Learning model generation method, information processing device, and program

Abstract

To provide a learning model generation method and the like that can suitably generate a learning model for classifying microorganisms.SOLUTION: In a learning model generation method, a computer executes processes of dividing an image containing microorganisms into a plurality of first division images, further dividing each of the first division images into a plurality of second division images, generating a combined image by combining the second division images with each other, generating training data in which the combined image is associated with classification information indicating the type of microorganism included in the combined image, and generating a learning model that outputs the classification information on the microorganisms when the first division image is input, based on the training data.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present invention relates to a method for generating a learning model, an information processing device, and a program. [Background technology]

[0002] In the testing of microorganisms such as bacteria, specimens are stained and then visually observed under a microscope to determine their species. However, determining bacterial species requires skilled laboratory technicians, and a method to perform this task mechanically is desirable.

[0003] For example, Patent Document 1 discloses an image processing device that detects bacteria from an image of a Gram-stained specimen using a learning model generated by machine learning, and if bacteria are detected, classifies the bacteria according to their type. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2022-20559 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in order to construct a learning model for classifying microorganisms as in the invention of Patent Document 1, a huge amount of training data is required.

[0006] In one aspect, an object of the present invention is to provide a learning model generation method that can suitably generate a learning model for classifying microorganisms. [Means for solving the problem]

[0007] In one aspect, a method for generating a learning model includes a computer executing a process in which an image containing microorganisms is divided into a plurality of first divided images, the first divided images are further divided into a plurality of second divided images, a combined image is generated by combining the second divided images, training data is generated that corresponds the combined image with classification information indicating the type of microorganism contained in the combined image, and a learning model is generated based on the training data that outputs classification information of the microorganism when the first divided image is input. [Effects of the Invention]

[0008] In one aspect, a learning model for classifying microorganisms can be suitably generated. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is an explanatory diagram illustrating an example of the configuration of an inspection system. [Figure 2] FIG. 2 is a block diagram illustrating an example of the configuration of a server. [Figure 3] FIG. 2 is a block diagram illustrating an example of the configuration of a terminal. [Figure 4] FIG. 1 is an explanatory diagram showing problems with bacterial testing using a machine learning model. [Figure 5] FIG. 10 is an explanatory diagram relating to a process for generating a combined image. [Figure 6] FIG. 10 is an explanatory diagram illustrating an example of an annotation screen. [Figure 7] FIG. 10 is an explanatory diagram regarding a learning model generation process. [Figure 8] FIG. 10 is an explanatory diagram showing the verification results of prediction accuracy. [Figure 9] 10 is a flowchart showing the steps of a learning model generation process. [Figure 10] 10 is a flowchart showing the procedure for classifying samples using a learning model. DETAILED DESCRIPTION OF THE INVENTION

[0010] The present invention will be described in detail below with reference to the drawings showing embodiments thereof. (Embodiment) Fig. 1 is an explanatory diagram showing an example of the configuration of an inspection system. In this embodiment, a learning model 50 is constructed that, when an image of Gram-stained bacteria is input, outputs classification information indicating the type of bacteria contained in the image, and an inspection system is described that uses this model to inspect the type of bacteria contained in a specimen. The inspection system includes an information processing device 1 and a terminal 2. Each device is connected to a network N such as the Internet.

[0011] In this embodiment, the test subject is described as bacteria, but microorganisms other than bacteria (for example, mold (fungi)) may also be the target. In other words, bacteria are an example of a microorganism, and other microorganisms may also be the target.

[0012] The information processing device 1 is an information processing device capable of various information processing and information transmission and reception, such as a server computer, a personal computer, a tablet computer, a smartphone, etc. In this embodiment, the information processing device 1 is assumed to be a server computer, and for simplicity, will be referred to as server 1 below. The server 1 generates the above-mentioned learning model 50 by learning training data in which classification information of the bacteria contained in each image is associated with a group of Gram-stained images of bacteria observed under a microscope 3. Specifically, as described below, the server 1 divides images containing only one type of bacteria into predetermined units and combines these to virtually generate a combined image containing multiple types of bacteria, which is used as training data.

[0013] Terminal 2 is a terminal device used by a user of this system (e.g., a medical technician), and is, for example, a personal computer, a tablet terminal, a smartphone, etc. As will be described later, when annotating a group of training images, the user operates terminal 2 to perform annotation (see Figure 6).

