KR20230078347A - Prediction method for autism trait of individual using brain waves and analysis apparatus - Google Patents

Abstract

A method for predicting an individual's autism spectrum propensity using an EEG signal includes receiving an EEG signal measured by an analysis device while a subject is performing a specific task, extracting features by frequency band from the EEG signal by the analysis device, and and predicting an autism spectrum propensity for the subject by inputting the characteristics into a predictive model built in advance by an analysis device. The predictive model is a model in which a correlation between EEG characteristics and autism spectrum tendencies is previously determined using EEG signals obtained while performing the specific task for a group having autism spectrum tendencies.

Description

Method and analysis device for predicting individual autism spectrum propensity using brain wave signals

The technology to be described below relates to a method of predicting the autism spectrum propensity of an ordinary person using an EEG signal.

Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by persistent deficits in social communication and interaction and repetitive behaviors. The hypothesis that autistic tendencies exist on a continuum is getting more and more attention and support, which means that normal people also show autistic traits to some extent. One of the most widely known clinical tools for examining the autism spectrum hypothesis is the Autism-Spectrum Quotient (AQ).

Korean Patent Publication No. 10-2020-0031496

It is difficult to use or interpret the autism spectrum tendencies of ordinary people only with conventional AQ tests. A technique for classifying autism spectrum disorder using brain waves has been studied, but the prior art has been a technique for classifying autism patients and normal people based on brain waves in a stable state.

The technology described below focuses on the fact that autism spectrum disorder basically accompanies impairment of social cognitive function, and uses the EEG of a subject performing a task of examining social cognitive function. To provide a technique for predicting autism spectrum tendencies .

A method for predicting an individual's autism spectrum propensity using an EEG signal includes receiving an EEG signal measured by an analysis device while a subject is performing a specific task, extracting features by frequency band from the EEG signal by the analysis device, and and predicting an autism spectrum propensity for the subject by inputting the characteristics into a predictive model built in advance by an analysis device.

An analysis device that predicts an individual's autism spectrum propensity using an EEG signal is an input device that receives an EEG signal measured while a subject is performing a specific task, and an input device obtained while performing the specific task for a group having an autism spectrum tendency. A storage device for storing a predictive model in which a correlation between EEG characteristics and autism spectrum propensity is determined using EEG signals, extracting features for each frequency band from the EEG signals, and inputting the features to the predictive model to provide information to the subject. It includes an arithmetic device for predicting the autism spectrum propensity for

The technology described below can predict the autism spectrum propensity of the general public by comprehensively using tasks that evaluate different social cognitive abilities. The technology described below can provide a model that can objectively evaluate the degree of autism spectrum propensity for the general public.

1 is an example of an EEG-based autism spectrum propensity prediction system.
2 is an example of a process of constructing an EEG-based autism spectrum propensity prediction model.
3 is an example of a process of performing a tuning task.
4 is an example of a process of performing a facial expression perception task.
5 is an example of a process of performing a common attention task.
6 is an example of an analysis device for predicting an EEG-based autism spectrum propensity.

Since the technology to be described below can have various changes and various embodiments, specific embodiments will be illustrated in the drawings and described in detail. However, this is not intended to limit the technology described below to specific embodiments, and it should be understood to include all modifications, equivalents, or substitutes included in the spirit and scope of the technology described below.

Terms such as first, second, A, B, etc. may be used to describe various elements, but the elements are not limited by the above terms, and are merely used to distinguish one element from another. used only as For example, without departing from the scope of the technology described below, a first element may be referred to as a second element, and similarly, the second element may also be referred to as a first element. The term and/or includes a combination of a plurality of related recited items or any one of a plurality of related recited items.

In terms used in this specification, singular expressions should be understood to include plural expressions unless clearly interpreted differently in context, and terms such as “comprising” refer to the described features, numbers, steps, operations, and components. , parts or combinations thereof, but it should be understood that it does not exclude the possibility of the presence or addition of one or more other features or numbers, step-action components, parts or combinations thereof.

Prior to a detailed description of the drawings, it is to be clarified that the classification of components in the present specification is merely a classification for each main function in charge of each component. That is, two or more components to be described below may be combined into one component, or one component may be divided into two or more for each more subdivided function. In addition, each component to be described below may additionally perform some or all of the functions of other components in addition to its main function, and some of the main functions of each component may be performed by other components. Of course, it may be dedicated and performed by .

