Brain image analysis data utilization system

Abstract

Provided is a brain image analysis data utilization system that is capable of generating an onset risk model to provide information evaluating a risk. An analysis service provision server 2000 receives tomographic brain image data obtained for a person to be evaluated, and returns an analysis result for atrophy of a prescribed region of interest. A data sharing service server 3000 stores the tomographic brain image data of the person to be evaluated and health information pertaining to the person to be evaluated which have been provided in a range permitted by the person to be evaluated and which are associated with each other by temporary identification information. A user server 4000 receives presentation of the temporary identification information owned by the person to be evaluated when the person to be evaluated subscribes to a unique service of a manager. The user server 4000 holds a unique database in which the temporary identification information is uniquely associated with data of the person to be evaluated in a user server, and analyzes data in the unique database to provide the unique service to the person to be evaluated.

Description

Brain image analysis data utilization system

[0001] The present invention relates to a brain image analysis data utilization system for utilizing data resulting from evaluation of the degree of brain atrophy using brain image data measured by a brain image data analysis device.

[0002] Dementia has become one of the most serious social issues in an aging society.

[0003] In this situation, although there have been reports on the development of new therapeutic drugs for the treatment of dementia (see Patent Document 1) and the exploration of new treatment methods such as ultrasound therapy (see Patent Document 2), it cannot be said that a fundamental treatment method for Alzheimer's disease has been established.

[0004] For this reason, current treatments are symptomatic treatments aimed at slowing the progression of symptoms. However, if the signs of dementia are noticed early and treatment is started at an early stage, it is possible to slow the progression while symptoms are still mild.

[0005] For example, according to the medical journal "Lancet" (see Non-Patent Document 1), it is known that as people age, the susceptibility to dementia changes due to the influence of past lifestyle habits (see Non-Patent Document 1). For example, education (1.6 times), high blood pressure (1.6 times), excessive drinking (1.2 times), obesity (1.6 times), smoking (1.6 times), depression (1.9 times), social isolation (1.6 times), lack of exercise (1.4 times), and diabetes (1.5 times).

[0006] FIG. 25 is a diagram showing the relationship between risk factors affecting the onset of dementia and each age group.

[0007] Figure 25 shows how much the probability of developing dementia would decrease if risk factors were eliminated. As shown in this figure, genetic factors are thought to account for only 5-10%, and it is known that lifestyle and social habits affect the risk of developing dementia.

[0008] Furthermore, according to the Japanese Society of Neurology's "2017 Dementia Disease Treatment Guidelines" (see Non-Patent Document 2), risk factors for lifestyle-related diseases such as hypertension, diabetes, and dyslipidemia can be risk factors that accelerate the progression of mild cognitive impairment to dementia, and therefore guidance on improving lifestyle habits and managing medication is necessary.

[0009] Therefore, at present, since dementia is thought to be closely related to lifestyle-related diseases, it is believed that detecting the signs as early as possible and reducing the risk of lifestyle-related diseases will indirectly lead to the prevention of Alzheimer's disease.

[0010] Specifically, it is believed that incorporating improvements to diet and regular exercise into daily life can help prevent Alzheimer's disease.

[0011] In addition, as part of the non-drug therapies introduced above, it is also said to be effective to make an effort to maintain a high level of intellectual activity by taking on new challenges on a daily basis, and to make music therapy and occupational therapy a habit.

[0012] If we could identify people at high risk of developing Alzheimer's disease before the symptoms appear, it might be possible to slow or even prevent the progression of the disease through intervention and prevention through lifestyle improvements such as diet, exercise, and brain-stimulating activities.

[0013] However, the most common early symptom of dementia is memory loss. However, as we all age, our memory deteriorates, so there are many cases where the person or their family does not notice the signs of dementia. And by the time they realize that "something is wrong," dementia has already progressed, and they may not be able to live without the support of those around them.

[0014] From this perspective, Patent Document 3 discloses a technology for detecting signs of dementia in which a user's exercise information is acquired by communicating with the user's information communication terminal via a network, and a statistical test is performed on the acquired exercise information with a data group in which the exercise information and dementia risk are associated, thereby assessing the user's dementia risk.

[0015] Furthermore, Patent Document 4 discloses a system that indicates the risk of dementia by using the value of hippocampal volume (hippocampal volume) as brain state data of a subject imaged with an MRI (Magnetic Resonance Imaging) device.

[0016] On the other hand, once dementia develops, it places a heavy burden not only on the individual but also on their family, and insurance companies are now offering so-called "dementia insurance" services.

[0017] In general, when providing insurance services, insurance is designed taking into account disease risks (see Patent Document 5).

[0018] Additionally, the company also provides support to health insurance associations in upgrading their health services by analyzing health checkup results and medical fee statement information (hereinafter referred to as "receipts").

[0019] Generally, life insurance companies assess whether or not to accept an insurance application based on the applicant's disclosure information and health examination results (this is called "risk selection").

[0020] For example, if insurance companies underwrite a large number of people with chronic illnesses or people who are at extremely high risk of developing illness or even dying in the future, the incidence of insured events will far exceed the initial expectations, making it difficult to maintain the insurance system. For this reason, risk selection is a very important function for insurance companies.

[0021] On the other hand, due to changes in the business environment surrounding insurance companies, such as changes in demographics, and for reasons such as fulfilling the social mission of private insurance, there is a need in the insurance industry to more precisely design traditional underwriting standards so that more people can take out insurance.

[0022] Patent No. 5033868 JP 2023-009246 A JP 2018-191722 A JP 2021-097988 A JP 2020-106882 A

[0023] Deborah E Barnes, Kristine Yaffe, “The projected effect of risk factor reduction on Alzheimer's disease prevalence”, Lancet Neurol 2011; 10: 819-28, www.thelancet.com / neurology Vol 13 August 2014https: / / www.neurology-jp.org / guidelinem / nintisyo_2017.html

[0024] However, conventional dementia risk assessments have not necessarily been sufficient to objectively assess the risk of developing dementia from an early stage. Furthermore, conventional dementia risk assessments using MRI imaging have problems in that the volume of the hippocampus accounts for less than 1% of the total brain volume, and therefore not only is there a lack of accuracy in the assessment, but the amount of data accumulated is also insufficient due to the need for high-resolution imaging, which is limited by the availability of MRI devices capable of such imaging.

[0025] For example, Alzheimer's disease is considered to progress very slowly among dementia types, and it is said that in many cases, it takes about 20 years for the full-blown symptoms to appear after the onset of premonition symptoms. Therefore, there are currently no established methods for early detection.

[0026] Here, when assessing the risk of developing other diseases from health checkup data, etc., it is common to generate a model for assessing the risk of developing a disease using health checkup data, data disclosed at the time of insurance enrollment, etc. as explanatory variables, on the assumption that longitudinal data can be obtained, from health checkup data when the subject is healthy to medical receipt data that can be obtained after the subject begins visiting a hospital for treatment. However, as mentioned above, when the time between the appearance of precursor symptoms in a healthy state and the onset of the disease is long, there is a problem in that it is difficult to generate such an onset risk model.

[0027] Furthermore, even if we were able to generate a dementia risk model, the question remains as to how to link this to behavioral changes such as improving diet and regular exercise. In particular, lifestyle changes must be implemented continuously over the long term, and there is the problem that it is difficult to motivate subjects to change their lifestyle habits.

[0028] The present invention has been made to solve such problems, and its purpose is to provide a brain image analysis data utilization system that can generate an onset risk model and provide information assessing the risk while utilizing brain tomographic image data accumulated as big data.

[0029] Another object of the present invention is to provide a brain image analysis data utilization system that can detect signs of dementia early, encourage behavioral changes in subjects, and provide information that can prevent the onset of the disease.

[0030] In order to achieve the above object, one aspect of the present invention is as follows: (1)

[0031] According to one aspect of the present invention, there is provided a brain image analysis data utilization system, comprising an analysis server that provides an analysis service that receives second brain tomographic image data captured for a subject to be evaluated and returns an analysis result on the atrophy of a predetermined region of interest in the brain, based on data obtained by three-dimensionally analyzing the degree of atrophy of a predetermined region of interest in the brain of a plurality of subjects, the first brain tomographic image data being captured by a tomographic image scanner device and being anonymized, the analysis server managing the second brain tomographic image data in association with temporary identification information that does not identify individuals, and further comprising a data sharing server that stores a shared database and is capable of data communication with an external party, The server stores the second brain tomography image data of the person being evaluated, provided by the analysis server within the scope of the person being evaluated's permission, and the person's health information, associated with temporary identification information, and further includes a user server that receives data from the data sharing server and uses the second brain tomography image data and health information.When the person being evaluated subscribes to the administrator's unique service, the user server receives the temporary identification information held by the person being evaluated, maintains a unique database in which the temporary identification information and the person being evaluated's data are uniquely associated on the user server, and provides the unique service to the person being evaluated by analyzing the data in the unique database.

[0032] (2) Preferably, in the configuration (1), the brain image analysis data utilization system further includes a service page providing server that executes user registration processing for the person being evaluated and returns analysis results in response to inquiries from the person being evaluated, wherein the service page providing server issues temporary identification information to the information processing device owned by the person being evaluated at the time of user registration, and the temporary identification information is stored in the memory unit of the information processing device, and the person being evaluated presents the temporary identification information to the imaging facility that will perform the imaging when capturing the second brain tomographic image data, and the imaging facility transmits the second brain tomographic image data to the analysis server in association with the temporary identification information, and the analysis server replies with the analysis results for the received second brain tomographic image data in association with the temporary identification information.

[0033] (3) Preferably, in the configuration of (1) or (2), when the person to be evaluated subscribes to a unique service operated by the administrator, the user server receives temporary identification information presented by the person to be evaluated via an information processing device.

[0034] (4) Preferably, in any one of the configurations (1) to (3), the information processing device further includes a display unit, and the temporary identification information is displayed as a two-dimensional code on the display unit.

[0035] (5) Preferably, in any one of the configurations (1) to (4), the user server calculates the assessment criteria when the subject applies for insurance by assessing the subject’s risk of developing dementia based on the second brain tomography image data and health information.

[0036] (6) Preferably, in any one of the configurations (1) to (5), the first brain tomographic image data of the multiple subjects is time-series data for each subject and includes clinical information on dementia associated with each time-series data, and the analysis server includes a memory device that stores a brain image risk model that uses machine learning or statistical analysis to input the degree of atrophy of the region of interest at a first time point based on the first brain tomographic image data of the multiple subjects and outputs the risk of dementia onset at a second time point after the first time point, and the analysis server receives the second brain tomographic image data as input and returns the risk of dementia calculated by the brain image risk model to the user server.

[0037] (7) Preferably, in any one of the configurations (1) to (6), the user server generates an onset risk model that predicts the onset risk using the onset risk output from the brain image risk model and the health check data as learning data, and generates insurance assessment criteria.

[0038] (8) Preferably, in any one of the configurations (1) to (7), the service page providing server updates and manages points for each person to be evaluated according to the frequency and amount of health information provided by the person to be evaluated, and provides discount services for imaging processing at imaging institutions according to the points.

[0039] (9) Preferably, in any one of the configurations (1) to (8), the administrator of the user server is a provider of services provided by the exercise facility, and the person being evaluated who subscribes to the unique service operated by the administrator provides health care information and dietary information obtained outside the exercise facility to the data sharing server, and the data sharing server calculates a recommended exercise menu and a recommended dietary menu for the person being evaluated based on the analysis results, information on the amount of exercise, health care information, and dietary information about the person being evaluated, and provides them to the user server.

[0040] (10) Preferably, in any one of the configurations (1) to (9), the administrator of the user server is a provider of counseling services to the person being evaluated, and the person being evaluated who subscribes to the unique service operated by the administrator registers the desired attributes of the counselor they desire via the user server, and the user server performs a matching process based on the desired attributes and the attributes of the counselor, i) conducting a test consultation without the matched counselor and the person being evaluated meeting face-to-face, and then ii) if the person being evaluated wishes, providing face-to-face counseling between the matched counselor and the person being evaluated.

[0041] (11) Preferably, in any one of the configurations (1) to (10), the analysis server includes a segmentation unit that performs segmentation to divide the entire brain tomographic image into each structural unit, an identification unit that identifies structural units related to the region of interest from the divided structural units, a volume calculation unit that calculates the brain volume of each identified structural unit, a calibration unit that calibrates the brain volume based on the input brain cross-sectional image in accordance with the imaging conditions of the tomographic image scanner device, a database containing age-related change data obtained by previously imaging multiple subjects and including brain volume data for each structural unit at each age under various imaging conditions, and a same-age ratio calculation unit that calculates the same-age ratio of the structural unit of the evaluation subject from the data for each structural unit at each age in the database, wherein the calibration unit calibrates the brain volume based on the age-related change data in the database, and the same-age ratio calculation unit, after calibration, calculates the same-age ratio of the brain volume of the structural unit of the evaluation subject for the multiple subjects based on the data for the structural unit at each age in the database, thereby analyzing the degree of atrophy in a specified region of interest in the brain.

