Determination device, determination method, and computer program

Abstract

To enable more efficient determination of a drug that may be approved as an approved drug.SOLUTION: A determination device comprises a control unit for determining the efficacy-related information of a drug to be determined by using: a determination model obtained by using multiple pieces of medication appearance information, which is information about the appearance of cells after administering existing approved drugs and discontinued drugs for a disease to be determined to a disease strain, which has been a cell affected by the disease to be determined, the determination model corresponding to the correlation between the medication appearance information and efficacy-related information, which is information about whether a drug will be an approved drug or a discontinued drug for the disease to be determined; and medication appearance information obtained by administering a drug to be determined for a disease to be determined to a disease strain of the disease to be determined.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a determination device, a determination method, and a computer program. [Background technology]

[0002] When obtaining approval for the manufacture of a drug, it is necessary to confirm the drug's efficacy, safety, etc. through testing in several phases and clear all of these tests. Only after the drug's efficacy and safety have been confirmed in all phases of testing can the drug be marketed as an approved drug. In such tests, for example, human genetic disease models using cells derived from human iPS cells, as described in Patent Document 1, are used. However, such tests still require a great deal of time and effort, and there has been a need for a technology that can more efficiently determine which drugs have the potential to be approved as drugs. Summary of the Invention [Problem to be solved by the invention]

[0003] The present invention has been made in consideration of the above-mentioned circumstances, and aims to provide a technology that enables more efficient determination of pharmaceuticals that have the potential to be approved as approved drugs. [Means for solving the problem]

[0004] One aspect of the present invention is a determination device that includes a control unit that determines the efficacy-related information of a drug to be determined by using a determination model obtained by using multiple medication appearance information, which is information about the appearance of each cell after administering existing approved drugs and discontinued drugs for a disease to be determined to a disease strain that is a cell affected by the disease, and efficacy-related information, which is information about whether a drug will be an approved drug or a discontinued drug for the disease to be determined, and the medication appearance information obtained by administering the drug to be determined for the disease to a disease strain of the disease to be determined. [Effects of the Invention]

[0005] The present invention makes it possible to more efficiently and simply and effectively determine which drugs have the potential to be approved as approved drugs. [Brief explanation of the drawings]

[0006] [Figure 1] FIG. 10 is a diagram showing a first specific example of medication appearance information used to generate a determination model. [Figure 2] FIG. 10 is a diagram showing a second specific example of medication appearance information used to generate a determination model. [Figure 3] FIG. 10 is a diagram illustrating a first specific example of a determination model. [Figure 4] FIG. 10 is a diagram illustrating a second specific example of a determination model. [Figure 5] FIG. 1 is a diagram illustrating an outline of a cluster. [Figure 6] FIG. 10 is a diagram illustrating a third specific example of a determination model. [Figure 7] FIG. 10 is a diagram showing an outline of a determination process using a first disease determination model generated using a first specific example of medication appearance information. [Figure 8] FIG. 10 is a diagram showing an outline of a determination process using a first disease determination model generated using a second specific example of medication appearance information. [Figure 9] 1 is a schematic block diagram showing the system configuration of a determination system 100 according to the present invention. [Figure 10] 2 is a schematic block diagram showing a specific example of the functional configuration of the terminal device 10. FIG. [Figure 11] 2 is a schematic block diagram showing a specific example of the functional configuration of the learning device 20. FIG. [Figure 12] 10 is a flowchart showing a specific example of processing by the learning device 20. [Figure 13] 2 is a schematic block diagram showing a specific example of the functional configuration of a determination device 30. FIG. [Figure 14] 10 is a flowchart showing a specific example of processing by the determination device 30. [Figure 15]FIG. 2 is a diagram illustrating an outline of an example of the hardware configuration of an information processing device 90 applied to the present embodiment. [Figure 16] FIG. 10 is a diagram showing a modified example of the determination device 30. [Figure 17] FIG. 10 is a diagram showing a modified example of the determination device 30. DETAILED DESCRIPTION OF THE INVENTION

[0007] [principle] The principle of the present invention will be described below. A drug that was previously approved during the drug manufacturing approval process (hereinafter referred to as an "approved drug") and a drug that was dropped without approval during that process (hereinafter referred to as a "dropped drug") are each administered to cells. Information regarding the appearance of cells after administration of the approved drug and the dropped drug (hereinafter referred to as "drug appearance information") is obtained for each. A model (hereinafter referred to as a "determination model") for distinguishing between the drug appearance information for the approved drug and the drug appearance information for the dropped drug is generated based on the difference between them. The drug to be evaluated (hereinafter referred to as a "target drug") is administered to cells to obtain drug appearance information regarding the target drug. Information regarding whether the target drug is effective as a drug for a target disease (hereinafter referred to as "effectiveness-related information") is determined using the drug appearance information of the target drug and the determination model. The effectiveness of an approved drug for a target disease is confirmed during the actual approval process. Therefore, by using a determination model generated using medication appearance information obtained by administering an approved drug, it is possible to determine the efficacy-related information described above. Such determination may be performed by a person or by an information processing device. When performed by an information processing device, the determination model may be generated by a learning process using training data, or may be generated by a method other than the learning process (such as a statistical method).

