WO2018124841A1 - Real-time seizure brainwave early detection method - Google Patents

Abstract

An embodiment of the present invention provides a real-time seizure brainwave early detection method including: a step for receiving pre-measured brainwave signals from each of a plurality of measurement areas of an examinee; a step for calculating power spectrum densities (PSDs) of the brainwave signals; a step for analyzing the PSDs of the brainwave signals to detect a specific frequency band including the frequency at which the separation width between a reference signal and a seizure signal is largest; a step for extracting features of the PSDs of the brainwave signals corresponding to the detected specific frequency band to classify one or more detection areas including effective features from among the plurality of measurement areas; and a step for measuring real-time brainwave signals corresponding to the one or more detection areas and the specific frequency band of the examinee, and determining whether the examinee is having seizure on the basis of the real-time brainwave signals.

Description

Real-time spasm EEG early detection method

Embodiments of the present invention relate to a method for detecting real-time spasm electroencephalograms measured from an intracranial EEG electrode or a scalp EEG electrode.

An epileptic seizure causes a change in behavior or consciousness, such as a loss of consciousness or shaking arms or legs for a few seconds or minutes during a seizure. Epilepsy is a disorder of the brain that is usually abnormally active. Epilepsy may involve unfamiliar feelings, emotions, or behaviors, or sometimes seizures, such as racing, muscle spasms, or loss of consciousness. In some patients with epilepsy, seizures occur only occasionally, while in others, more than 100 times a day. These seizures vary in number of seizures according to individual differences, and there is a difference in the risk of seizures. The risk of seizures is higher if the patient with epilepsy attacks has diseases such as hypoxia (chronic obstructive pulmonary disease, severe asthma), meningitis (meningitis), encephalitis, and brain tumors.

The number of epilepsy patients is reported to be more than 2 million in the United States, and 80% of these epilepsy patients are reported to be able to control seizures by medicine and surgery. However, the remaining 25-30% of patients continue to experience seizures. In the UK, there are 600,000 people with epilepsy. About 500 of these people are injured by sudden seizures and are killed by such injuries. Although there is no accurate statistics in Korea, it is estimated that there are about 30 to 400,000 epilepsy patients.

Meanwhile, EEG (EEG) is a measurement of the current flow in the living body caused by the synchronized activity of nerve cells occurring on the surface of the brain cortex using an electrode (electrode) attached to the skin of the scalp or surgically in the cranial cavity It can be measured by inserting an EEG electrode. Conventionally, epilepsy data analysis using EEG has been used for epilepsy diagnosis, seizure detection and prediction, but to accurately analyze EEG data at all or EEG frequencies on all EEG electrodes for the time before and after the seizure occurs. This takes a long time, making it difficult to detect convulsions early in the onset of convulsions.

In order to solve this problem, embodiments of the present invention automatically select a reference region and a frequency range capable of determining whether a seizure is detected from a measurement region of a plurality of intracranial EEG electrodes or scalp EEG electrodes in a fast time. The present invention aims to provide an early detection method for real-time spasm EEG.

According to an embodiment of the present invention, receiving pre-measured EEG signals from each of a plurality of measurement regions for a subject, calculating a power spectrum density (PSD) for the EEG signals, and the EEG Analyzing the power spectral density of the signals to detect a specific frequency band including a frequency having a largest separation between a baseline and a seizure signal, wherein the specific frequency band corresponds to the detected specific frequency band. Extracting a feature of the power spectral density (PSD) of the EEG signals to classify one or more detection areas including valid shapes of the plurality of measurement areas; and the one or more detection areas and the specific frequency band of the examinee. Measure real-time EEG signals corresponding to the test, and examine the blood based on the real-time EEG signals. That includes the steps of determining whether a seizure, and provides real-time EEG seizure early detection methods.

