KR102538720B1 - Apparatus and method for diagnosis of psychiatric disorders using mean amplitude-based features of event-related potential signal to improve diagnostic performances of computer-aided diagnosis system - Google Patents

Abstract

Disclosed are a mental disorder classification device and method based on an average amplitude of an event evoked potential signal for improving performance of a computer-assisted diagnosis system capable of helping diagnose mental disorder patients. The mental disorder classification method based on an average amplitude of an event evoked potential signal according to an embodiment of the present invention comprises the steps of: collecting brainwave data by measuring a brainwave evoked in a user while performing an auditory oddball method; extracting an event evoked potential signal within a section set based on the time of occurrence of a target stimulus; calculating an average amplitude for each of brainwave signals divided into units of a time window from the event evoked potential signal; and classifying mental disorders based on average amplitude characteristics calculated and obtained in units of the time window.

Description

Apparatus and method for diagnosis of psychiatric disorders using mean amplitude-based classification of psychiatric disorders based on the average amplitude of event evoked potential signals to improve the performance of computer-aided diagnostic systems that can help diagnose patients with psychiatric disorders features of event-related potential signal to improve diagnostic performances of computer-aided diagnosis system}

The present invention relates to an apparatus and method for classifying mental disorders based on the average amplitude of an event evoked potential signal to improve the performance of a computer-aided diagnostic system capable of helping diagnose patients with mental disorders.

Post-traumatic stress disorder (PTSD) and major depressive disorder (MDD) are very common diseases with lifetime prevalences of 7.8% and 7.7%, respectively, but both are diagnosed early and treated appropriately. It is a serious mental illness that can lead to suicide if not done. PTSD patients are a type of mental disorder that arises from psychological trauma that occurs as a result of directly experiencing or witnessing a traumatic traumatic event. PTSD patients repeatedly experience repetitive and invasive memories related to the event, and as a result, anxiety and arousal are higher than normal people, and they complain of difficulty in leading daily life due to extreme stress. MDD patients feel persistent depression, complain of symptoms such as decreased interest and motivation, changes in appetite and weight, and sleep disorders, and in severe cases, lead to suicide. In addition to the above unique clinical symptoms, both PTSD and MDD share similar symptoms such as emotional numbness, discomfort, and irritability, making it difficult to classify the two diseases.

Comorbid symptoms such as emotional numbness, dysphoria, and irritability are factors that impede accurate diagnosis of PTSD and MDD. Conventionally, the diagnosis of a patient with a mental illness has been made by the clinician's subjective judgment through a medical examination, and the similar clinical symptoms described above increase the misdiagnosis rate of the diagnosis of the two diseases. In addition, if the patient hides or deceives his or her symptoms, or if the patient does not recognize the symptoms and does not sufficiently share the symptoms with the clinician, the rate of misdiagnosis of the two diseases also increases.

Recently, in order to reduce diagnosis errors caused by ambiguous subjective diagnostic criteria, a feature based on the peak amplitude of the event evoked potential that can objectively classify PTSD patients and MDD patients is extracted and used to classify mental disorders. A computer-assisted diagnostic system based on a machine learning method has been developed, but its performance is still insufficient for practical use. Korean Intellectual Property Office Publication No. 10-2021-0048285 (published on May 3, 2021) discloses a method for providing information on mental disorders and a device for providing information on mental disorders using the same.

An object of the present invention is to provide an apparatus and method for classifying mental disorders that can help in accurately diagnosing mental disorders with a high co-morbidity rate by utilizing features based on average amplitude instead of features based on peak amplitude of event evoked potential.

In particular, the present invention can improve the performance of a computer-assisted diagnostic system that can help in accurate diagnosis by distinguishing patients with post-traumatic stress disorder (PTSD) and patients with major depressive disorder (MDD), which have a high co-morbidity rate, and Mental disorder classification that can improve diagnostic performance by classifying comorbid mental disorders more accurately than peak amplitude-based diagnostic systems using event-related potential (ERP)-based mean amplitude features It is to provide an apparatus and method.