[0008] FIG. 1 is a diagram showing a first specific example of medication appearance information used to generate a determination model. In this first specific example, a determination model is generated using medication appearance information obtained by administering an approved drug to a disease strain and medication appearance information obtained by administering a dropout drug to a disease strain. The disease strain may be cells obtained from a patient suffering from a specific disease, or may be cells that mimic a state of suffering from a specific disease. The cells that mimic a disease state may be, for example, disease model cells created by genetic manipulation.

[0009] The medication appearance information may be, for example, an image obtained by photographing the appearance of a cell. Such an image may be any type of image. For example, it may be an image obtained by photographing a visible light image without particularly staining the cell tissue. It may be an image obtained by staining the cell tissue and photographing light of a wavelength corresponding to the staining type. It may be an image obtained as a phase contrast image. Both an image of the disease strain before medication and an image of the disease strain after medication may be used as medication appearance information, or the image of the disease strain before medication may not be used and only the image of the disease strain after medication may be used as medication appearance information.

[0010] Instead of the image itself, a feature obtained from the image (hereinafter referred to as "appearance feature") may be used as medication appearance information. For example, a feature indicating the outline of a cell part, a feature indicating the shape of a cell part, a frequency component of the outline of a cell part, a feature related to the color of the region of a cell part, a feature indicating the size of the area of ​​a cell part, etc. may be used as the appearance feature.

[0011] A feature (hereinafter referred to as a "difference feature") indicating the difference between an image of a cell before administration (hereinafter referred to as a "pre-administration image") and an image of the cell after administration (hereinafter referred to as a "post-administration image") may be used as medication appearance information. The difference feature may be, for example, information indicating a difference in the outline of a cell portion, information indicating a difference in the shape of a cell portion, or information indicating a difference in the area size of a cell portion, or may be configured as other information.

[0012] The target site for medication appearance information may be the entire cell of the disease strain, or a specific part (such as a cell membrane or organelle). Specific examples of organelles include the nucleus, nucleolus, and mitochondria. The target cells include living cells. More specifically, mammalian cells, and even more specifically, primate cells. In a preferred embodiment, the target cells are human cells. Any living cell may be used, including, for example, nerve cells and glial cells (such as astrocytes). For example, iPSC-derived cells may be used as such cells.

[0013] The approved drug and dropped drug administered to the disease strain may be approved drugs and dropped drugs in a manufacturing approval for the disease that the disease strain is suffering from, or may be approved drugs and dropped drugs in a manufacturing approval for a disease other than the disease that the disease strain is suffering from. The disease that the cells are suffering from may be, for example, a neurological disease such as Parkinson's disease, or may be another disease. At least two pieces of medication profile information obtained in this way are used as one set of medication profile information in the first embodiment.

[0014] FIG. 2 shows a second specific example of medication profile information used to generate a discrimination model. In the second specific example, a discrimination model is generated using medication profile information obtained by administering an approved drug to a diseased strain and medication profile information obtained by administering a dropout drug to a diseased strain, as well as medication profile information obtained by administering an approved drug to a healthy strain and medication profile information obtained by administering a dropout drug to a healthy strain. The healthy strain may be cells obtained from a healthy individual (at least one not suffering from a specific disease) or iPSC-derived cells in a healthy state. The medication profile information obtained by administering an approved drug and a dropout drug to a healthy strain may be obtained in the same manner as the medication profile information regarding the diseased strain in the first specific example described above. Note that in the second specific example, it is preferable that the approved drug administered to the healthy strain and the approved drug administered to the diseased strain are the same approved drug. Similarly, it is preferable that the dropout drug administered to the healthy strain and the dropout drug administered to the diseased strain in the second specific example are the same dropout drug. At least four pieces of medication appearance information thus obtained are used as one set of medication appearance information in the second embodiment.

[0015] Next, a determination model generated using medication appearance information will be described. FIG. 3 is a diagram showing a first specific example of a determination model. The determination model of the first specific example is generated as a disease-specific determination model for each disease. For example, a determination model for a first disease is used to determine efficacy-related information related to the first disease. That is, it is used to determine whether or not manufacturing approval can be obtained when a drug to be determined (hereinafter referred to as a "drug to be determined") is used for a disease to be determined (hereinafter referred to as a "disease to be determined"). Therefore, in FIG. 3 showing the first specific example, the disease to be determined is the first disease, and such a determination model is shown as a determination model for the first disease. Similarly, when a second disease different from the first disease is set as the disease to be determined, a determination model used to determine efficacy-related information is shown as a determination model for the second disease.