[0014] 2 is a block diagram showing an example of the configuration of the server 1. The server 1 includes a control unit 11, a main memory unit 12, a communication unit 13, and an auxiliary memory unit . The control unit 11 has one or more arithmetic processing devices such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), a GPU (Graphics Processing Unit), etc., and performs various information processing, control processing, etc. by reading and executing a program P1 stored in the auxiliary storage unit 14. The main storage unit 12 is a temporary storage area such as an SRAM (Static Random Access Memory), a DRAM (Dynamic Random Access Memory), or a flash memory, and temporarily stores data necessary for the control unit 11 to execute arithmetic processing. The communication unit 13 is a communication module for performing communication-related processing, and transmits and receives information to and from the outside.

[0015] The auxiliary storage unit 14 is a non-volatile storage area such as a large-capacity memory or a hard disk, and stores a program P1 (program product) and other data necessary for the control unit 11 to execute processing. The auxiliary storage unit 14 also stores a learning model 50. The learning model 50 is a machine learning model that has learned training data as described above, such as a CNN (ResNet, VGG, DenseNet, MobileNet, or ResNeXt). The learning model 50 is expected to be used as a program module that constitutes part of artificial intelligence software.

[0016] The auxiliary storage unit 14 may be an external storage device connected to the server 1. The server 1 may be a multi-computer consisting of multiple computers, or may be a virtual machine virtually constructed by software.

[0017] Furthermore, in this embodiment, the server 1 is not limited to the above configuration, and may include, for example, an input unit that accepts operation input, a display unit that displays images, etc. Furthermore, the server 1 may be provided with a reading unit that reads a portable storage medium 1a such as a CD (Compact Disk)-ROM or a DVD (Digital Versatile Disc)-ROM, and may read and execute the program P1 from the portable storage medium 1a.

[0018] 3 is a block diagram showing an example of the configuration of the terminal 2. The terminal 2 includes a control unit 21, a main memory unit 22, a communication unit 23, a display unit 24, an input unit 25, and an auxiliary memory unit 26. The control unit 21 is an arithmetic processing unit such as one or more CPUs, MPUs, etc., and performs various information processing by reading and executing the program P2 stored in the auxiliary storage unit 26. The main storage unit 22 is a temporary storage area such as RAM, and temporarily stores data necessary for the control unit 21 to execute arithmetic processing. The communication unit 23 is a communication module for performing communication-related processing, and transmits and receives information to and from the outside. The display unit 24 is a display screen such as an LCD display, and displays images. The input unit 25 is an operation interface such as a keyboard or mouse, and accepts operation input from the user. The auxiliary storage unit 26 is a non-volatile storage area such as a hard disk or large-capacity memory, and stores the program P2 (program product) and other data necessary for the control unit 21 to execute processing.

[0019] The terminal 2 may be provided with a reading unit for reading a portable storage medium 2a such as a CD-ROM, and may read and execute the program P2 from the portable storage medium 2a.

[0020] 4 is an explanatory diagram showing the problems with bacterial testing using a machine learning model. An outline of this embodiment will be described below.

[0021] When building a machine learning model for bacterial testing, the first step is to identify individual bacteria through segmentation, as shown in Figure 4A. However, in clinical testing, which requires efficient testing of a large number of specimens, there are technical limitations to the magnification of the object observed using the microscope used. In images observed with a typical microscope, bacteria are small in size and have low resolution, making it difficult to identify individual bacterial species. Therefore, it is not possible to build a machine learning model with high predictive accuracy.

[0022] Next, as shown in Figure 4B, bacterial testing can be considered as an image classification problem, and images can be classified according to the type of bacteria contained in the image. However, when multiple types of bacteria are contained in a single image, it is extremely laborious for an examiner to determine the type of bacteria for each image and assign a label, making it difficult to build a model. For example, of the 20 types of bacterial species frequently observed in clinical testing, combinations of 1 to 4 types are frequently observed. 20 For each of the C4=4854 combinations, more than 100 images are required as training images, which is a huge number.

[0023] Therefore, in this embodiment, training data to be used as a learning target is created by virtually generating an image containing a mixture of multiple types of bacteria from an image containing only one type of bacteria.