[0042] (12) Preferably, in the configuration of (11), a structural unit selection unit is further provided that selects a structural unit that is related to atrophy of the region of interest, that has greater age-related changes than other structural units, and that has less change due to the influence of shooting conditions than other structural units, based on brain volume data stored in the database, and the identification unit identifies the structural unit selected by the structural unit selection unit.

[0043] (13) Preferably, in the configuration of (12), the predetermined brain region of interest is a structure known to be altered in dementia in the limbic system or medial temporal lobe, including the hippocampus.

[0044] (14) Preferably, in the configuration of (13), the structural unit selected by the structural unit selection unit is a structure selected based on the degree of age-related change and robustness to the influence of imaging conditions, and includes the ventricles.

[0045] According to the present invention, it is possible to generate a disease risk model and provide risk assessment information. Such information can be used, for example, to calculate insurance company assessment standards.

[0046] Furthermore, according to the present invention, it is possible to provide information that can detect signs of dementia early, encourage behavioral changes in subjects, and prevent the onset of dementia.

[0047] 1 is a functional block diagram of the brain image data analysis device of this embodiment. FIG. 1 is a block diagram for explaining the hardware configuration of the brain image data analysis device 1. FIG. 1 is a flowchart showing a pre-analysis of brain image data analysis processing. FIG. 2 is a flowchart showing analysis of brain image data analysis processing on a subject by the brain image data analysis device. FIG. 2 is a diagram showing segmentation training data at various structural levels. FIG. 3 is a diagram showing age-related changes in the mean value of ventricular volume and a predetermined variance. FIG. 4 is a diagram showing the frequency of ventricular atrophy in examinees by age (males). FIG. 5 is a diagram showing the frequency of ventricular atrophy in examinees by age (females). FIG. 6 is a diagram showing a histogram and log-normal distribution of the degree of atrophy (logarithmic value of ventricular volume) by age. FIG. 7 is a diagram showing the log-normal distribution of the degree of atrophy (logarithmic value of ventricular volume) by age. FIG. 8 is a diagram showing the difference between population analysis obtained from a normal distribution model and the AD patient area. FIG. 9 is a diagram showing the concept of data flow from data acquisition to shared use of data in order to achieve both pseudonymization of data and usability as longitudinal data. FIG. 10 is a conceptual diagram showing a configuration that enables use of pseudonymized information by both system operators and data users. 1 is a conceptual diagram showing a configuration that enables use by a system operator and a data user of pseudonym-processed information. FIG. 1 is a diagram showing an example of an "advice quick reference table" provided together with an "analysis result" notified to a subject as an analysis report from the service page operation server 1000. FIG. 2 is a flowchart for explaining operation of the system. FIG. 3 is an explanatory diagram showing an example of health check information and subject attribute information used to generate a development risk model in embodiment 1. FIG. 4 is an explanatory diagram showing an example of subject attribute information in embodiment 1. FIG. 5 is a functional block diagram for explaining the functional configuration of the data user server 4000. FIG. 6 is a diagram showing a first method for calculating a development risk model. FIG. 7 is a diagram showing a first method for calculating a development risk model. FIG. 8 is a diagram showing a second method for calculating a development risk model. FIG. 9 is a conceptual diagram for explaining the configuration of a data usage service. FIG. 10 is a diagram for explaining the flow of data in the data usage service. FIG. 11 is a conceptual diagram for explaining processing performed by the data sharing service server 3000.1 is a diagram showing an example of points given to members according to information collected via a fitness club or the service page operation server 1000. FIG. 2 is a diagram showing an example of points given to members according to information collected via a fitness club or the service page operation server 1000. FIG. 3 is a conceptual diagram for explaining the configuration of a data usage service. FIG. 4 is a diagram for explaining the processing flow in a counselor introduction service. FIG. 5 is a diagram showing the relationship between risk factors affecting the onset of dementia and each age group.

[0048] Hereinafter, a system for utilizing brain image analysis data according to an embodiment of the present invention will be described.

[0049] In this embodiment, the analysis service providing server functions as a brain image data analysis device that utilizes AI technology from Johns Hopkins University in the U.S. and software developed to analyze big data on 30,000 MRI brain images stored in Japan using AI to comprehensively evaluate brain atrophy and vascular changes, which are characteristics observed in patients with dementia and cerebral infarction.

[0050] Such brain image analysis technology developed by Johns Hopkins University in the United States is disclosed in the following documents, for example: Reference 1: U.S. Patent No. 10,074,173 Reference 2: U.S. Patent No. 10,535,133

[0051] In particular, Reference 1 discloses an automatic segmentation technique that can flexibly change the granularity level of the whole brain based on the hierarchical relationship of 254 structures defined in a brain atlas. Reference 2 discloses a technique that performs automatic segmentation of brain images and labels each structure using a generative model technique based on multiple atlases.

[0052] However, other segmentation methods and other labeling techniques may be used for segmenting brain images and labeling each structure.

[0053] In this specification, a "structural unit" of a brain image is described as referring to each of a plurality of structures (e.g., 505 structures) that have been automatically segmented based on one or more known brain atlases (brain maps), for example, using the method defined in Reference 1. However, as will be described later, the number of structures to be segmented is not limited to 505 structures, and for example, several segments may be integrated and considered as one "structural unit."

[0054] However, it is known that the structure of the brain is hierarchical and that its functions are localized, and the "structural units" obtained from brain images are not limited to those obtained using the method in Reference 1, but other divisions that can be distinguished from the information of the acquired images depending on the performance, type, imaging principle, imaging conditions, etc. of the scanner may also be used as "structural units."

[0055] As will be described later, a combination of multiple automatically segmented "structural units" may correspond to a single "brain structure." For example, the "brain structure" called the "ventricle" may consist of the lateral ventricle, the third ventricle, and the fourth ventricle, and the lateral ventricle may further have a hierarchical relationship consisting of an anterior portion, a posterior portion, and a lateral portion.

[0056] The brain image data analysis device of this embodiment utilizes the above-mentioned technology to quantify and evaluate ventricular expansion and the volume of the cerebral cortex in each part of the brain, and outputs reports such as the following: 1) "Comprehensive brain atrophy assessment" that evaluates the aging progression of the degree of brain atrophy by comparing with subjects of the same age.

[0057] 2) "Vascular lesion assessment" involves quantifying the volume of white matter lesion areas and comparing them with patients of the same age to assess the progression of vascular lesions in the brain over time.

[0058] The program that runs on the brain image data analysis device is a program for brain checkups and health centers that uses AI analysis of lesions or deviations from the average based on these evaluations to measure analytical values ​​that serve as supplementary or reference information for brain health risks and obtain information on the analytical results. <Principle of the analysis device>

[0059] First, the technical background and principles of the brain image data analysis device will be explained.

[0060] Brain image data analysis devices automatically quantify the degree of brain atrophy and vascular lesions from MRI images, enabling the extraction of information related to dementia. Conventional brain checkups have used brain MRI to detect and diagnose aneurysms, brain tumors, cerebral infarction, and intracerebral inflammation. However, they have not yet been able to predict the risk of developing dementia as a lifestyle-related disease and help reduce or prevent the onset of dementia. While existing technology includes programs for measuring the volume of the hippocampus, which plays a major role in cognitive function, the hippocampus accounts for only 0.3% of the human brain's volume. Volumetric measurements of the brain's parenchymal structure are prone to variability depending on the type of MRI equipment, resolution, and imaging method, making it difficult to accumulate useful information as stable tracking data. Furthermore, as discussed below, much of the remaining 99.7% of brain structure is important for estimating dementia risk, limiting the value of measuring the hippocampus alone.

[0061] Furthermore, measuring hippocampal atrophy requires a high-resolution brain MRI, but taking MRI images takes time, so high-resolution images are often not taken during brain checkups. Brain checkups in Japan have a long history of over 20 years, with over 1 million pieces of data already in existence, and some locations even having time-series data going back more than 15 years. However, for the reasons mentioned above, they are not necessarily able to reduce the risk of developing dementia or address risk management for dementia, which will become increasingly important in the future.

[0062] Analysis using the brain image data analysis device of this embodiment makes it possible to accumulate large amounts of big data and perform automatic AI analysis, which will greatly contribute to future dementia risk management. Analysis using this brain image data analysis device is already being conducted using existing data from brain checkup centers and research institutions. This analysis is a technology that is expected to lead to world-class big data analysis in the future, with existing data alone exceeding one million items.

[0063] By using this brain image data analysis, it is possible to improve the stability of measurements by measuring structures that are less affected by MRI image acquisition conditions, such as ventricular volume, and that sensitively reflect brain atrophy, and to realize a brain image analysis data utilization system that can more effectively utilize the results of analyzing brain atrophy data from brain checkup data. <Configuration>

[0064] 1 is a functional block diagram of the brain image data analysis device of this embodiment. As will be described later, the brain image data analysis device 1 also includes a configuration for setting parameters, conditions, etc. for analysis in advance based on existing big data in order to analyze the subject currently being evaluated.

[0065] As shown in Figure 1, the brain image data analysis device 1 of embodiment 1 has an image input unit 11, a segmentation unit 12, a brain structure volume / brain signal intensity calculation unit 31, an imaging condition calibration unit 15, a peer ratio calculation unit 16, a cerebrovascular disorder degree calculation unit 17, an analysis report creation unit 18, and an analysis report output unit 19.

[0066] The brain structure volume / brain signal intensity calculation unit 31 includes a brain structure selection unit 32 , an imaging condition influence evaluation unit 33 , a brain structure identification unit 34 , and a brain structure volume / signal intensity calculation unit 35 .

[0067] The "image input unit 11" accepts input of brain images (brain MRI images) acquired by an MRI scanner or brain images and imaging condition data transmitted from a PACS (Picture Archiving and Communication System), as well as attribute data of the imaged subject. The image input unit 11 corresponds to a communication interface that allows the brain image data analysis device 1, which is realized as a computer, to exchange data with the outside. The PACS may be a server on the cloud, and the brain image data analysis device itself may also be a server on the cloud.

[0068] The "segmentation unit 12" performs segmentation to divide the entire brain MRI image into structural units. Although not particularly limited, this segmentation can utilize the techniques described in the above-mentioned references 1 and 2, for example.

[0069] The 'imaging condition calibration unit 15' performs calibration in accordance with the imaging conditions when an image is captured by the MRI scanner, in order to minimize the influence of differences in the imaging conditions.

[0070] The 'same age ratio calculation unit 16' calculates the same age ratio of each structural unit of each subject from the structural unit data for each age in the database 2.

[0071] The 'cerebrovascular disorder degree calculation unit 17' calculates the degree of cerebrovascular disorder.

[0072] The analysis report creation unit 18 creates an analysis report based on the analysis results obtained.

[0073] The database 2a is a big data database and includes brain volume data for each structural unit at each age under various imaging conditions (including resolution, slice thickness, slice angle, slice direction, and sequence parameters (echo time, repetition time, flip angle, etc.)), age-structural unit volume data including brain signal intensity data, and age-structural unit signal intensity data. This database may be provided within an analysis center, or may be installed on a cloud server. Note that the data referred to as "big data" here may include not only existing data captured and integrated at multiple imaging sites, but also data that is newly acquired at multiple imaging sites, including non-existent ones, and that is accumulated sequentially.

[0074] As described above, the brain image data analysis device 1 is realized by executing arithmetic processing on a computer. Therefore, the segmentation unit 12, the brain structure volume / brain signal intensity calculation unit 31, the imaging condition calibration unit 15, the peer ratio calculation unit 16, the cerebrovascular disorder degree calculation unit 17, and the analysis report creation unit 18 correspond to functions executed by a CPU (Central Processing Unit) based on a program stored in memory, and this program includes modules that execute each function. In addition, the database 2a can also store information to be included in a report corresponding to the analysis results.

[0075] The imaging condition calibration unit 15 performs calibration using the age-structural unit volume data and age-structural unit signal intensity data in the database 2a. After calibration, the same-age ratio calculation unit 16 calculates the same-age ratio of each structural unit of each subject from the structural unit data for each age in the database 2, thereby analyzing the risk of developing dementia and obtaining auxiliary information or reference information.