[0016] Each disease determination model is generated using the disease strain of the target disease (disease to be determined) and medication appearance information obtained by administering an approved drug and a discontinued drug for the target disease to the disease strain. For example, to generate a first disease determination model, at least medication appearance information when an approved drug for the first disease is administered to a disease strain of the first disease and medication appearance information when a discontinued drug for the first disease is administered to a disease strain of the first disease are used.

[0017] According to this first specific example, a disease-specific determination model can be obtained for a certain disease (disease to be determined), which makes it possible to determine efficacy-related information with higher accuracy.

[0018] The first disease determination model may be generated by a learning process using multiple training data. In this case, for example, multiple pieces of medication appearance information obtained as a result of administering an approved drug for the first disease to a disease strain of the first disease (disease to be determined) are obtained. In this case, multiple pieces of medication appearance information may be obtained by administering multiple types of approved drugs for the disease to be determined to multiple disease strains prepared in different patients or by different processes. A predetermined learning process is performed using the multiple pieces of medication appearance information obtained in this way as training data. As a result, a trained model is obtained. The obtained trained model may be used as the first disease determination model.

[0019] FIG. 4 is a diagram showing a second specific example of a judgment model. The judgment model of the second specific example is generated as a judgment model dedicated to each cluster, for each group (cluster) of similar diseases. For example, a judgment model for a first cluster is used to judge efficacy-related information related to diseases belonging to the first cluster. Therefore, in FIG. 4 showing the second specific example, such a judgment model is shown as a judgment model for the first cluster. Similarly, a judgment model used to judge efficacy-related information related to diseases belonging to a second cluster different from the first cluster is shown as a judgment model for the second cluster.

[0020] The classification model for each cluster is generated using the disease strain of the disease belonging to that cluster and medication appearance information obtained by administering approved drugs and discontinued drugs for the disease targeted for that disease strain. For example, if at least a first disease and a second disease belong to a first cluster, medication appearance information for the first disease and medication appearance information for the second disease are used to generate the classification model for the first cluster.

[0021] FIG. 5 is a diagram showing an outline of clusters. Classification of each disease into clusters may be performed in any manner. For example, classification into clusters may be performed according to existing disease classifications. For example, classification may be performed based on the appearance of the disease strain, such that diseases with similar appearance characteristics are classified into the same cluster. For example, classification may be performed based on medication appearance information, such that diseases with similar medication appearance information are classified into the same cluster. Naturally, such medication appearance information will be compared between medication appearance information for different approved drugs or different withdrawn drugs for different diseases, but even in such cases, if a certain degree of similarity is recognized, the diseases may be classified into the same category.

[0022] In such classification using the appearance of disease strains and medication appearance information, classification may be performed by executing a clustering algorithm (e.g., k-means) using one or more feature amounts obtained from the appearance and medication appearance information. For example, in the example of FIG. 5, three clusters, a first cluster, a second cluster, and a third cluster, can be defined by defining boundaries to divide the region into three areas based on two feature amounts (first feature amount and second feature amount). Diseases are classified into multiple clusters in advance in this way, and a determination model for each cluster is generated by using multiple pieces of medication appearance information for diseases belonging to each cluster.

[0023] According to this second specific example, even if there is not much medication appearance information available for a certain disease, or if there is no approved drug for the disease, it is possible to determine efficacy-related information by using a determination model generated using medication appearance information for other similar diseases that are also classified into a raster.

[0024] The first cluster discrimination model may be generated by a learning process using multiple training data. In this case, for example, multiple pieces of medication profile information obtained as a result of administering an approved drug for the first disease to a disease strain of the first disease are obtained. Similarly, multiple pieces of medication profile information for disease strains of other diseases (e.g., a second disease) classified in the same cluster are also obtained. In this case, multiple pieces of medication profile information may be obtained by administering multiple types of approved drugs to multiple disease strains prepared in different patients or by different processes. A predetermined learning process is performed using the medication profile information for multiple diseases belonging to the same cluster obtained in this way as training data. As a result, a trained model is obtained. The obtained trained model may be used as the first cluster discrimination model.

[0025] Fig. 6 is a diagram showing a third specific example of a judgment model. The judgment model of the third specific example is generated as a common judgment model for unspecified types of diseases, regardless of the type or cluster of the disease. Therefore, in Fig. 6 showing the third specific example, such a judgment model is shown as a common disease judgment model. The common disease judgment model is generated using medication appearance information obtained for multiple types of diseases.

[0026] 7 is a diagram showing an outline of the judgment process by the first disease judgment model generated using the first specific example of medication appearance information. As described above, the judgment model using the first specific example of medication appearance information is a judgment model in which medication appearance information due to administration to a disease strain is used as an explanatory variable and efficacy relationship information is used as a response variable. Therefore, when using the first disease judgment model generated as such a judgment model, medication appearance information is obtained by administering the drug to be judged (hereinafter referred to as "target drug") to the disease strain of the first disease, and efficacy relationship information, which serves as a response variable, is obtained by providing this as an explanatory variable to the first disease judgment model.

