Information Processing System

Abstract

To provide an information processing system capable of estimating a subject's brain state using a simple inspection and estimating a state such as ill conditions of mind and body from the estimated brain state.SOLUTION: An information processing system comprises: a storage unit for storing inspection-brain state related information relating first inspection results, results of actions of a plurality of subjects to a predetermined inspection, with brain states of the subjects and second inspection results, results of states of mind and body of the plurality of subjects to the predetermined inspection; an input unit for receiving a first inspection result, a result of an action of a subject to the predetermined inspection; a control unit that estimates a brain state from the first inspection result on the basis of the inspection-brain state related information and estimates a state other than the brain state on the basis of the estimated brain state and the second inspection results; and an output unit for outputting the estimated state other than the brain state.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an information processing system for estimating the state of a subject.

Background Art

[0002] In the field of rehabilitation (hereinafter referred to as "rehab") for patients with cerebrovascular disorders (or stroke) such as cerebral infarction, improving the motivation of patients has been an issue. Therefore, it is desirable to provide feedback to patients on the rehab effect.

[0003] In addition, improving the efficiency of rehab has also been an issue. Therefore, it is required to provide timely feedback on the rehab effect to medical staff such as doctors, nurses, occupational therapists, physical therapists, and speech therapists (hereinafter referred to as "medical staff"), and reflect it in the rehab program.

[0004] Here, as background art in this technical field, for example, there is a report that the results of some simple tests (Clinical Assessment for Attention (CAT)) are related to the cerebral infarction site. For example, there is a report in Non-Patent Document 1.

Prior Art Documents

Non-Patent Documents

[0005]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] In order to provide early feedback on the rehabilitation effect to patients, medical staff, etc., it is an issue to visualize the brain state and improve the efficiency of rehabilitation by optimizing the rehabilitation program.

[0007] In order to provide feedback on the rehabilitation effect to patients and medical staff, for example, it would be good if the details of the patient's brain state could be grasped by a simple examination. For example, it would be good if the details of the brain state could be grasped by a simple examination without performing examinations such as MRI (Magnetic Resonance Imaging) or CT (Computed Tomography). Also, for example, it would be good if the brain state could be grasped in more detail than the information grasped only from the results of simple tests (such as CAT).

[0008] Furthermore, in addition to brain state estimation technology, it is important to estimate the state of discomfort or the like related to the physical and mental health of the subject from the estimated brain state, leading to early detection and early response to discomfort.

[0009] Therefore, an information processing system and an information processing method are provided that can estimate the brain state of a subject even with a simple examination and further estimate the state of physical and mental discomfort or the like from the estimated brain state.

Means for Solving the Problems

[0010] According to a first aspect of the present invention for solving the above problems, the following information processing system is provided. This information processing system includes a storage unit, an input unit, a control unit, and an output unit. The storage unit stores inspection-brain state related information that associates a first inspection result, which is the result of the actions of a plurality of subjects for a predetermined inspection, with the brain state of the subject, and a second inspection result, which is the result of the physical and mental states of a plurality of subjects for a predetermined inspection. The input unit receives a first inspection result, which is the result of the actions of a subject for a predetermined inspection. The control unit estimates the brain state from the first inspection result based on the inspection-brain state related information, and estimates a state other than the brain state based on the estimated brain state and the second inspection result. The output unit outputs the state other than the estimated brain state.

[0011] Further, according to a second aspect of the present invention for solving the above problems, the following information processing method is provided. This information processing method is a method for estimating the state of a subject by processing using an electronic computer. This information processing method includes: (1) obtaining inspection-brain state related information that associates a first inspection result, which is the result of the actions of a plurality of subjects for a predetermined inspection, with the brain state of the subject, and a second inspection result, which is the result of the physical and mental states of a plurality of subjects for a predetermined inspection; (2) obtaining a first inspection result, which is the result of the actions of a subject for a predetermined inspection; and (3) estimating the brain state from the first inspection result based on the inspection-brain state related information, and estimating a state other than the brain state based on the estimated brain state and the second inspection result.

Advantages of the Invention

[0012] According to the present invention, it is possible to provide an information processing system that can estimate a state (such as discomfort) that cannot be known only from inspection data related to the brain estimated from the brain state of a subject, even if the inspection is simple. This may lead to the early detection of discomfort.

Brief Description of the Drawings

[0013]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Embodiments for Carrying Out the Invention

[0014] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In different drawings, constituent blocks and constituent elements denoted by the same reference numerals indicate the same objects.

[0015] FIG. 1 shows a configuration example of an information processing system 1 according to an embodiment. The information processing system 1 has a configuration including, for example, a storage unit, an input unit, a control unit, and an output unit. Further, the information processing system 1 may further include a display unit.

[0016] In the example of FIG. 1, the storage unit has an inspection result holding unit 12, an inspection-brain state related information holding unit 11, an estimated brain state holding unit 14, and a brain state-state / intervention related information holding unit 15. Further, the input unit corresponds to the input unit 22, the control unit corresponds to the analysis means (13, 16), and the output unit corresponds to the brain state output unit, respectively.

[0017] The storage unit stores inspection-brain state related information that associates, for a plurality of subjects, an inspection result (a first inspection result) that is the result of the behavior of the subject for a predetermined inspection with the brain state of the subject. Details of this inspection-brain state related information will be described later with reference to FIG. 3. Here, an example of the predetermined inspection (an example of the first inspection) is an inspection by CAT (Clinical Attention Assessment, a standard attention inspection method), an inspection by biometric measurement such as measurement by a finger tapping device, etc. As an example of the predetermined inspection, it may include an inspection related to at least one of the functions of the brain related to movement, cognition, and attention.

[0018] In addition, the memory unit stores brain state-state correlation information that associates, for a plurality of subjects, a predetermined brain state with a test result (second test result) that is a result of a state other than the brain state of the plurality of subjects for a predetermined test. Details of this brain state-state correlation information will be described later with reference to FIG. 15.

[0019] Here, the predetermined test (the test corresponding to the second test) can be a test related to the physical and mental state (i.e., the mind and body) of the subject. The physical and mental state of the subject is evaluated, for example, by scores such as depression tendency, lack of motivation tendency, etc. using a standardized questionnaire, the result of a medical interview by a doctor, the degree of improvement in the score of ADL (Activity of Daily Living) in rehabilitation, and the like. Among these, it may include at least one or more scores or tests related to the functions of the brain.

[0020] The input unit 22 receives a test result (i.e., the first test result) that is a result of the behavior of the subject for a predetermined test. An example of the subject is a patient. For example, a patient undergoing rehabilitation is assumed to take a predetermined test for the purpose of providing feedback on the rehabilitation effect to the patient himself / herself or medical staff. Note that the subject may be a subject whose test results are included in test-brain state correlation information, brain state-state correlation information, etc., or a subject whose test results are not included.

[0021] The control unit estimates the brain state of the subject from the test result of the subject received from the input unit 22 based on the test-brain state correlation information. Here, the control unit may estimate the brain lesion site as an example of the brain state.

[0022] The output unit outputs the brain state estimated by the control unit. Here, the output unit may output an image indicating the brain lesion site (estimated brain lesion site) in the brain. For example, the output unit can visualize the brain state by outputting an image indicating the brain state to a display device (display or monitor) that operates as a display unit, and can provide feedback on the rehabilitation effect to the patient and medical staff. As a result, it is also possible to reflect the feedback result in the rehabilitation program (including changes to the rehabilitation program) and improve the patient's motivation. In this way, rehabilitation is made more efficient by providing effective and rapid information and feedback to the patient, medical staff, etc.

