Waveform generation identification method and program

By using a trained model to analyze distribution information from waveform data, the method accurately identifies sensors for IEDs, enhancing the precision of dipole estimation in epilepsy diagnosis.

JP2024166833AInactive Publication Date: 2024-11-29RICOH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
JP2023083209
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-19
Publication Date
2024-11-29
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing methods for identifying the sensor where characteristic waveform information of interictal epileptiform discharge (IED) appears in magnetoencephalography and electroencephalography data are inefficient and inaccurate, particularly due to high sampling frequencies and large datasets, leading to difficulties in manual identification and deviations from doctor's judgment criteria.

Method used

A method and program that utilize a trained model to identify the sensor by calculating distribution information from waveform data, using isomagnetic diagrams and sensor selection areas based on expert-labeled training data to accurately specify the sensors where IEDs occur.

Benefits of technology

This approach allows for precise identification of sensors exhibiting characteristic waveform information, aligning with doctor's judgment criteria and improving the reliability of dipole estimation in epilepsy diagnosis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024166833000001_ABST
    Figure 2024166833000001_ABST
Patent Text Reader

Abstract

To provide a waveform generation identification method and a program that allow a sensor in which characteristic waveform information on IED to be identified accurately, out of a plurality of sensors that has measured waveform data.SOLUTION: A waveform generation identification method includes: an acquisition step for acquiring waveform data on a biological signal measured by a plurality of sensors; a calculation step for calculating distribution information indicating distribution of values of the biological signals on the basis of the waveform data at the time when characteristic waveform information on IED appears of the acquired waveform data; and an identification step for identifying a sensor constituting a region of a dipole pattern on the basis of a selection region by giving, as input, the distribution information calculated in the calculation step to a model that has learned, as teacher data, the distribution information indicating the distribution of the values of the biological signals to which information on the sensor selected in analysis is imparted, and obtaining, as output, the selection region indicating the region of the dipole pattern.SELECTED DRAWING: Figure 4
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a waveform generation specifying method and a program. [Background technology]

[0002] In clinical epilepsy diagnosis using magnetoencephalography and electroencephalography, a technique called the equivalent current dipole method is used to evaluate the localization of epileptic lesions in the brain. In the equivalent current dipole method, the current source (dipole) that generates the magnetic field measured on the scalp is estimated. In order to estimate the dipole, it is necessary to identify the time when characteristic waveform information called interictal epileptic discharge (IED) appears and the sensor where the waveform information appears from the time series data detected by multiple sensors. However, although it is very important to analyze characteristic waveforms in magnetoencephalography and electroencephalography, the sampling frequency and the number of sensors during measurement are increasing due to technological advances, and as a result, the time required to visually search for characteristic waveform information of IEDs also tends to increase. Currently, these operations are performed manually by doctors, but because the data of magnetoencephalography is huge, it is difficult to manually and accurately identify the time and sensor for the characteristic waveform information of each IED.

[0003] As a technology for identifying the time and sensor at which an IED's characteristic waveform occurs, a technology has been disclosed in which a machine learning learning model is used to infer an IED's probability map and select the time and sensor at which the IED's characteristic waveform information appears (for example, Patent Document 1). Summary of the Invention [Problem to be solved by the invention]

[0004] However, in the technology described in Patent Document 1, the sensor selection is based on a machine learning model, which has the problem that it may not necessarily match the doctor's judgment criteria. It is desirable to realize an accurate sensor selection method that is close to the doctor's judgment criteria so that doctors can make more intuitive judgments.

[0005] The present invention has been made in consideration of the above, and aims to provide a waveform generation identification method and program that can accurately identify a sensor that exhibits characteristic waveform information of an IED among multiple sensors that measure waveform data. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems and achieve the object, the present invention is characterized by comprising an acquisition step of acquiring waveform data of biosignals measured by a plurality of sensors; a calculation step of calculating distribution information indicating a distribution of values ​​of the biosignals based on the waveform data at a time when characteristic waveform information of an IED (Interictal Epileptiform Discharge) appears among the acquired waveform data; and an identification step of providing the distribution information calculated in the calculation step as an input to a model trained using as teacher data the distribution information indicating the distribution of values ​​of the biosignals to which information of the sensors selected in analysis has been added, obtaining as an output a selection area indicating an area of ​​a dipole pattern, and identifying the sensors that constitute the area of ​​the dipole pattern based on the selection area. Effect of the Invention

[0007] According to the present invention, it is possible to accurately identify, from among a plurality of sensors that measure waveform data, a sensor that exhibits waveform information characteristic of an IED. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram showing an example of the overall configuration of a biological signal measuring system according to a first embodiment. [Diagram 2] FIG. 2 is a diagram illustrating an example of a functional block configuration of the server according to the first embodiment. [Diagram 3] FIG. 3 is a diagram illustrating an example of a hardware configuration of the information processing device according to the first embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of a functional block configuration of the information processing device according to the first embodiment. [Diagram 5] FIG. 5 is a diagram showing an example of an isomagnetic field diagram at a specific time. [Figure 6] FIG. 6 is a diagram showing an example of a sensor selection region corresponding to a dipole pattern. [Figure 7] FIG. 7 is a flowchart showing an example of the flow of the learning process of the information processing device according to the first embodiment. [Figure 8] FIG. 8 is a flowchart showing an example of the flow of a dipole estimation process in the information processing device according to the first embodiment. [Figure 9] FIG. 9 is a flowchart showing an example of the flow of a dipole pattern detection process of the information processing device according to the first embodiment. [Figure 10] FIG. 10 is a diagram illustrating an example of a functional block configuration of a server according to the second embodiment. As illustrated in FIG. [Figure 11] FIG. 11 is a diagram showing an example of a search range for an onset in waveform data. [Figure 12] FIG. 12 is a flowchart showing an example of the flow of a dipole estimation process of the information processing device according to the second embodiment. [Figure 13] FIG. 13 is a flowchart showing an example of the flow of a dipole pattern detection process of the information processing device according to the third embodiment. [Figure 14] FIG. 14 is a diagram showing an example of a plurality of sensor selection regions corresponding to a dipole pattern. [Figure 15] FIG. 15 is a flowchart showing an example of the flow of a dipole estimation process of the information processing device according to the fourth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, with reference to the drawings, an embodiment of a waveform generation specification method and a program according to the present invention will be described in detail. Furthermore, the present invention is not limited to the following embodiment, and the components in the following embodiment include those that a person skilled in the art can easily conceive, those that are substantially the same, and those that are within the scope of what is called equivalent. Furthermore, various omissions, substitutions, modifications, and combinations of the components can be made without departing from the spirit of the following embodiment.

