Information processing device, epilepsy surgery support device, program, information processing method

Abstract

To provide a method of determining a seizure onset zone and an epileptogenic zone around it on the basis of brain wave signals.SOLUTION: An information processing device for supporting a diagnosis of epilepsy comprises: an acquisition part which acquires brain wave signals from a plurality of electrodes installed on the head; a calculation part which calculates a time when high-frequency brain waves are detected by the respective electrodes, based on time frequency analysis, from the brain wave signals; and a determination part which determines whether respective portions of the brain respectively corresponding to the electrodes correspond to a seizure onset zone or an epileptogenic zone, based on the time.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an information processing apparatus and the like.

Background Art

[0002] In epilepsy surgery, in order to achieve the disappearance of epileptic seizures, appropriate resection of the seizure onset part (epileptic focus, seizure onset zone: SOZ) and the surrounding epileptogenic area (epileptogenic site, epileptogenic zone: EZ) is required. For the identification of the SOZ, a method of identifying the electrode corresponding to the SOZ in intracranial electroencephalogram examination during an epileptic seizure is generally used. For example, Patent Document 1 discloses an epilepsy determination device that determines whether an electroencephalogram of an epileptic seizure appears based on electroencephalogram data. Patent Document 2 also discloses a program that extracts the SOZ based on nuclear medicine images by utilizing the fact that cerebral blood flow in the SOZ increases during an epileptic seizure.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] The epilepsy seizure determination device described in Patent Document 1 can determine whether an epilepsy patient is having an epileptic seizure. However, the epilepsy seizure determination device described in Patent Document 1 does not mention specifying the SOZ and the EZ.

[0005] The program described in Patent Document 2 has a high probability of identifying the SOZ. However, it does not mention a method for identifying the EZ. In addition, the method in Patent Document 2 requires a relatively large imaging diagnostic device to acquire nuclear medicine images. Furthermore, acquiring nuclear medicine images during an epileptic seizure is time-consuming and laborious, making it difficult to identify the SOZ and EZ, for example, during epilepsy surgery.

[0006] This invention was made in view of the above-mentioned problems, and its objective is to propose a method that enables accurate (high-precision) and rapid (high-speed) determination of SOZ and EZ from electroencephalogram (EEG) signals. Furthermore, it aims to propose a method that enables determination of SOZ and EZ from EEG signals regardless of whether an epileptic seizure is occurring or not. [Means for solving the problem]

[0007] According to a first aspect of the present invention, an information processing device for assisting in the diagnosis of epilepsy includes: an acquisition unit that acquires electroencephalogram signals from a plurality of electrodes placed on the head; a calculation unit that calculates the time at which high-frequency electroencephalograms are detected at each of the plurality of electrodes based on time-frequency analysis from the electroencephalogram signals; and a determination unit that determines, based on the time, whether each brain region corresponding to each of the plurality of electrodes corresponds to the epileptic seizure initiation area or the epileptogenic region. According to a second aspect of the present invention, a program to be executed by a computer assisting in the diagnosis of epilepsy is to cause the computer to perform the following actions: acquire electroencephalogram (EEG) signals from a plurality of electrodes placed on the head; calculate the time at which high-frequency EEG signals are detected at each of the plurality of electrodes based on time-frequency analysis from the EEG signals; and determine, based on the time, whether each brain region corresponding to each of the plurality of electrodes corresponds to the epileptic seizure initiation site or the epileptogenic region. According to a third aspect of the present invention, the information processing method for an information processing device that supports epilepsy diagnosis includes: acquiring electroencephalogram (EEG) signals from a plurality of electrodes placed on the head; calculating the time at which high-frequency EEG signals are detected at each of the plurality of electrodes based on time-frequency analysis from the EEG signals; and determining, based on the time, whether each brain region corresponding to each of the plurality of electrodes corresponds to the epileptic seizure initiation site or the epileptogenic region. [Effects of the Invention]

[0008] According to the present invention, by performing time-frequency analysis of electroencephalogram (EEG) signals, it is possible to accurately and quickly determine, for example, the epileptic seizure initiation site and the epileptogenic region based on the time at which high-frequency EEG signals are detected. Furthermore, according to the present invention, the EEG signal may be either an EEG signal during an epileptic seizure or an EEG signal during a non-seizure state, thereby enabling the determination of the epileptic seizure initiation site and the epileptogenic region from the EEG signal regardless of whether it is during or after an EEG. [Brief explanation of the drawing]

[0009] [Figure 1] A block diagram showing an example of the functional configuration of an information processing device. [Figure 2] A diagram showing an example of a group of electrodes placed on the head. [Figure 3] A flowchart illustrating an example of the information processing flow. [Figure 4] A diagram showing specific examples of electroencephalogram (EEG) signals and detection results from the spike wave detection unit. [Figure 5] A diagram showing a specific example of the processing results in the time-frequency analysis unit. [Figure 6] A diagram showing a specific example of the processing results in the time-frequency analysis unit. [Figure 7] A diagram illustrating a specific example of processing results in the statistical analysis department. [Figure 8] A diagram illustrating a specific example of processing results in the statistical analysis department. [Figure 9] A diagram showing specific examples of processing results in the statistical analysis unit and the lesion area determination unit. [Figure 10] A diagram showing a specific example of the dependency relationship between the processing result of the spike detection unit and the processing result in the time-frequency analysis unit. [Figure 11] A diagram showing an example of information processing results. [Figure 12] A diagram showing an example of information processing results. [Figure 13] A diagram showing an example of information processing results. [Figure 14] A block diagram showing an example of the functional configuration of an epilepsy surgery support device. [Figure 15] A flowchart showing an example of the flow of electroencephalogram signal analysis processing. [Figure 16] A diagram showing an example of a display screen on the display unit of an epilepsy surgery support device. [Figure 17] A diagram showing an example of a display screen on the display unit of an epilepsy surgery support device. [Figure 18] A diagram showing an example of a display screen on the display unit of an epilepsy surgery support device. [Figure 19] A diagram showing an example of a display screen on the display unit of an epilepsy surgery support device. [Embodiments for Carrying Out the Invention]

[0010] Hereinafter, an example of an embodiment for carrying out the present invention will be described with reference to the drawings. In the description of the drawings, the same reference numerals may be assigned to the same elements, and redundant descriptions may be omitted. In addition, the components described in this embodiment are merely examples, and are not intended to limit the scope of the present invention thereto.

[0011] [Embodiment] Hereinafter, an example of an embodiment for realizing the information processing technology of the present invention will be described.

[0012] FIG. 1 is a block diagram showing an example of the functional configuration of an information processing apparatus 1 according to an aspect of this embodiment. The information processing apparatus 1 may also be referred to as an electroencephalogram analysis apparatus or an epilepsy region determination apparatus. The information processing device 1 includes, for example, an electroencephalogram (EEG) signal acquisition unit 110, a high-frequency oscillation (HFO) generation time calculation unit 120, and a lesion area determination unit 130. The high-frequency electroencephalogram generation time calculation unit 120 includes, for example, a spike wave detection unit 121, a bandpass filter unit 123, a time-frequency analysis unit 125, and a statistical analysis unit 127. The high-frequency electroencephalogram generation time calculation unit 120 may also be called a high-frequency rhythm generation time calculation unit or an HFO generation time calculation unit, and is sometimes referred to as the HFO generation time calculation unit 120. These can be, for example, functional units (functional blocks) of the unillustrated processing unit (processing unit) or control unit (control device) of the information processing device 1, and are configured to have processing circuits such as a CPU (Central Processing Unit), GPU (Graphics Processing Unit), DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), and FPGA (Field Programmable Gate Array).

[0013] The electroencephalogram (EEG) signal acquisition unit 110 receives, for example, an EEG signal acquired by the external head-mounted electrode group 5 of the information processing device 1 as input.

