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

Abstract

To reduce increase in a data storage amount associated with use of a generative model.SOLUTION: An information processing apparatus according to an embodiment includes a prompt generation unit, an acquisition unit, and a storage unit. The prompt generation unit generates a prompt. The acquisition unit inputs the prompt to a generative model and acquires generated information from the generative model. The storage unit stores, as storage target data, the prompt among the generated information and the prompt corresponding to the generated information.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The embodiments disclosed in this specification and the drawings relate to an information processing device, an information processing system, an information processing method, and a program. [Background technology]

[0002] In recent years, the use of generated information generated by generative AI (Artificial Intelligence) has been increasing. Prompt templates are also being stored (for example, Patent Document 1). With this technology, image data and the like can be easily generated by inputting a prompt into the generative AI. In particular, in the medical field, it is common for various pieces of information referenced by doctors for diagnosis to be stored for a certain period of time. However, if all of the generated generated information were to be subject to long-term storage, the amount of data to be stored would increase. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7416390 Summary of the Invention [Problem to be solved by the invention]

[0004] One of the problems to be solved by the embodiments disclosed in this specification and the drawings is to reduce the increase in data storage volume that accompanies the use of generative models. However, the problems to be solved by the embodiments disclosed in this specification and the drawings are not limited to the above problem. Problems corresponding to the effects of each configuration shown in the embodiments described below can also be positioned as other problems. [Means for solving the problem]

[0005] An information processing device according to an embodiment includes a prompt generation unit, an acquisition unit, and a storage unit. The prompt generation unit generates a prompt. The acquisition unit inputs the prompt into a generation model and acquires generation information from the generation model. The storage unit stores the prompt as data to be stored, out of the generation information and the prompt corresponding to the generation information. [Brief explanation of the drawings]

[0006] [Figure 1] FIG. 1 is a diagram illustrating an example of the overall configuration of an information processing system according to the first embodiment. [Figure 2] FIG. 2 is a diagram illustrating an example of input and output of a generative model according to the first embodiment. [Figure 3] FIG. 3 is a diagram showing an example of a display screen according to the first embodiment. [Figure 4] FIG. 4 is a flowchart showing an example of a processing flow regarding generation and storage of generation information according to the first embodiment. [Figure 5] FIG. 5 is a flowchart showing an example of the flow of a process for deleting data to be stored based on a deletion condition according to the first embodiment. [Figure 6] FIG. 6 is a diagram showing an example of a display screen according to the second embodiment. [Figure 7] FIG. 7 is a diagram showing an example of a state in which the user has pressed the save button in FIG. [Figure 8] FIG. 8 is a diagram showing an example of a display screen according to the third embodiment. [Figure 9] FIG. 9 is a diagram illustrating an example of the overall configuration of an information processing system according to the fourth embodiment. [Figure 10] FIG. 10 is a diagram showing an example of the relationship between the version of a generative model and the generation information 410 according to the sixth embodiment. [Figure 11] FIG. 11 is a diagram illustrating an example of the relationship between the version of a generative model and generation information when a prompt is re-entered according to the seventh embodiment. [Figure 12]FIG. 12 is a diagram illustrating an example of determining data to be stored based on the similarity of a plurality of pieces of generation information according to the ninth embodiment. [Figure 13] FIG. 13 is a diagram illustrating another example of determining data to be stored based on the similarity between a plurality of pieces of generation information according to the ninth embodiment. [Figure 14] FIG. 14 is a diagram illustrating an example of data to be stored according to the tenth embodiment. [Figure 15] FIG. 15 is a diagram illustrating an example of the overall configuration of an information processing system according to the eleventh embodiment. [Figure 16] FIG. 16 is a diagram illustrating an example of the hardware configuration of an X-ray CT apparatus according to the eleventh embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0007] Hereinafter, embodiments of an information processing device, an information processing system, an information processing method, and a program will be described in detail with reference to the drawings.

[0008] (First embodiment) Fig. 1 is a diagram showing an example of the overall configuration of an information processing system S according to the first embodiment. As shown in Fig. 1, the information processing system S includes, as an example, a first information processing device 100, a second information processing device 200, and a medical information system 300. The information processing system S is installed in, for example, a medical institution such as a hospital. The first information processing device 100, the second information processing device 200, and the medical information system 300 are communicably connected via a network 400 such as an in-hospital LAN (Local Area Network).

[0009] The medical information system 300 is a system that stores and manages medical information and includes, for example, one or more servers or PCs (Personal Computers). The medical information stored in the medical information system 300 includes, for example, a patient's electronic medical record, clinical information, interview results, test data, medical image data, and diagnosis results. The medical information system 300 may also include the functions of, for example, an electronic medical record system, a hospital information system (HIS), a laboratory information system (LIS), a radiology information system (RIS), or a medical image storage device. The medical image storage device is, for example, a PACS (Picture Archiving and Communication System) server device that stores medical image data in a format compliant with DICOM (Digital Imaging and Communications in Medicine). The medical information system 300 may also include multiple of the above systems and devices.

[0010] The first information processing device 100 and the second information processing device 200 are, for example, computers such as servers or PCs.

[0011] The second information processing device 200 includes a trained model of generative AI (Artificial Intelligence) (hereinafter referred to as generative model 40). The second information processing device 200 includes, for example, a storage circuit and a processor. The generative model 40 is stored in, for example, a storage circuit within the second information processing device 200 and is operated by the processor. Note that the configuration of the second information processing device 200 is not particularly limited, and a known configuration can be adopted.

[0012] When a prompt is input, the generative model 40 generates information corresponding to the prompt and outputs the generated information. The generative model 40 is, for example, an MMM (Multimodal Model) or an LLM (Large Language Model) that can handle learning-type data such as image data, and is an integrated AI model that can process multiple types of modalities (data types) such as text, images, voice, and numbers at once.

[0013] Information generated by the generative model 40 is referred to as generated information. The generated information includes, for example, image data. Note that the generated information may include not only image data but also text. Furthermore, depending on the content of the prompt, the generated information may not include image data. In this embodiment, since the information processing system S is used in the medical field, the generated information 410 includes, for example, information about a patient. The information about a patient is, for example, medical information. Furthermore, the information about a patient may be information about the patient's gender, age, and physique. Furthermore, the generated information is, for example, medical image data. Hereinafter, image data may be simply referred to as "image." Furthermore, medical images are also referred to as medical images.

[0014] The prompt input to the generative model 40 is generally text data. The prompt may also include image data. The prompt includes an instruction specifying the content of the generated information to be generated by the generative model 40. Since the information processing system S of this embodiment is used, for example, in the medical field, the prompt includes medical information. The prompt may also reference external information that is not described in the prompt text. The generative model 40 may search for external information, such as RAG (Retrieval-Augmented Generation), and use it to generate the generated information. Examples of external information to be searched include various types of medical information stored in the medical information system 300.

[0015] The first information processing device 100 is operated by a user and provides the user with the output result of the generative model 40 stored in the second information processing device 200. The user of the first information processing device 100 is, for example, a doctor or medical professional, who uses the generative model 40 via the first information processing device 100 to obtain information to refer to for diagnosing patients. The first information processing device 100 is an example of an information processing device in this embodiment.

[0016] The first information processing device 100 includes a network interface 110, a storage circuit 120, an input interface 130, a display 140, and a processing circuit 150.

[0017] The NW interface 110 is connected to the processing circuit 150, and controls the transmission and communication of various data between the first information processing device 100 and the second information processing device 200 and the medical information system 300. The NW interface 110 is realized by a network card, a network adapter, a NIC (Network Interface Controller), or the like.

[0018] The memory circuitry 120 stores in advance various types of information used by the processing circuitry 150. The memory circuitry 120 also stores various programs. The memory circuitry 120 is, for example, a non-volatile memory device such as an HDD (Hard Disk Drive), an SSD (Solid State Drive), or an integrated circuit memory device that stores various types of information. In addition to an HDD or SSD, the memory circuitry 120 may also be a drive device that reads and writes various types of information from / to portable storage media such as a CD (Compact Disc), a DVD (Digital Versatile Disc), or a flash memory, or a semiconductor memory element such as a RAM (Random Access Memory). The memory circuitry 120 is an example of a memory unit.

[0019] Furthermore, the memory circuitry 120 stores the generation information generated by the generative model 40 and the prompt corresponding to the generation information as data to be stored. The prompt corresponding to the generation information refers to a prompt input to the prompt for generating the generation information. In this embodiment, the memory circuitry 120 stores at least the prompt as data to be stored. Furthermore, when the generation information is designated as data to be stored by a user operation, the memory circuitry 120 may store the generation information as data to be stored.

[0020] In this embodiment, the data to be stored is not data that is temporarily stored during processing, but data that is to be stored long-term in the memory circuitry 120. In this embodiment, "storing as data to be stored" refers to storing data with the assumption that it will be stored long-term. Furthermore, simply "storing" includes both temporary storage and long-term storage. Furthermore, temporary storage during work until a doctor's diagnosis is confirmed is also different from storage as data to be stored. Data to be stored is also called data to be stored long-term.

[0021] The storage period for long-term storage is, for example, five years if the generated information is medical information used to diagnose a patient and is linked to an electronic medical record, but this period is an example and is not limited to this. Data to be stored may be stored for five years or more.

[0022] Furthermore, unlike cache data, which may be automatically erased or overwritten, the data to be stored in the storage circuitry 120 is not deleted unless a specified deletion condition is met or a deletion operation is performed by the user. Details of the prompts to the storage circuitry 120 and the saving and deletion of generated information will be described later.

[0023] The input interface 130 may be implemented by a mouse, a keyboard, a pen tablet combining a touch pen and a tablet for accepting user operations, a trackball, switch buttons, a touchpad for performing input operations by touching the operation surface, a touchscreen integrating a display screen and a touchpad, a non-contact input circuit using an optical sensor, and a voice input circuit. The input interface 130 may include multiple devices for accepting user operations. The input interface 130 is connected to the processing circuit 150 and converts input operations received from the user into electrical signals and outputs them to the processing circuit 150. Note that, in this specification, the input interface is not limited to those having physical operating components such as a mouse and a keyboard. For example, an electrical signal processing circuit that receives electrical signals corresponding to input operations from an external input device provided separately from the device and outputs these electrical signals to the processing circuit 150 is also included as an example of an input interface.

[0024] The display 140 displays various types of information under the control of the processing circuit 150. For example, the display 140 outputs a screen including the generation information generated by the generative model 40, a GUI (Graphical User Interface) for receiving various operations from the user, etc. Specifically, the display 140 is a liquid crystal display, a CRT (Cathode Ray Tube) display, etc. Note that the input interface 130 and the display 140 may be integrated. For example, the input interface 130 and the display 140 may be realized by a touch panel. The display 140 is an example of a display unit.

[0025] The processing circuitry 150 is a processor that reads out programs from the storage circuitry 120 and executes them to realize functions corresponding to each program. The processing circuitry 150 of this embodiment includes a reception function 151, a prompt generation function 152, a first acquisition function 153, a second acquisition function 154, a storage processing function 155, a display control function 156, and a deletion processing function 157. The reception function 151 is an example of a reception unit. The prompt generation function 152 is an example of a prompt generation unit. The first acquisition function 153 is an example of a first acquisition unit or an acquisition unit. The second acquisition function 154 is an example of a second acquisition unit or an acquisition unit. The storage processing function 155 is an example of a storage processing unit. The display control function 156 is an example of a display control unit. The deletion processing function 157 is an example of a deletion processing unit.

[0026] Here, for example, each of the processing functions of the processing circuitry 150, namely, the reception function 151, the prompt generation function 152, the first acquisition function 153, the second acquisition function 154, the storage processing function 155, the display control function 156, and the deletion processing function 157, is stored in the storage circuitry 120 in the form of a program executable by a computer. The processing circuitry 150 is a processor. For example, the processing circuitry 150 realizes the function corresponding to each program by reading the program from the storage circuitry 120 and executing it. In other words, the processing circuitry 150 in a state in which each program has been read has each of the functions shown in the processing circuitry 150 of FIG. 1. 1 has been described as realizing the processing functions performed by the reception function 151, prompt generation function 152, first acquisition function 153, second acquisition function 154, storage processing function 155, display control function 156, and deletion processing function 157 in a single processor, but it is also possible to combine multiple independent processors to configure the processing circuit 150 and have each processor execute a program to realize the function. Also, while it is described in FIG. 1 as realizing the single storage circuit 120 storing programs corresponding to each processing function, it is also possible to configure multiple storage circuits to be distributed and have the processing circuit 150 read out corresponding programs from individual storage circuits.

[0027] In the above description, an example has been described in which a "processor" reads and executes a program corresponding to each function from a storage circuit. However, the embodiment is not limited to this. The term "processor" refers to a circuit such as a central processing unit (CPU), a graphics processing unit (GPU), an application-specific integrated circuit (ASIC), or a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), or a field programmable gate array (FPGA)). If the processor is a CPU, for example, the processor realizes its function by reading and executing a program stored in a storage circuit. On the other hand, if the processor is an ASIC, instead of storing a program in the storage circuit 120, the function is directly incorporated into the processor circuit as a logic circuit. Note that each processor in this embodiment is not limited to being configured as a single circuit per processor, but may be configured as a single processor by combining multiple independent circuits to realize its function. Furthermore, multiple components in FIG. 1 may be integrated into a single processor to realize its function.

[0028] The reception function 151 receives various operations from the user via the input interface 130. For example, the reception function 151 receives a prompt input operation for inputting data to the generation model 40 by the user, an editing operation, and an operation for specifying data to be stored. For example, the reception function 151 receives a user operation for specifying the generation information 410 as data to be stored.

[0029] The prompt generation function 152 generates a prompt to be input to the generation model 40. For example, the prompt generation function 152 generates a prompt based on a user operation accepted by the acceptance function 151. The prompt generation function 152 may automatically generate some or all of the prompt in accordance with a predetermined rule. The prompt generation function 152 may also have a function for checking the description of the prompt 50 entered by the user.