[0010] [First embodiment] (Overall configuration of biosignal measurement system) Fig. 1 is a diagram showing an example of the overall configuration of a biosignal measurement system according to the first embodiment. Fig. 2 is a diagram showing an example of the functional block configuration of a server according to the first embodiment. The overall configuration of a biosignal measurement system 1 according to this embodiment will be described with reference to Figs. 1 and 2.

[0011] The biosignal measurement system 1 is a system that measures and displays multiple types of biosignals (e.g., magneto-encephalography (MEG) signals and electro-encephalography (EEG) signals) of a subject from a specific source (body part). Note that the biosignals to be measured are not limited to magneto-encephalography (MEG) signals and electro-encephalography (EEG) signals, but may be, for example, electrical signals generated in response to cardiac activity (electrical signals that can be expressed as an electrocardiogram).

[0012] As shown in Fig. 1, the biosignal measurement system 1 includes a measurement device 3 that measures one or more biosignals of a subject, a server 40 that records one or more types of biosignals measured by the measurement device 3, and an information processing device 50 that is a biosignal display device that analyzes one or more types of biosignals recorded in the server 40. The measurement device 3 is a magnetoencephalograph that measures, for example, a brain magnetic field or a magnetoencephalogram (an example of a biosignal) generated at the timing of application of a stimulus. Note that, although the server 40 and the information processing device 50 are depicted separately in Fig. 1, for example, at least a part of the functions of the server 40 may be incorporated into the information processing device 50.

[0013] In the example of FIG. 1, the subject (subject) lies on his / her back on the measurement table 4 with electrodes (or sensors) for measuring electroencephalograms attached to his / her head, and places his / her head in a recess 32 of a Dewar 31 of the measurement device 3. The Dewar 31 is a holding container in an extremely low temperature environment using liquid helium. A number of magnetic sensors (e.g., SQUID (Superconducting Quantum Interference Device) sensors) for measuring magnetoencephalograms are arranged inside the recess 32 of the Dewar 31. The measurement device 3 collects electroencephalogram signals from the electrodes and magnetoencephalogram signals from the magnetic sensors, and outputs time-series waveform data (hereinafter, sometimes simply referred to as waveform data) including the collected electroencephalogram signals and magnetoencephalogram signals to a server 40. The waveform data output to the server 40 is read by an information processing device 50 for display and analysis. In general, the Dewar 31 and the measurement table 4 incorporating the magnetic sensors are arranged in a magnetically shielded room, but for convenience, the magnetically shielded room is not shown in FIG. 1.

[0014] The information processing device 50 is a device that analyzes waveform data of MEG signals from multiple magnetic sensors and waveform data of EEG signals from multiple electrodes. For example, the information processing device 50 displays waveform data of MEG signals and EEG signals in synchronization on the same time axis. Here, the EEG signal is a signal that represents the electrical activity of nerve cells (the flow of ionic charges that occurs in the dendrites of neurons during synaptic transmission) as a voltage value between electrodes. Also, the MEG signal is a signal that represents minute electric field fluctuations caused by electrical activity of the brain.

[0015] As described above, in the conventional technology, the time when characteristic waveform information of an IED appears and the selection of the sensor are simultaneously realized by a model learned by machine learning. Among such models, for example, a segmentation model is a simple model for selecting the time and the sensor for dipole estimation, and the cost of learning and inference is not high. However, since it is machine learning, the inference result may not be desirable depending on the learning data used for learning. The selection of the sensor, which is usually performed in a rectangular shape, may result in a distorted selection area and a hollow selection area, which may deviate from the experience of a doctor who performs normal work, and there is a risk of damaging the reliability of the system. Above all, the dipole estimation solution may be estimated in an inappropriate location due to the influence of noise. In addition, if there are many cases in which sensor selection is not performed in the learning data, the inference result will also not select a sensor. The ACMEGS (American Clinical MEG Society) guidelines (see, for example, Non-Patent Document 1) recommend the selection of a sensor during dipole estimation for more accurate epilepsy focus diagnosis. The guidelines also state that when estimating a dipole, it is necessary to determine an appropriate time between the start and peak of an abnormal signal, particularly an IED. Here, the start refers to the part where the characteristic mountain-shaped waveform of an IED begins, and the peak refers to the top of the mountain-shaped waveform. Doctors calculate appropriate times between the start and peak based on the reliability of the current dipole and the amplitude of the magnetic field time series. In cases where an abnormal signal moves, it is necessary to select a sensor at each time. Although the method described in the above-mentioned Patent Document 1 also allows dynamic selection of a sensor at each time, there is a possibility that sensor selection can be improved by using the distribution of the magnetic field at each time. The information processing device 50 according to this embodiment improves the accuracy of sensor selection by using the distribution of the magnetic field at each time.

[0016] As shown in FIG. 2, the server 40 includes a data acquisition unit 401 and a data storage unit 402.

[0017] The data acquisition unit 401 is a functional unit that periodically acquires waveform data such as MEG signals and EEG signals measured by the measurement device 3. The waveform data includes time series data of each MEG signal measured by a plurality of magnetic sensors in the Dewar 31 of the measurement device 3, and time series data of each signal measured by a plurality of electrodes for measuring EEG attached to the head of the subject (person to be measured).

[0018] The data storage unit 402 is a functional unit that stores the waveform data acquired from the measurement device 3 .

[0019] 1 shows a configuration in which the measuring device 3 and the server 40 are directly connected, and the server 40 and the information processing device 50 are directly connected, but they may be configured to communicate data with each other via a network. The network connection method may be wired or wireless. The server 40 may also be configured to be a server on a network and connected to the cloud, for example.

[0020] (Hardware configuration of information processing device) 3 is a diagram showing an example of a hardware configuration of an information processing device according to the first embodiment. The hardware configuration of an information processing device 50 according to the present embodiment will be described with reference to FIG.

[0021] As shown in FIG. 3, the information processing device 50 includes a CPU (Central Processing Unit) 101, a RAM (Random Access Memory) 102, a ROM (Read Only Memory) 103, an auxiliary storage device 104, a network I / F 105, an input device 106, and a display device 107.

[0022] The CPU 101 is a calculation device that controls the overall operation of the information processing device 50. The RAM 102 is a volatile storage device that is used as a work area for the CPU 101. The ROM 103 is a non-volatile storage device that stores programs for the information processing device 50.