[0023] In the case of cerebrovascular disorders (or strokes) such as cerebral infarction, there are functions that are lost because a certain area of the brain becomes a lesion site (including damaged or missing sites) due to cerebral infarction or the like. Also, the functions lost vary depending on the brain lesion site. From this, regarding the test results, which are the results of the subject's actions in a predetermined test, there is a high possibility that the test results of subjects with common brain lesion sites will show the same or similar tendencies, and the test results of subjects with different brain lesion sites will show different tendencies. Focusing on this, in this embodiment, based on the test-brain state related information, the brain state (for example, the brain lesion site, etc.) of the subject is estimated from the test results of a predetermined test. This predetermined test may be a simple test such as measurement by a finger tapping device, and according to the present invention, it is possible to provide an information processing system that can estimate the brain state of the subject even with a simple test. Also, inefficient rehabilitation that only repeats rehabilitation training for the functions lost due to the brain lesion site is improved. By improving the efficiency of rehabilitation, a shortening of the treatment period by rehabilitation can be expected. Also, the workload of medical staff (medical practitioners such as doctors, occupational therapists, physical therapists, speech therapists, etc.) associated with rehabilitation can be reduced. Furthermore, it is possible to prevent overlooking the progression and recurrence of the disease state.

[0024] In the example of FIG. 1, the inspection result holding unit 12 holds inspection results obtained from a plurality of biological measurements such as CAT (Clinical Attention Assessment, a standard attention inspection method) and measurement by a finger tapping device. The analysis means 13 estimates a severe site in the probability map of the brain lesion site from the inspection results held by the inspection result holding unit 12 and the inspection-brain state related information held by the inspection-brain state related information holding unit 11. The estimated brain state holding unit 14 holds three-dimensional data of the brain image of the estimation result. Further, the three-dimensional data of the brain image of the estimation result may be output to the subject display unit 17 described later. The inspection-brain state related information may be held by the inspection-brain state related information holding unit 11 as a database. Here, the brain defect site, the cerebral infarction site, the brain atrophy site, etc. are collectively referred to as the brain lesion site. The analysis means 13 estimates the brain state by inputting one or more parameters obtained from one or more inspection results of the inspection result holding unit 12 into the inspection-brain state related information of the inspection-brain state related information holding unit 11.

[0025] FIG. 2 is a diagram showing an example of the hardware configuration of the information processing system 1. The information processing system 1 is composed of, for example, a storage device, an input device 25, an arithmetic device, etc. The storage device operates as a storage unit, the input device 25 operates as an input unit, and the arithmetic device operates as, for example, a control unit or an output unit. Note that the information processing system 1 may further include a display device 26 that operates as a display unit and a communication device that operates as a communication unit for communicating with an external device. Note that the arithmetic device may be composed of a processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), or may include a dedicated circuit for performing a specific process. Here, the dedicated circuit is, for example, an FPGA (Field Programmable Gate Array), an ASIC (Application Specific Integrated Circuit), a CPLD (Complex Programmable Logic Device), etc.

[0026] In this embodiment, an example in which the memory device is the memory 21 and the arithmetic device is the CPU 23 will be described. The memory 21 constitutes the inspection result holding unit 12, the inspection-brain state related information holding unit 11, and the estimated brain state holding unit 14, and the CPU 23 constitutes the analysis means (13, 16) and the brain state output unit. The brain state output from the CPU 23 may be displayed on the subject state display unit 17 composed of a display or a monitor. In the embodiment, the CPU 23 operates as a control unit and an output unit, and realizes functions such as the analysis means (13, 16) and the brain state output unit by executing the program stored in the memory 21. Incidentally, in the embodiment, the same applies to each function such as the analysis means (13, 16), the brain state output unit, the feature amount extraction unit 62, and the inspection-brain state related information learning unit 72 described as being configured by the CPU 23.

[0027] The input device 25 is, for example, an interface that receives data from an external device, a mouse, a keyboard, etc., and operates as the input unit 22. Note that the input device 25 may be an input / output interface (input / output IF). Further, the display device 26 is a display, a monitor, etc., and operates as the subject state display unit 17. Note that in front of the subject state display unit 17, there may be an input / output IF, a communication path, etc. related to the input / output of the subject state display unit 17.

[0028] The hardware configuration of the information processing system 1 may be composed of one or more computers (electronic computers). Note that each component of the hardware of the information processing system 1 described above may be singular or plural.

[0029] FIG. 3 is a diagram showing the contents of the examination-brain state related information database 50 as an example of the examination-brain state related information held by the examination-brain state related information holding unit 11, and is related information between an examination result (result of an examination or biological measurement, etc.) and a brain lesion position. The examination-brain state related information is information in which, for one or a plurality of subjects, an examination result (first examination result), which is the result of the subject's behavior in response to a predetermined examination, and the subject's brain state are associated. In the example of the examination-brain state related information in FIG. 3, identification information (subject number) regarding the subject (examined person), an examination result (examination output), which is the result of the subject's behavior in response to a predetermined examination, and the subject's brain state (lesion position) are associated. Here, examples of the "subject's brain state" in the examination-brain state related information include, for example, a brain state diagnosed by a doctor using the results of brain imaging examinations such as MRI (Magnetic Resonance Imaging) and CT (Computed Tomography). Also, examples of the "subject's brain state" include brain images, brain structures obtained from brain images, etc., and brain lesion information. Also, examples of the "subject's brain state" may include information estimated about the brain state based on brain images and various examination results, etc.

[0030] If the examination result is a finger tapping examination, for example, the information is such as the total movement distance, left-right balance, standard deviation of the contact time, standard deviation of the tap interval, standard deviation of the phase difference, etc. If the examination result is a CAT examination, for example, there are Digit span forward, Digit span backward, Visual cancellation, position stroop, etc. The examination result may also be, among other things, the score of MMSE (mini-mental state examination) or FIM (Functional Independence Measure). In the examination-brain state related information holding unit 11, those score information and feature quantity information, and brain structures, brain lesion information, etc. obtained by brain imaging examinations such as MRI (Magnetic Resonance Imaging) and CT (Computed Tomography) are stored. Thus, the examination-brain state related information holding unit 11 holds related information between various examination results and the brain state.

[0031] The actual location of the brain lesion may not necessarily be in one place and may not be represented by a single region name. The examination-brain state-related information may include the coordinate information of the cerebral infarction and the distribution information of the coordinate information. In the examination-brain state-related information holding unit 11, for example, information of a large number of subjects may be stored in advance as a database, and it may be configured to be able to search for the lesion location of the subject corresponding to the score information of a certain examination. In the embodiment, it is assumed that the main subject is a human, and the subject may also be referred to as the examinee.

[0032] As a specific method for calculating the total movement distance, for example, the distance between the thumb and the index finger may be acquired in time series, the sum of twice the maximum amplitude in each cycle may be calculated to obtain the total movement distance of the right hand and the left hand, and the sum of them may be calculated. That is, it is calculated from the physical position of the finger. The left-right balance may be calculated by taking the ratio of the total movement distances of the left and right. The contact time of the finger may be defined by the distance between the thumb and the index finger in the state of contact and separation, and the contact time may be used. The finger tap interval may be calculated by the interval between the contact start times of both fingers. The standard deviation of the phase difference may be obtained, for example, by applying the Hilbert transform or the like to the time series changes of the left and right finger tapping respectively to obtain the phase, and calculating the time series change of the phase difference between the left and right. The standard deviation of the time series change of the left-right phase difference may be calculated.

[0033] FIG. 4 is a flowchart showing the process of outputting the lesion site of the brain by the analysis means 13. This flowchart may be executed at a predetermined timing, such as the timing when the input unit receives an execution request from the administrator or the management device, or the timing when the input unit 22 receives the examination result which is the result of the action of the subject for a predetermined examination. First, the analysis means 13 reads out the examination result from the examination result holding unit 12 (step S401).