[0027] 8 is a diagram showing an outline of the judgment process by the first disease judgment model generated using the second specific example of medication appearance information. As described above, the judgment model using the second specific example of medication appearance information is a judgment model in which medication appearance information due to medication to each of the healthy strain and the diseased strain is used as an explanatory variable and efficacy relationship information is used as a response variable. Therefore, when using the first disease judgment model generated as such a judgment model, medication appearance information is obtained by administering the target drug to each of the diseased strain of the first disease and the healthy strain, and the medication appearance information is provided as an explanatory variable to the first disease judgment model, thereby obtaining efficacy relationship information as a response variable.

[0028] In the specific examples shown in Figures 7 and 8, the first disease judgment model is used as an example of the judgment model used. However, in the specific examples shown in Figures 7 and 8, a cluster judgment model may be used instead of the first disease judgment model, or a disease common judgment model may be used. In other words, the type of judgment model used in the explanation of Figures 7 and 8 may be any of the specific examples explained using Figures 3 to 6.

[0029] In Fig. 7, as long as a disease strain is used instead of a healthy strain in generating explanatory variables (medication appearance information) using the drug to be determined, any of the specific examples in Fig. 3 to Fig. 6 may be applied as the type of determination model. However, as training data used to generate the determination model, it is desirable to use medication appearance information using a disease strain instead of a healthy strain, as in Fig. 7.

[0030] In Fig. 8, as long as healthy and diseased strains are used in generating explanatory variables (medication appearance information) using the drug to be determined, any of the specific examples in Fig. 3 to Fig. 6 may be applied as the type of determination model. However, as training data used in generating the determination model, it is desirable to use medication appearance information using healthy and diseased strains, as in Fig. 8.

[0031] [system] A determination system for determining efficacy-related information by an information processing device using a determination model will be described below. Fig. 9 is a schematic block diagram showing the system configuration of a determination system 100 of the present invention. The determination system 100 is used to determine efficacy-related information of a drug to be determined based on medication appearance information of the drug to be determined.

[0032] The determination system 100 includes a terminal device 10, a learning device 20, and a determination device 30. The terminal device 10 and the determination device 30 are communicatively connected via a network 70. The learning device 20 and the determination device 30 may also be communicatively connected via the network 70. The network 70 may be a network using wireless communication or a network using wired communication. The network 70 may be configured using, for example, the Internet or a local area network (LAN). The network 70 may also be configured by combining multiple networks.

[0033] 10 is a schematic block diagram showing a specific example of the functional configuration of the terminal device 10. The terminal device 10 is configured using information devices such as a smartphone, tablet, personal computer, dedicated device, etc. The terminal device 10 includes a communication unit 11, an input unit 12, an output unit 13, a medication appearance information acquisition unit 14, a storage unit 15, and a control unit 16.

[0034] The communication unit 11 is a communication device. The communication unit 11 may be configured as, for example, a network interface. The communication unit 11 communicates data with other devices via the network 70 in accordance with the control of the control unit 16. The communication unit 11 may be a device that performs wireless communication or a device that performs wired communication.

[0035] The input unit 12 is configured using existing input devices such as a keyboard, a pointing device (mouse, tablet, etc.), buttons, a touch panel, etc. The input unit 12 is operated by a user when inputting user instructions to the terminal device 10. The input unit 12 may be an interface for connecting the input device to the terminal device 10. In this case, the input unit 12 inputs an input signal generated by the input device in response to a user input to the terminal device 10. The input unit 12 may be configured using a microphone and a voice recognition device. In this case, the input unit 12 acquires an acoustic signal generated by the user's speech, performs voice recognition on the words spoken by the user, and inputs character string information of the recognition result to the terminal device 10. The voice recognition processing may be performed by the control unit 16. The input unit 12 may be configured in any way as long as it is capable of inputting user instructions to the terminal device 10. The user may operate the input unit 12 to input information indicating, for example, a disease for which a target drug is to be determined (a target disease to be determined) to the terminal device 10. The information indicating the target disease to be determined may be transmitted to the determination device 30, for example, and used when selecting a determination model to be used in the determination process.

[0036] The output unit 13 outputs information in a form that can be recognized by the user. The output unit 13 may be, for example, an image display device such as a liquid crystal display or an organic EL (Electro Luminescence) display. The output unit 13 may be an interface for connecting an image display device to the terminal device 10. In this case, the output unit 13 generates a video signal for displaying image data and outputs the video signal to the image display device connected to the output unit 13. The output unit 13 may be a device for outputting sound, such as a speaker. The output unit 13 may be an interface for connecting an audio output device, such as a speaker or headphones, to the terminal device 10. In this case, the output unit 13 generates an audio signal for reproducing audio data and outputs the audio signal to the audio output device connected to the output unit 13. The output unit 13 may be configured as a touch panel integrated with the input unit 12.

