Information processing system, information processing apparatus, information processing method and program

Abstract

To provide an information processing system, an information processing apparatus, an information processing method and a program that generate highly accurate teacher data by performing annotation for learning data on the basis of medical information given to other medical image data corresponding to a subject.SOLUTION: An information processing system according to the present invention comprises: an acquisition unit that acquires first medical information corresponding to PCCT image data which is obtained using a photon counting type radiation detector for a subject and CT image data which is obtained using an energy integration type radiation detector for the subject; a teacher data acquisition unit that acquires first teacher data which is given a first correct label based on the first medical information for the CT image data; and a learning unit that learns a CT inference model which performs predetermined inference for the CT image data using the first teacher data.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an information processing system, an information processing device, an information processing method, and a program for generating highly accurate teacher data. [Background technology]

[0002] In general, training data used in training an inference device based on machine learning is generated based on the knowledge of annotators, such as experts, regarding the data.

[0003] For example, an inference machine that infers the presence or absence of lung cancer from medical image data such as CT images is generated by a learning process based on learning data in which correct labels indicating whether or not lung cancer is present are annotated by experts observing the CT image data.

[0004] Here, Patent Document 1 discloses that an annotator assigns a correct answer label to candidates for training data. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2016-76073 A Summary of the Invention [Problem to be solved by the invention]

[0006] However, in the technology described in Patent Document 1, the accuracy of annotations corresponding to correct labels depends on the annotator. For example, when a feature region that affects inference is overlooked or knowledge about the feature region is insufficient, the learning data may not be appropriately annotated.

[0007] The present invention aims to provide an information processing system, an information processing device, an information processing method, and a program that generate highly accurate teacher data by annotating learning data based on medical information assigned to other medical image data corresponding to a subject. [Means for solving the problem]

[0008] In order to solve the above problems, an information processing system according to one aspect of the present invention includes an acquisition unit that acquires first medical information corresponding to PCCT image data obtained by using a photon counting type radiation detector for a subject, and CT image data obtained by using an energy integral type radiation detector for the subject; a training data acquisition unit that acquires first training data in which a first correct answer label based on the first medical information is assigned to the CT image data; A learning unit that uses the teacher data to learn a CT inference model that performs a predetermined inference on the CT image data; has. Effect of the Invention

[0009] According to the present invention, highly accurate teacher data can be generated based on medical information added to other medical image data. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram showing an example of a functional configuration of an information processing system according to the present invention. [Diagram 2] FIG. 1 is a diagram showing an example of a hardware configuration of an information processing apparatus according to an embodiment of the present invention. [Diagram 3] FIG. 2 is a diagram showing an example of a teacher data creation flow of the information processing system according to the first embodiment. [Figure 4] FIG. 2 is a diagram showing an example of a teacher data creation flow of the information processing system according to the first embodiment. [Diagram 5] A diagram showing a schematic diagram of an example of an inference model relating to the second embodiment. [Figure 6]FIG. 11 is a diagram showing an example of a teacher data creation flow of an information processing system according to a second embodiment. [Figure 7] FIG. 11 is a diagram showing an example of a teacher data creation flow of an information processing system according to a second embodiment. [Figure 8] A diagram showing a schematic diagram of an example of an inference model relating to the third embodiment. [Figure 9] FIG. 13 is a diagram showing an example of an inference flow of an information processing system according to the third embodiment. [Figure 10] FIG. 13 is a diagram showing an example of an inference flow of an information processing system according to the third embodiment. [Figure 11] FIG. 13 is a diagram showing an example of an inference flow of an information processing system according to the third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] The present invention can be preferably applied to sinogram data acquired by a modality, medical image data generated by image data reconstruction from raw data, and the like.

[0012] The modality can be applied to, for example, CT data and PCCT data, which are data acquired by an X-ray CT device, photon counting CT (PCCT: Photon Counting Computed Tomography), etc., and CT image data and PCCT image data generated by image data reconstruction from the data.

[0013] It should be noted that PCCT image data is image data generated by reconstructing PCCT data obtained using a photon counting type radiation detector using a known reconstruction technique. Also, CT image data is image data generated by reconstructing CT data obtained using an energy integral type radiation detector. It should be noted that the energy integral type radiation detector and the photon counting type radiation detector are different radiation detectors.

[0014] Hereinafter, an information processing system, an information processing device (learning device, inference device) constituting the information processing system, an information processing method and a program executed by each device will be described with reference to the drawings. Note that the following description of the embodiment is merely an example, and is not limited to the following embodiment.

