[go: up one dir, main page]

WO2025243619A1 - Programme de prédiction de crise, support de stockage, dispositif de prédiction de crise et procédé de prédiction de crise - Google Patents

Programme de prédiction de crise, support de stockage, dispositif de prédiction de crise et procédé de prédiction de crise

Info

Publication number
WO2025243619A1
WO2025243619A1 PCT/JP2025/005202 JP2025005202W WO2025243619A1 WO 2025243619 A1 WO2025243619 A1 WO 2025243619A1 JP 2025005202 W JP2025005202 W JP 2025005202W WO 2025243619 A1 WO2025243619 A1 WO 2025243619A1
Authority
WO
WIPO (PCT)
Prior art keywords
seizure
information
subject
prediction
trained model
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
PCT/JP2025/005202
Other languages
English (en)
Japanese (ja)
Inventor
デービット・エー． リーブレック
智 古村
信和 中里
一敬 神
一司 浮城
舞美 小川
真由 藤川
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Emind Inc
Tohoku University NUC
Original Assignee
Emind Inc
Tohoku University NUC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Emind Inc, Tohoku University NUC filed Critical Emind Inc
Publication of WO2025243619A1 publication Critical patent/WO2025243619A1/fr
Pending legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems

Definitions

  • the present invention relates to a seizure prediction program, a storage medium, a seizure prediction device, and a seizure prediction method.
  • Patent Document 1 describes an epileptic seizure prediction technology that detects signs of an epileptic seizure from data generated from electrocardiogram signals. The technology described in Patent Document 1 detects signs of an epileptic seizure based on whether information based on heart rate exceeds a control limit that should be met when the patient is in an interictal period.
  • Patent Document 1 Technologies such as the one described in Patent Document 1 are known that detect epileptic seizures and the signs of seizures based on biological information such as heart rate, but these technologies require the acquisition of biological information. This requires the use of expensive equipment to measure the biological information and the placement of electrodes on the subject's body. Another problem is that signs of an epileptic seizure appear in biological information just before a seizure occurs, and even if signs are detected, there are limited countermeasures that can be taken.
  • the present invention aims to provide a novel technology that is highly useful for predicting seizures.
  • the present invention provides a seizure prediction program that causes a computer to function as an acquisition unit and a prediction unit.
  • the acquisition unit acquires log information that includes at least one of behavioral information indicating the subject's behavior, emotional information input based on the subject's subjective opinion, and seizure information related to records of seizures that have occurred in the subject.
  • the prediction unit predicts seizures associated with the subject's brain activity or mental activity based on the log information.
  • seizures can be predicted using log information without using biometric information. This makes it possible to predict seizures based on information that can be obtained without placing the burden of constantly wearing a device on the subject or interfering with the subject's body.
  • the acquisition unit acquires at least two of the behavioral information, the emotional information, and the seizure information as the log information, and the prediction unit predicts the seizure of the subject based on the multiple types of log information acquired by the acquisition unit.
  • the behavioral information includes information indicating the behavior and the timing of that behavior
  • the emotional information includes information indicating the emotion felt by the subject and the timing of that emotion
  • the seizure information includes information indicating the timing of a seizure that occurred in the subject
  • the prediction unit predicts a seizure based on the timing indicated by the log information.
  • the prediction unit estimates that a seizure will not occur within a specified time or period after the time indicated by the log information.
  • This configuration makes it possible to predict that seizures will not occur at a specified time. This can greatly improve the quality of life of epilepsy patients and others.
  • the prediction unit estimates that a seizure will not occur on the specified date based on the behavioral information or emotional information from the day before the specified date.
  • the subject can know that a seizure is unlikely to occur on the specified date.
  • a specific date as the specified period, it is expected that the subject will be able to plan their activities more easily. Specifically, on days when the possibility of a seizure is low, they will be able to go out and do other activities with peace of mind, which is expected to have a significant stress-reducing effect.
  • the prediction unit makes the prediction by estimating whether a seizure will occur on the specified date using the seizure interval, which indicates the period of time elapsed since the previous seizure, based on the timing of the seizure information recorded in the past immediately prior to the specified date.
  • the prediction unit makes the prediction by estimating whether a seizure will occur within a specified time or period after the time indicated by the log information.
  • This configuration makes it easier for the subject to take measures based on the prediction results. Specifically, during periods when seizures are likely to occur, the subject can take a variety of measures, such as using medication or avoiding dangerous objects or actions.
  • the prediction unit inputs the log information into a trained model generated by machine learning for the purpose of making the prediction, and obtains, as an output of the trained model, an estimated result of whether the subject will experience a seizure within a specified time or period after the time indicated by the log information, and the trained model is generated from training data including the subject's log information and information indicating whether the subject experienced a seizure within a specified time or period after the time indicated by the log information.
  • the trained model which has been trained based on past performance, can predict whether or not a seizure will occur within a specified time or period.
  • the trained model is generated by machine learning using a dataset including the training data for multiple subjects with the same seizure classification as the subject, and the training data for the subject.
  • This configuration makes it possible to achieve more appropriate predictions that reflect the subject's tendencies. Furthermore, if only the subject's data is used, the amount of training data tends to be small, but by combining data from multiple subjects and the subject's data, it is possible to ensure a large amount of training data while also reflecting the subject's characteristics.
  • the prediction unit predicts the subject's seizures based on the log information as well as the subject's seizure classification.
  • This configuration is expected to result in more accurate predictions.
  • the acquisition unit acquires the behavioral information as the log information, and the behavioral information includes sleep information related to the subject's sleep.
  • the acquisition unit acquires the behavioral information as the log information, and the behavioral information includes usage information acquired from background data on a device used by the subject.
  • This configuration allows behavioral information to be obtained without the subject having to enter it themselves, further reducing the burden on the subject.
  • the acquisition unit acquires the behavioral information as the log information, and the behavioral information includes movement information regarding the movement speed of the subject.
  • the acquisition unit further acquires evaluation information based on the results of an examination or ability test of the subject, and the prediction unit predicts a seizure in the subject based on the evaluation information in addition to the log information.