The real-time spasm EEG early detection method according to embodiments of the present invention detects a specific frequency band including a frequency having the largest separation between the reference signal and the seizure signal, and has an effective shape of power spectral density from a plurality of measurement regions. By classifying one or more detection areas and analyzing the real-time EEG signals of the examinee, it is possible to quickly and accurately detect the seizure at the initial time of the seizure occurrence of the examinee.

1 is a flow chart sequentially showing a real-time spasm EEG early detection method according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating power spectral density (PSD) for each frequency of the EEG signals measured from a plurality of subjects using the Welch method.

FIG. 3 is a conceptual diagram schematically illustrating a plurality of measurement regions in which an EEG electrode is disposed in the cranial cavity or the scalp of the examinee P. Referring to FIG.

FIG. 4 is a diagram illustrating the shapes of power spectral densities of reference signals and seizure signals extracted from the plurality of measurement areas of the examinee B of FIG. 2.

FIG. 5 is an exemplary diagram illustrating segregation of a seizure signal and a reference signal using a sparse probit classifier SPC from a feature of a power spectral density PSD extracted from one or more detection regions of an examinee.

According to an embodiment of the present invention, receiving pre-measured EEG signals from each of a plurality of measurement regions for a subject, calculating a power spectrum density (PSD) for the EEG signals, and the EEG Analyzing the power spectral density of the signals to detect a specific frequency band including a frequency having a largest separation between a baseline and a seizure signal, wherein the specific frequency band corresponds to the detected specific frequency band. Extracting a feature of the power spectral density (PSD) of the EEG signals to classify one or more detection areas including valid shapes of the plurality of measurement areas; and the one or more detection areas and the specific frequency band of the examinee. Measure real-time EEG signals corresponding to the test, and examine the blood based on the real-time EEG signals. That comprises determining whether a seizure, and provides real-time EEG seizure early detection methods.

In one embodiment of the present invention, the EEG signals corresponding to the at least one detection area and the specific frequency band of the examinee are separated into the reference signal and the seizure signal, and the reference signal and the seizure signal are separated. The method may further include selecting a reference range.

In an embodiment of the present disclosure, the determining of the seizure of the examinee may be determined to be a seizure when a first signal out of the reference range exists among the measured real-time EEG signals.

In one embodiment of the present invention, the step of classifying one or more detection areas by detecting a shape in the specific frequency band may be performed by using a sparse probit classifier (SPC). Signal and the seizure signal.

In one embodiment of the present invention, the detecting of the specific frequency band may detect the specific frequency band by using a Welch method.

In one embodiment of the present invention, the EEG signals corresponding to the plurality of measurement areas can be measured using EEG electrodes arranged in a matrix form measured in the intracranial electrode or the scalp electrode of the examinee. have.

An embodiment of the present invention provides a computer program stored in a recording medium for executing the method of any one of claims 1 to 6 using a computer.

Other aspects, features, and advantages other than those described above will become apparent from the following drawings, claims, and detailed description of the invention.

As the invention allows for various changes and numerous embodiments, particular embodiments will be illustrated in the drawings and described in detail in the written description. Effects and features of the present invention, and methods of achieving them will be apparent with reference to the embodiments described below in detail together with the drawings. However, the present invention is not limited to the embodiments disclosed below but may be implemented in various forms.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings, and the same or corresponding components will be denoted by the same reference numerals, and redundant description thereof will be omitted. .

In the following embodiments, the terms first, second, etc. are used for the purpose of distinguishing one component from other components rather than a restrictive meaning.

In the following examples, the singular forms "a", "an" and "the" include plural forms unless the context clearly indicates otherwise.

In the following examples, the terms including or having have meant that there is a feature or component described in the specification and does not preclude the possibility of adding one or more other features or components.

In the following embodiments, when a part such as a film, a region, a component, or the like is on or on another part, not only is it directly above the other part, but also another film, a region, a component, etc. is interposed therebetween. It also includes cases where there is.