[0015] [First embodiment] 1, an information processing system 1 including a learning device 2, which is an information processing device of the present invention, and an inference device 3 will be described. The information processing system 1 of the present invention is configured to include the learning device 2, the inference device 3, and a data server 4. The information processing system 1 also has an input unit 5 that accepts information input by a user to the information processing system 1, and a display unit 6 that displays information under display control from the information processing system 1.

[0016] Each device is configured to be connectable via a network. Each device and each functional configuration of the information processing system 1 may be configured from one device or multiple devices, and the device that performs the function of the learning device may be managed or operated by different people from the device that performs the function of the inference device. Each device may be connected by wire.

[0017] The information processing device 1 includes a learning device 2 that learns a CT inference model based on first medical information corresponding to PCCT image data obtained using a photon counting type radiation detector on a subject, and an inference device 3 that performs inference processing on the inference target data using the CT inference model learned by the learning device 2.

[0018] Here, the learning device 2 has an acquisition unit 21 that acquires first medical information corresponding to PCCT image data obtained on a subject using a photon counting type radiation detector, and CT image data obtained on the subject using an energy integration type radiation detector.

[0019] The learning device 2 also has a teacher data acquisition unit 22 that acquires first teacher data to which a first correct answer label based on first medical information has been assigned for the CT image data, and a judgment unit 23 that compares the first medical information with second medical information based on the CT image data and judges whether or not the first teacher data should be used for learning the CT inference model.

[0020] Furthermore, the learning device 2 includes a learning unit 24 that uses the teacher data to learn a CT inference model that performs a predetermined inference on the CT image data.

[0021] The acquisition unit 21 acquires first medical information corresponding to PCCT image data obtained for a subject using a photon counting type radiation detector, and CT image data obtained for the subject using an energy integral type radiation detector. Here, the PCCT image data is integral image data generated by reconstructing medical data obtained for a subject using a photon counting type radiation detector, or energy discriminated image data which is data after energy discrimination. The first medical information is document information such as a medical report generated by a doctor for reading or questioning, or a medical report generated by image processing, for the PCCT image data. Here, the first medical information may be acquired using an inference model of PCCT. The inference model of PCCT is an inference model learned based on teacher data to which a first correct answer label is assigned to the PCCT image data. The inference model of PCCT may be generated based on deep learning or machine learning technology.

[0022] The teacher data acquisition unit 22 acquires teacher data to which a first correct answer label based on the first medical information corresponding to the PCCT image data is assigned. The teacher data acquisition unit 22 extracts information that is to be a correct answer label from document information constituting the first medical information using a known technique such as natural language processing, sets the information as a first correct answer label for the CT image data of the same subject, and acquires teacher data that pairs the CT image data with the first correct answer label based on the first medical information.

[0023] When there is the first medical information and the second medical information for the CT image data, the determination unit 23 compares both information to determine whether or not the first teacher data based on the first medical information is used for learning the CT inference model. Note that the determination may be omitted as appropriate depending on the presence or absence of the second medical information or the user's settings. Here, the second medical information is medical information given to the CT image data, and is a medical report generated by a doctor through interpretation, questioning, etc., or document information described in a medical report generated by image processing for the CT image data.

[0024] The learning unit 24 acquires information on the inference model of the CT from the data server 4. Here, the information on the inference model includes structural data of the model, and further includes parameter information of the model if the inference model is a pre-trained model. Furthermore, if there is teacher data used in pre-training for the inference model, the teacher data may also be acquired.

[0025] The inference model is realized by a known deep learning or machine learning technique. For example, any inference model may be generated based on teacher data with a correct answer label, such as a deep learning technique such as CNN or Vision Transformer, or a machine learning technique such as SVM or RF, and a plurality of inference models may be combined as an ensemble model.

[0026] The learning unit 24 applies the acquired first teacher data to the inference model and performs a learning process to determine parameters for the inference model. If the inference model is a pre-trained inference model, additional learning may be performed on the inference model, or if teacher data used in the pre-training can be acquired, the parameters of the inference model may be newly learned using teacher data that combines the teacher data and the first teacher data.

[0027] When the learning process of the inference model is completed, the learning unit 24 transmits parameter information to the data server 4.