[0030] 2 is a diagram showing an example of input and output of the generative model 40 according to the first embodiment. As shown in Fig. 2, the prompt 50 includes, for example, text data input by the user, such as "Please generate and output similar case images based on the medical information of the subject," and information indicating the storage location of the referenced medical information.

[0031] The medical information of the subject (patient) may include, but is not limited to, the patient's electronic medical record, medical information, interview results, test data, medical image data, and diagnosis results. The storage location of the referenced medical information may be a path indicating a storage location within the medical information system 300 or a path indicating a storage location within the memory circuitry 120 of the first information processing device 100. The medical information may also be directly written in the prompt 50. The types of medical image data included in the medical information include, but are not limited to, X-ray image data, X-ray CT (Computed Tomography) image data, magnetic resonance image data, and ultrasound image data. Note that while FIG. 2 uses the term "subject's medical information" for the sake of explanation, the prompt 50 may actually include identification information, such as a patient ID, that identifies the subject.

[0032] It should be noted that the prompt 50 may include image data as well as text.

[0033] The content of the prompt 50 shown in FIG. 2 is merely an example and is not limited thereto. As another specific example, the prompt 50 may include medical image data of a patient photographed without contrast or the path to its storage location, and text data instructing the user to generate medical image data of contrast-enhanced image data based on the medical image data. Alternatively, the prompt 50 may include medical image data including the heart in its imaging range or the path to its storage location, and text data instructing the user to generate medical image data in a phase different from the phase in which the medical image data was actually photographed. More specifically, the prompt 50 may include medical image data of the heart photographed in diastole (ED) or the path to its storage location, and text data instructing the user to generate image data depicting the heart in systole (ES) based on the medical image data. The prompt 50 may also include medical image data of a patient's chest photographed or the path to its storage location, and text data instructing the user to generate medical image data at a different respiratory timing than the medical image data. The prompt 50 may also include medical image data of a patient photographed under certain imaging conditions or the path to its storage location, and text data instructing the user to generate medical image data under different imaging conditions than the medical image data. The prompt 50 may also include medical image data taken of a patient using a certain modality or the path to where it is saved, and text data instructing the generation of medical image data equivalent to the image taken using a modality different from the medical image data.

[0034] In this way, the generative model 40 generates image data without actually taking images, which reduces the patient's radiation exposure compared to, for example, performing multiple scans using X-ray CT. Also, compared to imaging using modalities that do not involve radiation exposure, there are advantages such as a shorter imaging time and a reduced burden on the patient.

[0035] Returning to FIG. 1, the first acquisition function 153 acquires medical information from the medical information system 300 via the network 400 and the NW interface 110. For example, the first acquisition function 153 may acquire medical information to be used in the prompt 50 from the medical information system 300 based on a user operation. Alternatively, the first acquisition function 153 may acquire predetermined medical information from the medical information system 300 and store it in the memory circuitry 120. The first acquisition function 153 may also be referred to as a medical information acquisition function (medical information acquisition unit).

[0036] The medical information acquired by the first acquisition function 153 is used by being included in or referenced in a prompt 50, as shown in Fig. 2. In the example shown in Fig. 2, medical information 21 related to a patient who is the subject of generation of a similar case image is used.

[0037] 1, the second acquisition function 154 inputs the prompt 50 generated by the prompt generation function 152 into the generative model 40 and acquires generation information 410 from the generative model 40. The second acquisition function 154 may also be referred to as a generative model control function (generative model control unit).

[0038] In the example shown in Fig. 2, when the generative model 40 receives a prompt 50, it outputs generated information 410. The prompt 50 in Fig. 2 includes the command "Please generate and output similar case images based on the medical information of the subject," and therefore the generated information 410 is similar case images. Note that in this embodiment, image data may be simply referred to as "images," as in the similar case images in Fig. 2.

[0039] Furthermore, the second acquisition function 154 restores (reproduces) the generated information 410 by re-inputting the prompt 50 stored in the memory circuitry 120 as data to be stored into the generative model 40 based on the user's operation received by the reception function 151. For example, assume that the prompt 50 generated by the generative model 40 is stored in the memory circuitry 120 as data to be stored by the memory processing function 155 (described later), and the generated information 410 is not stored as data to be stored. If the user subsequently wishes to check the generated information 410, the user executes an operation to re-input the prompt 50 stored in the memory circuitry 120 as data to be stored into the generative model 40. As a result, the generative model 40 re-generates the generated information 410 from the prompt 50 stored in the memory circuitry 120 as data to be stored. This re-generation of the generated information 410 using the same prompt 50 is referred to as restoration or reproduction of the generated information 410. Note that generated information 410 restored based on prompt 50 stored as data to be stored in storage circuitry 120 is referred to as "restored generated information" to distinguish it from the originally generated generated information 410.

[0040] 1 , the storage processing function 155 stores the prompt 50 of the generation information 410 and the prompt 50 corresponding to the generation information 410 as data to be stored in the storage circuitry 120. For example, when the prompt 50 is input to the generative model 40, the storage processing function 155 stores the prompt 50 as data to be stored in the storage circuitry 120. In other words, the storage processing function 155 automatically stores the prompt 50 used to generate the generation information 410 for a long period of time.

[0041] Furthermore, when the generated information 410 is designated as data to be stored by a user operation, the storage processing function 155 stores the generated information 410 as data to be stored in the storage circuitry 120. The operation of designating the data to be stored by the user is performed, for example, on a display screen displayed on the display 140 by the display control function 156 described later.

[0042] For example, the storage processing function 155 may store all prompts 50 input to the generative model 40 as data to be stored in the storage circuitry 120. Alternatively, for example, if a user performs trial and error, such as inputting a prompt 50 into the generative model 40 multiple times while editing the prompt 50, the storage processing function 155 may store the prompt 50 last input to the generative model 40 as data to be stored in the storage circuitry 120. Regarding the prompt 50, the storage processing function 155 automatically stores it in the storage circuitry 120 as data to be stored, regardless of the user's operation. In other words, of the generation information 410 and the prompt 50 corresponding to the generation information 410, the storage processing function 155 preferentially stores the prompt 50 in the storage circuitry 120 as data to be stored.

[0043] Furthermore, the storage processing function 155 stores the generated information 410 in the storage circuitry 120 as data to be stored in response to a user operation specifying the generated information 410 as data to be stored. Furthermore, if the user operation specifying the generated information 410 as data to be stored is not accepted, the storage processing function 155 does not store the generated information 410 in the storage circuitry 120 as data to be stored. In other words, the storage processing function 155 stores the generated information 410 in the storage circuitry 120 as data to be stored only when the user operation specifying the generated information 410 as data to be stored is accepted. Note that temporary storage associated with processing such as displaying the generated information 410 on the display 140 is not storage as data to be stored, and is therefore not particularly limited.

[0044] Even when the generation information 410 is not stored in the memory circuitry 120 as data to be stored, the prompt 50 is stored in the memory circuitry 120 as data to be stored, so that the user can re-obtain the generation information 410 as needed by re-inputting the prompt 50 into the generative model 40. Therefore, in this embodiment, it is not essential to set the generation information 410 as data to be stored, and the storage processing function 155 sets the generation information 410 as data to be stored only when instructed by the user.

[0045] The display control function 156 displays the generated information 410 on the display 140. The display control function 156 also displays an image button on the display 140 that accepts a user operation to specify whether the generated information 410 and / or the prompt 50 corresponding to the generated information 410 are to be stored.

[0046] 3 is a diagram showing an example of a display screen 141a according to the first embodiment. The display screen 141a is displayed on the display 140 by the display control function 156.

[0047] 3, the display screen 141a includes, for example, a medical image 501 of the subject referenced in the prompt 50, the generated information 410, a store button 61a, and a display prompt button 62. The display screen 141a may also include a back button 63, a done button 64, etc., as shown in FIG.

[0048] 3, the display screen 141a includes a first screen area 142 that displays the medical image 501, and a second screen area 143 that displays the generation information 410, a store button 61a, a prompt display button 62, a back button 63, and a complete button 64. Note that the screen layout of the display screen 141a is not limited to the example shown in FIG.

[0049] In FIG. 3, the generated information 410 is a similar case image, but the content of the generated information 410 changes depending on the content of the prompt 50, and is therefore not limited to the example of FIG.

[0050] The save button 61a, the prompt display button 62, the back button 63, and the done button 64 are image buttons that can accept user operations.

[0051] The storage button 61a accepts a user operation to designate the generated information 410 as data to be stored. For example, when the user operates the input interface 130, such as a mouse, to press the storage button 61a displayed on the display 140, the above-described reception function 151 accepts the operation. In this case, the storage processing function 155 stores the generated information 410 displayed on the display screen 141a in the storage circuitry 120 as data to be stored.

[0052] Furthermore, if the user does not press the store button 61a, the storage processing function 155 does not store the generation information 410 in the storage circuitry 120 as data to be stored.

[0053] The prompt display button 62 is operated by a user to display the prompt 50 used to generate the generation information 410 displayed on the display screen 141a. For example, the display control function 156 does not initially display the prompt 50 on the display screen 141a, but when the prompt display button 62 is pressed by the user, displays a prompt display field 65 as shown in FIG. 3. The prompt 50 is displayed in the prompt display field 65. The prompt display field 65 may be provided within the display screen 141a, or may be displayed as a separate screen superimposed on the display screen 141a.

[0054] The prompt display field 65 may allow the user to edit the prompt 50 .

[0055] Furthermore, the back button 63 accepts a user operation to return from the display screen 141a to the input screen for the prompt 50. For example, when the user presses the back button 63, the display control function 156 transitions the screen to the input screen for the prompt 50. On the input screen for the prompt 50, it is possible to edit the prompt 50 and input the edited prompt 50 into the generation model 40. Furthermore, when the user inputs the edited prompt 50 into the generation model 40, the display control function 156 displays the generation information 410 generated based on the edited prompt 50 on the display screen 141a.

[0056] The done button 64 accepts an operation by the user to close the display screen 141a.

[0057] When the user presses the back button 63, the storage processing function 155 may not store the prompt 50 in the storage circuitry 120 as data to be stored. For example, the user may press the back button 63 to return to the input screen for the prompt 50, edit the prompt 50 again, and re-input it into the generative model 40 until the desired generation information 410 is obtained. For this reason, when the user presses the back button 63, the storage processing function 155 may not store the prompt 50 in the storage circuitry 120 as data to be stored, but when the user finishes viewing the display screen 141a and presses the done button 64, the storage processing function 155 may store the latest prompt 50 at the time of pressing the done button 64 in the storage circuitry 120 as data to be stored. Furthermore, when the user presses the store button 61a, the storage processing function 155 may store the prompt 50 corresponding to the generation information 410 displayed on the display screen 141a in the storage circuitry 120 as data to be stored. Alternatively, the storage processing function 155 may store all prompts 50 input to the generative model 40 in the storage circuitry 120 as data to be stored.

[0058] 1, when a specified deletion condition is met or when a deletion operation is performed by the user, the deletion processing function 157 deletes the data to be stored in the storage circuitry 120. In this embodiment, only the generated information 410 is deleted based on the deletion condition, and the prompt 50 is not deleted.

[0059] The specified deletion conditions include, for example, the passage of a specified period of time since the data to be stored was stored, the remaining data capacity of the storage area for the data to be stored in the storage circuit 120, and the absence of user browsing for a specified period of time. User browsing of the generated information 410 may be counted, for example, based on the user's operation log or access log. The length of the specified period may be, for example, about one year, or may be arbitrarily determined by the administrator. Note that the specified deletion conditions are not limited to the above examples.

[0060] The deletion processing function 157 may determine the stored data to be deleted based on a combination of multiple deletion conditions. The multiple deletion conditions may be set by logical operations. The deletion conditions may be determined based on a predefined conditional expression using various logical operators (logical sum (OR operation), logical product (AND operation), exclusive logical sum (XOR operation), and negation (NOT operation)), or various logical operators may be presented to the user to allow them to set a conditional expression corresponding to the deletion condition. Alternatively, there may be only one deletion condition. The deletion condition may be set, for example, by an administrator of the first information processing device 100.

[0061] Next, the flow of processing executed by the first information processing device 100 configured as above will be described.

[0062] 4 is a flowchart showing an example of a process flow related to generation and storage of generation information 410 according to the first embodiment. The process of this flowchart is executed, for example, when the user performs an operation to open the input screen for prompt 50.

[0063] First, the display control function 156 causes the display 140 to display a prompt input screen (S1).

[0064] Then, the reception function 151 receives the input of the prompt 50 by the user (S2).

[0065] Also, if the user specifies medical information 21 to be acquired for use in creating the prompt 50 (S3 “Yes”), the first acquisition function 153 acquires the specified medical information 21 from the medical information system 300 (S4).

[0066] In addition, the user may generate the prompt 50 without using the medical information 21 in the medical information system 300. If the user does not specify the medical information 21 to be acquired to be used in creating the prompt 50 (S3 "No"), the process of S4 is not executed and the process proceeds to S5. In addition, if the generative model 40 of the second information processing device 200 can acquire the specified medical information 21 from the medical information system 300, the prompt 50 may include a path indicating the storage location within the medical information system 300 rather than the medical information 21 itself. In this case, the first acquisition function 153 may not need to download the medical information 21 from the medical information system 300 to the first information processing device 100.

[0067] Then, when the user has completed inputting the prompt 50 (S5 "Yes"), the prompt generation function 152 generates the prompt 50 based on the user's input. Then, the second acquisition function 154 inputs the generated prompt 50 to the generative model 40 (S6). Based on the input prompt 50, the generative model 40 generates generation information 410. Note that while the user has not completed inputting the prompt 50 (S5 "No"), the process of S6 is not executed and remains on standby.

[0068] Then, the second acquisition function 154 acquires the generation information 410 output from the generative model 40 (S7).

[0069] 3, the display control function 156 causes the display 140 to display the display screen 141a including the acquired generation information 410 (S8). Based on various operations performed by the user on the display screen 141a, the processes of S9 to S15 are executed.