[0034] Next, the analysis means 13 queries the inspection-brain state related information held by the inspection-brain state related information holding unit 11, and creates a three-dimensional probability map of the lesion site of the brain (step 402). FIG. 5 shows an example of a three-dimensional probability map of the lesion site of the brain. In the present embodiment, a three-dimensional probability map is described as an example, but the map format may be other formats as long as it is a map that shows the state of the brain. On the three-dimensional brain model (surface) 36, regions 38 with a high probability and regions 39 with a low probability of being the lesion site of the brain are displayed in different ways (for example, hatching pattern, color difference, shading difference). Here, the three-dimensional distributions of two types of probabilities are shown, but continuous probability values may be mapped with color shading or color, etc., and not limited to the surface of the brain model, and regions may also be mapped three-dimensionally in the three-dimensional brain model (inside) 37.

[0035] When creating a three-dimensional probability map, for example, the brain lesion site corresponding to a certain score is referred to from a database (examination-brain state-related information), and the actual brain lesion site of each subject is mapped (back-projected). This is done for all the subjects included in the database, and when the brain lesion sites of some or all of the subjects corresponding to a certain score are overlapped on a standard brain (e.g., the MNI (Montreal Neurological Institute) coordinate system), the frequency information is calculated, and the frequency information may be mapped onto a three-dimensional brain model (standard brain). Further, the information mapped onto the standard brain may be mapped onto the brain images previously acquired by MRI or CT for each subject. Thereby, a three-dimensional probability map corresponding to the frequency information of the brain lesion sites can be displayed, such as displaying the brain lesion sites (brain lesion sites observed with high frequency) that are highly likely to be common to multiple subjects corresponding to a certain score as a high-probability region 38 in the three-dimensional brain model. Here, the subject corresponding to a certain score does not necessarily have to be limited to the subject with the same score, and may be a subject whose scores are in the same range (predetermined range) or a subject whose score characteristics are in the same tendency, and the subject corresponding to the score as a predetermined criterion is selected. Also, it is expected that the accuracy of the probability map will increase as the number of subjects included in the database increases. Further, by displaying the probability map on the brain image of each subject, more explicit feedback regarding each subject becomes possible. Here, the brain image of each subject may be the one obtained by converting the standard brain structure according to the brain structure information and coordinate system of each subject.

[0036] Next, the analysis means 13 determines whether a predetermined test result has been read (step S403). If step S403 is NO, the process moves to step S401. Thereby, when there are a plurality of test results of the target subject, such as test results of finger tapping or CAT tests, as the predetermined test results, a three-dimensional probability map is created for each test.

[0037] If the answer in step S403 is YES, the three-dimensional probability maps of multiple lesion sites are superimposed by weighted addition to identify the critical regions of the brain (step S404). The display method may be the method shown in FIG. 5 (display example of three-dimensional probability map), and the critical regions may be emphasized and displayed. Here, the critical regions of the brain indicate high-probability brain lesion sites calculated using one or more lesion probability maps corresponding to each test result.

[0038] Next, the brain state output unit creates and outputs three-dimensional data indicating the critical regions of the brain (step S405). Note that the brain state output unit may output the three-dimensional probability map in S402. Also, in this flowchart, the steps of S403, S404, and S405 do not necessarily have to be executed. Thus, in this flowchart, some steps do not have to be executed, or additional steps may be executed.

[0039] FIG. 6 shows a diagram simply illustrating the data flow when creating the brain lesion and pre-residual function probability maps. The input unit 22 receives the scores 95 and intervention information, which are the results of the biological measurement 92, the results of the examination 93, and the results of the intervention 94, and records them in the examination result holding unit 12. Here, the biological measurement 92 is, for example, finger tapping, and the examination 93 is, for example, a CAT scan or the like. The score 95 is information such as a level representing the degree of intervention when performing a medical intervention or treatment.

[0040] Also, the examination-brain state related information may be information in which at least a part or all of the results of the biological measurement 92 (for example, biological signals), the results of the examination 93, the scores 95 and intervention information, etc., and the brain states of each subject are associated with each other for, for example, one or more subjects.

[0041] The analysis means 13 searches the database (DB) held in the examination-brain state related information holding unit 11 using the results of the biological measurement 92, the results of the examination 93, the scores 95 or level information, etc. as inputs for the target subject, and estimates the brain lesion map 41 obtained from each biological information (examination result).

[0042] As an example of the brain lesion map 41, it is a three-dimensional probability map of the brain lesion site as shown in FIG. 5. An example of the method for creating the brain lesion map 41 is the same as the method described in FIG. 4. For example, in the examination-brain state related information, one or more subjects having test results (results of biological measurement 92, results of test 93, scores 95, intervention information, etc.) corresponding to the test results of the subject to be examined and corresponding test results (the same or equivalent test results, or test results with the same or similar tendencies) are selected, and the brain lesion positions of the selected one or more subjects are overlaid and mapped on a standard brain (for example, the MNI (Montreal Neurological Institute) coordinate system), whereby the brain lesion map 41 may be created.

[0043] Also, the brain activity map 42 may be created, for example, by estimating, from the results of biological measurement 92 related to brain function (such as left hand movement) and intervention information among the brain lesion sites estimated in the subject to be examined, the brain regions where brain activity is expected and the brain regions where it is expected that the intervention will affect and cause activity, from the brain region-brain function database 51 and the intervention-brain region database 54, and mapping them to a brain model.

[0044] This brain activity map 42 may not be the one actually measured by brain image measurement of brain activity, but may be estimated from a database obtained from literature, biological measurement results, etc. In the brain activity map 42, when the brain regions estimated to be expected to have activity are displayed simultaneously with the estimated brain lesions, in this embodiment, they may be called the regions of preliminary and residual functions.

[0045] As the databases stored in the examination-brain state related information holding unit 11, it includes a probability map of brain lesions according to the levels of clinical examinations and biological signals (examination / biological signal-brain lesion database 50), a brain region-brain function database 51 that stores the activity regions related to the brain functions related to biological information, an intervention-brain region database 54 that stores the brain regions that are expected to be affected by the intervention and to have activity, and a brain function-rehabilitation database 52.

[0046] The analysis means 13 can create a plurality of brain lesion maps 41 and brain activity maps 42 using this information, and further create a fusion map 43 showing the brain lesion and the sites of reserve and residual functions by fusing these maps. The brain state output unit can output to the subject state display unit 17 or the like some or all of the lesion map 41, the brain activity map 42, and the fusion map 43. Hereinafter, maps showing brain states such as the lesion map 41, the brain activity map 42, and the fusion map 43 may be referred to as brain maps 34. The brain state output by the brain state output unit may include information on the brain lesion site and the sites of reserve and residual functions of the brain estimated by the analysis means 13. Thereby, an information processing system can be provided that can easily grasp the brain state by visualizing the brain lesion site and the reserve and residual functions.

[0047] Fig. 7 shows an example of the fusion map 43 showing the brain lesion and the sites of reserve and residual functions. For example, in the case of an area estimated to be a lesion and where brain activity is expected from the biological signal or the score of the intervention, it is displayed as the reserve and residual function site 45, and other places are displayed as the lesion site 44.

[0048] By this display of the reserve and residual functions, there is an effect that it becomes possible to know about the possibility that the alternative site is performing the function that should be performed by the actual lesion site. Also, when this estimation is performed over time, by displaying the lesion site and the reserve and residual function site over time, it becomes possible to visualize which site has changed due to rehabilitation or treatment, and to provide accurate and immediate feedback to medical staff and patients. There are effects such as being able to efficiently optimize the rehabilitation program.

[0049] In addition to outputting the fusion map 43 showing the brain lesion and the sites of reserve and residual functions, the brain state output unit may output information on brain functions related to the lesion site and the recommended rehabilitation plan as a report.

