JP2021043040A - Information processing device, program and information processing method - Google Patents

Abstract

To provide an information processing device or the like capable of estimating an advancing degree of medical treatment.SOLUTION: An information processing device 1 includes a first acquisition part for acquiring a diagnostic image obtained by photographing a subject administered with a first drug indicating a radionuclide by a radiopharmaceutical synthesizer or a formulation for a preparation before use by using a positron emission tomographic device 2, a first learning model for outputting lesion estimation information if the first acquisition part inputs the acquired diagnostic image, a second acquisition part for acquiring a medial treatment image of a medial treatment stage obtained by photographing the subject administered with a second drug indicating a radionuclide by the radiopharmaceutical synthesizer or the formulation for a preparation before use by using a medical treatment image photographing device 3 on the basis of the lesion estimation information outputted by the first learning model, and a second learning model for outputting lesion estimation information of a medical treatment stage if the second acquisition part inputs the acquired medical treatment image.SELECTED DRAWING: Figure 1

Description

The present invention relates to an information processing device, a program, and an information processing method.

Automation of lesion diagnosis using image recognition technology is being promoted. Patent Document 1 discloses a computer execution method in which a region of interest (object) is specified and divided in a digitized image, numerical values related to a classification task are created from the divided objects, and a lesion is diagnosed using a learning machine. Has been done.

Japanese Patent Application Laid-Open No. 2004-536367

However, the invention according to Patent Document 1 has a problem that the progress of treatment performed after the lesion diagnosis cannot be estimated.

One aspect is to provide an information processing device or the like capable of estimating the progress of treatment.

The information processing device is the first acquisition to acquire a diagnostic image taken by using a positron radiation tomography device of a subject to which the first drug labeled with a radionuclide is administered with a radiopharmaceutical synthesizer or a preparation preparation for use. Based on the unit, the first learning model that outputs the lesion estimation information when the diagnostic image acquired by the first acquisition unit is input, and the lesion estimation information output by the first learning model, the radiopharmaceutical synthesizer or A second acquisition unit for acquiring a treatment image of a treatment stage in which a subject to which a second drug labeled with a radionuclide in a preparation for use is administered is photographed using a medical imaging device, and the second acquisition unit. It is characterized by including a second learning model that outputs lesion estimation information at the treatment stage when the treatment image acquired by the patient is input.

In one aspect, it is possible to provide an information processing device or the like capable of estimating the progress of treatment.

It is explanatory drawing which shows the outline of the lesion estimation information system. It is a block diagram which shows the configuration example of a server. It is explanatory drawing which shows an example of the record layout of the subject DB. It is explanatory drawing which shows an example of the record layout of the diagnostic information DB. It is explanatory drawing which shows an example of the record layout of the treatment information DB. It is a block diagram which shows the structural example of a PET apparatus. It is a block diagram which shows the structural example of a medical apparatus. It is a block diagram which shows the configuration example of a terminal. It is explanatory drawing explaining the operation which outputs the lesion estimation information. It is explanatory drawing explaining the lesion estimation model for diagnosis. It is explanatory drawing explaining the therapeutic lesion estimation model. It is a flowchart which shows the processing procedure at the time of outputting the lesion estimation information at a diagnosis stage. It is a flowchart which shows the processing procedure at the time of outputting the lesion estimation information at the treatment stage. It is a flowchart which shows the processing procedure at the time of determining the end of treatment using 68Ga-PSMA after performing the nuclear medicine treatment. It is a flowchart which shows the processing procedure at the time of outputting the lesion estimation information of the treatment stage of breast cancer. It is a block diagram which shows the configuration example of the server of Embodiment 2. It is explanatory drawing about the learning process which converts a therapeutic image into an estimated image. It is a flowchart which shows the processing procedure at the time of outputting an estimated image. It is a block diagram which shows the configuration example of the server of Embodiment 3. It is explanatory drawing explaining the 2nd drug information output model. It is a flowchart which shows the processing procedure at the time of outputting the recommended dose of the 2nd drug. It is explanatory drawing explaining the 2nd drug information output model of the modification 1. It is a block diagram which shows the configuration example of the server of Embodiment 4. It is explanatory drawing explaining the treatment timing prediction model. It is a flowchart which shows the processing procedure at the time of outputting the prediction result of a treatment timing. It is a flowchart which shows the processing procedure at the time of determining a treatment timing by a contrast ratio. It is a block diagram which shows the configuration example of the server of Embodiment 5. It is explanatory drawing explaining the estimation treatment result output model. It is a flowchart which shows the processing procedure at the time of outputting the estimated treatment result. It is a flowchart which shows the processing procedure at the time of determining a treatment result by a contrast ratio. It is a functional block diagram which shows the operation of the server of Embodiments 1-5.

Hereinafter, the present invention will be described in detail with reference to the drawings showing the embodiments thereof.

(Embodiment 1)
The first embodiment relates to a mode in which lesion estimation information is output using artificial intelligence (AI). FIG. 1 is an explanatory diagram showing an outline of a lesion estimation information system. The system of the present embodiment includes an information processing device 1, a positron emission tomography device 2, a medical imaging device 3, and an information processing terminal 4, and each device transmits and receives information via a network N such as the Internet.

The information processing device 1 is an information processing device that processes, stores, and transmits / receives various types of information. The information processing device 1 is, for example, a server device, a personal computer, a general-purpose tablet PC (personal computer), or the like. In the present embodiment, the information processing device 1 is assumed to be a server device, and will be replaced with the server 1 below for the sake of brevity.

The positron emission tomography apparatus 2 is an apparatus for performing positron emission tomography. Positron Emission Tomography (PET) is a computed tomography technique that utilizes positron detection. In the present embodiment, the positron emission tomography apparatus 2 will be referred to as the PET apparatus 2 for the sake of brevity below. Similarly, an inspection using the PET device 2 is called a PET inspection. A PET scan can be used to photograph where and how large a lesion is on the whole body.

Specifically, the first drug is administered by infusion or injection to the body of the patient to be examined (hereinafter referred to as a subject). The first drug is a diagnostic drug used before the start of treatment. The first drug is a drug in which a ligand is labeled with a radionuclide, and is synthesized using a radiopharmaceutical synthesizer or a preparation preparation (kit) for use. Formulations that are easily altered should be prepared before use.

In the present embodiment, the radionuclide is a nuclide that emits positrons such as gallium-68. The ligand is a molecule that specifically binds to an antigen expressed at a lesion site such as cancer, and is, for example, a prostate specific Membrane Antigen (PSMA) -binding ligand, a monoclonal antibody against PSMA, a HER2 antibody, or the like. is there. Prostate-specific membrane antigen-binding ligands and monoclonal antibodies targeting PSMA bind to cancer cells such as prostate cancer. HER2 antibody (receptor tyrosine kinase) binds to cancer cells such as breast cancer.

By administering the first drug to the subject, the first drug accumulates at the lesion site and emits positrons and the like. When the PET examination is performed in this state, a portion that emits positrons and the like is visualized. Therefore, a lesion site such as prostate cancer or breast cancer can be found.

The medical imaging apparatus 3 is an apparatus for photographing a medical image, for example, a positron emission tomography apparatus (PET apparatus), a single photon emission tomography apparatus, a Compton camera, or the like. In the present embodiment, the medical imaging device 3 will be referred to as the medical device 3 in the following for the sake of brevity.

The single photon emission tomography device is a device for single photon emission tomography. Single photon emission computed tomography is one of the diagnostic imaging methods, and it is an application of scintigraphy to detect positrons emitted from radioisotopes administered into the body. The distribution is a tomographic image.

A Compton camera is a device that measures the incident direction, energy, and intensity of positrons, etc. by utilizing the Compton effect, which is a phenomenon in which high-energy electromagnetic waves such as positrons are scattered by electrons and become longer than the original wavelength. By simultaneously detecting the change in wavelength of scattered positrons and the recoil direction of electrons, it is possible to image positrons and the like.

When a lesion is diagnosed by PET examination, various treatments are performed depending on the diagnosis result. In this embodiment, nuclear medicine treatment, that is, nuclear medicine treatment using an unsealed radioisotope is performed.

Specifically, the second drug is administered to the subject. The second drug is a drug for the treatment of nuclear medicine. The second drug is a drug in which the same ligand as the first drug is labeled with a radionuclide. The radionuclide of the second drug is a nuclide that emits α-rays or β-rays and damages nearby cells.

By administering the second drug to the subject, the second drug accumulates at the lesion site and emits α-rays or β-rays. The cells at the lesion site are damaged by α or β rays and become necrotic. Therefore, the lesion can be treated.

After the second drug is administered, the progress of the treatment can be photographed by photographing the subject using the medical device 3. A specific example will be described below.

For example, lutetium 177 can be used as the radionuclide of the second drug. Lutetium 177 emits β-rays and γ-rays. Therefore, by using the medical device 3 capable of photographing γ-rays, it is possible to photograph the lesion site where the second drug is accumulated.

When a prostate-specific membrane antigen-binding ligand is used as the first drug, a prostate-specific membrane antigen-binding ligand (177Lu-PSMA) labeled with lutetium-177 is used as the second drug. When the HER2 antibody is used as the first drug, the HER2 antibody (177Lu-HER2) labeled with lutetium-177 is used as the second drug.

For example, astatine 211 can be used as the radionuclide of the second drug. Astatine 211 emits X-rays. For example, by using a medical device 3 such as a CT (Computed Tomography), a single photon emission tomography device, or a Computon camera, the site being treated can be imaged with a second drug.