  • This configuration is expected to result in more accurate predictions.
  • the present invention provides a seizure prediction device that includes an acquisition unit and a prediction unit.
  • the acquisition unit acquires log information that includes at least one of behavioral information indicating the behavior of a subject, emotional information input based on the subject's subjective opinion, and seizure information related to records of seizures that have occurred in the subject.
  • the prediction unit predicts seizures associated with the subject's brain activity or mental activity based on the log information.
  • the present invention provides a seizure prediction method in which a computer including an acquisition unit and a prediction unit acquires, via the acquisition unit, log information including at least one of behavioral information indicating the behavior of a subject, emotional information input based on the subject's subjective opinion, and seizure information related to records of seizures that have occurred in the subject, and predicts, via the prediction unit, a seizure associated with the subject's brain activity or mental activity based on the log information.
  • the present invention provides a seizure prediction method that acquires log information including at least one of the following: behavioral information indicating the subject's behavior and the timing of those actions; emotional information indicating the subject's subjective emotions and the timing of those emotions; and seizure information relating to a record of seizures that have occurred in the subject; and, based on the log information for the day before the specified date, estimates whether a seizure associated with brain activity or mental activity will occur on the specified date.
  • the present invention provides a novel technology that makes it easy to predict seizures.
  • FIG. 1 is a block diagram showing the configuration of a system according to an embodiment of the present invention.
  • FIG. 2 is a hardware configuration diagram of an information processing apparatus and a terminal device according to the embodiment.
  • 10 is a flowchart showing a seizure prediction procedure according to the present embodiment.
  • the present invention relates to technology for predicting seizures associated with brain activity or mental activity in a subject.
  • a seizure refers to a sudden symptom that occurs repeatedly in the same patient in association with brain activity or mental activity.
  • epileptic seizures and seizures similar to these (exhibiting similar symptoms), whose symptoms appear in the state of consciousness or motor state, are predicted.
  • Such seizures are classified into, for example, focal seizures, generalized seizures, and psychogenic non-epileptic seizures (PNES).
  • PNES psychogenic non-epileptic seizures
  • the term "subject" refers to the subject for whom a seizure is predicted.
  • the subject of the present invention is, in particular, a person who is at risk of having a seizure.
  • the subject refers to an epilepsy patient or a patient with psychogenic non-epileptic seizures who is the subject of seizure prediction.
  • the present invention is intended to present the results of seizure prediction, and is not intended as a therapeutic or diagnostic procedure.
  • this embodiment describes the configuration, operation, etc. of a seizure prediction device
  • similar effects can also be achieved by a system, method, method executed by a device, or computer program that causes a computer device to execute a method that has similar functions.
  • the program may be provided as a non-transitory computer-readable recording medium, or may be provided so that it can be downloaded from an external server.
  • Fig. 1 is a block diagram showing the configuration of a seizure prediction system including a seizure prediction device of this embodiment.
  • the seizure prediction system 0 of this embodiment includes a seizure prediction device 1 and a user terminal 2.
  • the seizure prediction device 1 and the user terminal 2 are configured to be able to communicate via a network NW.
  • the network NW is an IP (Internet Protocol) network, but there are no limitations on the type of communication protocol, type of network, etc.
  • the seizure prediction device 1 can be one or more information processing devices 10 (computer devices), such as a general-purpose server or personal computer.
  • the user terminal 2 can be a terminal device 9 (computer device), such as a personal computer, smartphone, or tablet terminal.
  • the seizure prediction device 1 is an information processing device 10 installed with a computer program (seizure prediction program) that executes a seizure prediction method.
  • FIG. 2(a) is a hardware configuration diagram of the information processing device 10. As shown in FIG. 2, the information processing device 10 has a control unit 101, a memory unit 102, and a communication unit 103, which are used to perform the functions of each unit and each process.
  • the control unit 101 has a processor such as a CPU that can execute an instruction set, and executes an OS and programs.
  • the storage unit 102 has a volatile memory such as a RAM capable of storing an instruction set, and a non-volatile recording medium such as an HDD or SSD capable of recording an OS, a seizure prediction program, a DBMS (database server), etc.
  • the communication unit 103 has an interface for physically connecting to a network, and controls communication with the network NW to input and output information.
  • Figure 2(b) is a hardware configuration diagram of the terminal device 9.
  • the terminal device 9 has a control unit 901, a memory unit 902, a communication unit 903, an input unit 904, and an output unit 905, which are used to perform the functions of each unit and each process.
  • the control unit 901 has a processor such as a CPU that can execute an instruction set, and executes an OS, programs, and the like.
  • the storage unit 902 includes a volatile memory such as a RAM capable of storing an instruction set, and a non-volatile recording medium such as an HDD or SSD capable of recording an OS, programs, and the like.
  • the communication unit 903 has an interface for physically connecting to a network, and controls communication with the network NW to input and output information.
  • the input unit 904 includes an operation input device capable of input processing, such as a touch panel or a keyboard, and an audio input device capable of audio input, such as a microphone.
  • the output unit 905 includes a display device capable of display processing, such as a display, and an audio output device, such as a speaker.
  • the seizure prediction device 1 has, as its functional configuration, an acquisition unit 11, a prediction unit 12, and a storage unit 13. This is a specific implementation of software-based information processing using hardware.
  • the user terminal 2 also has an input means for the user to input information, and an output means including means for outputting to the user and connecting to the network NW.
  • the user terminal 2 is used by the subject or another person who can observe the subject. In this embodiment, the user terminal 2 is mainly assumed to be a device such as a smartphone owned or carried by the subject.
  • the acquisition unit 11 acquires at least one of behavioral information indicating the behavior of the subject, emotional information input based on the subject's subjective opinion, and seizure information related to records of seizures that have occurred in the subject via the user terminal 2.
  • the acquisition unit 11 acquires behavioral information, emotional information, and seizure information via the user terminal 2.
  • the subject inputs this information via the user terminal 2 by selection or free input, and the user terminal 2 transmits the information to the seizure prediction device 1 together with the subject ID via the network NW.
  • log information records related to the subject that can be used to predict seizures, including seizure information, behavioral information, and emotional information.
  • Log information is linked to each subject's information by a subject ID that identifies the subject, and all information includes time information indicating the time corresponding to each type of record.