[0050] The brain state output unit outputs information such as left hand movement, language function, pain, inhibitory function, and attention function as information on brain functions related to the lesion site, using the brain region-brain function database 51 and the like. The brain state output unit outputs information such as daily plans, monthly plans, self-training plans, and recommended treatments as recommended rehabilitation plans for training a predetermined brain function (for example, the function carried out by the site of the patient's preliminary and residual functions), using the brain function-rehabilitation database 52 and the like. At this time, the brain state output unit may output frequency information such as actual rehabilitation records included in the brain function-rehabilitation database 52.

[0051] FIG. 8 shows an example of information included in the intervention-brain region database (hereinafter sometimes referred to as the brain region database) 54, the brain region-brain function database (hereinafter sometimes referred to as the brain function database) 51, and the brain function-rehabilitation database (hereinafter sometimes referred to as the rehabilitation database) 52.

[0052] The brain region database 54 includes, for example, information associating intervention information with the brain regions affected by the intervention. In the example of the brain region database 54 in FIG. 8, information such as the motor area as the activity area expected for the intervention (walking training) and the language area as the activity area expected for the intervention (language training) is held. The brain region database 54 and the brain function database 51 are constructed from various literature information, biological measurements, and the like.

[0053] The brain function database 51 holds information related to the types of brain functions corresponding to predetermined brain regions. In the example of the brain function database 51 in FIG. 8, for the right motor area, corresponding functions include left hand movement and the like. The examination-brain state related information holding unit 11, which is a storage unit, holds the brain function database 51 that associates brain regions with brain functions.

[0054] The rehabilitation database 52 stores examples of specific rehabilitation programs (e.g., language training, walking training) for training a predetermined brain function (e.g., speech function, left motor function). The rehabilitation database 52 is created from past experiences, literature, databases, etc. By maintaining such a database in abundance or making it available, the creation of the fusion map 43 of the brain lesion and the reserve / remaining function shown in FIG. 7 can be made more accurate, and it becomes possible to appropriately propose a rehabilitation program based on this. The examination-brain state-related information storage unit 11, which is a storage unit, holds the rehabilitation database 52 that associates a brain function with a rehabilitation program for training the brain region related to the brain function.

[0055] Through this process, it is possible to easily know the brain state without measuring the brain, and visualize the effects of treatments such as rehabilitation and medication. As a result, there is an effect that leads to the improvement of rehabilitation and treatment methods, and the motivation of patients and their families.

[0056] Also, if a database including the relationship between the effects of treatments such as rehabilitation and medication and rehabilitation programs, etc. is held, an optimal rehabilitation program can be generated according to the brain state, and as a result, there is an effect that can reduce the workload of medical staff such as doctors, occupational therapists, physical therapists, speech therapists, etc. involved in rehabilitation.

[0057] Furthermore, even after treatment by rehabilitation, medication, etc., by visualizing the brain state according to the present invention at regular intervals, there is an effect that it is possible to prevent overlooking progression and recurrence.

[0058] Next, FIG. 9 shows a diagram simply illustrating the data flow of the embodiment. The analysis means 13 compares the inspection results such as CAT stored in the inspection result holding unit 12 with the inspection-brain state related information held in the inspection-brain state related information holding unit 11, and estimates the brain lesion site (for example, a defective site of the brain, a preliminary / remaining function site, etc.) using, for example, pattern matching or machine learning inference methods, and constructs a brain map 34 (see FIG. 14). The brain state output unit can output, as an example of the brain map 34, three-dimensional coordinate data of the brain indicating the brain lesion site by the back-projected brain image, and control it to be displayed on the subject state display unit 17.

[0059] When the defective site of the brain and the preliminary / remaining function are grasped, the analysis means 13 refers to the brain function database 51 shown in FIG. 8, and specifies the brain function related to the estimated brain lesion site (for example, a defective site of the brain, a preliminary / remaining function site, etc.). The brain state output unit outputs a rehabilitation program associated with the specified brain function by referring to the rehabilitation database 52 from the specified brain function.

[0060] The brain state output unit can display the defective site of the brain and the preliminary / remaining function site estimated by a simple inspection by the analysis means 13 and the information of the specified rehabilitation program on the subject state display unit 17 simultaneously or separately.

[0061] The analysis means 13 estimates a brain region where the activity of the brain function has decreased or an abnormal region of the brain from the inspection results of the subject based on the inspection-brain state related information held in the inspection-brain state related information holding unit 11 and the brain function database 51. Further, the brain state output unit controls to display the estimated brain region where the activity of the brain function has decreased or the brain region with an abnormality in the brain function on the brain model.

[0062] The input unit 22 updates the database by putting the rehabilitation result into the rehabilitation database 52 as an index such as FIM (Functional Independence Measure) which is an index of daily operation (ADL: ACTIVITY Of Daily Living).

[0063] The brain state output unit can display the estimated brain defect site and the rehabilitation program on the subject state display unit 17 through a simple examination. By periodically having the subject perform the rehabilitation program and using the information processing system 1 to check the temporal changes in the brain map, the effectiveness of the rehabilitation program can be confirmed.

[0064] The analysis means 13 associates the examination result including the time when the subject was examined and the brain lesion site estimated from the examination result for the target subject, and records it in the examination result holding unit 12. The examination result holding unit 12 may hold not only the brain lesion site estimated in the latest examination but also the brain lesion sites estimated in past examinations for the target subject.

[0065] The analysis means 13 estimates each brain state based on the examination-brain state related information from the examination results of a predetermined examination at different timings performed on the target subject. The brain state output unit outputs each estimated brain state or the temporal change of the brain state. Here, the predetermined examination at different timings may be, for example, an examination performed on the target subject after a certain rehabilitation program is implemented and an examination performed before the implementation of the rehabilitation program. Thereby, the temporal change of the brain state before and after the implementation of the rehabilitation program can be visualized by the display (output) to the subject state display unit 17, and the patient and medical staff can confirm the effectiveness of the rehabilitation program. The brain state output unit may output the information of the rehabilitation program performed during the aforementioned different timings together with each estimated brain state or the temporal change of the brain state.

[0066] In addition, the brain state output unit has a function of outputting a predetermined examination (such as a cognitive test) necessary for estimating the brain state. For example, information on a predetermined examination necessary for estimating the brain state is stored in advance in the examination-brain state related information holding unit 11, and the brain state output unit may output information on a necessary predetermined examination such as a CAT for confirming the rehabilitation effect to the subject state display unit 17 or the like by referring to this information. The brain function database 51 may be one available on the web or one such as a table showing the relationship between coordinates and keywords.

[0067] FIG. 10 is a diagram showing an information processing system when a biological data acquisition unit 61 and a feature amount extraction unit 62 are added to the information processing system shown in FIG. 1. The feature amount extraction unit 62 is configured by the CPU 23, and the biological data acquisition unit 61 is configured by the input unit 22.

[0068] Biological data is acquired by the biological data acquisition unit 61, a feature amount is extracted from the biological data by the feature amount extraction unit 62, and the examination result holding unit 12 holds the biological data and the feature amount as examination results. Here, the feature amount includes, for example, the moving distance, the energy of finger movement, the contact time of the finger, the interval of finger tapping, the phase of finger tapping, etc. when using a finger tapping device, and includes the speed of finger tapping, the variation in the timing of finger tapping, the variation in the distance of finger tapping, etc. when using an image of finger tapping.

[0069] In the analysis means 13, using the examination results, the brain state such as the brain lesion site is estimated from the examination-brain state related information held by the examination-brain state related information holding unit 11, and the brain state output unit outputs the estimation result. Then, the estimated brain state holding unit 14 holds the estimation result of the brain state.

[0070] FIG. 11 is a diagram showing a configuration example of the information processing system 1 including the examination-brain state related information learning unit 72. The information processing system shown in FIG. 1 further includes an examination-brain state related information database holding unit 71 that holds a database of information associating examination results with brain states, and an examination-brain state related information learning unit 72. As an example of the information associating examination results with brain states, it may be the same or equivalent information as the examination-brain state related information shown in FIG. 3, and may include information such as brain images.