In the drawings, components may be exaggerated or reduced in size for convenience of description. For example, the size and thickness of each component shown in the drawings are arbitrarily shown for convenience of description, and thus the present invention is not necessarily limited to the illustrated.

In the case where an embodiment may be implemented differently, a specific process order may be performed differently from the described order. For example, two processes described in succession may be performed substantially simultaneously or in the reverse order of the described order.

In the following embodiments, when a film, a region, a component, or the like is connected, not only the film, the region, and the components are directly connected, but also other films, regions, and components are interposed between the film, the region, and the components. And indirectly connected. For example, in the present specification, when the film, the region, the component, and the like are electrically connected, not only the film, the region, the component, and the like are directly electrically connected, but other films, the region, the component, and the like are interposed therebetween. This includes indirect electrical connections.

1 is a flow chart sequentially showing a real-time spasm electroencephalogram early detection method according to an embodiment of the present invention, Figure 2 is a power signal for each frequency by using the Welch method for the EEG signals measured from a plurality of subjects It is a figure represented by spectral density (PSD). 3 is a conceptual diagram schematically illustrating a plurality of measurement regions in which the intracranial EEG or scalp EEG electrodes of the test subject P are disposed, and FIG. 4 is a reference signal and a seizure extracted from the plurality of measurement regions of the test subject B of FIG. 2. It is a figure which shows the shape of the power spectral density (PSD) of a signal. FIG. 5 is an exemplary diagram illustrating segregation of a seizure signal and a reference signal using a sparse probit classifier (SPC) from a feature of power spectral density extracted from one or more detection regions of an examinee.

Referring to FIG. 1, the method for early detection of real-time spasm EEG according to an embodiment of the present invention first receives pre-measured EEG signals from each of a plurality of measurement regions for a subject (S100). The measured EEG signals may be EEG signals measured when an actual seizure occurs. The EEG signals may be measured by inserting an EEG electrode into the cranial cavity of a portion estimated to be the subject's focus. However, the present invention is not limited to the measurement of EEG through intracranial electrodes, and in another embodiment, non-invasive scalp EEG by placing a sensor of the EEG electrode on the part of the subject suspected or more extensively outside the cranial cavity, especially the scalp. You can also measure the signal. In addition, the sensor for non-invasive measurement of the brain signal can measure not only brain waves but also various brain signals using a non-invasive method through the brain-imaging technique. For example, non-invasive brain imaging techniques include electroencephalography or electroencephalography (ElectroEncephaloGraphy, EEG), magnetoencephaloGraphy (MEG), near-infrared spectroscopy (NIRS), or functional magnetic resonance imaging. resonance imaging, fMRI), and the like.

Hereinafter, for convenience of explanation, the case where the EEG electrode in the cranial cavity is inserted to measure the EEG signal will be described. For example, in the case of measuring by inserting an EEG electrode into the cranial cavity, various sampling rates such as 200Hz, 256Hz, 516Hz, 1600Hz, etc. Can be measured. For example, when the EEG signal measured at a sampling rate of 1600 Hz is continuously analyzed in units of windows having a width of 1 sec and an overlapping size of 125 msec. The data can be verified by a cross validation method that trains the classification model into six parts, and classifies the classification model into six parts.

Next, a power spectrum density (PSD) of the EEG signals is calculated and analyzed to detect a specific frequency band including a frequency having the largest separation between the reference signal and the seizure signal (S200). For example, the specific frequency band may be a frequency band having a certain range based on the frequency having the largest separation width. The specific frequency band can be detected using the Welch method. However, the present invention is not limited thereto and may use various methods for detecting a specific frequency band.

Referring to FIGS. 1 and 2, it can be seen that there is a difference in a specific frequency band including a frequency where the baseline and seizure have the largest separation width for each subject. 15 Hz for subject A, 40 Hz for subject B, 20 Hz for subject C, 25 Hz for subject D, 200 Hz for subject E, 200 Hz for subject F, and 8 Hz for subject F It can be seen that the seizure signal is larger than the other frequencies. In this manner, by the Welch method, a specific frequency band capable of effectively separating the seizure signal from the reference signal can be detected.