A mental disorder classification method based on the average amplitude of an event evoked potential signal according to an embodiment of the present invention comprises the steps of collecting EEG data by measuring EEGs evoked in a user while performing an auditory bipolar stimulation method by an EEG data collection unit; extracting, by an event evoked potential signal extractor, an event evoked potential signal within a set interval based on the generation time of the target stimulus; Calculating, by an average amplitude calculation unit, average amplitudes of EEG signals divided in units of time windows from the event evoked potential signal; and classifying, by a mental disorder classification unit, a mental disorder based on the average amplitude feature calculated and obtained in units of the time window.

The calculating of the average amplitude may include dividing the event evoked potential signal into the EEG signals in units of the time window by an event evoked potential signal divider; adjusting a baseline of the EEG signals based on an average value of a section prior to occurrence of the target stimulus by a baseline adjusting unit; and calculating, by the average amplitude calculation unit, an average value for each baseline-adjusted EEG signal, and calculating average amplitude in units of the time window.

The calculating of the average amplitude may include calculating, by an average amplitude calculation unit, average amplitudes of EEG signals divided from the event evoked potential signal in units of time windows; selecting features using Fisher scores for each time window, by a Fisher score analysis unit; and determining an optimal time window length through a machine learning-based classifier and cross-validation using the selected features by a time window determiner.

Classifying the mental disorder may include classifying the mental disorder based on average amplitude characteristics of EEG signals calculated in units of a time window corresponding to the time window length determined by the time window determiner.

Classifying the mental disorder may include classifying mental disorders including post-traumatic stress disorder and depressive disorder with high co-morbidities.

According to an embodiment of the present invention, a computer program recorded on a computer-readable recording medium is provided to execute a mental disorder classification method based on the average amplitude of the event evoked potential signal.

An apparatus for classifying mental disorders based on the average amplitude of an event evoked potential signal according to an embodiment of the present invention includes an EEG data collection unit configured to collect EEG data by measuring EEGs evoked in a user while performing an auditory bipolar stimulation method; an event evoked potential signal extraction unit configured to extract an event evoked potential signal within a section set based on the generation time of the target stimulus; an average amplitude calculating unit configured to calculate average amplitudes of EEG signals divided in units of time windows from the event evoked potential signal; and a mental disorder classification unit configured to classify a mental disorder based on the average amplitude feature calculated and obtained in units of the time window.

An apparatus for classifying mental disorders based on the average amplitude of an event evoked potential signal according to an embodiment of the present invention includes an event evoked potential signal divider configured to divide the event evoked potential signal into the EEG signals in units of time windows; and a baseline adjuster configured to adjust a baseline of the EEG signals based on an average value of a section prior to occurrence of the target stimulus. The average amplitude calculation unit may calculate an average amplitude for each EEG signal whose baseline is adjusted by the baseline adjustment unit to calculate an average amplitude in units of the time window.

The average amplitude calculation unit may calculate average amplitudes of EEG signals divided in time window units from the event evoked potential signal. An apparatus for classifying mental disorders based on the average amplitude of an event evoked potential signal according to an embodiment of the present invention includes a Fisher score analyzer configured to select features using Fisher scores for various time windows; and a time window determination unit configured to determine an optimal time window length through a machine learning-based classifier and cross-validation using the features selected by the Fisher score analysis unit.

The mental disorder classification unit may classify mental disorders based on average amplitude characteristics of EEG signals calculated in units of time windows corresponding to the time window length determined by the time window determiner.

According to an embodiment of the present invention, an apparatus and method for classifying mental disorders that can help accurately diagnose mental disorders with a high co-morbidity rate by utilizing features based on average amplitude instead of features based on peak amplitude of event evoked potentials are provided. .