[0071] The control unit of the information processing system 1 includes the examination-brain state related information learning unit 72. The examination-brain state related information learning unit 72 learns the relationship between examination results and brain states using a database including the relationship between examination results and brain states held by the examination-brain state related information database holding unit 71. The examination-brain state related information holding unit 11 holds the learning result of the examination-brain state related information learning unit 72. The examination-brain state related information database holding unit 71 is constituted by the memory 21, and the examination-brain state related information learning unit 72 is constituted by the CPU 23.

[0072] The examination-brain state related information learning unit 72 learns the relationship between examination results (e.g., scores) and brain lesion sites by using a database including the relationship between examination results and brain states held by the examination-brain state related information database holding unit 71, so that it is possible to extract features (e.g., score trends) of examination results common to a plurality of subjects having the same brain lesion site, and records this feature as a learning result in the examination-brain state related information holding unit 11. Then, when estimating the brain state from the examination results of a certain subject, the analysis means 13 determines whether or not the subject has the same features as the features of the learning result, and if it has the same features, it can be estimated that there is the same brain lesion site. By performing such learning, an improvement in the estimation accuracy of the brain state can be expected. Also, by learning in advance and performing the estimation based on the learning result, it is possible to shorten the time required for estimating the brain state.

[0073] The input unit 22 receives clinical database 91, the results of examination 93, and the information of the brain function database 51. The clinical database 91 and the brain function database 51 may be stored in, for example, external storage means.

[0074] The clinical database 91 includes information on lesion sites such as CAT examinations, questionnaire-based examinations, and simple movement measurement examinations such as finger tapping examinations, and structural MRI (Magnetic Resonance Imaging) and CT (Computed Tomography) examinations. The examination results and the related information on the brain state (examination-brain state related information) are stored in the examination-brain state related information database holding unit 71 as the related information between the examination results and the brain state. The examination-brain state related information database holding unit 71 holds the brain images input by the input unit 22.

[0075] The examination results that are the results of examination 93 are recorded in the examination result holding unit 12.

[0076] The brain function database 51 includes the correspondence relationship of what functions correspond to each brain region and brain coordinates, and is recorded in the examination-brain state related information holding unit 11 as the related information between the brain region and the brain function.

[0077] FIG. 12 is a diagram showing an example of a feature extraction screen at the time of presenting a brain image based on finger tap performance. The brain state output unit displays a screen including the examination display unit 31 indicating information regarding the examination instruction on the subject state display unit 17. In the example of FIG. 12, an instruction is given to tap the fingers with both hands. When the information processing system 1 receives, through the input unit 22, a captured image (captured information) of finger tapping by a camera connected to the information processing system 1, the captured information is recorded in the examination result holding unit 12. The brain state output unit displays a screen including the captured information display unit 32 indicating the captured information on the brain state display unit 24 and provides feedback to the subject on whether the subject can tap the fingers well.

[0078] Furthermore, the inspection result holding unit 12 records the feature amounts including those of past inspections. Then, the brain state output unit displays on the subject state display unit 17 a screen including a feature amount display unit 33 that shows the time-series feature amounts, using the recorded feature amounts. Here, the tip coordinates of the index finger and the thumb are displayed in time series for the right hand (R) and the left hand (L), respectively.

[0079] FIG. 13 is a diagram showing an example of presenting a brain image based on finger tapping performance. The analysis means 13 determines, for example, the magnitude etc. of each feature amount using the speed of finger tapping, the balance of the sizes between the left and right, and the variation in the phase difference between the left and right, and estimates the presence or absence of a brain lesion based on the determination. The brain state output unit displays on the subject state display unit 17 a screen including a feature amount determination result display unit 35 that shows the determination result of each feature amount. When, as a result of the determination, each score regarding each feature amount is worse than a predetermined standard, the brain state output unit displays on the estimated brain map 34 the probability that the corresponding brain part is a brain lesion. At this time, the brain state output unit uses information such as the inspection - brain state related information holding unit 11 and the inspection·biological signal - brain lesion database 50. Although an example of displaying the brain lesion part has been shown here, not limited to the brain lesion part, it may be configured to display a pre - function part or a residual function part, or to display a fusion map 43 of the brain lesion part and the pre - residual function part.

[0080] The analysis means 13 creates the brain map 34 by calculating and mapping the lesion probability that has been standardized. When there are multiple examinations as predetermined examinations, the analysis means 13 may create the brain map 34 by superposition based on the lesion probabilities estimated from each examination. Also, the examination-brain state related information learning unit 72 may learn in advance the patterns of all examinations and reflect the learning results in the examination-brain state related information. In this way, by combining the results of multiple examinations, it becomes possible to construct a map with higher accuracy. The examination·bio-signal-brain lesion database 50 includes brain structure, blood components, actual brain images, brain lesion positions, histories of rehabilitation programs, information on staff corresponding to rehabilitation, etc., corresponding to the behavioral measurement·analysis indices (feature quantities), and it is possible to analyze how these factors affect the behavioral indices and the like.

[0081] FIG. 14 simply shows the flow of estimating the brain lesion position by calculating the total movement distance during left and right finger tapping. As shown in the feature quantity display unit 33, when the biological data acquisition unit 61 receives a captured image (captured information) of finger tapping by a camera connected to the information processing system 1, the feature quantity extraction unit 62 calculates the distance between the thumb and the index finger during left and right finger tapping from the captured image (step S1601). The analysis means 13 calculates the total movement distance for each of the left and right from this (step S1602). The analysis means 13 determines the magnitude etc. of the total movement distance for each of the left and right, which is this feature quantity, based on a predetermined criterion, and estimates the presence or absence of a brain lesion based on this determination. For example, when the total movement distance of the left hand is extremely small, etc., a functional deficit in the right motor area is estimated. Such an estimation is derived from the examination·bio-signal-brain lesion DB50 held in the examination-brain state related information holding unit 11 (step S1603).

[0082] Although this embodiment shows a simple example, even when multiple brain regions are related, etc., the information processing system 1 can output the estimated position of the brain lesion as a probability map by inverse projection of the brain map by database search.

[0083] According to the present embodiment, a database regarding the relationship between a simple examination (a cognitive test that can be implemented using paper, a tablet, etc.) and brain conditions such as the location of lesions and infarcts is held, for example, by the examination-brain state related information holding unit 11, and a state is made such that brain conditions can be estimated (back-projection from the result to the cause) from the simple examination by machine learning or the like.

[0084] In addition, the abnormal part of the brain and the effect of intervention (rehabilitation) or the like can be simultaneously displayed by a simple examination.

[0085] The analysis means 13 estimates the lesion site of the brain from a plurality of types of parameters of one or more simple tests. The analysis means 13 can estimate the lesion site of the brain from the superposition of probability maps or the like.

[0086] The analysis means 13 may obtain a "function suspected of being defective" from a database (a literature database such as Neurosynth) for the estimated lesion. Further, the analysis means 13 may obtain a "function expected to remain" borne by a site outside the lesion from the database. The information processing system 1 provides a simple examination regarding that function (function suspected of being defective). The analysis means 13 estimates a "remaining function" according to the test score of the examination. When the examination is performed before and after rehabilitation, events (such as a rehabilitation program) performed during the examination are also held in the examination result holding unit 12 as a database, and the analysis means 13 and the examination-brain state related information learning unit 72 evaluate the relationship between the change amount of the examination result and the rehabilitation program (correlation analysis, principal component analysis, machine learning, etc.), and it may be possible to visualize which rehabilitation program was effective for which improvement of brain function. The patient's rehabilitation time and work restraint time are similarly acquired and recorded in the examination result holding unit 12, and by analyzing with the analysis means 13 and the examination-brain state related information learning unit 72, the burden on medical staff may be scored and visualized. Note that the brain state output unit is visualized, for example, by outputting a screen to the subject state display unit 17.