[0023] The auxiliary storage device 104 is a storage device such as a hard disk drive (HDD) or a solid state drive (SSD) that stores various data, programs, and the like.

[0024] The network I / F 105 is an interface for communicating data with an external device such as the server 40 via a network. The network I / F 105 is, for example, a network interface card (NIC) that supports Ethernet (registered trademark) and is capable of wired or wireless communication in accordance with TCP (Transmission Control Protocol) / IP (Internet Protocol) or the like.

[0025] The input device 106 is a touch panel input function, a mouse, a keyboard, etc. for selecting letters, numbers, various instructions, and moving a cursor.

[0026] The display device 107 is a display configured with liquid crystal or organic EL (Electro-Luminescence) or the like, which displays various information such as a cursor, a menu, a window, characters, or an image.

[0027] The above-mentioned CPU 101, RAM 102, ROM 103, auxiliary storage device 104, network I / F 105, input device 106 and display device 107 are communicably connected to one another via a bus 108 such as an address bus and a data bus.

[0028] It should be noted that the hardware configuration of the information processing device 50 shown in FIG. 3 is merely an example, and it is not necessary for the information processing device 50 to include all of the components shown in FIG. 3, or it may include other components.

[0029] (Configuration and operation of functional blocks of information processing device) Fig. 4 is a diagram showing an example of the configuration of functional blocks of the information processing device according to the first embodiment. Fig. 5 is a diagram showing an example of an isomagnetic diagram at a specific time. Fig. 6 is a diagram showing an example of a sensor selection region corresponding to a dipole pattern. The configuration and operation of the functional blocks of the information processing device 50 according to this embodiment will be described with reference to Figs. 4 to 6.

[0030] As shown in FIG. 4, the information processing device 50 includes an acquisition unit 501, an IED detection unit 502, a calculation unit 503, a dipole pattern detection unit 504, and a dipole estimation unit 505.

[0031] The acquiring unit 501 is a functional unit that acquires waveform data of magnetoencephalographic signals and electroencephalographic signals from the server 40 via the network I / F 105 .

[0032] The IED detection unit 502 is a functional unit that detects the time when characteristic waveform information of an IED, called a spike, appears in the waveform data acquired by the acquisition unit 501. For the detection of the time by the IED detection unit 502, for example, the method of Patent Document 1 described above can be adopted.

[0033] The calculation unit 503 is a functional unit that calculates an isomagnetic field map (an example of distribution information) in which regions of equal magnetic field values ​​are expressed with the same luminance value, based on waveform data at the time detected by the IED detection unit 502.

[0034] An example of an isomagnetic field map calculated by the calculation unit 503 is shown in Fig. 5. In this way, an isomagnetic field map is usually calculated based on the magnetic field values ​​measured by a plurality of magnetoencephalograph sensors 602 arranged in a plane on a diagram of a human head 601. The isomagnetic field map as shown in Fig. 5 includes a source portion 603 which is a region where the magnetic field value is larger than the surroundings, and a sink portion 604 which is a region where the magnetic field value is smaller than the surroundings. That is, a magnetic field pattern including a source portion and a sink portion of the magnetic field (hereinafter, sometimes referred to as a dipole pattern) is observed in the isomagnetic field map at the time when the characteristic waveform information of an IED appears.

[0035] The dipole pattern detection unit 504 is a functional unit that detects a dipole pattern region from the isomagnetic map calculated by the calculation unit 503. Specifically, the dipole pattern detection unit 504 determines, for example, a rectangular region (hereinafter, may be referred to as a sensor selection region) surrounding the sensors that form the dipole pattern region from the isomagnetic map calculated by the calculation unit 503, thereby identifying the sensors that form the dipole pattern region. That is, the sensors included in the sensor selection region are the sensors identified by the dipole pattern detection unit 504. An example of the sensor selection region determined by the dipole pattern detection unit 504 is shown in FIG. 6. In FIG. 6, the sensor selection region determined by the dipole pattern detection unit 504 is shown as a sensor selection region 703. That is, the sensor 702 included in the sensor selection region 703 is the sensor identified by the dipole pattern detection unit 504, and the sensor not included in the sensor selection region 703 is the sensor not identified by the dipole pattern detection unit 504.

[0036] As shown in FIG. 4, the dipole pattern detection unit 504 has an area detection unit 5041, an aggregation unit 5042, a determination unit 5043, and a sensor identification unit 5044.

[0037] The area detection unit 5041 is a functional unit that provides the isomagnetic field calculated by the calculation unit 503 as an input to a trained model described later, and detects a sensor selection area as an output of the trained model. The trained model uses waveform data obtained by dipole analysis performed by a doctor or other specialist in epilepsy clinical practice, and trains the isomagnetic field of the waveform data and information of the sensor selected during the analysis as training data (teaching data). If a sensor selection area exists as information on the selected sensor, it is used, and if no sensor selection area exists, the sensor selection area may be restored from the selected sensor. That is, the model is trained using the isomagnetic field with the sensor selection area attached as a label as teaching data. Then, the trained model generated by this training inputs the isomagnetic field and outputs the sensor selection area. Specific processing of training such a model will be described in detail in FIG. 7.

[0038] In the example shown in Fig. 5, an isomagnetic field map in which three-dimensional sensors are arranged on a two-dimensional plane is shown, but it is also possible to perform learning with the three-dimensional sensor arrangement as it is. In that case, by performing learning using as input a three-dimensional spatial isomagnetic field map in which a sensor selection region, which is an ellipsoid or polyhedron region in which the selected sensor is included, is added as a label, it is possible to generate a trained model that similarly outputs a sensor selection region. However, in the following, the description will be continued assuming that the isomagnetic field map is two-dimensional.

[0039] In addition, as a model to be trained, a deep learning method such as EfficientDet (see, for example, Non-Patent Document 2) or YOLO (see, for example, Non-Patent Document 3), a method of training a dipole pattern template, or a method that combines machine learning methods such as HOG (Histogram of Oriented Gradients) features or SVM (Support Vector Machine) can also be used.

[0040] In the above description, a rectangular region is used as the sensor selection region, but an ellipse or a polygon, etc., may also be used as the sensor selection region.

[0041] Also, some doctors may perform dipole estimation using sensors included in multiple sensor selection areas. Even in this case, the multiple sensor selection areas can be used for learning as they are. Note that when using a sensor selection area other than a single rectangle, an object detection model such as EfficientDet cannot handle it, so it is possible to operate in the same way as object detection by using Semantic Segmentation such as U-Net (see, for example, Non-Patent Document 4).

