WO2023153490A1 - 情報処理システム、情報処理方法及び情報処理プログラム - Google Patents

情報処理システム、情報処理方法及び情報処理プログラム Download PDF

Info

Publication number
WO2023153490A1
WO2023153490A1 PCT/JP2023/004451 JP2023004451W WO2023153490A1 WO 2023153490 A1 WO2023153490 A1 WO 2023153490A1 JP 2023004451 W JP2023004451 W JP 2023004451W WO 2023153490 A1 WO2023153490 A1 WO 2023153490A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
new
processor
information processing
diagnostic
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/JP2023/004451
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
宏典 松政
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fujifilm Corp
Original Assignee
Fujifilm Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Fujifilm Corp filed Critical Fujifilm Corp
Priority to JP2023580326A priority Critical patent/JPWO2023153490A1/ja
Publication of WO2023153490A1 publication Critical patent/WO2023153490A1/ja
Priority to US18/791,450 priority patent/US20240395409A1/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS

Definitions

  • the present disclosure relates to an information processing system, an information processing method, and an information processing program.
  • image diagnosis is performed using medical images obtained by imaging devices such as CT (Computed Tomography) devices and MRI (Magnetic Resonance Imaging) devices.
  • medical images are analyzed by CAD (Computer Aided Detection/Diagnosis) using discriminators trained by deep learning, etc., and regions of interest including structures and lesions included in medical images are detected and/or Diagnosis is being made.
  • CT Computer Tomography
  • MRI Magnetic Resonance Imaging
  • Medical images may include personal information such as the subject's name, gender and age.
  • JP 2021-061042 discloses a system for anonymizing health data to protect patient privacy when communicating health data from one geographic area to another for data analysis. and methods are disclosed.
  • the present disclosure provides an information processing system, an information processing method, and an information processing program that can consistently manage learning and operation of learning models.
  • a first aspect of the present disclosure is an information processing system comprising: a first information processing device including at least one first processor; and a second information processing device including at least one second processor;
  • a first processor uses a combination of biometric information including at least a medical image and diagnostic information related to the biometric information to learn a learning model whose input is biometric information and whose output is diagnostic information. Acquiring a learned learning model from a first information processing device, and inputting new biometric information different from the biometric information used for learning of the learned learning model into the learned learning model to generate new biometric information Generate new diagnostic information about
  • the first information processing device further includes an input unit
  • the second information processing device further includes a storage unit in which biological information is stored
  • the first processor acquires the biological information stored in the storage unit from the second information processing device, displays the biological information on the display, and receives input via the input unit regarding diagnostic information related to the displayed biological information. good too.
  • a third aspect of the present disclosure is the first aspect or the second aspect, wherein the first processor acquires a combination of the new biological information and the new diagnostic information, and obtains the combination of the new biological information and the new diagnostic information. may be used to retrain the learning model.
  • the second information processing device further includes an input unit, and the second processor inputs the generated new diagnostic information to The new diagnostic information may be displayed on a display and may be modified via an input unit.
  • the first processor acquires a combination of the new biological information and the new diagnostic information, and if the new diagnostic information is corrected, the new biological information and The learning model may be retrained using the combination with the new diagnostic information after correction.
  • the biological information includes information about at least one of the subject that is the acquisition source and the imaging device used for the acquisition. Supplementary information to indicate may be attached.
  • a seventh aspect of the present disclosure is the sixth aspect, further comprising a third information processing apparatus including at least one third processor, wherein the third processor comprises at least one piece of supplementary information attached to the biological information. Part may be anonymized.
  • the second information processing device further includes a storage unit in which the biological information and the incidental information are stored
  • the third processor receives from the second information processing device Acquiring supplementary information, anonymizing at least part of the supplementary information, obtaining the supplementary information after anonymization from the third information processing device, and obtaining the supplementary information after anonymization from the third information processing device, the biometric information, the supplementary information before anonymization, and the anonymization and the anonymized incidental information may be associated with each other and stored in the storage unit, and the first processor may acquire the biometric information accompanied by the anonymized incidental information.
  • a ninth aspect of the present disclosure is the seventh aspect or the eighth aspect, wherein the third processor acquires additional information attached to the new biological information from the second information processing device, and obtains at least the additional information Part of it is anonymized, and the first processor acquires the new biometric information attached with the anonymized additional information, and uses the combination of the new biometric information and the new diagnostic information to relearn the learning model. good too.
  • a tenth aspect of the present disclosure is any one of the sixth to ninth aspects, wherein the second processor includes additional information attached to the biological information used for learning the learning model; The degree of similarity with additional information attached to the new biometric information may be estimated.
  • the second processor may present the generated new diagnostic information and the estimated degree of similarity.
  • a twelfth aspect of the present disclosure is the tenth aspect or the eleventh aspect, wherein the first processor obtains a combination of the new biological information, the new diagnostic information, and the degree of similarity, and based on the degree of similarity, A combination of the biological information and the new diagnostic information may be used to determine whether or not to relearn the learning model.
  • a thirteenth aspect of the present disclosure is any one of the sixth to twelfth aspects, wherein the incidental information is the name, gender, age, medical history and identification number of the subject, and acquisition of biological information may include information indicating at least one of the imaging conditions used in the above.
  • a fourteenth aspect of the present disclosure is any one of the first to thirteenth aspects, wherein the biometric information indicates information previously diagnosed in relation to a medical image included in the biometric information. including diagnostic information, the first processor uses a combination of the medical image and pre-diagnosis information included in the biometric information, and the diagnostic information related to the biometric information, learning that the input is biometric information and the output is diagnostic information You can train the model.