[0070] For example, if the user presses the back button 63 on the display screen 141a (S9 "Yes"), the display control function 156 transitions the display 140 to a prompt input screen and returns to the processing of S1. Note that if S1 is executed two or more times, the input screen may be displayed with the previously entered prompt 50 displayed. If the back button 63 is not pressed (S9 "No"), the display control function 156 continues to display the display screen 141a.

[0071] Furthermore, if the user presses the prompt display button 62 (S10 "Yes"), the display control function 156 displays the prompt display field 65 including the prompt 50 (S11). If the prompt display button 62 is not pressed (S10 "No"), the display control function 156 does not display the prompt 50.

[0072] Furthermore, if the user presses the storage button 61a (S12 "Yes"), the storage processing function 155 stores the generation information 410 displayed on the display screen 141a in the storage circuitry 120 as data to be stored (S13). If the storage button 61a is not pressed (S12 "No"), the storage processing function 155 does not store the generation information 410 in the storage circuitry 120 at least as data to be stored.

[0073] Even if the user presses the back button 63 after pressing the store button 61a, the storage processing function 155 stores the prompt 50 and the generated information 410 in the storage circuitry 120 as data to be stored.

[0074] The processes of S9 to S14 are repeated until the user presses the complete button 64 (S14 "No").

[0075] Furthermore, if the user presses the complete button 64 (S14 "Yes"), the storage processing function 155 stores the prompt 50 corresponding to the generation information 410 displayed on the display screen 141a in the storage circuitry 120 as data to be stored (S15). Furthermore, the display control function 156 may close the display screen 141a. At this point, the processing of this flowchart ends.

[0076] Next, the flow of the process of deleting data to be stored, which is executed by the first information processing device 100, will be described.

[0077] 5 is a flowchart showing an example of the flow of a process for deleting data to be stored based on a deletion condition according to the first embodiment. The process of this flowchart may be, for example, a scheduled periodic execution process, or may be executed by a user operation.

[0078] First, the deletion processing function 157 determines whether the remaining data capacity of the storage area for the data to be stored in the storage circuit 120 is less than a specified value. If the remaining data capacity of the storage area for the data to be stored in the storage circuit 120 is equal to or greater than the specified value (S101 "No"), the deletion processing is not executed, and the processing of this flowchart ends.

[0079] Furthermore, if the remaining data capacity of the storage area for data to be stored in the storage circuit 120 is less than a specified value (S101 "Yes"), the deletion processing function 157 determines whether or not there is any generated information 410 among the data to be stored stored in the storage circuit 120 that has been stored for more than five years. If there is no generated information 410 that has been stored for more than five years, the deletion processing is not executed and the processing of this flowchart ends (S102 "No"). In this case, the deletion processing function 157 may output a warning indicating that the remaining data capacity of the storage area for data to be stored is getting low.

[0080] Then, if there is any generated information 410 that has not been viewed by a user for a specified period of time among the generated information 410 that has been saved for more than five years (S103 "Yes"), the deletion processing function 157 deletes the generated information 410 as generated information 410 that satisfies the deletion condition (S104). "Not viewed by a user for a specified period of time" means that the generated information 410 has not been viewed by a user for a specified period of time from the time of the processing in FIG. 5.

[0081] If there is no generation information 410 that has not been viewed by a user for a specified period of time among the generation information 410 that has been saved for more than five years (S103 "No"), the deletion process is not executed and the process of this flowchart ends (S102 "No"). Even in this case, the deletion process function 157 may output a warning or the like indicating that the remaining data capacity of the storage area for data to be stored is getting low. Here, the process of this flowchart ends.

[0082] In the flowchart of Figure 5, the generated information 410 is subject to deletion when it meets all of the deletion conditions: "the remaining data capacity of the memory area for the data to be stored in the memory circuit 120 is less than a specified value," "more than five years have passed since storage," and "the data has not been viewed by a user for a specified period of time." However, the content and combination of the deletion conditions are not limited to these.

[0083] In addition to the process based on the deletion conditions shown in FIG. 5, any data to be stored may be deleted based on the operation of a user with administrator authority, for example.

[0084] As described above, the first information processing apparatus 100 of this embodiment inputs the generated prompt 50 to the generative model 40 and acquires the generation information 410 from the generative model 40. The first information processing apparatus 100 then stores the prompt 50 from the acquired generation information 410 and the prompt 50 corresponding to the generation information 410 in the storage circuitry 120 as data to be stored. Therefore, the first information processing apparatus 100 of this embodiment can reduce an increase in the amount of data stored due to the use of the generative model 40, compared to when all of the generated generation information 410 is subject to long-term storage. Furthermore, by storing the prompt 50 in the storage circuitry 120 as data to be stored, the first information processing apparatus 100 can re-obtain the generation information 410 even if it has not been saved, for example, by the user re-inputting the prompt 50 into the generative model 40 as needed.

[0085] Furthermore, when a prompt 50 is input to the generative model 40, the first information processing device 100 of this embodiment stores the prompt 50 as data to be stored in the memory circuitry 120. Therefore, according to the first information processing device 100 of this embodiment, the prompt 50 used to generate the generative information 410 is automatically stored for a long period of time, thereby reducing the occurrence of unintentional omission of the prompt 50 from being stored.

[0086] Furthermore, the first information processing device 100 of this embodiment accepts a user operation to designate the generated information 410 as data to be stored, and stores the generated information 410 in the storage circuitry 120 as data to be stored in response to the operation. Furthermore, the first information processing device 100 of this embodiment stores the prompt 50 in the storage circuitry 120 as data to be stored regardless of the user operation. Therefore, according to the first information processing device 100 of this embodiment, an increase in the amount of data to be saved can be reduced by basically designating only the prompt 50 as a storage target, and the generated information 410 can also be stored in the storage circuitry 120 as data to be stored when explicitly instructed by the user.

[0087] In addition, the first information processing device 100 of this embodiment displays a storage button 61a, which is an image button that can accept a user's operation to designate the generated information 410 as a storage target, on the display 140 together with the generated information 410, so that the user can operate the storage button 61a while checking the target generated information 410.

[0088] Furthermore, the first information processing device 100 of this embodiment includes a storage circuitry 120, which is a non-volatile storage device, and the prompt 50 and the generated information 410 stored in the storage circuitry 120 as data to be stored are not deleted unless a specified deletion condition is met or a deletion operation is performed by the user. Therefore, the first information processing device 100 of this embodiment can appropriately store data that requires long-term storage in the business in which the first information processing device 100 is used. Furthermore, the data to be stored is data that is stored long-term by the first information processing device 100, and the first information processing device 100 of this embodiment can reduce the increase in the amount of data stored that accompanies the use of the generative model 40 by limiting the target data to be stored as described above.

[0089] Furthermore, the generation information 410 that the first information processing device 100 of this embodiment causes the generative model 40 to generate includes image data. Generally, when the generation information 410 includes image data and the prompt 50 is text, the data volume of the generation information 410 is significantly larger than the data volume of the prompt 50. For this reason, by storing the prompt 50 as data to be stored in the memory circuitry 120 with priority, as in this embodiment, of the generation information 410 and the prompt 50, it is possible to effectively reduce the increase in the amount of data stored.

[0090] (Second embodiment) In the first embodiment described above, the first information processing device 100 stores the prompt 50 in the storage circuitry 120 as data to be stored regardless of the user's operation, and determines whether or not to designate the generated information 410 as data to be stored in response to the user's operation only. In contrast, in the present embodiment, it determines whether or not to designate both the prompt 50 and the generated information 410 as data to be stored in response to the user's operation.

[0091] The information processing system S of this embodiment includes, for example, a first information processing device 100, a second information processing device 200, and a medical information system 300, similar to the first embodiment described in Fig. 1. Also, the first information processing device 100 of this embodiment includes a NW interface 110, a memory circuitry 120, an input interface 130, a display 140, and a processing circuitry 150, similar to the first embodiment. Note that the same components as those in the first embodiment are denoted by the same reference numerals, and description thereof will be omitted.

[0092] The processing circuit 150 of this embodiment has a reception function 151, a prompt generation function 152, a first acquisition function 153, a second acquisition function 154, a storage processing function 155, a display control function 156, and a deletion processing function 157, similar to the first embodiment.

[0093] The reception function 151, the prompt generation function 152, the first acquisition function 153, the second acquisition function 154, and the deletion processing function 157 have the same functions as those in the first embodiment.

[0094] In addition to the functions of the first embodiment, the display control function 156 of this embodiment displays an image button on the display 140 that accepts a user operation to specify whether either or both of the generated information 410 and the prompt 50 corresponding to the generated information 410 are to be stored.

[0095] The storage processing function 155 of this embodiment stores, in response to a user operation, either or both of the generated information 410 and the prompt 50 corresponding to the generated information 410 in the storage circuitry 120 as data to be stored.

[0096] 6 is a diagram showing an example of a display screen 141b according to the second embodiment. For example, as shown in FIG. 6, the display screen 141b includes a medical image 501 of the subject referenced in the prompt 50, generated information 410, a store button 61b, a back button 63, and a done button 64. The display screen 141b may also include a prompt display button 62, similar to the display screen 141a of the first embodiment. The back button 63 and the done button 64 have the same functions as those in the first embodiment.

[0097] When the user presses the storage button 61b, the display control function 156 displays an image button on the display 140 that accepts the user's operation to specify whether to store either or both of the generated information 410 and the prompt 50 corresponding to the generated information 410.

[0098] Fig. 7 is a diagram showing an example of a state in which the user has pressed the store button 61b in Fig. 6. When the user has pressed the store button 61b, the display control function 156 causes a prompt store button 611 and a generation information store button 612 to be displayed on the display 140, as shown in Fig. 7. The prompt store button 611 and the generation information store button 612 may be displayed within the display screen 141b, or may be displayed on a separate screen superimposed on the display screen 141b.

[0099] When the user presses the prompt storage button 611, the storage processing function 155 stores the prompt 50 corresponding to the generation information 410 displayed on the display screen 141b as data to be stored in the storage circuitry 120. When the user does not press the prompt storage button 611, the storage processing function 155 does not store the prompt 50 corresponding to the generation information 410 displayed on the display screen 141b as data to be stored in the storage circuitry 120.

[0100] Furthermore, when the user presses the generation information storage button 612, the storage processing function 155 stores the generation information 410 displayed on the display screen 141b as data to be stored in the storage circuitry 120. Furthermore, when the user does not press the generation information storage button 612, the storage processing function 155 does not store the generation information 410 displayed on the display screen 141b as data to be stored in the storage circuitry 120.

[0101] In other words, the prompt store button 611 and the generation information store button 612 accept a user operation to select data to be selected. Therefore, the prompt store button 611 and the generation information store button 612 are examples of a selection unit. Note that the method of selection by the user is not limited to the prompt store button 611 and the generation information store button 612, and for example, other operation images such as a list box or an operation by a command may be adopted.

[0102] As described above, the first information processing device 100 of this embodiment displays an image button on the display 140 that accepts a user operation to specify whether to store either or both of the generated information 410 and the prompt 50 corresponding to the generated information 410, and stores either or both of the generated information 410 and the prompt 50 corresponding to the generated information 410 as data to be stored in the storage circuitry 120 in accordance with the user operation. Therefore, according to the first information processing device 100 of this embodiment, in addition to the effects of the first embodiment, it is possible to further reduce an increase in the amount of data stored.

[0103] (Third embodiment) In the above-described first and second embodiments, an example was described in which one piece of generation information 410 is output from the generation model 40. In this third embodiment, when a plurality of pieces of generation information 410 are output from the generation model 40, a user operation to designate each piece of generation information 410 as a storage target is accepted for each piece of generation information 410 output from the generation model 40.

[0104] The information processing system S of this embodiment includes, for example, a first information processing device 100, a second information processing device 200, and a medical information system 300, similar to the first embodiment described in FIG. 1. The generative model 40 of the second information processing device 200 of this embodiment generates and outputs multiple pieces of generated information 410. The generated information 410 of this embodiment includes image data. Also, similar to the first embodiment, the first information processing device 100 of this embodiment includes a NW interface 110, a memory circuitry 120, an input interface 130, a display 140, and a processing circuitry 150. Note that the same components as those of the first embodiment are denoted by the same reference numerals, and descriptions thereof will be omitted.

[0105] The processing circuit 150 of this embodiment has a reception function 151, a prompt generation function 152, a first acquisition function 153, a second acquisition function 154, a storage processing function 155, a display control function 156, and a deletion processing function 157, similar to the first embodiment.

[0106] The reception function 151, the prompt generation function 152, the first acquisition function 153, the second acquisition function 154, and the deletion processing function 157 have the same functions as those in the first embodiment.

[0107] In addition to the functions of the first embodiment, the display control function 156 of this embodiment displays an image button on the display 140 for each piece of generation information 410 output from the generation model 40, which accepts a user operation to designate the generation information 410 as a storage target.

[0108] More specifically, the generative model 40 of this embodiment generates and outputs multiple image data based on the input prompt 50. The display control function 156 displays, for each of the multiple image data output from the generative model 40, an image button on the display 140 that accepts a user operation to select the image data as a storage target.

[0109] FIG. 8 is a diagram illustrating an example of a display screen 141c according to the third embodiment. For example, as illustrated in FIG. 8, the display screen 141c includes a medical image 501 of a subject referenced in the prompt 50, multiple pieces of generated information 410a and 410b, a prompt storage button 620, a generated information A storage button 621a, a generated information B storage button 621b, a back button 63, and a complete button 64. The prompt storage button 620 does not necessarily have to be displayed. When the user selects the generated information A storage button 621a or the generated information B storage button 621b, the storage processing function 155 may store the corresponding prompt 50 in the storage circuitry 120, as in the first embodiment. Furthermore, the storage processing function 155 may be controlled to store the prompt 50 in the storage circuitry 120 when the user selects the complete button 64, regardless of whether the generated information A storage button 621a or the generated information B storage button 621b is pressed. The display screen 141c may also include a prompt display button 62, similar to the display screen 141a of the first embodiment. The back button 63 and the done button 64 have the same functions as those in the first embodiment. The prompt storage button 620 has the same function as the prompt storage button 611 of the second embodiment.