[0076] The brain structure volume / signal intensity calculation unit 31 will be further described below.

[0077] As described above, the brain structure volume / signal intensity calculation unit 31 corresponds to a function in which the CPU executes a corresponding module in a program stored in memory when the computer operates as the brain image data analysis device 1. The brain structure selection unit 32, the imaging condition influence evaluation unit 33, the brain structure identification unit 34, and the brain structure volume / signal intensity calculation unit 35 are implemented as submodules in the module of the brain structure volume / signal intensity calculation unit 31, each of whose functions is executed by the CPU.

[0078] The "brain structure volume / signal intensity calculation unit 31" first includes an imaging condition influence evaluation unit 33 that evaluates the influence of the imaging conditions of brain MRI images on brain structures based on big data before analyzing the current subject, and a brain structure selection unit 32 that, based on the evaluation results, selects brain structures that are highly effective in terms of being able to accurately calculate the volume of brain structures, etc. from an engineering perspective even if there are variations in imaging conditions and subject attributes (age, gender, etc.), and that exhibit large changes due to aging and disease. The "brain structure volume / signal intensity calculation unit 31" further includes a brain structure identification unit 34 that identifies brain structures that are less affected by the imaging conditions from the imaging data from the subject currently being evaluated, and a brain structure volume / signal intensity calculation unit 35 that calculates the volume and signal intensity of large structures, such as ventricles and lobar structures, from among the identified brain structures.

[0079] FIG. 2 is a block diagram for explaining the hardware configuration of the brain image data analysis device 1 shown in FIG.

[0080] In the following description, the brain image data analysis device 1 and the like are assumed to operate on a physical server 3000 that operates on the cloud. However, the brain image data analysis device 1 may also operate on an on-premise server, or may be installed at the location of the device that captures brain images, such as a workstation. The hardware configuration of the server 3000 is also common to servers that perform other functions.

[0081] As described above, the server 3000 may be configured such that a computing device (CPU: Central Processing Unit) within its own housing executes the arithmetic processing, or such that part of the program processing is executed on another server. In the following description, it is assumed that the arithmetic device within its own housing executes the arithmetic processing.

[0082] Referring to FIG. 2, the server 3000 comprises a computer device 3010, a network communication unit 3300 for communicating with a network, and a recording medium (e.g., a memory card) 3210 for recording data from the outside and providing the data to the computer device 3010.

[0083] For example, a USB memory, a memory card, or an external storage device can be used as the recording medium 3210. Furthermore, for example, a wired LAN or wireless LAN communication function can be used as the network communication unit 3100.

[0084] As shown in FIG. 2 , the computer main body constituting this computer device 3010 includes, in addition to a disk drive 3030 and a memory drive 3020, a CPU (Central Processing Unit) 3040, each connected to a bus 3050, memory including a ROM (Read Only Memory) 2060 and a RAM (Random Access Memory) 3070, a nonvolatile rewritable storage device 3080, and an input / output interface 3090 for communication via a network and for sending and receiving data with external devices. The nonvolatile storage device 3080 may be, for example, a hard disk drive (HDD) or a solid state drive (SSD). The following description will be given assuming that the drive is an SSD. An optical disk can be inserted into the disk 3030. A memory card 2210 can be inserted into the memory drive 3020 .

[0085] When the programs of the computer device 3010 operate, the data and programs that store the information that is the basis for the operation of the computer will be described as being stored in the SSD 3080.

[0086] 2, the medium capable of recording information such as a program to be installed in the computer main body may be, for example, a DVD-ROM (Digital Versatile Disc), a memory card, a USB memory, etc. In such cases, the computer main body is provided with a drive device (memory drive 3020, disk drive 3030) capable of reading these media.

[0087] The main components of the computer device 3010 are computer hardware and software executed by the CPU 3040. Generally, such software is stored in a storage medium and distributed or distributed via a network, and is obtained via the disk drive 3030 or the network communication unit 3300 and temporarily stored in the SSD 3080. The software is then read from the SSD 3080 into the RAM 3070 in the memory and executed by the CPU 3040. Note that when connected to a network, the software may be directly loaded into the RAM and executed without being stored in the SSD 3080.

[0088] When distributed, the program for functioning as computer device 3010 does not necessarily have to include an operating system (OS) that causes computer main body 3010 to execute functions of an information processing device, etc. The program only needs to include an instruction portion that calls appropriate functions (modules) in a controlled manner and achieves the desired results. How computer system 3010 operates is well known, and a detailed description will be omitted.

[0089] Furthermore, the CPU 3040 may be a single-core processor or a multi-core processor. That is, it may be a single-core processor or a multi-core processor. The server 3000 may also be configured with multiple servers to execute distributed processing.

[0090] 3 is a flowchart showing a pre-analysis of the brain image data analysis process, and FIG. 4 is a flowchart showing the analysis of the brain image data analysis process for a subject by the brain image data analysis device. (Pre-analysis using big data)

[0091] 3, before analysis of the current subject, the image input unit 11 executes an image input step of reading image data from the database 2a based on big data (step S501). The segmentation unit 12 executes a segmentation step for the image input in the image input step S501 (step S502). The segmentation in the segmentation step S502 divides the brain image into multiple structural units. Segmentation can also be performed by preparing, for example, training data that has already been segmented in the database 2a (which stores imaging condition data, age-structural unit volume data, and age-structural unit signal intensity data).

[0092] For example, by using training data such as that shown in US Pat. No. 10,074,173 in Reference 1, the structure of the brain can be defined at multiple structural levels.

[0093] Figure 5 shows segmentation training data at various structural levels. In the example of Figure 5, the coarseness (granularity) of the brain structure definition is defined in five levels: 287 structures, 137 structures, 54 structures, 19 structures, and 8 structures. Finer structure definitions increase the amount of information about brain regions, but reduce data reproducibility and increase dependency on imaging conditions. On the other hand, coarser structure definitions decrease the amount of information about brain regions, but increase data reproducibility and decrease dependency on imaging conditions. As a result of segmentation, all of the defined structures can serve as indicators (volume values, signal intensity values) for measuring the brain's health status. The brain image data analysis device 1 can select structures that are more suitable as health indicators from these many structures, and by performing preprocessing, it is possible to obtain a final indicator with high suitability.

[0094] Here, the correspondence between the "structural units" and the "brain structures" used in the analysis is as follows:

[0095] That is, for example, segmentation can be performed in the smallest unit, and a larger structure can be defined by combining multiple units.

[0096] 3, next, the brain structure volume / signal intensity calculation unit 31 executes an imaging condition influence evaluation step S503 and a brain structure selection step S504. Each processing step executed by the brain structure volume / signal intensity calculation unit 31 will be described in detail below.

[0097] First, as a module of the imaging condition influence evaluation program, the imaging condition influence evaluation unit 33 executes an imaging condition influence evaluation step (step S503). The imaging condition influence evaluation step S503 is a step for evaluating the influence of various imaging conditions from an engineering perspective. The imaging condition influence evaluation unit 33 selects structures that are highly robust from an engineering perspective and evaluates them based on the data in the database 2a. In this step S503, the influence of brain MRI images taken under various conditions on the definition of the structure (e.g., volume) is evaluated. This is because, for example, when MRI images are taken at different resolutions, the volume of a specific structure may be measured as being larger.

[0098] For example, there are nearly 100 parameters for MRI imaging conditions, and it cannot be assumed that imaging conditions are the same between imaging sites (e.g., hospitals). All of these parameters have the potential to affect the volume.

[0099] As a result, strictly speaking, it is impossible to compare data between facilities as it is, and each hospital must collect hundreds or thousands of data points before it is possible to make an assessment such as, "The average hippocampal volume of a 50-year-old is XX, and this person's hippocampus is smaller in comparison." This process must be repeated even if one imaging condition is changed.

[0100] Therefore, when, for example, 505 brain structures are divided by segmentation, the imaging condition influence evaluation unit 33 performs a process of evaluating and separating those that are sensitive to differences in these parameters, with their volume values ​​changing when the parameters are changed, from those that are insensitive, with their volume values ​​remaining almost unchanged regardless of the parameters. The latter is an excellent index from an engineering perspective. Therefore, the imaging condition influence evaluation unit 33 evaluates "how much the volume values ​​change depending on the imaging parameters" based on, for example, changes in the CV (Coefficient of Variation) value or the volume-age correlation curve due to the imaging conditions.

[0101] Next, as a module of the brain structure selection program, the brain structure selection unit 32 executes a brain structure selection step (step S504). In the brain structure selection step S504, brain structures with high suitability for brain structure volume and signal calculation processing are selected. The brain structures selected here include structures with known medical and biological functions, or structures known to be medically important for the disease risk being evaluated. For example, the hippocampus, which is known to be important for memory. Furthermore, from an engineering perspective, structures suitable as indicators require different suitabilities, and in the brain structure selection step 504, brain structures with a large effect on variability from an engineering perspective are selected.

[0102] Below, we will explain the significance of "brain structures with a large effect despite variability." From an engineering perspective, the three most important suitability factors are good reproducibility (small variability when multiple measurements are taken under the same conditions), small influence from imaging conditions (robustness; similar values ​​are obtained even when measurements are taken under different conditions), and high sensitivity to the biological changes we want to detect. In this context, "sensitivity" refers to the ratio of the biological change (effect), its magnitude, and reproducibility; even if the effect is high, if the variability is large, it cannot be considered an excellent indicator from an engineering perspective.

[0103] Therefore, if the risk of dementia is to be evaluated, for example, the brain structure selection unit 32 will determine, from the standpoints of sensitivity and reproducibility, as will be described later, to use only the volume of the ventricles as an index from the 505 structures obtained by segmentation from the imaging data of the subject, or to use an index that is a combination of the ventricles and other brain structures. The brain structure selection unit 32 stores the brain structures selected in this way, for example, in database 2a, in association with the disease to be evaluated, as selected brain structure data.

[0104] The method of defining the brain structure at multiple levels using the training data described above, as shown in Figure 5, is extremely effective in finding robust structures.

[0105] Thereafter, the brain image data analysis device 1 of the first embodiment executes processing by the brain structure volume / signal intensity calculation unit 35 (S505) and processing by the imaging condition calibration unit 15 (step S506). <Processing by the imaging condition calibration unit 15>

[0106] The imaging condition calibration unit 15 first performs a "brain structure volume-age correlation creation process" in which, prior to analysis of the current subject, it creates a volume-age correlation curve for the measurement results of brain structures of multiple people of different ages (each subject) based on existing big data.

[0107] Next, the imaging condition calibration unit 15 performs a "brain structure volume-age correlation parameter extraction process" to extract, for each imaging condition, volume-age correlation parameters that define the correlation curve created by the brain structure volume-age correlation creation process, as well as statistical parameters (such as average values).

[0108] Next, the imaging condition calibration unit 15 performs an "imaging condition parameter matching process" to determine calibration coefficients to match or minimize the volume and signal intensity of each brain structure so that the differences in the parameters of the volume-age correlation curves and statistical parameters extracted in the "brain structure volume-age correlation parameter extraction process" are minimized even when the imaging conditions are different. Here, the "calibration coefficient" is not particularly limited, but refers to, for example, a coefficient used to calibrate the volume of a specific brain structure so that it matches between imaging conditions A and B. Similarly, the "calibration coefficient" refers to a coefficient used to calibrate the signal intensity so that it matches between imaging conditions A and B. Note that, for example, the "brain structure volume-age correlation parameter" can be the average value and standard deviation of the structure at each age. Therefore, calibration is a correction between group A and group B (for example, two groups imaged using MRI devices with different resolutions). If we assume that the average for Group A is 10% larger than the average for Group B at age 30, and Group A is considered the standard, then by reducing the volume of the 30-year-old patients in Group B by 10%, the data from Group B can be compared with the data from Group A.

[0109] To calibrate the difference in volume values ​​due to different shooting conditions, the shooting condition calibration unit 15 can also take images of the same person under different shooting conditions and measure the influence of each shooting condition (for example, resolution) to perform calibration. Steps S701 to S703 show one method of calibrating the difference in volume values ​​due to different shooting conditions (when not using the same person).

[0110] For example, there are two methods for determining the calibration coefficients:

[0111] In the first method, for example, if the average value and standard deviation of a 30-year-old person under condition A are average AA and deviation SA, and under condition B they are average AB and deviation SB, then by setting volume data B of the person photographed under condition B to B+(average AA-average AB), the average value of data B photographed under condition B can be compared with condition A. In other words, (average AA-average AB) becomes the calibration coefficient for B→A conversion. In this case, it is assumed that there is a certain amount of measurement data under conditions A and B, and that the averages of each can be calculated.