In addition, according to an embodiment of the present invention, the performance of a computer-assisted diagnostic system that can help in accurate diagnosis by distinguishing patients with post-traumatic stress disorder (PTSD) and major depressive disorder (MDD), which have a high co-morbidity rate, is improved. It can use the event-related potential (ERP)-based mean amplitude feature to more accurately classify comorbid mental disorders than the peak amplitude-based diagnosis system, thereby improving the diagnostic performance of mental disorders. can increase

In addition, according to an embodiment of the present invention, unlike the diagnostic system based on the peak amplitude of the event evoked potential, the average-based feature is used using the entire waveform of the event evoked potential signal, and information processing other than cognitive function (e.g., emotion and Since it is classified using the degree of activation of the brain related to exercise), it can be helpful in diagnosing patients with various mental diseases as well as mental diseases with cognitive decline.

In addition, according to an embodiment of the present invention, by using objective biosignal-based characteristics to help diagnose patients with two mental illnesses with high co-morbidities, it is possible to reduce the rate of misdiagnosis in the traditional questionnaire method, and furthermore, other co-morbid mental illnesses. It can also be applied to the patient's EEG data to help diagnose overall mental disorders.

1 is a flowchart of a mental disorder classification method based on the average amplitude of an event evoked potential signal according to an embodiment of the present invention.
2 is a block diagram of an apparatus for classifying mental disorders based on the average amplitude of an event evoked potential signal according to an embodiment of the present invention.
3 is a flowchart of a mental disorder classification method based on the average amplitude of an event evoked potential signal according to another embodiment of the present invention.
4 is a conceptual diagram for explaining a mental disorder classification method based on the average amplitude of an event evoked potential signal according to an embodiment of the present invention.
5 is an exemplary view showing event evoked potential signals of PTSD patients and MDD patients.

Advantages and features of the present invention, and methods of achieving them, will become clear with reference to the detailed description of the following embodiments taken in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only these embodiments make the disclosure of the present invention complete, and common knowledge in the art to which the present invention belongs. It is provided to completely inform the person who has the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numbers designate like elements throughout the specification.

In this specification, when a certain component is said to "include", it means that it may further include other components, not excluding other components unless otherwise stated. '~ unit' used in this specification is a unit that processes at least one function or operation, and may mean, for example, software, an FPGA, or a hardware component. Functions provided by '~unit' may be performed separately by a plurality of components or may be integrated with other additional components. '~unit' in this specification is not necessarily limited to software or hardware, and may be configured to be in an addressable storage medium or configured to reproduce one or more processors. Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

An apparatus and method for classifying mental disorders based on the average amplitude of event evoked potential signals according to an embodiment of the present invention calculates average amplitudes of EEG signals divided in units of time windows from event evoked potential signals extracted from EEGs evoked by a user. Calculate and classify the mental disorder based on the average amplitude feature calculated and obtained in units of time windows. According to an embodiment of the present invention, it is possible to help in accurate diagnosis by distinguishing patients with post-traumatic stress disorder (PTSD) and patients with major depressive disorder (MDD), which have a high co-morbidity rate, and a peak amplitude-based diagnosis system. The performance of diagnosis can be improved by more accurately classifying comorbid mental disorders.

1 is a flowchart of a mental disorder classification method based on the average amplitude of an event evoked potential signal according to an embodiment of the present invention. Referring to FIG. 1, while an auditory oddball task is being performed on the user by the auditory bipolar performance unit in order to classify mental disorders, the EEG data collection unit measures the EEG caused by the user. EEG data of can be collected (S10). When the EEG data is collected, the EEG data pre-processing unit applies a band-pass filter to the EEG data and performs pre-processing to remove motion noise, thereby removing noise caused by the movement of the user's eyes and the external environment (S20). .

When noise is removed from the brain wave data, an event evoked potential signal within a set interval based on the generation time of the target stimulus generated by the auditory bipolar performance unit can be extracted using the event evoked potential signal extractor (S30). When the event evoked potential signal is extracted by the event evoked potential signal extractor, the event evoked potential signal may be divided into EEG signals in units of time windows by the event evoked potential signal divider (S40).