[0110] The plurality of pieces of generated information 410a and 410b are both image data (images A and B).

[0111] The generated information A storage button 621a is an image button corresponding to the generated information 410a (generated information A (image A)). The generated information B storage button 621b is an image button corresponding to the generated information 410b (generated information B (image B)). Hereinafter, when there is no need to distinguish between the generated information A storage button 621a and the generated information B storage button 621b, they will simply be referred to as the generated information storage button 621. The generated information storage button 621 is an example of a selection section.

[0112] 8 shows an example in which two pieces of generation information 410a, 410b are output from the generative model 40, but the number of pieces of generation information 410 output may be three or more. The display control function 156 displays multiple generation information storage buttons 621 corresponding to each piece of output image data according to the number of pieces of image data output from the generative model 40. In other words, the display control function 156 dynamically generates the generation information storage buttons 621 according to the number of pieces of image data output from the generative model 40.

[0113] When the user presses the generation information A storage button 621a, the storage processing function 155 stores the generation information 410a displayed on the display screen 141c as data to be stored in the storage circuitry 120. When the user does not press the generation information A storage button 621a, the storage processing function 155 does not store the generation information 410a as data to be stored in the storage circuitry 120.

[0114] Furthermore, when the user presses the generation information B storage button 621b, the storage processing function 155 stores the generation information 410b displayed on the display screen 141c as data to be stored in the storage circuitry 120. Furthermore, when the user does not press the generation information B storage button 621b, the storage processing function 155 does not store the generation information 410b as data to be stored in the storage circuitry 120.

[0115] The method of selection by the user is not limited to the prompt storage button 620 and the generated information storage button 621, but other operation images such as a list box or an operation by a command may be adopted.

[0116] In this way, the first information processing device 100 of this embodiment displays on the display 140, for each piece of image data output from the generative model 40, an image button that accepts a user operation to designate the image data as a storage target. Therefore, according to the first information processing device 100 of this embodiment, when the generation information 410 includes multiple pieces of image data, the user can individually designate whether or not to store the pieces of image data for a long period of time, thereby effectively reducing the increase in the amount of data stored compared to when all image data are uniformly designated as storage targets.

[0117] (Fourth embodiment) In the first to third embodiments described above, the first information processing device 100 determines the data to be stored in response to a user operation. In this fourth embodiment, the first information processing device 100 evaluates the necessity of storing the generated information 410, and determines whether or not to set the generated information 410 as data to be stored based on the evaluation result.

[0118] 9 is a diagram showing an example of the overall configuration of an information processing system S according to the fourth embodiment. The information processing system S according to this embodiment includes, for example, a first information processing device 100, a second information processing device 200, and a medical information system 300, similar to the first embodiment. The first information processing device 100 according to this embodiment also includes a NW interface 110, a storage circuitry 120, an input interface 130, a display 140, and a processing circuitry 150, similar to the first embodiment. Note that the same components as those in the first embodiment are denoted by the same reference numerals, and description thereof will be omitted.

[0119] The processing circuit 150 of this embodiment includes a reception function 151, a prompt generation function 152, a first acquisition function 153, a second acquisition function 154, a storage processing function 155, a display control function 156, a deletion processing function 157, an evaluation function 158, and a determination function 159. The evaluation function 158 is an example of an evaluation unit. The determination function 159 is an example of a determination unit.

[0120] The reception function 151, the prompt generation function 152, the first acquisition function 153, the second acquisition function 154, and the display control function 156 have the same functions as those in the first embodiment.

[0121] The evaluation function 158 evaluates the necessity of storing the generated information 410. The necessity of storing refers to the necessity of storing the generated information 410 for a long period of time in the memory circuit 120 as data to be stored. For example, the necessity of storing increases as the importance of the generated information 410 increases. In this embodiment, the information processing system S is used in the medical field, and the generated information 410 includes medical information about patients, so the evaluation function 158 evaluates the clinical importance of the generated information 410. Note that the evaluation criteria for the necessity of storing may differ depending on the business field in which the information processing system S is used. Clinical importance is an example of a property of the generated information 410.

[0122] For example, the stronger the association between the generated information 410 and the diagnostic result for a patient for whom the generated information 410 was referenced during diagnosis, the higher the evaluation function 158 evaluates the clinical importance of the generated information 410. In other words, the stronger the association between the generated information 410 and the diagnostic result for a patient for whom the generated information 410 was referenced during diagnosis, the higher the evaluation function 158 evaluates the need to store the generated information 410. The strength of the association with the diagnostic result is stronger, for example, when the symptoms of the disease name included in the diagnostic result are directly expressed in the image data of the generated information 410. Furthermore, the evaluation function 158 may evaluate the association between the generated information 410 and the diagnostic result as stronger the fewer other medical information 21 the doctor referenced during diagnosis. Note that the strength of the association with the diagnostic result may be determined by the generative model 40 or another generative model.

[0123] The evaluation function 158 may express the degree of clinical importance numerically or as a level such as "low," "medium," or "high."

[0124] Furthermore, the evaluation criteria for the clinical importance of the generated information 410 are not limited to the strength of the relevance to the diagnosis result. For example, other examples of evaluation criteria include "if there is no other test data, evaluate the clinical importance higher", "the less test data other than the generated information 410 there is, the higher the clinical importance is evaluated", "the more times (or the time spent checking, or the number of checkers) the doctor checks the generated information 410 at the time of diagnosis, the higher the clinical importance is evaluated", and "if the results are inconsistent with other medical information, evaluate the clinical importance higher", etc.

[0125] "Other test data" and "other medical information" are test data and test results not included in or referenced by prompt 50.

[0126] The generated information 410 being inconsistent with other medical information 21 refers to, for example, a difference between the medical condition indicated by the generated information 410 and the medical condition indicated by other test results or captured medical image data. Specifically, this applies when cancer is clearly depicted in the image data that is the generated information 410, but the tumor marker test value is not positive. This also applies when the size of a tumor depicted in medical image data captured by a modality of a patient differs from the size of a tumor depicted in the image data that is the generated information 410. Note that the generated information 410 being inconsistent with other medical information does not necessarily mean that either the generated information 410 or the other medical information is incorrect, but also includes cases where information not detected in a test or the like is obtained in the generated information 410. Doctors may derive diagnostic results from multiple perspectives by referring to multiple pieces of conflicting information during diagnosis. Therefore, storing multiple pieces of conflicting information may be useful for patient follow-up and future verification.

[0127] On the other hand, there may be cases where the generated information 410 clearly contains errors compared to other medical information. For this reason, the evaluation criteria may include a criterion that "the more likely the generated information 410 is to contain errors, the lower its clinical importance should be evaluated."

[0128] The evaluation function 158 may also calculate the reliability (likelihood) of the generated information 410, and evaluate the clinical importance higher as the likelihood increases. A known method can be used to calculate the likelihood of the generated information 410. For example, the evaluation function 158 may use another generative model for calculating the likelihood of the generated information 410. The other generative model may be provided in the first information processing device 100 or in an external device. The evaluation function 158 may also use the other generative model for calculating the likelihood of the generated information 410. Alternatively, the generative model 40 may have an automatic evaluation function for calculating the likelihood of the generated information 410 that it has generated. In this case, the evaluation function 158 evaluates the clinical importance according to the likelihood calculated by the generative model 40.

[0129] The doctor may also determine the clinical significance of the generated information 410. In this case, the evaluation function 158 obtains the clinical significance of the generated information 410 input by the doctor.

[0130] The evaluation function 158 may determine the level of clinical importance by combining multiple evaluation criteria. The evaluation criteria may be predetermined, or may be selected by an administrator of the first information processing device 100. The evaluation criteria for clinical importance are not limited to the above examples.

[0131] The determination function 159 determines whether or not to select the generated information 410 as data to be stored based on the evaluation result by the evaluation function 158. For example, if the level of clinical importance of the generated information 410 determined by the evaluation function 158 is equal to or greater than a specified standard, the determination function 159 determines that the generated information 410 is data to be stored. Furthermore, if the level of clinical importance of the generated information 410 is less than the specified standard, the determination function 159 determines that the generated information 410 is not data to be stored. The specified standard may be set in advance or by an administrator of the first information processing device 100.

[0132] In addition to the functions of the first embodiment, the storage processing function 155 of this embodiment stores the prompt 50, or both the generated information 410 and the prompt 50, in the storage circuitry 120 as data to be stored based on the judgment result by the judgment function 159.

[0133] More specifically, when the determination function 159 determines that the generated information 410 is to be stored as data to be stored, the storage processing function 155 stores the generated information 410 and the prompt 50 corresponding to the generated information 410 as data to be stored in the storage circuitry 120. Furthermore, when the determination function 159 determines that the generated information 410 is not to be data to be stored, the storage processing function 155 stores the prompt 50 as data to be stored in the storage circuitry 120, but does not store the generated information 410 as data to be stored in the storage circuitry 120.

[0134] In this embodiment, the storage processing function 155 stores the prompt 50 in the storage circuitry 120 as data to be stored, regardless of the determination result. The evaluation function 158 may also have a function for evaluating the necessity of storing the prompt 50. The evaluation criteria for evaluating the necessity of storing the prompt 50 may be different from the evaluation criteria for evaluating the necessity of storing the generated information 410.

[0135] Furthermore, in addition to the functions of the first embodiment, the deletion processing function 157 of this embodiment may use, as a specified deletion condition, criteria similar to the evaluation criteria used by the above-described evaluation function 158, or the evaluation results by the evaluation function 158. For example, the specified deletion condition may include the strength of the association between the generated information 410 and the diagnosis result, or the evaluation result of clinical importance being less than a specified criterion.

[0136] As described above, the first information processing device 100 of this embodiment evaluates the necessity of storing the generated information 410 and determines whether or not to designate the generated information 410 as data to be stored based on the evaluation result. Furthermore, the first information processing device 100 of this embodiment stores the prompt 50 or both the generated information 410 and the prompt 50 as data to be stored in the storage circuitry 120 based on the determination result. Therefore, in addition to the effects of the first embodiment, the first information processing device 100 of this embodiment can automatically designate the generated information 410 as data to be stored depending on the necessity of storing the generated information 410. This reduces the effort required by the user to determine the necessity of storing each piece of generated information 410. Furthermore, it is possible to reduce the likelihood of generated information 410 that is highly necessary to be stored being lost without being stored.

[0137] (Fifth embodiment) In the first to fourth embodiments described above, the first information processing device 100 stores the generated information 410 or the prompt 50 as data to be stored. In this fifth embodiment, the medical information 21 related to the generated information 410 is also stored as data to be stored.

[0138] The information processing system S of this embodiment includes, for example, a first information processing device 100, a second information processing device 200, and a medical information system 300, similar to the first and fourth embodiments described with reference to Figures 1 and 9. Also, the first information processing device 100 of this embodiment includes a NW interface 110, a memory circuitry 120, an input interface 130, a display 140, and a processing circuitry 150, similar to the first and fourth embodiments. Note that the same components as those in the first embodiment are denoted by the same reference numerals, and description thereof will be omitted.

[0139] The processing circuit 150 of this embodiment has a reception function 151, a prompt generation function 152, a first acquisition function 153, a second acquisition function 154, a storage processing function 155, a display control function 156, a deletion processing function 157, an evaluation function 158, and a judgment function 159, similar to the fourth embodiment.

[0140] The reception function 151, the prompt generation function 152, the second acquisition function 154, the deletion processing function 157, the evaluation function 158, and the determination function 159 have the same functions as those in the fourth embodiment.

[0141] The evaluation function 158 of this embodiment evaluates the relevance between various pieces of medical information 21 stored in the medical information system 300 and the generated information 410 in addition to the functions of the fourth embodiment.

[0142] The medical information 21 that is highly related to the generated information 410 is, for example, the medical information 21 included in the prompt 50 corresponding to the generated information 410, or the medical information 21 referenced by the prompt 50. Alternatively, the generative model 40 may identify the medical information 21 that is highly related to the generated information 410. In this case, the evaluation function 158 acquires the identification result of the medical information 21 that is highly related to the generated information 410 from the generative model 40. For example, the generative model 40 may identify the medical information 21 that was referenced by the RAG when generating the generated information 410 as the medical information 21 that is highly related to the generated information 410.

[0143] Alternatively, the conditions for the medical information 21 that are highly relevant to the generated information 410 may be specified in advance. For example, if it is determined that the generated information 410 will be generated based on medical image data of a patient as an application of the generative model 40, the medical image data that is the source of the generated information 410, the imaging conditions of the modality of the original medical image data, and the patient information of the subject of imaging may be specified as the medical information 21 that is highly relevant to the generated information 410.

[0144] The evaluation function 158 may evaluate the strength of the relevance between the generated information 410 and the medical information 21 using a numerical value or the like, or may evaluate it in two stages, "relevant" or "not relevant."

[0145] In addition to the functions of the first and fourth embodiments, the first acquisition function 153 of this embodiment acquires medical information from the medical information system 300 based on the evaluation result of the relevance with the generated information 410 by the evaluation function 158. For example, the first acquisition function 153 may acquire medical information 21 whose relevance with the generated information 410 has been evaluated to be higher than a specified standard. Note that if the medical information 21 to be acquired has already been acquired for generating the prompt 50, the reacquisition process by the first acquisition function 153 may be omitted. Note that if the relevance with the generated information 410 is evaluated in two stages, "yes" and "no," the first acquisition function 153 may acquire medical information 21 that has been evaluated to be relevance with the generated information 410.