[0037] The medication appearance information acquisition unit 14 accepts medication appearance information data of the drug to be assessed, which is input to the terminal device 10. The medication appearance information acquisition unit 14 may read medication appearance information data recorded on a recording medium such as a CD-ROM or a USB memory (Universal Serial Bus Memory). The medication appearance information acquisition unit 14 may acquire images of healthy and diseased strains captured by a still camera or video camera from the camera. When images are acquired from a camera, wired communication via a communication cable such as a USB cable or a LAN cable, or wireless communication such as a wireless LAN or Bluetooth, may be performed. The medication appearance information acquisition unit 14 may be configured as an imaging device such as a still camera or a video camera. The medication appearance information acquisition unit 14 may receive image data from another information processing device via a network. The medication appearance information data input by the medication appearance information acquisition unit 14 may be stored in the storage unit 15.

[0038] The storage unit 15 is configured using a storage device such as a magnetic hard disk drive or a semiconductor storage device. The storage unit 15 stores data used by the control unit 16. The storage unit 15 stores data required when the control unit 16 performs processing.

[0039] The control unit 16 is configured using a processor such as a CPU (Central Processing Unit) and a memory (main storage device). The control unit 16 functions when the processor executes a program. Note that all or part of the functions of the control unit 16 may be realized using hardware such as an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array). The program may be recorded on a computer-readable recording medium. Examples of computer-readable recording media include portable media such as flexible disks, magneto-optical disks, ROMs, CD-ROMs, and semiconductor storage devices (e.g., SSDs: Solid State Drives), as well as storage devices such as hard disks and semiconductor storage devices built into computer systems. The program may be transmitted via a telecommunications line.

[0040] The control unit 16 may execute, for example, an application installed on its own device (terminal device 10). A specific example of such an application is an application provided to the terminal device 10 as a dedicated application for the determination system 100. Another specific example of such an application is a web browser application. Such an application may be pre-installed on the terminal device 10, or may be downloaded each time a determination process is executed. For example, when implemented as a web browser application, the terminal device 10 may download and execute the application from a device specified by a specific web server (for example, the web server itself or another server) in response to the terminal device 10 connecting to the web server. The control unit 16 operates according to the program of the application being executed.

[0041] The control unit 16 controls the terminal device 10 in response to user operations and information received from the determination device 30. For example, the control unit 16 transmits medication appearance information selected by the user operating the input unit 12 and medication appearance information acquired by the medication appearance information acquisition unit 14 to the determination device 30 using the communication unit 11. For example, when information transmitted from the determination device 30 is received by the communication unit 11 via the network 70, the control unit 16 generates screen data based on the received information and causes the output unit 13 to display the screen data. Such screen data includes images and text indicating the information transmitted from the determination device 30. For example, when information transmitted from the determination device 30 is received by the communication unit 11 via the network 70, the control unit 16 generates audio data based on the received information and causes the output unit 13 to output the audio data.

[0042] 11 is a schematic block diagram showing a specific example of the functional configuration of learning device 20. Learning device 20 is configured using an information processing device such as a personal computer or a server device. Learning device 20 includes a communication unit 21, a storage unit 22, and a control unit 23.

[0043] The communication unit 21 is a communication device. The communication unit 21 may be configured as, for example, a network interface. The communication unit 21 communicates data with other devices via the network 70 in accordance with the control of the control unit 23. The communication unit 21 may be a device that performs wireless communication or a device that performs wired communication.

[0044] The storage unit 22 is configured using a storage device such as a magnetic hard disk drive or a semiconductor storage device. The storage unit 22 stores data used by the control unit 23. The storage unit 22 may function as, for example, a teacher data storage unit 221 and a trained model storage unit 222.

[0045] The training data storage unit 221 stores training data used in the learning process executed by the learning device 20. The training data stored in the training data storage unit 221 is a combination of medication appearance information obtained by administering at least an existing approved drug and a dropped drug to a disease strain, and label information indicating whether the medication appearance information was obtained for an approved drug or a dropped drug. The medication appearance information used in the training data may include not only medication appearance information obtained by administering a drug to a disease strain as described above, but also medication appearance information obtained by administering a drug to a healthy strain. Furthermore, medication appearance information related to only a specific disease may be used as training data. In this case, a trained model dedicated to that specific disease is generated. Furthermore, medication appearance information related to only diseases belonging to a specific cluster may be used as training data. In this case, a trained model dedicated to the diseases belonging to that specific cluster is generated. Furthermore, medication appearance information related to unspecified diseases may be used as training data. In this case, a trained model common to all diseases is generated.

[0046] The trained model storage unit 222 stores a trained model obtained by a learning process using the training data stored in the training data storage unit 221.

[0047] The control unit 23 is configured using a processor such as a CPU and a memory. The control unit 23 functions as an information control unit 231 and a learning control unit 232 by the processor executing a program. All or part of the functions of the control unit 23 may be realized using hardware such as an ASIC, PLD, or FPGA. The above program may be recorded on a computer-readable recording medium. Examples of computer-readable recording media include portable media such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, and a semiconductor storage device (e.g., an SSD), as well as storage devices such as a hard disk or semiconductor storage device built into a computer system. The above program may be transmitted via a telecommunications line.

