KR20200097192A - Method and program for calculating brain injury index of status epilepticus - Google Patents

Abstract

Provided is a method for calculating a brain injury index of status epilepticus. The method comprises: a data acquisition step of acquiring EEG data and quantified EEG data of a first patient for a predetermined period of time by a computer, wherein the EEG data and the quantified EEG data are divided into predetermined periods to be mutually matachable; and a step of deriving a first status epilepticus EEG data section corresponding to status epilepticus on the EEG data based on waveform information on the EEG data by the computer.

Description

Calculation method and program of brain damage index of persistent state of epilepsy {METHOD AND PROGRAM FOR CALCULATING BRAIN INJURY INDEX OF STATUS EPILEPTICUS}

The present invention relates to a method and a program for calculating an index of brain damage in a persistent state of epilepsy.

A'status epilepticus' is a condition in which epilepsy attacks persist. This is a completely different disease from “epilepsy,” a chronic disease with recurrent seizures. In general, most epilepsy seizures stop spontaneously within a few minutes and recover after showing a refractory period for a certain period of time after the seizure, whereas persistent epilepsy refers to a state in which seizures persist or epilepsy attacks recur without recovery of consciousness. Long-lasting epilepsy attacks not only cause brain damage, but also have high morbidity and mortality rates approaching 20-30%. Persistent epilepsy occurs in patients with epilepsy, but it also occurs in many cases. In particular, it often occurs in various diseases such as stroke, encephalitis, and dementia that cause brain damage in an aging society. When the persistent state of epilepsy first appears in such a situation, it may progress to epilepsy, but in other cases, epilepsy and persistent state of epilepsy can be viewed as completely different diseases.

Conventionally, for general epilepsy, there are many techniques for determining whether an epileptic seizure occurs based on a convulsive waveform or a patient's body movement. However, since there is no motor seizure in the persistent state of epilepsy, it is difficult to know whether the epilepsy seizure continues even when the patient is directly visually checked. Therefore, an expert's visual analysis using EEG data plays a decisive role in diagnosis. In addition, at the current technical level, it is difficult to know how long the EEG findings suggesting an epileptic seizure must last for brain damage.

An object to be solved by the present invention is to provide a method and a program for determining whether or not a persistent state of brain damage has occurred using EEG data, quantified EEG data, near-infrared spectroscopy data, clinical data, and the like.

In addition, the problem to be solved by the present invention is to provide a brain damage index calculation method and program that derives the level of brain damage according to the persistent state of epilepsy through brain image data or the patient's consciousness state data, and derives even the brain damage index. I want to.

In addition, the present invention reflects the shape of the seizure wave in the persistent state of epilepsy, the number of occurrences, frequency, and time intervals of the seizure wave, and accurately calculates the level of brain damage of the patient according to the continuation state of epilepsy. The purpose of this study is to provide a method and program for calculating persistent brain damage indicators.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems that are not mentioned will be clearly understood by those skilled in the art from the following description.

In an embodiment of the present invention for solving the above-described problem, a method for calculating a brain damage index in a continuous state of epilepsy includes a computer obtaining EEG data and quantified EEG data of a first patient for a predetermined period of time, wherein the EEG The data and the quantified EEG data are divided into predetermined periods so as to be mutually matchable, the data acquisition step and the computer based on the waveform information on the EEG data, 1 It includes the step of deriving the EEG data section of the persistent state of epilepsy.

The method for calculating the brain damage index of the persistent state of epilepsy according to an embodiment of the present invention for solving the above-described problem is a section in which the computer does not derive whether it corresponds to the persistent state of epilepsy from the waveform information on the EEG data. For, based on the information on the quantified EEG data, the EEG data section matching the section corresponding to the continuation state of epilepsy is extracted, and the EEG data section is referred to as the second epilepsy sustained state EEG data Further comprising the step of deriving as.