When a prostate-specific membrane antigen-binding ligand is used as the first drug, a prostate-specific membrane antigen-binding ligand (211 At-PSMA) labeled with astatine 211 is used as the second drug. When the HER2 antibody is used as the first drug, the HER2 antibody (211 At-HER2) labeled with astatine 211 is used as the second drug.

After a series of radionuclide nuclear medicine treatments are completed, a second diagnostic agent may be administered and the treatment status may be observed. In this case, the same drug as the first drug is used for the second drug, and a device capable of photographing γ-rays is used for the medical device 3.

The information processing terminal 4 is a terminal device that receives or displays lesion estimation information, and is, for example, an information processing device such as a smartphone, a mobile phone, a tablet, or a personal computer terminal. In the following, for the sake of brevity, the information processing terminal 4 will be read as the terminal 4.

The server 1 according to the present embodiment acquires a diagnostic image taken by using the PET device 2. The server 1 inputs the acquired diagnostic image and outputs the lesion estimation information by using the diagnostic lesion estimation model described later. The server 1 acquires a treatment image of the treatment stage taken by using the medical device 3 based on the lesion estimation information output by the diagnostic lesion estimation model. The server 1 inputs the acquired treatment image using the therapeutic lesion estimation model described later, and outputs the lesion estimation information at the treatment stage.

FIG. 2 is a block diagram showing a configuration example of the server 1. The server 1 includes a control unit 11, a storage unit 12, a communication unit 13, an input unit 14, a display unit 15, a reading unit 16, and a large-capacity storage unit 17. Each configuration is connected by bus B.

The control unit 11 includes arithmetic processing units such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), and a GPU (Graphics Processing Unit), and reads and executes the control program 1P stored in the storage unit 12. , Performs various information processing, control processing, etc. related to the server 1. Although the control unit 11 is described as a single processor in FIG. 2, it may be a multiprocessor.

The storage unit 12 includes memory elements such as a RAM (Random Access Memory) and a ROM (Read Only Memory), and stores a control program 1P or data required for the control unit 11 to execute processing. In addition, the storage unit 12 temporarily stores data and the like necessary for the control unit 11 to execute arithmetic processing. The communication unit 13 is a communication module for performing processing related to communication, and transmits / receives information to / from the PET device 2, the medical device 3, or the terminal 4 via the network N.

The input unit 14 is an input device such as a mouse, a keyboard, a touch panel, and a button, and outputs the received operation information to the control unit 11. The display unit 15 is a liquid crystal display, an organic EL (electroluminescence) display, or the like, and displays various information according to the instructions of the control unit 11.

The reading unit 16 reads a portable storage medium 1a including a CD (Compact Disc) -ROM or a DVD (Digital Versatile Disc) -ROM. The control unit 11 may read the control program 1P from the portable storage medium 1a via the reading unit 16 and store it in the large-capacity storage unit 17. Further, the control unit 11 may download the control program 1P from another computer via the network N or the like and store it in the large-capacity storage unit 17. Furthermore, the control unit 11 may read the control program 1P from the semiconductor memory 1b.

The large-capacity storage unit 17 includes, for example, a recording medium such as an HDD (Hard disk drive) or an SSD (Solid State Drive). The large-capacity storage unit 17 includes a subject DB (database) 171, a diagnostic information DB 172, a treatment information DB 173, a diagnostic lesion estimation model (first learning model) 174, and a therapeutic lesion estimation model 175 (second learning model). including.

The subject DB 171 stores information about the subject. The diagnostic information DB 172 stores the diagnostic information of the subject. The treatment information DB 173 stores the treatment information of the subject. The diagnostic lesion estimation model 174 is an estimator that estimates lesion information based on a diagnostic image, and is a trained model generated by machine learning. The therapeutic lesion estimation model 175 is an estimator that estimates lesion information at the treatment stage based on a treatment image, and is a trained model generated by machine learning.

In this embodiment, the storage unit 12 and the large-capacity storage unit 17 may be configured as an integrated storage device. Further, the large-capacity storage unit 17 may be composed of a plurality of storage devices. Furthermore, the large-capacity storage unit 17 may be an external storage device connected to the server 1.

In the present embodiment, the server 1 is described as one information processing device, but it may be distributed and processed by a plurality of servers, or it may be configured by a virtual machine.

FIG. 3 is an explanatory diagram showing an example of the record layout of the subject DB 171. The subject DB 171 includes a subject ID row, a lesion type row, a diagnosis ID row, and a treatment ID row. The subject ID column stores the ID of the subject uniquely identified in order to identify each subject. The lesion type column stores the lesion type information of the subject. The diagnostic ID column stores the ID of the diagnostic information that identifies the diagnostic information. The treatment ID column stores the ID of the treatment information that identifies the treatment information.

FIG. 4 is an explanatory diagram showing an example of the record layout of the diagnostic information DB 172. The diagnostic information DB 172 includes a diagnostic ID column, a subject ID column, a diagnostic image sequence, an imaging date / time sequence, an administered drug sequence, and a lesion information sequence. The diagnostic ID column stores the ID of the diagnostic information uniquely identified in order to identify each diagnostic information (diagnosis data). The subject ID column stores the subject ID that identifies the subject.

The diagnostic image sequence stores the diagnostic image. The shooting date / time column stores information on the date and time when the diagnostic image was taken. The administered drug sequence stores the name of the first drug administered to the subject. The lesion information column stores diagnostic information for the lesion of the subject.

FIG. 5 is an explanatory diagram showing an example of the record layout of the treatment information DB 173. The treatment information DB 173 includes a treatment ID column, a sub ID column, a subject ID column, a treatment image column, an imaging date / time column, an administered drug column, a lesion information column, and a remarks column. The treatment ID column stores the ID of the treatment information uniquely specified in order to identify each treatment information (treatment data). The sub ID column stores the child ID of the treatment ID. In the sub-ID column, an identifier for identifying each treatment information according to the number of treatments is stored for the same treatment course (treatment ID). The subject ID column stores the subject ID that identifies the subject.

The treatment image sequence stores the treatment image of the treatment stage taken by using the medical device 3. The shooting date / time column stores information on the date and time when the treatment image was taken. The administered drug sequence stores the name of the second drug administered to the subject. The lesion information column stores the lesion information of the subject at the treatment stage. The remarks column stores information such as comments or supplements for treatment information.

FIG. 6 is a block diagram showing a configuration example of the PET device 2. The PET device 2 includes a control unit 21, a storage unit 22, a communication unit 23, an input unit 24, a display unit 25, a photographing unit 26, a detector 27, and a unit board 28. Each configuration is connected by bus B.

The control unit 21 includes arithmetic processing units such as a CPU and an MPU, and performs various information processing, control processing, and the like related to the PET device 2 by reading and executing the control program 2P stored in the storage unit 22. Although the control unit 21 is described as a single processor in FIG. 6, it may be a multiprocessor. The storage unit 22 includes memory elements such as RAM and ROM, and stores the control program 2P or data required for the control unit 21 to execute the process. In addition, the storage unit 22 temporarily stores data and the like necessary for the control unit 21 to execute arithmetic processing.

The communication unit 23 is a communication module for performing processing related to communication, and transmits / receives information to / from the server 1 or the like via the network N. The input unit 24 may be an operation button, a keyboard or a mouse. The display unit 25 is a liquid crystal display, an organic EL display, or the like, and displays various information according to the instructions of the control unit 21.

The photographing unit 26 is, for example, a photographing device such as a CCD (Charge Coupled Device) camera or a CMOS (Complementary Metal Oxide Semiconductor) camera. The photographing unit 26 may be composed of a plurality of photographing devices. The detector 27 is a semiconductor γ-ray detector for detecting positrons and the like emitted from a subject. A large number of semiconductor γ-ray detectors can be provided on the unit substrate 28.

The unit substrate 28 has an integrated circuit (ASIC) for measuring γ-ray detection energy and detection time, and measures the detected γ-ray detection energy and detection time, or detects γ-rays. The detected γ-ray information is output by detecting the address of the device 27.

FIG. 7 is a block diagram showing a configuration example of the medical device 3. The medical device 3 includes a control unit 31, a storage unit 32, a communication unit 33, an input unit 34, a display unit 35, an imaging unit 36, a detector 37, and a unit board 38. Each configuration is connected by bus B.

The control unit 31 includes arithmetic processing units such as a CPU and an MPU, and performs various information processing, control processing, and the like related to the medical device 3 by reading and executing the control program 3P stored in the storage unit 32. Although the control unit 31 is described as a single processor in FIG. 7, it may be a multiprocessor. The storage unit 32 includes memory elements such as RAM and ROM, and stores the control program 3P or data required for the control unit 31 to execute the process. In addition, the storage unit 32 temporarily stores data and the like necessary for the control unit 31 to execute arithmetic processing.

The communication unit 33 is a communication module for performing processing related to communication, and transmits / receives information to / from the server 1 or the like via the network N. The input unit 34 may be an operation button, a keyboard or a mouse. The display unit 35 is a liquid crystal display, an organic EL display, or the like, and displays various information according to the instructions of the control unit 31.

The photographing unit 36 is a photographing device such as a CCD camera or a CMOS camera. The photographing unit 36 may be composed of a plurality of photographing devices. The detector 37 is a semiconductor radiation detector for detecting radiation emitted from a subject (for example, β-rays or γ-rays). A large number of semiconductor radiation detectors can be provided on the unit substrate 38.

The unit substrate 38 has an integrated circuit (ASIC) for measuring the radiation detection energy and the detection time, and measures the detected radiation detection energy and the detection time, and the detector 37 that detects the radiation. It detects the address and outputs the detected radiation information.