[0028] Here, the inference device 3 in the information processing system 1 performs inference processing using the inference model learned by the learning device 2. Specifically, the inference device 3 includes an inference target data acquisition unit 31 that acquires medical image data of an inference target, and a model acquisition unit 32 that acquires a CT inference model learned using teacher data that pairs a correct answer label based on medical information assigned to PCCT image data obtained for a subject using a photon counting type radiation detector and CT image data obtained for the subject using an energy integral type radiation detector. The inference device 3 further includes an inference unit 33 that performs a predetermined inference on the medical image data of the inference target using the CT inference model, and a display control unit 34 that causes the display unit 6 to display the inference result by the inference unit 33.

[0029] The data server 4 stores various data. For example, it stores medical image data such as teacher data and inference target data, various inference models, and information on parameters of the inference models, and transmits data via a network in response to requests from each device.

[0030] The input unit 5 is made up of input devices such as a mouse and a keyboard, and receives various instructions from the user.

[0031] The display unit 6 is a display attached to the information processing system 1, a mobile terminal of a medical worker via an external server, or the like.

[0032] 2 shows an example of a specific configuration of the information processing system 1 or the learning device 2 and the inference device 3 that constitute the information processing system 1. In this example, the information processing device 1 has a CPU 20, a GPU 21, a RAM 22, a ROM 23, and a storage device 24, which are connected via a system bus 25.

[0033] Hereinafter, an example of learning an inference model by the learning device 2 of this embodiment will be described with reference to FIG.

[0034] In step S30, the acquisition unit 21 acquires first medical information corresponding to PCCT image data obtained on the subject using a photon counting type radiation detector, and CT image data obtained on the subject using an energy integral type radiation detector, and then transmits the data to the teacher data acquisition unit 22 and proceeds to the next step.

[0035] In step S31, the teacher data acquisition unit 22 acquires first teacher data in which the CT image data and the first correct label are paired by extracting a first correct label from the acquired first medical information using a known natural language processing technique, etc. After transmitting information about the first teacher data to the learning unit 24, the teacher data acquisition unit 22 proceeds to the next step.

[0036] In step S32, the learning unit 24 acquires information on the CT inference model from a storage unit such as the data server 4, and performs learning processing using the first teacher data set.

[0037] The CT inference model is a model that performs a predetermined inference using CT image data as input, and the task performed may be any of classification, detection, extraction, etc. The class to be classified may be an area depicted in an image such as an organ or a lesion site, or the degree of abnormality of a predetermined area may be inferred from information depicted in an image. The model may also perform either classification or regression.

[0038] When the learning process for the inference model using the first teacher data is completed, the learning unit 24 transmits the parameter information determined by the learning to the data server 4, and ends the flow.

[0039] With this configuration, it is possible to efficiently generate training data for learning a CT inference model that performs CT inference based on medical information assigned to medical image data that is different from the input of the inference model, such as PCCT image data.

[0040] Furthermore, PCCT image data is image data with a higher resolution and higher definition than CT image data. In addition to integral image data by PCCT, the use of energy discrimination image data enables highly accurate interpretation. Therefore, compared to CT image data, PCCT image data has a larger amount of information, and therefore the accuracy of the medical information generated tends to be higher. Due to these characteristics, the information processing system 1 can create highly accurate teacher data by using the first medical information for the PCCT image data as a correct answer label for the CT image data. In addition, it is possible that the characteristic area that has become easier to interpret by taking the PCCT image is locally depicted on the CT image data. In such a case, more accurate inference is possible by learning a CT inference model using as teacher data CT image data to which a correct answer label based on medical information obtained from the PCCT image data has been added.

[0041] (Variation 1) Next, a first modified example of learning an inference model by the learning device 2 of this embodiment will be described with reference to FIG.

[0042] In the first modification, when second medical information is added to CT image data, the determination unit 23 compares the first medical information with the second medical information to determine whether or not to use the first medical information as first teacher data to which a first correct answer label based on the first medical information is added. Specifically, even if the medical information added to the CT image data is overlooked or annotated based on the oversight, it is possible to create highly accurate teacher data. Hereinafter, the same configuration as the above-mentioned embodiment is given the same figure number, and the description will be omitted as appropriate.

[0043] In step 40, the acquisition unit 21 further acquires second medical information for the CT image data. The second medical information is medical information assigned to the CT image data, and is a medical report generated by a doctor through image interpretation, questioning, etc., or document information described in a medical report generated by image processing of the CT image data. The acquisition unit 21 transmits the acquired second information to the determination unit 23 and proceeds with the process.