In an embodiment of the present invention for solving the above-described problem, the method for calculating the brain damage index of the persistent state of epilepsy includes the computer using the first near-infrared spectroscopy data, the first epilepsy continuing state EEG data, and the second epilepsy. Matching with the persistent EEG data, deriving the characteristics of the first near-infrared spectroscopy data for the persistent state of epilepsy, and the second near-infrared spectroscopy data obtained from the second patient based on the characteristics of the first near-infrared spectroscopy data. As a basis, further comprising the step of determining whether the second patient is in a persistent state of epilepsy, wherein the data acquisition step comprises: the computer using the near-infrared spectroscopy method for the predetermined period of time of the first patient. Further comprising acquiring near-infrared spectroscopy data, wherein the first near-infrared spectroscopy data is divided into a predetermined section so as to mutually match the EEG data and the quantified EEG data.

According to an embodiment of the present invention for solving the above-described problem, the method for calculating the brain damage index of the persistent state of epilepsy includes, by the computer, the first epilepsy continuing state brainwave data and the second epilepsy, with respect to the brainwave data. Further comprising learning the persistent state EEG data as a feature value of the persistent state of epilepsy.

According to an embodiment of the present invention for solving the above-described problem, the method for calculating the brain damage index of the persistent state of epilepsy includes the computer based on the EEG data, the first epilepsy continuing state EEG data, and the second epilepsy. The method further includes learning the continuous EEG data and the derived first near-infrared spectroscopy data for the persistent state of epilepsy as a feature value.

In an embodiment of the present invention for solving the above-described problem, a method for calculating a brain damage index in a continuous state of epilepsy includes the step of obtaining, by the computer, at least one of brain image data or consciousness state data for the predetermined period of time, and As the computer learns by matching at least one of the brain image data or consciousness state data with the first epilepsy persistence state EEG data and the second epilepsy persistence state EEG data, the level of brain damage to the EEG data Predicting the brain damage level further comprises a step of predicting.

The method for calculating the brain damage index of the persistent state of epilepsy according to an embodiment of the present invention for solving the above-described problems is a clinical change after the occurrence of a specific EEG section of the persistent state of epilepsy derived by the learning step by the computer. Learning to generate and present the level of brain damage as an index, the step of presenting the level of brain damage as an index further comprises.

In an embodiment of the present invention for solving the above-described problem, the method for calculating a brain damage index in a continuing state of epilepsy according to an embodiment of the present invention comprises the steps of determining and presenting, by the computer, treatment and measures for the damaged brain according to the predicted brain damage level. It includes more.

The computer program for calculating the brain damage index of the persistent state of epilepsy according to an embodiment of the present invention for solving the above-described problems is combined with a computer, which is hardware, and stored in a medium to execute any one of the above-described methods. .

Other specific details of the present invention are included in the detailed description and drawings.

According to the present invention as described above, it is possible to confirm the occurrence of a persistent state of epilepsy that was difficult to confirm with only EEG data.

In addition, according to the present invention as described above, it is possible to guide the incidence level of brain damage according to the occurrence of the persistent state of epilepsy to the medical staff.

In addition, through the present invention, by guiding to a numerical index, the medical staff can intuit the current state of the patient and influence the decision-making of the medical staff.

In addition, through the present invention, the medical staff can quickly take action on the patient by calculating and guiding the cause disease of the persistent state of epilepsy.

In addition, through the present invention, it is possible to predict the neurological prognosis after recovery of the persistent state of epilepsy.

The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a flow chart of a method for deriving a duration of epilepsy using EEG data according to an embodiment of the present invention.
FIG. 2 is a flowchart of a method of deriving a duration of epilepsy condition using quantified EEG data for a section that is not derived only from EEG data according to another embodiment of the present invention.
3 is a flow chart of a method of determining whether or not the epilepsy persists using near-infrared spectroscopy data according to another embodiment of the present invention.
4 is a view showing the result of matching the quantified EEG data graph and the near-infrared spectroscopy data graph for the same patient.
5 is a flowchart of a method of predicting a brain damage level according to another embodiment of the present invention.
6 is a flow chart of a method of generating and presenting a brain damage level as an index according to another embodiment of the present invention.
7 is a flowchart of a method of determining and presenting treatment and measures for a damaged brain according to a brain damage level according to another embodiment of the present invention.