[0146] In addition to the functions of the first and fourth embodiments, the storage processing function 155 of this embodiment stores medical information 21 as data to be stored in the storage circuitry 120 based on the evaluation result of the relevance with the generated information 410 by the evaluation function 158. For example, when there is medical information 21 whose relevance with the generated information 410 is evaluated to be higher than a specified standard, the storage processing function 155 may associate the medical information 21 with the prompt 50 and store the medical information 21 as data to be stored in the storage circuitry 120. Note that when the relevance with the generated information 410 is evaluated in two stages, "yes" and "no," the storage processing function 155 may store the medical information 21 evaluated to be relevance with the generated information 410 as data to be stored in the storage circuitry 120.

[0147] In addition, if there is no medical information 21 that is evaluated to have a relevance to the generated information 410 higher than a specified standard, or if there is no medical information 21 that is evaluated to have a relevance to the generated information 410, the memory processing function 155 may not store the medical information 21, and may store only the prompt 50 as data to be stored.

[0148] In addition to the functions of the first and fourth embodiments, the display control function 156 of this embodiment displays the medical information 21 on the display 140 based on the evaluation result of the relevance with the generated information 410 by the evaluation function 158. For example, the display control function 156 may display medical information 21 that has been evaluated to have a relevance with the generated information 410 higher than a specified standard, or medical information 21 that has been evaluated to have a relevance with the generated information 410, together with the generated information 410 on the display screens 141a to 141c.

[0149] In this way, the first information processing device 100 of this embodiment associates the medical information 21 with the prompt 50 based on the evaluation result of the relevance between the generated information 410 and the medical information 21, and stores the medical information 21 as data to be stored in the memory circuitry 120. Therefore, according to the first information processing device 100 of this embodiment, it is possible to reduce the occurrence of a situation where, for example, when the prompt 50 is re-input into the generative model 40 at a later date, the generated information 410 cannot be reproduced because the medical information 21 referenced by the prompt 50 does not exist, or the validity of the generated information 410 cannot be verified because the related medical information 21 is unknown.

[0150] (Sixth embodiment) In the sixth embodiment, the first information processing apparatus 100 further stores past versions of the generative model 40.

[0151] The information processing system S of this embodiment includes, for example, a first information processing device 100, a second information processing device 200, and a medical information system 300, similar to the first embodiment described in Fig. 1. The first information processing device 100 of this embodiment also includes a NW interface 110, a memory circuitry 120, an input interface 130, a display 140, and a processing circuitry 150, similar to the first and fourth embodiments. Note that the same components as those in the first embodiment are denoted by the same reference numerals, and description thereof will be omitted.

[0152] The processing circuit 150 of this embodiment has a reception function 151, a prompt generation function 152, a first acquisition function 153, a second acquisition function 154, a storage processing function 155, a display control function 156, and a deletion processing function 157, similar to the first embodiment.

[0153] The reception function 151, the prompt generation function 152, the first acquisition function 153, the second acquisition function 154, the display control function 156, and the deletion processing function 157 have the same functions as those in the first embodiment.

[0154] In addition to the functions of the first embodiment, the storage processing function 155 of this embodiment stores, when storing a prompt 50 in the storage circuitry 120 as data to be stored, version information of the generative model 40 that input the prompt 50 in association with the prompt 50. The version information of the generative model 40 is, for example, a number indicating the version of the generative model 40.

[0155] Generally, the version of the generative model 40 is updated periodically. For example, a new version of the generative model 40 may have updated parameters or changed functions as a result of additional training performed on a previous version of the generative model 40. Different versions of the generative model 40 may output different generation information 410 even when the same prompt 50 is input.

[0156] The previous version of the generative model 40 may be stored in the storage circuitry 120 of the first information processing device 100 or in the second information processing device 200.

[0157] 10 is a diagram showing an example of the relationship between the version of the generative model 40a according to the sixth embodiment and the generation information 410. For example, suppose that the second acquisition function 154 inputs a prompt 50 into the generative model 40a to obtain the generation information 410. If the version of the generative model 40a at the time of generating this generation information 410 is "ver. 3," the same generation information 410 may not be obtained even if the prompt 50 is input into a version released after "ver. 3."

[0158] Therefore, the storage processing function 155 associates the version of the generative model 40a ("ver. 3" in the example shown in FIG. 10) with the prompt 50 and stores them as data to be stored in the storage circuitry 120. In this case, if it becomes necessary to reproduce the generation information 410 using the prompt 50 at a later date, the version of the generative model 40a stored in association with the prompt 50 can be identified as the input destination for the prompt 50.

[0159] (Seventh embodiment) In the seventh embodiment, in addition to the sixth embodiment, when a prompt 50 input to a previous version of a generative model 40 is re-input, the prompt 50 is input to multiple versions of the generative model 40.

[0160] The information processing system S of this embodiment includes, for example, a first information processing device 100, a second information processing device 200, and a medical information system 300, similar to the first embodiment described in Fig. 1. The first information processing device 100 of this embodiment also includes a NW interface 110, a memory circuitry 120, an input interface 130, a display 140, and a processing circuitry 150, similar to the first and fourth embodiments. Note that the same components as those in the first embodiment are denoted by the same reference numerals, and description thereof will be omitted.

[0161] The processing circuit 150 of this embodiment has a reception function 151, a prompt generation function 152, a first acquisition function 153, a second acquisition function 154, a storage processing function 155, a display control function 156, and a deletion processing function 157, similar to the first embodiment.

[0162] The reception function 151, the prompt generation function 152, the first acquisition function 153, the storage processing function 155, and the deletion processing function 157 have the same functions as those in the first or sixth embodiment.

[0163] In addition to having the same functions as the first and sixth embodiments, the second acquisition function 154 of this embodiment inputs a prompt 50 that was input to a previous version of the generative model 40 into multiple versions of the generative model 40 when re-inputting the prompt 50.

[0164] 11 is a diagram showing an example of the relationship between the versions of the generative models 40a, 40b and the generation information 410c, 410d when a prompt 50 according to the seventh embodiment is re-input. For example, the second acquisition function 154 inputs the prompt 50 to both the generative model 40a of version "ver. 3" saved in association with the prompt 50 and the generative model 40b of the latest version "ver. 4." In this case, the second acquisition function 154 acquires the generation information 410c (generation information X) generated by the generative model 40a and the generation information 410b (generation information Y) generated by the generative model 40b.

[0165] In addition to the same functions as those of the first and sixth embodiments, the display control function 156 of this embodiment causes the generation information 410c, 410d generated in different versions to be associated with version information and displayed on the display 140. The display control function 156 may also identify and display the difference between the generation information 410c, 410d generated in different versions.

[0166] In this way, the first information processing device 100 of this embodiment can provide the user with generated information 410 generated by multiple versions of the generative model 40, thereby providing information for verifying the contents of past generated information 410 or for reconsidering a patient's diagnosis, etc.

[0167] In addition, the first information processing device 100 of this embodiment can assist the user in reviewing past diagnostic results taking into account changes in the generation information 410d due to the latest generation model 40b by displaying the generation information 410c generated by an older version of the generation model 40a and the generation information 410d generated by the latest version of the generation model 40b in a comparable manner.

[0168] (Eighth embodiment) In the eighth embodiment, the first information processing apparatus 100 inputs one prompt 50 into the same generative model 40 multiple times to obtain multiple pieces of generative information 410.

[0169] In general, even if the version of the generative model 40 is the same, the content of the generated information 410 may change each time the prompt 50 is entered. Such changes in the generated information 410 are also called "fluctuations." Due to the occurrence of such fluctuations, if only the prompt 50 is to be stored and the generated information 410 is not, there is an issue with the reproducibility of the generated information 410.

[0170] The information processing system S of this embodiment includes, for example, a first information processing device 100, a second information processing device 200, and a medical information system 300, similar to the first embodiment described in Fig. 1. Also, the first information processing device 100 of this embodiment includes a NW interface 110, a memory circuitry 120, an input interface 130, a display 140, and a processing circuitry 150, similar to the first embodiment. Note that the same components as those in the first embodiment are denoted by the same reference numerals, and description thereof will be omitted.

[0171] The processing circuit 150 of this embodiment has a reception function 151, a prompt generation function 152, a first acquisition function 153, a second acquisition function 154, a storage processing function 155, a display control function 156, and a deletion processing function 157, similar to the first embodiment.

[0172] The reception function 151, the prompt generation function 152, the first acquisition function 153, and the deletion processing function 157 have the same functions as those in the first or sixth embodiment.

[0173] In addition to the same functions as those of the first embodiment, the second acquisition function 154 of this embodiment inputs the same prompt 50 multiple times to the same generative model 40 to acquire multiple pieces of generation information 410. For example, if the prompt 50 instructs the generation of similar case image data, the generative model 40 outputs multiple pieces of similar case image data corresponding to the multiple inputs of the prompt 50.

[0174] Furthermore, when the plurality of pieces of generated information 410 include non-image data such as numerical values, the second acquisition function 154 may calculate statistical information of the plurality of pieces of generated information 410. The statistical information may be, for example, average information or most frequent information of the plurality of pieces of generated information 410.

[0175] In addition to the same functions as in the first embodiment, the display control function 156 of this embodiment may also adopt the largest number of pieces of generated information 410 as the display target when the same pieces of generated information 410 are included among multiple pieces of generated information 410 output from the same generative model 40.

[0176] Furthermore, the display control function 156 may display statistical information of multiple pieces of generated information 410 on the display 140. Furthermore, when the generated information 410 is a region image showing a region such as a lesion in the image data, the display control function 156 may display an overlapping region.

[0177] Furthermore, the display control function 156 may cause the display 140 to display a message that provides feedback to the user who input the prompt 50 about the magnitude of fluctuations in the plurality of pieces of generated information 410 .

[0178] For example, the second acquisition function 154 may determine the magnitude of the fluctuation based on the magnitude of the difference between the multiple pieces of generated information 410. In this case, if the fluctuation is greater than or equal to a specified standard, the display control function 156 may cause the display 140 to display a message urging the user to change the prompt 50.

[0179] Generally, when the content of the prompt 50 is vague or when the prompt 50 is too short and does not contain enough information, there is likely to be significant fluctuation in the plurality of pieces of generated information 410. For this reason, for example, the display control function 156 may display a warning on the display 140 when the prompt 50 stored in the memory circuitry 120 as data to be stored is vague and the reproducibility of the generated information 410 is low.

[0180] Furthermore, the display control function 156 may display a warning on the display 140 if the number of characters included in the prompt 50 is less than a specified number of characters, or if the prompt 50 does not satisfy a specified condition. The specified number of characters is, for example, 50 characters, but is not limited to this value. The specified condition is, for example, a constraint on the generation of the generation information 410. Furthermore, if the prompt 50 instructs the generation of image data, the display control function 156 may set a specified condition that the prompt 50 includes the original image data or the path to the destination where the original image data is saved.

[0181] The display control function 156 displays the warning when the prompt 50 is stored in the storage circuitry 120 as data to be stored. The trigger for displaying the warning may be, for example, the user pressing the prompt display button 62 on the display screen 141a shown in FIG. 3. Alternatively, the trigger may be the user pressing the complete button 64. Furthermore, when the display control function 156 displays the prompt storage buttons 611, 620 on the display 140 as shown in FIGS. 7 and 8, the trigger for displaying the warning may be the user pressing the prompt storage button 611, 620.

[0182] In addition to the same functions as those of the first embodiment, the storage processing function 155 of this embodiment may set data to be stored based on the magnitude of fluctuations in the plurality of pieces of generated information 410. For example, the storage processing function 155 may determine data to be stored based on a rule such as "if the fluctuation is equal to or greater than a specified standard, the prompt 50 is set as data to be stored, and if the fluctuation is less than the specified standard, the prompt 50 and the generated information 410 are set as data to be stored," or "if the fluctuation is less than the specified standard, the prompt 50 is set as data to be stored, and if the fluctuation is equal to or greater than the specified standard, the prompt 50 and the generated information 410 are set as data to be stored."

[0183] Furthermore, if the user edits the prompt 50 after the display control function 156 displays a warning, the storage processing function 155 stores the edited prompt 50 as data to be stored in the storage circuitry 120. Furthermore, if the user edits the prompt 50 after the warning is displayed, the second acquisition function 154 may input the edited prompt 50 to the generative model 40 and verify the reproducibility of the generation information 410. In this case, after the display control function 156 displays on the display 140 the generation information 410 generated in the generative model 40 by the edited prompt 50, the storage processing function 155 may store the edited prompt 50 in the storage circuitry 120 as data to be stored if the user performs an operation to specify the edited prompt 50 as data to be stored.

[0184] In this way, the first information processing device 100 of this embodiment has a function to reduce fluctuations in the generated information 410, so even if only the prompt 50 is stored and the generated information 410 is not stored, the generated information 410 can be reproduced with high accuracy using the stored prompt 50.

[0185] (Ninth embodiment) In the above-mentioned eighth embodiment, the first information processing device 100 inputs one prompt 50 into the same generative model 40 multiple times to obtain multiple pieces of generated information 410, but the first information processing device 100 may also determine the data to be stored based on fluctuations in the multiple pieces of generated generated information 410.

[0186] The information processing system S of this embodiment includes, for example, a first information processing device 100, a second information processing device 200, and a medical information system 300, similar to the first embodiment described in Fig. 1. Also, the first information processing device 100 of this embodiment includes a NW interface 110, a memory circuitry 120, an input interface 130, a display 140, and a processing circuitry 150, similar to the first embodiment. Note that the same components as those in the first embodiment are denoted by the same reference numerals, and description thereof will be omitted.

[0187] The processing circuitry 150 of this embodiment includes, as in the first embodiment, a reception function 151, a prompt generation function 152, a first acquisition function 153, a second acquisition function 154, a storage processing function 155, a display control function 156, and a deletion processing function 157. Furthermore, the processing circuitry 150 of this embodiment may include a determination function 159, as in the fourth embodiment described with reference to FIG.

[0188] For example, the determination function 159 may determine whether to store the prompt 50 and each of the plurality of pieces of generated information as data to be stored in the memory circuitry 120 based on the similarity between the plurality of pieces of generated information generated by inputting the same prompt 50 multiple times to the generative model 40. In other words, the determination function 159 may determine the data to be stored in the memory circuitry 120 as data to be stored.