When the event evoked potential signal is divided into EEG signals by the event evoked potential signal division unit, the baseline adjustment unit adjusts the baseline of the EEG signals based on the average value of the interval before the occurrence of the target stimulus (baseline correction) Processing can be performed (S50). Thereafter, the average amplitude calculation unit may calculate average amplitudes of the EEG signals divided in time window units from the event evoked potential signal (S60). In this case, the average amplitude calculator may calculate an average amplitude for each time window by calculating an average value for each baseline-adjusted EEG signal.

When the average amplitude is calculated for each time window, the mental disorder classifier may classify the mental disorder based on the obtained average amplitude feature calculated for each time window (S70). In this case, the mental disorder classification unit may classify the mental disorder based on the average amplitude characteristics of EEG signals calculated in units of time windows corresponding to the time window length (window size) determined by the time window determination unit.

2 is a block diagram of an apparatus for classifying mental disorders based on the average amplitude of an event evoked potential signal according to an embodiment of the present invention. Referring to FIGS. 1 and 2 , the apparatus 100 for classifying mental disorders based on the average amplitude of an event evoked potential signal according to an embodiment of the present invention includes an auditory bipolar performance unit 110, an EEG data collection unit 120, and an EEG data collection unit 120. Data preprocessing unit 130, event evoked potential signal extraction unit 140, event evoked potential signal division unit 150, time window adjustment unit 160, Fisher score analysis unit 170, time window determination unit 180 , a baseline adjustment unit 190, an average amplitude calculation unit 200, and a mental disorder classification unit 210.

The auditory quantum stimulation performing unit 110 may perform auditory quantum stimulation schemes to the user. The brain wave data collection unit 120 may collect brain wave data by measuring brain waves generated by the user while the auditory bipolar stimulation method is performed by the auditory bipolar stimulation unit 110 . In an embodiment, the EEG data collecting unit 120 may include a wearable EEG collecting device worn by a user in the form of a headset on the ear, around the ear, or on the head to collect EEG. The EEG data collection unit 120 may include one or more EEG collection electrodes for collecting EEG.

The EEG data pre-processing unit 130 applies a band pass filter (eg, a 1 to 30 Hz band pass filter) to the EEG data collected by the EEG data collection unit 120 and performs pre-processing to remove motion noise. can do. Noise cancellation can be performed by removing physiological artifacts (eg, signals with a magnitude greater than 75 μV) from the EEG signals collected at each EEG collection electrode.

The event evoked potential signal extractor 140 may extract the event evoked potential signal within a section set based on the generation time of the target stimulus generated by the auditory bipolar performance unit 110 . The event evoked potential signal divider 150 may divide the event evoked potential signal extracted by the event evoked potential signal extractor 140 into EEG signals in units of time windows having a predetermined time window length.

The time window adjusting unit 160 controls the event evoked potential signal so that the event evoked potential signal is divided according to various time windows (eg, time windows having various window sizes such as 50 ms, 100 ms, 200 ms, and 300 ms). The division unit 150 is controlled. The Fisher score analysis unit 170 may select features using Fisher scores for each of various time windows. The time window determiner 180 may determine an optimal time window length through a machine learning-based classifier and cross-validation using the features selected by the Fisher score analyzer 170 .

The baseline adjustment unit 190 may adjust the baseline of the EEG signals based on an average value of a section prior to the occurrence of the target stimulus. The average amplitude calculation unit 200 may calculate average amplitudes of EEG signals divided in units of time windows from the event evoked potential signal. The average amplitude calculation unit 200 may calculate an average amplitude for each time window by calculating an average value for each baseline-adjusted EEG signal.

The mental disorder classification unit 210 may classify the mental disorder based on the average amplitude feature calculated and obtained in units of time windows. The mental disorder classification unit 210 may classify the mental disorder based on the average amplitude characteristics of EEG signals calculated in units of time windows corresponding to the time window length determined by the time window determiner 180 .

3 is a flowchart of a mental disorder classification method based on the average amplitude of an event evoked potential signal according to another embodiment of the present invention. 4 is a conceptual diagram for explaining a mental disorder classification method based on the average amplitude of an event evoked potential signal according to an embodiment of the present invention. 5 is an exemplary view showing event evoked potential signals of PTSD patients and MDD patients.