[0087] The visualization result is not limited to a three-dimensional brain lesion / brain defect map (brain map), and may be numerical data (score) such as the volume in each brain region. Shading according to the activity level may be added to the display method of the brain map.

[0088] The information processing system 1 of the present embodiment can estimate and output the severe part of the brain or the location / remaining function of an infarction (lesion) using a simple test without necessarily measuring the brain. Further, in the present embodiment, the brain state estimated by the analysis means 13 is held in the brain state holding unit 14.

[0089] Further, the information processing system 1 of the present embodiment can also cause the subject state display unit 17 to display information related to the recovery prediction of the subject by the analysis means 16. This process is performed using the brain state-state related information held in the brain state-state / intervention related information holding unit 15 and the brain state held in the estimated brain state holding unit 14. Next, an explanation regarding the recovery prediction of the subject will be given.

[0090] First, with reference to FIG. 15, the brain state-state related information will be described. FIG. 15 is a diagram showing an example of the brain state-state related information. The brain state-state related information is held in the brain state-state intervention related holding unit 15, and as shown in FIG. 15, it is data that summarizes the degree of brain injury and symptom scores before and after recovery (in this example, at the time of admission and discharge) of a plurality of subjects. This brain state-state related information includes a brain injury degree table and a symptom score table. Here, the degree of brain injury is evaluated for each brain area, and in this example, a value based on a percentage is stored. Note that the higher the numerical value, the higher the degree of injury. On the other hand, the symptom score is the total score of the test based on the results of one process of examination, and the higher the score, the better it indicates. And in the symptom score table, the symptom scores are stored for each different process of examination (described as "test" in FIG. 15). Note that the examination (test) is related to the physical and mental state of the subject (that is, related to the second examination). Therefore, the symptom score is data based on the results of the second examination, and the symptom score is evaluated, for example, by scores such as the depression tendency and the lack of motivation tendency by a standardized questionnaire, the results of a doctor's interview, and the degree of improvement of the ADL (Activity of Daily Living) score in rehabilitation.

[0091] Note that the example in FIG. 15 is data with inpatients before and after hospitalization as the subjects, but the brain state-state related information may be data appropriately evaluated for before and after the recovery of the subject, and the subject is not limited to inpatients. For example, the subject may be a person who regularly uses a rehabilitation facility.

[0092] The analysis means 16 performs a process of predicting recovery using the brain state-state related information described above. Next, an example of the process of predicting recovery will be described with reference to FIG. 16. FIG. 16 is a flowchart for explaining an example of the process of displaying the recovery prediction of the subject.

[0093] The analysis means 16 reads out the symptom score table of the brain state-state related information from the brain state-state intervention related holding unit 15 (S601). Then, the analysis means 16 performs data-driven analysis (for example, appropriate clustering) using the symptom scores in the symptom score table, and generates a plurality of groups classified from the perspective of the recovery trend (that is, the pattern of changes in symptom scores before and after recovery) (S602). As a result, a plurality of groups with different recovery trends of the subject are generated, and a plurality of subjects with similar recovery trends belong to one group.

[0094] In the present embodiment, in the process of S602, the analysis means 16 generates three groups (group 1 to group 3) with different recovery trends of the subject. That is, the analysis means 16 generates group 1 to which the subject with a large difference between the symptom score before recovery and the symptom score after recovery (that is, the subject with a good recovery trend and a remarkable recovery) belongs. The analysis means 16 generates group 3 to which the subject with little recovery trend and a small difference between the symptom scores before and after recovery belongs. The analysis means 16 generates group 2 to which the subject with a difference in symptom scores before and after recovery being intermediate between the cases of group 1 and group 3, having a generally good recovery trend and showing recovery belongs.

[0095] Next, the analysis means 16 reads out the brain injury degree table of the brain state-state related information from the brain state-state intervention related holding unit 15 (S603). Then, the analysis means 16 specifies the brain regions of the subjects belonging to each group generated in S602 (S604). Specifically, the analysis means 16 acquires the brain injury degree (before recovery, and in this example, the brain injury degree at the time of hospitalization) of the subject belonging to the group from the brain injury degree table. Then, the analysis means 16 specifies the brain regions common to each subject within the group (the brain regions where brain injury is recognized) using an appropriate method (for example, data matching processing).

[0096] Next, the analysis means 16 creates a brain region-group table (S605). An example of the brain region-group table will be described with reference to FIG. 17. FIG. 17 is a diagram showing an example of the brain region-group table.

[0097] As shown in FIG. 17, the brain region-group table is data that summarizes which group the brain regions of the subject specified in S604 correspond to. In the example of FIG. 17, data showing the specified brain regions by images is stored, but instead of or in addition to this, data on the degree of brain damage (numerical value) of the specified brain regions may be stored. Also, in addition to the information for identifying the group corresponding to the brain region, a radar chart regarding the symptom scores before and after recovery may be generated and stored by the analysis means 16. Note that the radar chart may not be stored in the brain region-group table (that is, may be omitted from the brain region-group table) and may be generated in S606 described later.

[0098] Here, the radar chart will be described. The numerical values of the radar chart correspond to the tests (that is, test in the symptom score table), and in the example of FIG. 17, 15 types of tests are shown. The radar chart shows the magnitude of the symptom score. That is, as going from the center of the radar chart to the outside, the value of the symptom score increases. Also, the magnitude of the symptom score before recovery is shown by a dotted line on the radar chart, and the magnitude of the symptom score after recovery is shown by a solid line on the radar chart.

[0099] In the radar chart, in the example of FIG. 17, 15 types of tests are shown, but the types of tests can be appropriately selected from the tests included in the symptom score table. Also, the magnitude of the symptom score in the radar chart is determined by an appropriate method. The magnitude of the symptom score may be, for example, the average value of the symptom scores of each subject belonging to the group. For example, the magnitude of the symptom score regarding testA may be the average value of testA of each subject belonging to the group.

[0100] The analysis means 16 reads out the brain state from the estimated brain state holding unit 14. Then, the analysis means 16 refers to the brain region-group table, selects a group of brain regions close to the brain state read out from the estimated brain state holding unit 14 from the brain region-group table, and displays the radar chart corresponding to the selected group on the subject state display unit 17 as a recovery prediction (S606). If the radar chart is not stored in the brain region-group table, the analysis means 16 generates and displays the radar chart corresponding to the selected group.

[0101] In addition to the radar chart, the analysis means 16 may display on the subject state display unit 17 a message explaining the recovery tendency of the subject. The message may be composed of, for example, a sentence explaining the current state and a sentence explaining the expected recovery in the future. On the subject state display unit 17, for example, a message such as "Your brain is in the state shown in the figure, and your physical function is not good at present, but there is an expectation of recovery as shown in the figure." and an image (a brain map and a radar chart of the subject corresponding to the figure of the message) may be displayed.

[0102] In this embodiment, the analysis means 16 generates three groups, but it is sufficient if the recovery tendency can be appropriately classified. For example, four or more groups may be generated. Also, in the above processing, the brain state is read out from the estimated brain state holding unit 14 in S606, but the brain state may be read out earlier than this.

[0103] New subject data may be added to the brain state-state related information. For example, the brain state-state related information may be updated at the timing when data of a plurality of subjects are collected. Then, the analysis means 16 may perform the processing (S601 to S606) using the updated brain state-state related information.

[0104] Here, when the size of the symptom score in the radar chart is the average value of the symptom scores of each subject belonging to the group, the analysis means 16 can assign the data of the new subject to the existing group by a method that does not rely on data-driven analysis. Then, the analysis means 16 may update the brain state-group table according to the update of the brain state-state related information. Next, the assignment process by the analysis means 16 will be described.