[0014] The head-mounted electrode group 5 comprises one or more electrodes 5x that are placed on the patient's head. For example, if the number of electrodes in the head-mounted electrode group 5 is "N" (where "N" is a natural number), then the head-mounted electrode group 5 comprises N electrodes 5A, 5B, 5C, ..., 5N. Figure 2 shows an example of a head-mounted electrode group 5 placed inside the skull. In this figure, for example, each electrode 5x is placed at equal intervals on the subdural surface of the brain as a subdural electrode. Note that each electrode 5x in the head-mounted electrode group 5 may be a deep electrode, or a combination of subdural and deep electrodes may be used. Furthermore, each electrode 5x in the head-mounted electrode group 5 is not limited to intracranial electrodes. For example, some or all of each electrode 5x may be a scalp electrode.

[0015] In the following, the electroencephalogram (EEG) signal acquired by electrode 5A will be referred to as the "channel 1 signal," the EEG signal acquired by electrode 5B as the "channel 2 signal," ..., and the EEG signal acquired by electrode 5N as the "channel N signal." Note that "channel" may be abbreviated as "Ch." Also, the electrode that acquires the "channel x signal" may be referred to as "electrode Ch.x." Furthermore, each electrode may be referred to as a "channel."

[0016] The electroencephalogram (EEG) signal acquisition unit 110, for example, receives a multi-channel EEG signal of "N" channel from the head-mounted electrode group 5 as input, buffers the EEG signal for a predetermined period (for example, "90 minutes"), and has the function of outputting the EEG signal within the predetermined period to the HFO (High Focus Occurrence) time calculation unit 120.

[0017] The spike detection unit 121, for example, receives electroencephalogram (EEG) signals acquired by the EEG signal acquisition unit 110 as input, detects spikes from each channel signal, and outputs the time at which the spikes were detected (hereinafter referred to as the "tagging time") to the time-frequency analysis unit 125.

[0018] The bandpass filter unit 123, for example, when it receives an electroencephalogram (EEG) signal as input, applies a bandpass filter to each channel signal and outputs the filtered EEG signal (hereinafter referred to as the "bandpass EEG signal") to the time-frequency analysis unit 125.

[0019] The time-frequency analysis unit 125, for example, when it receives a bandpass electroencephalogram signal and a tagging time as input, performs time-frequency analysis on the bandpass electroencephalogram signal of each channel signal and converts it into a frequency spectrum. The time-frequency analysis unit 125 then has a function to calculate, for example, the average of the frequency spectra (hereinafter referred to as "electrode-specific frequency spectrum") for each electrode over a predetermined window period (for example, "500 milliseconds") before and after the tagging time, based on the tagging time.

[0020] The statistical analysis unit 127, for example, receives the frequency spectrum calculated by the time-frequency analysis unit 125 and the electrode-specific frequency spectrum as input, and analyzes the time period and frequency band in which a statistically significant increase in the power (amplitude) of the electrode-specific frequency spectrum is observed for each electrode. Based on the analysis results, the statistical analysis unit 127 calculates, for example, the power increase latency of the electrode-specific frequency spectrum (hereinafter referred to as "HFO power increase latency") and has the function of outputting the HFO power increase latency for each electrode to the lesion area determination unit 130.

[0021] The lesion area determination unit 130, for example, when it receives the HFO power increase latency as input from the statistical analysis unit 127, has the function of determining whether the site where each electrode is placed corresponds to the SOZ, the EZ, or neither the SOZ nor the EZ, and outputting the determination result.

[0022] Here, the "output" of the judgment result may include not only the display of the judgment result on the device itself (display output), but also, for example, the output of the judgment result to other functional units of the device (internal output), or the output (external output) or transmission (external transmission) of the judgment result to devices other than the device itself (external devices).

[0023] [Information Processing Procedures] Figure 3 is a flowchart showing an example of the information processing procedure in this embodiment. The process shown in the flowchart of Figure 3 is realized, for example, by the processing unit of the information processing device 1 reading the program code stored in a memory unit (not shown) into a RAM (not shown) and executing it.

[0024] In the flowchart in Figure 3, each symbol S represents a step. The flowchart described below is merely an example of the information processing procedure in this embodiment, and other steps may be added or some steps may be deleted. Furthermore, some of the steps in the flowchart may be rearranged before execution.

[0025] First, the electroencephalogram (EEG) signal acquisition unit 110 performs EEG signal acquisition processing (S110). In EEG signal acquisition processing, for example, the EEG signal acquisition unit 110 acquires EEG signals from the head-mounted electrode group 5 and stores the EEG signals in a buffer (not shown). Furthermore, the electroencephalogram (EEG) signals acquired using this method may be those obtained during an epileptic seizure or those obtained during a non-seizure period.

[0026] Then, for example, when the electroencephalogram (EEG) signal acquisition unit 110 acquires EEG signals for a predetermined period (for example, "90 minutes"), the spike wave detection unit 121 performs spike wave tagging (S120).

[0027] In the spike wave tagging process, the spike wave detection unit 121 records the time at which a spike wave is detected from the channel x signal for a specific electrode Ch.x as the tagging time and tags it. The electrode Ch.x may be specified manually, for example, or the electrode in which the spike wave was detected first among all channel signals may be used. Alternatively, the electrode Ch.x may be the electrode in which the most prominent spike wave was detected among all electrodes.

[0028] Here, the following methods can be used to detect spikes from electroencephalogram (EEG) signals. - When the amplitude of the signal exceeds the first threshold relative to the baseline of the channel x signal, or when that condition persists for the first set period. - If the derivative of the channel x signal exceeds the second threshold, or if that condition persists for the second set period. Furthermore, in order to obtain the signal used as a reference in the HFO power increase latency calculation process described later, if other spikes are detected within a predetermined window period (e.g., "500 milliseconds") before and after the detected spike, those spikes may be excluded from tagging. Furthermore, if other tagged spikes exist within a third set period (e.g., "5 seconds") before or after the detected spike, the detected spike may be excluded from tagging. The third set period may be set, for example, so that the number of spikes tagged for one electrode (a specific channel x signal) is a predetermined number (e.g., "30 to 50").

[0029] The spike wave detection unit 121 may terminate the spike wave tagging process when the number of tagged spike waves reaches a predetermined number (for example, "30 to 50"). Furthermore, if the number of spikes tagged from the channel x signal over a predetermined period (e.g., 90 minutes) for a specific electrode Ch.x is less than a predetermined number (e.g., 30 to 50), the spike tagging process may be performed on an electrode different from electrode Ch.x.

[0030] Furthermore, the detection of spike waves may be performed manually.

[0031] Figure 4 shows an example of the results when spike wave detection processing is performed on the electroencephalogram (EEG) signal acquired by the EEG signal acquisition unit 110 and the EEG signal from electrode Ch.1. For example, if a spike is detected in the channel 1 signal at time "T1", that time "T1" is recorded as the tagging time. Although a spike may be observed again approximately 1.5 seconds later, a small spike-like baseline fluctuation is observed in the 500 milliseconds immediately preceding this spike, so the tagging time is not recorded (tagging of the spike). Then, for example, if a spike is detected again at time "T2" outside the third set period, that time "T2" is recorded as the tagging time. In this example, during the spike wave tagging process, the times "T1" and "T2" are tagged as the tagging times (spike wave detection times) for electrode Ch.1. In the following, spike detection processing is performed on the channel x signal, and the tag tagged at tagging time "t" is denoted as "tag "x(t)".