[0048] The information control unit 231 controls the input and output of information. For example, the information control unit 231 acquires training data from other devices (information processing devices or storage media) and records the training data in the training data storage unit 221. For example, the information control unit 231 transmits the trained model stored in the trained model storage unit 222 to another device (for example, the determination device 30).

[0049] The learning control unit 232 executes a learning process using the training data stored in the training data storage unit 221. Specific examples of such learning processes include supervised learning for classification, such as support vector machines, random forests, and neural networks. The learning control unit 232 generates a trained model for outputting efficacy-related information based on input medication appearance information, for example, by performing supervised learning. The learning control unit 232 records the generated trained model in the trained model storage unit 222. The trained model obtained by the learning control unit 232 may be transmitted to the determination device 30 and recorded in the determination model storage unit 321 of the determination device 30. Such a trained model can obtain efficacy-related information of the drug to be determined as an output by providing medication appearance information of the drug to be determined as an input.

[0050] 12 is a flowchart showing a specific example of processing by the learning device 20. First, the information control unit 231 acquires training data (step S101). The training data may be input by a user, acquired via communication from another information device, or acquired from a recording medium connected to the learning device 20, for example. The learning control unit 232 executes a learning process using the training data and records a trained model in the trained model storage unit 222 (step S102).

[0051] 13 is a schematic block diagram showing a specific example of the functional configuration of the determination device 30. The determination device 30 is configured using an information processing device such as a personal computer or a server device. The determination device 30 includes a communication unit 31, a storage unit 32, and a control unit 33.

[0052] The communication unit 31 is a communication device. The communication unit 31 may be configured as, for example, a network interface. The communication unit 31 communicates data with other devices via the network 70 in accordance with the control of the control unit 33. The communication unit 31 may be a device that performs wireless communication or a device that performs wired communication.

[0053] The storage unit 32 is configured using a storage device such as a magnetic hard disk drive or a semiconductor storage device. The storage unit 32 stores data used by the control unit 33. The storage unit 32 may function as a determination model storage unit 321, for example.

[0054] The judgment model storage unit 321 stores a judgment model used by the judgment unit 332 when performing the judgment process. The judgment model may be configured using information of a trained model generated in advance by a learning process, for example. Such a learning process may be executed by, for example, another device (e.g., the learning device 20) or by the device itself (the judgment device 30). The judgment model does not necessarily have to be generated by a learning process. The judgment model may be configured using, for example, a lookup table that associates medication appearance information with efficacy-related information, or may be configured in another manner.

[0055] The control unit 33 is configured using a processor such as a CPU and a memory. The control unit 33 functions as an information control unit 331 and a determination unit 332 by the processor executing a program. All or part of the functions of the control unit 33 may be realized using hardware such as an ASIC, PLD, or FPGA. The above program may be recorded on a computer-readable recording medium. Examples of computer-readable recording media include portable media such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, and a semiconductor storage device (e.g., an SSD), as well as storage devices such as a hard disk or semiconductor storage device built into a computer system. The above program may be transmitted via a telecommunications line.

[0056] The information control unit 331 acquires medication appearance information of the drug to be determined from another device such as the terminal device 10. The information control unit 331 transmits information indicating the determination result obtained by the determination unit 332 to another device such as the terminal device 10. Such exchange of information between the information control unit 331 and another device may be performed by communication using the communication unit 31, for example.

[0057] The determination unit 332 performs a determination process using the determination model stored in the determination model storage unit 321 and the medication appearance information of the drug to be determined. Through the determination process, the effectiveness-related information of the drug to be determined is determined.

[0058] 14 is a flowchart showing a specific example of the processing of the determination device 30. First, the information control unit 331 acquires medication appearance information of the drug to be determined from the terminal device 10 (step S201). The determination unit 332 performs a determination process using at least the medication appearance information of the drug to be determined, and acquires efficacy-related information of the drug to be determined (step S202). The determination unit 332 transmits efficacy-related information indicating the determination result to the terminal device 10 (step S203).

[0059] The determination system 100 configured in this manner can more accurately determine the efficacy-related information of a target drug by using the medication appearance information of the target drug. Specifically, the determination system 100 performs a learning process using the medication appearance information of multiple existing approved drugs and discontinued drugs to obtain a trained model. Using the medication appearance information of existing approved drugs and discontinued drugs, it is possible to learn the information that appears in the appearance of the effects of approved drugs and discontinued drugs on cells. Using a trained model obtained based on such information, efficacy-related information, which is the objective variable when the medication appearance information of the target drug is used as the explanatory variable, is estimated. This makes it possible to accurately predict whether a target drug will be approved or discontinued based on the appearance characteristics that occur in a disease strain due to medication. In other words, there is a certain correlation between whether a drug will be approved or discontinued in its manufacturing approval and the appearance characteristics that occur in the disease strain when the drug is administered to the disease strain. Therefore, by learning the medication appearance information of the target drug as the explanatory variable, it is possible to accurately estimate the information indicating whether the drug will be approved (efficacy-related information), which is the objective variable.