Advantages and features of the present invention, and a method of achieving them will become apparent with reference to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in a variety of different forms, only the present embodiments are intended to complete the disclosure of the present invention, It is provided to fully inform the technician of the scope of the present invention, and the present invention is only defined by the scope of the claims.

The terms used in the present specification are for describing exemplary embodiments and are not intended to limit the present invention. In this specification, the singular form also includes the plural form unless otherwise specified in the phrase. As used in the specification, “comprises” and/or “comprising” do not exclude the presence or addition of one or more other elements other than the mentioned elements. Throughout the specification, the same reference numerals refer to the same elements, and "and/or" includes each and all combinations of one or more of the mentioned elements. Although "first", "second", and the like are used to describe various elements, it goes without saying that these elements are not limited by these terms. These terms are only used to distinguish one component from another component. Therefore, it goes without saying that the first component mentioned below may be the second component within the technical idea of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used in the present specification may be used as meanings that can be commonly understood by those of ordinary skill in the art to which the present invention belongs. In addition, terms defined in a commonly used dictionary are not interpreted ideally or excessively unless explicitly defined specifically.

In this specification, a'computer' includes all various devices capable of visually presenting a result to a user by performing arithmetic processing. For example, computers are not only desktop PCs and notebooks, but also smart phones, tablet PCs, cellular phones, PCS phones, and synchronous/asynchronous systems. A mobile terminal of the International Mobile Telecommunication-2000 (IMT-2000), a Palm Personal Computer (PC), a personal digital assistant (PDA), and the like may also be applicable. Further, the computer may correspond to medical equipment that acquires or observes medical images. In addition, the computer may correspond to a server computer connected to various client computers. In addition, the computer may consist of more than one device.

In the present specification, the'continuous state of epilepsy' is determined based on the continuation or repetition of abnormal neuronal activity causing brain damage, regardless of whether it is externally expressed as an action.

In the present specification, the'brain wave data' refers to a current generated according to brain activity or data derived and amplified and recorded. 'EEG data' is acquired in the form of a waveform graph by electroencephalography (EEG), which records the electrical activity of the brain by attaching electrodes to the scalp.

In the present specification,'quantified EEG data' refers to data obtained by extracting EEG extracted through a scalp sensor as an analog EEG signal and converting it into a digital signal. 'Quantitative EEG data' is obtained in the form of a brain map by quantitative electroencephalography (qEEG).

In the present specification,'near-infrared spectroscopy (NIRS) data' refers to data obtained by measuring changes in blood flow by providing near-infrared rays to the scalp surface.

Hereinafter, a method and a program for calculating an index of brain damage in an epileptic persistence state according to embodiments of the present invention will be described with reference to the drawings.

A deep neural network (DNN) according to embodiments of the present invention refers to a system or network that constructs one or more layers in one or more computers to perform a determination based on a plurality of data. For example, the deep neural network may be implemented as a set of layers including a convolutional pooling layer, a locally-connected layer, and a fully-connected layer. The convolutional pooling layer or the local access layer can be configured to extract features in the image. The fully connected layer may determine a correlation between features of an image. In some embodiments, the overall structure of the deep neural network may be in a form in which a local access layer is connected to the convolutional pooling layer and a fully connected layer is connected to the local access layer. The deep neural network may include various criteria (ie, parameters), and may add new criteria (ie, parameters) through analysis of an input image.

The deep neural network according to the embodiments of the present invention is a structure called a convolutional neural network suitable for image analysis, and a feature extraction layer that self-learns features having the highest discriminative power from given image data. ) And a prediction layer that learns a prediction model to produce the highest prediction performance based on the extracted features may be configured in an integrated structure.

The feature extraction layer spatially integrates a feature map and a convolution layer that creates a feature map by applying a plurality of filters to each area of an image, thereby providing features that are invariant to changes in position or rotation. It may be formed in a structure in which a pooling layer that enables extraction is alternately repeated several times. Through this, features of various levels can be extracted from low-level features such as points, lines, and faces to complex and meaningful high-level features.