[0044] In step S41, the determination unit 23 compares the first medical information with the second medical information and determines whether the two pieces of information match. The match may not be a perfect match, but may be calculated as a score representing the degree of match, such as a partial match. If the first medical information and the second medical information do not match, the determination unit 23 advances the process to step S31. On the other hand, if the first medical information and the second medical information match, the process is terminated.

[0045] With this configuration, even if there is an oversight in the second medical information, a correct answer label based on the first medical information can be added, so that highly accurate teacher data can be created. On the other hand, if the second medical information and the first medical information match, it is not necessary to create a correct answer label based on the first medical information, so no processing is required.

[0046] [Second embodiment] An example of a flow of creating teacher data in the information processing system 1 in this embodiment will be described below with reference to Fig. 5 and Fig. 6. Note that the same components as those in the first embodiment are given the same reference numerals, and the description will be omitted as appropriate.

[0047] In this embodiment, the inference model of the CT to be learned is a pre-trained inference model and is stored in the data server 4.

[0048] The data server 4 stores the pre-trained pre-CT inference model 410 in association with second teacher data 415, which is teacher data used to train the pre-CT inference model. The learning device 2 transmits the trained CT inference model 420 from the learning device 2 to the data server 4. Here, the transmitted information may be updated parameter information of the CT inference model. When the model structure is different from the pre-CT inference model 410, the model structure information of the CT inference model 420 is stored in association with parameter information.

[0049] Hereinafter, the learning flow of the learning device 2 in this embodiment will be described with reference to FIG.

[0050] In step S61, the acquisition unit 21 further acquires second teacher data used to learn the pre-CT inference model 410 as second medical information, and transmits the second medical information to the judgment unit 23, and then proceeds to the next step.

[0051] In step S62, the determination unit 61 determines whether the first correct label based on the first medical information and the second correct label based on the second medical information match. The match may be determined by calculating the degree of match. If the determination determines that the correct labels match, this learning flow ends. On the other hand, if it is determined that the correct labels are different, information on the first correct label is sent to the teacher data acquisition unit 22, and processing proceeds.

[0052] With this configuration, even if there is an annotation error in the second training data 415 used to train the pre-CT inference model 410, highly accurate training data can be created in which the first correct label based on the first medical information assigned to the PCCT image data is used as the correct label for the CT image data.

[0053] (Variation 2) Next, a second modified example of the learning of an inference model by the learning device 2 of this embodiment will be described with reference to Fig. 7. Configurations similar to those in other embodiments and the above-mentioned examples will be given the same reference numerals, and descriptions thereof will be omitted as appropriate.

[0054] In this modified example, the inference results for the CT image data by the pre-CT classifier are obtained as second medical information and compared with the first medical information for the PCCT image data. If the first medical information and the second medical information differ, a first correct answer label based on the first medical information is assigned to the CT image data, and training data is generated.

[0055] The learning process in the second modification will be described below with reference to FIG.

[0056] In step S71, the acquisition unit 21 further acquires an inference model of the pre-CT scan, performs inference processing on the CT image data using the acquired inference model of the pre-CT scan, transmits the inference result to the judgment unit 23 as second medical information, and proceeds to the next step.

[0057] In step S72, the determination unit 23 determines whether the first medical information and the second medical information match, and if the two pieces of information are different, the process proceeds to step S31. On the other hand, if the two pieces of information are different, the process ends.

[0058] With this configuration, it is possible to create training data with correct labels based on the first medical information for CT image data that the pre-CT inference model cannot classify well, and by using this training data to train the CT inference model, highly accurate inference is possible.

[0059] (Variation 3) The acquisition unit 21 may further acquire an acquisition interval between the first medical information and the second medical information. If the information acquisition interval is large, the determination unit 23 determines not to assign the first correct label to the CT image data. In addition, the determination unit 23 may determine whether to assign the first correct label to the CT image data based on the diagnosis or findings described in the first medical information and the information acquisition interval.

[0060] [Third embodiment] Using the inference model generated by the learning method described in the first and second embodiments, inference processing is performed on medical image data that is the subject of inference.

[0061] An example of an inference flow of the information processing system 1 in this embodiment will be described below with reference to FIGS.

[0062] 8 shows a schematic diagram of the configuration of the inference device 3 and the relationship between the data server 4. The model acquisition unit 31 acquires from the data server 4 an inference model 420 of a CT trained in either the first embodiment or the second embodiment.