[0042] The aggregation unit 5042 is a functional unit that, when a plurality of sensor selection regions are detected from one isomagnetic map by the region detection unit 5041, removes and aggregates the plurality of sensor selection regions. Specifically, when a plurality of sensor selection regions overlap, the aggregation unit 5042 aggregates the sensor selection regions into one sensor selection region, and when the detection confidence of the detected sensor selection region is low, the aggregation unit 5042 removes the sensor selection region. Note that the aggregation unit 5042 may perform at least one of aggregation and removal. Examples of algorithms for removing and aggregating a plurality of sensor selection regions include Non-Maximum Suppression (NMS), Non-Maximum Weighted (NMW), Weighted Boxes Fusion (WBF), and the like (see, for example, Non-Patent Document 5).

[0043] The determination unit 5043 is a functional unit that determines whether or not a sensor selection area remains as a result of the removal / aggregation of the sensor selection areas performed by the aggregation unit 5042 .

[0044] The sensor identifying unit 5044 is a functional unit that, when the determining unit 5043 determines that a sensor selection region remains, identifies the sensors included in the sensor selection region as sensors that form a region of a dipole pattern.

[0045] The dipole estimation unit 505 is a functional unit that performs dipole estimation using the time at which characteristic waveform information of the IED identified by the IED detection unit 502 appears and the waveform data corresponding to the sensor identified by the dipole pattern detection unit 504.

[0046] The above-mentioned acquisition unit 501, IED detection unit 502, calculation unit 503, dipole pattern detection unit 504, and dipole estimation unit 505 are realized by executing a program by CPU 101 shown in Fig. 3. Note that some or all of the functional units of the acquisition unit 501, IED detection unit 502, calculation unit 503, dipole pattern detection unit 504, and dipole estimation unit 505 may be realized by a hardware circuit (integrated circuit) such as a field-programmable gate array (FPGA) or an application specific integrated circuit (ASIC) instead of a software program.

[0047] Note that the functions of each functional unit of the information processing device 50 shown in Fig. 4 are conceptually shown, and the configuration is not limited to this. For example, the multiple functional units illustrated as independent functional units in the information processing device 50 shown in Fig. 4 may be configured as one functional unit. On the other hand, the function of one functional unit in the information processing device 50 shown in Fig. 4 may be divided into multiple functions and configured as multiple functional units.

[0048] (Learning process flow of information processing device) 7 is a flowchart showing an example of the flow of the learning process of the information processing device according to the first embodiment. The flow of the learning process of the information processing device 50 according to the present embodiment will be described with reference to FIG.

[0049] <Step S11> The acquiring section 501 acquires waveform data (here, waveform data of a magnetoencephalographic signal) that serves as teacher data for learning a model in the area detecting section 5041 of the dipole pattern detecting section 504. Then, the process proceeds to step S12.

[0050] <Step S12> The waveform data is obtained by performing dipole analysis in epilepsy clinical practice by a doctor or other specialist, and therefore includes the time when the characteristic waveform information of the IED appeared and sensor selection information. If there is no sensor selection information, the sensor selection area may be restored from the selected sensor as described above. Note that the waveform data is not limited to clinical data, and may be newly prepared for learning purposes or collected auxiliary using AI. Then, the process proceeds to step S13.

[0051] <Step S13> Then, the area detection unit 5041 registers the isomagnetic field of the waveform data to which the sensor selection area is added as a label as learning data (teacher data). Note that the isomagnetic field of the waveform data may be calculated by the function of the calculation unit 503. Then, the process proceeds to step S14.

[0052] <Step S14> Then, the area detection unit 5041 performs learning using the registered learning data (teacher data), and proceeds to step S15.

[0053] <Step S15> As a result, the area detection unit 5041 inputs the isomagnetic field diagram and generates a trained model that outputs a sensor selection area.

[0054] (Flow of dipole estimation process of information processing device) Fig. 8 is a flowchart showing an example of the flow of a dipole estimation process of the information processing device according to the first embodiment. Fig. 9 is a flowchart showing an example of the flow of a dipole pattern detection process of the information processing device according to the first embodiment. The flow of a dipole estimation process of the information processing device 50 according to this embodiment will be described with reference to Figs. 8 and 9.

[0055] <Step S21> The acquiring unit 501 acquires waveform data of the magnetoencephalographic signal from the server 40 via the network I / F 105. Then, the process proceeds to step S22.

[0056] <Step S22> The IED detection unit 502 detects a time when characteristic waveform information of an IED, called a spike, appears in the waveform data acquired by the acquisition unit 501. Then, the process proceeds to step S23.

[0057] <Step S23> The calculation unit 503 calculates an isomagnetic field diagram in which regions of equal magnetic field values ​​are expressed with the same luminance value, based on the waveform data at the time detected by the IED detection unit 502. Then, the process proceeds to step S24.

[0058] <Step S24> The dipole pattern detection unit 504 detects a dipole pattern area from the isomagnetic field map calculated by the calculation unit 503. That is, the dipole pattern detection unit 504 determines a sensor selection area surrounding the sensors that configure the dipole pattern area from the isomagnetic field map calculated by the calculation unit 503, thereby identifying the sensors that configure the dipole pattern area. Specifically, the following steps S241 to S246 are performed.

[0059] <<Step S241>> The area detection unit 5041 of the dipole pattern detection unit 504 receives the isomagnetic field calculated by the calculation unit 503. Then, the process proceeds to step S242.

[0060] <<Step S242>> The area detection unit 5041 provides the isomagnetic field as an input to the trained model, and detects the sensor selection area as an output of the trained model. Then, the process proceeds to step S243.

[0061] <<Step S243>> When a plurality of sensor selection regions are detected by the region detection unit 5041, the aggregation unit 5042 of the dipole pattern detection unit 504 removes and aggregates the plurality of sensor selection regions, and then the process proceeds to step S244.

[0062] <<Step S244>> The determination unit 5043 of the dipole pattern detection unit 504 determines whether or not a sensor selection area remains as a result of the removal and aggregation of the sensor selection areas by the aggregation unit 5042. If a sensor selection area remains (step S244: Yes), the process proceeds to step S245, and if not (step S244: No), the process proceeds to step S246.

[0063] <<Step S245>> If the determination unit 5043 determines that a sensor selection region remains, the sensor identification unit 5044 of the dipole pattern detection unit 504 identifies the sensors included in the sensor selection region as sensors that form the region of the dipole pattern. Then, step S24 ends, and the process proceeds to step S25.