The convolutional layer obtains a feature map by taking a nonlinear activation function to the dot product of the filter and the local receptive field for each patch of the input image. In comparison, CNN is characterized by using a filter with sparse connectivity and shared weights. This connection structure reduces the number of parameters to be learned, and makes learning through the backpropagation algorithm more efficient, resulting in improved prediction performance.

The integrated layer (Pooling Layer or Sub-sampling Layer) creates a new feature map by using the area information of the feature map obtained from the previous convolutional layer. In general, the feature map newly created by the integration layer is reduced to a smaller size than the original feature map. Typical integration methods include maximum pooling, which selects the maximum value of the corresponding area in the feature map, and corresponding within the feature map. There are average pooling, which calculates the average value of an area. In general, the feature map of the integrated layer may be less affected by the location of an arbitrary structure or pattern existing in the input image than the feature map of the previous layer. That is, the integrated layer can extract features that are more robust to regional changes such as noise or distortion in the input image or previous feature map, and these features can play an important role in classification performance. Another role of the integrated layer is to reflect the features of a wider area as it goes up from a deeper structure to the upper learning layer.As the feature extraction layer is piled up, the lower layer reflects local features and rises to the upper layer. It is possible to create features that reflect the more abstract features of the entire image.

In this way, the features finally extracted through the repetition of the convolutional layer and the integration layer are classified by a classification model such as Multi-layer Perception (MLP) or Support Vector Machine (SVM). -connected layer) and can be used for classification model training and prediction.

However, the structure of the deep neural network according to the embodiments of the present invention is not limited thereto, and may be formed as a neural network having various structures.

1 is a flow chart of a method for deriving a duration of epilepsy using EEG data according to an embodiment of the present invention.

Referring to FIG. 1, the method of calculating a brain damage index in a continuing state of epilepsy according to an embodiment of the present invention includes a step of obtaining, by a computer, EEG data of a first patient (S111), and a first brain attack on the EEG data by a computer. It includes a step (S130) of deriving the increased sustained EEG data section.

In the step S111 of obtaining the EEG data of the first patient by the computer, the computer obtains EEG data of the first patient for a predetermined period of time, but the EEG data is divided into predetermined periods.

The predetermined period may be a minimum unit time length for determining the continuation state of epilepsy, and may be a predetermined period arbitrarily designated by the user.

However, the predetermined period should be divided into a predetermined period identical to the division period of the EEG data for the quantified EEG data or near-infrared spectroscopy data to be described later.

The step of deriving, by the computer, the first epilepsy continuation state EEG data section from the EEG data (S130), the first epilepsy continuation state corresponding to the continuation state of epilepsy on the EEG data by the computer based on waveform information on the EEG data. It is to derive the EEG data section.

When the waveform on the EEG data is visually checked, the section corresponding to the continuing state of epilepsy can be identified. The first epileptic persistence state corresponding to the epilepsy persistence state by receiving the section of the waveform corresponding to the epilepsy persistence state from outside (e.g., medical staff, etc.) or through the section of the waveform previously stored on the computer It is derived as an EEG data section.

Receiving the section of the waveform corresponding to the persistent state of epilepsy from the outside means that medical staff, etc., labeling the section including the waveform corresponding to the previously known state of epilepsy, and the computer labeled The section is extracted and classified. For example, the computer extracts an EEG data section including a waveform image that has already been identified by an image recognition method.

The EEG data section corresponding to the continuation state of epilepsy is derived from the medical staff using the waveform labeled with the section corresponding to the continuation state of epilepsy with respect to the EEG data, but the EEG data alone makes it difficult to determine the section corresponding to the continuation state of epilepsy. There is a section that does not work.

Therefore, for a section that is not derived only from the EEG data, it is possible to derive whether there is a section corresponding to the continuing state of epilepsy by using supplemental additional data.

FIG. 2 is a flowchart of a method of deriving a duration of epilepsy condition using quantified EEG data for a section that is not derived only from EEG data according to another embodiment of the present invention.