[0032] Returning to Figure 3, for example, when the electroencephalogram tagging process is performed, the bandpass filter unit 123 performs the bandpass filtering process (S130). In the bandpass filtering process, the bandpass filter unit 123 applies a bandpass filter, for example, from "0.08Hz" to "300Hz", to each channel signal of the electroencephalogram (EEG) signal acquired by the EEG signal acquisition unit 110, to generate a bandpass EEG signal. Note that the cutoff frequency of a band-pass filter is not limited to "0.08Hz" to "300Hz". For example, it could be "1Hz" to "200Hz". Also, for example, the filter used in band-pass filtering may be a low-pass filter.

[0033] Here, we will describe clinical findings regarding the relationship between high-frequency electroencephalography (HFO), which serves as the basis for determining the cutoff frequency of band-pass filters, and SOZ / EZ. In SOZ, high-frequency oscillations (HFOs) of 80 Hz or higher have been reported to be detected in intracranial electroencephalography (EEG) (Zijlmans M, Jiruska P, Zelmann R, Leijten FS, Jefferys JG, Gotman J. High-frequency oscillations as a new biomarker in epilepsy. Ann Neurol. 2012 Feb;71(2):169-178.). Similarly, HFOs detected during non-seizure periods have also been reported to be an indicator of SOZ (van Klink N, Frauscher B, Zijlmans M, Gotman J. Relationships between interictal epileptic spikes and ripples in surface EEG. Clin Neurophysiol. 2016 Jan;127(1):143-149.).

[0034] This method aims to determine SOZ and EZ by analyzing HFOs associated with epileptic spikes. Therefore, the cutoff frequency of the band-pass filter may be set to analyze the frequency range of HFOs, from approximately 80Hz to 150Hz.

[0035] For example, when the spike wave tagging process and the bandpass filtering process are performed, the time-frequency analysis unit 125 performs the time-frequency analysis process (S140). In the time-frequency analysis process, the time-frequency analysis unit 125 first applies a short-time Fourier transform to each channel of the bandpass electroencephalogram signal, for example, at discrete time intervals, to calculate a time-continuous frequency spectrum (frequency power spectrum). Alternatively, a wavelet transform may be used to calculate the frequency spectrum.

[0036] For example, once the bandpass electroencephalogram signals are converted into frequency spectra for all channels, the time-frequency analysis unit 125 extracts frequency spectra for all channel signals within a predetermined time window period, using, for example, the tagging time for each tagged electrode Ch.x in the spike wave tagging process as a reference. Then, the time-frequency analysis unit 125 calculates electrode-specific frequency spectra by, for example, averaging the frequency spectra extracted for each electrode.

[0037] Figures 5 and 6 show specific examples of time-frequency analysis processing. In the following, the predetermined window period before and after the tag "x(t)" of the bandpass electroencephalogram signal will be referred to as the "tag "x(t) time window." The frequency spectrum within the tag "x(t)" time window will be referred to as the "tag "x(t) frequency spectrum."

[0038] Figure 5 shows an example of frequency spectrum calculation in time-frequency analysis processing. In this figure, the left side shows the bandpass electroencephalogram (EEG) signals for each channel in the tag "x(T1)" time window, which is a predetermined window period (500 milliseconds in this example) before and after the tag "x(T1)" detected at tagging time "T1", as a result of performing spike wave detection processing on electrode Ch.x. The right side shows the corresponding tag "x(T1)" frequency spectrum calculated from the bandpass EEG signals of each channel in the tag "x(T1)" time window. Perform this process for all tags, for example.

[0039] Furthermore, the conversion to the tag "x(t)" frequency spectrum in the tag "x(t)" time window may be performed by first converting the entire bandpass electroencephalogram signal into a frequency spectrum and then extracting the tag "x(t)" frequency spectrum in the tag "x(t)" time window, or by first extracting the bandpass electroencephalogram signal in the tag "x(t)" time window and then calculating the tag "x(t)" frequency spectrum.

[0040] Figure 6 shows an example of calculating the electrode-specific frequency spectrum at electrode Ch.1 from the tag "x(t)" frequency spectrum ("t" = "1,···,M") when, for example, "M" spikes are tagged at electrode Ch.x. In this figure, the left side shows the channel 1 signal of the tag "x(t)" frequency spectrum for each tag, starting with the channel 1 signal of the tag "x(T1)" shown on the right side of Figure 5, arranged vertically. Each tag "x(t)" frequency spectrum is aligned with its tagging time "t" as the reference point (time axis "0").

[0041] When epileptic spikes are detected, there is a possibility that subsequent HFOs may be generated due to the influence of these epileptic spikes. Since the frequency spectra of each tag "x(t)" are aligned on the time axis based on the detection of spikes, performing analysis based on multiple frequency spectra is expected to have the effect of preventing false detections of artifacts and physiologically occurring HFOs, compared to, for example, simply detecting and counting HFOs.

[0042] On the right side of the figure, the frequency spectra of each electrode in electrode Ch.1 are shown, calculated by, for example, using the tagging time "t" of the spectrum of channel 1 signal shown on the left as the reference time "0" and applying an arithmetic mean to the power. This process is performed, for example, for all electrodes.

[0043] Furthermore, the method for calculating the electrode-specific frequency spectrum for a given electrode Ch.y is not limited to the summation average of multiple tag "x(t)" frequency spectra tagged to electrode Ch.x. For example, the geometric mean or harmonic mean may also be used. Furthermore, it is not limited to using the "x(t)" frequency spectrum for all tags. For example, if the "x(ext)" frequency spectrum of one tag is significantly different from the "x(t)" frequency spectrum of another tag (and is judged as an outlier), the "x(ext)" frequency spectrum may be excluded from the analysis (not included in the calculation of the average).

[0044] Returning to Figure 3, for example, when the time-frequency analysis process is executed, the statistical analysis unit 127 performs the HFO power increase latency calculation process (S150). In the HFO power increase latency calculation process, for example, the statistical analysis unit 127 first performs a statistical analysis on the electrode-specific frequency spectrum calculated in the time-frequency analysis process for each channel. In the statistical analysis, the statistical analysis unit 127 uses the frequency spectral power during a time period in the same channel when no spikes occur (for example, approximately 300 to 500 milliseconds immediately before a tagged spike) as a reference to calculate the time-frequency range in which the electrode-specific frequency spectral power is determined to be statistically significant. Hereinafter, the calculated determination result expressed in terms of frequency spectrum will be referred to as the "electrode-specific significance determination spectrum".

[0045] The statistical analysis method used here may include, for example, the bootstrap method (resampling method). This is expected to improve the accuracy of the analysis results even when it is difficult to estimate the shape of the distribution (population) because the number of samples (frequency spectrum) obtainable from the electroencephalogram signal is limited.

[0046] It should be noted that the statistical analysis methods used here are not limited to nonparametric bootstrap methods. For example, the analysis may be performed by assuming a normal distribution or the like and using maximum likelihood estimation. This is expected to improve the speed of the analysis. Furthermore, if accuracy is prioritized over processing speed, for example, the multiscale bootstrap method may be used.

[0047] Subsequently, the statistical analysis unit 127 determines, for example, based on the electrode-specific significance determination spectrum, which divides the grid into predetermined frequency bands and time zones, a grid mesh in which the HFO power increase is statistically significant. Here, the mesh divisions may be set to, for example, "5 milliseconds" for time zones and "10 hertz" for frequency bands. For example, the frequency band and time zone of the mesh divisions may be matched to the resolution (frequency band and time zone) used for significance determination in statistical analysis methods. Furthermore, the statistical analysis unit 127 may, for example, determine a segmented mesh as a segmented mesh in which an increase in HFO power is observed if the segmented mesh is determined to be significant in the electrode-specific significance determination spectrum. In other words, the segmented mesh can be said to be a partial expansion (or discrete representation) of the electrode-specific significance determination spectrum.