[0024] Fig. 5 is an explanatory diagram of the process of generating a combined image. Fig. 5 illustrates how an image containing a first bacterium (e.g., Staphylococcus aureus) and an image containing a second bacterium (e.g., Escherichia coli) are divided and then randomly combined to generate a combined image.

[0025] In this embodiment, images containing only one type of bacteria are used as training images. For example, as shown in FIG. 5, the server 1 uses an image containing only a first type of bacteria and an image containing only a second type of bacteria as training images. The server 1 then divides the training image into a predetermined number of images. The example on the left side of FIG. 5 illustrates the image being divided into nine. Note that dividing the image into nine images is just an example, and the number of divisions may be eight or less, or ten or more.

[0026] In the following description, the images obtained by dividing the original image into a predetermined number of divided images will be referred to as "first divided images."

[0027] The server 1 further divides the first divided image into a predetermined number of images into a predetermined number of images. The example in the center of Fig. 5 illustrates how the image is divided into two. Note that dividing the image into two images is just an example, and the number of divided images may be three or more. The number of divided images is preferably determined according to the number of types of bacteria that may be contained in one image (in this example, two types: Staphylococcus aureus and Escherichia coli).

[0028] In the following description, an image obtained by further dividing the first divided image will be referred to as a "second divided image."

[0029] Through the above process, the original image is divided into 9 × 2 = 18 pieces. The server 1 combines each second divided image with a second divided image that is different from the second divided image of the pair before division, to generate a combined image of the same size as the original first divided image. Specifically, the server 1 randomly combines and combines two second divided images from the 18 second divided images obtained by dividing the image including the first bacterium and the 18 second divided images obtained by dividing the image including the second bacterium, to generate a greater number of combined images than the original first divided images. An example of a combined image is shown on the right side of Figure 5. As illustrated on the right side of Figure 5, three patterns of combined images are generated: one that includes both the first bacterium and the second bacterium, one that includes only the first bacterium, and one that includes only the second bacterium.

[0030] The server 1 generates training data by assigning classification information to each combined image indicating the type of bacteria contained in the combined image. In the example on the right side of Figure 5, the first combined image from the top is assigned classification information indicating the classification result as a first bacterium and a second bacterium. Next, the second combined image is assigned classification information indicating the classification result as the first bacterium. The third combined image is assigned classification information indicating the classification result as the second bacterium. In this way, the server 1 assigns classification information according to the combination of bacteria contained in each of the second segmented images that make up the combined image. In this way, the server 1 generates training data to be used for machine learning.

[0031] In the following description, training data in which a combined image formed by combining a second segmented image containing a first bacterium and a second segmented image containing a second bacterium is associated with classification information indicating that the classification result is the first bacterium and the second bacterium is referred to as "first training data." Furthermore, training data in which a combined image formed by combining second segmented images containing the first bacterium is associated with classification information indicating that the classification result is the first bacterium is referred to as "second training data." Furthermore, training data in which a combined image formed by combining second segmented images containing the second bacterium is associated with classification information indicating that the classification result is the second bacterium is referred to as "third training data."

[0032] It is preferable that the server 1 generates more combined images (first training data) containing a mixture of multiple types of bacteria than combined images (second training data and third training data) containing a single type of bacteria, thereby learning more situations in which multiple types of bacteria are mixed, and improving prediction accuracy.

[0033] FIG. 6 is an explanatory diagram showing an example of an annotation screen. FIG. 6 illustrates an operation screen displayed by terminal 2 when generating training data as described above. This screen includes a first number designation field 61, a second number designation field 62, a bacterial species designation field 63, an image designation field 64, and a divided image 65. The first number designation field 61 is a designation input field for designating the number of first divided images into which the original image is to be divided. The second number designation field 62 is a designation input field for designating the number of second divided images into which the first divided image is to be further divided. The bacterial species designation field 63 is a designation input field for designating the type of bacteria (classification information) contained in the image. The image designation field 64 is an operation input field for designating the image to be divided. The divided image 65 is a display field for displaying the divided image.

[0034] Terminal 2 displays the screen and accepts various setting inputs from the user. That is, terminal 2 accepts an input specifying the number of images to be used when dividing the original image into first divided images in response to an operation input into a first number specification field 61. Next, terminal 2 accepts an input specifying the number of images to be used when dividing the first divided image into second divided images (the number of bacterial species that may be included in the image) in response to an operation input into a second number specification field 62. Next, terminal 2 accepts an input specifying the type of bacteria included in the image (annotation) in response to an operation input into a bacterial species specification field 63.