FIG. 8 is a block diagram showing a configuration example of the terminal 4. The terminal 4 includes a control unit 41, a storage unit 42, a communication unit 43, an input unit 44, and a display unit 45. Each configuration is connected by bus B.

The control unit 41 includes arithmetic processing units such as a CPU and an MPU, and performs various information processing, control processing, and the like related to the terminal 4 by reading and executing the control program 4P stored in the storage unit 42. Although the control unit 41 is described as a single processor in FIG. 8, it may be a multiprocessor. The storage unit 42 includes memory elements such as RAM and ROM, and stores the control program 4P or data required for the control unit 41 to execute the process. In addition, the storage unit 42 temporarily stores data and the like necessary for the control unit 41 to execute arithmetic processing.

The communication unit 43 is a communication module for performing processing related to communication, and transmits / receives information to / from the server 1 or the like via the network N. The input unit 44 may be a keyboard, a mouse, or a touch panel integrated with the display unit 45. The display unit 45 is a liquid crystal display, an organic EL display, or the like, and displays various information according to the instructions of the control unit 41.

FIG. 9 is an explanatory diagram illustrating an operation of outputting lesion estimation information. The control unit 11 of the server 1 acquires a diagnostic image taken by the PET device 2 via the communication unit 13. In the present embodiment, an example in which the diagnostic image is acquired from the PET device 2 via the network has been described, but the present invention is not limited to this. For example, the diagnostic image may be transferred to the server 1 via the memory card.

The control unit 11 inputs the acquired diagnostic image to the diagnostic lesion estimation model 174 that outputs the lesion estimation information when the acquired diagnostic image is input, and outputs the estimation result of estimating the lesion information of the subject. The lesion information is, for example, the shape of the cancer, the extent of the spread (invasion or metastasis) of the cancer, or the degree of malignancy. The following describes an example of the extent of the spread of prostate cancer. The extent to which prostate cancer has spread can be categorized by the TNM classification, which assesses how widespread the cancer is.

The TNM classification is a standard for determining the degree of cancer progression, and is a staging system adopted by the Union for International Cancer Control (UICC), which indicates the degree of cancer progression. T indicates the size, spread, and depth (infiltration) of the cancer that has developed, N indicates the degree of spread of the cancer to the surrounding lymph nodes, and M indicates the presence or absence of distant metastasis. T is 5 levels from 0 to 4, N is 2 levels from 0 to 1, and M is 2 levels from 0 to 1. 0 means "nothing".

An example of classifying the extent of prostate cancer into T0, T1, T2, T3, T4, N0, N1, M0 and M1 by the above-mentioned TNM classification will be described. T0 is the absence of cancer. T1 is a localized cancer found by chance or by biopsy. T2 is a cancer that resembles a round prostate in half and has a small area in one lobe, a cancer that covers more than half of one leaf, or a cancer that spreads to both lobes. T3 is a cancer that crosses the membrane that covers the prostate or has seminal vesicle infiltration. T4 is a cancer that invades the bladder and rectum near the prostate. N is the presence or absence of regional lymph node metastasis, N0 is no metastasis, and N1 is metastasis. M is the presence or absence of metastasis (distant metastasis) to a distant place (for example, liver, lung or bone), M0 is no metastasis, and M1 is metastasis.

The process of estimating lesion information will be described later. The control unit 11 stores the lesion estimation information output by the diagnostic lesion estimation model 174 in the subject DB 171 and the diagnostic information DB 172 of the large-capacity storage unit 17. The control unit 11 transmits the lesion estimation information to the terminal 4 via the communication unit 13. In the present embodiment, an example in which the lesion estimation information is transmitted to the terminal 4 has been described, but the present invention is not limited to this. For example, the control unit 11 may display the lesion estimation information on the display unit 15.

The control unit 11 acquires a treatment image of the treatment stage taken by using the medical device 3 based on the lesion estimation information output by the diagnostic lesion estimation model 174.

The doctor who uses the terminal 4 decides the treatment policy based on the displayed lesion estimation information. In the following description of this embodiment, the treatment stage when it is decided to perform RI internal therapy will be described.

In the treatment phase, the second agent is administered multiple times. In the present embodiment, the dose, the number of administrations, and the like are determined by the doctor based on his / her specialized knowledge. By administering the second drug, nuclear medicine treatment is performed in which the cells at the lesion site are necrotic by radiation.

After administering the second drug, the doctor appropriately takes a treatment image using the medical device 3. For example, if there are many systemic metastases or if the lesion is large, the doctor will take a treatment image every time the second drug is administered two or three times. If there are few systemic metastases or the lesions are small, the doctor will take a treatment image after each dose or two doses of the second drug.

The doctor administers the same drug as the first drug as needed and takes a treatment image. Since a clear treatment image can be taken, lesion information at the treatment stage can be estimated with high accuracy. For example, by administering the same drug as the first drug and taking a treatment image immediately before determining the end of treatment, it is possible to avoid overlooking small lesions.

The control unit 11 inputs the acquired treatment image to the therapeutic lesion estimation model 175 that outputs the lesion estimation information at the treatment stage when the acquired treatment image is input, and estimates the lesion information at the treatment stage. Output the result. The process of estimating lesion information at the treatment stage will be described later. The control unit 11 stores the lesion estimation information at the treatment stage output by the therapeutic lesion estimation model 175 in the subject DB 171 and the treatment information DB 173 of the large-capacity storage unit 17. The control unit 11 transmits the lesion estimation information at the treatment stage to the terminal 4 via the communication unit 13.

Subsequently, the lesion information estimation process using the diagnostic lesion estimation model 174 will be described. FIG. 10 is an explanatory diagram illustrating a diagnostic lesion estimation model 174. The diagnostic lesion estimation model 174 is used as a program module that is part of artificial intelligence software. The diagnostic lesion estimation model 174 is an estimator for which a neural network has been constructed that inputs a diagnostic image taken by using the PET device 2 and outputs the result of estimating the lesion information.

The control unit 11 of the server 1 constructs (generates) a diagnostic lesion estimation model 174 as a diagnostic lesion estimation model 174 by performing deep learning to learn the feature amount of the lesion portion in the diagnostic image. For example, the diagnostic lesion estimation model 174 is a CNN (Convolution Neural Network), which includes an input layer that accepts input of diagnostic images, an output layer that outputs the result of estimating lesion information, and an intermediate layer that has been trained by backpropagation. And have.

The input layer has a plurality of neurons that receive input of the pixel value of each pixel included in the diagnostic image, and passes the input pixel value to the intermediate layer. The intermediate layer has a plurality of neurons for extracting the feature amount of the diagnostic image, and the extracted image feature amount is passed to the output layer. For example, the case where the diagnostic lesion estimation model 174 is CNN will be described as an example. The intermediate layer is composed of a convolution layer that convolves the pixel values of each pixel input from the input layer and a pooling layer that maps the pixel values convoluted by the convolution layer. Finally, the feature amount of the image is extracted while compressing the information. The intermediate layer then estimates (predicts) lesion information from the fully connected layer whose parameters have been learned by backpropagation. The estimation result is output to the output layer having a plurality of neurons.

The diagnostic image may be input to the input layer after the feature amount is extracted by passing through the convolution layer and the pooling layer which are alternately connected.

Instead of CNN, RCNN (Regions with Convolutional Neural Network), Fast RCNN, Faster RCNN, Mask RCNN, SSD (Single Shot Multibook Detector), YOLO (You Only Look Once), SVM (Support Vector Machine), Bayesian Network , Or any object detection algorithm such as a regression tree may be used.

For example, the control unit 11 acquires an identification result for identifying a lesion site (for example, stomach, large intestine, prostate, etc.) using a learned site identification model constructed by an SSD object detection algorithm based on a diagnostic image. Based on the lesion site identification result, the control unit 11 inputs a diagnostic image into the learned lesion estimation model constructed by the SSD object detection algorithm for each site, and outputs the estimation result of estimating the lesion. Since the lesion location is known in the diagnostic image, the image of the lesion region cut out from the diagnostic image may be input to the learned lesion estimation model described above. By using the image of the lesion area, it is possible to improve the accuracy of the estimation result of estimating the lesion information and prevent the lesion part from being overlooked.

At the time of learning, a large amount of past diagnostic data (for example, diagnostic image) existing in a medical institution is used as teacher data. The control unit 11 of the server 1 performs learning by using a combination of the diagnostic image and the teacher data associated with the lesion information corresponding to the diagnostic image. The teacher data is data in which lesion information (for example, T1, T2, T3, etc.) corresponding to the diagnostic image is labeled on the diagnostic image. The control unit 11 inputs the diagnostic image, which is the teacher data, to the input layer, performs arithmetic processing in the intermediate layer, and acquires the estimation result of estimating the lesion information from the output layer.

The estimation result output from the output layer may be a value discretely indicating the presence or absence of a lesion (for example, a value of "0" or "1"), and a continuous probability value (for example, "0" to "1"). It may be a value in the range up to 1 ”). The control unit 11 acquires, for example, an estimation result of estimating the shape, size, spread degree, malignancy, etc. of the cancer as the estimation result output from the output layer.

As the teacher data, a combination of teacher data stored in advance in the storage unit 12 or the large-capacity storage unit 17 may be acquired. Alternatively, the control unit 11 may generate teacher data for learning the lesion information by labeling the stage information for each of the diagnostic images in the plurality of diagnostic images. Specifically, the control unit 11 generates teacher data in which, for example, the name of the extent of the spread of prostate cancer (for example, T1 or T2, etc.) is associated with each diagnostic image. The control unit 11 inputs the diagnostic image, which is the teacher data, to the input layer, performs arithmetic processing in the intermediate layer, and acquires the estimation result of estimating the lesion information corresponding to the diagnostic image from the output layer.