1 to 5 , the time window controller 160 controls the event evoked potential signal divider 150 to divide the event evoked potential signal P300 according to various time windows. An event evoked potential signal such as P300 corresponds to a stimulus presentation time associated with cognitive processing such as attention (eg, among stimuli corresponding to reference numerals 11 to 15 in FIG. A positive going wave may be generated at about 300 ms after the stimulus occurs).

The event evoked potential signals measured for MDD patients and PTSD patients (eg, P300 shown as a solid line in FIG. 4 ) are both measured for normal subjects (eg, shown as a dotted line in FIG. 4 ). Compared to P300), the size of P300 is lowered and the time of occurrence is delayed.

The Fisher score analysis unit 170 may select features using Fisher scores for each of various time windows (S110). Since the Fisher score algorithm is well known, a detailed description of the Fisher score algorithm is omitted so as not to obscure the subject matter of the present invention. The time window determiner 180 may determine an optimal time window length ΔT through cross-validation with a machine learning-based classifier using the features selected by the Fisher score analyzer 170 (S120).

The mental disorder classification unit 210 calculates the average amplitude characteristics of EEG signals calculated in units of time windows (TW) corresponding to the time window length (ΔT) determined by the time window determination unit (a plurality of signals divided in units of time windows). The mental disorder may be classified based on the average amplitude values for each section) (S130 and S140).

According to the embodiment of the present invention as described above, the performance of a computer-assisted diagnostic system that can help in accurate diagnosis by distinguishing patients with post-traumatic stress disorder (PTSD) and major depressive disorder (MDD) with a high co-morbidity rate is improved. Specifically, by using the event-related potential (ERP)-based mean amplitude feature, more accurately than the peak amplitude-based diagnosis system, comorbid mental disorders are classified and diagnosed. performance can be improved.

According to the embodiment of the present invention, unlike the diagnostic system based on the peak amplitude of the event evoked potential, the average-based feature is used using the entire waveform of the event evoked potential signal, and information processing other than cognitive function (e.g., emotion and movement) Since it is classified using the degree of activation of the brain related to , it can be helpful in diagnosing patients with various mental diseases as well as mental diseases showing cognitive decline.

In addition, according to an embodiment of the present invention, when diagnosing two mentally ill patients with high co-morbidities using objective bio-signal-based characteristics, the misdiagnosis rate of the traditional questionnaire method can be reduced, and furthermore, other co-morbid mental diseases It can also be applied to the patient's EEG data to help diagnose overall mental disorders.

Hereinafter, an experiment for verifying the performance of the apparatus and method for classifying mental disorders based on the average amplitude of an event evoked potential signal according to an embodiment of the present invention as described above will be described. For the present invention, EEG data measured from 51 PTSD patients, 67 MDD patients, and 39 normal controls were measured.

In general, it is known that when two different stimuli are presented, attention is high for a stimulus with a relatively low presentation rate, and at this time, brain activation (P300) related to attention appears. The method is an oddball task.

In the case of the auditory oddball task used in the present invention, it consists of two different types of sound stimuli (target - 1,500 Hz; standard - 1,000 Hz), and the presentation ratio of each sound stimulus is 15% and 85%. . Stimuli were presented a total of 400 times. EEGs evoked while performing the auditory stimulation method were measured using a 62-electrode EEG measurement system based on the international 10-20 electrode system of the International Federation (sampling rate - 1,000 Hz).

After measuring the EEG, eye movement, noise due to movement, and large noises from unknown external environments were removed, and then the EEG signal was divided from 100 ms before the presentation of the target stimulus to 600 ms after the presentation of the target stimulus based on the time of presentation of the target stimulus. . If the divided signal contained an amplitude exceeding ± 75 μV, the signal was excluded from further analysis.

Each divided signal was subjected to baseline correction by subtracting the average value of the time point before the target stimulus (-100 ms - 0 ms), and then the event evoked potential signal was calculated by averaging all divided signals. Five different time windows were used to extract features based on average amplitude [50 ms, 100 ms, 200 ms, and 300 ms]. The features were extracted by averaging the EEG signals of each channel using each time criterion. In the present invention, three different classification combinations were constructed as follows.