[0035] Furthermore, terminal 2 reads out the image for which the bacterial species was specified above in response to an operation input to image specification field 64. Terminal 2 divides the image up to the second divided image and displays it as divided image 65. The corresponding bacterial species name is registered (annotated) in the displayed divided area, except for the second divided image (blank area) that does not contain bacteria.

[0036] In this case, terminal 2 binarizes the image using Otsu's method or the like, identifies second divided images (blank areas) that do not contain bacteria, and registers (annotates) the blank areas in the identified divided areas instead of the bacterial species name, and displays them in a different display mode (for example, a different display color) from other areas related to the second divided images. In the example of FIG. 6, the areas related to the second divided images that do not contain bacteria are shown hatched. The second divided images (blank areas) that do not contain bacteria can be added or deleted as selection targets using the mouse pointer or keyboard directional keys. Note that in the example of FIG. 6, the second divided image (blank area) that does not contain bacteria is selected, but instead of selecting a blank area that does not contain bacteria, a divided area that includes the bacterial species specified in the bacterial species specification field 63 may be selected.

[0037] Terminal 2 excludes the second segmented images that do not contain bacteria from the second segmented images used to generate the combined image, thereby making it possible to avoid learning areas that do not contain bacteria.

[0038] The terminal 2 accepts the above operation for each image, divides the image, and accepts setting input (annotation) of classification information (type of bacteria) for each second divided image. The server 1 combines these second divided images to generate a combined image, and assigns classification information to the combined image according to the combination of the second divided images. In this way, the server 1 generates training data.

[0039] The server 1 generates a learning model 50 that outputs classification information when the first divided image is input based on the training data. Figure 7 is an explanatory diagram regarding the generation process of the learning model 50. The left side of Figure 7 illustrates how a combined image of the first to third training data is given to the learning model 50 and learning is performed.

[0040] The learning model 50 is a machine learning model constructed by deep learning, such as CNN (ResNet, etc.). The learning model 50 may be a machine learning model other than CNN, such as Vision Transformer. The server 1 inputs training combined images into the learning model 50 to output classification information. The server 1 compares the output classification information with a correct answer value and adjusts parameters such as weights so that the two are close to each other. The server 1 provides each combined image related to the first to third training data to the learning model 50 to optimize the parameters, and finally generates the learning model 50.

[0041] Furthermore, the server 1 verifies the prediction accuracy of the learning model 50 generated above, as shown on the right side of Figure 7. To verify the prediction accuracy, an image containing a mixture of multiple types of bacteria is used. Here, we consider the case where an image containing a mixture of two types of bacteria, a first bacterium (Staphylococcus aureus) and a second bacterium (Escherichia coli), is used.

[0042] First, the server 1 divides the image into first divided images. The server 1 outputs classification information by inputting the first divided images into a learning model 50, and verifies the prediction accuracy by comparing the output classification information with a correct answer value.

[0043] Furthermore, the server 1 further divides the first divided image to generate a second divided image. The server 1 then randomly combines the generated second divided images to generate a combined image. The server 1 also uses this combined image to verify the prediction accuracy. The server 1 inputs the combined image into the learning model 50 to output classification information, and verifies the prediction accuracy by comparing the output classification information with the correct answer value.

[0044] The reason why not only the first divided image but also the combined image is used to verify the prediction accuracy is to avoid the influence of the boundary between the two second divided images that exists in the combined image used for learning. By using not only the first divided image but also the combined image for verification, the prediction accuracy of the learning model 50 can be suitably verified.

[0045] When actually performing an examination, the server 1 makes a prediction using the learning model 50 generated above. Specifically, the server 1 acquires an image of the sample from the terminal 2 and divides it into first divided images. The server 1 then inputs the divided first divided images into the learning model 50, thereby outputting classification information. The server 1 returns the output classification information to the terminal 2 and presents it to the user (laboratory technician).

[0046] 8A to 8C are explanatory diagrams showing the results of verification of prediction accuracy, each of which illustrates the results of an experiment conducted to verify the prediction accuracy of the learning model 50 according to this embodiment.