The control unit 11 compares the estimation result output from the output layer with the information labeled for the diagnostic image in the teacher data, that is, the correct answer value, and intermediates the output value from the output layer so as to approach the correct answer value. Optimize the parameters used for arithmetic processing in the layer. The parameters are, for example, the weight between neurons (coupling coefficient), the coefficient of the activation function used in each neuron, and the like. The method of optimizing the parameters is not particularly limited, but for example, the control unit 11 optimizes various parameters by using the backpropagation method.

The control unit 11 performs the above processing on each diagnostic image included in the teacher data to generate a diagnostic lesion estimation model 174. Thereby, for example, the control unit 11 can build a model capable of estimating the lesion information corresponding to the diagnostic image by learning the lesion estimation model 174 for diagnosis using the teacher data.

When the control unit 11 acquires the diagnostic image, the control unit 11 inputs the acquired diagnostic image into the diagnostic lesion estimation model 174. The control unit 11 performs arithmetic processing for extracting the feature amount of the diagnostic image in the intermediate layer of the diagnostic lesion estimation model 174. The control unit 11 inputs the extracted feature amount to the output layer of the lesion estimation model 174 for diagnosis, and acquires the estimation result of estimating the lesion information corresponding to the diagnostic image as an output.

As shown in the figure, as an example of the lesion estimation information, the estimation result output from the output layer indicates the extent of the spread of prostate cancer, and is a continuous probability value of each of T, N, and M. Specifically, with respect to the diagnostic image, the probability values of "T0", "T1", "T2", "T3", and "T4" are "0.02", "0.02", and "0. The estimation results of "05", "0.02", and "0.89" are output. The estimation result in which the probability value of "N0" is "0.10" and the probability value of "N1" is "0.90" is output. The estimation result in which the probability value of "M0" is "0.85" and the probability value of "M1" is "0.15" is output.

Further, the estimation result may be output by using a predetermined threshold value. For example, in the control unit 11, the probability value (0.89) of “T4” is equal to or higher than a predetermined threshold value (for example, 0.85), and the probability value (0.90) of “N1” is a predetermined threshold value (for example, 0. When it is determined that the value is 85) or more and the probability value (0.85) of “M0” is equal to or more than a predetermined threshold value (for example, 0.80), the degree of spread of prostate cancer corresponding to the diagnostic image is “T4”. , "N1" and "M0". That is, the control unit 11 gives an estimated result of "a cancer infiltrating the bladder or rectum close to the prostate, having lymph node metastasis, and no distant metastasis (liver, lung, bone, etc.)" to the diagnostic image. Output. In addition, instead of using the above-mentioned threshold value, the highest probability value of each of T, N, and M may be output as an estimation result from the probability values of T, N, and M estimated by the diagnostic lesion estimation model 174. ..

A diagnostic lesion estimation model for each of T, N, and M may be constructed. For example, when an N estimation model for estimating the presence or absence of regional lymph node metastasis is constructed, the control unit 11 inputs the acquired diagnostic image into the N estimation model. The control unit 11 performs arithmetic processing for extracting the feature amount of the diagnostic image in the intermediate layer of the N estimation model. The control unit 11 inputs the extracted feature amount to the output layer of the N estimation model, and acquires the estimation result of estimating the presence or absence of regional lymph node metastasis corresponding to the diagnostic image as an output.

Note that FIG. 10 has described an example in which a diagnostic image is input to the diagnostic lesion estimation model 174, but the present invention is not limited to this. For example, in addition to the diagnostic image, information about the patient to which the diagnostic image belongs may be input to the diagnostic lesion estimation model 174. Information about the patient is, for example, the patient's gender, age, past diagnosis history, and the like.

The lesion information estimation process is not limited to the identification process by the machine learning described above. For example, local feature extraction methods such as A-KAZE (Accelerated KAZE), SIFT (Scale Invariant Feature Transform), SURF (Speeded-Up Robust Features), ORB (Oriented FAST and Rotated BRIEF), and HOG (Histograms of Oriented Gradients). The lesion information may be estimated using.

Note that FIG. 10 has described an example in which the diagnostic lesion estimation model 174 outputs an estimation result of the extent of prostate cancer spread, but the present invention is not limited to this. It can also be applied to other types of cancer (eg, gastric cancer, breast cancer or colorectal cancer, etc.). In addition to the extent of cancer spread, the shape, size, malignancy, etc. of the cancer may be output.

Subsequently, the process of estimating the lesion information at the treatment stage using the therapeutic lesion estimation model 175 will be described. FIG. 11 is an explanatory diagram illustrating a therapeutic lesion estimation model 175. The therapeutic lesion estimation model 175 is utilized as a program module that is part of artificial intelligence software. The therapeutic lesion estimation model 175 is an estimator for which a neural network has been constructed that inputs a treatment image of a treatment stage taken by using a medical device 3 and outputs a result of estimating lesion information at the treatment stage. Since the lesion information at the treatment stage is the same as the lesion information at the diagnosis stage, the description thereof will be omitted.

The control unit 11 of the server 1 constructs (generates) the therapeutic lesion estimation model 175 as the therapeutic lesion estimation model 175 by performing deep learning to learn the feature amount of the lesion portion in the treatment image. For example, the therapeutic lesion estimation model 175 is a CNN, which includes an input layer that accepts input of a treatment image, an output layer that outputs the result of estimating lesion information at the treatment stage, and an intermediate layer that has been learned by back propagation. Has. Since the configuration of the therapeutic lesion estimation model 175 is the same as the configuration of the diagnostic lesion estimation model 174, the description thereof will be omitted.

At the time of learning, a large amount of past treatment data (for example, treatment image) existing in a medical institution is used as teacher data. The control unit 11 performs learning by using a combination of the treatment image and the teacher data in which the lesion information at the treatment stage corresponding to the treatment image is associated with each other. The teacher data is data in which the treatment image is labeled with the lesion information at the treatment stage corresponding to the treatment image. The control unit 11 inputs the treatment image, which is the teacher data, to the input layer, performs arithmetic processing in the intermediate layer, and acquires the estimation result of estimating the lesion information at the treatment stage from the output layer.

The control unit 11 compares the estimation result output from the output layer with the information labeled for the treatment image in the teacher data, that is, the correct answer value, and intermediately so that the output value from the output layer approaches the correct answer value. Optimize the parameters used for arithmetic processing in the layer. The control unit 11 performs the above processing on each treatment image included in the teacher data, and generates a treatment lesion estimation model 175. Thereby, for example, the control unit 11 can learn the therapeutic lesion estimation model 175 using the teacher data, and can construct a model capable of estimating the lesion information at the treatment stage corresponding to the therapeutic image.

When the control unit 11 acquires the treatment image, the control unit 11 inputs the acquired treatment image into the therapeutic lesion estimation model 175. The control unit 11 performs arithmetic processing for extracting the feature amount of the treatment image in the intermediate layer of the treatment lesion estimation model 175. The control unit 11 inputs the extracted feature amount to the output layer of the therapeutic lesion estimation model 175, and acquires the estimation result of estimating the lesion information at the treatment stage corresponding to the therapeutic image as an output.

Since the lesion location is known in the treatment image, an image of the lesion region cut out from the treatment image may be input to the treatment lesion estimation model 175. In addition to the treatment image, the control unit 11 may input information about the patient to which the treatment image belongs into the treatment lesion estimation model 175. Information about the patient is, for example, the patient's gender, age, treatment history, and the like. A therapeutic lesion estimation model for each of T, N, and M may be constructed.

Instead of CNN, any object detection algorithm such as RCNN, Fast RCNN, Faster RCNN, Mask RCNN, SSD, YOLO, SVM, Bayesian network, or regression tree may be used.

In the following, a processing procedure for outputting prostate cancer lesion estimation information will be described as an example.

FIG. 12 is a flowchart showing a processing procedure when outputting lesion estimation information at the diagnostic stage. The control unit 11 of the server 1 acquires a diagnostic image taken by using the PET device 2 of the subject to which the first drug has been administered via the communication unit 13 (step S101). 68Ga-PSMA is used as the first drug. The control unit 11 estimates the lesion information of the subject by using the diagnostic lesion estimation model 174 that outputs the lesion estimation information when the acquired diagnostic image is input (step S102).

The control unit 11 stores the estimated lesion estimation information in the large-capacity storage unit 17 (step S103). Specifically, the control unit 11 allocates a diagnostic ID, and records the subject ID, the diagnostic image, the shooting date and time of the diagnostic image, the drug administered to the subject, and the lesion estimation information as one record in the diagnostic information DB 172. Remember. The control unit 11 stores the lesion type (for example, prostate cancer, breast cancer, etc.) and the diagnosis ID as one record in the subject DB 171 in association with the subject ID.

The control unit 11 transmits the lesion estimation information to the terminal 4 via the communication unit 13 (step S104). The control unit 41 of the terminal 4 receives the lesion estimation information transmitted from the server 1 by the communication unit 43 (step S401), displays the received lesion estimation information by the display unit 45 (step S402), and ends the process. ..