Thereafter, a feature of the power spectral density of the EEG signals corresponding to the detected specific frequency band is extracted to classify one or more detection areas including valid shapes among the plurality of measurement areas (S300).

1, 3, and 4, EEG signals corresponding to the plurality of measurement regions may be measured using an EEG electrode arranged in a matrix (M) form in the cranial cavity of the examinee P. Can be. In FIG. 3, the EEG electrodes E are arranged to be somewhat exaggerated for convenience of explanation, but the electrodes inserted into the cranial cavity of the examinee P may be smaller than this. In addition, although an 8 × 16 matrix form is illustrated as an example in the drawings, the present invention is not limited thereto, and electrodes may be arranged in various shapes or numbers that may represent a plurality of measurement regions.

Electroencephalogram electrode (E) may be surgically inserted into the intracranial electrode in order to measure in detail the part suspected to be the lesion of the subject (P), the method of attaching the electrode to the scalp without using an invasive method such as surgery It can also be arranged to measure the EEG signal. At this time, the shape of the power spectral density (PSD) of the EEG signals corresponding to the detected specific frequency band among the EEG signals obtained from the plurality of measurement areas is extracted as shown in FIG. The dotted line represents the feature of the reference signal, and the solid line represents the feature of the seizure signal. As such, when features that are easily separated in some measurement areas are concentrated, analysis of some measurement areas may be more efficient and accurate than analysis of seizure signals using the entire measurement area. Therefore, the present invention can classify one or more detection areas including shapes effective for the seizure signal among the plurality of measurement areas and use them for analysis.

Meanwhile, in the classifying of one or more detection regions, the EEG signals may be separated into a reference signal and a seizure signal using a sparse probit classifier (SPC). Sparse probit classifiers (SPCs) are designed in mathematics and are also used in the field of EEG. The present invention can effectively separate seizure and reference signals from one or more detection regions classified using them. Can be.

Next, at least one detection region of the examinee and real-time EEG signals corresponding to the specific frequency band are measured (S500), and whether the examinee has seizures is determined based on the real-time EEG signals (S700). At this time, among the pre-measured EEG signals, one or more detection areas and EEG signals corresponding to a specific frequency band of the examinee may be separated into a reference signal and a seizure signal, and a reference range for dividing the reference signal and the seizure signal may be selected ( S400).

1 and 5, it can be seen that one or more detection regions classified from a plurality of measurement regions and EEG signals corresponding to a specific frequency band are effectively separated into seizure signals and reference signals based on a significant dividing line SL. .

Here, the dividing line SL may be a reference range for dividing the examinee's reference signal and the seizure signal. According to an examinee, one or more detection areas classified from a plurality of measurement areas may be different, and specific frequency bands in which seizure signals and reference signals are best separated may be different. Therefore, the reference range derived according to one or more detection regions and specific frequency regions may vary from subject to subject. The present invention enables patient-specific seizure detection by analyzing real-time EEG signals using this reference range. The determining of the seizure of the examinee may be determined to be a seizure when there is a first signal out of a reference range among the measured real-time EEG signals. In other words, by measuring the real-time EEG signals in real time from the examinee, and extracting the shape of the power spectral density (PSD) from the real-time EEG signals, it is determined to be a seizure when there is a first signal outside the reference range by comparing with the reference range can do. Since the seizure can be determined from the EEG signal measured in real time using the reference range, the present invention can detect the seizure at the initial stage of the seizure.

Table 1 shows the accuracy of predicting the moment when seizure occurs from the EEG signals of a plurality of subjects.