  • a fifteenth aspect of the present disclosure is any one of the first to fourteenth aspects, wherein the diagnostic information includes information indicating the position and size of the region of interest included in the medical image; Information indicating findings may be included at least one of.
  • a sixteenth aspect of the present disclosure is an information processing method, wherein a first processor uses a combination of biometric information including at least a medical image and diagnostic information related to the biometric information to input biometric information, A learning model whose output is diagnostic information is learned, a second processor acquires a learned learning model learned by the first processor, and new biological information different from the biological information used for learning the learned learning model By inputting the biometric information into a learned learning model, new diagnostic information relating to the new biometric information is generated.
  • a seventeenth aspect of the present disclosure is an information processing program, in which a first processor uses a combination of biometric information including at least a medical image and diagnostic information related to the biometric information to input biometric information, causing the second processor to acquire the learned learning model trained by the first processor, and the biometric information used for learning the learned learning model; By inputting new biometric information different from the above into a learned learning model, processing for generating new diagnostic information related to the new biometric information is executed.
  • the information processing system, information processing method, and information processing program of the present disclosure can consistently manage learning and operation of learning models.
  • FIG. 1 is a diagram showing an example of a schematic configuration of an information processing system according to first and second exemplary embodiments;
  • FIG. It is a block diagram which shows an example of the hardware constitutions of a research server.
  • FIG. 4 is a block diagram showing an example of functional configurations of a research server and a clinical server in a learning phase; It is a figure which shows an example of biometric information.
  • It is a figure which shows an example of the screen displayed on a display.
  • 4 is a block diagram showing an example of functional configurations of a research server and a clinical server in an operation phase;
  • FIG. It is a figure which shows an example of the screen displayed on a display.
  • FIG. 4 is a block diagram showing an example of functional configurations of a research server and a clinical server in a relearning phase; 10 is a flowchart showing an example of 1-1 information processing; 10 is a flowchart showing an example of 1-2 information processing; FIG. 11 is a flowchart showing an example of 2-1 information processing; FIG. FIG. 11 is a flow chart showing an example of 2-2 information processing; FIG. 13 is a flowchart showing an example of 3-1 information processing; 13 is a flowchart illustrating an example of 3-2 information processing; FIG. 11 is a block diagram showing an example of functional configurations of a research server and a clinical server according to the second exemplary embodiment; FIG.
  • FIG. 11 is a diagram showing an example of a schematic configuration of an information processing system according to a third exemplary embodiment;
  • FIG. It is a block diagram which shows an example of the hardware constitutions of an anonymization server.
  • 4 is a block diagram showing an example of functional configurations of a research server, a clinical server, and an anonymization server;
  • FIG. It is a figure which shows an example of the anonymized additional information.
  • FIG. 1 is a diagram showing a schematic configuration of an information processing system 1.
  • the information processing system 1 includes a research server 100 and a clinical server 200 .
  • the research server 100 and the clinical server 200 are connected via a wired or wireless network 9 so as to be communicable with each other.
  • the network 9 is, for example, a LAN (Local Area Network), a WAN (Wide Area Network), or the like.
  • the research server 100 is an example of the first information processing device of the present disclosure
  • the clinical server 200 is an example of the second information processing device of the present disclosure.
  • the clinical server 200 is a known medical information system such as PACS (Picture Archiving and Communication System) 2, RIS (Radiology Information System) 3 and HIS (Hospital Information System) 4. is connected with Further, the clinical server 200 may be connected to an imaging device 5 (modality) that generates a medical image representing the diagnostic target site by imaging the diagnostic target site of the subject.
  • the imaging device 5 is, for example, a plain X-ray imaging device, a CT (Computed Tomography) device, an MRI (Magnetic Resonance Imaging) device, a PET (Positron Emission Tomography) device, or the like.
  • the information processing system 1 of the present disclosure utilizes biometric information obtained in a production environment to learn a diagnostic model M for diagnosing biometric information.
  • FIG. 1 As the research server 100 and the clinical server 200, for example, server computers, personal computers, and the like can be applied as appropriate.
  • the research server 100 includes a CPU (Central Processing Unit) 121, a nonvolatile storage unit 122, and a memory 123 as a temporary storage area.
  • the research server 100 also includes a display 124 such as a liquid crystal display, an input unit 125 such as a keyboard and mouse, and a network I/F (Interface) 126 .
  • a network I/F 126 is connected to the network 9 and performs wired or wireless communication.
  • the CPU 121, the storage unit 122, the memory 123, the display 124, the input unit 125, and the network I/F 126 are connected via a bus 129 such as a system bus and a control bus so as to exchange various information with each other.
  • a bus 129 such as a system bus and a control bus
  • the storage unit 122 is implemented by a storage medium such as a HDD (Hard Disk Drive), SSD (Solid State Drive), flash memory, or the like.
  • the storage unit 122 stores an information processing program 127 in the research server 100 .
  • the CPU 121 reads out the information processing program 127 from the storage unit 122 , expands it in the memory 123 , and executes the expanded information processing program 127 .
  • the CPU 121 is an example of the first processor of the present disclosure.
  • learning data DB (DataBase) 128 and diagnostic model M are stored in storage unit 122 .
  • the learning data DB 128 stores learning data used for learning the diagnostic model M (details will be described later).
  • the storage unit 122 may be realized by, for example, a cloud server or the like.
  • the diagnostic model M is a model whose input is biological information and whose output is information related to diagnosis of the biological information (hereinafter referred to as "diagnostic information").
  • the diagnostic model M consists of neural networks, such as CNN (Convolutional Neural Network) and RNN (Recurrent Neural Network). Diagnostic model M is an example of a learning model of the present disclosure.
  • the clinical server 200 includes a CPU 221, a nonvolatile storage unit 222, and a memory 223 as a temporary storage area.
  • the clinical server 200 also includes a display 224 such as a liquid crystal display, an input unit 225 such as a keyboard and mouse, and a network I/F 226 .