FIG. 13 is a flowchart showing a processing procedure when outputting lesion estimation information at the treatment stage. The control unit 11 of the server 1 acquires a treatment image of the treatment stage taken by using the medical device 3 for the subject to which the second drug is administered (step S111). As the second drug, 177Lu-PSMA is used. The control unit 11 estimates the lesion information at the treatment stage by using the treatment lesion estimation model 175 that outputs the lesion estimation information at the treatment stage when the acquired treatment image is input (step S112).

The control unit 11 stores the lesion estimation information at the estimated treatment stage in the large-capacity storage unit 17 (step S113). Specifically, when the treatment ID does not exist, the control unit 11 allocates the treatment ID. If a treatment ID exists for the same treatment course, the control unit 11 identifies the treatment ID. The control unit 11 assigns a sub ID in association with the treatment ID, and sets the subject ID, the treatment image, the shooting date and time of the treatment image, the drug administered to the subject, the lesion estimation information at the treatment stage, and remarks as one. It is stored in the treatment information DB 173 as a record. The control unit 11 stores the treatment ID in the subject DB 171 in association with the subject ID and the diagnosis ID.

The control unit 11 transmits the lesion estimation information at the treatment stage to the terminal 4 via the communication unit 13 (step S114). The control unit 41 of the terminal 4 receives the lesion estimation information at the treatment stage transmitted from the server 1 by the communication unit 43 (step S411), and displays the received lesion estimation information at the treatment stage by the display unit 45 (step S411). Step S412).

FIG. 14 is a flowchart showing a processing procedure when determining the end of treatment using 68Ga-PSMA after performing nuclear medicine treatment. The contents overlapping with FIG. 13 are designated by the same reference numerals and the description thereof will be omitted.

The control unit 11 receives an instruction as to whether or not to execute the treatment end determination via the communication unit 13 or the input unit 14 (step S115). When the control unit 11 receives an instruction not to execute the treatment end determination (NO in step S115), the control unit 11 ends the process. When the control unit 11 receives an instruction to execute the treatment end determination (YES in step S115), the control unit 11 acquires a treatment image of the treatment stage taken by using the medical device 3 for the subject to which 68Ga-PSMA has been administered. (Step S116).

The control unit 11 estimates the lesion information at the treatment stage using the therapeutic lesion estimation model 175 (step S117). The control unit 11 stores the estimated lesion estimation information at the treatment stage in the subject DB 171 and the treatment information DB 173 of the large-capacity storage unit 17 (step S118). The control unit 11 transmits the lesion estimation information at the treatment stage to the terminal 4 via the communication unit 13 (step S119). The control unit 41 of the terminal 4 executes step S411.

In the following, a processing procedure for outputting breast cancer lesion estimation information will be described as an example. As the first drug for obtaining a diagnostic image, a HER2 antibody labeled Ga-68 with a radiopharmaceutical synthesizer and kit is used. The control unit 11 of the server 1 acquires a diagnostic image of the subject to which the first drug has been administered, taken by using the PET device 2. Since the process of outputting the lesion estimation information at the diagnosis stage is the same as the process procedure shown in FIG. 12, the description thereof will be omitted.

FIG. 15 is a flowchart showing a processing procedure when outputting lesion estimation information at the treatment stage of breast cancer. The contents overlapping with FIG. 13 are designated by the same reference numerals and the description thereof will be omitted. The control unit 11 of the server 1 acquires a treatment image of the treatment stage taken by using the medical device 3 for the subject to which the second drug is administered (step S121). 211At-HER2 is used as the second drug. After that, the control unit 11 executes step S112.

Further, the control unit 11 of the server 1 can simultaneously display the diagnostic image and the treatment image via the display unit 15. Furthermore, the control unit 11 can transmit the diagnostic image and the treatment image to the terminal 4 via the communication unit 13.

According to this embodiment, it is possible to output lesion estimation information using the diagnostic lesion estimation model 174 based on the diagnostic image.

According to this embodiment, it is possible to output lesion estimation information at the treatment stage using the therapeutic lesion estimation model 175 based on the treatment image.

According to this embodiment, it is possible to solve the problem of shortage of doctors by outputting lesion estimation information using artificial intelligence.

According to the present embodiment, it is possible to improve the estimation accuracy of the lesion information by performing deep learning for learning the feature amount for the diagnostic image or the therapeutic image.

(Embodiment 2)
The second embodiment relates to a form of acquiring an estimated image obtained by estimating an image of a subject to which the first drug has been administered using a PET device 2 based on a therapeutic image using artificial intelligence. The description of the contents overlapping with the first embodiment will be omitted.

FIG. 16 is a block diagram showing a configuration example of the server 1 of the second embodiment. The contents overlapping with FIG. 2 are designated by the same reference numerals and the description thereof will be omitted. The conversion model 176 is stored in the large-capacity storage unit 17. The conversion model (third learning model) 176 is a converter that converts (estimates) a subject to which the first drug is administered into an image taken by using the PET device 2 based on the treatment image, and is machine learning. Is a trained model generated by.

FIG. 17 is an explanatory diagram regarding a learning process for converting a treatment image into an estimated image. In the present embodiment, the server 1 learns the treatment image by using the method of GAN (Generative Adversarial Network) and generates the conversion model 176. FIG. 17 conceptually illustrates the configuration of the GAN.

The GAN is composed of a generator that generates output data from input data and a discriminator that identifies the authenticity of the data generated by the generator. The generator accepts random noise (latent variables) or therapeutic image input and produces output data. The discriminator learns the truth of the input data by using the true data given for learning and the data given by the generator. In GAN, the generator and the discriminator compete with each other for learning, and finally the network is constructed so that the loss function of the generator is minimized and the loss function of the discriminator is maximized.

In the present embodiment, the server 1 uses the generator constructed by the GAN method as a converter constituting the conversion model 176, and is administered a Lu-177-labeled PSMA-binding ligand (second drug). From the therapeutic image (177Lu-PSMA) obtained by photographing the subject, the image (68Ga-PSMA) obtained by administering the subject to which the PSMA-binding ligand (first drug) labeled with Ga-68 was administered using the PET device 2 ) (Estimated image).

In this embodiment, an example of converting a therapeutic image (177Lu-PSMA) into an estimated image (68Ga-PSMA) has been described, but the present invention is not limited to this. It can also be applied to images taken of subjects to which a drug labeled with another radionuclide has been administered.

In the present embodiment, an example in which the treatment image is input to the conversion model 176 has been described, but the present invention is not limited to this. For example, in addition to the treatment image, information about the patient to which the treatment image belongs may be input into the conversion model 176.

In this embodiment, GAN is used as the generation (learning) method of the conversion model 176, but the conversion model 176 is not limited to the trained model related to GAN, and other U-NET (U-shaped neural network) and the like. It may be a trained model by a learning method such as deep learning or decision tree.

The processing when U-NET is used will be described. The U-NET is composed of an encoder including a convolution layer and a pooling layer for downsampling, and a decoder including a convolution layer and a pooling layer for upsampling. The control unit 11 acquires a large number of combinations of the treatment image of the subject and the teacher image of the subject taken by the PET device 2 at the same time as training data. The control unit 11 inputs the treatment image to the encoder and trains the parameters of the convolutional layer by using an error backpropagation method or the like so that the output of the decoder approaches the teacher image. When the learning is completed, the control unit 11 acquires a new treatment image and inputs the acquired treatment image to the U-NET. The control unit 11 acquires an estimated image output from the U-NET and displays it on the display unit 15.

FIG. 18 is a flowchart showing a processing procedure when outputting an estimated image. The control unit 11 of the server 1 acquires a treatment image of the treatment stage taken by using the medical device 3 for the subject to which 177Lu-PSMA has been administered (step S131). When the acquired treatment image is input, the control unit 11 uses a conversion model 176 that outputs an estimated image obtained by estimating the image taken by the PET device 2 from the subject to which the first drug is administered. Acquire an estimated image (step S132).

The control unit 11 transmits the acquired estimated image to the terminal 4 by the communication unit 13 (step S133). The control unit 41 of the terminal 4 receives the estimated image transmitted from the server 1 by the communication unit 43 (step S431), displays the received estimated image by the display unit 45 (step S432), and ends the process.

According to the present embodiment, based on the treatment image, it is possible to output an estimated image obtained by estimating the image of the subject to which the first drug is administered by using the PET device 2 by using the conversion model 176. Become.

According to the present embodiment, by outputting the estimated image (68Ga-PSMA) using the conversion model 176, it is possible to reduce the need to carry out the treatment end determination by 68Ga-PSMA.

(Embodiment 3)
The third embodiment relates to a mode in which a recommended dose of a second drug is output using artificial intelligence. The description of the contents overlapping with the first and second embodiments will be omitted.

FIG. 19 is a block diagram showing a configuration example of the server 1 of the third embodiment. The contents overlapping with FIG. 16 are designated by the same reference numerals and the description thereof will be omitted. The second drug information output model (fourth learning model) 177 is stored in the large-capacity storage unit 17. The second drug information output model 177 is a predictor that predicts the recommended dose of the second drug to be administered to the subject based on the diagnostic image, and is a learned model generated by machine learning.

FIG. 20 is an explanatory diagram illustrating the second drug information output model 177. The second drug information output model 177 is used as a program module that is a part of artificial intelligence software. The second drug information output model 177 is a predictor for which a neural network has been constructed that inputs a diagnostic image and outputs a recommended dose of the second drug to be administered to the subject.

The control unit 11 of the server 1 constructs (generates) the second drug information output model 177 as the second drug information output model 177 by performing deep learning to learn the feature amount of the lesion portion in the diagnostic image. For example, the second drug information output model 177 is CNN, and is intermediate between an input layer that accepts the input of a diagnostic image, an output layer that outputs the result of predicting the recommended dose of the second drug, and an output layer that has been learned by backpropagation. Has a layer. Since the configuration of the second drug information output model 177 is the same as the configuration of the diagnostic lesion estimation model 174, the description thereof will be omitted.

