KR20240002544A - Method for determining suicide risk group using fNIRS signal - Google Patents

Abstract

The present invention relates to a method for determining a suicide risk group using an fNIRS signal, which includes obtaining fNIRS signal data measuring changes in hemoglobin concentration according to changes in blood flow in the subject's brain using a functional near-infrared spectroscopy device, and obtaining fNIRS signal data. It includes the steps of extracting feature data to be input into the suicide risk group discrimination model as a basis, and inputting the extracted feature data into the suicide risk group discrimination model to output the suicide risk group classification result for the measured person. If the fNIRS signal data is acquired during the rest period before the subject performs the cognitive task, feature data is extracted from the result of the brain default mode network analysis of the subject's brain obtained by analyzing the fNIRS signal data. If fNIRS signal data is acquired during a task performance period in which the subject performs a predetermined cognitive task, the fNIRS signal data is divided into predetermined time intervals to obtain one or more statistical values of mean, slope, median, maximum, variance, kurtosis, and skewness. Extract feature data. The suicide risk group discrimination model can be learned with a training data set in which the suicidal ideation test scale test results are labeled with feature data extracted from fNIRS signal data. According to the present invention, the risk of suicide for a subject can be determined and predicted in advance using the fNIRS signal that determines and predicts the risk of suicide in advance.

Description

Method for determining suicide risk group using fNIRS signal}

The present invention relates to a method for determining suicide risk groups using fNIRS signals.

Conventionally, methods such as MRI, CT, angiography, or electroencephalography were mainly used to classify patients by monitoring information about the biological condition of the human head. However, although this method has the advantage of high measurement accuracy, it has the problem of being expensive and providing only information about the moment of measurement, making continuous monitoring difficult.

As an alternative to this problem, functional near-infrared spectroscopy (fNIRS) has recently been attracting attention. fNIRS is a method of non-invasively measuring changes in blood flow in the brain using near-infrared rays. It is based on measurement of the degree of hemoglobin oxygenation in cerebral blood flow.

Oxidized hemoglobin concentration (Oxy-Hb) and reduced hemoglobin concentration (Deoxy-Hb) can be measured by irradiating near-infrared rays to a living body and detecting and processing the fNIRS signal that has penetrated the tissue. Activated areas of the brain consume oxygen transported by Oxy-Hb in the blood, and Oxy-Hb is converted to Deoxy-Hb. These two hemoglobins have optical properties in the visible and near-infrared regions, and their concentrations measured by functional near-infrared spectroscopy can be used as a measure of brain activity.

Meanwhile, the total number of patients with mood disorders at high risk of self-harm and suicide has recently exceeded 1 million and is continuously increasing, but there is almost no clinical research and evaluation using neurophysiological signals from the brain on patients with mood disorders. The situation is that it is not being done. In addition, it is necessary to determine and predict the risk of suicide in advance and present guidelines for psychopharmacological treatment and behavioral treatment according to the risk of suicide.

Korean Patent Publication No. 10-2016-0121348 (October 19, 2016)

Therefore, the technical problem to be solved by the present invention is to provide a method for determining suicide risk groups using fNIRS signals to determine and predict suicide risk in advance.

The method of determining a suicide risk group using an fNIRS signal using a computing device according to the present invention includes the steps of using a functional near-infrared spectroscopy device to obtain fNIRS signal data measuring the change in hemoglobin concentration according to the change in blood flow in the subject's brain, A step of extracting feature data to be input into a suicide risk group determination model based on fNIRS signal data, and inputting the extracted feature data into the suicide risk group discrimination model to output a suicide risk group classification result for the measured person. do.

The suicide risk group discrimination model can be learned with a training data set in which the Scale for Suicide Ideation (SSI) test results are labeled with feature data extracted from fNIRS signal data.

The fNIRS signal data may include at least one of the concentration of oxidized hemoglobin and the concentration of reduced hemoglobin in the cerebral blood flow of the subject.

When the fNIRS signal data is acquired during a rest period before the subject performs a cognitive task, the feature data can be extracted from the result of a brain default mode network analysis of the subject's brain obtained by analyzing the fNIRS signal data.

When the fNIRS signal data is acquired during a task performance period in which the subject performs a predetermined cognitive task, the fNIRS signal data is divided into predetermined time intervals to select one of the mean value, slope, median, maximum, variance, kurtosis, and skewness. The feature data can be extracted using the above statistical values.

The training data set may further include one or more of the task type, task difficulty, and task performance results of the cognitive task performed by the subject.

One or more of the task type, task difficulty, and task performance results of the cognitive task performed by the measured person are input into the suicide risk group discrimination model along with the characteristic data extracted for the measured person, resulting in a suicide risk group classification for the measured person. can be output.

According to the present invention, the risk of suicide for a subject can be determined and predicted in advance using the fNIRS signal that determines and predicts the risk of suicide in advance.

Figure 1 is a configuration diagram of a system for determining suicide risk groups using fNIRS signals according to an embodiment of the present invention.
FIG. 2 is a block diagram showing the configuration of the computing device of FIG. 1 in more detail.
Figure 3 is a flowchart of a method for determining a suicide risk group using fNIRS signals according to an embodiment of the present invention.

Then, with reference to the attached drawings, embodiments of the present invention will be described in detail so that those skilled in the art can easily implement the present invention.