[0063] The inference flow of the inference device 3 in this embodiment will be described with reference to FIG.

[0064] In step S90, the inference target data acquisition unit 31 acquires medical image data of the inference target, transmits the acquired information to the inference unit 33, and proceeds to the next step.

[0065] In step S91, the model acquisition unit 32 acquires a CT inference model trained using training data that pairs a correct answer label based on medical information assigned to PCCT image data obtained for a subject using a photon counting type radiation detector with CT image data obtained for the subject using an energy integral type radiation detector, and transmits the information about the inference model to the inference unit 33, and then proceeds to the next step.

[0066] In step S92, the inference unit 33 performs a predetermined inference process on the acquired medical image data to be inferred using a CT inference model, transmits the inference result to the display control unit 34, and then proceeds to the next step.

[0067] In step S93, the display control unit 34 causes the display unit 6 to display the inference result, and ends the inference flow.

[0068] With this configuration, inference can be performed using a learned CT inference model based on highly accurate training data.

[0069] (Variation 1) Hereinafter, a first modified example of the inference flow of the information processing system 1 in this embodiment will be described with reference to FIG.

[0070] In this modified example, the inference unit 33 determines whether the medical information added to the CT image data matches the inference result based on the CT inference model, and if they do not match, the display control unit 34 causes the display unit 6 to display the added medical information and the inference result so that they can be compared.

[0071] In step S1003, the inference unit 33 compares the medical information added to the CT image data with the inference result for the CT image data. If the medical information and the inference result differ, the inference unit 33 transmits the medical information and the inference result to the display control unit 34 and proceeds to the next step.

[0072] In step S1004, the display control unit 34 causes the display unit 6 to display the medical information and the information on the inference result in a comparative manner.

[0073] With this configuration, it is possible to determine whether the medical information assigned to CT image data by a doctor or other person differs from the inference results of the CT inference model, and if they differ, a request can be made to confirm the differences in the medical information.

[0074] (Variation 2) Hereinafter, the second modified example of the inference flow of the information processing system 1 in this embodiment will be described with reference to FIG.

[0075] In this modified example, the inference unit 33 acquires an inference model 410 of the pre-CT and an inference model 420 of the CT generated through the learning flow of the above-mentioned embodiment, and performs inference using both inference models.

[0076] Furthermore, when the inference results from both inference models are different, the display control unit 34 associates both inference results with difference information between the two inference results, and causes the display unit 6 to perform display processing.

[0077] In step S1101, the inference unit 33 acquires the inference model 410 of the pre-CT and the inference model 420 of the CT generated through the above-mentioned learning process from the data server 4, and then proceeds to the next step.

[0078] In step S1102, the inference unit 33 performs a first inference process using the pre-CT inference model 410 and a second inference process using the CT inference model 420 on the medical data to be inferred, and then proceeds to the next step.

[0079] In step S1103, the inference unit 33 determines whether the first inference result from the first inference process and the inference result from the second inference process match, and if they differ, transmits the first and second inference results and, if necessary, difference information between the inference results to the display control unit 34, and proceeds to step S1104. If they match, the inference flow ends.

[0080] In step S1104, the display control unit 34 causes the display unit 6 to display the first and second inference results and the difference information acquired from the inference unit 33.

[0081] With this configuration, when the inference results based on the inference model of the pre-CT and the inference model of the CT differ, the user can check information on the differences.

[0082] (Variation 3) Note that the inference unit 33 may acquire pixel information that contributed to the inference result using a visualization technique such as Grad-CAM, and the display control unit 34 may display the acquired pixel information on the display unit 6. Here, the configuration implemented by the inference unit 33 has been described, but this modified example may be applied to verification of the inference model learned by the learning unit 24 of the first and second embodiments.

[0083] With this configuration, the user can understand which pixel of the CT image the CT inference model performed the inference process by looking at.

[0084] (Variation 4) Based on first medical information corresponding to PCCT image data obtained using a photon counting radiation detector, a first correct answer label may be assigned to the pseudo-generated CT image data from the PCCT data obtained using the radiation detector.

[0085] With this configuration, training data for a CT inference model can be created even from information obtained by a single image of the subject using a photon counting radiation detector.

[0086] (Other Examples) The present invention can also be realized by executing the following process: That is, software (programs) that realize the functions of the above-described embodiments are supplied to a system or device via a network or various storage media, and the computer (or CPU, MPU, etc.) of the system or device reads and executes the programs.