[0112] The statistical parameters of both conditions may be matched based on either condition A or condition B. For example, if the condition is the strength of the measurement magnetic field, the statistical parameters may be matched to the measurement data under the condition of the highest magnetic field strength, which is thought to provide the highest resolution. Alternatively, if the slice thickness is different between condition A and condition B, the statistical parameters may be matched to the measurement data under the condition of a thinner slice pressure, which allows for the observation of a finer structure.

[0113] The second method is to use the statistical value of A as prior information when there is a large amount of data for condition A and a small amount of data for condition B.

[0114] A more rigorous implementation of this concept can be achieved using so-called Bayesian estimation.

[0115] In the above explanation, calibration is performed by comparing the average value and standard deviation for each age group to achieve a match. However, for example, when the average value is in the 20s to 80s range, the average values ​​for ages 20 to 80 under conditions A and B may be applied to a polynomial, and the calibration coefficients may be calculated so that the polynomials match. In this case, differences in imaging conditions may be added as covariates.

[0116] The above-mentioned Bayesian estimation will be explained in more detail below.

[0117] That is, for example, let us say that the average of structure X at age 30, found from 30,000 cases under imaging condition A, is θ, and the standard deviation is γ. Let us say that under imaging condition B, the results of volume measurement of structure X for n people are average μ and standard deviation σ. A Bayesian model can be used to estimate the posterior distribution. f(θ|data) ∝ f(θ)f(d|θ) For example, if both the prior distribution f(θ) and likelihood f(d|θ) follow normal distributions, the posterior distribution (estimation of the average under imaging condition B) is: Mean: θγ 2 / (nγ 2 +σ 2 ) + μnγ / (nγ 2 +σ 2 ) Variance: γ 2 σ 2 / (nγ 2 +σ 2 ) is a normal distribution. (Analysis for current subjects)

[0118] Next, with reference to FIG. 4, the process of analysis performed by the brain structure volume / signal intensity calculation unit 31 based on the brain image data and imaging condition data of the subject currently being evaluated will be described.

[0119] After the image input step (step S510) and the segmentation step (step S512), the brain structure volume / signal intensity calculation unit 31 executes a brain structure identification step S514, a brain structure volume / signal intensity calculation step S516, processing by the imaging condition calibration unit 15 (step S518), and processing by the same age ratio calculation unit 16 (step S520).

[0120] First, for analysis of the current subject, the image input unit 11 executes an image input step in which brain image data and imaging condition data of the subject to be evaluated are transmitted from the imaging site (the facility where the imaging device is installed and imaging is performed) (step S510). The segmentation unit 12 executes a segmentation step for the image input in the image input step S510 (step S512). The segmentation in the segmentation step S512 divides the brain image into multiple structural units (e.g., 505 structural units). Again, the segmentation can be performed with reference to the segmentation training data in the database 2a described above.

[0121] Next, as a module of the brain structure identification program, the brain structure identification unit 34 performs labeling of brain structures for the structural units divided by segmentation, and executes a brain structure identification step of identifying structures that are less affected by the imaging conditions according to the prior evaluation by the imaging condition influence evaluation unit 33 (step S514). In the brain structure identification step S514, structures are identified that are robust and less affected by the imaging conditions, and that can be minimized by calibration, and that are selected taking into account their sensitivity to the disease risk of the evaluation target. As described above, examples of robust structures include the ventricles and large lobe structures (e.g., the frontal lobe) when dementia risk is the target.

[0122] Next, as a module of the brain structure volume / signal intensity calculation program, the brain structure volume / signal intensity calculation unit 35 calculates the volume and signal intensity for each brain structure (step S516). That is, the volume and signal intensity of the brain structure are calculated from the structures with large volumes, such as the ventricles and lobe structures, identified in step S514.

[0123] Here, for example, the volume value of the identified brain structure is calculated by multiplying the number of voxels contained in the structure by the volume of one voxel. Furthermore, the signal intensity calculation is equivalent to the average value of the signal intensities of all voxels contained in the structure. In this case, the signal intensity calculation is not limited to the average value, and may be a median, moment, or histogram.

[0124] Here too, the method of defining the brain structure at multiple levels using the training data described above, as shown in FIG. 5, is extremely effective in finding robust structures.

[0125] Thereafter, the brain image data analysis device 1 of the first embodiment executes processing by the imaging condition calibration unit 15 (step S518) and processing by the same age ratio calculation unit 16 (step S520). <Processing by the same age ratio calculation unit 16>

[0126] The peer ratio calculation unit 16 executes an age-brain volume distribution creation process, an age-brain volume distribution parameterization process, and an age percentile calculation process.

[0127] Here, the peer age ratio calculation unit 16 corresponds to a function in which the CPU executes the corresponding module in the program stored in the memory when the computer operates as the brain image data analysis device 1. The age-brain volume distribution creation process, age-brain volume distribution parameterization process, and age percentile calculation process are implemented as sub-modules in the module of the peer age ratio calculation unit 16, each of which is executed by the CPU. <Configuration Description>

[0128] The age-based brain volume distribution generation process generates a population distribution of the volumes of selected brain structures at each age, where "population distribution" refers to the distribution of the mean and standard deviation of volumes obtained from three or more subjects.

[0129] The age-brain volume distribution parameterization process calculates parameters of the volume distribution of brain structures.

[0130] The age percentile calculation process uses the calculated parameters to calculate an age percentile for each subject based on their age.

[0131] The age-brain volume distribution generation process generates a population distribution of the volume of the selected structures at each age.

[0132] Next, the age-brain volume distribution parameterization process parameterizes the population distribution. Experimental results showed that using the logarithm of ventricular volume on the horizontal axis approaches a normal distribution. Therefore, for example, if a normal distribution model is used, the distribution can be parameterized using two values: the mean and standard deviation.

[0133] FIG. 6 is a diagram showing age-dependent changes in the mean value of ventricular volume and a predetermined variance.

[0134] As an alternative to the parameterization described above, where the horizontal axis represents the logarithm of the ventricular volume and the normal distribution is used, a method (e.g., Bayesian statistics) can be used to directly calculate percentiles from raw data, as shown by the dashed line in Figure 6.

[0135] Next, the age percentile calculation process can use the determined parameters to determine percentiles based on the age of each individual.

[0136] The cerebrovascular disorder degree calculation unit 17 calculates the degree of cerebrovascular disorder.

[0137] First, to calculate the volume of brain structures containing abnormal white matter signal regions, we incorporate the abnormal white matter signal regions into the segmentation training data.

[0138] The same age ratio is calculated from the age correlation of the brain white matter signal abnormality area in the database 2a.

[0139] Next, as a simultaneous calculation process for cerebral atrophy and cerebral white matter signal abnormality regions, the degree of cerebral atrophy and cerebrovascular disease are calculated simultaneously from the same MRI image.

[0140] As described above, the volume of each structure can be used to determine the shape of the brain and the degree of atrophy. Meanwhile, the degree of cerebrovascular damage can also be measured using the structural definitions provided in the training data. For example, by defining abnormal signal intensity regions in the white matter of a specific brain structure using the training data, the volume of the abnormal signal regions can be automatically calculated for each individual in the process of calculating the volume of the abnormal signal regions in the brain white matter signal.

[0141] 7A and 7B are graphs showing the incidence of ventricular atrophy in examinees by age group.

[0142] Figure 7(a) shows the frequency of ventricular volume atrophy in male examinees (by age group: 30s, 40s, 50s, 60s, and 70s), and Figure 7(b) shows the frequency of ventricular volume atrophy in female examinees (by age group: 30s, 40s, 50s, 60s, and 70s).

[0143] 8A and 8B are diagrams showing a histogram and log-normal distribution of age-specific atrophy (logarithmic value of ventricular volume).

[0144] Figures 8(a) and 8(A) are histograms of the degree of atrophy (logarithmic value of ventricular volume) by age, and Figures 8(b) and 8(B) show a log-normal distribution. In the log-normal distribution, for example, the average degree of atrophy for people in their 70s and the position where atrophy has progressed by one standard deviation from the average degree of atrophy are also shown. 68.2% of people in their 70s fall within the "average of people in their 70s + 1 standard deviation."

[0145] Conversely, for example, if a person in their 60s falls into the range exceeding the "average for people in their 70s plus one standard deviation," it is highly likely that their atrophy is more severe than that of people of the same age.

[0146] FIG. 9 is a diagram showing the difference between the population analysis obtained from the normal distribution model and the AD patient region.

[0147] As shown in the figure, population analysis can identify the "AD patient region" by comparing the degree of ventricular atrophy with the average value for the same age. If the degree of ventricular atrophy falls within this AD patient region, supplementary information can be provided indicating a high risk of dementia.

[0148] Alternatively, by using statistical analysis or machine learning, it is possible to generate a "brain imaging risk model" that takes the degree of ventricular atrophy as input and outputs the risk of developing dementia in multiple stages or as a numerical probability.

[0149] Furthermore, the ventricles are brain structures with a relatively large volume, and volume evaluation can be performed not only by imaging with an MRI device but also by imaging with an X-ray CT device.

[0150] In patients with Alzheimer's disease, there is a strong positive correlation between the volumetric atrophy of the hippocampus, which has been reported to shrink, and the volumetric atrophy of the limbic system. Furthermore, the volume of the ventricles tends to expand in accordance with the volumetric atrophy of the hippocampus.

[0151] Here, the limbic system is a collective term for the paleocortex (hippocampus, fornix, dentate gyrus), paleocortex (olfactory lobe, piriform lobe), intermediate cortex (cingulate gyrus, hippocampal gyrus), and subcortical nuclei (amygdala, septum, mammillary bodies).

[0152] The ventricles are cavities in the brain where cerebrospinal fluid is produced, and in humans there are a total of four ventricles: a pair of lateral ventricles on the left and right, and one third ventricle and one fourth ventricle in the midline. (Data sharing structure using temporary IDs)

[0153] The above explanation has been given of a system and method for assessing "disease risk," such as "risk of dementia," by evaluating the volume of specific brain structures using an apparatus capable of capturing cross-sectional images of the human body, such as an MRI apparatus or an X-ray CT apparatus.

[0154] However, the imaging data acquired by such imaging devices, the disease risk assessed from the imaging data, and other clinical information about the subject are extremely valuable "medical data" and "healthcare data."

[0155] Therefore, the following describes a configuration for making such "medical data" and "healthcare data" available for sharing with third parties.

[0156] The following are prerequisites:

[0157] 1) Both "medical data" and "healthcare data" are personal information that requires special care in handling. In particular, information related to a subject's disease history is required to be handled with the strictest of care under the Personal Information Protection Act. In Japan, this information is treated as sensitive personal information.

[0158] 2) On the other hand, from the perspective of effective use of valuable data, progress is being made in establishing a system that allows data to be provided to third parties by "anonymizing" each piece of data. However, while the true significance of "medical data" and "healthcare data" lies in the time series data (longitudinal data) of a specific individual, "data that is longitudinally related to the same person" is not necessarily desirable from the perspective of "anonymization."

[0159] 3) Therefore, one possible way to achieve both the protection of personal information and the use of data is to pseudonymize the data, process it so that it cannot easily identify a specific individual, notify the subject himself / herself of the purpose of use, etc., and use the data with his / her consent.

[0160] FIG. 10 is a diagram showing the concept of data flow from data acquisition to data joint use, in order to achieve both kana processing of data and usability as longitudinal data.

[0161] First, as will be described later, it is assumed that the system operator who provides the platform for shared data use issues identification information CP-ID as a shared temporary ID to each subject. The shared temporary ID is used in both health checkups for healthy individuals and diagnosis and treatment for patients.

[0162] Although not particularly limited, for example, the identification information CP-ID is converted into a two-dimensional code and stored in the subject's smartphone, and is configured to be displayed on the smartphone screen only when the user performs special access.

[0163] For example, a subject who has been issued such a shared temporary ID is to present this shared temporary ID to a hospital or clinic when undergoing imaging at an imaging site.

[0164] In the system operator's system, the shared temporary ID and the subject's personal information are associated and stored in a secure manner for security reasons.

[0165] Then, at the imaging site that is the data provider (for example, a data provider such as a hospital or clinic), the subject's name, a pseudonymized ID issued to each subject for the operation of the imaging site, information indicating whether consent has been obtained from the subject for data sharing, and information indicating whether the subject has requested the deletion of their data are associated with the shared temporary ID and stored in a storage device as a "personal identification information DB."