[0048] In addition, during the HFO power increase latency calculation process, the statistical analysis unit 127 may not calculate electrode-specific significance determination spectra through statistical analysis, but instead directly calculate the segmented mesh in which electrode-specific frequency spectral power is determined to be statistically significant.

[0049] When a mesh section in which an increase in HFO power is observed is determined, the statistical analysis unit 127 calculates the HFO power increase latency, which is the latency at which a significant increase in HFO power is observed, based on the time of spike wave generation. For example, the statistical analysis unit 127 calculates the HFO power increase latency as the starting point of the region in the mesh section where an increase in HFO power is observed for 20 milliseconds or more and with a bandwidth of 40 Hz or more.

[0050] Furthermore, a method such as the complex demodulation method may be used to realize time-frequency analysis. In this embodiment, as an example, an example of the results obtained by performing analysis using this complex demodulation method is shown.

[0051] Figure 7 shows an example of a significant determination spectrum for each electrode, calculated based on the frequency spectra of each electrode. In this figure, the left side shows the electrode-specific frequency spectra at electrode Ch.1 as shown in Figure 6, and the right side shows the electrode-specific significance determination spectra at electrode Ch.1 after applying the bootstrap method. In the electrode-specific significance determination spectra, statistically significant portions of the electrode-specific frequency spectra, shown at power levels from 0 to 100%, are highlighted in black, while portions that were not statistically significant are highlighted in gray.

[0052] Figure 8 shows a specific example of calculating the HFO power increase latency based on the significant determination spectrum for each electrode. In this figure, the top row shows the electrode-specific significance determination spectra, and the middle row shows the mesh sections marked with circles in the frequency band from "80 Hz" to "150 Hz," which is effective for HFO detection, where an increase in HFO power was determined from the electrode-specific significance determination spectra within a latency period of "-40 milliseconds" to "65 milliseconds" based on each tagging time (spike wave generation time). Furthermore, the lower part of the figure shows a dashed line enclosing the region where an increase in HFO power is observed for a duration of 20 milliseconds or more with a bandwidth of 40 Hz or more. In this example, a significant increase in HFO power was determined in the mesh section with a bandwidth of 40 Hz or more, between -30 milliseconds and 25 milliseconds. Therefore, the latency of the HFO power increase at electrode Ch.1 is calculated to be -30 milliseconds.

[0053] Note that the frequency band and time zone used to calculate the mesh divisions are not limited to the values ​​mentioned above. Also, the divisions of the mesh divisions are not limited to "10 Hz - 5 milliseconds". These values ​​may be adjustable depending on the balance between processing speed and accuracy.

[0054] Furthermore, the bandwidth and duration used to determine the HFO power increase latency are not limited to a range of "40Hz bandwidth" or greater and a duration of "20 milliseconds" or greater. For example, it could be "20Hz bandwidth" or "30Hz bandwidth," or it could be a duration of "30 milliseconds" or greater, or a duration of "40 milliseconds" or greater.

[0055] Furthermore, if it is determined that a segmented mesh shows a continuous increase in HFO power along the frequency axis, the bandwidth condition may be deemed to be met, or the bandwidth condition may be deemed to be met by the sum of the frequency widths of segmented meshes where a discrete increase in HFO power is observed.

[0056] This process is repeated for all electrodes, for example, to calculate the HFO power increase latency for each electrode.

[0057] Returning to Figure 3, for example, once the HFO power increase latency at each electrode is calculated, the lesion area determination unit 130 performs SOZ determination processing (S160). In the SOZ determination process, the lesion area determination unit 130 determines, for example, that the area corresponding to the electrode with the minimum HFO power increase latency among the HFO power increase latency at each electrode is an SOZ.

[0058] Subsequently, the lesion area determination unit 130 performs EZ determination processing (S170). In the EZ determination process, the lesion area determination unit 130 determines the area corresponding to the electrode where the HFO power increase latency at each electrode is less than or equal to the minimum HFO power increase latency (i.e., the HFO power increase latency in SOZ) + threshold latency, as an EZ. In addition, during the EZ determination process, the lesion area determination unit 130 may determine the area corresponding to the electrode where the HFO power increase latency at each electrode is smaller than the minimum HFO power increase latency + threshold latency as an EZ.

[0059] Here, the threshold latency is a parameter that corresponds to the spread of the EZ starting from the SOZ, and for example, the threshold latency may be set to "10 milliseconds". Alternatively, the threshold latency may be set to "15 milliseconds" or "5 milliseconds".

[0060] Figure 9 shows an example of calculating the HFO power increase latency at each electrode, and an example of determining SOZ and EZ. In this figure, for example, the upper section shows a divisional mesh based on the significant determination spectra for each electrode, from electrode Ch.1 to electrode Ch.4. In each divisional mesh, for example, the table in the lower section shows the HFO power increase latency calculated based on the region where an increase in HFO power is observed for a continuous duration of 20 milliseconds or more at a bandwidth of 40 Hz or more, indicated by the dashed line. By referring to the table below, for example, electrode Ch.1, which has the minimum HFO power increase latency, can be determined to be SOZ. Also, for example, if the threshold latency is "10 milliseconds", electrode Ch.2, whose HFO power increase latency is less than or equal to "-20 milliseconds" (after the threshold latency of "10 milliseconds" has elapsed from the HFO power increase latency in SOZ, which is "-30 milliseconds"), can be determined to be EZ.

[0061] Returning to Figure 3, once the series of processes from S120 to S170 are executed, the SOZ and EZ are determined when the spike wave detection process is performed on a specific electrode Ch.x. For example, when the EZ determination process is executed, the information processing device 1 may specify an electrode Ch.y different from the electrode Ch.x used in the previous spike wave detection process and execute the series of processes from S120 to S170 again. As a result, the SOZ and EZ are determined when the spike wave detection process is performed again on electrode Ch.y. The above process may be repeated so that spike wave tagging is performed on all electrodes. Furthermore, it may be possible to specify electrodes within an arbitrary range as the electrodes to be targeted for loop processing following the above spike wave tagging process. Furthermore, in the spike wave tagging process, tagging may be performed first on electrodes within an arbitrary range, and then the series of processes from S130 to S170 may be executed.

[0062] For example, when the SOZ determination process and the EZ determination process are executed, the information processing device 1 determines whether or not to terminate the process (S180). For example, if the system determines to terminate processing based on user operation (user input) to an unillustrated operation unit (input unit), the information processing device 1 terminates processing (S180: YES). If it is determined that the process should not be terminated (S180:NO), the information processing device 1 returns the process to, for example, the electroencephalogram signal acquisition process.

[0063] This method calculates a stable HFO power increase latency by averaging (additive averaging) EEG recordings in time windows based on multiple tagged epileptic spikes and applying statistical analysis. Furthermore, HFOs associated with tagged epileptic spikes are considered to be highly correlated with epilepsy. Therefore, by using the electroencephalogram signals immediately preceding these epileptic spikes as a reference for statistical analysis, for example, we have achieved the efficient exclusion of artifacts such as physiological HFOs associated with eye movements.