[0189] Fig. 12 is a diagram illustrating an example of determining data to be stored based on the similarity of a plurality of pieces of generated information according to the ninth embodiment. Note that Fig. 12 shows an overview of the process, and the second acquisition function 154 and the like are omitted.

[0190] In addition to the same processing as in the first embodiment, the second acquisition function 154 of this embodiment inputs the same prompt 50 to the same generative model 40 multiple times to acquire multiple pieces of generation information 410. For example, in FIG. 12, the second acquisition function 154 inputs the prompt 50 to the generative model 40 twice.

[0191] In FIG. 12, the prompt 50 includes, as an example, the text "#input sentence: Please generate and output a report based on the subject's medical information." First, the second acquisition function 154 inputs the prompt 50 to the generative model 40 in the first input. In this case, the generative model 40 outputs first generated information 410e. The generative model 40 may also output a summary 420 of the first generated information 410e together with the first generated information 410e.

[0192] Then, in the second input, the second acquisition function 154 inputs the same prompt 50 as the first input to the generative model 40 for verification of the first generation information 410e. In this case, the generative model 40 outputs the second generation information 410f based on the prompt 50 input the second time.

[0193] In the example shown in FIG. 12, the prompt 50 includes a command to output a report, and therefore the first generated information 410e and the second generated information 410f are reports on the subject's medical information. Therefore, the first generated information 410e is also referred to as a first report. The second generated information 410f is also referred to as a second report. The first report and the second report include text data explaining, for example, the diagnosis for the subject, the location of the lesion, etc. Depending on the content of the prompt 50, the first generated information 410e and the second generated information 410f may be, for example, image data indicating the location of the lesion.

[0194] If the prompt 50 is ambiguously written, even if the same prompt 50 is input twice to the same generative model 40, the output data may differ between the first and second times. Furthermore, even if there is fluctuation in the generative model 40, different data may be output even when the same prompt 50 is input. Fluctuation in the generative model 40 may occur, for example, when the amount of training data for the generative model 40 is insufficient or when the content of the training data for the generative model 40 is biased. Therefore, the first generation information 410e and the second generation information 410f shown in FIG. 12 may be the same or different. Furthermore, when the first generation information 410e and the second generation information 410f are different, the difference between the first generation information 410e and the second generation information 410f may be slight or large.

[0195] Therefore, the judgment function 159 of this embodiment compares the first generated information 410e and the second generated information 410f to determine the similarity, and based on the similarity, determines whether or not to store the prompt 50 in the memory circuitry 120 as data to be stored.

[0196] For example, if the similarity between the first generation information 410e and the second generation information 410f is equal to or greater than a specified standard, the prompt 50 is likely to be an appropriate prompt that will stably obtain similar output data from the generation model 40. For this reason, if the similarity between the first generation information 410e and the second generation information 410f is equal to or greater than a specified standard, the determination function 159 of this embodiment determines to store the prompt 50 as data to be stored in the memory circuitry 120. In this case, the storage processing function 155 of this embodiment stores the prompt 50 as data to be stored in the memory circuitry 120, as shown in FIG. 12 .

[0197] The method by which the determination function 159 determines the similarity between the first generated information 410e and the second generated information 410f is not particularly limited. For example, the determination function 159 may determine the similarity so that the greater the similarity of the medical report content, such as the common diagnosis or lesion location included in the first generated information 410e and the second generated information 410f, the higher the similarity. Alternatively, the determination function 159 may determine the similarity between the first generated information 410e and the second generated information 410f based on the proportion of common words. Furthermore, when the first generated information 410e and the second generated information 410f are image data, the determination function 159 may determine the similarity based on the position, number, size, etc. of the lesions shown in the image data.

[0198] The specified criteria for the similarity between the first generated information 410e and the second generated information 410f, which is the criterion for determining whether or not the prompt 50 is to be stored data, may be set in advance or may be specified by the user.

[0199] Between the first and second input processes, the prompt 50, the first generation information 410e, and the second generation information 410f may be temporarily stored in a volatile memory or the storage circuitry 120, but are not stored as data to be stored in the storage circuitry 120. That is, until the data to be stored is determined by the determination function 159, none of the prompt 50, the first generation information 410e, and the second generation information 410f is stored as data to be stored in the storage circuitry 120. After the data to be stored is determined by the determination function 159, data other than the data to be stored is deleted.

[0200] FIG. 13 is a diagram illustrating another example of determining data to be stored based on the similarity of a plurality of pieces of generation information according to the ninth embodiment.

[0201] A low degree of similarity between the first generated information 410e and the second generated information 410f output from the generative model 40 in response to the input of the same prompt 50 indicates that it is difficult to reproduce the first generated information 410e using the prompt 50. Therefore, in this embodiment, if the degree of similarity between the first generated information 410e and the second generated information 410f is less than a specified standard, the determination function 159 determines not to store the prompt 50 in the memory circuitry 120 as data to be stored. In this case, the storage processing function 155 stores the first generated information 410e in the memory circuitry 120 as data to be stored. The storage processing function 155 may also store a summary of the first generated information 410e in the memory circuitry 120 as data to be stored. In this case, the storage processing function 155 does not store the prompt 50 in the memory circuitry 120.

[0202] In addition, when the amount of data is small, the summary of the first generated information 410e may be stored in the memory circuitry 120 even if the memory circuitry 120 stores the prompt 50 but does not store the first generated information 410e, as described in FIG. 12.

[0203] 12 and 13 may be divided and executed by a plurality of functions. For example, the evaluation function 158 may determine the similarity between the first generated information 410e and the second generated information 410f. The evaluation function 158 may also evaluate the degree of fluctuation of the plurality of pieces of generated information 410 or the ambiguity of the prompt 50. The determination function 159 may then determine the data to be stored based on the evaluation result of the degree of fluctuation by the evaluation function 158.

[0204] 12 and 13, the determination function 159 determines the data to be stored. However, the user may make the final determination based on the determination result by the determination function 159. For example, the display control function 156 may display the similarity between the first generated information 410e and the second generated information 410f determined by the determination function 159 on the display 140. The display control function 156 may also display the first generated information 410e and the second generated information 410f together with the similarity on the display 140. In this case, the reception function 151 receives a user's selection operation to select which of the prompt 50, the first generated information 410e, and the second generated information 410f to store in the storage circuitry 120 as data to be stored. The storage processing function 155 stores the data selected by the user from the prompt 50, the first generated information 410e, and the second generated information 410f in the storage circuitry 120.

[0205] Furthermore, the user may edit the prompt 50 according to the similarity between the first generation information 410e and the second generation information 410f displayed on the display 140. For example, the display control function 156 displays the similarity between the first generation information 410e and the second generation information 410f and an editing screen for the prompt 50 on the display 140. In this case, the second acquisition function 154 inputs the prompt 50 edited by the user into the generative model 40 twice to obtain new first generation information 410e and new second generation information 410f based on the edited prompt 50. Then, the determination function 159 calculates the similarity between the new first generation information 410e and the new second generation information 410f. The display control function 156 displays the newly calculated similarity on the display 140. In this case, the editing of the prompt 50 and the process of determining the similarity between the first generated information 410e and the second generated information 410f may be repeated until the user decides to store the prompt 50 in the memory circuitry 120 as data to be stored.

[0206] Furthermore, the display control function 156 may display, on the display 140, example prompts that have small fluctuations in the output of the generative model 40, along with the editing screen for the prompt 50. The example prompts that have small fluctuations in the output of the generative model 40 may be, for example, prompts 50 that have a high degree of similarity between the first generation information 410e and the second generation information 410f, among multiple prompts 50 previously input to the generative model 40. In other words, the prompts 50 that have a high degree of similarity between the first generation information 410e and the second generation information 410f are prompts 50 that have a high probability of reproducing the first generation information 410e. The example prompts may be stored in, for example, the memory circuitry 120. In this case, the display control function 156 reads out the example prompts from the memory circuitry 120 and displays them on the display 140. In this case, the user can edit the prompt 50 by referring to the displayed example prompts.

[0207] Note that the memory circuitry 120 may store a plurality of example prompts. In this case, the display control function 156 may select one of the example prompts stored in the memory circuitry 120 that is similar to the prompt 50 before being edited by the user, and display the selected example prompt on the display 140. For example, if the prompt 50 includes an instruction to generate and output a report, the display control function 156 may select one of the example prompts that includes an instruction to generate and output a report, and display the selected example prompt on the display 140. Also, if the prompt 50 includes an instruction to generate and output image data, the display control function 156 may select one of the example prompts that includes an instruction to generate and output image data, and display the selected example prompt on the display 140.

[0208] In addition to manual editing of the prompt 50 by the user, automatic editing of the prompt 50 by the first information processing device 100 may be possible. For example, the prompt generation function 152 of the processing circuitry 150 of the first information processing device 100 may edit the prompt 50 so that the fluctuation is smaller, based on an example of a prompt with small fluctuation in the output of the generative model 40 stored in the memory circuitry 120.

[0209] For example, the prompt generation function 152 may automatically edit the prompt 50 using a trained model that associates a prompt with high variability with a prompt edited by a user to reduce the variability of the high-variability prompt. The second acquisition function 154 may input the automatically edited prompt 50 into the generation model 40 twice to obtain new first generation information 410e and new second generation information 410f. In this case, the determination function 159 calculates the similarity between the new first generation information 410e and the new second generation information 410f. If the similarity is equal to or greater than a specified standard, the storage processing function 155 stores the automatically edited prompt 50 in the storage circuitry 120 as data to be stored. Furthermore, the automatic editing of the prompt 50 and the process of determining the similarity between the first generation information 410e and the second generation information 410f may be repeated until the similarity is equal to or greater than a specified standard.

[0210] (Tenth embodiment) In the tenth embodiment, existing medical information similar to the generated information 410 output from the generative model 40 is used as a backup in case the generated information 410 is not accurately reproduced.

[0211] The information processing system S of this embodiment includes, for example, a first information processing device 100, a second information processing device 200, and a medical information system 300, similar to the first embodiment described in Fig. 1. The medical information system 300 of this embodiment has, for example, a PACS function. Therefore, the medical information system 300 stores multiple medical image data.

[0212] Similarly to the first embodiment, the first information processing device 100 of this embodiment includes a NW interface 110, a storage circuit 120, an input interface 130, a display 140, and a processing circuit 150. Note that the same components as those in the first embodiment are denoted by the same reference numerals, and descriptions thereof will be omitted.

[0213] The processing circuit 150 of this embodiment has a reception function 151, a prompt generation function 152, a first acquisition function 153, a second acquisition function 154, a storage processing function 155, a display control function 156, and a deletion processing function 157, similar to the first embodiment.

[0214] FIG. 14 is a diagram illustrating an example of data to be stored according to the tenth embodiment.

[0215] 14 includes, as an example, the text "#input sentence: Please generate and output a reference image (reference) based on the medical information of the subject." In this embodiment, the prompt 50 includes an instruction to output a reference image.

[0216] The second acquisition function 154 of this embodiment inputs a prompt 50 to the generative model 40. In this case, the generative model 40 outputs a reference image as generation information 410 based on the prompt 50. In this embodiment, the reference image output from the generative model 40 is also referred to as a first reference image to distinguish it from a second reference image described below.

[0217] As described above, the medical information system 300 of this embodiment stores a plurality of medical image data (hereinafter referred to as medical images 411a-411n). In this embodiment, among the plurality of medical images 411a-411n, one similar to the generated information 410 output from the generative model 40 is used as a second reference image that serves as a backup for the generated information 410. In the example shown in FIG. 14, the medical image 411a is the second reference image similar to the generated information 410.

[0218] The generated information 410 and the medical image 411a being similar means, for example, that the organs to be imaged, the cross-sectional positions and cross-sectional directions in the subject, and the positions, sizes, and numbers of depicted abnormal parts (e.g., tumors) in the generated information 410 and the medical image 411a are the same or similar. Note that the definition of similarity between the generated information 410 and the medical image 411a is not limited to this.

[0219] The multiple medical images 411a-411n are images generated in advance based on, for example, medical images (medical images) of subjects other than the "subject" in the prompt 50. In other words, the multiple medical images 411a-411n are not images dedicated to the prompt 50 relating to a certain subject, but are commonly referenced each time the generative model 40 is used. In other words, the medical images 411a-411n are stored in the medical information system 300 as a database of reference medical images.

[0220] In addition to the same processing as in the first embodiment, the first acquisition function 153 of this embodiment acquires a medical image 411a similar to the generated information 410 from the medical information system 300 via the network 400 and the NW interface 110. For example, the first acquisition function 153 may identify the medical image 411a to 411n that is most similar to the generated information 410 by comparing the generated information 410 with the medical images 411a to 411n using known image processing.

[0221] The display control function 156 of this embodiment causes the generated information 410 and a medical image 411 a similar to the generated information 410 to be displayed on the display 140 .

[0222] Furthermore, the storage processing function 155 of this embodiment stores the prompt 50 and an address indicating the storage location of the medical image 411a in the medical information system 300 as storage target data in the storage circuitry 120. The address indicating the storage location of the medical image 411a in the medical information system 300 is also referred to as storage location information of the medical image 411a.

[0223] The storage processing function 155 stores an address indicating the storage destination of a medical image 411a similar to the generated information 410 in the storage circuitry 120 as data to be stored, but does not store the medical image 411a itself as data to be stored in the storage circuitry 120. Furthermore, the storage processing function 155 does not store the generated information 410 as data to be stored in the storage circuitry 120. Therefore, the first information processing device 100 of this embodiment can suppress an increase in the amount of data to be stored in a single input of the prompt 50 to the generation model 40.