	One	2	3	4	5	6	7
Patient 1	98.73	100	100	100	100	100	98.42
Patient 2 (data 1)	100	100	100	100	100	100	98.94
Patient 2 (data 2)	100	98.72	100	100	100	100	100
Patient 3	100	96.97	100	100	100	100	100
Patient 4	100	100	100	100	100	100	100
Patient 5	100	100	100	100	100	100	100
Patient 6	100	100	100	87.50	100	100	100

Referring to Table 1, it can be seen that the seizure signal and the reference signal are almost accurately separated from the EEG signals of six epileptic patients with an average accuracy of 99.58 ± 0.00185%.

As described above, the real-time spasm EEG early detection method according to an embodiment of the present invention detects a specific frequency band having the largest separation between the reference signal and the seizure signal, the power spectrum density (PSD) of the plurality of measurement areas By classifying one or more detection areas having an effective shape and analyzing the real-time EEG signals of the examinee, it is possible to quickly and accurately detect the seizure of the examinee.

Embodiments according to the present invention described above may be implemented in the form of a computer program that can be executed through various components on a computer, such a computer program may be recorded on a computer readable medium. At this time, the media may be magnetic media such as hard disks, floppy disks and magnetic tape, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and ROMs. Hardware devices specifically configured to store and execute program instructions, such as memory, RAM, flash memory, and the like.

On the other hand, the computer program is specifically designed and configured for the present invention, but may be known and available to those skilled in the computer software field. Examples of computer programs may include not only machine code generated by a compiler, but also high-level language code executable by a computer using an interpreter or the like.

Particular implementations described in the present invention are examples and do not limit the scope of the present invention in any way. For brevity of description, descriptions of conventional electronic configurations, control systems, software, and other functional aspects of the systems may be omitted. In addition, the connection or connection members of the lines between the components shown in the drawings by way of example shows a functional connection and / or physical or circuit connections, in the actual device replaceable or additional various functional connections, physical It may be represented as a connection, or circuit connections. In addition, unless specifically mentioned, such as "essential", "important" may not be a necessary component for the application of the present invention.

As described above, the present invention has been described with reference to one embodiment shown in the drawings, which is merely exemplary, and it will be understood by those skilled in the art that various modifications and embodiments may be made therefrom. Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

According to an embodiment of the present invention, there is provided a real-time spasm EEG early detection method. In addition, embodiments of the present invention can be applied to the detection of cramped brain waves used in industry.

Claims (7)

1. Receiving pre-measured EEG signals from each of the plurality of measurement regions for the subject;

   Calculating a power spectrum density (PSD) for the brain wave signals;

   Analyzing the power spectral density (PSD) of the EEG signals to detect a specific frequency band including a frequency having a largest separation between a reference signal and a seizure signal;

   Extracting a feature of a power spectral density (PSD) of the EEG signals corresponding to the detected specific frequency band to classify one or more detection areas including valid shapes among the plurality of measurement areas; And

   Measuring real time EEG signals corresponding to the at least one detection region and the specific frequency band of the examinee, and determining whether the examinee has a seizure based on the real time EEG signals; .
2. According to claim 1,

   After the classifying step, the reference range for separating the EEG signals corresponding to the at least one detection area and the specific frequency band of the examinee into the reference signal and the seizure signal, and distinguishes the reference signal from the seizure signal Further comprising the step of selecting, real-time spasm EEG early detection method.
3. The method of claim 2,

   The determining of the seizure of the examinee is determined to be a seizure when the first signal out of the reference range of the measured real-time EEG signals, real-time spasm EEG early detection method.
4. The method of claim 2,

   The classifying the at least one detection region may include separating the EEG signals into the reference signal and the seizure signal using a sparse probit classifier.
5. According to claim 1,

   The detecting of the specific frequency band may include detecting the specific frequency band by using a Welch method.
6. According to claim 1,

   The EEG signals corresponding to the plurality of measurement areas are measured using EEG electrodes arranged in a matrix form in the intracranial or scalp of the examinee, real-time spasm EEG early detection method.
7. A computer program stored in a recording medium for executing the method of any one of claims 1 to 6 using a computer.

Images