Referring to FIG. 2, the method of deriving the duration of the epilepsy state section includes the step of obtaining, by a computer, EEG data and quantified EEG data of the first patient (S112), and the computer derives the first EEG continuous state EEG data section. A step (S130) and a step (S150) of deriving the EEG data section of the second epilepsy sustained state by the computer.

In the step of obtaining the first patient's EEG data and quantified EEG data by the computer (S112), the computer obtains the first patient's EEG data and quantified EEG data for a predetermined time, but the EEG data and the quantified EEG data are mutually matched. It is divided into predetermined periods as possible.

The step of deriving the EEG data section of the second epilepsy sustained state by the computer (S150) is, as described above, for the section in which the computer does not derive whether or not it corresponds to the epilepsy persistence state only by the waveform information on the EEG data, additional data Based on the information on the quantified EEG data, the EEG data section matching the section identified as the section in the continuous state of epilepsy is extracted from the quantified EEG data, and the EEG data section is referred to as the second epilepsy section. It is derived as continuous EEG data.

Quantified EEG data can be analyzed by various analysis methods such as spectrum, nonlinear dynamic analysis, fractal analysis, neurometrics analysis, and the information on the quantified EEG data may include all information on the quantified EEG data obtained by various analysis methods. .

Based on the information on the quantified EEG data, by deriving the EEG data section corresponding to the continuation state of epilepsy that cannot be grasped with the EEG data alone, the waveform information of the EEG data can be supplemented to more accurately identify whether the EEG persistence state is applicable. have.

In addition, the EEG data and the quantified EEG data are complementary to each other, and for a section that is not recognized as information on the quantified EEG data, whether the EEG data corresponds to the continuation state of epilepsy based on the waveform information of the EEG data. Can be derived.

If the epilepsy persistence status is derived based on the quantified EEG data, the section corresponding to the epilepsy persistence status is derived based on the information of the quantified EEG data, and in the case of the part that is not derived from the information of the quantified EEG data In the EEG data, the quantified EEG data section matching the section identified as a continuation state of epilepsy from the EEG data may be extracted and derived using waveform information of the EEG data.

The derived data can be used as learning data for deriving the persistent state of epilepsy. Therefore, the computer uses the first epilepsy continuation state EEG data and the second epilepsy continuation state EEG data to the EEG persistent state. Learn with the feature values of

With repetitive learning, it is possible to accurately grasp whether each section corresponds to the continuing state of epilepsy with only the EEG data.

3 is a flow chart of a method of determining whether or not the epilepsy persists using near-infrared spectroscopy data according to another embodiment of the present invention.

Referring to FIG. 3, the method of determining whether or not the epilepsy persists using near-infrared spectroscopy data includes the step of obtaining, by a computer, EEG data, quantified EEG data, and first near-infrared spectroscopy data of a first patient (S113), a computer A step of deriving a first epilepsy sustained state EEG data section (S130), a computer deriving a second epilepsy sustained state EEG data section (S150), a computer first near-infrared spectroscopy data for the persistent state of epilepsy A step of deriving a characteristic of (S170) and a step (S190) of determining whether the computer continues to have epilepsy with respect to the second patient.

Step (S113) of the computer obtaining EEG data, quantified EEG data, and first near-infrared spectroscopy data of the first patient, by the computer using EEG data, quantified EEG data, and near-infrared spectroscopy for a predetermined period of time of the first patient. The first near-infrared spectroscopy data is obtained, but the EEG data, the quantified EEG data, and the first near-infrared spectroscopy data are divided into a certain section so as to be mutually matchable.

The step of deriving, by the computer, the features of the first near-infrared spectroscopy data for the persistent state of epilepsy (S170), the computer, with respect to the first near-infrared spectroscopy data, the first epilepsy continuing state EEG data and the second epilepsy continuing state EEG By matching the data, the characteristics of the first near-infrared data for the persistent state of epilepsy are derived.

That is, the characteristic of the first near-infrared spectroscopy data for the persistent state of epilepsy is obtained by matching the duration of the epilepsy state section derived using the EEG data and the quantified EEG data.