In addition, in performing a method or method of operation, each process constituting the method may occur in a different order from the specified order unless a specific order is clearly described in context. That is, each process may occur in the same order as specified, may be performed substantially simultaneously, or may be performed in the reverse order.

Hereinafter, it will be described that the analysis device analyzes the subject's brain wave signal (EEG signal) using a certain predictive model. The analysis device may predict the subject's autism spectrum propensity based on the current subject's brain wave signal. The analysis device may be implemented with various devices capable of processing data. For example, the analysis device may be implemented as a PC, a server on a network, a smart device, a chipset in which a dedicated program is embedded, and the like.

1 is an example of an EEG-based autism spectrum propensity prediction system 100. 1 illustrates an example in which the analysis device is a computer terminal 130 and a server 140 .

The brain wave measuring device 110 measures an brain wave (EEG) signal of a subject. A subject performs a certain task, and the EEG measurement device 110 measures EEG while the subject performs the task. The subject's EEG signal may be stored in the EMR (Electronic Medical Record, 120).

Meanwhile, a separate computer device 105 may present a specific task and receive feedback from the subject. Alternatively, the brain wave measuring device 110 may be an integrated device that presents a certain task and receives an input from a user.

In FIG. 1 , the user A may predict the autism spectrum propensity using the EEG signal of the subject using the computer terminal 130 . The computer terminal 130 receives the EEG signal of the subject. The computer terminal 130 may receive an EEG signal from the EEG measuring device 110 or the EMR 120 through a wired or wireless network. In some cases, the computer terminal 130 may be a device physically connected to the brain wave measuring device 110 .

The computer terminal 130 may pre-process the subject's EEG signal constantly. The computer terminal 130 may extract EEG signal features from the EEG signal and predict the autism spectrum propensity for the subject based on the EEG signal features using a previously prepared prediction model. A detailed analysis process will be described later. User A can check the analysis result.

The server 140 may receive the subject's EEG signal from the EEG measuring device 110 or the EMR 120 . The server 140 may process the subject's EEG signal constantly. The server 140 may extract EEG signal features from the EEG signal and predict the autism spectrum propensity for the subject based on the EEG signal features using a previously prepared prediction model. A detailed analysis process will be described later. The server 140 may transmit the analysis result to the terminal of user A. User A can check the analysis result.

The computer terminal 130 or the server 140 does not use information input by the subject while performing the task, but uses EEG data generated while the subject performs the task as an analysis target. The computer terminal 130 and/or the server 140 may store the analysis result in the EMR 120.

2 is an example of a process 200 of constructing an EEG-based autism spectrum propensity prediction model. The process of building a predictive model may be performed through a separate computer device rather than an analysis device. Hereinafter, it is assumed that a computer device builds a predictive model.

First, the brain wave measurement device records the brain waves of the subject performing the above-described task (210). The task is at least one of an alignment task, a facial expression perception task, and a common attention task. Details of each task will be described later.

The computer device receives brain wave signals (EEG signals) of the subject measured while performing the task. In this case, the brain wave data may be data measured while performing at least one of a tuning task, a facial expression perception task, and a common attention task. That is, data measured while performing a specific task or a plurality of tasks may be used as EEG data used for model construction.

The computer device may pre-process the EEG signal to a constant (220). EEG signal preprocessing may include at least one of various preprocessing processes used in EEG signal processing. For example, the computer device may filter signals of a specific frequency band (1 to 100 Hz) from the previously acquired EEG signal. The computer device may remove noise included in the EEG signal. For example, the computer device may remove signals such as a safety level signal, an EMG signal, and an ECG signal included in the signals by using independent component analysis (ICA).

Thereafter, the computer device may extract certain features from the EEG signal (230). First, the computer device may extract a signal fragment (epoch) of a certain length to be a feature extraction unit. The length of the fragment may be determined in consideration of the experimental environment or the size of the data. Thereafter, the computer device may convert the EEG signal into a frequency band (Fast Fourier Transform) and distinguish the signal for each frequency band. For example, the computer device may extract signal features for each delta, theta, alpha, beta, and gamma band.