During learning, a large amount of past diagnostic data existing in medical institutions is used as teacher data. The control unit 11 performs learning by using a combination of the diagnostic image and the teacher data in which the dose of the second drug corresponding to the diagnostic image is associated with each other. The teacher data is data in which the dose of the second drug corresponding to the diagnostic image is labeled on the diagnostic image. The control unit 11 inputs the diagnostic image, which is the teacher data, to the input layer, performs arithmetic processing in the intermediate layer, and acquires a prediction result (recommended dose) for predicting the recommended dose of the second drug from the output layer.

The control unit 11 compares the recommended dose of the second drug output from the output layer with the dose of the second drug labeled with respect to the diagnostic image in the teacher data, that is, the correct answer value, and outputs from the output layer. Optimize the parameters used for arithmetic processing in the intermediate layer so that the values approach the correct values. The control unit 11 performs the above processing on each diagnostic image included in the teacher data, and generates a second drug information output model 177. As a result, for example, the control unit 11 can build a model capable of predicting the recommended dose of the second drug corresponding to the diagnostic image by learning the second drug information output model 177 using the teacher data. it can.

When the control unit 11 acquires the diagnostic image, the control unit 11 inputs the acquired diagnostic image to the second drug information output model 177. The control unit 11 performs arithmetic processing for extracting the feature amount of the diagnostic image in the intermediate layer of the second drug information output model 177. The control unit 11 inputs the extracted feature amount to the output layer of the second drug information output model 177, and acquires the prediction result of predicting the recommended dose of the second drug corresponding to the diagnostic image as an output.

As shown, the prediction result output from the output layer is a continuous probability value indicating the dose of the second drug. As an example, for the diagnostic image, the probability values of "0 cycle", "1 cycle", "2 cycle", "3 cycle", "4 cycle", and "5 cycle" are "0.01" and "", respectively. The prediction results of "0.03", "0.02", "0.92", "0.01", and "0.01" are output. A cycle is a unit of dose of a second drug.

The generation process of the second drug information output model 177 is not limited to the above-mentioned process. For example, the control unit 11 may generate a second drug information output model 177 for each site of the subject. In this case, the control unit 11 inputs the site information and the diagnostic image into the second drug information output model 177, and outputs a prediction result that predicts the recommended dose of the second drug corresponding to the diagnostic image.

In this embodiment, an example in which a diagnostic image is input to the second drug information output model 177 has been described, but the present invention is not limited to this. For example, in addition to the diagnostic image, information about the patient to which the diagnostic image belongs may be input to the second drug information output model 177.

FIG. 21 is a flowchart showing a processing procedure when outputting a recommended dose of the second drug. The control unit 11 of the server 1 acquires a diagnostic image taken by using the PET device 2 (step S141). The control unit 11 acquires the recommended dose of the second drug by using the second drug information output model 177 that outputs the recommended dose of the second drug when the acquired diagnostic image is input (step S142).

The control unit 11 transmits the acquired recommended dose of the second drug to the terminal 4 by the communication unit 13 (step S143). The control unit 41 of the terminal 4 receives the recommended dose of the second drug transmitted from the server 1 by the communication unit 43 (step S441). The control unit 41 displays the recommended dose of the received second drug on the display unit 45 (step S442), and ends the process.

According to this embodiment, it is possible to output the recommended dose of the second drug using the second drug information output model 177 based on the diagnostic image.

<Modification example 1>
The process of outputting the recommended type of the second drug and the recommended dose or the recommended administration order of the second drug for each type will be described using the second drug information output model 177.

FIG. 22 is an explanatory diagram illustrating the second drug information output model 177 of the first modification. The second drug information output model 177 is a predictor for which a neural network has been constructed that inputs a diagnostic image and outputs a recommended type of the second drug and a recommended dose of the second drug for each type.

The control unit 11 performs learning by using a combination of the diagnostic image, the type of the second drug corresponding to the diagnostic image, and the teacher data in which the dose of the second drug for each type is associated with each other. The control unit 11 inputs the diagnostic image, which is the teacher data, to the input layer, performs arithmetic processing in the intermediate layer, and acquires the prediction result of predicting the type of the second drug and the recommended dose for each type from the output layer. ..

The control unit 11 uses the recommended second drug type and the recommended dose for each type of the second drug output from the output layer, and the dose information for each type and type of the second drug labeled on the diagnostic image in the teacher data. That is, the parameters used for the arithmetic processing in the intermediate layer are optimized so that the output value from the output layer approaches the correct answer value in comparison with the correct answer value. The control unit 11 performs the above processing on each diagnostic image included in the teacher data, and generates a second drug information output model 177. As a result, for example, the control unit 11 learns the second drug information output model 177 using the teacher data, so that the type of the second drug corresponding to the diagnostic image and the recommended dose for each type can be predicted. Can be built.

In addition to the above-mentioned processing, the control unit 11 receives the teacher data in which the diagnostic image, the type of the second drug corresponding to the diagnostic image, and the drug cycle for each type (for example, the number of administrations) are associated with each other. Learning may be performed using a combination. In this case, for example, the control unit 11 learns the second drug information output model 177 using the teacher data to obtain a model capable of predicting the type of the second drug corresponding to the diagnostic image and the drug cycle for each type. Can be built.

When the control unit 11 acquires the diagnostic image, the control unit 11 inputs the acquired diagnostic image to the second drug information output model 177. The control unit 11 performs arithmetic processing for extracting the feature amount of the diagnostic image in the intermediate layer of the second drug information output model 177. The control unit 11 inputs the extracted feature amount to the output layer of the second drug information output model 177, and outputs the prediction result of predicting the type of the second drug corresponding to the diagnostic image and the recommended dose for each type. get.

In addition, the second drug information output model 177 can be used to output the recommended administration order of the second drug instead of the recommended dose of the second drug for each type. Furthermore, the combination of the recommended type of the second drug, the recommended dose of the second drug for each type, and the recommended administration order of the second drug can be output from the second drug information output model 177.

According to this modification, it is possible to output the recommended type of the second drug and the recommended dose or the recommended administration order of the second drug for each type by using the second drug information output model 177 based on the diagnostic image. It will be possible.

(Embodiment 4)
The fourth embodiment relates to a mode in which a treatment timing is determined based on a diagnostic image. The description of the contents overlapping with the first to third embodiments will be omitted. The treatment timing can be classified based on lesion information (for example, the extent of cancer spread, malignancy, etc.). For example, the treatment timing may be classified into "no treatment required", "immediately", "1 month later", "2 months later", "3 months later" or "half a year later". For example, when the treatment timing is "1 month later", it is necessary to start the treatment by "1 month later".

FIG. 23 is a block diagram showing a configuration example of the server 1 of the fourth embodiment. The contents overlapping with FIG. 19 are designated by the same reference numerals and the description thereof will be omitted. A treatment timing prediction model (fifth learning model) 178 is stored in the large-capacity storage unit 17. The treatment timing prediction model 178 is a predictor that predicts the treatment timing based on the diagnostic image, and is a learned model generated by machine learning.

FIG. 24 is an explanatory diagram illustrating a treatment timing prediction model 178. The treatment timing prediction model 178 is used as a program module that is part of artificial intelligence software. The treatment timing prediction model 178 is a predictor for which a neural network has been constructed that inputs a diagnostic image and outputs a prediction result that predicts the treatment timing.

The control unit 11 of the server 1 constructs (generates) the treatment timing prediction model 178 as the treatment timing prediction model 178 by performing deep learning to learn the feature amount of the lesion portion in the diagnostic image. For example, the treatment timing prediction model 178 is a CNN, and has an input layer that accepts input of a diagnostic image, an output layer that outputs a result of predicting the treatment timing, and an intermediate layer that has been learned by back propagation. Since the configuration of the treatment timing prediction model 178 is the same as that of the diagnostic lesion estimation model 174, the description thereof will be omitted.

During learning, a large amount of past diagnostic data existing in medical institutions is used as teacher data. The control unit 11 performs learning using a combination of the diagnostic image and the teacher data associated with the treatment timing corresponding to the diagnostic image. The teacher data is data in which the treatment timing corresponding to the diagnostic image is labeled with respect to the diagnostic image. The control unit 11 inputs the diagnostic image, which is the teacher data, to the input layer, performs arithmetic processing in the intermediate layer, and acquires the prediction result of predicting the treatment timing from the output layer.

The control unit 11 compares the predicted treatment timing output from the output layer with the treatment timing labeled with respect to the diagnostic image in the teacher data, that is, the correct answer value, so that the output value from the output layer approaches the correct answer value. , Optimize the parameters used for arithmetic processing in the intermediate layer. The control unit 11 performs the above processing on each diagnostic image included in the teacher data, and generates a treatment timing prediction model 178. As a result, for example, the control unit 11 can build a model capable of predicting the treatment timing corresponding to the diagnostic image by learning the treatment timing prediction model 178 using the teacher data.

When the control unit 11 acquires the diagnostic image, the control unit 11 inputs the acquired diagnostic image into the treatment timing prediction model 178. The control unit 11 performs arithmetic processing for extracting the feature amount of the diagnostic image in the intermediate layer of the treatment timing prediction model 178. The control unit 11 inputs the extracted feature amount to the output layer of the treatment timing prediction model 178, and acquires the prediction result of predicting the treatment timing corresponding to the diagnostic image as an output.