[0224] Furthermore, for example, at a later date, a user may re-input the prompt 50 stored in the memory circuitry 120 as storage target data into the generative model 40 in order to restore the generated information 410. In this case, the display control function 156 of this embodiment displays, on the display 140, the medical image 411a indicated by the address stored in the memory circuitry 120 as storage target data, along with the generated information 410 generated based on the re-input prompt 50. For example, due to fluctuations in the generative model 40, the generated information 410 generated based on the re-input prompt 50 may differ from the generated information 410 generated when the prompt 50 was initially input. Even in this case, the display control function 156 displays, on the display 140, a medical image 411a similar to the generated information 410 generated when the prompt 50 was initially input, allowing the user to confirm the medical image 411a similar to the initial generated information 410. In other words, the medical image 411a serves as a backup in case the reproducibility of the generated information 410 is low due to fluctuations in the generative model 40.

[0225] In this way, according to the first information processing device 100 of this embodiment, by storing an address indicating the storage location of a medical image 411a similar to the generated information 410 in the memory circuitry 120 as data to be stored, it is possible to suppress the increase in the amount of data to be stored and to ensure a guarantee in case the reproducibility of the generated information 410 is low in the future.

[0226] (Eleventh embodiment) In the above-described first to tenth embodiments, the first information processing device 100, which is a computer such as a server or a PC, is an example of an information processing device. In this eleventh embodiment, an X-ray CT device executes processing as an example of an information processing device.

[0227] Fig. 15 is a diagram showing an example of the overall configuration of an information processing system S according to the eleventh embodiment. As shown in Fig. 15, in the information processing system S of this embodiment, as an example, an X-ray CT apparatus 1001, a second information processing apparatus 200, and a medical information system 300 are communicably connected via a network 400 such as an in-hospital LAN. The second information processing apparatus 200 and the medical information system 300 have the same functions as those of the first embodiment.

[0228] FIG. 16 is a diagram showing an example of the hardware configuration of an X-ray CT apparatus 1001 according to the eleventh embodiment. The X-ray CT apparatus 1001 irradiates an object (subject) P with X-rays from an X-ray tube 1011 and detects the irradiated X-rays with an X-ray detector 1012. The X-ray CT apparatus 1001 generates a CT image of the object P based on the output from the X-ray detector 1012. There are various types of X-ray CT apparatuses 1001, such as third-generation CT, fourth-generation CT, and fifth-generation CT, and any of these types is applicable to this embodiment. Here, the third-generation CT is a rotate / rotate type in which the X-ray tube and the detector rotate together around the object. The fourth-generation CT is a stationary / rotate type in which a large number of X-ray detection elements arranged in a ring shape are fixed, and only the X-ray tube rotates around the object. The fifth-generation system includes a focus coil that focuses the electron beam generated from the electron gun, a deflection coil that electromagnetically deflects it, and a target ring that surrounds half of the subject P and generates X-rays when the deflected electron beam collides with it.

[0229] As shown in FIG. 16, the X-ray CT apparatus 1001 has a gantry 1010, a bed 1030, and a console 1040. For convenience of explanation, a plurality of gantry 1010 are depicted in FIG. 16. The gantry 1010 is a scanning apparatus having a configuration for performing X-ray CT imaging on a subject P. The bed 1030 is a transport device on which the subject P to be subjected to X-ray CT imaging is placed and for positioning the subject P. The console 1040 is a computer that controls the gantry 1010. For example, the gantry 1010 and the bed 1030 are installed in a CT examination room, and the console 1040 is installed in a control room adjacent to the CT examination room. The gantry 1010, the bed 1030, and the console 1040 are connected to each other by wire or wirelessly so that they can communicate with each other.

[0230] The console 1040 does not necessarily have to be installed in a control room. For example, the console 1040 may be installed in the same room as the gantry 1010 and the bed 1030. Alternatively, the console 1040 may be incorporated into the gantry 1010.

[0231] In this embodiment, the rotation axis of the rotating frame 1013 in the non-tilted state or the longitudinal direction of the tabletop 1033 of the bed 1030 is defined as the Z-axis direction, the axis direction perpendicular to the Z-axis direction and horizontal to the floor surface is defined as the X-axis direction, and the axis direction perpendicular to the Z-axis direction and perpendicular to the floor surface is defined as the Y-axis direction.

[0232] As shown in FIG. 16, the gantry 1010 includes an X-ray tube 1011, an X-ray detector 1012, a rotating frame 1013, an X-ray high voltage device 1014, a control device 1015, a wedge 1016, a collimator 1017, and a data acquisition system (DAS) 1018.

[0233] The X-ray tube 1011 is a vacuum tube having a cathode (filament) that generates thermoelectrons and an anode (target) that generates X-rays upon impact of the thermoelectrons. The X-ray tube 1011 irradiates the subject P with X-rays by irradiating the thermoelectrons from the cathode to the anode using a high voltage supplied from an X-ray high voltage device 1014.

[0234] It should be noted that the hardware that generates X-rays is not limited to the X-ray tube 1011. For example, instead of the X-ray tube 1011, X-rays may be generated using a fifth generation system.

[0235] The X-ray detector 1012 detects X-rays emitted from the X-ray tube 1011 and passing through the subject P, and outputs an electrical signal corresponding to the detected X-ray dose to the DAS 1018. The X-ray detector 1012 has, for example, an X-ray detection element row in which a plurality of X-ray detection elements are arranged in the channel direction along an arc centered on the focal point of the X-ray tube 1011. The X-ray detector 1012 has, for example, a structure in which a plurality of X-ray detection elements are arranged in the slice direction (column direction, row direction) in the channel direction. The X-ray detector 1012 is, for example, an indirect conversion type detector having a grid, a scintillator array, and a photosensor array. The scintillator array has a plurality of scintillators. The scintillator has scintillator crystals that output light with an amount of light corresponding to the dose of incident X-rays. The grid is arranged on the X-ray incident surface side of the scintillator array and has an X-ray shielding plate that has the function of absorbing scattered X-rays. The grid is sometimes called a collimator (one-dimensional collimator or two-dimensional collimator). The photosensor array has a function of converting light from the scintillator into an electrical signal corresponding to the amount of light. For example, a photomultiplier tube (PMT) is used as the photosensor. The X-ray detector 1012 may be a direct conversion type detector having a semiconductor element that converts incident X-rays into an electrical signal. Here, the X-ray detector 1012 is an example of a detection unit.

[0236] The rotating frame 1013 is an annular frame that supports the X-ray tube 1011 and the X-ray detector 1012 so as to face each other, and rotates the X-ray tube 1011 and the X-ray detector 1012 using a control device 1015 (described later). An image field of view (FOV) is set at an opening in the rotating frame 1013. For example, the rotating frame 1013 is made of aluminum casting. Note that the rotating frame 1013 can further support an X-ray high voltage generator 1014, a wedge 1016, a collimator 1017, a DAS 1018, and the like in addition to the X-ray tube 1011 and the X-ray detector 1012. The rotating frame 1013 can also further support various components not shown in FIG. 16 .

[0237] The X-ray high voltage device 1014 has a high voltage generator and an X-ray control device. The high voltage generator has electrical circuits such as a transformer and a rectifier, and generates a high voltage to be applied to the X-ray tube 1011 and a filament current to be supplied to the X-ray tube 1011. The X-ray control device controls the output voltage according to the X-rays emitted by the X-ray tube 1011. The high voltage generator may be of a transformer type or an inverter type. The X-ray high voltage device 1014 may be provided on a rotating frame 1013 in the gantry 1010, or on a fixed frame (not shown) in the gantry 1010. The fixed frame is a frame that rotatably supports the rotating frame 1013.

[0238] The control device 1015 includes a driving mechanism such as a motor and an actuator, and a processing circuit having a processor, memory, etc. that controls the driving mechanism. The control device 1015 receives input signals from the input interface 1043 or other input interfaces provided on the gantry 1010, and controls the operation of the gantry 1010 and the bed 1030. Examples of the operation control by the control device 1015 include control to rotate the rotating frame 1013, control to tilt the gantry 1010, and control to operate the bed 1030. Note that the control to tilt the gantry 1010 is realized by the control device 1015 rotating the rotating frame 1013 around an axis parallel to the X-axis direction based on inclination angle (tilt angle) information input via the input interface 1043 attached to the gantry 1010. Note that the control device 1015 may be provided on the gantry 1010 or on the console 1040.

[0239] The wedge 1016 is a filter for adjusting the amount of X-rays irradiated from the X-ray tube 1011. Specifically, the wedge 1016 is a filter that transmits and attenuates the X-rays irradiated from the X-ray tube 1011 so that the X-rays irradiated from the X-ray tube 1011 to the subject P have a predetermined distribution. For example, the wedge 1016 is a wedge filter or a bow-tie filter, and is made by processing aluminum or the like to have a predetermined target angle and a predetermined thickness.

[0240] The collimator 1017 limits the irradiation range of the X-rays that have passed through the wedge 1016. The collimator 1017 slidably supports multiple lead plates that shield the X-rays, and adjusts the shape of the slits formed by the multiple lead plates. The collimator 1017 is sometimes called an X-ray aperture.

[0241] The DAS 1018 reads out from the X-ray detector 1012 an electrical signal corresponding to the X-ray dose detected by the X-ray detector 1012. The DAS 1018 amplifies the read electrical signal and integrates (adds up) the electrical signal over a view period to collect detection data having a digital value corresponding to the X-ray dose over the view period. The detection data is called projection data. The DAS 1018 is realized, for example, by an application specific integrated circuit (ASIC) equipped with circuit elements capable of generating projection data. The projection data is transmitted to the console 1040 via a non-contact data transmission device or the like. Here, the DAS 1018 is an example of a detection unit.

[0242] The detection data generated by the DAS 1018 is transmitted by optical communication from a transmitter having a light emitting diode (LED) provided on the rotating frame 1013 to a receiver having a photodiode provided on a non-rotating portion of the gantry 1010 (for example, a fixed frame; not shown in FIG. 16), and then transferred to the console 1040. The method of transmitting the detection data from the rotating frame 1013 of the rotating portion to the non-rotating portion of the gantry 1010 is not limited to the optical communication described above, and any method of non-contact data transfer may be used.

[0243] In this embodiment, an X-ray CT device 1001 equipped with an integral type X-ray detector 1012 will be described as an example, but the technology according to this embodiment can also be realized as an X-ray CT device 1001 equipped with a photon counting type X-ray detector.

[0244] The bed 1030 is a device on which the subject P to be scanned is placed and moved. The bed 1030 has a base 1031, a bed drive device 1032, a top plate 1033, and a support frame 1034. The base 1031 is a housing that supports the support frame 1034 so that it can move in the vertical direction. The bed drive device 1032 is a drive mechanism that moves the top plate 1033, on which the subject P is placed, in the longitudinal direction of the top plate 1033. The bed drive device 1032 includes a motor, an actuator, etc. The top plate 1033 is a plate on which the subject P is placed. The top plate 1033 is provided on the upper surface of the support frame 1034. The top plate 1033 can protrude from the bed 1030 toward the gantry 1010 so that the entire body of the subject P can be imaged. The tabletop 1033 is formed of, for example, carbon fiber reinforced plastic (CFRP), which has good X-ray transparency and physical properties such as rigidity and strength. The inside of the tabletop 1033 is hollow. The support frame 1034 supports the tabletop 1033 so that it can move in the longitudinal direction of the tabletop 1033. The bed driving device 1032 may move the support frame 1034 in addition to the tabletop 1033 in the longitudinal direction of the tabletop 1033.

[0245] The console 1040 has a memory 1041, a display 1042, an input interface 1043, a processing circuit 1044, and a NW interface 1045. Data communication between the memory 1041, the display 1042, the input interface 1043, the NW interface 1045, and the processing circuit 1044 is performed via a bus. Note that although the console 1040 will be described as being separate from the stand 1010, the stand 1010 may include the console 1040 or some of the components of the console 1040.

[0246] The memory 1041 is realized by, for example, a semiconductor memory element such as a ROM, a RAM, or a flash memory, a hard disk, an optical disk, or the like. For example, the memory 1041 stores various types of data to be stored. The memory 1041 also stores projection data and reconstructed image data. For example, the memory 1041 also stores various programs. The storage area of ​​the memory 1041 may be located within the X-ray CT device 1001 or may be located in an external storage device connected via a network. Here, the memory 1041 is an example of a storage unit in this embodiment.

[0247] The display 1042 is a liquid crystal display, a CRT display, etc. that displays various information. The display 1042 is an example of a display unit in this embodiment.

[0248] The display 1042 may be installed anywhere in the control room. Alternatively, the display 1042 may be installed on the stand 1010. The display 1042 may be a desktop type, or may be configured as a tablet terminal or the like capable of wireless communication with the main body of the console 1040. Alternatively, one or more projectors may be used as the display 1042.

[0249] The input interface 1043 receives various input operations from an operator, converts the received input operations into electrical signals, and outputs the electrical signals to the processing circuit 1044 .

[0250] As the input interface 1043, for example, a mouse, keyboard, trackball, switch, button, joystick, touchpad, touch panel display, etc. can be used as appropriate. Note that in this embodiment, the input interface 1043 is not limited to one having these physical operation components. For example, an example of the input interface 1043 also includes an electrical signal processing circuit that receives an electrical signal corresponding to an input operation from an external input device provided separately from the device and outputs this electrical signal to the processing circuit 1044. The input interface 1043 may also be provided on the stand 1010. The input interface 1043 may also be configured as a tablet terminal or the like that is capable of wireless communication with the console 1040 main body. Here, the input interface 1043 is an example of an input unit.

[0251] The NW interface 1045 is connected to the processing circuitry 1044, and controls the transmission and communication of various data between the X-ray CT apparatus 1001 and the second information processing apparatus 200 and the medical information system 300. The NW interface 1045 is realized by a network card, a network adapter, a NIC, or the like.