Meanwhile, the model construction process generally uses EEG signals obtained from a plurality of subjects. Accordingly, the computer device may uniformly normalize EEG signals between individual subjects.

The computer device may perform regression analysis by selecting the extracted features and using information representing the autism spectrum propensity of the subject (240). The information representing the autism spectrum tendency may be a result of a test previously performed on the subject. In the case of the autism spectrum, the AQ (Autism-Spectrum Quotient) test can be used to evaluate the autism spectrum propensity. The researcher also used the AQ test results as reference information. The computer device may derive a linear model to which a stepwise variable selection method is applied to the frequency characteristics of the subject's brain waves.

Through this process, the computer device may build a predictive model (multiple linear regression model) for predicting the autism spectrum propensity using the EEG signal of the subject performing the specific task.

Furthermore, other techniques may be utilized for the predictive model. For example, a prediction model may be built through a machine learning model such as a support vector machine (SVM) or a deep learning model.

In the process of constructing the above-described autism spectrum propensity prediction model, subjects perform certain tasks. The EEG measuring device measures EEG of a subject performing a corresponding task. The task proceeds in the form of receiving feedback on the subject's response or evaluation of certain content displayed on the screen. The task may be provided through an EEG measuring device or a separate computer device. Hereinafter, it is assumed that a separate computer device provides a certain task.

Explain the task used by the researcher. Of course, even in the process of predicting the autism spectrum propensity for a specific subject, the EEG measurement device measures the EEG of the subject performing the same task.

3 to 5, three problems will be described. 3 to 5, black or white rectangles represent screens displayed on the screen. A plurality of screens corresponds to a progress process in the order of time from top to bottom.

3 is an example of a process of performing a tuning task. The alignment task consists of a two-step task. 3(A) is an example for the first evaluation, and FIG. 3(B) is an example for the second evaluation. The alignment task is a task to see the alignment effect in which individual opinions are adjusted in the direction that agrees with the majority of evaluations in the second evaluation compared to the first evaluation. The stronger the autistic tendency, the less the conformity effect.

Referring to Figure 3 (A), the computer device presents a fixation on the screen for 0.5 seconds in the first evaluation, and presents a specific picture in the center of the screen for 1 second (1s). The fixation point refers to a point for fixing the gaze of the subject. As for the picture, any picture among a plurality of picture candidates may be displayed. Thereafter, the computer device receives an evaluation score for the picture presented by the subject. Here, the evaluation means an evaluation of whether the picture was drawn well or not, and can be evaluated using a certain score system. The researcher was asked to enter a score between 1 and 10 using the keyboard. After that, the computer device shows the score given by the subject for 2 seconds on the screen, then shows the evaluation of the group (many) for 3 seconds, and finally shows the evaluation of the subject and the majority at once for 3 seconds. The group evaluation score may be an average or a median value of scores evaluated by a plurality of people.

Referring to FIG. 3(B), in the second evaluation, the subject inputs a score once again for the picture he or she has scored. The computer device first presents a fixation on the screen for 0.5 seconds, presents the same picture used for the first evaluation in the center of the screen for 1 second, and finally receives the subject's evaluation.

4 is an example of a process of performing a facial expression perception task. The facial expression recognition task is a task to respond to a point where an emotion is felt while watching a facial expression whose emotional intensity gradually changes, and to name the corresponding emotion. The stronger the autistic tendency, the lower the sensitivity to facial expressions.

The progress of the task is as follows. First, the computer device may output information describing the performance of the corresponding task on the screen. The computer device presents a fixed point for 0.5 seconds and presents 26 faces that change from expressionless to extreme expressions every 0.3 seconds. In the experiment, the researcher repeatedly presented facial expressions for 6 emotions (fear, sadness, surprise, joy, anger, disgust) 5 times for each gender (male and female) (a total of 60 trials). In this process, the computer device received input from the subject about the facial expression that felt the emotion. The subject can select an emotion by pressing the space bar at the time of feeling the emotion and inputting which emotion was detected from among the six emotions with a number key.

5 is an example of a process of performing a common attention task. The common attention task is a task that evaluates the ability of the common attention effect to more quickly and accurately detect stimuli in positions consistent with gaze. The stronger the autistic tendency, the less sensitively responds to gaze information, reducing the common attention effect.