As shown, the prediction result output from the output layer is a continuous probability value indicating the treatment timing. As an example, for the diagnostic image, the probability values of "no treatment required", "immediately", "1 month later", "2 months later", "3 months later", and "half a year later" are "0.01". , "0.03", "0.02", "0.92", "0.01", "0.01".

In this embodiment, an example in which a diagnostic image is input to the treatment timing prediction model 178 has been described, but the present invention is not limited to this. For example, in addition to the diagnostic image, information about the patient to which the diagnostic image belongs may be input to the treatment timing prediction model 178.

FIG. 25 is a flowchart showing a processing procedure when outputting a prediction result of treatment timing. The control unit 11 of the server 1 acquires a diagnostic image taken by using the PET device 2 (step S151). The control unit 11 acquires the treatment timing by using the treatment timing prediction model 178 that outputs the prediction result of predicting the treatment timing when the acquired diagnostic image is input (step S152).

The control unit 11 transmits the acquired treatment timing to the terminal 4 by the communication unit 13 (step S153). The control unit 41 of the terminal 4 receives the treatment timing transmitted from the server 1 by the communication unit 43 (step S451), displays the received treatment timing by the display unit 45 (step S452), and ends the process.

Further, the treatment timing determination process is not limited to the determination (prediction) process using the treatment timing prediction model 178 described above. The treatment timing can be determined based on the contrast ratio of the lesion site. The contrast ratio of the lesion site is the ratio of the brightness (brightness) of the lesion site to the brightness of the normal site with respect to the diagnostic image. The brightness is the brightness in each pixel, that is, ((R + G + B) / 3). R is red, G is green, and B is the pixel value of the blue sub-pixel.

The control unit 11 of the server 1 calculates the contrast ratio of the lesion site based on the diagnostic image. The control unit 11 compares the calculated contrast ratio of the lesion site with the diagnostic reference contrast ratio corresponding to the lesion site. The diagnostic reference contrast ratio may be stored in the storage unit 12 or the large-capacity storage unit 17 in advance, or may be acquired from an external normal database (NDB: normal database). Further, regarding the reference contrast ratio for diagnosis, a plurality of reference contrast ratios can be provided depending on the treatment timing. For example, diagnostic reference contrast ratios for "no treatment required", "immediately", "1 month later", "2 months later", "3 months later", and "half a year later" may be provided.

Specifically, the control unit 11 compares the calculated contrast ratio of the lesion site with the respective diagnostic reference contrast ratios provided by the treatment timing described above. For example, when the control unit 11 determines that the calculated contrast ratio of the lesion site is equal to or greater than the diagnostic reference contrast ratio "after 1 month" and equal to or less than the diagnostic reference contrast ratio "after 2 months", " The treatment timing "one month later" is output as the judgment result.

FIG. 26 is a flowchart showing a processing procedure when determining the treatment timing based on the contrast ratio. The control unit 11 acquires a diagnostic image taken by using the PET device 2 (step S161). The control unit 11 calculates the contrast ratio of the lesion site based on the brightness of the acquired diagnostic image (step S162).

The control unit 11 acquires the diagnostic reference contrast ratio corresponding to the lesion site from the external normal database via the communication unit 13 (step S163). The control unit 11 compares the calculated contrast ratio of the lesion site with the acquired reference contrast ratio for diagnosis (step S164), determines the corresponding treatment timing (step S165), and ends the process.

According to this embodiment, it is possible to output the prediction result of the treatment timing using the treatment timing prediction model 178 based on the diagnostic image.

According to this embodiment, it is possible to determine the treatment timing based on the diagnostic reference contrast ratio stored in the normal database or the like.

(Embodiment 5)
The fifth embodiment relates to an embodiment that outputs an estimated treatment result including the number of remaining treatments based on the treatment image. The description of the contents overlapping with the first to fourth embodiments will be omitted. Estimated treatment results include the number of remaining treatments (eg, complete cure, once or twice, etc.). In addition to the number of remaining treatments, the estimated treatment result may include recommended drugs to be administered for treatment, advice, and the like.

FIG. 27 is a block diagram showing a configuration example of the server 1 of the fifth embodiment. The contents overlapping with FIG. 23 are designated by the same reference numerals and the description thereof will be omitted. The large-capacity storage unit 17 stores an estimated treatment result output model (sixth learning model) 179. The estimated treatment result output model 179 is an estimator that estimates the treatment result including the number of remaining treatments based on the treatment image, and is a learned model generated by machine learning.

FIG. 28 is an explanatory diagram illustrating an estimated treatment result output model 179. The estimated treatment result output model 179 is used as a program module that is part of artificial intelligence software. The estimated treatment result output model 179 is an estimator for which a neural network has been constructed that inputs a treatment image and outputs an estimated result that estimates the treatment result including the remaining number of treatments.

The control unit 11 of the server 1 constructs (generates) the estimated treatment result output model 179 as the estimated treatment result output model 179 by performing deep learning to learn the feature amount of the lesion portion in the treatment image. For example, the estimated treatment result output model 179 is CNN, which is an input layer that accepts input of a treatment image, an output layer that outputs a result of estimating a treatment result including the number of remaining treatments, and an intermediate layer that has been learned by back propagation. And have. Since the configuration of the estimated treatment result output model 179 is the same as the configuration of the diagnostic lesion estimation model 174, the description thereof will be omitted.

During learning, a large amount of past treatment data existing in medical institutions is used as teacher data. The control unit 11 performs learning using a combination of the treatment image and the teacher data in which the treatment result including the remaining number of treatments is associated with each other. The teacher data is data in which the treatment result including the remaining number of treatments is labeled on the treatment image. The control unit 11 inputs the treatment image, which is the teacher data, to the input layer, performs arithmetic processing in the intermediate layer, and acquires an estimation result that estimates the treatment result including the remaining number of treatments from the output layer.

The control unit 11 compares the treatment result including the remaining number of treatments output from the output layer with the treatment result labeled with respect to the treatment image in the teacher data, that is, the correct answer value, and the output value from the output layer is the correct answer value. The parameters used for the arithmetic processing in the intermediate layer are optimized so as to approach. The control unit 11 performs the above processing on each treatment image included in the teacher data, and generates an estimated treatment result output model 179. As a result, for example, the control unit 11 can learn the estimated treatment result output model 179 using the teacher data, thereby constructing a model corresponding to the treatment image and capable of estimating the treatment result including the remaining number of treatments. it can.

When the control unit 11 acquires the treatment image, the control unit 11 inputs the acquired treatment image to the estimation treatment result output model 179. The treatment image may be a treatment image of a subject to which 68Ga-PSMA has been administered (for example, a treatment image for determining the end of treatment), or a treatment image of a subject to which 177Lu-PSMA has been administered. You may. Furthermore, the treatment image may be a combination of the treatment image of the subject to which 68Ga-PSMA has been administered and the treatment image of the subject to which 177Lu-PSMA has been administered.

The control unit 11 performs arithmetic processing for extracting the feature amount of the treatment image in the intermediate layer of the estimation treatment result output model 179. The control unit 11 inputs the extracted feature amount to the output layer of the estimated treatment result output model 179, and acquires the estimated result of estimating the treatment result including the remaining number of treatments corresponding to the treatment image as an output.

As shown in the figure, the prediction result output from the output layer is a continuous probability value indicating the number of remaining treatments. As an example, for the treatment image, the probability values of "complete cure", "1 time", "2 times", and "3 times" are "0.01", "0.03", "0.91", respectively. The estimation result of "0.05" is output.

In this embodiment, an example in which the treatment image is input to the estimation treatment result output model 179 has been described, but the present invention is not limited to this. For example, in addition to the treatment image, information about the patient to which the treatment image belongs may be input to the estimation treatment result output model 179.

FIG. 29 is a flowchart showing a processing procedure when outputting an estimated treatment result. The control unit 11 of the server 1 acquires a treatment image from the medical device 3 (step S171). The treatment image may be a treatment image of a subject to which 177Lu-PSMA has been administered, a treatment image of a subject to which 68Ga-PSMA has been administered, or a treatment image of a combination of both. The control unit 11 acquires the estimated treatment result by using the estimated treatment result output model 179 that outputs the estimated treatment result including the remaining number of treatments when the acquired treatment image is input (step S172).

The control unit 11 transmits the acquired estimated treatment result to the terminal 4 by the communication unit 13 (step S173). The control unit 41 of the terminal 4 receives the estimated treatment result transmitted from the server 1 by the communication unit 43 (step S471), displays the received estimated treatment result on the display unit 45 (step S472), and ends the process. ..

Further, the determination process of the treatment result is not limited to the determination (estimation) process using the above-mentioned estimation treatment result output model 179. Based on the contrast ratio of the lesion site, the treatment result including the number of remaining treatments can be determined.

The control unit 11 of the server 1 calculates the contrast ratio of the lesion site based on the treatment image. The control unit 11 compares the calculated contrast ratio of the lesion site with the therapeutic reference contrast ratio corresponding to the lesion site. The therapeutic reference contrast ratio may be stored in the storage unit 12 or the large-capacity storage unit 17 in advance, or may be acquired from an external normal database. Further, regarding the therapeutic reference contrast ratio, a plurality of therapeutic reference contrast ratios can be provided depending on the number of remaining treatments. For example, the reference contrast ratios of "complete cure", "1 time", "2 times", and "3 times" may be provided.