  • a network I/F 226 is connected to the network 9 and performs wired or wireless communication.
  • the CPU 221, the storage unit 222, the memory 223, the display 224, the input unit 225, and the network I/F 226 are connected via a bus 229 such as a system bus and a control bus so as to exchange various information with each other.
  • the storage unit 222 is implemented by storage media such as HDD, SSD, and flash memory, for example.
  • An information processing program 227 in the clinical server 200 is stored in the storage unit 222 .
  • the CPU 221 reads the information processing program 227 from the storage unit 222 , expands it in the memory 223 , and executes the expanded information processing program 227 .
  • the CPU 221 is an example of the second processor of the present disclosure.
  • the storage unit 222 stores a biometric information DB 228 .
  • the biometric information DB 228 stores biometric information obtained from various external devices such as the PACS 2, RIS 3, HIS 4, and imaging device 5 (details will be described later). That is, the storage unit 222 stores biometric information.
  • the storage unit 222 may be realized by, for example, a cloud server or the like.
  • FIG. 4 research server 100 includes acquisition unit 130 , learning unit 132 and control unit 134 .
  • the CPU 121 functions as an acquisition unit 130 , a learning unit 132 and a control unit 134 by executing the information processing program 127 by the CPU 121 .
  • the clinical server 200 includes an acquisition unit 230, a generation unit 232, and a control unit 234.
  • FIG. The CPU 221 functions as an acquisition unit 230 , a generation unit 232 and a control unit 234 by executing the information processing program 227 by the CPU 221 .
  • functions of the research server 100 and the clinical server 200 will be described separately for the learning phase, operation phase, and re-learning phase of the diagnostic model M.
  • FIG. 4 functions of the research server 100 and the clinical server 200 will be described separately for the learning phase, operation phase, and re-learning phase of the diagnostic model M.
  • FIG. 4 also illustrates various types of information input to and output from the research server 100 and the clinical server 200 in the learning phase.
  • biometric information acquired by the clinical server 200 is used to input correct labels and learn the diagnostic model M in the research server 100 .
  • the acquisition unit 230 of the clinical server 200 acquires biological information from various external devices such as the PACS 2, RIS 3, HIS 4 and imaging device 5.
  • the control unit 234 of the clinical server 200 stores the biometric information acquired by the acquisition unit 230 in the biometric information DB 228 of the storage unit 222 . Also, the control unit 234 outputs the biometric information stored in the biometric information DB 228 to the research server 100 .
  • biometric information B includes at least a medical image B1.
  • a medical image B1 shown in FIG. 5 is a diagram schematically showing an example of a medical image acquired by the imaging device 5, and represents a tomographic plane including lungs.
  • a medical image may include regions of structures showing various organs of the human body (eg, organs such as lungs and liver) and various tissues (eg, blood vessels, nerves, muscles, etc.) that make up the various organs.
  • Medical images may also include areas of abnormal shadowing caused by lesions (eg, nodules, tumors, injuries, defects, inflammation, etc.). At least one of the region of the structure and the region of the abnormal shadow is hereinafter referred to as a “region of interest”.
  • a region of abnormal shadow due to a nodule in the medical image B1 is illustrated as a region of interest R.
  • FIG. Note that a single medical image may include a plurality of regions of interest.
  • the biological information B is accompanied by supplementary information B2 indicating information on at least one of the subject from which the biometric information is acquired and the imaging device 5 used for acquisition.
  • the supplementary information B2 includes at least one of the subject's name, gender, age, medical history and identification number ("subject ID (identification)"), and imaging conditions used to acquire biometric information. It may contain information indicating Imaging conditions include, for example, imaging date and time, type of imaging device 5 used for imaging medical images, manufacturer, identification number (“imaging device ID”), imaging site, imaging protocol, imaging sequence, imaging method, and contrast agent. is used or not.
  • the acquisition unit 130 of the research server 100 acquires biological information stored in the storage unit 222 (biological information DB 228) from the clinical server 200.
  • the control unit 134 of the research server 100 performs control to display the biological information acquired by the acquisition unit 130 on the display 124 .
  • the control unit 134 receives input via the input unit 125 for diagnostic information (that is, correct label) related to the biological information displayed on the display 124 .
  • the diagnostic information includes, for example, at least one of coordinate information indicating the position and size of the region of interest included in the medical image included in the biological information, and findings information indicating findings of the region of interest.
  • FIG. 6 shows an example of the screen D1 displayed on the display 124 by the control unit 134.
  • the screen D1 includes a medical image B1 including the region of interest R and input fields 91 and 92 for diagnostic information about the medical image B1.
  • the input field 91 is a field for inputting coordinate information A1 indicating the position and size of the region of interest R included in the medical image B1 as an example of diagnostic information.
  • the control unit 134 may accept specification of the coordinate information A1 in the form of a bounding box, a mask, or the like by operating the mouse pointer 90 in the input field 91 .
  • the input field 92 is a field for inputting finding information A2 regarding the medical image B1 and/or the region of interest R as an example of diagnostic information.
  • the finding information A2 may include, for example, information indicating at least one of the name (type), properties, measured values, positions, and estimated disease name of the medical image B1 and/or the region of interest R.
  • names (types) include names of structures such as “lung” and “liver” and names of abnormal shadows such as “nodule”. Characteristic mainly means the characteristics of the abnormal opacity.
  • absorption values such as “solid” and “frosted”
  • margins such as “clear/unclear”, “smooth/irregular”, “spicular”, “lobed” and “serrated”
  • Shape and findings indicative of overall shape such as “nearly circular” and “irregular” are included.
  • the relationship with surrounding tissues such as “pleural contact” and “pleural indentation”, and findings regarding the presence or absence of contrast enhancement and washout.
  • a measured value is a value that can be quantitatively measured from a medical image, and includes, for example, size (major axis, minor axis, volume, etc.), CT value in units of HU, and the number of regions of interest when there are multiple regions of interest. and the distance between regions of interest. Also, the measured values may be replaced with qualitative expressions such as “large/small” and “large/small”.
  • location is meant anatomical location, location in a medical image, and relative positional relationship with other regions of interest such as “inside,” “marginal,” and “periphery,” and the like.
  • Anatomical location may be indicated by organ names such as “lung” and “liver”, or lungs may be designated as “right lung”, “upper lobe”, and apical segment (“S1”). It may be represented by a subdivided expression.