1) PTSD patients vs. normal control group

2) MDD patients vs. normal control group

3) PTSD patients vs. MDD patient

In order to classify the two groups of each combination, important features from 1 to 20 were selected using Fisher's score for each time criterion, and the machine learning-based support vector machine (support vector machine) was used using the selected features. machine, SVM) classifier and leave-one-out cross validation method were used. Since the number of subjects in each group recruited for the present invention is disproportionate, the classification performance was derived using balanced accuracy ([specificity + sensitivity]*0.5) to reduce errors in accuracy due to unbalanced subjects. . The measured classification accuracy is shown in Table 1.

1) Highest classification accuracy when classifying PTSD patients and normal controls - 84.84% when using 50 ms time criterion-based features

2) Highest classification accuracy when classifying MDD patients and normal controls - 77.69% when using 200 ms time criterion-based features

3) Highest classification accuracy in classifying PTSD patients and MDD patients - 79.97% when using 100 ms time base-based features

In particular, the accuracy of classifying MDD patients and normal controls using average amplitude-based features increased by about 17% over the previous peak amplitude-based classification result (balanced accuracy - 60.16%), and PTSD patients and MDD patients were classified. (balanced accuracy - 70.83%), it was confirmed that the classification accuracy improved by about 10% or more.

Through the above results, when diagnosing patients with mental illness, the average amplitude-based features containing relatively more information than the amplitude-based features were used by using the premise event evoked potential waveform rather than the peak amplitude-based features extracted based on fragmentary information. It can be inferred that this is more suitable. As described above, an embodiment of the present invention introduces a new feature based on average amplitude to improve the performance of an event evoked potential EEG-based computer-aided diagnostic system that can help diagnose two diseases with high co-morbidity. It was verified that it can.

Embodiments of the present invention can induce not only P300 event evoked potentials related to cognitive function, but also event evoked potentials related to emotions (early posterior negativity; EPN, late positive potential; LPP) and movement-related cortical potentials (MRCP). A variety of possible stimulation paradigms can be used. In addition, the embodiments of the present invention are useful in diagnosing not only PTSD patients and MDD patients, but also mental disorders (eg, schizophrenia, bipolar disorders) that share similar clinical symptoms, making accurate diagnosis difficult. In addition, it can be used to classify mental disorders such as generalized anxiety disorder or panic disorder with similar clinical characteristics.

At least some of the configurations of the embodiments described above may be implemented as hardware components, software components, and/or a combination of hardware components and software components. For example, the devices, methods and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate (FPGA) array), programmable logic units (PLUs), microprocessors, or any other device capable of executing and responding to instructions.

A processing device may run an operating system and one or more software applications running on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of software. For convenience of understanding, there are cases in which one processing device is used, but those skilled in the art know that a processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It will be understood that it can include

For example, a processing device may include a plurality of processors or a processor and a controller. Also, other processing configurations are possible, such as a parallel processor. Software may include a computer program, code, instructions, or a combination of one or more of the foregoing, which configures a processing device to operate as desired or processes independently or collectively. You can command the device.

Software and/or data may be any tangible machine, component, physical device, virtual equipment, computer storage medium or device, intended to be interpreted by or provide instructions or data to a processing device. can be embodied in Software may be distributed on networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer readable media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer readable medium. Computer readable media may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the embodiment or may be known and usable to those skilled in computer software.

Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CDROMs and DVDs, and ROMs, RAMs, and flash memories. hardware devices specially configured to store and execute program instructions, such as; Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, as well as machine language codes such as those produced by a compiler. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

As described above, although the embodiments have been described with limited examples and drawings, those skilled in the art can make various modifications and variations from the above description. For example, the described techniques may be performed in an order different from the method described, and/or components of the described system, structure, device, circuit, etc. may be combined or combined in a different form than the method described, or other components may be used. Or even if it is replaced or substituted by equivalents, appropriate results can be achieved. Therefore, other implementations, other embodiments, and equivalents of the claims are within the scope of the following claims.