[0166] On the other hand, the data provider deletes information that can identify individuals, such as names and health insurance card numbers (including MAN numbers in Japan and Social Security numbers in the United States), and stores the subject's gender, age (coarse-grained to years, etc., as necessary), test results (imaging data, disease risk assessment indicators based on imaging), drug prescription information, and other clinical data in a storage device separate from the personal identification information DB, associated with the pseudonymized ID and shared temporary ID as a "pseudonymized processed information DB."

[0167] In this specification, "anonymously processed information" means "information about an individual obtained by processing personal information so that a specific individual cannot be identified, and the personal information cannot be restored." "Pseudonymously processed information" means "information about an individual obtained by processing personal information so that a specific individual cannot be identified unless it is compared with other information."

[0168] In this case, i) personally identifiable information will be stored separately from pseudonymized information, ii) it will be managed so that it cannot be compared with the personally identifiable information database unless certain requirements are met, and iii) only data for which the individual has given consent for data sharing will be registered in the pseudonymized information database.

[0169] When data is used for its intended purpose at the time of acquisition, such as health checkups or medical treatment, within a hospital or other facility, it will be permitted to be used as personal information, provided that certain requirements are met and matching with a pseudonymized ID is possible.

[0170] On the other hand, for joint purposes with the consent of the individual, only pseudonymized information will be shared.

[0171] With the above configuration, it is possible to share data in a manner that makes it difficult for anyone other than the data source to easily identify the subject, and within the scope of the subject's consent.

[0172] In the following, the operation of the system will be described on the assumption that imaging data and clinical data are stored from the imaging site in a PACS on the cloud.

[0173] 11 and 12 are conceptual diagrams showing a configuration that enables the use of the kana-processed information as explained in FIG. 10 by the system operator and by the data user.

[0174] Here, although not limited to, it is assumed that the system operator manages and operates the service page operation server 1000 and the analysis service providing server 2000. Furthermore, while a company's health insurance association or human resources department is also assumed as a data user, here, an insurance company is assumed as an example. Insurance companies that handle life insurance, dementia insurance, and nursing care insurance for individuals, as well as accident insurance for companies, are assumed as insurance companies.

[0175] Here, the service page management server 1000 provides a service such as authenticating the request as being from a subject 100 who is a registered member, for example, on a smartphone 200 owned by the subject, and then providing a service page containing the results of the health check, imaging data related to the results of the diagnosis, clinical data, advice information and recommended information for the subject, etc., which are displayed on the screen of the smartphone 200.

[0176] Furthermore, the service page operation server 1000 issues identification information CP-ID as a shared temporary ID as described in FIG. 10 to the subject (user 100) who is a member.

[0177] FIG. 13 is a diagram showing an example of an "advice chart" provided together with the "analysis results" notified to the subject from the service page operation server 1000 as an analysis report.

[0178] As described above, the analysis service providing server 2000 notifies the subject of the results of an assessment of, for example, the "risk of dementia" based on volumetric evaluation of specific brain structures in the brain, based on imaging data from an apparatus capable of capturing cross-sectional images of the human body, such as an MRI apparatus or an X-ray CT apparatus.

[0179] As shown in FIG. 13, for example, the "risk of dementia" is presented in four levels, "A good," "B standard," "C caution," and "D caution," depending on the degree of ventricular atrophy.

[0180] Here, the criteria for dividing the risk into stages are not particularly limited, but one possible method is to divide the risk into stages according to the degree of atrophy of brain structures, based on a predetermined standard of the magnitude of the relative risk.

[0181] As shown in FIG. 13, recommended information is displayed regarding actions recommended to the subject depending on the stage of the evaluation result and the age range (20 to 45 years, 46 to 65 years, 65 to 80 years).

[0182] If the patient's condition is "A Good," regular checkups every two to three years are recommended for all age groups.

[0183] For the "B Standard" category, people aged 46 to 65 and 65 to 80 are recommended to undergo regular medical checkups every two to three years and to improve their lifestyle habits.

[0184] If the patient is rated "C Caution," then a follow-up visit in two years is recommended for those aged 20 to 45, and lifestyle improvements are strongly recommended for those aged 46 to 65 and 65 to 80. For example, improvements in negative lifestyle factors such as drinking, high blood pressure, obesity, and high blood sugar are recommended.

[0185] If the rating is "D Caution," those aged 20 to 45 are recommended to return for a follow-up visit in two years, and those aged 46 to 65 and 65 to 80 are recommended to visit a memory clinic and receive counseling.

[0186] FIG. 14 is a flowchart for explaining the operation of the system shown in FIGS.

[0187] The imaging site 5000 where health checkups and medical examinations are performed is, for example, a hospital or a clinic.

[0188] 14, 12, and 13, first, a user 100 who will be a subject accesses the service page operation server 1000 and performs user registration (S900). Then, the service page operation server 1000 issues a CP-ID, which is a shared temporary ID that can be used both for health checkups and medical treatment at medical institutions (S902). After user registration, the user 100 is called a member 100.

[0189] The storage device 1100 connected to the service page operation server 1000 stores big data that associates the above-described brain structure images with healthcare information for each subject. The "healthcare information" includes information on dementia health checkups (hereinafter referred to as "health checkup information") and medical information such as prescriptions. While not particularly limited, the "prescription data" here is provided by the member 100 or the imaging site with the member's consent. However, as described below, the "prescription data" may be in other formats as long as it contains information that can at least identify the onset of dementia. Furthermore, the storage device 1100 receives data from the service page operation server 1000 and stores the data in association with a shared temporary ID and personal information such as the member's name. As described below, if the member has given consent, and within the scope of that consent, the storage device 1100 also stores information on the captured images, the contents of the analysis report, and other related healthcare information and clinical information after the member's health checkup or medical examination.

[0190] The smartphone 200 of the member 100 that receives the shared temporary ID stores the shared temporary ID as, for example, a two-dimensional code, although this is not particularly limited.

[0191] When member 100 visits imaging site 5000 (hospital or clinic) for a health check (brain checkup) or medical treatment, member 100 displays a two-dimensional code of the shared temporary ID on the screen of his / her smartphone 200 and has it read into the system of imaging site 5000 (S904).

[0192] However, the method for passing the shared temporary ID to the system of the imaging site 5000 is not limited to using such a two-dimensional code, and may also be configured to transmit it via encrypted wireless communication using technology such as short-range communication.

[0193] Then, the member 100 is photographed at the imaging site 5000 (S906).

[0194] Image data captured at the imaging site 5000 is stored and registered in, for example, the cloud PACS 2100 (S908).

[0195] At this time, although not limited to this, for example, when storing in the cloud PACS 2100, personal identification information may be managed separately at the imaging site 5000, or if the cloud PACS is managed with the same level of security as the in-hospital system, the data registered in the cloud PACS 2100 is associated with information on the shared temporary ID and in-hospital ID, as well as image information, personal information such as patient names, and information on the member 100's consent to the shared use of the data.

[0196] Although not limited to this, data stored in cloud PACS 2100 may be encrypted to ensure stricter security, and furthermore, the data may be divided and stored on multiple servers, and the data stored on each server may be divided and stored using technology such as electronic tallying, thereby further improving security.

[0197] After the image data is stored in the cloud PACS 2100, the analysis service providing server 2000 executes an analysis process on the data in the cloud PACS 2100 (S910) and creates an analysis report (S912).

[0198] Here, "analysis processing" refers to the analysis described above, and includes analysis for "assessing the risk of dementia."

[0199] The analysis service providing server 2000 returns the analysis report to the system of the imaging site 5000 (S920), and the report is displayed on the screen of the system's display device (S922). In a hospital, this displayed report can be viewed by a doctor and used as auxiliary diagnostic information. In addition, in the case of a health checkup, the analysis report can be printed out and provided to the member 100.

[0200] Meanwhile, the analysis service providing server 2000 transmits the analysis report to the data sharing server 3000 , and the data sharing server 3000 stores the analysis report in a shared database (hereinafter, shared DB) 3100 .

[0201] The shared DB 3100 stores image information, medical examination information, clinical information, analysis reports, etc. in association with the shared temporary ID, within the scope of the consent of the member 100. As a result, the shared DB 3100 does not store any personally identifying information.

[0202] The shared DB 3100 also stores, as a recommended behavior database, behaviors recommended to the member 100 as behavioral changes in response to medical checkup information, clinical information, analysis report results, and health-related information.

[0203] Here, the "health-related information" includes information collected by questionnaires about lifestyle habits and the like administered to members 100.

[0204] Therefore, when returning the analysis report to the member 100, the service page operation server 1000 may refer to the recommended action database and include the recommended action in the analysis report returned to the member 100.

[0205] In addition, in response to access from a member, the service page management server 1000 may display the member's analysis report, brain images, etc. in chronological order or selectively on the display screen of the member's 100 smartphone 200.

[0206] Furthermore, the service server operation server 1000 can be configured to provide points to users when they receive health-related information from members 100, depending on the content and amount of information provided. The points can be applied to discounts on brain checkup (head MRI scan) fees and affiliated service fees. This provides an incentive for members 100 to sign up for such services.

[0207] By accessing the shared DB 3100, the data user server 4000 stores in the user data server 4100, in association with the shared temporary ID, the temporary subscriber ID for the user's service (here, life insurance), and "health care information" of clinical information and analysis reports to the extent that the member 100 has agreed to the data user's use. The data user separately manages a correspondence table between personal identification information and temporary subscriber IDs. As a result, the personal identification information is not stored in the user data server 4100.

[0208] With the above configuration, brain image data, health check information, clinical information, and analysis report data captured at the imaging site 5000 can be shared within the scope of the member's 100 consent, and in a manner that makes it difficult to identify individuals without directly associating them with personal identification information.

[0209] As a result, as the data stored in the shared DB 3100 accumulates, not only will the accuracy of assessing the risk of dementia and other conditions improve, but data users will also be able to provide healthcare services based on the subject's healthcare information.

[0210] Thus, according to this embodiment and other embodiments, it is possible to realize a brain image analysis data utilization system that minimizes the effects of uneven imaging conditions seen in brain checkups, enables accurate and reproducible measurement of atrophy in various parts of the brain, and enables shared use of the analysis results of a brain image data analysis device that analyzes the risk of developing dementia.

[0211] Brain imaging data can also be used to assess a subject's risk of disease and provide information for behavioral change (processing at the insurance company, which is the data user).

[0212] As described above, it is envisioned that insurance companies will use the data in shared DB 3100 to calculate assessment criteria for "risk selection."

[0213] Generally, insurance risk events include hospitalization, surgery, and onset of illness. However, if it becomes possible to predict the risk of onset of dementia, this will provide many insights into optimizing underwriting standards in risk selection.

[0214] This is because one of the conditions for dementia insurance payment is the payment of a lump sum if a person is diagnosed with dementia. Furthermore, after being diagnosed with dementia, the person will be certified as needing nursing care, and will be required to pay a lump sum nursing care payment and a nursing care pension.

[0215] Therefore, as an example of an insurance assessment standard for dementia, we will explain an example of constructing a "model for the risk of developing dementia."

[0216] When taking out life insurance or dementia insurance, certain disclosures are required, and these disclosures contain a lot of information about your health.

[0217] On the other hand, insurance companies can obtain information on the occurrence of diseases over a certain period of time after enrollment through multi-year insurance contracts.

[0218] Therefore, by analyzing this information, along with health checkup and prescription data, it is possible to build a model that predicts the future risk of developing dementia based on the health condition at the time of enrollment. Therefore, for insurance companies, it will also be possible to obtain "analyzed time-series data of brain health conditions with dementia flags" for future insurance product development.

[0219] However, compared to other diseases, dementia takes a long time to develop from the appearance of early symptoms, so this characteristic must also be taken into consideration.

[0220] The information handled during the initial health check can be broadly divided into two types: general health check information and disclosure items specific to private insurance.

[0221] Of these, the health checkup information includes various numerical items that can be determined through blood tests, etc., while the notification items include a huge amount of information such as past medical history.

[0222] The challenge in predicting the future onset of dementia is to identify information that is truly related to the disease from this information. Therefore, we first convert various types of information into a format that is easy to analyze, and then create a continuous or binary feature dataset.

[0223] Next, in order to derive factors that contribute to the risk of lifestyle-related diseases from the vast amount of information, we will combine machine learning and statistical analysis to extract and model risk factors that contribute to the onset of dementia.