  • the estimated disease name is the evaluation result estimated based on the abnormal shadow, for example, disease names such as “cancer” and “inflammation”, and “negative / positive”, “benign / malignant” and " evaluation results such as "mild/severe”.
  • the learning unit 132 of the research server 100 learns the diagnostic model M using a combination of the biometric information acquired by the acquisition unit 130 and the diagnostic information received by the control unit 134 regarding the biometric information.
  • the control unit 134 stores the combination of the biological information and diagnostic information used for learning the diagnostic model M in the learning data DB 128 of the storage unit 122 .
  • FIG. 7 also illustrates various types of information input to and output from the research server 100 and the clinical server 200 in the operation phase.
  • the research server 100 uses the learned diagnostic model M to diagnose biological information newly obtained in the production environment.
  • the control unit 134 of the research server 100 outputs the learned diagnostic model M to the clinical server 200 in the learning phase.
  • the acquisition unit 230 of the clinical server 200 acquires the trained diagnostic model M from the research server 100 .
  • the acquisition unit 230 acquires new biometric information different from the biometric information used for learning of the learned diagnostic model M in the learning phase from various external devices such as the PACS 2, RIS 3, HIS 4, and imaging device 5.
  • the generating unit 232 of the clinical server 200 inputs the new biological information acquired by the acquiring unit 230 to the learned diagnostic model M acquired by the acquiring unit 230, thereby generating new diagnostic information related to the new biological information. .
  • the control unit 234 of the clinical server 200 performs control to display the new diagnostic information generated by the generation unit 232 on the display 224 .
  • FIG. 8 shows an example of a screen D2 displayed on the display 224 by the controller 234. As shown in FIG. On the screen D2, coordinate information A1 of a region of interest R as an example of new diagnostic information is displayed on a medical image B5 (new biological information). The screen D2 also displays a medical image B5 and/or finding information A2 of the region of interest R as an example of new diagnostic information. Further, the control unit 234 associates the new biometric information with the new diagnostic information and stores them in the biometric information DB 228 of the storage unit 222 .
  • control unit 234 may accept correction by the user via the input unit 225 for the new diagnostic information displayed on the display 224 .
  • Screen D2 displays a button 94 for selecting whether or not to accept the correction.
  • the control unit 234 controls the display 224 to display a screen for accepting the correction of the new diagnostic information.
  • FIG. 9 shows an example of a screen D3 displayed on the display 224 by the control unit 234.
  • the screen D3 includes an input field 95 for receiving correction of the coordinate information A1 and an input field 96 for receiving correction of the finding information A2.
  • the control unit 234 stores the corrected new diagnostic information and the new biometric information in the biometric information DB 228 of the storage unit 222 in association with each other.
  • FIG. 10 also illustrates various types of information input to and output from the research server 100 and the clinical server 200 in the relearning phase.
  • a combination of the new biometric information and the new diagnostic information is used to re-learn the diagnostic model M that has already been trained in the research server 100 .
  • the control unit 234 of the clinical server 200 outputs the new diagnostic information and the new biological information stored in the biological information DB 228 to the research server 100.
  • Acquisition unit 130 of research server 100 acquires a combination of new biological information and new diagnostic information from clinical server 200 .
  • the learning unit 132 of the research server 100 re-learns the diagnostic model M using the combination of the new biological information and the new diagnostic information acquired by the acquisition unit 130 .
  • the new diagnostic information may be modified by the user in the clinical server 200. If the new diagnostic information has been revised, the learning unit 132 of the research server 100 may re-learn the diagnostic model M using a combination of the new biometric information and the revised new diagnostic information.
  • the control unit 134 of the research server 100 When re-operating the diagnostic model M in the clinical server 200 after re-learning the diagnostic model M, the control unit 134 of the research server 100 outputs the re-learned diagnostic model M to the clinical server 200 in the re-learning phase. .
  • the clinical server 200 uses the retrained diagnostic model M acquired from the research server 100 to generate new diagnostic information. In this way, the diagnostic model M is re-learned using a combination of the new biometric information and the new diagnostic information generated in the operation phase, and the re-learned diagnostic model M is operated repeatedly. The accuracy of model M can be improved.
  • FIG. 11 to 16 the actions of the research server 100 and the clinical server 200 according to this exemplary embodiment will be described with reference to FIGS. 11 to 16.
  • FIG. Hereinafter, the actions of the research server 100 and the clinical server 200 will be described separately for the learning phase, operation phase, and re-learning phase of the diagnostic model M.
  • FIG. 11 to 16 the actions of the research server 100 and the clinical server 200 will be described separately for the learning phase, operation phase, and re-learning phase of the diagnostic model M.
  • the CPU 121 executes the information processing program 127 to execute the 1-1 information processing shown in FIG.
  • the CPU 221 executes the information processing program 227 to execute the 1-2 information processing shown in FIG.
  • the 1-2 information process is executed, for example, when the user gives an instruction to start execution via the input unit 225, and the 1-1 information process is executed subsequent to the 1-2 information process. .
  • step S ⁇ b>121 the acquisition unit 230 acquires biological information from various external devices such as the PACS 2 , RIS 3 , HIS 4 and imaging device 5 .
  • step S ⁇ b>122 the control unit 234 stores the biological information acquired in step S ⁇ b>121 in the storage unit 222 .
  • step S123 the control unit 234 outputs the biological information stored in the storage unit 222 in step S121 to the research server 100, and ends this 1-2 information processing.
  • step S ⁇ b>111 the acquisition unit 130 acquires biological information from the clinical server 200 .
  • the control unit 134 controls the display 124 to display the biological information acquired in step S111.
  • control unit 134 receives an input of the diagnostic information related to the biological information displayed on display 124 in step S112.
  • step S114 the learning unit 132 uses a combination of the biological information acquired in step S111 and the diagnostic information received in step S113 to learn the diagnostic model M, and performs the present 1-1 information processing. finish.
  • the CPU 121 executes the information processing program 127 to execute the 2-1 information processing shown in FIG.
  • the CPU 221 executes the information processing program 227 to execute the 2-2 information processing shown in FIG.
  • the 2-1 information process is executed, for example, when the user gives an instruction to start execution via the input unit 125, and the 2-2 information process is executed subsequent to the 2-1 information process. .