  • the time corresponding to each type of record is, for example, date and time information indicating the date and time of the behavior, emotion, seizure, etc. indicated by each record.
  • the acquisition unit 11 acquires log information within a specific time range a predetermined time before the time at which prediction is to be made, and passes it to the prediction unit 12, which will be described later.
  • time information may be information that specifies the day to which it belongs, without including the time of day.
  • a “day” begins when the “subject (target person)” wakes up, and a single day here refers to the period from when the subject wakes up until when they wake up the next day.
  • “behavior” refers to observable movements or reactions exhibited by a subject. Therefore, biological information that cannot be observed without using measuring equipment, such as heart rate, brain waves, and breathing, is not included in the behavioral information of the present invention.
  • the behavioral information in the log information includes, for example, sleep information about the sleep of the subject, movement information about the movement of the subject, usage information about the usage history of the user terminal 2 used by the subject, exercise information about the exercise performed by the subject, diet information about the meals performed by the subject, medication information about the medication history of the subject, free time and working hours or the ratio thereof, etc.
  • information indicating any behavior may be included in the behavioral information in the present invention.
  • Each type of behavioral information includes a unique behavior log ID, time information indicating the date and time of the behavior, and behavior content indicating the type of behavior represented by the behavioral information. Furthermore, the behavioral information is linked to information about the subject who performed the behavior by a subject ID.
  • Sleep information which is part of the behavioral information, includes the time of bedtime (date and time) and time of wake-up (date and time) as the aforementioned time information. Sleep duration can be calculated from the time of bedtime and time of wake-up, but sleep duration may also be input separately. Sleep may also be divided into types such as "real sleep,” which occurs mainly at night, and "naps," such as afternoon naps, and input separately.
  • bedtime and wake-up time refer to the actual start and end times of sleep, but the planned start time of sleep (planned time of bedtime) and planned end time of sleep (planned time of wake-up) may also be acquired. The difference between the planned start time of sleep and the actual start time of sleep, and the difference between the planned end time of sleep and the actual end time of sleep may also be used as sleep information.
  • the boundary between sleep days is set to 5:00 a.m. Specifically, sleep that started between 5:00 a.m. on the previous day and 4:59:59 a.m. on the current day is treated as sleep on the "previous day.” If a person wakes up during the night, the sleep information up to the point where the person feels they have woken up is counted as one piece of sleep information, without separating the information before and after the awakening. However, the time of the awakening during the night may be recorded as sleep information, and the time when the awakening occurred may be excluded from the sleep time.
  • Sleep information also includes information about sleep quality. Specifically, information indicating the feelings upon waking up can be used as information about sleep quality.
  • the feelings upon waking up refer to the subjective feelings about sleep that the subject has. For example, it is expected that the subject will input their feelings upon waking up in the form of selecting from multiple options such as well-rested, fairly well-rested, average, not very well-rested, not at all rested, etc.
  • information relating to sleep quality may also be information representing other indicators, such as brain waves during sleep (for example, the time or percentage at which alpha waves, beta waves, theta waves, and delta waves are detected during sleep), waking up during sleep, body movements during sleep, etc.
  • the information relating to sleep quality be information that can be obtained without measuring brain waves, body movements, etc.
  • the movement information included in the behavioral information includes movement speed.
  • the movement speed of the user terminal 2 can be determined using a GPS (Global Positioning System) receiver or sensor equipped in the user terminal 2, and this can be used as movement information.
  • GPS Global Positioning System
  • the average movement speed, maximum movement speed, minimum movement speed, etc. over a predetermined period, such as a day (from waking up to going to bed) can be used as movement information.
  • background data acquired on the user terminal 2 can be used as usage information relating to the usage history of the user terminal 2 used by the subject.
  • usage information relating to the usage history of the user terminal 2 used by the subject.
  • time and duration of use (total usage time in a specified period) for each app (or each app category), the message sending history and number of replies for a messaging app, etc. can be used as usage information.
  • app categories refer to the division of each app according to its function or purpose, such as tools, SNS, games, music, etc.
  • usage information include screen-off time, screen-on time, number of photos saved, number of videos saved, number of calendar events registered, etc.
  • the exercise information included in the behavioral information includes information such as the type, intensity, and duration of exercise performed by the subject. Specifically, for example, the number of times exercised over a specified period, the total time, the number of steps, the distance walked and run, stride length, the number of floors climbed, the time spent on both feet, stability, speed, active energy (energy consumed during exercise), energy consumed at rest, the number of falls, and other information can be used as exercise information.
  • the exercise information included in the behavioral information includes information such as the type, intensity, and duration of exercise performed by the subject. Specifically, for example, the number of times exercised over a specified period, the total time, the number of steps, the distance walked and run, stride length, the number of floors climbed, the time spent on both feet, stability, speed, active energy (energy consumed during exercise), energy consumed at rest, the number of falls, and other information can be used as exercise information.
  • any information that can be input by the user themselves or that can be obtained by a wearable device may be used.
  • the dietary information includes information such as the content and amount of meals eaten by the subject.
  • the number of meals eaten and calorie intake can also be used as exercise information or dietary information.
  • whether or not a meal was eaten for breakfast, lunch, snacks, dinner, and a late-night snack may be accepted.
  • the medication information included in the behavioral information includes information such as the timing at which the subject took medication (time of day, whether it was morning, noon, or night, or whether it was before or after a meal, etc.), the type and amount of medication, etc. Furthermore, for example, the number of times a medication was taken per day may also be used as medication information.
  • the medication information is input by the subject or an observer via the user terminal 2, and is acquired by the acquisition unit 11.
  • behavioral information described above is an example, and the various types of behavioral information may include any information other than the above.
  • all of the above is used to predict seizures, but only some of the information may be used for prediction.
  • “emotions” refer to feelings that a subjective individual experiences. Therefore, things that can be inferred from biological information such as brain activity information or heart rate are not included in the "emotions" of this invention.