[0064] For the reasons stated above, it is considered that the selection of electrodes used in the electroencephalogram tagging process in this method has very little effect on the determination of SOZ and EZ. Figure 10 illustrates the correspondence between the electrodes used in the electroencephalogram (EEG) tagging process for a given EEG signal and the electrode-specific significance determination spectra calculated from the results of each tagging process. In this figure, the left side shows the electroencephalogram (EEG) signals before and after the spike wave that were subjected to spike wave tagging (electrodes Ch.B4 to Ch.B14). In the EEG signals, the vertical gray lines correspond to the spike wave detection times when spike wave tagging was performed on the electrodes indicated by the accompanying arrows. In this case, from the EEG signals for the same time period, spike waves were detected on electrodes Ch.B5, Ch.B6, Ch.B12, and Ch.B14, in that order from left to right. To the right of the electroencephalogram (EEG) signal, a list of electrode-specific significance determination spectra obtained for each electrode used in the EEG tagging process is shown. Hereafter, the electrode-specific significance determination spectrum obtained by performing spike tagging on the channel x signal will be referred to as the "electrode-specific significance determination spectrum based on Ch.x". In this figure, the first row from top to bottom shows the electrode-specific significance determination spectra based on Ch.B5, the second row shows the electrode-specific significance determination spectra based on Ch.B6, the third row shows the electrode-specific significance determination spectra based on Ch.B12, and the fourth row shows the electrode-specific significance determination spectra based on Ch.B14. Each column, from left to right, shows the calculated significance determination spectra for electrodes Ch.B5, Ch.B6, Ch.B12, and Ch.B14, respectively. When calculating the HFO power increase latency for each electrode from these electrode-specific significance determination spectra, the smallest HFO power increase latency was found in electrode Ch.B5 of the first column, regardless of which electrode was subjected to EEG tagging.

[0065] Furthermore, this method allows for the determination of SOZ and EZ using electroencephalogram (EEG) signals during the interictal period of epileptic seizures, eliminating the need to wait for an epileptic seizure to occur when acquiring EEG signals. This enables analysis even in children and patients with intellectual disabilities where long-term intracranial electrode placement is difficult. In addition, it is possible to place new intracranial electrodes as needed during epilepsy surgery and perform analysis in real time. Furthermore, it is possible to perform analysis using non-invasive scalp electroencephalography (EEG) recording instead of intracranial electrodes, which are invasive and burdensome for patients.

[0066] [Information Processing Results] Figures 11-13 show examples of SOZ and EZ determination results in patients who underwent intracranial electrode placement and electroencephalogram (EEG) recording for the purpose of epilepsy surgery. In these patients, seizures had disappeared for more than two years after partial cortical resection, suggesting that the resection area included the seizure focus. The processing time required for the determination is approximately 5 minutes, for example, when analyzing 90 minutes of electroencephalogram (EEG) signals and automatically tagging all electrodes with spike waves (approximately 30-50 tags per electrode).

[0067] Figure 11 shows an example of the SOZ and EZ determination results for the first target patient. In the following figures, each electrode is represented by a circle, with a number inside the circle to identify the electrode. In addition, the suffixes "A" and "B" are placed near the electrodes to identify the subdural electrodes. Furthermore, the seizure focus used to determine the resection area during surgery is represented by a double circle, and the resection area is represented by a dashed line. The seizure focus used to determine the extent of resection during surgery was the seizure focus estimated from the electroencephalogram (EEG) during a seizure using existing methods. In the legend, the SOZ determined from the HFO power increase latency calculated based on the electrode-specific significance determination spectrum based on Ch.x is referred to as "SOZ based on electrode x". Furthermore, the EZ determined to have a threshold latency of "L" in the EZ determination process, based on the HFO power increase latency calculated based on the electrode-specific significance determination spectrum based on Ch.x, is referred to as "EZ based on electrode x at threshold latency L".

[0068] This figure shows that the SOZ determined by the information processing device 1 overlaps with the seizure focus, and the EZ extending from the SOZ is located within the surgical resection area. It also shows that by appropriately increasing the threshold latency "L", the extent of the EZ can be made to better match the surgical resection area.

[0069] Figure 12 shows an example of the SOZ and EZ determination results for the second patient group. In the second patient group, electroencephalogram signals were acquired using subdural electrodes "A" and "B," as well as deep electrodes "a" and "b." In this diagram, the determination results for SOZ and EZ within the range of subdural electrode "A" are included in a portion of the surgical resection area, but within the range of subdural electrode "B" they do not coincide with the surgical resection area. Therefore, Figure 13 shows the processing results when the electrode used for spike wave tagging is changed from "Ch.A3" to "Ch.A18". In this figure, it can be seen that the electrode determined to be an SOZ based on electrode A3 was Ch.A3, while the electrode determined to be an SOZ based on electrode A18 was Ch.B46. Also, the spread of the EZ differs depending on the position of the SOZ, so it is different from the example in Figure 12. In the second group of patients, multiple brain malformations that caused epilepsy were found, which is thought to be why multiple SOZs and associated EZs were identified. Combining the assessment results in Figures 12 and 13, it can be seen that the areas identified as SOZs and EZs closely match the surgically resected areas. Thus, this method makes it possible to accurately determine SOZ even when it is distributed across multiple regions of the brain.

[0070] [Effects and Effects of the Embodiment] The information processing device 1 in this embodiment (for example, an example of an information processing device that supports epilepsy diagnosis) comprises: an electroencephalogram (EEG) signal acquisition unit 110 (for example, an example of an acquisition unit) that acquires EEG signals from a group of head-mounted electrodes 5 (for example, an example of a group of electrodes mounted on the head); a high-frequency EEG generation time calculation unit 120 (for example, an example of a calculation unit) that calculates the HFO power increase latency (for example, an example of the time at which high-frequency EEGs are detected in each of the multiple electrodes) from the EEG signals based on a bandpass filter and a Fourier transform (for example, an example of a time-frequency analysis); and a lesion area determination unit 130 (for example, an example of a determination unit) that determines, based on time, whether each brain region corresponding to each of the multiple electrodes corresponds to an epileptic seizure initiation area or an epileptogenic region. This configuration allows for accurate and rapid determination (estimation) of the epileptic seizure initiation site and epileptogenic region based on the time at which high-frequency brain waves are detected, obtained by performing time-frequency analysis of the electroencephalogram (EEG) signal. Furthermore, as mentioned above, the EEG signal can be either an EEG signal from an epileptic seizure or an EEG signal from a non-seizure state, allowing for determination (estimation) of the epileptic seizure initiation site and epileptogenic region from the EEG signal regardless of whether it is during or after an EEG.

[0071] Furthermore, the information processing device 1 shows an example configuration in which the spike wave detection unit 121 (for example, an example of a calculation unit) detects the tagging time (for example, an example of the rising edge of a spike wave) from the electroencephalogram signal, and performs analysis in time-frequency analysis using a time window based on the rising edge. This allows for more robust detection of noise, such as high-frequency brain waves caused by factors other than epilepsy, in the electroencephalogram (EEG) signal by using a time window based on the rise time of the spike wave.

[0072] Furthermore, the information processing device 1 shows an example configuration in which the statistical analysis unit 127 (for example, an example of a calculation unit) calculates time based on the bootstrap method (for example, an example of a statistical method). As a result, by incorporating statistical methods, the validity of the calculated time can be further improved.

[0073] Furthermore, the information processing device 1 shows an example configuration in which the statistical analysis unit 127 (for example, an example of a calculation unit) calculates time based on determining whether the electrode-specific frequency spectrum (for example, an example of the results of time-frequency analysis) is statistically significant in a segmented mesh (for example, an example of a set time width and frequency width). This allows for improved accuracy in calculating time by determining statistical significance between the set time range and frequency range.

[0074] Furthermore, the information processing device 1 shows an example configuration in which, in the lesion area determination unit 130 (for example, an example of a determination unit), if the HFO power increase latency (for example, an example of time) is the shortest among multiple electrodes, the brain region corresponding to the electrode is determined to be the epileptic seizure initiation site, and if the time is less than or equal to the HFO power increase latency in SOZ + threshold latency (for example, an example of a set value), the brain region corresponding to the electrode is determined to be an epileptogenic region. According to this method, the epileptic seizure initiation site and the epileptogenic region can be easily and quickly determined based on time.

[0075] [First variation] In the above embodiment, the information processing device 1 specifies a particular electrode for spike wave detection in the spike wave tagging process and outputs the SOZ·EZ determination result based on that electrode, but is not limited to this.