  • step S211 the control unit 134 outputs the diagnostic model M learned in the 1-1 information process to the clinical server 200, and ends the 2-1 information process.
  • step S ⁇ b>221 the acquisition unit 230 acquires the learned diagnostic model M from the research server 100 .
  • step S ⁇ b>222 the acquisition unit 230 acquires new biometric information from various external devices such as the PACS 2 , RIS 3 , HIS 4 and imaging device 5 .
  • step S223 the generating unit 232 generates new diagnostic information related to the new biological information by inputting the new biological information obtained in step S222 to the learned diagnostic model M obtained in step S221.
  • step S224 the control unit 234 performs control to display the new diagnostic information generated in step S223 on the display 224, and accepts correction by the user for the new diagnostic information. If the correction of the new diagnostic information is received (that is, if step S224 is Y), the process proceeds to step S225, and the new biometric information acquired in step S222 and the corrected new diagnostic information received in step S224 are combined. They are stored in the storage unit 222 in association with each other.
  • step S224 the process proceeds to step S226, and the new biometric information acquired in step S222, the new diagnostic information generated in step S223, are associated with each other and stored in the storage unit 222 .
  • steps S225 and S226 are completed, this 2-2 information processing is terminated.
  • the CPU 121 executes the information processing program 127 to execute the 3-1 information processing shown in FIG.
  • the CPU 221 executes the information processing program 227 to execute the 3-2 information processing shown in FIG.
  • the 3-2 information process is executed, for example, when the user gives an instruction to start execution via the input unit 225, and the 3-1 information process is executed following the 3-2 information process. .
  • step S321 the control unit 234 outputs the new diagnostic information and the new biological information stored in the storage unit 222 in the 2-2nd information processing to the research server 100, and ends the 3-2nd information processing. do.
  • step S ⁇ b>311 the acquiring unit 130 acquires a combination of new biological information and new diagnostic information from the clinical server 200 .
  • step S312 the learning unit 132 re-learns the diagnostic model M using the combination of the new biometric information and the new diagnostic information acquired in step S311, and ends this 3-1 information processing.
  • the information processing system 1 includes a first information processing device including at least one first processor and a second information processing device including at least one second processor.
  • the first processor uses a combination of biometric information including at least a medical image and diagnostic information related to the biometric information to learn a learning model whose input is biometric information and whose output is diagnostic information.
  • the second processor acquires the learned learning model from the first information processing device, and inputs new biometric information different from the biometric information used for learning of the learned learning model to the learned learning model. generates new diagnostic information about the new biometric information.
  • learning and operation of learning models can be consistently managed.
  • the research server 100 re-learns the diagnostic model M based on the new diagnostic information in which the error has been corrected. Therefore, the accuracy of the diagnostic model M can be improved.
  • the biometric information may not be attached with additional information.
  • the information processing system 1 includes additional information attached to each of the biometric information used for learning the diagnostic model M and the new biometric information generated in the operation phase. It has the function of estimating the degree of similarity. This is because the degree of similarity of incidental information may be correlated with the accuracy of new diagnostic information.
  • the information processing system 1 according to the second exemplary embodiment will be described below, but descriptions of the same configurations and functions as those of the first exemplary embodiment will be omitted as appropriate.
  • FIG. 17 shows an example of functional configurations of the research server 100 and the clinical server 200 according to this exemplary embodiment.
  • the research server 100 includes an acquisition unit 130, a learning unit 132, and a control unit 134, and has the same configuration as that of the first exemplary embodiment.
  • the clinical server 200 according to this exemplary embodiment further includes an estimation unit 236 in addition to the acquisition unit 230, generation unit 232, and control unit 234 similar to those of the first exemplary embodiment.
  • the CPU 221 functions as an acquisition unit 230 , a generation unit 232 , a control unit 234 and an estimation unit 236 by executing the information processing program 227 by the CPU 221 .
  • the estimating unit 236 of the clinical server 200 calculates the degree of similarity (hereinafter referred to as simply called "similarity").
  • the incidental information is stored in the biometric information DB 228 together with the biometric information and the new biometric information, for example.
  • FIG. 18 shows an example of supplementary information of learning data and new biometric information.
  • the estimation unit 236 may estimate whether or not the contents are similar for each item of the supplementary information, and estimate the average as the overall degree of similarity. Further, for example, the estimating unit 236 may estimate a weighted average for each item of supplementary information as a comprehensive degree of similarity.
  • the estimating unit 236 estimates the degree of similarity between the incidental information of the new biometric information and the incidental information of each of the plurality of learning data, and calculates the representative value such as the average value and the median value as the comprehensive degree of similarity. can be estimated as
  • the control unit 234 of the clinical server 200 may present the new diagnostic information generated by the generation unit 232 and the degree of similarity estimated by the estimation unit 236 .
  • FIG. 19 shows an example of a screen D4 displayed on the display 224 by the controller 234. As shown in FIG. The screen D4 displays the degree of similarity estimated by the estimation unit 236 in addition to the new diagnostic information similar to the screen D2 (see FIG. 8). The higher the degree of similarity, the more suitable the new biological information to be diagnosed is for the diagnostic model M, and the higher the degree of accuracy of the new diagnostic information. can be grasped. Further, the control unit 234 associates the new biometric information, the new diagnosis information, and the degree of similarity, and stores them in the biometric information DB 228 of the storage unit 222 .
  • the control unit 234 of the clinical server 200 outputs the new diagnosis information, the new biometric information and the degree of similarity stored in the biometric information DB 228 to the research server 100 .
  • Acquisition unit 130 of research server 100 acquires a combination of new biological information, new diagnostic information, and a degree of similarity. Based on the degree of similarity acquired by the acquiring unit 130, the learning unit 132 of the research server 100 determines whether or not to relearn the diagnostic model M using a combination of the new biological information and the new diagnostic information. good.