[0224] For example, the construction of "onset risk models," although they are models for the onset risk of cardiovascular disease, diabetes, cerebrovascular disease, and other diseases other than dementia, is disclosed in the following publicly available documents. Publicly available document 1: Takehiko Baba et al., "Study on disease-specific onset risk prediction models using health checkups and medical insurance claims," ​​National Health Insurance Association, https: / / www.kyoukaikenpo.or.jp / ~ / media / Files / honbu / cat740 / houkokusho / h30 / 08tokyo3.pdf Publicly available document 2: Mitsuru Tsunekawa et al., "Predicting the onset of lifestyle-related diseases using health checkup data," Materials from the AIM Joint Research Meeting of the Japanese Society for Medical Informatics and the Japanese Society for Artificial Intelligence, SIG-AIMED-007-10 Publicly available document 3: Fumie Yaegashi et al., "Predicting the onset of lifestyle-related diseases using medical insurance claims data," The 33rd Annual Conference of the Japanese Society for Artificial Intelligence, 2019. 1H3-J-13-05

[0225] Here, "receipt data" refers to the detailed medical fee statement that a medical institution bills an insurer for the medical treatment a patient has received. For example, this includes basic information such as the patient's gender, age, and the date of treatment, as well as the name of the diagnosed illness, medical treatment, and prescribed medications.

[0226] For example, in Reference 2, the existence of prescription data means that medical treatment has begun at a hospital or other facility, and is therefore used as information to identify the time when the target disease was first diagnosed.

[0227] In other words, assuming that longitudinal data is collected on a specific subject, from their health status to the onset of dementia, and given certain "health checkup information," the medical insurance claims data is used to assess the probability that the subject corresponding to that "health checkup information" will develop the disease in the future.

[0228] In Reference 2, positive case data (data of healthy people who will develop the disease and be diagnosed with the target disease within a certain number of years) from a time before the disease has yet to develop, and negative case data (data of healthy people who will not be diagnosed with the target disease within a certain number of years) are prepared as training data by undersampling and bagging. Then, a model is generated by machine learning to identify whether or not the disease will develop within a certain period of time using health checkup data as input.

[0229] In addition, disease onset risk models generated by machine learning are not limited to such binary (or multi-valued) classification models; machine learning can also be used to generate models that predict the probability of onset.

[0230] 15A and 15B are explanatory diagrams showing an example of health check information and subject attribute information used to generate a disease development risk model in embodiment 1. FIG.

[0231] Such data is stored as newly acquired data 1120 in the storage device 1100 shown in Fig. 11. It is also stored as brain image big data / analysis data 1110, and at least a part of it can be used as needed as the shared DB 3100. Therefore, to generate a development risk model, the brain image big data / analysis data 1110 and the newly acquired data 1120 can be used together, directly or indirectly, as learning data.

[0232] 15(a) and 15A, the health checkup information is information relating to the results of a health checkup (medical examination) received by a subject, and includes the subject ID, the year of medical examination indicating the year in which the subject underwent the medical examination, BMI (Body Mass Index), systolic blood pressure, triglycerides, fasting blood glucose, the results of the medical interview, and the degree of atrophy of brain structure based on brain cross-sectional images. The results of the medical interview include, for example, information indicating whether or not the subject has a drinking habit, whether or not the subject has a physical exercise habit, etc. The above items from BMI to fasting blood glucose are representative examples, and in reality, the health checkup information does not need to include all of these items, and may also include information on items other than these (e.g., diastolic blood pressure, etc.).

[0233] In the case of brain image big data / analysis data 1110, the "subject ID" is a temporary ID at the time of data acquisition, and in the case of newly acquired data 1120, it can be a shared temporary ID.

[0234] The health checkup information is an example of information indicating a person's health condition (health condition information), and other information indicating a person's health condition may be used instead of or in addition to the health checkup information. Examples of other information indicating a person's health condition may include information about the person's health condition measured by a device worn by the person (for example, a so-called wearable device, etc.) (for example, the person's amount of exercise, sleep time, pulse rate, blood pressure, non-invasively measured blood glucose level, etc.).

[0235] Here, as a method for measuring blood glucose levels non-invasively, for example, when blood glucose levels rise, the band of light incident on the skin changes, and the amount absorbed in the wavelength range close to blue decreases. This change causes the color of the reflected light to change, and the intensity of the detected light changes accordingly. Methods for measuring blood glucose levels non-invasively by detecting this change have been put into practical use.

[0236] Specifically, it has been reported that people with diabetes are about 1.5 times more likely to develop Alzheimer's disease and about 2.5 times more likely to develop vascular dementia in old age than those without diabetes. It has also been reported that severe hypoglycemia, a side effect of diabetes treatment, increases the risk of developing dementia. Therefore, to prevent dementia, it is important to maintain stable blood sugar levels within a range that does not cause hypoglycemia. At the same time, it is also an important factor in predicting the risk of developing the disease.

[0237] FIG. 5B(b) is an explanatory diagram showing an example of attribute information related to health information.

[0238] Attribute information is information about the subject's attributes, including the subject ID, gender, date of birth, occupational category, and date of membership registration. For example, when registering as a group at a workplace, the date of joining the workplace's health insurance association (health insurance) may be registered instead of or in addition to the date of membership registration. This makes it possible to match the date of joining the health insurance association with the health insurance association's medical receipt data, thereby improving the accuracy of identifying the date of dementia onset, including for mid-career employees. For example, consider a case where dementia-related medical data is first recorded in the medical receipt data after joining the health insurance association. In this case, if the period from the date of the first dementia-related medical receipt data to the date of the next dementia-related medical receipt data is shorter than the period from the first dementia-related medical receipt data after joining the health insurance association by a certain amount, the date on which dementia-related medical data first appears in the medical receipt data can be identified as the "date of dementia onset." If we imagine a case where a company takes out accident insurance to prepare for accidents that occur during work, insurance companies could use this type of prescription data to obtain information on the "date of onset of dementia."

[0239] In addition, when individuals take out life insurance, dementia insurance, nursing care insurance, etc., premiums are billed based on the onset of dementia, so in this case too, insurance companies will be able to obtain information on the timing of the onset of dementia.

[0240] 16 is a functional block diagram for explaining the functional configuration of the data user server 4000. As described above, it is assumed here that the data user server 4000 is managed by an insurance company.

[0241] Referring to Figure 16, the data user server 4000 includes a communication interface 4002 for exchanging data with the data sharing service server 3000 via the network 2, an arithmetic processing unit 4100, a memory 4200 for temporarily storing data, and a storage unit 4300 for storing data in a non-volatile manner.

[0242] The storage unit 4300 stores data used in the process of generating the disease risk model and calculating the insurance assessment standard.

[0243] Generally speaking, a "database" can be a database with a well-defined schema and capable of using transactions, such as a relational database or an object-relational database, or an object database or column-oriented database management system, in which data in the same column is aggregated and stored in a physically close area.

[0244] The data stored in the memory unit 4300 includes brain image big data analysis data 4310, health data 4320, notification data 4330 submitted by the insured person when taking out insurance, brain image risk model analysis information data 4340, disease onset risk model information data 4350, and insurance assessment standard data 4360.

[0245] Here, as an example, the brain image big data analysis data 4310 is data that models the risk of a subject developing dementia after a certain number of years (or a certain period of time) based on the degree of atrophy of a specific brain structure using statistical analysis or machine learning of already acquired brain image big data. Therefore, with this data, even if the longitudinal data of medical insurance claims data such as those described above does not exist, it is possible to evaluate the risk of a current insured person developing dementia after a certain number of years from the brain image data. Furthermore, the brain image risk model analysis information data 4340 is, for example, data that evaluates the degree of atrophy of a specific brain structure by analyzing the insured person's own cross-sectional brain image data submitted by the insured person using the analysis service providing server 2000, and the atrophy degree is recorded in association with the insured person's subject ID.

[0246] The onset risk model information data 4350 stores, for example, model parameters for identifying the onset risk model used to set assessment standards for the insured, calculated by one of the following methods.

[0247] 17A and 17B are diagrams showing a first method for calculating a development risk model.

[0248] 17 is basically based on the premise that longitudinal data has been acquired for a plurality of specific subjects, from their health status to the onset of dementia, as in Publicly Known Document 2. Given certain "health checkup information," medical examination data from a medical institution is used to evaluate the probability that the subject corresponding to such "health checkup information" will develop a disease in the future.

[0249] 16, 17(a), and 17B, when processing starts (S600) and data of a new member is registered in the data sharing service server 3000, the learning dataset generation module 4110 acquires health information such as medical checkup data and brain image data obtained from a brain checkup from the shared DB 3100 in association with the shared temporary ID, and stores these in the brain image data 4310 and the health data 4320, respectively (S602). Note that the health data is assumed to be associated with the user's attribute information (age, gender, etc.).

[0250] The stored brain image data may be the captured brain image data itself, or may be converted from the brain cross-sectional image data by the above-mentioned method into data on the volume of a predetermined brain structure calculated and then stored. In addition, the atrophy degree of a predetermined brain structure evaluated by the analysis service server 2000 for the brain image data is also stored as brain image risk model analysis information data 4340. The same applies hereinafter to the storage of brain image data 4310 in the storage device 4300.

[0251] In addition, when a user who holds a shared temporary ID applies for subscription to an insurance company's services, the learning dataset generation module 4110 associates the notification data submitted upon subscription with the shared temporary ID and stores it in the notification data 4330 (S604).

[0252] In addition, when a user who is subscribed to an insurance company's service and holds a shared temporary ID visits a medical institution for dementia-related issues and applies for insurance payment along with information about the results, the learning dataset generation module 4110 associates the received data with the shared temporary ID and stores it in the health data 4320 (S606).

[0253] Furthermore, when applying for insurance premium payment, the user has a brain tomographic image taken by a medical institution, and the brain image data 4310 is imported into the data user server 4000 via the user or via the shared DB 3100 and stored in the brain image data 4310. Here, too, the data may be converted into volume data of the brain structure before being stored. Furthermore, the degree of atrophy of a specified brain structure evaluated and provided by the analysis service server 2000 is stored in the brain image risk model analysis information data 4340.

[0254] In this way, when a predetermined number or more of the policyholder's data from health status to the onset of dementia has been accumulated, the important risk factor extraction module 4140 extracts important risk factors for the risk of onset from the health data 4320, notification data 4330, and brain image risk analysis information 4340 (S608). Here, in addition to well-known methods such as the "filter method" and "wrapper method" for so-called feature selection, the so-called "embedded method" may also be used, in which feature selection is performed in parallel when the onset risk model is generated in the onset risk model generation module 4150 described below.

[0255] The onset risk model generation module 4150 receives the selected feature values ​​as input and generates an onset risk model that outputs the onset risk in stages (S610). Note that the output of the onset risk model is not limited to a configuration that outputs one of multiple levels as described above, and may also be a model that outputs the onset probability as a numerical value.

[0256] As shown in Figure 17B, once the onset risk model is generated, the insurance assessment standard calculation module 4160 calculates the insurance assessment standard by using the output from the onset risk model to "evaluate the probability of an event causing insurance payment occurring."

[0257] Here, "insurance assessment criteria" refers to information held by an insurer (such as an insurance company) as criteria for assessing whether or not an applicant can enroll in an existing insurance product.

[0258] Although not particularly limited, for example, the following documents disclose methods for calculating insurance assessment standards. Document: JP 2021-196696 A Document: JP 2019-153179 A

[0259] In addition, if the data user server 4000 is operated by a company's health insurance association, or if data provided by the company's health insurance association can also be used, the medical examination data can be configured to use prescription data, as described above.

[0260] 18A and 18B are diagrams showing a second method for calculating the development risk model.

[0261] 18A and 18B show a method for generating an "onset risk model" for a plurality of specific subjects at a stage where longitudinal data from the health state to the onset of dementia has not necessarily been acquired. However, as explained in FIGS. 17A and 17B, it is of course possible to update the onset risk model by re-learning as longitudinal data from the health state to the onset of dementia is accumulated.

[0262] 18(b) and 18B, first, the degree of atrophy of a specific brain structure is calculated from brain tomographic image data based on "brain image big data," and data from this brain image big data that contains data on a specific subject up to the onset of dementia is used as training data. It is assumed that, through such machine learning or statistical analysis, a "brain image risk model" is generated that assesses the risk of developing dementia using the brain tomographic image data as an explanatory variable.

[0263] Again, the output of the "brain imaging risk model" may be a probability value of the risk of developing dementia after a certain number of years (or after a certain period of time) as the objective variable, or a discrete output of the level of risk evaluated at multiple stages. In the following, the "brain imaging risk model" will be described as one that takes the volume of a specified brain structure and the attributes of the subject (age, sex) as input and outputs risk in stages.