[0252] The processing circuitry 1044 controls the overall operation of the X-ray CT apparatus 1001. The processing circuitry 1044 has a processor and memories such as ROM and RAM as hardware resources. The processing circuitry 1044 executes various system processes using the processor that executes programs loaded into the memory. The processing circuitry 1044 includes, for example, a reception function 1051, a prompt generation function 1052, a first acquisition function 1053, a second acquisition function 1054, a storage processing function 1055, a display control function 1056, and a deletion processing function 1057. The reception function 1051, the prompt generation function 1052, the first acquisition function 1053, the second acquisition function 1054, the storage processing function 1055, the display control function 1056, and the deletion processing function 1057 have functions similar to those of the reception function 151, the prompt generation function 152, the first acquisition function 153, the second acquisition function 154, the storage processing function 155, the display control function 156, and the deletion processing function 157 in the processing circuit 150 of the first embodiment.

[0253] The processing circuitry 1044 further includes an imaging function 1058 and an image processing function 1059. The imaging function 1058 and the image processing function 1059 are functions for capturing X-ray CT images.

[0254] For example, the imaging function 1058 controls a CT scan performed by the gantry 1010 to obtain detection data of X-rays detected by the X-ray detector 1012. The imaging function 1058 also performs reconstruction processing on the X-ray detection data to obtain X-ray CT image data obtained by imaging the subject P. Examples of reconstruction processing include filtered back projection, iterative reconstruction, and machine learning. The image processing function 1059 converts the X-ray CT image data into tomographic image data of an arbitrary cross section or three-dimensional image data using a known method, based on an input operation received from an operator via the input interface 1043.

[0255] Furthermore, for example, the prompt generation function 1052 of this embodiment may generate a prompt 50 including the text "Please generate and output similar case images based on the medical information of the subject," as described in FIG. 2 of the first embodiment, based on a user operation accepted by the acceptance function 1051. Furthermore, the second acquisition function 1054 of this embodiment inputs the prompt 50 generated by the prompt generation function 152 into the generation model 40 stored in the second information processing device 200, and acquires generation information 410 from the generation model 40.

[0256] The medical information of the subject referenced by the prompt 50 in this embodiment is, for example, CT image data captured by the X-ray CT device 1001. The prompt 50 may also reference medical information such as the patient's electronic medical record and examination data stored in the medical information system 300 in addition to the CT image data captured by the X-ray CT device 1001.

[0257] The X-ray CT device 1001 of this embodiment has a function of generating generation information 410 in the generation model 40, so that, for example, it is possible to generate multiple different reference images from CT image data generated by a single scan by the X-ray CT device 1001. This has the advantage of reducing the imaging time and the burden on the patient.

[0258] Note that the medical information of the subject referenced in the prompt 50 may be limited to CT image data captured by the X-ray CT device 1001. In this case, the processing circuit 1044 of this embodiment may not have the first acquisition function 1053 that acquires medical information from the medical information system 300. Also, in this case, the information processing system S of this embodiment may not include the medical information system 300. Also, the generative model 40 may be provided in the X-ray CT device 1001 instead of the second information processing device 200. In this case, the X-ray CT device 1001 can use the generative model 40 standalone.

[0259] Although the X-ray CT apparatus 1001 is used as an example of an information processing apparatus in this embodiment, other medical image diagnostic apparatuses may also be used as an example of an information processing apparatus. For example, various medical image diagnostic apparatuses, such as an X-ray diagnostic apparatus, an MRI (Magnetic Resonance Imaging) apparatus, an ultrasound diagnostic apparatus, a SPECT (Single Photon Emission Computed Tomography) apparatus, a PET (Positron Emission Computed Tomography) apparatus, a SPECT-CT apparatus in which a SPECT apparatus and an X-ray CT apparatus are integrated, and a PET-CT apparatus in which a PET apparatus and an X-ray CT apparatus are integrated, may be provided with the functions of the processing circuit 1044 shown in FIG. 16 .

[0260] (Variation 1) In each of the above-described embodiments, the information processing system S is installed in a medical institution such as a hospital, but it may also be installed in a company or the like other than a medical institution. Furthermore, in each of the above-described embodiments, an example in which the information processing system S is used in the medical field has been described. However, the field of use of the information processing system S is not limited to the medical field, and the information processing system S can be used in various businesses, research and development, and the like that utilize generative AI. Note that, when the information processing system S is used in a field other than the medical field, it does not need to include the medical information system 300. For example, the information processing system S may include the first information processing device 100 and the generative model 40. Furthermore, the information processing system S may include another database management device or the like instead of or in addition to the medical information system 300.

[0261] (Variation 2) In the above-described embodiments, an example has been described in which the first information processing device 100, the second information processing device 200, and the medical information system 300 are connected via a network 400 such as an in-hospital LAN, as shown in FIG. 1 and other figures. However, the configuration of the information processing system S is not limited to this. For example, the second information processing device 200 may be provided outside a medical institution. As a specific example, the second information processing device 200 may be a server of a business that provides generation AI as a service. In this case, the second information processing device 200 may be connected to the first information processing device 100 and the medical information system 300 via the Internet or the like, rather than via the in-hospital LAN.

[0262] (Variation 3) The generative model 40 may be provided in the first information processing device 100, rather than in the second information processing device 200. For example, the generative model 40 may be stored in the memory circuitry 120 of the first information processing device 100 and used by the second acquisition function 154 described above. Alternatively, the second acquisition function 154 may be configured to include the generative model 40.

[0263] (Variation 4) In each of the above-described embodiments, the generative model 40 is a multimodal model, but the technology for constructing the generative model 40 is not limited to this, and various AI models can be adopted.

[0264] (Variation 5) In the above-described embodiments, as shown in FIG. 1 etc., the storage circuitry 120 of the first information processing device 100 stores the data to be stored, but the data to be stored may also be stored outside the first information processing device 100. For example, the information processing system S may include an information processing device dedicated to storing the data to be stored. In this case, the external storage device of the first information processing device 100 is an example of a storage unit.

[0265] (Variation 6) In the first embodiment described above, only the generated information 410 is subject to deletion based on the deletion conditions, and the prompt 50 is not deleted. However, the prompt 50 may also be subject to deletion based on the deletion conditions. Furthermore, the deletion conditions for the generated information 410 and the deletion conditions for the prompt 50 may be different.

[0266] (Variation 7) In the above-described fourth embodiment, the judgment function 159 judges whether the generated information 410 is to be stored based on the evaluation criteria for the clinical significance of the generated information 410, but the judgment criteria are not limited to this.

[0267] For example, the determination function 159 may determine the size of the data volume of the generated information 410 and determine whether to set the generated information 410 as data to be stored based on the data volume of the generated information 410. The data volume of the generated information 410 is an example of the property of the generated information 410 in this modification. Specifically, the determination function 159 may determine that both the prompt 50 and the generated information 410 are data to be stored when the data volume of the generated information 410 is less than a specified standard, and may determine that only the prompt 50 is data to be stored when the data volume of the generated information 410 is equal to or greater than the specified standard. Note that the evaluation function 158, rather than the determination function 159, may evaluate the size of the data volume of the generated information 410.

[0268] Furthermore, if the prompt 50 includes image data or a large amount of medical information 21, the data volume may be larger than that of the generated information 410. For this reason, the determination function 159 may determine that only the generated information 410 is data to be stored when the data volume of the prompt 50 is equal to or larger than that of the generated information 410, or may determine that only the prompt 50 is data to be stored when the data volume of the prompt 50 is less than a specified standard.

[0269] Based on the determination result by the determination function 159, the storage processing function 155 stores only the prompt 50, only the generated information 410, or both the generated information 410 and the prompt 50 in the storage circuitry 120 as data to be stored.

[0270] It is also possible to combine multiple of the above-described embodiments and modified examples. For example, any of the first to third and fifth to eleventh embodiments may be combined with the fourth embodiment. In this case, the storage processing function 155 may store, in the storage circuitry 120, the generated information 410 or the prompt 50 designated as data to be stored by a user operation and the generated information 410 or the prompt 50 determined as data to be stored based on the evaluation result of the necessity of storage.

[0271] Furthermore, the first embodiment and the third embodiment may be combined. For example, the first information processing device 100 may include the storage processing function 155 of the first embodiment, which has a function of automatically storing the prompt 50 input to the generative model 40 in the storage circuitry 120 as data to be stored, and the display control function 156 of the third embodiment, which has a function of displaying, for each piece of image data output from the generative model 40, an image button on the display 140 that accepts a user operation to designate the image data as data to be stored.

[0272] The various data handled in this specification are typically digital data.

[0273] According to at least one of the embodiments described above, it is possible to reduce the increase in the amount of data stored that accompanies the use of generative models.

[0274] Although several embodiments have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, modifications, and combinations of embodiments can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as within the scope of the invention and its equivalents as defined in the claims. [Explanation of symbols]

[0275] 21 Medical Information 40,40a,40b Generative Model 50 prompts 61a, 61b Memory button 62 Prompt display button 63 Back button 64 Finish button 65 Prompt display field 100 First information processing device 110,1045 Network Interface 120 Memory circuit 130,1043 Input Interface 140,1042 displays 141a~141c Display screen 142 First Screen Area 143 Second Screen Area 150,1044 processing circuit 151,1051 Reception function 152,1052 Prompt generation function 153,1053 First Acquisition Function 154,1054 Second acquisition function 155,1055 Amnestics 156,1056 Display control function 157,1057 Deletion processing function 158 Rating Function 159 Judgment Function 200 Second information processing device 300 Medical Information Systems 400 Network 410, 410a~410d Generation information 410e First generation information 410f Second generation information 411a~411n,501 Medical Imaging 611,620 Prompt Memory Button 612,621 Generation information memory button 621a Generated information A memory button 621b Generated information B memory button 1001 X-ray CT device 1010 Mounting stand 1011 X-ray tube 1012 X-ray detector 1013 Rotating Frame 1014 X-ray high voltage device 1015 Control device 1016 Wedge 1017 Collimator 1018 DAS 1030 berth 1031 Foundation 1032 Bed drive unit 1033 Top plate 1034 Support Frame 1040 Console 1041 memory 1058 shooting function 1059 Image Processing Function P Subject S Information Processing System

Claims

1. a prompt generator that generates a prompt; an acquisition unit that inputs the prompt into a generative model and acquires generation information from the generative model; a storage unit configured to store the prompt as data to be stored, out of the generation information and the prompt corresponding to the generation information; An information processing device comprising:

2. the prompt generator generates a prompt including medical information; The generated information is a medical image. The information processing device according to claim 1 .

3. a storage processing unit that stores the prompt as the data to be stored in the storage unit when the prompt is input to the generative model, The information processing device according to claim 1 .

4. a receiving unit that receives a user's operation to designate the generation information as the data to be stored; the storage processing unit stores the generated information as the data to be stored in the storage unit in response to an operation by the user, and stores the prompt as the data to be stored in the storage unit regardless of an operation by the user. The information processing device according to claim 3 .

5. a display control unit that causes a selection unit to be displayed on a display unit, the selection unit receiving a user's operation for selecting whether to set the generation information and / or the prompt corresponding to the generation information as the data to be stored; a storage processing unit configured to store, in the storage unit, one or both of the generation information and the prompt corresponding to the generation information in response to an operation by the user, as the data to be stored.

3. The information processing device according to claim 1.

6. the display control unit causes the display unit to display a prompt storage button for storing the prompt corresponding to the generation information as the data to be stored in the storage unit. The information processing device according to claim 5 .

7. the display control unit causes the display unit to display a generation information storage button for storing the generation information in the storage unit as the storage target data. The information processing device according to claim 5 .

8. the display control unit causes the display unit to display a prompt display button that causes the display unit to display the prompt stored in the storage unit. The information processing device according to claim 5 .

9. the generated information includes image data; the generative model outputs a plurality of the image data; a display control unit that causes a selection unit to be displayed on a display unit for each of the image data output from the generative model, the selection unit receiving a user operation to select the image data as the data to be stored; 4. The information processing device according to claim 1.

10. an evaluation unit that evaluates the necessity of storing the generated information; a determination unit that determines whether or not to set the generation information as the data to be stored based on an evaluation result by the evaluation unit, a storage processing unit that stores the prompt, or both the generation information and the prompt, as the data to be stored in the storage unit based on a determination result by the determination unit. The information processing device according to claim 1 .

11. the generated information includes information about the patient; the evaluation unit evaluates that the stronger the association between the generated information and a diagnosis result for the patient, the higher the need to store the generated information. The information processing device according to claim 10.

12. a determination unit that determines whether or not the generated information is to be the data to be stored according to a data amount of the generated information; a storage processing unit that stores the prompt, or both the generation information and the prompt, as the data to be stored in the storage unit based on a determination result by the determination unit. The information processing device according to claim 1 .

13. the storage unit is a non-volatile storage device, The data to be stored stored in the storage unit will not be deleted unless a specified deletion condition is satisfied or a deletion operation is performed by a user.

3. The information processing device according to claim 1.

14. a storage processing unit that sets the data to be stored based on the magnitude of fluctuation of the plurality of pieces of generation information generated by the generation model based on the prompt, 3. The information processing device according to claim 1.

15. a determination unit that determines whether or not the generated information is to be the data to be stored based on an evaluation result of the magnitude of the fluctuation; The information processing device according to claim 14.

16. the acquisition unit inputs the prompt stored in the storage unit as the data to be stored into the generative model, and thereby acquires reconstructed generation information reconstructed based on the prompt from the generative model. The information processing device according to claim 1 .

17. a generative model that receives a prompt and outputs information corresponding to the prompt; an information processing device, The information processing device includes: a prompt generator for generating the prompt; an acquisition unit that inputs the prompt into a generative model and acquires generation information from the generative model; a storage unit configured to store the prompt as data to be stored out of the generation information and the prompt corresponding to the generation information, Information processing system.

18. a prompt generation step for generating a prompt; an acquisition step of inputting the prompt into a generative model and acquiring generative information from the generative model; a storage processing step of storing the prompt, out of the generation information and the prompt corresponding to the generation information, in a storage unit as data to be stored; An information processing method including:

19. a prompt generation step for generating a prompt; an acquisition step of inputting the prompt into a generative model and acquiring generative information from the generative model; a storage processing step of storing the prompt, out of the generation information and the prompt corresponding to the generation information, in a storage unit as data to be stored; A program that causes a computer to execute the following.

Images