  • Emotional information in the log information is information that indicates the type and intensity of an emotion, expressed in any way, such as the four types of joy, anger, sadness, and pleasure, the eight basic emotions of joy, anticipation, anger, disgust, sadness, surprise, fear, and trust (Plutchik's wheel of emotions), or indicators of pleasure, displeasure, and arousal (Russell's circumplex model).
  • the type and intensity of an emotion selected from multiple pre-set options may be used as emotional information.
  • only the type of emotion may be used as emotional information, without using the intensity.
  • Emotion information may be input only once a day, or may be input at any time (multiple times) when the subject feels an emotion.
  • the acquisition unit 11 acquires emotion information selected subjectively by the user from nine categories - "happy,” “fun,” “calm,” “tranquil,” “bored,” “anxious,” “tired,” “angry,” and “sad” - regarding the overall emotion of the day.
  • Seizure information is one type of log information in the present invention, and is information relating to past seizures that have occurred in the subject. Seizure information includes the date on which the seizure occurred, the time the seizure started and ended, the type of seizure, etc. For example, the type of seizure may be the presence or absence of consciousness and/or the presence or absence of convulsions. Seizure information is input by the subject or an observer via the user terminal 2, and is acquired by the acquisition unit 11.
  • the acquisition unit 11 of this embodiment also acquires environmental information related to the environment surrounding the subject, and biometric information of the subject acquired by a wearable device communicatively connected to the user terminal 2.
  • the environmental information and biometric information are linked to the subject information of each subject by a subject ID that identifies the subject, and both include information indicating the time period corresponding to each type of record.
  • Environmental information may include information such as illuminance, temperature, air pressure, weather, and humidity.
  • the environmental information may be acquired by a sensor equipped on the user terminal 2, or may be acquired from an external database that provides weather data based on the location information of the user terminal 2 obtained by GPS or the like.
  • location information indicating the area of stay and activity range information indicating the range of activity may be used as environmental information. Both location information and activity range information can be obtained based on the position information of the user terminal 2 obtained by GPS or the like. Location information is assumed to be classified by area, such as downtown, outskirts, residential areas, etc., and information on which area classification the location indicated by the location information falls within. Furthermore, for example, the difference between the activity range on weekdays and weekends may be used as activity range information.
  • Biometric information can be any information that can be acquired by sensors equipped in the wearable device. Examples of biological information include heart rate, activity level, and blood pressure. As mentioned above, the present invention primarily predicts seizures based on log information, and biological information can be used as a supplement.
  • the acquisition unit 11 also acquires subject information that has been pre-stored in the storage unit 13.
  • the subject information includes information such as a subject ID that uniquely identifies the subject, the subject's name, seizure classification, attributes, diagnostic results for the subject, and psychological evaluation results.
  • Seizure classification refers to the classification of seizures exhibited by a subject. Seizures associated with brain or mental activity are classified based on the brain activity at the time of the seizure. For example, epileptic seizures cause excessive electrical excitation in the brain, and are divided into “focal seizures (also called partial seizures)” and “generalized seizures” depending on the location and spread of this excitation.
  • seizures that do not show abnormal epileptic waves on the EEG and therefore are not technically "epileptic” seizures, but which suddenly cause mental and physical states similar to those of epileptic seizures are called psychogenic non-epileptic seizures (PNES).
  • PNES psychogenic non-epileptic seizures
  • “seizures” include psychogenic non-epileptic seizures, and "psychogenic non-epileptic seizures” are treated as one type of seizure classification.
  • Seizures associated with brain activity or mental activity are classified according to the patient, and in this embodiment, the seizure that occurs in the subject is registered as the seizure classification in the subject information from the three categories described above: "focal seizures,” “generalized seizures,” and “psychogenic non-epileptic seizures.” Note that seizure classifications are not limited to the above three categories; focal seizures and generalized seizures may be further subdivided, and a seizure classification called "unclassifiable" may also be established.
  • Attributes include the attributes of the subject, such as age or generation, sex, family structure (whether married or not, or whether family members live together, etc.), occupation, years of employment, working hours per week, drinking and smoking habits, recent (e.g., one month) health condition, whether or not the subject is receiving care, housework and exercise habits, etc.
  • the diagnostic results are the diagnosis given to the subject and test data obtained through tests at the hospital.
  • the psychological evaluation results are the results of interviews and psychological evaluation tests administered to the subject, and are an example of the evaluation information of the present invention. For example, evaluation results based on the subject's responses regarding depressive symptoms, QOL, stigma, etc., and ability evaluation results based on the subject's ability test results regarding verbal IQ, memory, etc., are included in the evaluation information.
  • the prediction unit 12 predicts a seizure in the subject based on the log information acquired by the acquisition unit 11.
  • the prediction unit 12 combines the various types of information acquired by the acquisition unit 11 described above and uses them to predict a seizure.
  • the combination of information to be used can be determined arbitrarily, but it is particularly preferable to use a combination of at least two of behavioral information, emotional information, and seizure information to predict a seizure, such as a combination of behavioral information and emotional information, a combination of behavioral information and seizure information, or a combination of emotional information and seizure information.
  • the log information, environmental information, and biological information contain date and time information indicating the date and time as information indicating the time period.
  • the prediction unit 12 predicts a seizure at a specified time based on the date and time information. More specifically, the prediction unit 12 predicts a seizure that will occur later than the time indicated by the information used for the prediction (log information).
  • the prediction unit 12 makes a prediction using information from a predetermined time before the specified time or specified period for which the prediction is to be made.
  • the specified period may be a specific day
  • the prediction unit 12 may predict whether or not a seizure will occur in the subject on the specified day (specified period) using log information, environmental information, and biological information of the subject on the day before the specified day that is the target of prediction.
  • the period of information used for the prediction and the range of time periods for which the prediction is to be made may be changed as desired. For example, information from the week immediately preceding a specified date may be used to predict the presence or absence of a seizure on a specified date. Information from one week may also be used to predict the presence or absence of a seizure in the following week.