[0105] Each symptom score on the radar chart is represented by the average value of each subject. Therefore, each symptom score can be considered as a probability density distribution considering the variation. Thus, the analysis means 16 evaluates the similarity between the symptom score on the radar chart and the data (symptom score) of the new subject. Note that the method for evaluating the similarity is not particularly limited, and for example, the following method can be used.

[0106] That is, the analysis means 16 inputs the symptom score of the new subject into the probability density distribution of the symptom score and calculates the distance from the average value (that is, the distance from the center of the probability density distribution). Note that the symptom score of the probability density distribution and the symptom score of the new subject relate to the same test.

[0107] The analysis means 16 uses the calculated distance to obtain the total sum of the probability density around the average value by integration. That is, when the distance from the average value is σa, the analysis means 16 performs integration in the range from -σa to +σa and obtains the integration value. The analysis means 16 performs the same process for all symptom scores in one group and obtains the integration values of all symptom scores. Note that the smaller the distance from the average value, the higher the similarity (that is, the symptom score of the new subject is closer to the average value), so the smaller the integration value, the higher the similarity. Then, the analysis means 16 aggregates the integration values of all symptom scores and performs a process to obtain the aggregation result.

[0108] The analysis means 16 performs all of the above processes for all groups. Then, the analysis means 16 compares the aggregation results (the values obtained by aggregating the integrated values) for each group, selects the group with the smallest value, and assigns a new subject to the selected group. In this way, the analysis means 16 performs the process of assigning a new subject to an existing group.

[0109] Next, a second embodiment will be described. In the second embodiment, an information processing system that outputs an appropriate intervention menu (i.e., a rehabilitation menu) in addition to recovery prediction will be described. Note that the content described in the above embodiments may be omitted.

[0110] In the present embodiment, an intervention menu table is stored in the brain state - state / intervention related information holding unit 15, and the analysis means 16 performs processing using the intervention menu table. First, the intervention menu table will be described with reference to FIG. 18. FIG. 18 is an example of an intervention menu table.

[0111] As shown in FIG. 18, the intervention menu table contains data indicating the effects of interventions when interventions were performed on a plurality of subjects from before recovery to after recovery (in this embodiment, from the time of admission to the time of discharge). In the intervention menu table, the interventions (x1, x2 ··· xn) each represent a rehabilitation menu with a different configuration, and the number of interventions performed on a plurality of subjects from before recovery to after recovery is stored in the intervention menu table. The effects (x1, x2 ··· xn) are parameters for evaluating the degree of recovery achieved as a result of the interventions (x1, x2 ··· xn), and in the intervention menu table, the effects (x1, x2 ··· xn) are stored for each intervention (x1, x2 ··· xn) on each subject. Note that the intervention menu table may be updated at an appropriate timing (for example, when data for a new subject is aggregated).

[0112] An example of the processing of the analysis means 16 in this embodiment will be described. FIG. 19 is a flowchart for explaining an example of the processing of displaying an effective intervention menu.

[0113] In this embodiment, the same processing as the processing (S601 to S605) of the analysis means 16 in the first embodiment is performed. That is, the analysis means 16 uses the symptom score table to generate a plurality (three in this embodiment) of groups with different recovery tendencies, and uses the brain damage degree table to identify the brain regions of the subjects belonging to each group, and creates a brain region-group table (S701).

[0114] Next, the analysis means 16 reads the intervention menu table. Then, the analysis means 16 acquires, from the intervention menu table, an intervention (rehabilitation menu) that is highly effective for the subjects belonging to each group in the brain region-group table. Here, the analysis means 16 acquires a highly effective intervention for each group in the brain region-group table (S702).

[0115] The analysis means 16 may evaluate and acquire a highly effective intervention from the intervention menu table based on an appropriate method. For example, the analysis means 16 may quantitatively evaluate the magnitude of the effect and acquire a representative intervention with a magnitude of the effect equal to or greater than a predetermined value from the intervention menu table. Further, the analysis means 16 may also consider the number of interventions and acquire an intervention with a number of interventions equal to or less than a predetermined value and a magnitude of the effect equal to or greater than a predetermined value.

[0116] The analysis means 16 may obtain highly effective interventions by the following method. That is, the analysis means 16 classifies the subjects in the intervention menu table from the same perspective as the processing in the first embodiment (i.e., the perspective of the recovery tendency based on the symptom score), and generates a plurality of groups classified from the perspective of the recovery tendency. The analysis means 16 refers to each subject belonging to the generated group and obtains highly effective interventions for each group. Here, the analysis means 16 may obtain highly effective interventions based on an appropriate method. For example, the analysis means 16 may obtain representative interventions with an effect magnitude equal to or greater than a predetermined value. Further, the analysis means 16 may obtain the intervention that results in the most effects equal to or greater than a predetermined value for each subject. Further, the analysis means 16 may also consider the number of interventions and obtain interventions with the number of interventions equal to or less than a predetermined value and an effect magnitude equal to or greater than a predetermined value.

[0117] The analysis means 16 reads the brain state from the estimated brain state holding unit 14. Then, the analysis means 16 refers to the brain region-group table, selects a group of brain regions close to the brain state read from the estimated brain state holding unit 14 from the brain region-group table, and displays the intervention corresponding to the selected group on the subject state display unit 17 (S703). Note that the analysis means 16 may obtain a radar chart corresponding to the selected group from the brain region-group table, or generate a radar chart corresponding to the selected group, and display a radar chart showing the recovery prediction in addition to the intervention on the subject state display unit 17.

[0118] In addition, the analysis means 16 may display a message explaining the recovery tendency and intervention of the subject. The message may be composed of, for example, a sentence explaining the current state, a sentence explaining the expected future recovery, and intervention information. On the subject state display unit 17, for example, a message such as "Your brain is in the state shown in the figure, and your physical function is not good at present, but there is a possibility of recovery as shown in the figure. It is advisable to perform intervention Xn every day." and an image (a brain map and a radar chart of the subject corresponding to the figure in the message) may be displayed. Here, the intervention Xn is the intervention acquired in S703. Also, a message for the rehabilitation staff (for example, a message such as "Please execute intervention Xn.") may be displayed.

[0119] Next, a third embodiment will be described. In the third embodiment, an information processing system that can estimate states other than the brain state (for example, mental state and visceral sensation) in addition to estimating the brain state will be described. Note that the content described in the above embodiments may be omitted.

[0120] In this embodiment, the behavioral examination table is held in the examination-brain state related information holding unit 11, and the brain injury degree information is held in the examination-brain state related information holding unit 11. The analysis means 13 performs processing using the behavioral examination table and the brain injury degree information.

[0121] In the behavioral examination table, the results (second examination results) of the mental and physical states (that is, mental and physical) of a plurality of subjects for a predetermined examination are stored. In the behavioral examination table, data (behavioral examination scores) in which the examination results are scored are stored. The behavioral examination scores may include, for example, an anxiety score, a depression score, a stress score, and the like.

[0122] The brain injury degree information is data in which the brain injury degree is evaluated numerically (for example, as a percentage) for each brain region for a plurality of subjects. Specifically, the brain injury degree information is data in which the brain is divided into 116 regions by the AAL (Automated Anatomical Labeling) method and the brain injury degree for each brain region is evaluated, and is stored in a database. Note that the brain injury degree information may be the brain injury degree table described in the second embodiment above, which includes data for 116 brain regions (areas).

[0123] Next, the processing of the analysis means 13 will be described with reference to FIG. 20. FIG. 20 is a flowchart for explaining an example of the process of estimating the brain state.