[0064] <<Step S246>> Due to factors such as a distorted dipole pattern, all sensor selection areas may be removed as a result of removal and aggregation of sensor selection areas by the aggregation unit 5042. In this case, there is a high possibility that noise is included in the waveform data on which the isomagnetic field calculated by the calculation unit 503 is based, and dipole estimation is not performed, and processing moves to waveform data of IEDs at other times.

[0065] <Step S25> The dipole estimation unit 505 performs dipole estimation using the time when the characteristic waveform information of the IED identified by the IED detection unit 502 appears and the waveform data corresponding to the sensor identified by the dipole pattern detection unit 504. This ends the dipole estimation process.

[0066] As described above, in the information processing device 50 according to the present embodiment, the acquisition unit 501 acquires waveform data of the brain magnetic signal measured by a plurality of sensors, the calculation unit 503 calculates an isomagnetic field diagram showing the distribution of the values ​​of the brain magnetic signal based on the waveform data at the time when the characteristic waveform information of the IED appears among the waveform data acquired by the acquisition unit 501, and the dipole pattern detection unit 504 provides the isomagnetic field diagram calculated by the calculation unit 503 as an input to a model trained using the isomagnetic field diagram showing the distribution of the values ​​of the brain magnetic signal and the information of the sensor selected in the analysis as teacher data, obtains a sensor selection area showing the area of ​​the dipole pattern as an output, and identifies the sensor that constitutes the area of ​​the dipole pattern based on the sensor selection area. In this way, by having the model train using teacher data of the isomagnetic field diagram with the information of the sensor selected in the analysis by a specialist such as a doctor as a label, it is possible to identify a sensor close to the one selected by the specialist. Therefore, it is possible to more accurately identify the sensor in which the characteristic waveform information of the IED appears among the plurality of sensors that have measured the waveform data.

[0067] In the above, the operation of outputting the sensor selection area by the trained model using the isomagnetic field based on the waveform data of the magnetoencephalogram signal has been described, but the present invention is not limited to this. For example, in the case of the waveform data of the electroencephalogram signal, it is possible to train the model using a current distribution map in which areas of equal current values ​​are expressed with the same brightness value, to provide the current distribution map (an example of distribution information) of the waveform data of the electroencephalogram signal as an input to the trained model, and to obtain the sensor selection area as an output. In this case, the electrodes that measure the electroencephalogram signal correspond to the above-mentioned "sensor".

[0068] [Second embodiment] The biosignal measurement system 1 according to the second embodiment will be described with a focus on differences from the biosignal measurement system 1 according to the first embodiment. In the above-mentioned first embodiment, an operation using the time at which characteristic waveform information of an IED detected by the IED detection unit 502 appears has been described. As described above, the method of detecting the time by the IED detection unit 502 is to detect using the machine learning model described in Patent Document 1, and the time detected using such a model is often inferred to be the time near the peak part of the characteristic waveform of an IED. In addition, for example, in the method described in JP-A-2023-020273 (Patent Document 2), a method of searching for the peak part of the characteristic waveform of an IED is disclosed. However, depending on the epileptic lesion, the position of the dipole may not change between the time (time of the peak part, etc.) detected by the IED detection unit 502 as described above and the time at which an IED actually occurs from the epileptic lesion, but may conversely vary greatly. In this case, the part where an IED actually occurs in the waveform data is referred to as an onset, and the time of the part is referred to as an onset time. That is, the dipole pattern at the time of the peak portion may differ from the dipole pattern at the onset portion. Therefore, even if a stable dipole is obtained from the waveform data at the time of the peak portion with a good S / N ratio, even if the brain area is removed, the epilepsy seizure may not disappear. Therefore, the information that a doctor or other specialist wants to know is information about which area of ​​the brain the characteristic waveform of an IED originates from, and it is important for dipole estimation at the onset in the waveform data to diagnose epilepsy. In this embodiment, an operation of detecting the onset time by searching for the onset based on the time detected by the IED detection unit 502 and estimating the dipole from the waveform data at the onset time will be described. Note that the overall configuration of the biosignal measurement system 1 according to this embodiment and the hardware configuration of the information processing device are the same as those described in the first embodiment.

[0069] (Configuration and operation of functional blocks of information processing device) Fig. 10 is a diagram showing an example of the configuration of the functional blocks of the server according to the second embodiment. Fig. 11 is a diagram showing an example of a search range for an onset in waveform data. The configuration and operation of the functional blocks of the information processing device 50a according to the present embodiment will be described with reference to Figs. 10 and 11.

[0070] 10, the information processing device 50a includes an acquisition unit 501, an IED detection unit 502, an onset search unit 506, a calculation unit 503, a dipole pattern detection unit 504, and a dipole estimation unit 505. That is, the information processing device 50a includes an onset search unit 506 in addition to the functional units included in the information processing device 50 according to the first embodiment described above.

[0071] The onset search unit 506 is a functional unit that performs dipole estimation for at least one of the waveform data portions before and after the time detected by the IED detection unit 502, calculates index values ​​such as GoF (Goodness of Fit) of the estimated solution, confidence volume (CV), waveform amplitude, and dipole distance, and searches for onsets based on the index values. Specifically, the onset search unit 506 detects (searches for) the onset time based on the index values. At this time, the onset search unit 506 can also normalize the sensor selection area for each time using the sensor selection area for at least one of the times before and after the time detected by the IED detection unit 502. This can be expected to provide a robust effect against sudden noise or artifacts.

[0072] Here, the operation of the onset search unit 506 will be described using a specific waveform 801 with reference to FIG. 11. The waveform 801 is a waveform when a certain sensor observes characteristic waveform information of an IED. The peak of the waveform where characteristic waveform information appears in the waveform 801 is shown as a peak portion 802. As described above, the time detected by the IED detection unit 502, that is, the time detected by the method described in Patent Document 1 and JP-A-2023-020273, is detected as a time near this peak portion 802. However, since it is known that the onset, which is the part where an IED actually occurs in the waveform 801, is a time before the time of the peak portion 802, the onset search unit 506 searches for the onset in a search range 803, which is a predetermined section before the time of the peak portion 802. In this case, the onset search unit 506 performs dipole estimation with waveform data at all times included in the search range 803, calculates the above-mentioned index value, and comprehensively detects the onset time.

[0073] The calculation unit 503 calculates an isomagnetic field diagram in which regions of equal magnetic field values ​​are expressed by the same brightness value, based on the waveform data at the onset time detected by the onset search unit 506 .