  • the prediction unit 12 of this embodiment outputs a prediction result that "a seizure will not occur” (is unlikely to occur) particularly when the possibility of a seizure occurring during a predetermined period, for example, on a specified date, is low.
  • the prediction unit 12 predicts seizures using a trained model generated using machine learning technology and given training data.
  • the prediction unit 12 inputs information acquired by the acquisition unit 11 into the trained model, receives an estimated result of whether or not a seizure has occurred at a specified time as the output of the trained model, and outputs a prediction result based on this.
  • the memory unit 13 stores the trained model in advance, and the prediction unit 12 inputs various information into the trained model stored in the memory unit 13 and receives the presence or absence of a seizure as an output. The procedure for generating the trained model will be described later.
  • the memory unit 13 stores the subject information, log information, environmental information, biometric information, etc. described above. Subject information is pre-registered for each subject before log information and other information is accepted. The memory unit 13 then accepts log information, environmental information, biometric information, etc., along with the subject ID from the registered subject at any time, and stores this information in association with the subject information. In this way, various types of information are linked to the subject and registered along with the date and time, and can be used to predict seizures.
  • the storage unit 13 receives and registers various types of information as needed, and the acquisition unit 11 acquires information on the timing required for seizure prediction from that information and passes it on to the prediction unit 12.
  • the procedure for acquiring information can be changed as desired.
  • various types of information can be stored in the user terminal 2, and the user terminal 2 can transmit the information required for the specified timing required for prediction to the seizure prediction device 1 each time.
  • the storage unit 13 also stores the trained model described below.
  • each device may be located in another device, and various functions may be realized by multiple devices working together.
  • the functions of the seizure prediction device 1 and the user terminal 2 may be provided in a single device, and the input and reception of various information such as log information and the prediction of seizures may be performed by a single device.
  • machine learning is performed on a different model for each seizure classification, and a trained model is generated for each seizure classification. That is, the seizure prediction device 1 generates a trained model that has learned only training data related to focal seizures, a trained model that has learned only training data related to generalized seizures, and a trained model that has learned only training data related to psychogenic non-epileptic seizures, and stores them in the memory unit 13.
  • any known classification model can be used.
  • models such as logistic regression, k-nearest neighbor (KNN), decision tree, support vector machine (SVM), and artificial neural network (ANN) can be used. Note that all of these models are widely used in classification problems, and those skilled in the art will understand how they work, so a detailed explanation will be omitted.
  • seizures are predicted using a prediction model that outputs a prediction result of whether or not a seizure will occur.
  • the prediction model in this embodiment uses, as explanatory variables, subject information as well as log information, environmental information, and biological information within a certain range prior to the specified period (or specified time) for which the prediction is to be made, to estimate whether or not a seizure will occur within the specified period, and outputs the result.
  • the prediction unit 12 then receives the estimation result and outputs the prediction result based on it.
  • pairs of information used as input (explanatory variables) for the model and seizure information that serves as training data for the output are obtained as learning data.
  • Machine learning techniques are widely used, and those skilled in the art will be able to understand the learning procedure, so a detailed explanation will be omitted.
  • a model is generated that outputs a prediction result of whether or not a seizure will occur on a specified day, using as explanatory variables subject information, the seizure interval determined based on seizure information over a specified period, and behavioral information, emotional information, environmental information, and biological information from the day before the specified day. Therefore, the training data obtained includes a set of subject information (particularly attributes, diagnostic results, and psychological evaluation results), seizure information for a specific subject on a certain day, seizure information for the most recent specified period (or the period since the previous seizure), behavioral information, emotional information, environmental information, and biological information from the day before.
  • a “day” in this embodiment begins when the "subject (target person)" wakes up, and a day here refers to the period from when the subject wakes up to when they wake up the next day.
  • training data data from multiple different subjects is used as training data. While it is possible to generate a trained model specifically for a single subject, in this embodiment, from the perspective of ensuring a sufficient amount of training data, it is decided to train pairs of subject information, log information, environmental information, and biological information, as well as seizure information, for multiple subjects assigned the same seizure classification.
  • a trained model may be generated by performing machine learning using a dataset that includes training data from the subject in addition to the training data described above. Also, for example, a trained model may be generated by prioritizing training data from times of day when the subject is most likely to experience seizures.
  • a trained model may first be generated using training data on multiple subjects other than the target individual, and then the model may be further trained using additional training data using the target individual, thereby adjusting the model to suit the target individual.
  • the explanatory variables can be not only seizure information, behavioral information, or emotional information, but also any combination of seizure information, behavioral information, emotional information, environmental information, biological information, and subject information.
  • the period of time for which the information used as an explanatory variable is to be a certain period or longer is preferable in order to reflect the subject's seizure tendencies.
  • the information used for adjustments to suit the subject (seizure information, behavioral information, or emotional information) be for a period longer than the period of the information (seizure information, behavioral information, or emotional information) used in the aforementioned learning involving multiple subjects.
  • learning is performed using information from the previous day as the explanatory variable and the presence or absence of a seizure on a specified day as the objective variable
  • additional learning can be performed using information from one month prior to the specified day as the explanatory variable and the presence or absence of a seizure on the specified day as the objective variable.
  • all available past data can be used without any particular time period being limited. In this way, sufficient information about the subject's seizures can be learned, enabling highly accurate predictions that better reflect the subject's seizure tendencies.
  • seizures are predicted using the trained models that correspond to the three seizure classifications of focal seizures, generalized seizures, and psychogenic non-epileptic seizures and that have been further adjusted for each subject using the method described above.