Claims

1. An acquisition unit that acquires first medical information corresponding to PCT image data obtained by using a photon-counting type radiation detector for a subject, and CT image data obtained by using an energy-integrating type radiation detector for the subject; A teacher data acquisition unit that acquires first teacher data in which a first correct label based on the first medical information is assigned to the CT image data; A learning unit that learns a CT inference model that performs a predetermined inference on CT image data using the first teacher data; An information processing system, characterized by comprising the above.

2. The information processing system according to claim 1, wherein the first medical information is medical information generated by a user for the PCT image data.

3. The information processing system according to claim 1, wherein the first medical information is information based on a result inferred using a PCT inference model for the PCT image data.

4. The information processing system according to claim 1, wherein the CT inference model is an inference model pre-trained using second teacher data based on second medical information corresponding to the CT image data.

5. The information processing system according to claim 4, wherein the learning unit further learns the CT inference model using the first teacher data acquired by the teacher data acquisition unit.

6. The information processing system according to claim 1, wherein the acquisition unit further acquires second medical information corresponding to the CT image data.

7. The information processing system according to claim 6, wherein the second medical information is medical information generated by a user for the CT image data.

8. The information processing system according to claim 6, wherein the second medical information is information based on a result inferred using the pre-trained CT inference model.

9. The information processing system according to any one of claims 6 to 8, further comprising a determination unit that determines whether to use the first teacher data for learning the CT inference model by comparing the first medical information and the second medical information.

10. The determination unit determines not to use the first teacher data for learning the inference model of the CT when the first correct label based on the first medical information and the second correct label based on the second medical information match. The information processing system according to claim 9, characterized in that.

11. The determination unit determines to use the first teacher data for learning the inference model of the CT when the first correct label based on the first medical information and the second correct label based on the second medical information are different. The information processing system according to claim 9, characterized in that.

12. The determination unit determines not to use the first teacher data for learning the inference model of the CT when the first correct label based on the first medical information and the second correct label, which is the correct label based on the second medical information and used for pre-learning the inference model of the CT, match. The information processing system according to claim 9, characterized in that.

13. The determination unit determines to use the first teacher data for additional learning of the pre-learned inference model of the CT when the first correct label based on the first medical information and the second correct label, which is the correct label based on the second medical information and used for pre-learning the inference model of the CT, are different. The information processing system according to claim 9, characterized in that.

14. An inference target data acquisition unit that acquires medical image data to be inferred, A model acquisition unit that acquires the inference model of the CT generated by the learning by the learning unit, An inference unit that performs a predetermined inference process on the medical image data to be inferred using the learned inference model of the CT, The information processing system according to claim 1, further comprising:

15. The information processing system according to claim 14, further comprising a display control unit that causes the display unit to display information based on the inference result by the inference unit.

16. The inference unit further acquires pixel information that contributed to the predetermined inference, The display control unit displays the pixel information in association with the medical image data to be inferred. The information processing system according to claim 15, characterized in that.

17. The information processing system according to claim 1, wherein the energy integration type radiation detector and the photon counting type radiation detector are different radiation detectors.

18. An inference target data acquisition unit that acquires medical image data to be inferred, A model acquisition unit that acquires a CT inference model learned using teacher data that pairs a correct label based on medical information given to PCCCT image data obtained by using a photon counting type radiation detector for a subject, and CT image data obtained by using an energy integration type radiation detector for the subject, An inference unit that performs a predetermined inference on the medical image data to be inferred using the CT inference model, A display control unit that displays the inference result, An information processing apparatus characterized by comprising:

19. An acquisition step of acquiring, for a subject, first medical information corresponding to PCCCT image data obtained by using a photon counting type radiation detector, and CT image data obtained by using an energy integration type radiation detector for the subject, A teacher data acquisition step of acquiring first teacher data in which a first correct label based on the first medical information is assigned to the CT image data, A learning step of learning a CT inference model that performs a predetermined inference on the CT image data using the first teacher data, An information processing method characterized by comprising:

20. A program for causing a computer to execute the information processing method according to claim 19.

21. An acquisition unit that acquires first image data obtained by using a photon counting type radiation detector for a subject, and second image data obtained by using an energy integration type radiation detector for the subject, A teacher data acquisition unit that acquires first teacher data in which a first correct label based on the first image data is assigned to the second image data, A learning unit that learns an inference model that performs a predetermined inference on the second image data using the first teacher data, An information processing system characterized by comprising:

Images