[0074] The processes of the other functional units of the information processing device 50a are similar to those of the information processing device 50 according to the above-described first embodiment.

[0075] (Flow of dipole estimation process of information processing device) 12 is a flowchart showing an example of the flow of the dipole estimation process of the information processing device according to the second embodiment. The flow of the dipole estimation process of the information processing device 50a according to the present embodiment will be described with reference to FIG.

[0076] <Steps S21a and S22a> The processes in steps S21a and S22a are similar to those in steps S21 and S22 (FIG. 8) described above, respectively, and then the process proceeds to step S23a.

[0077] <Step S23a> The onset search unit 506 performs dipole estimation for at least one of the waveform data portions before and after the time detected by the IED detection unit 502 (a predetermined search range before the time when the time is the peak time), calculates index values ​​such as the GoF of the estimated solution, the confidence volume (CV), the waveform amplitude, and the dipole distance, and performs onset search based on the index values. Specifically, the onset search unit 506 detects the onset time based on the index values. Then, the process proceeds to step S24a.

[0078] <Step S24a> The calculation unit 503 calculates an isomagnetic field diagram in which regions of equal magnetic field values ​​are expressed with the same brightness value, based on the waveform data at the onset time detected by the onset search unit 506. Then, the process proceeds to step S25a.

[0079] <Steps S25a and S26a> The processes in steps S25a and S26a are similar to the processes in steps S24 and S25 (FIG. 8) described above, respectively, and then the dipole estimation process is completed.

[0080] By estimating not only one dipole but also evaluating the spread of multiple dipoles, it is possible to evaluate information such as whether resection is possible and, if so, which area of ​​the brain it is in. As described above, in cases where the dipole pattern moves between the onset and peak parts, the process using the time automatically detected by the IED detection unit 502 may not be able to properly evaluate the dipole cluster, but the evaluation is improved by detecting the onset time by the onset search unit 506 as in this embodiment.

[0081] In order to detect the onset time, the onset search unit 506 performs dipole estimation at each time point during a search period that includes the onset time. Therefore, the dipole estimation unit 505 may use the result of dipole estimation already performed at the onset time detected by the onset search unit 506 as the final dipole estimation result.

[0082] [Third embodiment] The biosignal measurement system 1 according to the third embodiment will be described with a focus on differences from the biosignal measurement system 1 according to the first embodiment. Some epilepsy cases may have multiple epileptic lesions. In that case, if it is assumed that there is one current source in dipole estimation, they cannot be separated by automatic detection, and the dipole cannot be properly estimated. Therefore, when multiple epileptic lesions are suspected, it is necessary to perform sensor selection separately for each and then perform dipole estimation. Therefore, in this embodiment, when it is determined that multiple sensor selection areas exist as a result of the removal and aggregation process of multiple sensor selection areas by the aggregation unit 5042, the operation of identifying sensors in each sensor selection area will be described assuming that multiple dipole patterns exist. Note that the overall configuration of the biosignal measurement system 1 according to this embodiment, and the hardware configuration and functional block configuration of the information processing device 50 are the same as those described in the first embodiment.

[0083] (Flow of dipole estimation process of information processing device) FIG. 13 is a flowchart showing an example of a flow of a dipole pattern detection process of an information processing device according to a third embodiment. FIG. 14 is a diagram showing an example of a plurality of sensor selection regions corresponding to a dipole pattern. A flow of a dipole estimation process of the information processing device 50 according to this embodiment will be described with reference to FIG. 13 and FIG. 14. In particular, in FIG. 13, a flow of a dipole pattern detection process by the dipole pattern detection unit 504 of the information processing device 50 according to this embodiment will be described. The dipole pattern detection unit 504 of the information processing device 50 according to this embodiment executes detection processes S241a to S248a of a dipole pattern region shown in FIG. 13, instead of step S24 of FIG. 8, specifically, detection processes S241 to S246 of a dipole pattern region shown in FIG. 9.

[0084] <Steps S241a to S243a> The processes in steps S241a to S243a are similar to the processes in steps S241 to S243 (FIG. 9) described above, respectively, and then the process proceeds to step S244a.

[0085] <Step S244a> The determination unit 5043 of the dipole pattern detection unit 504 determines whether or not a sensor selection area remains as a result of the removal and aggregation of the sensor selection areas by the aggregation unit 5042. If a sensor selection area remains (step S244a: Yes), the process proceeds to step S245a, and if not (step S244a: No), the process proceeds to step S248a.

[0086] <Step S245a> The determination unit 5043 determines whether or not there are multiple remaining sensor selection areas as a result of the removal / aggregation by the aggregation unit 5042. If there are multiple sensor selection areas (step S245a: Yes), the process proceeds to step S246a, and if there is a single sensor selection area (step S245a: No), the process proceeds to step S247a.

[0087] <Step S246a> When the determination unit 5043 determines that a plurality of sensor selection regions exist, the sensor identification unit 5044 of the dipole pattern detection unit 504 identifies the sensors included in each sensor selection region as sensors that constitute the region of each dipole pattern. Here, Fig. 14 shows an example in which the determination unit 5043 determines that two sensor selection regions 901 and 902 exist. In this case, the sensor identification unit 5044 identifies the sensors included in the sensor selection regions 901 and 902 as sensors that constitute the region of each dipole pattern. Then, step S24 ends, and the process proceeds to the above-mentioned step S25 (Fig. 8).

[0088] <Step S247a> If the determination unit 5043 determines that a single sensor selection region exists, the sensor identification unit 5044 identifies the sensors included in the sensor selection region as sensors that configure the region of the dipole pattern. Then, step S24 ends, and the process proceeds to step S25.

[0089] <Step S248a> The process of step S248a is similar to the process of step S246 (FIG. 9) described above.

[0090] As described above, in the information processing device 50 according to this embodiment, when a sensor selection area remains as a result of processing by the aggregation unit 5042, the determination unit 5043 determines whether or not there are multiple remaining sensor selection areas, and when the determination unit 5043 determines that multiple sensor selection areas exist, the sensor identification unit 5044 identifies the sensors included in each sensor selection area as sensors that constitute the area of ​​each dipole pattern. This allows sensors to be selected separately when multiple epileptic lesions are suspected, making it possible to perform dipole estimation.

[0091] In addition, if it is determined that multiple sensor selection areas exist as a result of the removal / aggregation process of multiple sensor selection areas by the aggregation unit 5042 in this embodiment, it is determined that multiple dipole patterns exist, and in this case, the operation of identifying sensors in each sensor selection area can also be applied to the second embodiment described above.