[0060] Furthermore, the determination system 100 may be used to determine whether an approved drug (hereinafter referred to as an "approved drug") that has already been approved for a certain disease can be applied to other diseases (hereinafter referred to as "other diseases"). For example, the approved drug is administered to a strain of another disease for which approval has not been obtained as the drug to be determined, and medication appearance information obtained thereby is acquired. The terminal device 10 may transmit the medication appearance information acquired in this manner to the determination device 30, thereby acquiring efficacy-related information when the other disease is the target disease to be determined as the determination result. By performing such a determination, it becomes possible to more easily determine whether an approved drug can be applied to other diseases.

[0061] FIG. 15 is a diagram illustrating an outline of an example of the hardware configuration of an information processing device 90 applied to this embodiment. The information processing device 90 includes a processor 91, a main memory device 92, a communication interface 93, an auxiliary memory device 94, an input / output interface 95, and an internal bus 96. The processor 91, the main memory device 92, the communication interface 93, the auxiliary memory device 94, and the input / output interface 95 are communicably connected to each other via the internal bus 96. The information processing device 90 may be applied to, for example, the learning device 20 and the determination device 30. In this case, for example, the communication units 21 and 31 may be configured using the communication interface 93. For example, the memory units 22 and 32 may be configured using the auxiliary memory device 94. Furthermore, the control units 23 and 33 may be configured using the processor 91 and the main memory device 92.

[0062] (Variation) The determination unit 332 of the determination device 30 may be configured to determine the cluster to which the disease for which the target drug is used belongs and to perform determination processing using a determination model for the determined cluster. In this case, information used to determine the cluster to which the disease for which the target drug is used belongs needs to be acquired from the terminal device 10 or the like. In this case, for example, a user of the terminal device 10 may operate the input unit 12 to input direct information indicating the cluster to the terminal device 10, and the terminal device 10 may transmit the information to the determination device 30. Alternatively, an image of a disease strain for which the target drug is used before administration may be input to the terminal device 10, and the terminal device 10 may transmit the image data to the determination device 30. In this case, the determination unit 332 of the determination device 30 may determine the cluster to which the disease belongs based on feature quantities obtained from the image data of the disease strain. In this case, information serving as a determination criterion is pre-stored in the storage unit 32.

[0063] In this embodiment, the terminal device 10 and the determination device 30 are configured as separate devices, but they may also be configured as an integrated device. FIG. 16 is a diagram showing a modified example of the determination device 30 configured in this manner. The determination device 30 shown in FIG. 16 includes an input unit 34, an output unit 35, and a medication appearance information acquisition unit 36. The input unit 34, the output unit 35, and the medication appearance information acquisition unit 36 ​​of the determination device 30 shown in FIG. 16 function in the same manner as the input unit 12, the output unit 13, and the medication appearance information acquisition unit 14 of the terminal device 10, respectively. The control unit 33 operates in response to an operation on the input unit 34, performs a determination process using the medication appearance information acquired by the medication appearance information acquisition unit 36, and outputs information using the output unit 35.

[0064] In this embodiment, the learning device 20 and the determination device 30 are configured as separate devices, but they may also be configured as an integrated device. FIG. 17 is a diagram showing a modified example of the determination device 30 configured in this manner. The memory unit 32 of the determination device 30 shown in FIG. 17 also functions as a teacher data memory unit 322. The control unit 33 of the determination device 30 shown in FIG. 17 also functions as a learning control unit 333. The teacher data memory unit 322 functions in the same way as the teacher data memory unit 221 of the learning device 20. The learning control unit 333 functions in the same way as the learning control unit 232 of the learning device 20.

[0065] The learning device 20 may be implemented using a plurality of information processing devices. For example, the learning device 20 may be implemented using a device such as a cloud. For example, in the learning device 20, the memory unit 22 and the control unit 23 may each be implemented in different information processing devices. For example, the memory unit 22 of the learning device 20 may be distributed and implemented across a plurality of information processing devices. The determination device 30 may be implemented using a plurality of information processing devices. For example, the determination device 30 may be implemented using a device such as a cloud. For example, in the determination device 30, the memory unit 32 and the control unit 33 may each be implemented in different information processing devices. For example, the memory unit 32 of the determination device 30 may be distributed and implemented across a plurality of information processing devices.

[0066] The application running on the terminal device 10 may be an application that performs other processing using the determination result of the validity-related information and provides the user with information obtained by the processing. Such an application may be, for example, an application that performs processing using an API (Programming Interface) provided by the determination device 30. In this case, the output unit 13 may output to the terminal device 10 other information obtained by processing using the validity-related information, instead of the information itself (validity-related information) obtained from the determination device 30.