  • the learning unit 132 determines whether or not to relearn the diagnostic model M using a combination of the new biometric information and the new diagnostic information by comparing the degree of similarity with a predetermined threshold. You may For example, when the degree of similarity is equal to or greater than the threshold, the combination of the new biometric information and the new diagnostic information may be similar to the learning data (combination of biometric information and diagnostic information) already used for learning the diagnostic model M. It can be assumed that the On the other hand, if the degree of similarity is less than the threshold, it can be estimated that there is a high possibility that the combination of the new biological information and the new diagnostic information is not similar to the learning data that has already been used for learning the diagnostic model M.
  • the learning unit 132 sets a condition as to whether to use for re-learning when the degree of similarity is equal to or greater than the threshold or whether to use for re-learning when the degree of similarity is less than the threshold, according to the characteristics of the diagnostic model M to be learned. can be different.
  • the learning unit 132 re-learns the diagnostic model M using a combination of the new biological information and the new diagnostic information.
  • the learning unit 132 uses the combination of the new biological information and the new diagnostic information to re-learn the diagnostic model M. You don't have to use it.
  • a more versatile model can be created by re-learning the diagnostic model M using a combination of new biometric information and new diagnostic information whose degree of similarity is less than the threshold.
  • the diagnostic model M is re-learned using a combination of new biometric information and new diagnostic information whose degree of similarity is equal to or higher than the threshold, the versatility of the model may decrease.
  • the learning unit 132 re-learns the diagnostic model M using a combination of the new biological information and the new diagnostic information. You may let That is, when the degree of similarity acquired by the acquisition unit 130 indicates that the degree of similarity is equal to or greater than a predetermined threshold, the learning unit 132 uses the combination of the new biological information and the new diagnostic information to re-learn the diagnostic model M. You don't have to use it.
  • the condition used by the learning unit 132 (whether to use for re-learning if the degree of similarity is equal to or greater than the threshold, or whether to use for re-learning if the degree of similarity is less than the threshold) is associated with the diagnostic model M, for example, and is determined in advance. Alternatively, the user may arbitrarily select. Further, for example, the learning unit 132 may perform re-learning of the diagnostic model M using a combination of the new biometric information and the new diagnostic information with weighting according to the degree of similarity.
  • the information processing system 1 anonymizes at least part of the additional information attached to the biometric information used for learning the diagnostic model M and the additional information attached to the new biometric information used for re-learning. It has the function of converting As shown in FIG. 5, the supplementary information may include personal information such as the subject's name, sex, and age. From the viewpoint of privacy protection, personal information that does not affect diagnosis should not be output to the research server 100. is preferred.
  • the information processing system 1 according to the third exemplary embodiment will be described below, but descriptions of the same configurations and functions as those of the first and second exemplary embodiments will be omitted as appropriate.
  • FIG. 20 is a diagram showing a schematic configuration of the information processing system 1 according to this exemplary embodiment.
  • the information processing system 1 according to this exemplary embodiment further includes an anonymization server 300 in addition to the information processing system 1 according to the first exemplary embodiment (see FIG. 1).
  • the research server 100, the clinical server 200, and the anonymization server 300 are connected to communicate with each other via a wired or wireless network 9.
  • FIG. The anonymization server 300 is an example of the third information processing device of the present disclosure.
  • the anonymization server 300 includes a CPU 321, a nonvolatile storage unit 322, and a memory 323 as a temporary storage area.
  • the anonymization server 300 also includes a display 324 such as a liquid crystal display, an input unit 325 such as a keyboard and mouse, and a network I/F 326 .
  • a network I/F 326 is connected to the network 9 and performs wired or wireless communication.
  • the CPU 321, the storage unit 322, the memory 323, the display 324, the input unit 325, and the network I/F 326 are connected via a bus 329 such as a system bus and a control bus so as to exchange various information with each other.
  • the storage unit 322 is realized by storage media such as HDD, SSD, and flash memory, for example.
  • the storage unit 322 stores an information processing program 327 in the anonymization server 300 .
  • the CPU 321 reads the information processing program 327 from the storage unit 322 , expands it in the memory 323 , and executes the expanded information processing program 327 .
  • the CPU 321 is an example of the third processor of the present disclosure.
  • the research server 100 includes an acquisition unit 130, a learning unit 132, and a control unit 134, and has the same configuration as that of the first exemplary embodiment.
  • the clinical server 200 includes an acquisition unit 230, a generation unit 232, and a control unit 234, and has the same configuration as that of the first exemplary embodiment.
  • the anonymization server 300 includes an acquisition unit 330 , an anonymization unit 332 and a control unit 334 .
  • the CPU 321 functions as an acquisition unit 330 , an anonymization unit 332 and a control unit 334 by executing the information processing program 327 by the CPU 321 .
  • FIG. 22 also illustrates various types of information input to and output from the research server 100, the clinical server 200, and the anonymization server 300 in the learning phase.
  • the acquisition unit 230 of the clinical server 200 acquires biological information from various external devices such as the PACS 2, RIS 3, HIS 4, and imaging device 5.
  • the biometric information is accompanied by incidental information that may include personal information such as the subject's name, sex and age.
  • the control unit 234 of the clinical server 200 stores the biometric information acquired by the acquisition unit 230 in the biometric information DB 228 of the storage unit 222 together with the incidental information. Also, the control unit 234 outputs the biometric information stored in the biometric information DB 228 to the anonymization server 300 together with the incidental information.
  • the acquisition unit 330 of the anonymization server 300 acquires biological information and incidental information from the clinical server 200.
  • the anonymization unit 332 of the anonymization server 300 anonymizes at least part of the supplementary information attached to the biometric information acquired by the acquisition unit 330 .
  • FIG. 23 shows an example of supplementary information partly anonymized. As shown in FIG. 23, the anonymization unit 332 may hide personal information that does not affect diagnosis, such as the subject ID and name. On the other hand, the anonymization unit 332 may not anonymize information that can be used for diagnosis, such as gender, age, and medical history, even if it is personal information. Which item of the incidental information is to be anonymized may be determined in advance, or may be arbitrarily designated via the input unit 325, for example.