[0092] [Fourth embodiment] The biosignal measurement system 1 according to the fourth embodiment will be described with a focus on differences from the biosignal measurement system 1 according to the first embodiment. In the above-mentioned first embodiment, the operation of automatically detecting the time when characteristic waveform information of an IED appears by the IED detection unit 502 has been described. In actual analysis, there are cases where the characteristic waveform of an IED is visually detected from waveform data of a magnetoencephalograph or the like, or where the characteristic waveform of an IED is searched for using an electroencephalograph measured simultaneously with a magnetoencephalograph as a guide. In other words, in cases where the time when dipole estimation is performed visually is determined in advance, only sensor selection is necessary. Therefore, in this embodiment, the operation of manually detecting the time when characteristic waveform information of an IED appears and appropriately specifying a sensor based on the time will be described. Note that the overall configuration of the biosignal measurement system 1 according to this embodiment, and the hardware configuration and functional block configuration of the information processing device 50 are the same as those described in the first embodiment.

[0093] (Configuration and operation of functional blocks of information processing device) In the information processing device 50 according to this embodiment, the IED detection unit 502 detects a time manually designated by an expert such as a doctor as the time when characteristic waveform information of an IED appears. For example, the time when the expert visually determines that characteristic waveform information of an IED appears after checking the waveform data corresponding to each sensor displayed on the display device 107 is designated via the input device 106, and the IED detection unit 502 detects the time designated by the designation operation as the time when characteristic waveform information of an IED appears.

[0094] The processing of the other functional units of the information processing device 50 according to this embodiment is similar to that of the information processing device 50 according to the above-described first embodiment.

[0095] (Flow of dipole estimation process of information processing device) 15 is a flowchart showing an example of the flow of the dipole estimation process of the information processing device according to the fourth embodiment. The flow of the dipole estimation process of the information processing device 50 according to this embodiment will be described with reference to FIG.

[0096] <Step S21b> The process of step S21b is the same as the process of step S21 (FIG. 8) described above, and then the process proceeds to step S22b.

[0097] <Step S22b> The IED detection unit 502 detects the time that is manually designated by an expert such as a doctor as the time when characteristic waveform information of an IED appears, and then proceeds to step S23b.

[0098] <Steps S23b to S25b> The processes in steps S23b to S25b are similar to the processes in steps S23 to S25 (FIG. 8) described above, respectively. This completes the dipole estimation process.

[0099] As described above, in the information processing device 50 according to this embodiment, the IED detection unit 502 detects the time when the characteristic waveform information of the IED appears as the time when the expert manually designates the time. This makes it possible to appropriately select a sensor with high accuracy even in a case where the time when the dipole estimation is performed is determined by visual inspection in advance and only sensor selection is required.

[0100] In addition, the operation of detecting the time manually specified by the IED detection unit 502 in this embodiment as the time at which characteristic waveform information of an IED appears can also be applied to the second and third embodiments described above.

[0101] In each of the above-mentioned embodiments, when at least one of the functional units of the information processing device 50, 50a is realized by executing a program, the program is provided by being incorporated in advance in a ROM or the like. In each of the above-mentioned embodiments, the program executed by the information processing device 50, 50a may be provided by being recorded in a computer-readable recording medium such as a CD-ROM (Compact Disc Read Only Memory), a flexible disk (FD), a CD-R (Compact Disk-Recordable), or a DVD (Digital Versatile Disc) in an installable or executable format. In each of the above-mentioned embodiments, the program executed by the information processing device 50, 50a may be provided by being stored on a computer connected to a network such as the Internet and downloaded via the network. In each of the above-mentioned embodiments, the program executed by the information processing device 50, 50a may be provided or distributed via a network such as the Internet. Furthermore, in each of the above-mentioned embodiments, the program executed by the information processing device 50, 50a has a modular configuration including at least any one of the above-mentioned functional units, and in terms of actual hardware, the CPU 101 reads out and executes the program from the above-mentioned storage device (e.g., ROM 103 or auxiliary storage device 104, etc.), so that each of the above-mentioned functional units is loaded and generated on the main storage device (RAM 102).

[0102] The aspects of the present invention are as follows. <1> An acquisition step of acquiring waveform data of biological signals measured by a plurality of sensors; A calculation step of calculating distribution information indicating a distribution of values ​​of the biological signal based on the waveform data at a time when characteristic waveform information of an IED (Interictal Epileptiform Discharge) appears among the acquired waveform data; a step of providing the distribution information calculated in the calculation step as an input to a model trained using as training data distribution information indicating a distribution of values ​​of the biosignals and information on the sensors selected in the analysis, obtaining a selection area indicating a dipole pattern area as an output, and identifying the sensors constituting the dipole pattern area based on the selection area; The method for identifying waveform occurrence includes the steps of: <2> The identifying step includes: a region detection step of inputting the calculated distribution information to the model and obtaining the selected region as an output; an aggregation step of at least one of removing and aggregating the plurality of selected regions when the plurality of selected regions are obtained as an output of the model; a determination step of determining whether or not the selected area remains as a result of the processing of the aggregation step; a sensor identification step of identifying the sensors included in the selected region as the sensors constituting the region of the dipole pattern when it is determined that the selected region remains; The above-mentioned <1> This is a waveform occurrence identification method described in <3> The method further includes a detection step of automatically detecting a time when the characteristic waveform information appears from the acquired waveform data, In the calculation step, the distribution information is calculated based on the waveform data at the time detected in the detection step among the acquired waveform data. <1> or <2> This is a waveform occurrence identification method described in <4> a detection step of detecting a time designated by an operation on an input unit as a time at which the characteristic waveform information appears from the acquired waveform data, In the calculation step, the distribution information is calculated based on the waveform data at the time detected in the detection step among the acquired waveform data. <1> or <2> This is a waveform occurrence identification method described in <5> A search step is further provided for calculating a predetermined index value for at least one of the waveform data portions before and after the time detected in the detection step, and searching for an onset time at which the IED occurred in the waveform data based on the index value; In the calculation step, the distribution information is calculated based on the waveform data of the onset time searched for in the search step among the acquired waveform data. <3> or <4> This is a waveform occurrence identification method described in <6> In the determination step, if the selected area remains as a result of the processing in the aggregation step, it is determined whether or not there are a plurality of the selected areas remaining; In the sensor identification step, when it is determined in the determination step that a plurality of the selected regions exist, the sensors included in each of the selected regions are identified as the sensors constituting each region of the dipole pattern. <2> This is a waveform occurrence identification method described in <7> In the acquiring step, the waveform data of a magnetic brain signal is acquired as the biological signal, In the calculation step, an isomagnetic field diagram is calculated as the distribution information based on the waveform data at a time when the characteristic waveform information appears among the acquired waveform data. <1> ~ <6> 13. A waveform occurrence identification method according to claim 12. <8> In the acquiring step, the waveform data of an electroencephalogram signal is acquired as the biological signal, In the calculation step, a current distribution map is calculated as the distribution information based on the waveform data at a time when the characteristic waveform information appears among the acquired waveform data. <1> ~ <6> 13. A waveform occurrence identification method according to claim 12. <9> The method further includes the step of performing a dipole estimation using the waveform data corresponding to the time when the characteristic waveform information appears and the specified sensor. <1> ~ <8> 13. A waveform occurrence identification method according to claim 12. <10> On the computer, An acquisition step of acquiring waveform data of biological signals measured by a plurality of sensors; A calculation step of calculating distribution information indicating a distribution of values ​​of the biological signal based on the waveform data at a time when characteristic waveform information of an IED (Interictal Epileptiform Discharge) appears among the acquired waveform data; a step of providing the distribution information calculated in the calculation step as an input to a model trained using as training data distribution information indicating a distribution of values ​​of the biosignals and information on the sensors selected in the analysis, obtaining a selection area indicating a dipole pattern area as an output, and identifying the sensors constituting the dipole pattern area based on the selection area; This is a program for executing the above. [Explanation of symbols]