[0076] The information processing device 1, for example, repeats steps S120 to S170 in Figure 3 at an arbitrary threshold latency, and determines, for example, the SOZ and EZ when spike wave tagging processing is performed on each electrode. The information processing device 1 may then determine whether SOZ and EZ are present in all electroencephalogram signals based on the SOZ and EZ determination results for each electrode, for example, by taking a majority vote for each electrode.

[0077] [Second variation] In the above embodiment, in the time-frequency analysis process, the time-frequency analysis unit 125 calculated the electrode-specific frequency spectrum based on the frequency spectrum during a predetermined window period before and after the tag "x(t)" of the bandpass electroencephalogram signal, but is not limited to this. For example, the time-frequency analysis unit 125 may calculate the electrode-specific frequency spectrum based on the relative power increase value from the bandpass electroencephalogram signal in the reference interval, which is 300 milliseconds to 500 milliseconds immediately preceding the tagged spike wave, within a predetermined window period before and after the tag "x(t)" of the bandpass electroencephalogram signal.

[0078] For example, the power of the frequency spectrum of each electrode at a predetermined time frequency may be calculated according to the following formula. Power of the frequency spectrum of each electrode at a given time frequency = Average power of the frequency spectrum at a given time frequency ÷ Average power of the reference interval

[0079] A more concrete example would be a case where, for instance, in a certain Ch.x, tags T1, T2, and T3 are assigned, and in each of these tags, • Tag T1: Power in the reference section = "1", power in the frequency spectrum at a given time frequency = "2.5" • Tag T2: Power in the reference section = "10", power in the frequency spectrum at a given time frequency = "20" • Tag T3: Power in the reference section = "100", power in the frequency spectrum at a given time frequency = "150" In this case, the average power value of the reference interval is "(1+10+100)÷3" = "37", and the summed average power of the frequency spectrum at a given time frequency is "(2.5+20+150)÷3" = "57.5", therefore, Power of the frequency spectrum of each electrode at a given time frequency = "57.5 ÷ 37" = "1.554" This is the result. In other words, in this case, the power in the electrode-specific frequency spectrum indicates that the power increase from the reference interval at a given time frequency is approximately "155%".

[0080] [Examples] Next, embodiments of epilepsy surgery support devices, terminals, and electronic devices (electronic equipment) to which the above-described information processing device 1 is applied, or which are equipped with the above-described information processing device 1, will be described. Here, as an example, an embodiment of an epilepsy surgery support device will be described. However, it goes without saying that the embodiments to which the present invention can be applied are not limited to this embodiment.

[0081] Figure 14 shows an example of the functional configuration of the epilepsy surgery support device 10. The epilepsy surgery support device 10 includes, for example, a processing unit 100, a storage unit 200, a group of head-mounted electrodes 5, an operation unit 310, a display unit 320, and a communication unit 330.

[0082] The processing unit 100 is a processing unit that comprehensively controls each part of the epilepsy surgery support device 10 according to various programs such as system programs stored in the memory unit 200, and performs various processing related to video editing, and is configured to have processing circuits such as a CPU, GPU, DSP, ASIC, FPGA, etc.

[0083] The processing unit 100 includes, as its main functional units, an electroencephalogram (EEG) signal acquisition unit 110, an HFO (High Frequency Occurrence) time calculation unit 120, a lesion area determination unit 130, and a display control unit 140. The HFO time calculation unit 120 includes, as its functional units, a spike wave detection unit 121, a bandpass filter unit 123, a time-frequency analysis unit 125, and a statistical analysis unit 127. The display control unit 140, for example, has the function of receiving the output (judgment result) from the lesion area determination unit 130, visualizing the judgment result, and displaying it on the display unit 320. The other functional units correspond to, for example, the functional units provided by the information processing device 1 in Figure 1.

[0084] The storage unit 200 is a storage device that includes memory circuits such as ROM (Read Only Memory), RAM (Random Access Memory), and flash memory, as well as a hard disk drive and a magneto-optical disk drive.

[0085] The memory unit 200 stores, for example, an electroencephalogram (EEG) analysis application program 210 and an EEG signal temporary storage unit 220.

[0086] The electroencephalogram (EEG) analysis application program 210 is a program that is read by the processing unit 100, for example, and executed as an EEG analysis application process.

[0087] The electroencephalogram (EEG) signal temporary storage unit 220 is a buffer that stores, for example, the EEG signals acquired by the EEG signal acquisition unit 110.

[0088] The operation unit 310 is configured to have input devices for the user to perform various operations on the epilepsy surgery support device 10, such as operation buttons and operation switches. The operation unit 310 may also have a touch panel (not shown) that is integrally configured with the display unit 320, and this touch panel may function as an input interface between the user and the epilepsy surgery support device 10. The operation unit 310 may output operation signals to the processing unit 100 according to user operations, for example. An input device that accepts sound (including voice) input as user input may also be configured.

[0089] The display unit 320 is a display device configured with, for example, an LCD (Liquid Crystal Display) or an OLED (Organic Electro-luminescence Display), and performs various displays based on the display signals output from the display control unit 140.

[0090] The communication unit 330 is a communication device for sending and receiving information used within the device to and from an external information processing device. Various communication methods can be applied to the communication unit 330, including wired connection via a cable compliant with a predetermined communication standard such as Ethernet or USB (Universal Serial Bus), wireless connection using wireless communication technology compliant with a predetermined communication standard such as Wi-Fi (registered trademark) or 5G (fifth-generation mobile communication system), and connection using short-range wireless communication such as Bluetooth (registered trademark).

[0091] The processing unit 100 of the epilepsy surgery support device 10 performs electroencephalogram (EEG) signal analysis processing according to, for example, the EEG analysis application program 210 stored in the memory unit 200.

[0092] Figure 15 is a flowchart showing an example of the electroencephalogram signal analysis procedure in this embodiment. When the electroencephalogram (EEG) signal acquisition unit 110 acquires EEG signals from, for example, the head-mounted electrode group 5 (S110), the HFO generation time calculation unit 120 calculates the HFO power increase latency at each electrode, for example, according to steps S120 to S150 in Figure 3.

[0093] The electroencephalogram (EEG) signal acquisition unit 110 may acquire EEG signals by receiving EEG signals acquired by an external processing device (not shown) via a communication unit 330, for example. Furthermore, the processing unit 100 may perform some of the processing steps of each step via the communication unit 330 using an external processing device (not shown).

[0094] Furthermore, the HFO generation time calculation unit 120 may, for example, perform a spike tagging process on a specified electrode based on a user operation on the operation unit 310 and calculate the HFO power increase latency at each electrode based on the processing result, or it may perform a spike tagging process on all electrodes and calculate the HFO power increase latency at each electrode based on the processing result.

[0095] Then, the lesion area determination unit 130 performs SOZ determination processing (S160), and stores, for example, the electrode determined to be SOZ and the electrode used for spike wave tagging processing in the storage unit 200.

[0096] Next, the processing unit 100 executes the threshold latency setting process (S210). In the threshold latency setting process, the processing unit 100 receives the threshold latency based, for example, on user operation on the operation unit 310. Alternatively, in the threshold latency setting process, the processing unit 100 may select the threshold latency to be used in the EZ determination process from a plurality of pre-set threshold latencies (for example, "5 milliseconds", "10 milliseconds", and "15 milliseconds").

[0097] Then, the lesion area determination unit 130 performs an EZ determination process (S170) based on the threshold latency set in the threshold latency setting process, and stores, for example, the electrode determined to be EZ, the electrode used for the spike wave tagging process, and the threshold latency in the storage unit 200.