  • prediction can be performed by any method based on the correlation between seizure information, behavioral information, or emotional information and the presence or absence of a seizure, and the prediction method is not limited to this.
  • FIG. 3 is a flowchart showing the processing steps for predicting a seizure by the prediction unit 12.
  • the acquisition unit 11 acquires information to be used for predicting a seizure.
  • the acquired information includes subject information, log information about the subject, environmental information, and biological information, which are the same types of information used to generate the trained model.
  • the timing of the log information, environmental information, and biological information may also be acquired within the ranges used as explanatory variables in generating the trained model.
  • the user terminal 2 transmits log information, environmental information, and biological information for each subject as needed and stores them in the memory unit 13.
  • the acquisition unit 11 acquires information for the required period from the memory unit 13 depending on the period to be predicted.
  • the acquisition unit 11 acquires log information, environmental information, and biological information, as well as subject information, for the same period as the period used for the aforementioned adjustment to suit the subject.
  • the acquisition period may be shorter than the period used for adjusting the model to suit the subject.
  • the acquisition unit 11 acquires log information, environmental information, and biological information for the day before the specified day, as well as subject information, to estimate the presence or absence of a seizure on a specified day using information from the previous day.
  • step S2 the prediction unit 12 receives the information acquired in step S1 from the acquisition unit 11 and inputs it into the trained model.
  • the prediction unit 12 references the subject ID and inputs subject information (particularly attributes, diagnosis results and psychological evaluation results), log information, environmental information and biological information into the corresponding trained model. Note that if adjustment for each subject is not made, the prediction unit 12 references the seizure classification in the subject's subject information and inputs subject information (particularly attributes, diagnosis results and psychological evaluation results), log information, environmental information and biological information into the trained model corresponding to that seizure classification.
  • the prediction unit 12 receives the probability that a seizure will occur in the subject on the specified day and the probability that a seizure will not occur in the subject on the specified day as the output of the trained model.
  • the prediction unit 12 outputs the higher probability as the prediction result.
  • the user terminal 2 receives the output from the prediction unit 12 and outputs the result to the subject by displaying it on a screen, speaking, or the like.
  • the output format may be determined arbitrarily, but for example, the user terminal 2 may display the prediction result as a symbol, such as a "sunny” symbol if the prediction result is that a seizure will occur, or a "rainy” symbol if the prediction result is that a seizure will not occur.
  • the seizure prediction device of this embodiment can predict seizures based on easily available information, such as behavioral information or emotional information.
  • the seizure prediction method of the present invention is a method in which a computer executes the above-mentioned procedures, or a method in which any subject executes the above-mentioned procedures.
  • the seizure prediction method of the present invention is intended solely to provide a subject with information to help guide their behavior by predicting seizures, and is not intended for diagnosis or treatment.
  • the seizure prediction device of this embodiment can predict the presence or absence of a seizure within a certain range, such as the current day, rather than just before the seizure occurs.
  • a seizure was actually predicted using the seizure prediction device, seizure prediction program, and seizure prediction method of the present invention.
  • trained models were generated for each seizure classification using the log information, environmental information, and biological information of multiple subjects using the method described in ⁇ 3. Training process>.
  • the subjects were 12 epilepsy patients aged 18 years or older who required hospitalization for detailed examination to diagnose epilepsy and determine treatment options.
  • the data collection period was 667 days, of which seizures associated with brain activity or mental activity were observed on 82 days.
  • explanatory variables including behavioral information such as sleep duration, wake-up time, average daily movement speed, and maximum daily movement speed; emotional information input by the subject at any time; most recent seizure information indicating the date and time of the previous seizure; biometric information measured by a wristwatch-type wearable device; and attribute information indicating the subject's attributes. Note that the period since the previous seizure may be used instead of or in addition to the most recent seizure information.
  • explanatory variables are as shown in the above-mentioned embodiment. In this example, learning is performed for each seizure classification to generate a trained model, and no adjustments are made to suit each subject.
  • LightGBM Light Gradient Boosting Machine
  • LightGBM is a supervised learning method that classifies explanatory variables according to the target variable.
  • Example 2 the inventors generated trained models under different detailed conditions, added training data, and conducted experiments to confirm accuracy. Below, we will explain the differences between Example 2 and Example 1. Explanations of parts common to Example 1 will be omitted.
  • trained models were generated for each seizure classification using the log information, environmental information, and biological information of multiple subjects using the method described in ⁇ 3. Training process>.
  • the subjects were 23 epilepsy patients aged 18 years or older who required hospitalization for detailed examination to diagnose epilepsy and determine treatment options.
  • the data collection period was 1,590 days, of which seizures associated with brain activity or mental activity were observed on 126 days.
  • 270 items were used as explanatory variables, including behavioral information, emotional information, seizure information, biological information, and attribute information. Specifically, from the 368 items used in Example 1, items that were substantially overlapping or meaningless on their own were excluded or combined, and items indicating the presence or absence of abnormalities in MRI scans and medication history were added.
  • Example 2 adjustments were made for missing data and imbalanced data during learning. Specifically, median imputation was performed for missing data. Furthermore, based on the frequency of seizures, the learning data was imbalanced data in that there were more days when no seizures occurred than days when seizures occurred. Therefore, in Example 2, adjustments for imbalanced data were made by oversampling.
  • Example 2 Compared to the results of Example 1, precision was particularly improved. Here, precision and recall are in a trade-off relationship. In Example 2, precision and recall were well balanced at a high level, and the F-value, which indicates the harmonic mean of precision and recall, was improved compared to Example 1. From this, it can be said that accuracy was improved by adjusting explanatory variables, filling in missing data, oversampling, and other adjustments.
  • Seizure prediction system 1 Seizure prediction device 2: User terminal 11: Acquisition unit 12: Prediction unit 13: Storage unit 10: Information processing device 101: Control unit 102: Storage unit 103: Communication unit 9: Terminal device 901: Control unit 902: Storage unit 903: Communication unit 904: Input unit 905: Output unit