[0103] 1. Biosignal measurement system 3. Measuring Equipment 4 Measurement Table 31 Duwa 32 Depression 40 Servers 50, 50a Information processing device 101 CPU 102 RAM 103 ROM 104 Auxiliary storage 105 Network I / F 106 Input Devices 107 Display device 108 Bus 401 Data Acquisition Department 402 Data storage unit 501 Acquisition Department 502 IED Detector 503 Calculation Department 504 Dipole pattern detector 505 Dipole Estimation Unit 506 Onset Search Unit 601 Head 602 Sensor 603 Outlet 604 Sunken section 701, 702 Sensor 703 Sensor selection area 801 Waveform 802 Peak part 803 Search range 901, 902 Sensor selection area 5041 Region detection unit 5042 Aggregation unit 5043 Judgment unit 5044 Sensor identification unit

Prior art documents

Patent documents

[0104]

Patent Document 1

Patent Document 2

Non-patent documents

[0105]

Non-patent Document 1

Non-patent Document 2

Non-patent Document 3

Non-Patent Document 4

Non-Patent Document 5

Claims

1. An acquisition step of acquiring waveform data of biological signals measured by a plurality of sensors; a calculation step of calculating distribution information indicating a distribution of values ​​of the biological signal based on the waveform data at a time when characteristic waveform information of an IED (Internal Epileptiform Discharge) appears among the acquired waveform data; a step of providing the distribution information calculated in the calculation step as an input to a model trained using as training data distribution information indicating a distribution of values ​​of the biosignals and information on the sensors selected in the analysis, obtaining a selection area indicating a dipole pattern area as an output, and identifying the sensors constituting the dipole pattern area based on the selection area; The method for determining waveform occurrence comprising:

2. The identifying step includes: a region detection step of inputting the calculated distribution information to the model and obtaining the selected region as an output; an aggregation step of at least one of removing and aggregating the plurality of selected regions when the plurality of selected regions are obtained as an output of the model; a determination step of determining whether or not the selected area remains as a result of the processing of the aggregation step; a sensor identification step of identifying the sensors included in the selected region as the sensors constituting the region of the dipole pattern when it is determined that the selected region remains; 2. The method of claim 1, further comprising:

3. The method further includes a detection step of automatically detecting a time when the characteristic waveform information appears from the acquired waveform data, 3. The waveform occurrence identification method according to claim 1, wherein in the calculating step, the distribution information is calculated based on the waveform data at the time detected in the detecting step among the acquired waveform data.

4. a detection step of detecting a time designated by an operation on an input unit as a time at which the characteristic waveform information appears from the acquired waveform data, 3. The waveform occurrence identification method according to claim 1, wherein in the calculating step, the distribution information is calculated based on the waveform data at the time detected in the detecting step among the acquired waveform data.

5. A search step of calculating a predetermined index value for at least one of the waveform data portions before and after the time detected in the detection step, and searching for an onset time at which the IED occurred in the waveform data based on the index value, 4. The waveform occurrence identification method according to claim 3, wherein in the calculating step, the distribution information is calculated based on the waveform data of the onset time searched for in the searching step among the acquired waveform data.

6. In the determination step, if the selected area remains as a result of the processing in the aggregation step, it is determined whether or not there are a plurality of the selected areas remaining; 3. The waveform generation identification method according to claim 2, wherein, in the sensor identification step, if it is determined in the determination step that a plurality of the selected areas exist, the sensors included in each of the selected areas are identified as the sensors that constitute each area of ​​the dipole pattern.

7. In the acquiring step, the waveform data of a magnetic brain signal is acquired as the biological signal, 3. The waveform occurrence identification method according to claim 1, wherein in the calculation step, an isomagnetic field diagram is calculated as the distribution information based on the waveform data at a time when the characteristic waveform information appears among the acquired waveform data.

8. In the acquiring step, the waveform data of an electroencephalogram signal is acquired as the biological signal, 3. The waveform occurrence identification method according to claim 1 or 2, wherein in the calculation step, a current distribution map is calculated as the distribution information based on the waveform data at a time when the characteristic waveform information appears among the acquired waveform data.

9. 3. The method for identifying waveform occurrence according to claim 1, further comprising an estimation step of performing a dipole estimation using the time at which the characteristic waveform information appears and the waveform data corresponding to the identified sensor.

10. On the computer, An acquisition step of acquiring waveform data of biological signals measured by a plurality of sensors; a calculation step of calculating distribution information indicating a distribution of values ​​of the biological signal based on the waveform data at a time when characteristic waveform information of an IED (Internal Epileptiform Discharge) appears among the acquired waveform data; a step of providing the distribution information calculated in the calculation step as an input to a model trained using as training data distribution information indicating a distribution of values ​​of the biosignals and information on the sensors selected in the analysis, obtaining a selection area indicating a dipole pattern area as an output, and identifying the sensors constituting the dipole pattern area based on the selection area; A program for executing the above.

Citation Information

Patent Citations

  • Waveform generation specification method and program

    JP2021069929A

  • Information processing device, program and information processing method

    JP2023020273A