The progress of the task is as follows. The computer device presents a fixed point for 1 second and a face facing the front for 2 seconds. In the computer device, the gaze of the face displayed on the screen moves slowly for 0.34 seconds, and presents a target circular stimulus at intervals equal to SOA (Stimulus onset asynchrony). 5 shows target circular stimuli as colored dots. The computer device receives a selection as to whether the subject's gaze and the target stimulus coincide or do not match. The target circular stimulus may be given as a correct answer or an incorrect answer, respectively. In the experiment, the researcher set the SOA at 0.25 second, 0.5 second, and 0.75 second, repeated 10 times each, and presented the entire target circular stimulus 60 times while receiving feedback from the subject. When a selection is input from the subject, the computer device removes the target stimulus, presents the current gaze stimulus for 2 seconds, and then returns to the front gaze for 0.34 seconds. Finally, the computer device presents a frontal gaze for 1 second.

6 is an example of an analysis device 300 for predicting autism spectrum tendencies based on brain waves. The analysis device 300 corresponds to the above-described analysis devices (130 and 140 in FIG. 1). The analysis device 300 may be physically implemented in various forms. For example, the analysis device 300 may have a form of a computer device such as a PC, a network server, and a chipset dedicated to data processing.

First, an EEG measuring device measures EEG while a subject performs a certain task (at least one of an alignment task, a facial expression perception task, and a common attention task). The process of performing the task is as described in FIGS. 3 to 5 .

The analysis device 300 may include a storage device 310, a memory 320, an arithmetic device 330, an interface device 340, a communication device 350, and an output device 360.

The storage device 310 may store the aforementioned prediction model (regression analysis model or other machine learning model).

The storage device 310 may store the subject's EEG signal.

The storage device 310 may store a program or code for processing the subject's EEG data.

The storage device 310 may store analysis results.

The memory 320 may store data and information generated during the process of processing the EEG signal by the analysis device 300 and the process of evaluating the autism spectrum propensity using a predictive model.

The interface device 340 is a device that receives certain commands and data from the outside. The interface device 340 may receive an EEG signal of an analysis target from a physically connected input device or an external storage device. At this time, the EEG signal is measured while performing the above-described task (at least one of the congruence task, the facial expression perception task, and the common attention task). The interface device 340 may transmit the analysis result to an external object.

The communication device 350 refers to a component that receives and transmits certain information through a wired or wireless network. The communication device 350 may receive a dynamic EEG signal of an analysis target from an external object. Alternatively, the communication device 350 may transmit the analysis result to an external object such as a user terminal.

Since the interface device 340 and the communication device 350 are configured to send and receive certain data from a user or other physical object, they can also be collectively referred to as input/output devices. If the function of receiving an EEG signal is limited, the interface device 340 and the communication device 350 may be referred to as input devices.

The output device 360 is a device that outputs certain information. The output device 360 may output interfaces and analysis results necessary for data processing.

The arithmetic device 330 may receive the subject's EEG signal using commands or program codes stored in the storage device 310 to evaluate the subject's autism spectrum propensity.

The arithmetic device 330 may pre-process the EEG signal constantly using the above-described process and corresponding program, and extract features for each frequency band. Signal processing and feature extraction for each frequency band are the same as or similar to the process described in FIG. 2 .

The computing device 330 may predict the autism spectrum propensity of the target person by inputting the extracted features for each frequency band to a prediction model built in advance. The predictive model may output an autism spectrum propensity probability value for the subject. The computing device 330 may determine the autism spectrum tendency (normal or abnormal) of the subject by comparing a pre-set threshold with an output value of the predictive model.

The arithmetic device 330 may be a device such as a processor, an AP, or a chip in which a program is embedded that processes data and performs certain arithmetic operations.

In addition, the above-described EEG processing method, EEG-based predictive model construction method, and autism spectrum propensity prediction method using the predictive model may be implemented as a program (or application) including an executable algorithm that can be executed on a computer. The program may be stored and provided in a temporary or non-transitory computer readable medium.