[0047] 8A shows the results of verifying the prediction accuracy of learning model 50, which was trained not on combined images but on images containing only one type of bacteria. As shown in FIG. 8A, when predictions were made on images containing only one type of bacteria, the hit rate (precision), sensitivity (recall), and F-value (F1 score) all showed high values. However, because the model was not trained on images containing a mixture of multiple types of bacteria, the prediction accuracy was low when predictions were made on images containing a mixture of multiple types of bacteria.

[0048] Figures 8B and 8C show the results of verifying the prediction accuracy of learning model 50 when a combined image containing a mixture of different bacteria is used as the learning target. More specifically, Figure 8B shows the results of verification using a virtual combined image containing multiple types of bacteria, while Figure 8C shows the results of verification using an actual image (first segmented image). In both Figures 8B and 8C, it can be seen that the prediction accuracy for images containing multiple types of bacteria (third row of the table) is significantly improved compared to Figure 8A. In this way, by learning a combined image formed by combining multiple second segmented images, it is now possible to accurately detect the type of bacteria even when multiple types of bacteria are present.

[0049] 9 is a flowchart showing the procedure of the process of generating the learning model 50. The process of generating the learning model 50 by machine learning will be described with reference to FIG.

[0050] The control unit 11 of the server 1 receives setting inputs, such as the number of images to be divided into first divided images, the number of images to be divided into second divided images, and the type of bacteria (classification information) contained in the images, via the annotation screen displayed on the terminal 2 (step S11). The control unit 11 acquires a group of images for training (step S12). The control unit 11 divides each acquired image into a plurality of first divided images according to the set number (step S13). The control unit 11 further divides the first divided images into a plurality of second divided images (step S14). After division into second divided images, second divided images that do not contain bacteria are excluded from subsequent processing.

[0051] The control unit 11 randomly combines and combines the divided second divided images to generate a combined image (step S15). Specifically, the control unit 11 generates a combined image by combining a second divided image with a second divided image that is different from the second divided image of the pair before division, so that the combined image has the same size as the first image when divided into the first number of images. In this case, the control unit 11 randomly combines the second divided images to generate a greater number of combined images than the original first divided images. Furthermore, it is preferable that the control unit 11 generates more combined images (first training data) by combining second divided images of different types of bacteria than combined images (second training data and third training data) of second divided images of the same type of bacteria.

[0052] The control unit 11 generates training data that associates the generated combined image with classification information for the bacteria in the combined image (step S16). That is, the control unit 11 assigns classification information to the combined image according to the combination of bacteria contained in each of the second segmented images that make up the combined image. In this way, the control unit 11 generates first training data in which the bacteria contained in the combined image are different between the first bacteria and the second bacteria, second training data in which the bacteria contained in the combined image are common to the first bacteria, and third training data in which the bacteria contained in the combined image are common to the second bacteria.

[0053] Based on the training data, the control unit 11 generates a learning model 50 that outputs classification information for bacteria in the first segmented image when the first segmented image is input (step S17). For example, the control unit 11 generates a CNN (ResNet) as the learning model 50. The control unit 11 outputs classification information by inputting a training combined image into the learning model 50, and compares the output classification information with the correct value of the classification information associated with the training combined image. The control unit 11 optimizes parameters such as weights so that the two are similar to each other, and generates the learning model 50.

[0054] The control unit 11 acquires a group of test images (step S18). The test image group is an image containing a mixture of multiple types of bacteria. The control unit 11 divides the test images into first divided images, and further divides them into second divided images (step S19). The control unit 11 generates a combined image by combining the divided second divided images (step S20). The control unit 11 then inputs the first divided image and the combined image by combining the second divided images into the learning model 50 to output classification information, and verifies the prediction accuracy of the learning model 50 by comparing the output classification information with the correct answer value (step S21). The control unit 11 ends the series of processes.

[0055] Fig. 10 is a flowchart showing the procedure for classifying a specimen using the learning model 50. The processing contents for classifying bacteria using the learning model 50 will be described with reference to Fig. 10. The procedure for classifying a specimen may be either Fig. 10 or the processing contents for the test image group (the process of outputting classification information from step S18 to step S21).