100: Mental disorder classification device based on the average amplitude of the event evoked potential signal
110: Auditory bipolar performance unit
120: EEG data collection unit
130: EEG data pre-processing unit
140: event evoked potential signal extraction unit
150: event evoked potential signal divider
160: time window control unit
170: Fisher score analysis unit
180: time window determining unit
190: baseline adjustment unit
200: average amplitude calculation unit
210: mental illness classification unit

Claims (10)

A computer program recorded on a computer-readable recording medium to execute a mental disorder classification method based on the average amplitude of an event evoked potential signal,
The average amplitude-based mental disorder classification method of the event evoked potential signal is:
collecting brain wave data by measuring brain waves caused by a user while performing an auditory bipolar stimulation method by an brain wave data collecting unit;
extracting, by an event evoked potential signal extraction unit, an event evoked potential signal within a set interval based on a target stimulus generation time point;
calculating, by an average amplitude calculation unit, average amplitudes of EEG signals divided in units of time windows from the event evoked potential signal; and
Classifying, by a mental disorder classification unit, a mental disorder based on an average amplitude feature calculated and obtained in units of the time window;
The step of calculating the average amplitude is:
selecting features using Fisher scores for each time window, by a Fisher score analysis unit;
determining an optimal time window length through a machine learning-based classifier and cross-validation using the selected features by a time window determiner;
dividing the event evoked potential signal into the EEG signals in units of the time window corresponding to the length of the time window, by an event evoked potential signal divider;
adjusting a baseline of the EEG signals based on an average value of a section prior to occurrence of the target stimulus by a baseline adjustment unit; and
Based on the average amplitude of the event evoked potential signal, comprising calculating, by the average amplitude calculation unit, the average amplitude of each baseline-adjusted EEG signal for the EEG signals divided in units of time windows from the event evoked potential signal. A computer program recorded on a computer-readable recording medium to execute a mental disorder classification method. delete delete The method of claim 1,
Classifying the mental disorder comprises classifying the mental disorder based on average amplitude characteristics of the EEG signals calculated in units of a time window corresponding to the time window length determined by the time window determiner. A computer program recorded on a computer-readable recording medium to execute a mental disorder classification method based on the average amplitude of an evoked potential signal. The method of claim 1,
Classifying the mental disorder is read by a computer to execute a mental disorder classification method based on the average amplitude of the event evoked potential signal, including the step of classifying mental disorders including post-traumatic stress disorder and depressive disorder with high co-morbidities. A computer program recorded on a recordable medium. delete an EEG data collection unit configured to collect EEG data by measuring an EEG caused by a user while performing an auditory quantum stimulation method;
an event evoked potential signal extraction unit configured to extract an event evoked potential signal within a section set based on the generation time of the target stimulus;
a Fisher score analysis unit configured to select features using Fisher scores for various time windows; and
a time window determination unit configured to determine an optimal time window length through a machine learning-based classifier and cross-validation using the features selected by the Fisher score analysis unit;
an event evoked potential signal divider configured to divide the event evoked potential signal into EEG signals in units of time windows corresponding to the time window length;
a baseline adjustment unit configured to adjust baselines of the EEG signals based on an average value of a section prior to occurrence of the target stimulus;
an average amplitude calculating unit configured to calculate average amplitudes of the EEG signals divided in units of time windows from the event evoked potential signal; and
A mental disorder classification unit configured to classify a mental disorder based on the average amplitude feature calculated and obtained in units of the time window,
The average amplitude calculation unit calculates an average value of each EEG signal baseline-adjusted by the baseline adjusting unit from the event evoked potential signal to calculate the average amplitude in units of the time window, and classifies mental disorders based on average amplitude of the event evoked potential signal. Device. delete delete The method of claim 7,
The mental disorder classification unit classifies the mental disorder based on the average amplitude characteristics of the EEG signals calculated in units of a time window corresponding to the time window length determined by the time window determination unit, based on the average amplitude of the event evoked potential signal. Mental illness classification device.

Images