[0264] 16, 18(b), and 18B, when processing starts (S620) and data of a new member is registered in the data sharing service server 3000, the learning dataset generation module 4110 acquires health information such as medical checkup data and brain image data obtained from a brain checkup, in association with the shared temporary ID from the shared DB 3100, and stores the health information in brain image data 4310 and health data 4320, respectively. Note that here too, the health data is assumed to be associated with the user's attribute information (age, gender, etc.) (S622).

[0265] Furthermore, the stored brain image data may be the captured brain image data itself, data on the volume of a specified brain structure, or the degree of atrophy of a specified brain structure, and may be stored in the brain image data 4310 or brain image risk model analysis information data 4340 of the storage device 4300.

[0266] In addition, when a user who holds a shared temporary ID applies for subscription to an insurance company's services, the learning dataset generation module 4110 associates the notification data submitted upon subscription with the shared temporary ID and stores it in the notification data 4330 (S624).

[0267] Next, the important risk factor extraction module 4140 extracts important risk factors for the risk of developing the disease from the health data 4320, notification data 4330, and brain image risk analysis information 4340, in the same manner as in FIG. 17 (S626).

[0268] Here, risk factors may include the subject's attributes, disclosure data, health data, and the degree of atrophy of specific brain structures based on brain imaging data.

[0269] Furthermore, the risk of developing a disease from brain images output from the brain image risk model can be used as correct answer data and also as learning data.

[0270] The onset risk model generation module 4150 receives the selected feature values ​​as input and generates an onset risk model that outputs the onset risk in stages (S628). Here, multiple levels of values ​​are used as the correct answer data, so the onset risk model is expected to output one of these multiple levels, but depending on the output of the brain image risk model, it may also be a model that outputs the onset probability as a numerical value.

[0271] With this configuration, when generating an onset risk model, it is possible to generate the onset risk model before longitudinal data is accumulated for the period until a specific subject actually develops dementia, and then perform the calculation of the insurance assessment standard according to the same procedure as described in Figure 17.

[0272] As mentioned above, as longitudinal data is accumulated, from health status to the onset of dementia, the onset risk model can be updated through re-learning, thereby gradually improving the accuracy of the onset risk model.

[0273] With the above configuration, the member 100 or the insured person can periodically visualize their brain health status and receive high-quality information to maintain and improve it, thereby reducing the risk of developing dementia.

[0274] On the other hand, current insurance products such as dementia insurance still have many unknown aspects regarding the risk of developing dementia, which may result in premiums being set at higher prices for policyholders. However, this will also enable insurance companies to avoid the risk of unexpected increases in insurance payments due to a future increase in the number of dementia patients.

[0275] In contrast, the above configuration accumulates the results of analysis of brain images over time for each subscriber, obtained as part of the brain health visualization service, and when the subscriber ultimately claims insurance (i.e. develops dementia), that information is recorded as information to identify the time of onset.

[0276] This allows time-series data of brain image analysis over time to be accumulated for each shared temporary ID, with information on sorting dementia patients into those who have not. In addition, it is desirable to accumulate data related to health conditions such as health status, presence or absence of disease, exercise, diet, blood pressure, and sleep for each shared temporary ID. This accumulated time-series data is collectively referred to as "dementia-flagged brain health status analysis time-series data."

[0277] Based on this data, insurance companies can understand the objective dementia risk factors of their policyholders and develop innovative insurance products that can adjust premiums to appropriate levels or pay incentives to policyholders in good condition. With the above structure, it will be possible to present appropriate evidence when releasing variable-rate insurance, which will also be advantageous in obtaining approval from regulatory authorities.

[0278] In other words, by accumulating analyzed time-series data on brain health status with dementia flags as described above, insurance companies can build evidence for estimating future dementia risk based on brain health status, based on statistically significant data that stratifies the time-series patterns of dementia from those that do not.This evidence can be used to develop new insurance products that include variable premiums and incentives based on brain health status.

[0279] Furthermore, by working with the administrator of the service page operation server 1000, the insurance company can provide a solution for visualizing and improving the brain health status of the insured person in order to encourage the delay and prevention of dementia in the insured person, thereby reducing the amount of insurance payments made by the insurance company. [Embodiment 2]

[0280] In the second embodiment, a case will be described in which a fitness club uses data as an administrator of a data user server 4000 for a data sharing server 3100 as described with reference to FIGS. 11 and 12. FIG.

[0281] FIG. 19 is a conceptual diagram for explaining the configuration of the data use service according to the second embodiment.

[0282] FIG. 20 is a diagram for explaining the flow of data in the data utilization service.

[0283] Although not shown in FIG. 19, it is assumed that a medical examination for a brain checkup is performed by a medical institution 5000 and brain checkup information is stored in the shared DB 3100 as shown in FIG.

[0284] 19 and 20, a fitness club and an insurance company may coexist as data users, as described in embodiment 1. In this case, as described below, the insurance company may simultaneously provide services such as collaboration with the fitness club, and as described in embodiment 1, the insurance company may develop innovative insurance products that can adjust insurance premiums to appropriate levels based on information such as the amount of exercise of the insured person, or pay incentives to insured persons in good health.

[0285] Referring to Figures 19 and 20, the user 100 transmits the exercise performed at the fitness club 5000 to the fitness club server (data user server) 4000, for example, using a tablet terminal 5020 held by the instructor 5010, and the data is stored in the user database 4100.

[0286] The method of collecting information about the exercise of user 100 is not limited to the above method, and it is also possible to configure the system to automatically collect data from fitness machines and training machines used in fitness clubs (or sports gyms).

[0287] Although not particularly limited, a configuration for automatically collecting data is disclosed in the following document, for example: Document: JP 2019-32822 A

[0288] Meanwhile, at home, the user 100 transmits data such as daily blood pressure measurements taken with the sphygmomanometer 210 and daily weight and body fat percentage measurements taken with the scale / body composition meter 220 to the service page operation server 1000 via the smartphone 200. It is also possible to acquire data related to blood glucose levels using the non-invasive measurement method described above. While not particularly limited, data from the sphygmomanometer 210, the scale / body composition meter 220, and other blood glucose level measuring devices (not shown) can be transmitted to the smartphone 200 via near-field wireless communication or the like. Furthermore, the smartphone 200 can be used as a pedometer to transmit data on the amount of exercise, such as walking or jogging near the user's home, to the service page operation server 1000 via the smartphone 200. The service page operation server 1000 associates the received data with a shared temporary ID and stores it as "health-related information" in the storage device 1100. The amount of exercise performed during walking or jogging may be measured using a dedicated device (such as a pedometer or smartwatch) and transmitted to the service page operation server 1000 via the smartphone 200 .

[0289] Furthermore, at home, the user 100 takes an image of a meal using the camera of the smartphone 200 and transmits the image and the time of the meal to the service page operation server 1000. The service page operation server 1000 may be configured to calculate information on calorie intake and nutritional balance from the image using artificial intelligence technology, associate the information with the shared temporary ID, and store it in the storage device 1100 as health information together with the time of the meal.

[0290] The technology for acquiring information on calorie intake and nutritional balance from image data of meals is well known, as disclosed in the following documents: Document: JP 2021-18567 A Document: JP Patent No. 6598930 A

[0291] In the service page management server 1000, as in embodiment 1, brain image information from the imaging site 5000 is stored in the storage device 1100 in association with a shared temporary ID, as well as information regarding meals and, as will be described later, point information about the user 100.

[0292] The data sharing service server 3000 stores the brain image information, analysis reports, and clinical information from the imaging site 5000 in the storage device 3100, in association with the shared temporary ID, within the scope of consent from the user 100. Furthermore, the data sharing service server 3000 stores the health information about the user 100, transmitted via the service page operation server 1000, in the storage device 3100, in association with the shared temporary ID.

[0293] Here, health-related information includes health care information such as blood pressure, weight, and body composition, as well as data on dietary habits and daily exercise habits.

[0294] Furthermore, the data sharing service server 3000 may further classify the dietary status into a plurality of ranks based on the calorie intake, nutritional balance, and meal times, depending on the attributes of the user 100, such as age, and the degree of atrophy of the brain structure based on brain image information, and store these ranks. Similarly, the data sharing service server 3000 may classify the exercise status into a plurality of ranks.

[0295] Then, depending on such eating and exercise situations, the data sharing service server 3000 may select predetermined recommended meal contents and information regarding predetermined recommended daily exercise, associate them with the shared temporary ID, and store them in the storage device 3100.

[0296] The fitness club server 4000 associates the temporary subscriber ID assigned to the user when joining the fitness club with the shared temporary ID presented by the user 100 when joining the fitness club, and further associates with the shared temporary ID the health care information and daily exercise status data received from the shared DB 3100, as well as the activity status at the fitness club (the exercise menu being performed and its exercise load) and stores them in the user database 4100.

[0297] In this way, the operator of the service page operation server 1000 can introduce services linked to fitness clubs to the user 100.

[0298] By having user 100 exercise at a fitness club, it becomes possible to provide services that reduce "social isolation" and "lack of exercise," which are factors that increase the "risk of dementia" among middle-aged and elderly people, as part of the user's service menu.

[0299] Therefore, by utilizing the information stored in the shared database, the operator of the service page operation server 1000 and the manager of the fitness club will be able to provide the following services to reduce the "dementia risk" factors.

[0300] i) Providing users with regular brain scans and information on their brain health. ii) Providing recommended actions to reduce "dementia risk" factors. ・Exercise menus and exercise load at fitness clubs ・Amount of exercise in home life and daily life (number of steps per day) ・Meal menu recommendations

[0301] That is, although not limited to, the service page operation server 1000, the data sharing service server 3000, and the fitness club server 4000 each share and execute the following processes: 1) Service page operation server 1000

[0302] i) A "shared temporary ID" is issued to the member (user 100).

[0303] ii) When "image data of a meal" is received from a member, the "image" is sent to the data sharing service server. Note that the service page operation server 1000 may generate data that determines the "contents of the meal (calories, nutrition)" from the "image" and send it to the data sharing service server.

[0304] iii) Receives data on blood pressure, weight, body composition, and daily exercise from members and sends it to the data sharing service server. 2) Data sharing service server 3000

[0305] i) Collect brain imaging data and assess (rank) "dementia risk."

[0306] ii) Collection of healthcare information (blood pressure, weight, body composition) data via the service page operation server 1000.

[0307] iii) Collection of meal content data via the service page operation server 1000. Data is generated and saved by determining the "meal content (calories, nutrition)" from the "image."

[0308] iv) Collection of daily exercise data via the service page operation server 1000.

[0309] v) Select recommended diet and exercise programs by analyzing the following data: Input data: Dementia risk rank from brain images "Healthcare information" for each dementia risk rank "Dietary content data" for each dementia risk rank "Exercise data" for each dementia risk 3) Fitness club server 4000

[0310] i) Collecting data on each member's exercise menu and exercise load at fitness clubs.

[0311] ii) Provide the exercise menu and exercise load data to the data sharing service server.

[0312] iii) Recommended meals and exercise menus are received from the data sharing service server and provided to the member.

[0313] The following daily life data related to dementia risk factors will be collected from members and fitness clubs: a) Records of exercise volume at fitness clubs b) Records of exercise volume in daily life (pedometers on smartphones, etc.) c) Weight measurement data, body fat measurement data (body composition monitors) d) Dietary data (photographs of meals taken with a smartphone and uploaded)

[0314] FIG. 21 is a conceptual diagram for explaining the processing executed by the data sharing service server 3000.

[0315] That is, as described above, FIG. 21 shows the flow of processing in which the data sharing service server 3000 calculates recommended diet and exercise menus.

[0316] The following processing may be performed by statistical processing based on the collected data, or a model to perform the processing may be generated by supervised learning using artificial intelligence such as a neural network, with the target value (hereinafter referred to as the target weight) as the correct answer data.

[0317] In the following description, it is assumed that a trained model of the artificial intelligence has been generated as described above. In this process, the computing device of the data sharing service server 3000 generates a menu for "optimizing body weight" for the fitness service in accordance with a program stored in the memory.

[0318] First, the target weight calculation processing unit 4200 calculates the target weight using the dementia onset risk rank based on the brain image data and current health care information as input.

[0319] Here, the dementia onset risk rank is calculated by the analysis service providing server 2000, and can be referenced by the data sharing service server 3000 as data in the analysis report.

[0320] Next, the recommended diet and recommended exercise menu calculation processing unit 4210 receives the calculated target weight, current diet information, and current exercise information as input, and outputs a recommended diet menu and a recommended exercise menu.

[0321] For example, a recommended meal menu is output that matches the recommended daily calorie intake and nutrient intake based on the calories to be reduced and nutrients to be obtained according to a pre-set model. Although not particularly limited, for example, menu data to be referenced in advance may be created as a table corresponding to the calorie intake and nutrient intake.