[0056] The control unit 11 of the server 1 acquires an image of a specimen to be inspected from the terminal 2 (step S31). The control unit 11 divides the acquired image into first divided images (step S32). The control unit 11 inputs the divided images into the learning model 50, thereby outputting classification information of the bacteria (step S33). The control unit 11 ends the series of processes.

[0057] As described above, according to this embodiment, the training image is divided into second divided images, which are then combined, and the resulting combined image is learned to generate a learning model 50. This reduces the amount of training data that needs to be manually created, while enabling accurate prediction of bacterial species even for specimens containing a mixture of multiple types of bacteria.

[0058] The embodiments disclosed herein are to be considered as illustrative in all respects and not restrictive. The scope of the present invention is defined by the claims, not by the above meaning, and is intended to include all modifications within the meaning and scope of the claims.

[0059] The matters described in the embodiments can be combined with each other. In addition, the independent claims and dependent claims described in the claims can be combined with each other in any combination, regardless of the reference format. Furthermore, the claims use a format in which a claim references two or more other claims (multiple claim format), but this is not limited to this. A multiple claim (multi-multi claim) that references at least one other multiple claim may also be used. [Explanation of symbols]

[0060] 1. Server (information processing device) 11 Control section 12 Main memory 13 Communications Department 14 Auxiliary storage P1 Program 2. Devices 21 Control section 22 Main memory 23 Communications Department 24 Display section 25 Input section 26 Auxiliary storage P2 Program

Claims

1. Dividing the image containing the microorganism into a plurality of first divided images; further dividing the first divided image into a plurality of second divided images; generating a combined image by combining the second divided images; generating training data in which the combined image is associated with classification information indicating the types of microorganisms contained in the combined image; A learning model is generated based on the training data, which outputs classification information of microorganisms when the first segmented image is input. A method for generating a learning model in which processing is performed by a computer.

2. The image contains one type of microorganism, The image including the first microorganism and the image including the second microorganism are each divided into the second divided images and combined to generate the combined image. The method for generating a learning model according to claim 1 .

3. first training data in which the combined image obtained by combining the second divided image including the first microorganism and the second divided image including the second microorganism is associated with the classification information for the first microorganism and the second microorganism; second training data in which the combined image obtained by combining the second divided images including the first microorganism is associated with the classification information identifying the first microorganism; and generating third training data in which the combined image obtained by combining the second divided images containing the second microorganism is associated with the classification information identifying the second microorganism. The method for generating a learning model according to claim 2 .

4. generating the learning model based on the first training data, the second training data, and the third training data; The method for generating a learning model according to claim 3 .

5. generating the first training data more than the second training data and the third training data; The method for generating a learning model according to claim 3 .

6. The second divided image is combined with a second divided image different from the second divided image of the pair before the division to generate a combined image having the same size as the first divided image. The method for generating a learning model according to claim 1 .

7. The second divided images are randomly combined to generate a greater number of combined images than the first divided images. The method for generating a learning model according to claim 1 .

8. generating a first divided image obtained by dividing an image including a plurality of types of microorganisms and a combined image obtained by combining second divided images obtained by further dividing the first divided image; The generated first divided image and combined image are input into the learning model, respectively, to verify the prediction accuracy. The method for generating a learning model according to claim 1 .

9. An information processing device including a control unit, The control unit Dividing the image containing the microorganism into a plurality of first divided images; further dividing the first divided image into a plurality of second divided images; generating a combined image by combining the second divided images; generating training data in which the combined image is associated with classification information indicating the types of microorganisms contained in the combined image; A learning model is generated based on the training data, which outputs classification information of microorganisms when the first segmented image is input. Information processing device.

10. Dividing the image containing the microorganism into a plurality of first divided images; further dividing the first divided image into a plurality of second divided images; generating a combined image by combining the second divided images; generating training data in which the combined image is associated with classification information indicating the types of microorganisms contained in the combined image; A learning model is generated based on the training data, which outputs classification information of microorganisms when the first segmented image is input. A program that causes a computer to perform a process.

11. Acquire an image containing microorganisms, Dividing the acquired image into first divided images; The training data, which associates the divided first divided image with a combined image obtained by combining the second divided images obtained by dividing the first divided image, and classification information indicating the types of microorganisms contained in the combined image, is input into a trained learning model, and the classification information is output. A program that causes a computer to perform a process.

Images