  • the control unit 334 of the anonymization server 300 outputs to the clinical server 200 the biometric information attached with the additional information anonymized by the anonymization unit 332 .
  • the acquisition unit 230 of the clinical server 200 acquires the biometric information attached with the anonymized incidental information from the anonymization server 300 .
  • the control unit 234 of the clinical server 200 stores the biometric information acquired by the acquisition unit 230, the additional information before anonymization, and the additional information after anonymization in association with each other in the biometric information DB 228 of the storage unit 222. .
  • the acquisition unit 130 of the research server 100 acquires from the clinical server 200 biometric information accompanied by anonymized additional information. After that, the research server 100 receives the diagnostic information and trains the diagnostic model M in the same manner as in the first exemplary embodiment.
  • the acquisition unit 330 of the anonymization server 300 acquires additional information attached to the new biometric information from the clinical server 200 .
  • the anonymization unit 332 of the anonymization server 300 anonymizes at least part of the supplementary information attached to the new biometric information acquired by the acquisition unit 330 .
  • the control unit 334 of the anonymization server 300 outputs to the clinical server 200 the new biometric information attached with the additional information anonymized by the anonymization unit 332 .
  • the acquisition unit 230 of the clinical server 200 acquires the new biometric information attached with the anonymized incidental information from the anonymization server 300 .
  • the control unit 234 of the clinical server 200 associates the new biometric information acquired by the acquisition unit 230, the additional information before anonymization, the additional information after anonymization, and the new diagnosis information, and stores them in the storage unit 222. Stored in the biometric information DB 228 .
  • the acquisition unit 130 of the research server 100 acquires from the clinical server 200 new biometric information accompanied by anonymized additional information. After that, the learning unit 132 of the research server 100 re-learns the diagnostic model M using the combination of the new biometric information and the new diagnostic information in the same manner as in the first exemplary embodiment.
  • the personal information in the incidental information acquired by the research server 100 is anonymized. Therefore, learning and operation of the learning model can be consistently managed while protecting privacy.
  • the study server 100 acquires the biometric information and the new biometric information attached with the anonymized additional information from the clinical server 200, but the present invention is not limited to this. .
  • the research server 100 may directly acquire the biometric information and the new biometric information attached with the post-anonymization additional information from the anonymization server 300 .
  • the input is biometric information including medical images
  • the diagnostic model M is output as diagnostic information.
  • the technology of the present disclosure may be applied to a learning model that performs so-called multimodal learning that outputs diagnostic information based on other information about the subject in addition to medical images.
  • the biometric information that is the input of the learning model may include pre-diagnosis information indicating information previously diagnosed in relation to the medical image included in the biometric information.
  • the pre-diagnosis information may be, for example, information described in an electronic medical record relating to the subject from which the medical image is acquired, past examination results, information generated by a known CAD system, and the like.
  • information includes information indicating the position and feature amount (eg, volume and major axis) of the region of interest R included in the medical image, information indicating the grade of disease (eg, cancer grade), and processing results of texture analysis for medical images.
  • the learning unit 132 of the research server 100 uses a combination of the medical image and pre-diagnosis information included in the biometric information and the diagnostic information related to the biometric information to set the input as biometric information and output You may learn the learning model which makes diagnostic information.
  • various processes such as the acquisition units 130, 230 and 330, the learning unit 132, the generation unit 232, the anonymization unit 332, the control units 134, 234 and 334, and the estimation unit 236 are performed.
  • Various processors shown below can be used as the hardware structure of the processing unit to be executed.
  • the various processors include, in addition to the CPU, which is a general-purpose processor that executes software (programs) and functions as various processing units, circuits such as FPGAs (Field Programmable Gate Arrays), etc.
  • Programmable Logic Device PLD which is a processor whose configuration can be changed, ASIC (Application Specific Integrated Circuit) etc. Circuits, etc. are included.
  • One processing unit may be composed of one of these various processors, or a combination of two or more processors of the same or different type (for example, a combination of a plurality of FPGAs, or a combination of a CPU and an FPGA). combination). Also, a plurality of processing units may be configured by one processor.
  • a single processor is configured by combining one or more CPUs and software.
  • a processor functions as multiple processing units.
  • SoC System on Chip
  • a processor that realizes the function of the entire system including multiple processing units with a single IC (Integrated Circuit) chip. be.
  • various processing units are configured using one or more of the above various processors as a hardware structure.
  • the information processing programs 127, 227 and 327 are pre-stored (installed) in the storage units 122, 222 and 322, respectively, but the present invention is not limited to this.
  • the information processing programs 127, 227 and 327 are recorded in recording media such as CD-ROM (Compact Disc Read Only Memory), DVD-ROM (Digital Versatile Disc Read Only Memory), and USB (Universal Serial Bus) memory.
  • CD-ROM Compact Disc Read Only Memory
  • DVD-ROM Digital Versatile Disc Read Only Memory
  • USB Universal Serial Bus
  • the information processing programs 127, 227 and 327 may be downloaded from an external device via a network.
  • the technology of the present disclosure extends to a storage medium that non-temporarily stores an information processing program in addition to the information processing program.
  • the technology of the present disclosure can also appropriately combine the exemplary embodiments described above.
  • the description and illustration shown above are detailed descriptions of the parts related to the technology of the present disclosure, and are merely examples of the technology of the present disclosure.
  • the above descriptions of configurations, functions, actions, and effects are descriptions of examples of configurations, functions, actions, and effects of portions related to the technology of the present disclosure. Therefore, unnecessary parts may be deleted, new elements added, or replaced with respect to the above-described description and illustration without departing from the gist of the technology of the present disclosure. Needless to say.