[0322] Similarly, the recommended exercise menu is estimated based on the calories burned through exercise and the member's attributes (input information can include age, sex, and also, for example, the member's usual exercise menu). Again, the menu data to be referenced may be prepared in advance as a table.

[0323] 22A and 22B are diagrams showing examples of points given to members according to information collected via the fitness club or the service page operation server 1000. FIG.

[0324] As described above, such point information may be stored in storage device 1100 of service page operation server 1000, or may be stored in storage device 4100 by fitness club server 4000. Alternatively, service page operation server 1000 and fitness club server 4000 may jointly operate the point-granting service.

[0325] 22(a) and 22(A) show examples in which points are awarded according to the degree of achievement of a recommended meal menu or a recommended exercise menu. If the degree of achievement is less than a predetermined level (e.g., 70%), no points are awarded, but if the degree of achievement is above that level, the points awarded increase in stages.

[0326] 22(b) and 22B show examples of points awarded for providing data.

[0327] Every time health care information, exercise information, or diet information is provided within a predetermined period (for example, three months or one year), the information is updated and stored in the storage device 1100 in the service page operation server 1000 .

[0328] With this configuration, members are given points based on the content of the information provided, the amount of information, and behavioral changes (the degree to which the amount of exercise and the content of the activities conform to the recommended content), and can receive discounts on brain checkup fees and fitness club fees.

[0329] As a result, the operator of the data sharing service server 3000 can generate revenue by analyzing information from the collected data (using AI) and providing such analyzed data.

[0330] Moreover, with such a configuration, the identity of the collected data is guaranteed, and as more data is collected, the value of the data itself increases.

[0331] In the third embodiment, a case will be described in which the introducer of the counselor uses data as an administrator of the data user server 4000 for the data sharing server 3100 as described with reference to FIGS. 11 and 12. FIG.

[0332] FIG. 23 is a conceptual diagram for explaining the configuration of the data use service according to the third embodiment.

[0333] FIG. 24 is a diagram for explaining the flow of processing in the counselor introduction service.

[0334] Although not shown in FIG. 23, it is assumed that a medical examination for a brain checkup is performed by a medical institution 5000 and brain checkup information is stored in the shared DB 3100 as shown in FIG.

[0335] Also, although not shown in Figure 23, as explained in embodiment 1, data users may include fitness clubs and insurance companies in addition to the counselor's introducer.

[0336] In Figure 23, the same parts as in Figures 19 and 20 are given the same reference numerals and description thereof will not be repeated.

[0337] The storage device 4100 of the counselor introduction server 4000, which is a data user server, has information on counselors 6000.1, 6000.2, . . . registered in advance.

[0338] The counselor information is not particularly limited, but may include, for example, gender, age, available days and times, years of experience, and evaluations from past members.

[0339] Referring to FIG. 24, first, the counselor himself / herself registers the counselor's attributes via counselor introduction server 4000 (S700).

[0340] Next, the member 100 registers his / her desired conditions for the member's counselor via the service page operation server 1000 (S710).

[0341] Here, desired conditions can be set such as gender, age, desired interview date and time, years of experience, etc.

[0342] Next, the counselor introduction server 4000 adjusts the probability of successful matching based on the points given to each member, as described in the second embodiment (S720).

[0343] For example, the higher the points of a member, the higher the chance of matching, at a predetermined rate. As an example, when two or more people have the same requests, priority may be given to matching with members with higher points. However, the process of adjusting the chance of successful matching based on points is not a required process, and whether or not to perform it may be changed depending on, for example, the congestion of counseling requests.

[0344] Next, as explained in the second embodiment, the counselor introduction server 4000 adjusts the probability of successful matching based on the dementia risk rank obtained as an analysis result for each member (S730). Here, too, a predetermined ratio is set so that the higher the risk of a member, the higher the probability of matching. As an example, when multiple requests with the same conditions are made, priority may be given to matching with members with higher risk.

[0345] Members with high risk should receive counseling early, so the system is designed to change the matching probability based on risk.

[0346] Next, the counselor introduction server 4000 executes a process of matching members with counselors (S740).

[0347] Next, the counselor introduction server 4000 conducts a consultation (test consultation) between the member and the counselor without them meeting face to face, for example, through a voice-to-text conversion chat (S750).

[0348] Rather than meeting face-to-face right away, members can first chat via text (but also via voice input) before deciding whether to have a face-to-face consultation, making it easier for members to judge compatibility with a counselor, which cannot be determined based on desired conditions alone.

[0349] Next, the counselor introduction server 4000 receives an instruction from the member as to whether or not to proceed to face-to-face consultation (S760). If the member does not proceed to face-to-face consultation, the process returns to the matching process again (N in S760).

[0350] If the member proceeds to a face-to-face consultation (Y in S760), the counselor introduction server 4000 sets up a consultation between the counselor and the member via the web with audio and video or an actual face-to-face consultation (S770).

[0351] The process of the online consultation with audio and images may be carried out using another external service rather than the counselor introduction server 4000 itself. In this case, the counselor introduction server 4000 issues a URL for the consultation to the member.

[0352] As a counselor, when the consultation is conducted over the web with audio and video, it is possible to set the date and time of the interview with a relatively high degree of freedom.

[0353] The contents of the shared DB may be configured so that the counselor can also refer to the information on the member who is the subject of the consultation.

[0354] This structure allows members to consult with a counselor online about dementia-related issues, which can motivate members to continue making behavioral changes to prevent dementia.

[0355] Furthermore, if services such as those provided through collaboration with insurance companies and fitness clubs are simultaneously provided in addition to the counselor introducer services, as explained in embodiment 1, insurance companies can develop innovative insurance products that allow them to adjust insurance premiums to appropriate levels based on information such as the amount of exercise performed by policyholders, or to pay incentives to policyholders in good condition.Furthermore, fitness clubs can provide counseling services to their members.

[0356] The embodiments disclosed herein are merely examples of configurations for specifically implementing the present invention, and do not limit the technical scope of the present invention. The technical scope of the present invention is defined by the claims, not by the description of the embodiments, and is intended to include modifications within the literal scope of the claims and within the scope of equivalent meanings.

[0357] 1 Brain image data analysis device, 2a Database, 11 Image input unit, 12 Segmentation unit, 31 Brain structure volume / brain signal intensity calculation unit, 15 Shooting condition calibration unit, 16 Peer age ratio calculation unit, 17 Cerebrovascular disorder degree calculation unit, 18 Analysis report creation unit 19 Analysis report output unit, 1000 Service page operation server, 2000 Analysis service provision server, 3000 Data sharing service provision server, 4000 Data user server.

Claims

1. The system includes an analysis server that provides an analysis service that, based on data obtained by analyzing the degree of atrophy of a predetermined region of interest in the brain in three dimensions from first brain tomography image data of multiple subjects, which have been anonymized and captured by a tomography scanner, receives second brain tomography image data captured from the subject being evaluated and returns the analysis results regarding the atrophy of the predetermined region of interest. The analysis server manages the second brain tomography image data in association with non-personally identifiable provisional identification information. Furthermore, it includes a data sharing server that stores a shared database and can communicate data with the outside world. The aforementioned data sharing server is The second brain tomography image data of the person being evaluated, provided from the analysis server within the scope of the person being evaluated's permission, and the health information of the person being evaluated are stored in association with the provisional identification information. The system further comprises a user server that receives data from the data sharing server and utilizes the second brain tomography image data and the health information, When the person being evaluated subscribes to the administrator's proprietary service, the user server receives the provisional identification information held by the person being evaluated, maintains a proprietary database in which the provisional identification information and the data of the person being evaluated are independently associated, and provides the proprietary service to the person being evaluated by analyzing the data in the proprietary database. The aforementioned analysis server, A segmentation unit that performs segmentation, dividing the entire brain tomographic image of the second brain tomographic image data into individual structural units, An identification unit for identifying structural units related to the region of interest from among the divided structural units, A volume calculation unit that calculates the brain volume of each identified structural unit, A calibration unit for calibrating the brain volume based on the input brain tomography image, according to the imaging conditions of the tomography scanner device, A database containing age-related change data, including brain volume data of each structural unit at each age under various imaging conditions, obtained by imaging multiple subjects in advance, The system includes a same-age ratio calculation unit that calculates the same-age ratio of the structural units of the person being evaluated from the data of each structural unit at each age in the database, The calibration unit performs calibration of the brain volume based on the age-related change data in the database. A brain image analysis data utilization system in which, after calibration, the age-group ratio calculation unit calculates the age-group ratio of the brain volume of the structural units of the subject being evaluated to the number of subjects, based on the structural unit data for each age in the database, thereby analyzing the degree of atrophy of a predetermined region of interest in the brain.

2. The system further includes a service page server that performs the user registration process for the person being evaluated and returns the analysis results in response to an inquiry from the person being evaluated. The service page providing server issues the provisional identification information to the information processing device owned by the person being evaluated during user registration, and the provisional identification information is stored in the storage unit of the information processing device. The person being evaluated presents the provisional identification information to the imaging facility performing the imaging when acquiring the second brain tomography image data. The imaging device transmits the second brain tomography image data to the analysis server in association with the provisional identification information. The brain image analysis data utilization system according to claim 1, wherein the analysis server returns the analysis result for the received second brain tomography image data in association with the provisional identification information.

3. The brain image analysis data utilization system according to claim 2, wherein when the person being evaluated subscribes to the proprietary service operated by the administrator, the user server receives the provisional identification information presented by the person being evaluated via the information processing device.

4. The brain image analysis data utilization system according to claim 3, wherein the information processing device further includes a display unit, and the provisional identification information is displayed on the display unit as a two-dimensional code.

5. The brain image analysis data utilization system according to any one of claims 1 to 4, wherein the user server calculates the assessment criteria for when the person to be evaluated applies for insurance by evaluating the risk of developing dementia of the person to be evaluated based on the second brain tomography image data and the health information.

6. The first brain tomography image data of the aforementioned multiple subjects is time-series data for each subject and includes clinical information related to dementia associated with each time-series data, The analysis server includes a memory device that stores a brain image risk model that outputs the risk of developing dementia at a second time point after the first time point, based on the degree of atrophy of the region of interest at a first time point, using machine learning or statistical analysis, based on the first brain tomography image data of the plurality of subjects, The brain image analysis data utilization system according to claim 1, which takes the received second brain tomography image data as input and returns the disease risk calculated by the brain image risk model to the user server.

7. The brain image analysis data utilization system according to claim 6, wherein the user server generates a disease risk model that predicts the disease risk, using the disease risk output from the brain image risk model and health check data as training data, and generates insurance assessment criteria.

8. The brain image analysis data utilization system according to claim 2, wherein the service page providing server updates and manages points for each person being evaluated according to the frequency and amount of health information it receives from the person being evaluated, and provides a discount service for imaging processing at the imaging facility according to the points.

9. The administrator of the aforementioned user server is the provider of the sports facility service, The person being evaluated who has subscribed to the proprietary service operated by the aforementioned administrator shall provide healthcare information and dietary information obtained outside the exercise facility to the data sharing server. The brain image analysis data utilization system according to claim 3, wherein the data sharing server calculates a recommended exercise menu and a recommended meal menu for the person being evaluated based on the analysis results, exercise volume information, healthcare information and dietary information for the person being evaluated, and provides them to the user server.

10. The administrator of the user server is the provider of counseling services to the person being evaluated, The person being evaluated who has subscribed to the proprietary service operated by the aforementioned administrator registers the desired attributes of the counselor via the user server. The user server performs a matching process based on the desired attributes and the counselor's attributes. i) After conducting a test consultation in which the matched counselor and the person to be evaluated do not meet in person, ii) If the person being evaluated wishes, the brain image analysis data utilization system according to claim 3, further comprising providing face-to-face counseling between the matched counselor and the person being evaluated.

11. The system further comprises a structural unit selection unit that, based on the brain volume data stored in the database, selects structural units related to atrophy of the region of interest, where age-related changes are greater than other structural units and changes due to imaging conditions are smaller than other structural units. The brain image analysis data utilization system according to claim 1, wherein the identification unit targets the structural unit selected by the structural unit selection unit for identification.

12. The brain image analysis data utilization system according to claim 11, wherein the predetermined brain region of interest is a structure known to change in dementias of the limbic system, including the hippocampus, or the medial temporal lobe.

13. The brain image analysis data utilization system according to claim 12, wherein the structural unit selected by the structural unit selection unit is a structure selected based on the degree of age-related changes and robustness to the influence of imaging conditions, and includes a ventricle.