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Biomedical Technology (AREA)
  • Pathology (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Epidemiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Medical Treatment And Welfare Office Work (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

La présente invention aborde le problème de la fourniture d'une nouvelle technique qui est très utile pour prédire des crises. Un programme selon la présente invention amène un ordinateur à fonctionner en tant qu'unité d'acquisition et unité de prédiction, l'unité d'acquisition acquérant des informations de journal comprenant des informations de comportement indiquant un comportement d'un sujet et/ou des informations d'émotion entrées en fonction d'une opinion subjective du sujet et/ou des informations de crise relatives à un enregistrement de crises qui se sont produites chez le sujet ; et l'unité de prédiction prédit des crises associées à l'activité cérébrale ou à l'activité mentale du sujet en fonction des informations de journal.
PCT/JP2025/005202 2024-05-20 2025-02-17 Programme de prédiction de crise, support de stockage, dispositif de prédiction de crise et procédé de prédiction de crise Pending WO2025243619A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2024-081762 2024-05-20
JP2024081762A JP7534745B1 (ja) 2024-05-20 2024-05-20 発作予測プログラム、記憶媒体、発作予測装置および発作予測方法

Publications (1)

Publication Number Publication Date
WO2025243619A1 true WO2025243619A1 (fr) 2025-11-27

