefore, by performing learning using herbs with a high degree of correspondence to indications and constraints, it is possible to propose Kampo herbal preparations that are more appropriate to the patient's symptoms.

[0040] (3) The indications, limitations, and crude drugs used in the apparatus 1 of this embodiment are based on the information in the package inserts for medical Kampo preparations and the manufacturing and marketing approval standards for Kampo preparations for general use established by the Ministry of Health, Labour and Welfare. In other words, because official documents are used without using proprietary data, it is easier to create data and to output Kampo crude drug preparations with high accuracy compared to using data based on the subjective judgment of individual physicians (such as past diagnostic results or publications).

[0041] <5. Variation> While embodiments of the present invention have been described above, the above description is merely illustrative in every respect, and it goes without saying that various improvements and modifications can be made without departing from the scope of the present invention. For example, the following modifications are possible. Furthermore, the following modifications can be combined as appropriate.

[0042] <5-1> In the above embodiment, in order to output recommended herbal medicine preparations, 532 indications, 284 limiting conditions, and 173 herbs are specified from the package insert information for medical herbal medicine preparations and the manufacturing and marketing approval standards for general-use herbal medicine preparations for 442 herbal medicine preparations. However, it is not necessary to use all of these, and multiple of them can be used as needed. Furthermore, the herbs, indications, and limiting conditions are not extracted from a single package insert information for medical herbal medicine preparations and manufacturing and marketing approval standards for general-use herbal medicine preparations, but can also be extracted from, for example, multiple similar package insert information for medical herbal medicine preparations and manufacturing and marketing approval standards for general-use herbal medicine preparations.

[0043] Furthermore, the herbal medicine preparations, herbs, indications, and limiting conditions used in this decision support device can also be extracted from documents other than the package insert information for medical herbal medicine preparations and the manufacturing and marketing approval standards for over-the-counter herbal medicine preparations, such as documents issued by medical institutions involved in herbal medicine and classical texts related to herbal medicine.

[0044] <5-2> To produce more accurate Kampo herbal medicine preparations, it is also possible to select the outputted Kampo herbal medicine preparations. For example, physical strength conditions can be included in the constraints, and these conditions can be used to further select from the outputted Kampo herbal medicine preparations. For example, as physical strength conditions, the following can be extracted from the constraints in the package insert information for medical Kampo preparations and the manufacturing and sales approval standards for general-use Kampo preparations by reinterpreting them: "1. Weak physical strength," "2. Slightly weak physical strength," "3. Moderate physical strength," "4. Relatively strong physical strength," and "5. Full physical strength." In this example, 1 to 5 are considered to represent increasing physical strength. Then, as shown in Figure 12, when physical strength conditions are included in the constraints of the input information, if the physical strength conditions included in this input information do not match the physical strength conditions specified in the outputted Kampo herbal medicine preparations, the Kampo herbal medicine preparations that do not match can be excluded from the output.

[0045] The determination of whether or not a condition matches can be made, for example, if the input information includes "2. Slightly weak physical strength," then if the physical strength condition specified by the output herbal medicine preparation is not "2. Slightly weak physical strength," it can be excluded from the output. However, other judgments may also be made; for example, if the physical strength condition specified by the output herbal medicine preparation for the above input information is between 3 and 5, it can also be excluded from the output.

[0046] <5-3> The calculation unit 111 uses a learner 6 having a neural network as described above, but the configuration of the learner 6 is not particularly limited. It is not particularly limited as long as it can take crude drugs as input and output indications and limiting conditions. Therefore, the type of learner is not particularly limited, and in addition to neural networks, it can be a learner that learns using, for example, a support vector machine or a naive Bayes classifier.

[0047] <5-4> In the above embodiment, the learning device 6 and relational model 7 described above are used to take input information including limiting conditions and indications as input and output information including Kampo herbal medicine preparations that match this input information as output. However, it is also possible to use a calculation unit that uses machine learning or the like to take input information including limiting conditions and indications as input and output information including Kampo herbal medicine preparations that match this input information as output. Therefore, for example, it is also possible to use a calculation unit other than the one described above to configure a decision support device, decision support method, and decision support program related to the 6th to 11th viewpoints described above. [Explanation of Symbols]

[0048] 1 Decision support device 111 Calculation Unit 121 Decision Support Program 6 Learning devices

Claims

1. The system includes a calculation unit that takes input information including restrictive conditions and indications based on the diagnosis of traditional Chinese medicine as input, and outputs output information including traditional Chinese herbal medicine preparations that match the input information. The calculation unit described above, The system has a relational model that defines the relationship between the model input and the model output, which is generated based on a learner that has been trained to use the crude drugs constituting the Kampo herbal medicine preparation as model input, and to use limiting conditions based on the Kampo medicine diagnosis that are suitable for the model input, and information including the indications, as model output. Based on the aforementioned relationship model, the system is configured to output a corresponding model input from the model output that is highly related to the input information, and to output at least one herbal medicine preparation composed of the herbal medicines included in the output model input as output information. A device to assist in the selection of herbal medicine preparations.

2. The Kampo herbal medicine preparation determination support device according to claim 1, wherein the composition of the crude drugs in the Kampo herbal medicine preparation, the limiting conditions, and the indications are extracted from the package insert information for medical Kampo medicine preparations and the manufacturing and marketing approval standards for Kampo medicine preparations for general use.

3. The aforementioned input information restriction conditions include physical fitness conditions, The aforementioned physical condition related to the aforementioned herbal medicine preparation is defined, The calculation unit is configured to exclude a herbal medicine preparation from the output information when the physical condition included in the input information does not match the physical condition related to the herbal medicine preparation included in the output information, as described in claim 1, for the determination support device for herbal medicine preparations.

4. The system includes a step of inputting the input information to a calculation unit that takes input information including restrictive conditions and indications based on the diagnosis of traditional Chinese medicine as input, and outputs output information including traditional Chinese herbal medicine preparations that match the input information, thereby outputting the output information. The calculation unit described above, The system has a relational model that defines the relationship between the model input and the model output, which is generated based on a learner that has been trained to use the crude drugs constituting the Kampo herbal medicine preparation as model input, and to use limiting conditions based on the Kampo medicine diagnosis that are suitable for the model input, and information including the indications, as model output. Based on the aforementioned relationship model, the system is configured to output a corresponding model input from the model output that is highly related to the input information, and to output at least one herbal medicine preparation composed of the herbal medicines included in the output model input as output information. Methods for supporting the selection of herbal medicine preparations.

5. On the computer, A calculation unit takes input information including restrictive conditions and indications based on the diagnosis of traditional Chinese medicine as input, and outputs output information including herbal medicine preparations that match the input information. By inputting the input information, the calculation unit is made to execute the step of outputting the output information. The calculation unit described above, The system has a relational model that defines the relationship between the model input and the model output, which is generated based on a learner that has been trained to use the crude drugs constituting the Kampo herbal medicine preparation as model input, and to use limiting conditions based on the Kampo medicine diagnosis that are suitable for the model input, and information including the indications, as model output. Based on the aforementioned relationship model, the system is configured to output a corresponding model input from the model output that is highly related to the input information, and to output at least one herbal medicine preparation composed of the herbal medicines included in the output model input as output information. A program to support the selection of herbal medicine preparations.

Images