The step of determining whether the computer continues to have epilepsy with respect to the second patient (S190) is the second patient with respect to the second near-infrared spectroscopy data obtained from the second patient by the computer based on the characteristics of the first near-infrared spectroscopy data. It is to determine whether or not the epilepsy persists.

Based on the characteristics of the first near-infrared spectroscopy data for the derived epilepsy persistence state, it is possible to determine whether the epilepsy persists state by using only the second near-infrared spectroscopy data of a new patient.

The derived data can be used as learning data for deriving the persistent state of epilepsy. Therefore, the computer can use the EEG data, the first epilepsy continuation state EEG data, the second epilepsy continuation state EEG data, and the epilepsy continuation state. The characteristics of the first near-infrared spectroscopy data for are learned as the characteristic values of the persistent state of epilepsy.

With repetitive learning, it is possible to accurately grasp whether each section corresponds to the continuing state of epilepsy with only the EEG data.

4 is a view showing the result of matching the quantified EEG data graph and the near-infrared spectroscopy data graph for the same patient.

4A and 4B are graphs for different sections, respectively, and are diagrams to show that the same is applied to each section rather than one section. Therefore, it will be described based on the drawing (a) of FIG. 4.

The drawing disclosed above in FIG. 4A corresponds to a near-infrared spectroscopy data graph, and the drawing disclosed below in FIG. 4A corresponds to a quantified EEG data graph.

On the near-infrared spectroscopy data graph, data 10 indicating an oxyhemoglobin value and data 20 indicating a deoxyhemoglobin value are shown.

In each graph, the portion indicated in the form of a band corresponds to the portion indicated as a matching portion.

Values represented by solid lines on the quantified EEG data graph represent EEG values of the right and left hemispheres as quantified EEG data (qEEG) values expressed as PeakEnvelope.

Looking at the value in the band on the near-infrared spectroscopy data graph, it corresponds to the portion where the oxyhemoglobin value increased, and looking at the value in the band on the quantified EEG data graph, the quantified brain wave data (qEEG) value expressed as a PeakEnvelope is It can be seen that it corresponds to the rising part.

It was confirmed that hemodynamic changes were actually occurring in the brain in the period in which the quantified EEG data and near-infrared spectroscopy data corresponding to the bands on each graph rises.

This result is confirmed that the repetitive results are derived in the ascending section in FIG. 4B as well, and by checking the section in which the quantified EEG data value and near-infrared spectroscopy data value rises, it is determined whether or not it corresponds to the epilepsy persistence state. You can accurately grasp.

In addition, even without a quantified EEG data value, only the near-infrared spectroscopy data value, it is possible to accurately determine whether or not the epilepsy persists through the rising section.

5 is a flowchart of a method of predicting a brain damage level according to another embodiment of the present invention.

Referring to Figure 5, the method of predicting the level of brain damage, the computer obtaining EEG data and quantified EEG data of the first patient (S112), the computer deriving the first EEP continuous state EEG data section (S130), the computer deriving the second epilepsy continuous state EEG data section (S150), the computer obtaining at least one of the brain image data or the consciousness state data (S210), and the computer brain for the EEG data It includes a step of predicting the damage level (S230).

In step S210 of obtaining at least one of the brain image data or the consciousness state data by the computer (S210), the computer acquires at least one of the brain image data or the consciousness state data for a predetermined period of time.

Brain image data is image data obtained by Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET), and Magnetoencephalography (MEG). Can include all.

The state of consciousness data includes the result of using a tool to evaluate the state of consciousness (Glasgow coma scale, GCS).

In the step S230 of the computer predicting the level of brain damage for the EEG data, the computer performs at least one of brain image data or consciousness state data, the first epilepsy sustained state EEG data, and the second epilepsy sustained state EEG data. By matching and learning, the level of brain damage to the EEG data is predicted.

The method of predicting the level of brain damage can determine the brain condition based on how many features correspond to the continuous state of epilepsy per hour.

Using at least one of brain image data that can visually check the brain state and consciousness state data that numerically represents the state of consciousness, the EEG data is matched to the section corresponding to the continuing state of epilepsy and is more accurate for the brain state. I can judge.