[0098] Subsequently, the processing unit 100 performs brain image acquisition processing (S220). In brain image acquisition processing, the processing unit 100 acquires brain image diagnostic information from the image diagnostic device 400 via the communication unit 330, for example. Here, the diagnostic imaging device 400 may use various equipment such as MRI (Magnetic Resonance Imaging), CT (Computed Tomography), SPECT (Single Photon Emission Computed Tomography), and PET (Positron Emission Tomography). The processing unit 100 may, prior to processing, acquire brain image diagnostic information from the image diagnostic device 400 and store it in the storage unit 200.

[0099] Then, the display control unit 140 executes the lesion area display processing (S230). In the lesion area display processing, the display control unit 140 superimposes, for example, brain image diagnostic information and the SOZ / EZ judgment results stored in the memory unit 200 to generate an image for determining the epilepsy resection area. The display control unit 140 then displays the image for determining the epilepsy resection area on the display unit 320.

[0100] Figures 16 and 17 show an example of a display screen including an image for determining the epilepsy resection area displayed on the display unit 320. Figure 16 shows an example of the result display screen for an electroencephalogram (EEG) analysis application, with the application name "Ictal Region Analyzer" displayed at the top. The left pane of the judgment result display screen is configured to show, for example, the name and age of the patient, and below that, the judgment result display area IJR1 for displaying the image for determining the epilepsy resection area is displayed. The right pane is configured to display a status display showing the analysis status, a tagging electrode selection area TSR1 for selecting electrodes to be targeted for spike wave tagging, and a threshold latency setting area LSR1 for setting threshold latencies in the threshold latency setting process. Below that, a legend display area CIR1 is displayed for showing the legend of the judgment results in the judgment result display area IJR1.

[0101] This diagram shows that in the tagging electrode selection region TSR1, the spikes observed on the "Ch.A12" electrode are targeted for tagging. For example, by tapping the arrows displayed on the left and right of the tagging electrode selection region TSR1, you can change the electrode used in the spike tagging process. Furthermore, in the threshold latency setting area LSR1, it is possible to select a threshold latency from pre-set values ​​such as "5 milliseconds," "10 milliseconds," "15 milliseconds," and "20 milliseconds," and it is currently displayed that "5 milliseconds" is selected. The judgment result display area IJR1 displays an image for determining the epilepsy resection area based on the results of the SOZ and EZ judgment processes when the selected electrode for spike wave tagging is "Ch.A12" and the threshold latency is "5 milliseconds". In this figure, for example, superimposed on the MRI image, the SOZ marker (e.g., a star), the EZ marker (e.g., an upward triangle), and other markers not determined to be SOZ or EZ (e.g., white circles) are displayed in the area corresponding to each electrode position, as shown in the legend display area CIR1. Note that the markers may be distinguishable by color rather than shape. Furthermore, in the judgment result display area IJR1, the recommended brain resection area boundary, calculated based on the SOZ and EZ judgment results, is superimposed on the MRI image as a dashed line. This recommended brain resection area boundary may be determined, for example, based on the clustering boundary (discrimination boundary) between markers judged as SOZ / EZ and markers that are not judged.

[0102] For example, if the threshold latency setting area LSR1 is tapped, the display will transition to the judgment result display screen shown in Figure 17. On this screen, in the judgment result display area IJR1, more EZ markers are displayed near the SOZ markers. The user of the epilepsy surgery support device 10 can appropriately and quickly identify the brain resection area by referring to the determination result display area IJR1, which displays the electrodes to be tagged and the threshold latency in the tagging electrode selection area TSR1 and the threshold latency setting area LSR1.

[0103] Furthermore, the brain resection area boundary may be changed and displayed when the electrodes and threshold latency to be tagged are selected. Additionally, the electrodes to be tagged may be input by, for example, tapping the tagging electrode selection area TSR1. The threshold latency may be input by, for example, tapping the threshold latency setting area LSR1.

[0104] [Effects and Effects of the Examples] The epilepsy surgery support device 10 of this embodiment can be used to obtain the same functions and effects as those of the previously described embodiment.

[0105] Furthermore, the epilepsy surgery support device 10 of this embodiment (for example, an example of an epilepsy surgery support device that assists in epilepsy surgery) shows an example configuration comprising an information processing device 1 and a display control unit 140 (for example, an example of a display unit) that displays the determination result of a lesion area determination unit 130 (for example, an example of a determination unit). This configuration allows users of the epilepsy surgery support device to efficiently understand the determination of the epileptic seizure initiation site and the epileptogenic region by checking the judgment results displayed on the display unit.

[0106] Furthermore, this embodiment shows an example of a configuration in which the display unit displays an image for determining the epilepsy resection area (for example, an example of a determination result) superimposed on brain image diagnostic information (for example, an example of brain image diagnostic information). According to this, by checking the judgment results displayed on the screen, users of the epilepsy surgery support device can efficiently understand the judgment results regarding the epileptic seizure initiation site and the epileptogenic region.

[0107] Furthermore, this embodiment shows an example of a configuration that includes a touch panel (for example, an example of an input unit) that accepts set values. With this configuration, users of the epilepsy surgery support device can input desired settings through the input unit, allowing them to adjust and confirm the brain regions identified as the epileptic seizure initiation site and epileptogenic areas.

[0108] Furthermore, this embodiment shows an example of a configuration in which the calculation unit detects the rising edge of epileptic spikes from the electroencephalogram signal, the time-frequency analysis is performed using a time window based on the rising edge, and the system includes a touch panel (for example, an example of an input unit) that accepts electrodes for the electroencephalogram signal detecting the rising edge. This configuration allows users of the epilepsy surgery support device to arbitrarily specify the electrodes that detect the rise of spike waves, and to check the judgment results according to the state of the electroencephalogram signal.

[0109] [Example 1: Modified Version] In the above embodiment, the user is allowed to specify (select) an arbitrary threshold latency, but the invention is not limited to this. Figure 18 shows an example of the result display screen in this modified example. On this screen, the threshold latency setting area LSR1, which was configured to be displayed in the right pane, is not shown. The legend display area CIR2 shows the SOZ marker (e.g., a star), the EZ marker when the threshold latency is set to "5 milliseconds" (e.g., a pentagon), the EZ marker when the threshold latency is set to "10 milliseconds" (e.g., an upward-pointing triangle), the EZ marker when the threshold latency is set to "15 milliseconds" (e.g., a square), the EZ marker when the threshold latency is set to "20 milliseconds" (e.g., a downward-pointing triangle), and other markers that are not determined to be SOZ or EZ (e.g., a white circle). The judgment result display area IJR2 shows the judgment result in accordance with the legend shown in the legend display area CIR2, where the EZ expands as the threshold latency increases from SOZ. In this way, by displaying the judgment results for multiple threshold latencies together in the judgment result display area IJR2, users can more intuitively grasp the extent of the EZ (Easy Zone).

[0110] In this modified example, it may be possible to set upper and lower limits and time intervals for the threshold latency used in the determination.

[0111] Furthermore, the legend can be displayed using color changes, or a combination of color and shape. For example, SOZ could be displayed in red, and EZ could change from purple to blue as the threshold latency increases.

[0112] This modified example shows an example configuration in which the determination unit determines the brain region corresponding to an electrode as the epileptic seizure initiation site if the time is the shortest among multiple electrodes, and determines the brain region corresponding to an electrode as an epileptogenic region if the time is less than or equal to a set value, and the display unit displays a pentagon mark (for example, an example of the first display mode) as a legend marker indicating EZ (for example, an example of the determination result) when the set value is a threshold latency of "5 milliseconds" (for example, an example of the first set value), and displays the determination result as an upward-pointing triangle mark (for example, an example of the second display mode) which is different from the first display mode when the set value is a threshold latency of "10 milliseconds" (for example, an example of the second set value). According to this, users of the epilepsy surgery support device can easily understand the epilepsy-prone areas and corresponding brain regions determined for each set value by checking the manner of the judgment results displayed on the display unit.