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Public Health (AREA)
  • Medical Informatics (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Epidemiology (AREA)
  • Data Mining & Analysis (AREA)
  • Pathology (AREA)
  • Databases & Information Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Biophysics (AREA)
  • Physics & Mathematics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Veterinary Medicine (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)
  • Medical Treatment And Welfare Office Work (AREA)
PCT/JP2023/004451 2022-02-09 2023-02-09 情報処理システム、情報処理方法及び情報処理プログラム Ceased WO2023153490A1 (ja)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2023580326A JPWO2023153490A1 (https=) 2022-02-09 2023-02-09
US18/791,450 US20240395409A1 (en) 2022-02-09 2024-08-01 Information processing system, information processing method, and information processing program

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2022019036 2022-02-09
JP2022-019036 2022-02-09

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US18/791,450 Continuation US20240395409A1 (en) 2022-02-09 2024-08-01 Information processing system, information processing method, and information processing program

Publications (1)

Publication Number Publication Date
WO2023153490A1 true WO2023153490A1 (ja) 2023-08-17

Family

ID=87564491

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2023/004451 Ceased WO2023153490A1 (ja) 2022-02-09 2023-02-09 情報処理システム、情報処理方法及び情報処理プログラム

Country Status (3)

Country Link
US (1) US20240395409A1 (https=)
JP (1) JPWO2023153490A1 (https=)
WO (1) WO2023153490A1 (https=)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7795180B1 (ja) * 2024-10-15 2026-01-07 株式会社プレシジョン 情報処理装置、情報処理方法及び情報処理プログラム

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2018036836A (ja) * 2016-08-31 2018-03-08 株式会社ジェイマックシステム 医用画像ファイル管理装置、医用画像ファイル管理方法および医用画像ファイル管理プログラム
JP2020086519A (ja) * 2018-11-15 2020-06-04 キヤノンメディカルシステムズ株式会社 医用画像処理装置、医用画像処理方法、およびプログラム
JP2021039748A (ja) * 2019-08-30 2021-03-11 キヤノン株式会社 情報処理装置、情報処理方法、情報処理システム及びプログラム

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2018036836A (ja) * 2016-08-31 2018-03-08 株式会社ジェイマックシステム 医用画像ファイル管理装置、医用画像ファイル管理方法および医用画像ファイル管理プログラム
JP2020086519A (ja) * 2018-11-15 2020-06-04 キヤノンメディカルシステムズ株式会社 医用画像処理装置、医用画像処理方法、およびプログラム
JP2021039748A (ja) * 2019-08-30 2021-03-11 キヤノン株式会社 情報処理装置、情報処理方法、情報処理システム及びプログラム

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7795180B1 (ja) * 2024-10-15 2026-01-07 株式会社プレシジョン 情報処理装置、情報処理方法及び情報処理プログラム

Also Published As

Publication number Publication date
US20240395409A1 (en) 2024-11-28
JPWO2023153490A1 (https=) 2023-08-17

Similar Documents

Publication Publication Date Title
CN112885453A (zh) 用于标识后续医学图像中的病理变化的方法和系统
US20200342990A1 (en) Medical image processing apparatus, method, and program, and diagnosis support apparatus, method, and program
US11669960B2 (en) Learning system, method, and program
JP5676269B2 (ja) 脳画像データの画像解析
JP6957214B2 (ja) 情報処理装置、情報処理システム、情報処理方法及びプログラム
JP2007530160A (ja) 医用画像のための自動決定支援を提供するシステムおよび方法
US20230197253A1 (en) Medical image processing apparatus, method, and program
CN107072613B (zh) 基于纵向特征对感兴趣组织的健康状态的分类
US11756292B2 (en) Similarity determination apparatus, similarity determination method, and similarity determination program
EP4252663B1 (en) Image display device, method, and program
JP7524334B2 (ja) 画像表示装置、方法およびプログラム
US20230360213A1 (en) Information processing apparatus, method, and program
JPWO2019176407A1 (ja) 学習支援装置、学習支援方法、学習支援プログラム、関心領域判別装置、関心領域判別方法、関心領域判別プログラム及び学習済みモデル
WO2020209382A1 (ja) 医療文書作成装置、方法およびプログラム
WO2021157705A1 (ja) 文書作成支援装置、方法およびプログラム
JP7007469B2 (ja) 医療文書作成支援装置、方法およびプログラム、学習済みモデル、並びに学習装置、方法およびプログラム
WO2020202822A1 (ja) 医療文書作成支援装置、方法およびプログラム
Bhattacharya et al. Gazediff: A radiologist visual attention guided diffusion model for zero-shot disease classification
US20240395409A1 (en) Information processing system, information processing method, and information processing program
WO2020241857A1 (ja) 医療文書作成装置、方法およびプログラム、学習装置、方法およびプログラム、並びに学習済みモデル
EP4358022B1 (en) Medical image diagnostic system, medical image diagnostic method, and program
JP2023067186A (ja) 情報処理装置、情報処理方法及び情報処理プログラム
US12608801B2 (en) Information processing apparatus, information processing method, and information processing program for supporting interpretation of medical image
JP7376715B2 (ja) 経過予測装置、経過予測装置の作動方法および経過予測プログラム
JP7368592B2 (ja) 文書作成支援装置、方法およびプログラム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23752952

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2023580326

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 23752952

Country of ref document: EP

Kind code of ref document: A1