In addition, the level of brain damage may be calculated by reflecting the shape of the seizure wave, the number of occurrences, frequency, and time intervals of the seizure wave according to the continuing state of epilepsy.

Furthermore, the level of brain damage may be presented as an index for the determined brain condition.

In one embodiment, the index of the degree of brain damage, which indicates the level of brain damage as an index, is based on the degree of hemodynamic change during the seizure period, and may be expressed by Equation 1 below.

<Equation 1>

NIRSt is the degree of hemodynamic change measured by near-infrared spectroscopy, and t corresponds to the duration of the change.

The NIRSt value is the sum of the results of seizures based on a predetermined time standard. The predetermined predetermined time may be based on a time criterion that can be most accurately determined, and preferably may be based on 10 seconds.

6 is a flowchart of a method of generating and presenting a brain damage level as an index according to another embodiment of the present invention.

Referring to Figure 6, the method of generating and presenting the level of brain damage as an indicator is the step of generating and presenting the level of brain damage as an index by learning the clinical change after the occurrence of a specific EEG section in the continuous state of epilepsy by a computer ( S250) is further included.

Based on the prediction for the level of brain damage described above in FIG. 5, the level of brain damage is generated as an index.

In other words, the computer constructs a learning model that can calculate and provide brain damage indicators for new patients' EEG data. In one embodiment, the brain damage indicator may be calculated as a value within a predetermined numerical range. For example, if the state of no brain damage is set to 0, and the state that the entire brain or the entire specific range of the brain is damaged is set to 100, the computer determines the brain damage state from 0 to 100 based on the EEG data of each patient. Build a learning model to be calculated and provided by value.

7 is a flowchart of a method of determining and presenting treatment and measures for a damaged brain according to a brain damage level according to another embodiment of the present invention.

Referring to FIG. 7, the method of determining and presenting treatment and measures for the damaged brain according to the level of brain damage further comprises a step S270 of determining and presenting the treatment and measures for the brain according to the level of brain damage by a computer. Include.

After predicting the level of brain damage, various treatments or measures for the brain according to the level of brain damage are presented. The treatment and measures provided by the computer may be treatments or measures according to the level of brain damage previously input from the medical staff, and the most appropriate treatment or measures are provided by learning the result values for treatment or measures according to each brain damage level. May include.

For example, a computer can determine the amount of anesthetic drug to be added based on the level of brain damage.

In addition, the method for calculating the brain damage index of the persistent state of epilepsy according to an embodiment of the present invention includes the step of generating, by a computer, a criterion for the disease causing the persistent state of epilepsy according to the EEG data, and the EEG data continuously acquired by the computer Based on the learning model, it further includes the step of determining and providing the cause of the persistent state of epilepsy.

The persistent state of epilepsy occurs not only in general epilepsy but also in various diseases such as stroke, encephalitis (especially autoimmune encephalitis), dementia, and hepatic coma. If the persistent state of epilepsy persists for a long time, serious brain damage will occur, so quick action is required for the cause of the occurrence. Therefore, it is possible to learn the EEG data of a patient who already knows the cause of the persistent state of epilepsy with a deep neural network, and predict the cause of the persistent state of epilepsy based on the EEG data of a new patient.

To this end, the computer further includes data on the cause of occurrence of the persistent state of epilepsy determined based on the cause of each patient's visit to the first epilepsy sustained state EEG data and the second epilepsy sustained state EEG data for the EEG data. You can learn as.

In one embodiment, the computer may construct a model for predicting the cause of the persistent state of epilepsy by using a learning dataset including data on the cause of occurrence of the persistent state of epilepsy independently of the calculation of the brain damage index. Since the cause of the persistence of epilepsy is not known for all patients, there may be a difference between the amount of data used to calculate the brain damage index and the amount of data available to the model for predicting the cause of the persistent state of epilepsy, so separate learning can be performed. have.

In addition, it is possible to construct a learning model for the prognosis by learning not only the cause of occurrence, but also the prognosis after the occurrence of the persistent state of epilepsy as learning data.

The prognosis after the occurrence of the persistent state of epilepsy can be classified and used according to a predetermined time unit.