[0113] [Second variation of the example] The above embodiment shows an example in which the user specifies one electrode to be used for the spike wave tagging process, but it is not limited to this. Figure 19 shows an example of the result display screen in this modified example. On this screen, in the right pane, the Tagging Electrode Selection Area TSR2 is configured to display electrodes in which spikes were detected during the spike tagging process, for example, in a vertical arrangement. Furthermore, in the Tagging Electrode Selection Area TSR2, tapping a displayed electrode will add a check mark to the left of the electrode display. This screen shows that, for example, spike wave tagging was successful for electrodes "Ch.A12", "Ch.A17", ..., "Ch.B4", and that "Ch.A12" and "Ch.A17" were selected based on user input.

[0114] Furthermore, in the threshold latency setting region LSR2, for example, the threshold latency can be set individually for each of the one or more electrodes selected in the tagging electrode selection region TSR2 (in this screen, "Ch.A12" and "Ch.A17").

[0115] The legend display area CIR3 is configured to display a legend based on the selection results in the tagged electrode selection area TSR2 and the threshold latency setting area LSR2. On this screen, for example, the SOZ marker based on "Ch.A12" (e.g., a star), the SOZ marker based on "Ch.A17" (e.g., a diagonal star), the EZ marker based on "Ch.A12" with a threshold latency of "5 milliseconds" (e.g., an upward-pointing triangle), the EZ marker based on "Ch.A17" with a threshold latency of "10 milliseconds" (e.g., a downward-pointing triangle), and other markers that are not determined to be SOZ or EZ (e.g., a white circle) are displayed.

[0116] The judgment result display area IJR3 shows the judgment results for SOZ and EZ, in accordance with the legend shown in the legend display area CIR3. In addition, in areas where the judgment results overlap, multiple markers are displayed, for example, by overlapping or side by side.

[0117] In this way, by displaying the judgment results based on the spike wave tagging process for multiple electrodes together in the judgment result display area IJR3, users can verify the SOZ and EZ judgment results more appropriately and efficiently.

[0118] Furthermore, the legend may be displayed using, for example, a combination of color and shape. For instance, the shape of electrodes may be changed to distinguish them in the spike wave tagging process, and the color may be changed to distinguish them in the threshold latency. For example, on this screen, markers based on "Ch.A12" may be represented by an "upper triangle," markers based on "Ch.A17" by a "lower triangle," SOZ may be displayed in "red," EZ with a threshold latency of up to "5 milliseconds" in "green," and EZ with a threshold latency of up to "10 milliseconds" in "blue."

[0119] [Third modified example of the embodiment] The above embodiment shows an example where the display unit is configured as a touch panel, but it is not limited to this. For example, the display unit 320 may be an AR (Augmented Reality) glasses. In this case, for example, the processing unit 100 of the epilepsy surgery support device 10 acquires intracranial images using an imaging unit (not shown) mounted on AR glasses during epilepsy surgery. The display control unit 140 may then, for example, in the lesion area display processing, display the SOZ / EZ determination results and recommended brain resection area boundaries within the AR glasses based on the positions of each electrode acquired based on the intracranial images.

[0120] By displaying the information in this way, surgeons can perform surgery while confirming the SOZ / EZ assessment results and recommended brain resection area boundaries projected onto the actual intracranial view. This is expected to improve seizure resolution rates in epilepsy surgery and enhance safety by shortening surgical time. [Explanation of symbols]

[0121] 1. Information Processing Device 5 Head-mounted electrode group 10. Epilepsy surgery support device 110 Electroencephalogram (EEG) signal acquisition unit 120 HFO generation time calculation unit 130 Lesion area determination unit

Claims

1. An information processing device that supports the diagnosis of epilepsy, An acquisition unit that acquires electroencephalogram (EEG) signals from multiple electrodes placed on the head, A calculation unit calculates the time at which high-frequency brain waves in a predetermined frequency band of 80 Hz or higher are detected in each of the plurality of electrodes, based on time-frequency analysis of the electroencephalogram signals. A determination unit that determines, based on the aforementioned time, whether each brain region corresponding to each of the plurality of electrodes corresponds to the epileptic seizure initiation site or epileptogenic region, Equipped with, The determination unit determines that the brain region corresponding to the electrode is the epileptic seizure initiation site if the time is the shortest among the plurality of electrodes, and determines that the brain region corresponding to the electrode is an epileptogenic region if the time is less than or equal to a set value. Information processing device.

2. An information processing apparatus according to claim 1, The calculation unit detects the rising edge of a spike wave from the electroencephalogram signal, The aforementioned time-frequency analysis is performed using a time window based on the rising edge. Information processing device.

3. An information processing apparatus according to claim 1, The calculation unit calculates the time based on a statistical method. Information processing device.

4. An information processing apparatus according to claim 3, The calculation unit calculates the time based on determining whether the results of the time-frequency analysis are statistically significant in the set time width and frequency width. Information processing device.

5. An epilepsy surgery support device that assists in epilepsy surgery, The information processing apparatus according to claim 1, A display unit that displays the determination result of the determination unit, An epilepsy surgery support device equipped with [feature / feature].

6. The epilepsy surgery support device according to claim 5, The display unit displays the judgment result superimposed on the brain image diagnostic information. Epilepsy surgery support device.

7. An epilepsy surgery support device according to claim 5 or claim 6, The display unit displays the determination result in a first display mode if the setting value is a first setting value, and displays the determination result in a second display mode different from the first display mode if the setting value is a second setting value different from the first setting value. Epilepsy surgery support device.

8. The epilepsy surgery support device according to claim 7, It includes an input unit that accepts the aforementioned set value, Epilepsy surgery support device.

9. The epilepsy surgery support device according to claim 7, The calculation unit detects the rising edge of epileptic spikes from the electroencephalogram signal, The aforementioned time-frequency analysis is performed using a time window based on the rising edge. The system includes an input unit that receives the electrodes for detecting the rise of the electroencephalogram signal, Epilepsy surgery support device.

10. A program to be run by a computer that assists in the diagnosis of epilepsy, Acquiring electroencephalogram (EEG) signals from multiple electrodes placed on the head, Based on time-frequency analysis of the electroencephalogram signals, the time at which high-frequency electroencephalograms in a predetermined frequency band of 80 Hz or higher are detected in each of the plurality of electrodes is calculated. Based on the aforementioned time, it is determined whether each brain region corresponding to each of the multiple electrodes corresponds to the epileptic seizure initiation site or the epileptogenic region. If the aforementioned time is the shortest among the multiple electrodes, the brain region corresponding to the electrode is determined to be the epileptic seizure initiation site, and if the aforementioned time is less than or equal to a set value, the brain region corresponding to the electrode is determined to be the epileptogenic region. A program to cause the aforementioned computer to execute.

11. An information processing method for an information processing device that supports the diagnosis of epilepsy, The aforementioned information processing device acquires electroencephalogram signals from multiple electrodes placed on the head, The information processing device calculates, based on time-frequency analysis, the time at which high-frequency brain waves in a predetermined frequency band of 80 Hz or higher are detected in each of the plurality of electrodes from the electroencephalogram signal. The information processing device determines, based on the time, whether each brain region corresponding to each of the plurality of electrodes corresponds to the epileptic seizure initiation site or the epileptogenic region, The information processing device determines that if the time is the shortest among the plurality of electrodes, the brain region corresponding to the electrode is the epileptic seizure initiation site, and if the time is less than or equal to a set value, the brain region corresponding to the electrode is the epileptogenic region. Information processing methods including

Images