A non-transitory readable medium is not a medium that stores data for a short moment, such as a register, cache, or memory, but a medium that stores data semi-permanently and can be read by a device. Specifically, the various applications or programs described above are CD, DVD, hard disk, Blu-ray disk, USB, memory card, ROM (read-only memory), PROM (programmable read only memory), EPROM (Erasable PROM, EPROM) Alternatively, it may be stored and provided in a non-transitory readable medium such as EEPROM (Electrically EPROM) or flash memory.

Temporary readable media include static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), and enhanced SDRAM (Enhanced SDRAM). SDRAM, ESDRAM), Synchronous DRAM (Synclink DRAM, SLDRAM) and Direct Rambus RAM (DRRAM).

This embodiment and the drawings accompanying this specification clearly represent only a part of the technical idea included in the foregoing technology, and those skilled in the art can easily understand it within the scope of the technical idea included in the specification and drawings of the above technology. It will be obvious that all variations and specific examples that can be inferred are included in the scope of the above-described technology.

Claims (10)

receiving, by an analysis device, an EEG signal measured while the subject is performing a specific task;
extracting, by the analysis device, features for each frequency band from the EEG signal; and
Including the step of predicting the autism spectrum propensity for the subject by inputting the feature into a predictive model built in advance by the analysis device,
The predictive model is an individual's autism spectrum using EEG signals, which is a model in which the correlation between EEG characteristics and autism spectrum tendencies is determined in advance using EEG signals obtained while performing the specific task for a group having autism spectrum tendencies. Propensity prediction method. According to claim 1,
The method of predicting an individual's autism spectrum propensity using an EEG signal in which the specific task is at least one of an alignment task, a facial expression perception task, and a common attention task. According to claim 2,
The tuning task includes a first evaluation process and a second evaluation process,
The first evaluation task includes a process of outputting a random picture on a screen and presenting a result of evaluating the picture by the subject and evaluation results of other people,
The secondary rating task is a method of predicting an individual's autism spectrum propensity using an EEG signal comprising a process of re-evaluating the picture by the subject. According to claim 2,
The facial expression perception task is a method of predicting an individual's autism spectrum propensity using an EEG signal comprising a process of selecting an emotion currently felt from the subject while outputting a facial expression representing other emotions from an expressionless face on a screen. According to claim 2,
The common attention task outputs a face facing the front on a screen, outputs a face that changes to a gaze in a certain direction for a certain period of time, and outputs target circular stimuli at different intervals by a certain SOA (Stimulus Onset Asynchrony). A method for predicting an individual's autism spectrum propensity using an EEG signal comprising receiving feedback from the subject on whether or not the gaze in the certain direction matches the target circular stimulus. an input device that receives an EEG signal measured while the subject is performing a specific task;
a storage device for storing a predictive model in which a correlation between EEG characteristics and autism spectrum tendencies is determined using EEG signals obtained while performing the specific task for a group having autism spectrum tendencies; and
Analysis of predicting an individual's autism spectrum propensity using an EEG signal including an arithmetic unit that extracts features for each frequency band from the brain wave signal and inputs the feature to the prediction model to predict the autism spectrum propensity for the subject. Device. According to claim 6,
The specific task is an analysis device for predicting an individual's autism spectrum propensity using an EEG signal that is at least one of a tuning task, a facial expression perception task, and a common attention task. According to claim 7,
The tuning task includes a first evaluation process and a second evaluation process,
The first evaluation task includes a process of outputting a random picture on a screen and presenting a result of evaluating the picture by the subject and evaluation results of other people,
The second evaluation task is an analysis device for predicting an individual's autism spectrum propensity using an EEG signal comprising a process of re-evaluating the picture by the subject. According to claim 7,
The facial expression perception task predicts the individual's autism spectrum propensity using an EEG signal including a process of receiving a selection of the emotion currently felt from the subject while outputting a facial expression representing other emotions from an expressionless face on the screen Analysis device. According to claim 7,
The common attention task outputs a face facing the front on a screen, outputs a face that changes to a gaze in a certain direction for a certain period of time, and outputs target circular stimuli at different intervals by a certain SOA (Stimulus Onset Asynchrony). An analysis device for predicting an individual's autism spectrum propensity using an EEG signal comprising receiving feedback from the subject on whether or not the gaze in the certain direction and the target circular stimulus match.

Images