[0067] Although an embodiment of the present invention has been described above in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and includes designs within the scope of the gist of the present invention.

[0068] The present invention includes the following aspects. [1] A determination device comprising: a control unit that determines the efficacy-related information of a drug to be determined by using a determination model obtained by using multiple medication appearance information, which is information regarding the appearance of each cell after administering existing approved drugs and discontinued drugs for a disease to be determined to a disease strain, which is a cell affected by the disease to be determined, and a determination model that corresponds to the correlation between efficacy-related information, which is information regarding whether a drug will be an approved drug or a discontinued drug for the disease to be determined, and the medication appearance information, and the medication appearance information obtained by administering a drug to be determined for the disease to be determined to a disease strain of the disease to be determined. [2] The medication appearance information further includes information on the appearance of each cell after administering the existing approved drug and the dropped drug to a healthy cell line, which is a healthy cell line; The control unit determines the efficacy-related information of the drug to be evaluated by further using medication appearance information obtained by administering the drug to be evaluated to the healthy strain. [3] A step of obtaining a determination model according to the correlation between efficacy-related information, which is information on whether a drug will be an approved drug or a discontinued drug for the target disease, and the medication appearance information, by using multiple medication appearance information, which is information on the appearance of each cell after administering existing approved drugs and discontinued drugs for the target disease to a disease strain, which is a cell affected by the target disease; A step of administering a drug to be determined for a disease to be determined to a disease strain of the disease to be determined and acquiring medication appearance information; determining efficacy-related information of the drug to be evaluated by using the medication appearance information of the drug to be evaluated and the evaluation model; A determination method having the following. [4] A computer program for causing a computer to function as a determination device having a control unit that determines the efficacy-related information of a drug to be determined by using a determination model obtained by using multiple medication appearance information, which is information about the appearance of each cell after administering existing approved drugs and discontinued drugs for the disease to be determined to a disease strain, which is a cell affected by the disease to be determined, and a determination model that corresponds to the correlation between efficacy-related information, which is information about whether a drug will be an approved drug or a discontinued drug for the disease to be determined, and the medication appearance information, and medication appearance information obtained by administering a drug to be determined for the disease to be determined to a disease strain of the disease to be determined. [Explanation of symbols]

[0069] 100...Determination system, 10...Terminal device, 11...Communication unit, 12...Input unit, 13...Output unit, 14...Medication appearance information acquisition unit, 15...Memory unit, 16...Control unit, 20...Learning device, 21...Communication unit, 22...Memory unit, 221...Teacher data memory unit, 222...Learned model memory unit, 23...Control unit, 231...Information control unit, 232...Learning control unit, 30...Determination device, 31...Communication unit, 32...Memory unit, 321...Determination model memory unit, 33...Control unit, 331...Information control unit, 332...Determination unit [Prior art documents] [Patent documents]

[0070] [Patent Document 1] Japanese Patent Application Publication No. 2023-140198

Claims

1. A determination device comprising: a control unit that determines the efficacy-related information of a drug to be determined by using a determination model obtained by using multiple medication appearance information, which is information regarding the appearance of each cell after administering existing approved drugs and discontinued drugs for a disease to be determined to a disease strain, which is a cell affected by the disease to be determined, and efficacy-related information, which is information regarding whether a drug will be an approved drug or a discontinued drug for the disease to be determined, and the medication appearance information obtained by administering the drug to be determined for the disease to be determined to a disease strain of the disease to be determined.

2. The medication appearance information further includes information on the appearance of each cell after administering the existing approved drug and the dropped drug to a healthy cell line, which is a healthy cell line; The determination device according to claim 1 , wherein the control unit determines the efficacy-related information of the drug to be determined by further using medication appearance information obtained by administering the drug to the healthy strain.

3. a step of obtaining a determination model according to the correlation between efficacy-related information, which is information on whether a drug will be an approved drug or a discontinued drug for the target disease, and the medication appearance information, by using a plurality of medication appearance information, which is information on the appearance of each cell after administering existing approved drugs and discontinued drugs for the target disease to a disease strain, which is a cell affected by the target disease; A step of administering a drug to be determined for a disease to be determined to a disease strain of the disease to be determined and acquiring medication appearance information; determining efficacy-related information of the drug to be evaluated by using the medication appearance information of the drug to be evaluated and the evaluation model; A determination method having the following.

4. A computer program for causing a computer to function as a determination device that includes a control unit that determines the efficacy-related information of a drug to be determined by using a determination model obtained by using multiple medication appearance information, which is information about the appearance of each cell after administering existing approved drugs and discontinued drugs for the disease to be determined to a disease strain that is a cell affected by the disease to be determined, and medication appearance information obtained by administering the drug to be determined for the disease to be determined to a disease strain of the disease to be determined.The determination model is obtained by using multiple medication appearance information, which is information about the appearance of each cell after administering existing approved drugs and discontinued drugs for the disease to be determined to a disease strain that is a cell affected by the disease to be determined, and

Images