[0124] The analysis means 13 reads the behavioral examination table from the examination-brain state related information holding unit 11 (S801). Then, the analysis means 13 performs data-driven analysis (for example, cluster analysis) on the behavioral examination scores to classify the behavioral examination scores of each subject into a plurality of groups (S802). The analysis means 13 generates, for example, four groups from the viewpoints of the anxiety score and the stress score. In this case, the analysis means 13 uses the magnitude relationship of the scores to generate a group with a high anxiety score and a high stress score, a group with a high anxiety score and a low stress score, a group with a low anxiety score and a high stress score, and a group with a low anxiety score and a low stress score.

[0125] The analysis means 13 reads the brain injury degree information (S803). Then, for the subjects belonging to each group classified in S802, the brain regions in which brain injury is recognized are specified (S804). In this embodiment, since the classification is made into four groups in S802, four brain regions (Z1 to Z4) corresponding to each group are specified. Here, the process of specifying the brain regions in which brain injury is recognized is executed as follows as an example.

[0126] That is, anxiety disorders and stress disorders (i.e., symptoms related to the second examination) are known to be related to damage in certain brain regions. As an example, the right Rolandic operculum included in the brain region is known as a site related to emotion processing, and it is considered that emotion processing may malfunction due to damage to this site. Therefore, it is considered likely that damage has occurred in a common brain region in each subject within the group classified in S802.

[0127] Therefore, the analysis means 13 identifies the common brain region where brain damage is recognized in each subject within the group. The method for identifying the brain region within the group is not particularly limited as long as it can be appropriately executed. The analysis means 13 may, for example, extract a common brain region where the degree of brain damage is higher than a threshold value for each subject belonging to the group, and identify a common brain region where the degree of brain damage is higher than a certain value. Note that the brain regions (Z1 to Z4 in this embodiment) identified in this way may be referred to as "brain damage regions".

[0128] The analysis means 13 reads out the test result of the subject (test result related to the first test) from the test result holding unit 12 (S805). Here, the brain state of the subject may be obtained by the same processing as in the case of the first embodiment, but in this embodiment, the analysis means 13 compares the brain region where damage is recognized in the brain state of the subject with the brain damage region, and determines whether there is a brain region corresponding to the brain damage region in the brain state of the subject. When there is a corresponding brain damage region, the analysis means 13 selects the corresponding brain damage region, and the estimated brain state holding unit 14 holds a brain state including information on the brain damage region (S806).

[0129] Incidentally, as described above, anxiety disorders and stress disorders (i.e., symptoms based on the second test result) are known to be related to damage in certain brain regions. Therefore, when the estimated brain state includes a brain damage region, it can be considered that there is a problem with the second test result.

[0130] Therefore, when the estimated brain state includes a brain damage area, a state other than the brain state may be output from the brain state output unit. For example, when the subject's brain state includes a brain damage area and the brain damage area includes the right Rolandic operculum, it may be output that there is a problem with the subject's emotion processing. Also, the right Rolandic operculum is also known as a part related to the state of visceral sensation (e.g., the state of the intestine). Therefore, it may also be output that there is a problem with the state of visceral sensation. And, as an example, the message of the output result may be displayed on the subject state display unit 17.

[0131] Therefore, according to the present embodiment, based on the brain state of the subject, the state related to the mind of the subject and the state related to the body of the subject can be estimated, and based on the brain state, the state of the subject other than the brain state can be estimated.

[0132] Needless to say, all the concepts of the present invention represented by the content described in the above embodiments may be applied to inspections and state grasping of the living body other than the brain. The present invention is not limited to the above-described embodiments and includes various modifications. For example, the above-described embodiments have been described in detail for better understanding of the present invention, and are not necessarily limited to those having all the configurations described.

Explanation of Reference Numerals

[0133] 1: Information processing system, 11: Inspection-brain state related information holding unit, 12: Inspection result holding unit, 13: Analysis means, 14: Estimated brain state holding unit, 15: Brain state-state / intervention related information holding unit, 17: Subject state display unit, 21: Memory, 22: Input unit, 31: Inspection display unit, 32: Photographing information display unit, 33: Feature amount display unit, 50: Inspection-biological signal-brain lesion database, 51: Brain Region - Brain Function Database, 52: Brain Function - Rehabilitation Database, 54: Intervention - Brain Region Database, 61: Biological Data Acquisition Unit, 62: Feature Extraction Unit, 71: Examination - Brain State - Related Information Database Holding Unit, 72: Examination - Brain State - Related Information Learning Unit, 91: Clinical Database, 94: Intervention.

Claims

1. A storage unit that stores test-brain state association information correlating first test results, which are results of the behaviors of a plurality of subjects in a predetermined test, with the brain states of the subjects, and second test results, which are results of the mental and physical states of the plurality of subjects in a predetermined test; an input unit that receives a first test result that is a result of a subject's behavior with respect to the predetermined test; a control unit that estimates a brain state from the first examination result based on the examination-brain state related information and estimates a state other than the brain state; an output unit that outputs a state other than the estimated brain state; Equipped with The control unit is classifying the second test results into groups, extracting common brain regions for each subject belonging to each group in which the degree of brain damage is higher than a threshold, and identifying the common brain regions in which the degree of brain damage is higher than a certain value as brain damaged regions; The brain damage area is compared with the estimated brain state, and a mental state is estimated as a state other than the brain state. An information processing system comprising:

2. 2. The information processing system according to claim 1, The control unit is When the estimated brain state includes a brain region corresponding to the brain damage region related to the right Rolandic operculum, a state related to the internal organs is estimated as a state other than the brain state. An information processing system comprising:

3. 2. The information processing system according to claim 1, The storage unit is storing brain state-state association information associating the subject's brain state before and after recovery with the second test results before and after recovery; The control unit is predicting recovery of the subject using the brain state-state association information and the brain state estimated from the first test result; An information processing system comprising:

4. 4. The information processing system according to claim 3, The control unit is classifying the brain state-state related information based on the recovery tendency of the subject, and generating a plurality of groups to which a plurality of subjects having similar recovery tendencies belong; Identifying a group corresponding to a brain state estimated from the first test result from the plurality of groups; performing a recovery prediction corresponding to the identified group; An information processing system comprising:

5. 5. The information processing system according to claim 4, A display unit is provided, The display unit displays a radar chart showing the recovery prediction. An information processing system comprising:

6. 6. The information processing system according to claim 5, The radar chart includes data based on the second test results before and after recovery. An information processing system comprising:

7. 4. The information processing system according to claim 3, A display unit is provided, The storage unit is storing an intervention menu table that associates a plurality of different intervention menus with effects of each of the intervention menus on a plurality of subjects; The control unit is selecting an intervention menu having a high recovery effect on the subject for which the recovery prediction is performed from the intervention menu table; displaying the selected intervention menu and the recovery prediction on the display unit; An information processing system comprising:

8. A method for estimating a state of a subject by processing using a computer, comprising: Obtaining test-brain state association information correlating first test results, which are results of the behaviors of a plurality of subjects in a predetermined test, with the brain states of the subjects, and second test results, which are results of the mental and physical states of the plurality of subjects in a predetermined test; obtaining a first test result that is a result of a subject's behavior in response to the predetermined test; Inferring a brain state from the first examination result based on the examination-brain state related information, and estimating a state other than the brain state; In the estimation of a state other than the brain state, classifying the second test results into groups, extracting common brain regions for each subject belonging to each group in which the degree of brain damage is higher than a threshold, and identifying the common brain regions in which the degree of brain damage is higher than a certain value as brain damaged regions; The brain damage area is compared with the estimated brain state, and a mental state is estimated as a state other than the brain state.

23. An information processing method comprising:

9. 9. The information processing method according to claim 8, When the estimated brain state includes a brain region corresponding to the brain damage region related to the right Rolandic operculum, a state related to the internal organs is estimated as a state other than the brain state.

23. An information processing method comprising:

10. A program for causing an electronic computer to execute the information processing method according to claim 8.

Images