Family

ID=92261394

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2025/005202 Pending WO2025243619A1 (fr) 2024-05-20 2025-02-17 Programme de prédiction de crise, support de stockage, dispositif de prédiction de crise et procédé de prédiction de crise

Country Status (2)

Country Link
JP (1) JP7534745B1 (fr)
WO (1) WO2025243619A1 (fr)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190246990A1 (en) * 2017-07-25 2019-08-15 Seer Medical Pty Ltd Methods and systems for forecasting seizures
WO2019221252A1 (fr) * 2018-05-17 2019-11-21 一般社団法人認知症高齢者研究所 Dispositif de traitement d'informations, procédé de traitement d'informations et programme
JP2020523283A (ja) * 2017-05-16 2020-08-06 エボゲン，インコーポレーテッド 発作およびてんかんの検出のためのバイオマーカーおよび方法
JP2022512503A (ja) * 2018-12-13 2022-02-04 リミナル サイエンシズ インコーポレイテッド 注釈が付けられた信号データに基づきトレーニング済の統計モデルを利用するデバイス用のシステムおよび方法
JP2022098175A (ja) * 2020-12-21 2022-07-01 ノックオンザドア株式会社 症状の情報を管理するプログラム、症状の情報を管理する装置、及び症状の情報を管理する方法
US20230141496A1 (en) * 2020-04-03 2023-05-11 The Children's Medical Center Corporation Computer-based systems and devices configured for deep learning from sensor data non-invasive seizure forecasting and methods thereof
US20230248302A1 (en) * 2022-02-09 2023-08-10 The Alfred E. Mann Foundation For Scientific Research Systems and methods for vagus nerve monitoring and stimulation
CN117373663A (zh) * 2023-10-08 2024-01-09 重庆邮电大学 基于多特征选择与伪三维网络的癫痫预测装置

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2020523283A (ja) * 2017-05-16 2020-08-06 エボゲン，インコーポレーテッド 発作およびてんかんの検出のためのバイオマーカーおよび方法
US20190246990A1 (en) * 2017-07-25 2019-08-15 Seer Medical Pty Ltd Methods and systems for forecasting seizures
WO2019221252A1 (fr) * 2018-05-17 2019-11-21 一般社団法人認知症高齢者研究所 Dispositif de traitement d'informations, procédé de traitement d'informations et programme
JP2022512503A (ja) * 2018-12-13 2022-02-04 リミナル サイエンシズ インコーポレイテッド 注釈が付けられた信号データに基づきトレーニング済の統計モデルを利用するデバイス用のシステムおよび方法
US20230141496A1 (en) * 2020-04-03 2023-05-11 The Children's Medical Center Corporation Computer-based systems and devices configured for deep learning from sensor data non-invasive seizure forecasting and methods thereof
JP2022098175A (ja) * 2020-12-21 2022-07-01 ノックオンザドア株式会社 症状の情報を管理するプログラム、症状の情報を管理する装置、及び症状の情報を管理する方法
US20230248302A1 (en) * 2022-02-09 2023-08-10 The Alfred E. Mann Foundation For Scientific Research Systems and methods for vagus nerve monitoring and stimulation
CN117373663A (zh) * 2023-10-08 2024-01-09 重庆邮电大学 基于多特征选择与伪三维网络的癫痫预测装置

Also Published As

Publication number Publication date
JP2025175585A (ja) 2025-12-03
JP7534745B1 (ja) 2024-08-15

Similar Documents

Publication Publication Date Title
US20250139578A1 (en) Method to increase efficiency, coverage, and quality of direct primary care
US11875895B2 (en) Method and system for characterizing and/or treating poor sleep behavior
US10265028B2 (en) Method and system for modeling behavior and heart disease state
US8622900B2 (en) Calculating and monitoring the efficacy of stress-related therapies
US8622901B2 (en) Continuous monitoring of stress using accelerometer data
US8725462B2 (en) Data aggregation platform
US9189599B2 (en) Calculating and monitoring a composite stress index
US9173567B2 (en) Triggering user queries based on sensor inputs
US20180096738A1 (en) Method for providing health therapeutic interventions to a user
US8540629B2 (en) Continuous monitoring of stress using a stress profile created by renal doppler sonography
US8529447B2 (en) Creating a personalized stress profile using renal doppler sonography
US20210391083A1 (en) Method for providing health therapeutic interventions to a user
EP3218836B1 (fr) Procédé et processeur permettant de déterminer la santé d'un individu
Kikhia et al. Utilizing ambient and wearable sensors to monitor sleep and stress for people with BPSD in nursing homes
US20210358628A1 (en) Digital companion for healthcare
JP2018509270A (ja) 精神的苦痛モニタリングシステム
US20210121082A1 (en) Method and a processor for determining health of an individual
JP7534745B1 (ja) 発作予測プログラム、記憶媒体、発作予測装置および発作予測方法
US20250078984A1 (en) Information processing apparatus, information processing method, and non-transitory recording medium
US20250174337A1 (en) Artificial intelligence-based personalized predictive treatment system
WO2023234317A1 (fr) Système de traitement d'informations, procédé de traitement d'informations, programme et support d'enregistrement
JP2025155464A (ja) ヘルスケアアプリの情報提供システム
WO2021260846A1 (fr) Dispositif de génération vocale, procédé de génération vocale et programme de génération vocale