The above-described method for calculating brain damage indicators in a continuous state of epilepsy according to an exemplary embodiment of the present invention may be implemented as a program (or application) to be executed by being combined with a computer, which is hardware, and stored in a medium.

The steps of the method or algorithm described in connection with the embodiment of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or a combination thereof. Software modules include Random Access Memory (RAM), Read Only Memory (ROM), Erasable Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), Flash Memory, hard disk, removable disk, CD-ROM, or It may reside on any type of computer-readable recording medium well known in the art to which the present invention pertains.

In the above, embodiments of the present invention have been described with reference to the accompanying drawings, but those skilled in the art to which the present invention pertains can be implemented in other specific forms without changing the technical spirit or essential features. You can understand. Therefore, the embodiments described above are illustrative in all respects, and should be understood as non-limiting.

Claims (9)

The computer acquires EEG data and quantified EEG data of the first patient for a predetermined time, wherein the EEG data and the quantified EEG data are divided into predetermined periods so as to be mutually matchable; And
Including the step of deriving, by the computer, a first epilepsy sustained state EEG data section corresponding to an epileptic persistence state from the EEG data based on waveform information on the EEG data,
How to calculate the brain damage index of the persistent state of epilepsy. The method of claim 1,
EEG data matched with the section corresponding to the continuation state of epilepsy, based on the information on the quantified EEG data, for a section in which the computer does not derive whether or not it corresponds to the continuation state of epilepsy from the waveform information on the EEG data Extracting a section, further comprising the step of deriving the EEG data section as the second epilepsy sustained state EEG data, which is an epilepsy sustained state section,
How to calculate the brain damage index of the persistent state of epilepsy. The method of claim 2,
The computer matching the first near-infrared spectroscopy data with the first epileptic persistence state EEG data and the second epilepsy sustained state EEG data to derive characteristics of the first near-infrared spectroscopy data for the sustained epilepsy state; And
Based on the second near-infrared spectroscopy data obtained from the second patient based on the characteristics of the first near-infrared spectroscopy data, further comprising the step of determining whether or not the epilepsy persists for the second patient,
The data acquisition step,
The computer further comprises acquiring first near-infrared spectroscopy data using near-infrared spectroscopy for the predetermined period of time of the first patient, wherein the first near-infrared spectroscopy data is correlated with the EEG data and the quantified EEG data. It is divided into a certain section so that it can be matched,
How to calculate the brain damage index of the persistent state of epilepsy. The method of claim 2,
The computer further comprising the step of learning the first epilepsy sustained state EEG data and the second epilepsy sustained state EEG data as characteristic values of the epileptic persistence state with respect to the EEG data,
How to calculate the brain damage index of the persistent state of epilepsy. The method of claim 3,
The computer learns features of the first epilepsy persistence state EEG data, the second epilepsy persistence state EEG data, and the derived first near-infrared spectroscopy data for the epilepsy persistence state as a feature value for the EEG data Further comprising the step of,
How to calculate the brain damage index of the persistent state of epilepsy. The method of claim 2,
Obtaining, by the computer, at least one of brain image data and consciousness state data for the predetermined period of time; And
As the computer learns by matching at least one of the brain image data or consciousness state data with the first epileptic persistence state EEG data and the second epilepsy sustained state EEG data, the level of brain damage to the EEG data Predicting, further comprising a brain injury level prediction step,
How to calculate the brain damage index of the persistent state of epilepsy. The method of claim 6,
The step of presenting the level of brain damage as an index by learning the clinical change after the occurrence of a specific EEG section of the persistent state of epilepsy derived by the learning step by the computer to generate and present the level of brain damage as an index. More included,
How to calculate the brain damage index of the persistent state of epilepsy. The method of claim 6,
The computer further comprises the step of determining and presenting treatment and measures for the damaged brain according to the predicted brain damage level,
How to calculate the brain damage index of the persistent state of epilepsy. A computer program for calculating brain damage indicators in a persistent state of epilepsy, which is combined with a computer that is hardware and stored in a medium to execute any one of claims 1 to 8.

Images