Specifically, the control unit 11 compares the calculated contrast ratio of the lesion site with the reference contrast ratio for each treatment provided by the number of remaining treatments described above. For example, when the control unit 11 determines that the calculated contrast ratio of the lesion site is equal to or more than the treatment reference contrast ratio of "2 times" and equal to or less than the treatment reference contrast ratio of "3 times", the number of remaining treatments. The treatment result is determined twice.

FIG. 30 is a flowchart showing a processing procedure when determining a treatment result based on the contrast ratio. The control unit 11 acquires a therapeutic image of a subject to which the second drug has been administered (step S181). The control unit 11 calculates the contrast ratio of the lesion site based on the brightness of the acquired treatment image (step S182).

The control unit 11 acquires the therapeutic reference contrast ratio corresponding to the lesion site from the external normal database via the communication unit 13 (step S183). The control unit 11 compares the calculated contrast ratio of the lesion site with the acquired reference contrast ratio for treatment (step S184), determines the treatment result including the number of remaining treatments (step S185), and ends the process.

According to this embodiment, it is possible to output the estimated treatment result using the estimated treatment result output model 179 based on the treatment image.

According to this embodiment, it is possible to determine the treatment result based on the treatment reference contrast ratio stored in the normal database or the like.

(Embodiment 6)
FIG. 31 is a functional block diagram showing the operation of the server 1 of the first to fifth embodiments. When the control unit 11 executes the control program 1P, the server 1 operates as follows.

The first acquisition unit 10a acquires a diagnostic image taken by using the PET device 2. The first learning model 10b outputs lesion estimation information when the diagnostic image acquired by the first acquisition unit 10a is input. The second acquisition unit 10c acquires a treatment image of the treatment stage taken by using the medical device 3 based on the lesion estimation information output by the first learning model 10b.

The second learning model 10d outputs lesion estimation information at the treatment stage when the treatment image acquired by the second acquisition unit 10c is input. When the treatment image is input, the third learning model 10e outputs an estimated image in which the image of the subject to which the first drug is administered is estimated by using the PET device 2. The display unit 10f simultaneously displays the diagnostic image and the treatment image. The fourth learning model 10g outputs the recommended dose of the second drug when the diagnostic image is input. In addition, the fourth learning model 10g can output the recommended type of the second drug and the recommended dose or the recommended administration order of the second drug for each type when the diagnostic image is input.

The fifth learning model 10h outputs a prediction result that predicts the treatment timing when a diagnostic image is input. The first calculation unit 10i calculates the contrast ratio of the lesion site based on the diagnostic image. The first determination unit 10j determines the treatment timing by comparing the contrast ratio calculated by the first calculation unit 10i with the diagnostic reference contrast ratio corresponding to the lesion site.

The sixth learning model 10k outputs an estimated treatment result including the number of remaining treatments when the treatment image is input. The second calculation unit 10l calculates the contrast ratio of the lesion site based on the treatment image. The second determination unit 10m determines the treatment result including the remaining number of treatments based on the comparison between the contrast ratio calculated by the second calculation unit 10l and the treatment reference contrast ratio corresponding to the lesion site.

The sixth embodiment is as described above, and the other parts are the same as those of the first to fifth embodiments. Therefore, the corresponding parts are designated by the same reference numerals and detailed description thereof will be omitted.

The embodiments disclosed this time should be considered as exemplary in all respects and not restrictive. The scope of the present invention is indicated by the scope of claims, not the above-mentioned meaning, and is intended to include all modifications within the meaning and scope equivalent to the scope of claims.

1 Information processing device (server)
11 Control unit 12 Storage unit 13 Communication unit 14 Input unit 15 Display unit 16 Reading unit 17 Large-capacity storage unit 171 Subject DB
172 Diagnostic information DB
173 Treatment information DB
174 Diagnostic lesion estimation model (first learning model)
175 Therapeutic lesion estimation model (second learning model)
176 conversion model (third learning model)
177 2nd drug information output model (4th learning model)
178 Treatment timing prediction model (fifth learning model)
179 Estimated treatment result output model (6th learning model)
1a Portable storage medium 1b Semiconductor memory 1P control program 2 Positron emission tomography equipment (PET equipment)
21 Control unit 22 Storage unit 23 Communication unit 24 Input unit 25 Display unit 26 Imaging unit 27 Detector 28 Unit board 2P control program 3 Medical imaging device (medical device)
31 Control unit 32 Storage unit 33 Communication unit 34 Input unit 35 Display unit 36 Imaging unit 37 Detector 38 Unit board 3P control program 4 Information processing terminal (terminal)
41 Control unit 42 Storage unit 43 Communication unit 44 Input unit 45 Display unit 4P Control program 10a 1st acquisition unit 10b 1st learning model 10c 2nd acquisition unit 10d 2nd learning model 10e 3rd learning model 10f Display unit 10g 4th learning Model 10h 5th learning model 10i 1st calculation unit 10j 1st judgment unit 10k 6th learning model 10l 2nd calculation unit 10m 2nd judgment unit

Claims (16)

A first acquisition unit that acquires diagnostic images taken by using a positron emission tomography device of a subject to which a first drug labeled with a radionuclide is administered with a radiopharmaceutical synthesizer or a preparation preparation for use.
A first learning model that outputs lesion estimation information when a diagnostic image acquired by the first acquisition unit is input, and
Based on the lesion estimation information output by the first learning model, a subject to which a second drug labeled with a radionuclide was administered with a radiopharmaceutical synthesizer or a preparation preparation for use was photographed using a medical imaging device. The second acquisition department that acquires the treatment image of the treatment stage
An information processing device including a second learning model that outputs lesion estimation information at the treatment stage when a treatment image acquired by the second acquisition unit is input. The information processing apparatus according to claim 1, wherein the second agent contains a prostate-specific membrane antigen-binding ligand labeled with lutetium 177. The information processing apparatus according to claim 1, wherein the second agent contains a prostate-specific membrane antigen-binding ligand labeled with gallium-68. The information processing device according to any one of claims 1 to 3, wherein the medical imaging device is a positron emission tomography device. The first agent comprises a gallium-68 labeled HER2 antibody.
The second agent comprises an astatine 211 labeled HER2 antibody.
The information processing apparatus according to claim 1, wherein the medical imaging apparatus is a single photon emission tomography apparatus or a Compton camera. A claim comprising a third learning model that outputs an estimated image obtained by estimating an image taken by using the positron emission tomography apparatus of the subject to which the first drug is administered when the treatment image is input. The information processing apparatus according to any one of claims 1 to 5. The information processing apparatus according to any one of claims 1 to 6, further comprising a display unit that simultaneously displays the diagnostic image and the therapeutic image. The information processing apparatus according to any one of claims 1 to 7, further comprising a fourth learning model that outputs a recommended dose of the second drug when the diagnostic image is input. The fourth learning model is characterized in that, when the diagnostic image is input, the recommended type of the second drug and the recommended dose or the recommended administration order of the second drug for each type are output. Item 8. The information processing apparatus according to item 8. The information processing apparatus according to any one of claims 1 to 9, further comprising a fifth learning model that outputs a prediction result that predicts the treatment timing when the diagnostic image is input. Based on the diagnostic image, the first calculation unit that calculates the contrast ratio of the lesion site and
Any of claims 1 to 9, further comprising a first determination unit that determines the treatment timing by comparing the contrast ratio calculated by the first calculation unit with the diagnostic reference contrast ratio corresponding to the lesion site. The information processing device described in one. The information processing apparatus according to any one of claims 1 to 11, further comprising a sixth learning model that outputs an estimated treatment result including the number of remaining treatments when the treatment image is input. In the sixth learning model, a therapeutic image of a subject to which a gallium-68-labeled prostate-specific membrane antigen-binding ligand was administered and a subject to which a lutetium-177-labeled prostate-specific membrane antigen-binding ligand was administered. The information processing apparatus according to claim 12, wherein an estimated treatment result including the number of remaining treatments is output when the treatment image of the above is input. A second calculation unit that calculates the contrast ratio of the lesion site based on the treatment image,
Claim 1 includes a second determination unit that determines a treatment result including the number of remaining treatments based on a comparison between the contrast ratio calculated by the second calculation unit and the treatment reference contrast ratio corresponding to the lesion site. The information processing apparatus according to any one of claims 11. A diagnostic image of a subject to which the first drug labeled with a radionuclide was administered with a radiopharmaceutical synthesizer or a preparation preparation for use was obtained using a positron emission tomography device.
Using the first learning model that outputs the lesion estimation information when the acquired diagnostic image is input, the lesion estimation information is output.
Based on the output lesion estimation information, a subject to whom a second drug labeled with a radionuclide was administered with a radiopharmaceutical synthesizer or a preparation preparation for use was photographed using a medical imaging device for treatment at the treatment stage. Get the image,
A program that causes a computer to execute a process of outputting lesion estimation information at the treatment stage using a second learning model that outputs lesion estimation information at the treatment stage when the acquired treatment image is input. A diagnostic image obtained by using a positron emission tomography device to obtain a diagnostic image of a subject to which the first drug labeled with a radionuclide was administered with a radiopharmaceutical synthesizer or a preparation preparation for use was obtained.
Using the first learning model that outputs the lesion estimation information when the acquired diagnostic image is input, the lesion estimation information is output.
Based on the output lesion estimation information, a subject to whom a second drug labeled with a radionuclide was administered with a radiopharmaceutical synthesizer or a preparation preparation for use was photographed using a medical imaging device for treatment at the treatment stage. Get the image,
An information processing method that outputs lesion estimation information at the treatment stage using a second learning model that outputs lesion estimation information at the treatment stage when the acquired treatment image is input.

Images