The terminology used herein is for describing embodiments and is not intended to limit the invention. As used herein, singular forms also include plural forms, unless specifically stated otherwise in the context. As used in the specification, “comprises” and/or “comprising” does not exclude the presence or addition of one or more other elements in addition to the mentioned elements. Like reference numerals refer to like elements throughout the specification, and “and/or” includes each and every combination of one or more of the referenced elements. Although “first”, “second”, etc. are used to describe various components, these components are of course not limited by these terms. These terms are merely used to distinguish one component from another. Therefore, it goes without saying that the first component mentioned below may also be a second component within the technical spirit of the present invention.

In this specification, “computing device” includes all various devices that can perform computational processing and provide results to the user. For example, computing devices include not only desktop PCs, laptop computers, and server computers, but also smart phones, tablet PCs, cellular phones, and PCS phones (Personal Communication Service phones). ), synchronous/asynchronous IMT-2000 (International Mobile Telecommunication-2000) mobile terminal, Palm Personal Computer (Palm PC), personal digital assistant (PDA), etc. may also be applicable.

Figure 1 is a configuration diagram of a system for determining suicide risk groups using fNIRS signals according to an embodiment of the present invention.

Referring to FIG. 1, the suicide risk group determination system using fNIRS signals according to the present invention may include a functional near-infrared spectroscopy device 100, a computing device 200, and an audio-visual device 300.

The functional near-infrared spectroscopy device 100 can irradiate near-infrared rays (wavelength: 650 to 1000 nm) to the subject's head and detect near-infrared signals that have penetrated the subject's brain tissue. In addition, the functional near-infrared spectroscopy device 100 can process the detected near-infrared signal through functional near-infrared spectroscopy to non-invasively obtain fNIRS signal data measuring the change in hemoglobin concentration according to the change in blood flow in the brain.

The fNIRS signal data may include time series data of the concentration of cerebral blood oxidized hemoglobin and time series data of the concentration of reduced hemoglobin.

The functional near-infrared spectroscopy device 100 includes a plurality of light sources (light emitting/laser diodes) that irradiate near-infrared rays to the head of the subject and a plurality of optical sensors (light reception) that detect near-infrared rays reflected or scattered from the subject's head. fNIRS signal data can be acquired from multiple channels, including diodes). For example, to measure fNIRS signal data, 16 pairs of light emitting/laser diodes and light receiving diodes can be placed on the forehead corresponding to the prefrontal cortex of the subject to obtain fNIRS signal data in 16 channels. Of course, the number of channels may vary depending on the embodiment. The part of the subject's head where fNIRS signal data is measured may be at a location other than the forehead or may be at multiple locations.

Meanwhile, the functional near-infrared spectroscopy device 100 can be implemented to measure fNIRS signal data containing a direct current drift component of 0.01 Hz or less, a hemodynamic response signal of approximately 0.01 to 0.1 Hz, and physiological noise components near 0.1, 0.5, and 1 Hz. there is. The photoelectrode of the light emitting diode required to measure such fNIRS signal data can generate light in the near-infrared region modulated at a frequency of at least 2 Hz, and the generated near-infrared ray can pass through the prefrontal cortex and be detected by the light receiving diode. Of course, in addition to what is described here, it is also possible to implement the functional near-infrared spectroscopy device 100 to generate light in the near-infrared component region of other frequency bands and measure the resulting hemodynamic response signal.

The functional near-infrared spectroscopy device 100 is implemented to remove physiological noise, optical noise, and white noise included in the fNIRS signal data and then transmit it to the computing device 200, or the computing device 200 removes the physiological noise, optical noise, and white noise included in the fNIRS signal data. It can be implemented to remove noise components.

To remove noise components included in fNIRS signal data, physiological noise using trend removal technology, digital filters such as smoothing filter, IIR filter, FIR filter, application filter such as Kalman filter, Wiener filter, and short channel-long channel combination. Various types of filters that can remove signals in the noise frequency region, such as removal filters, can be used.

The audio-visual device 300 can be implemented as an audio-visual device for presenting a cognitive task to the subject, for example, a display device such as a TV, monitor, or projector. A cognitive task can be presented to the person being measured through the audio-visual device 300.

While the subject is performing a predetermined cognitive task, the functional near-infrared spectroscopy device 100 can be implemented to measure fNIRS signal data of the subject.

The cognitive task presented to the subject is Nbaek, which evaluates the functions of the prefrontal cortex, such as executive functions, working memory, cognitive inhibition, and inhibitory control. It may be a neuropsychological test such as an N-back test, a Stroop test, or a verbal fluency test, but is not limited to these.

fNIRS signal data can be obtained by distinguishing between a task performance period in which the subject performs a predetermined cognitive task and a rest period in which the subject does not perform the cognitive task.

For example, the fNIRS signal data of the subject can be measured during the rest period before presenting the cognitive task. In addition, the cognitive task is presented with a sound or notification indicating the task is presented, and the person being measured performs the cognitive task for a predetermined period of time, and then stops performing the task with a sound or notification indicating that the task is to be stopped. The fNIRS signal data of the subject can be measured during the task performance period. In order to distinguish between starting and stopping task performance, a separate synchronization signal can be generated in accordance with the task performance time and measured together.

The computing device 200 extracts feature data from the fNIRS signal data acquired for the person being measured during the cognitive task performance period in the functional near-infrared spectroscopy device 100, and inputs this into a suicide risk group discrimination model to result in suicide risk group classification for the person being measured. can be output.

The suicide risk group discrimination model is a training data set consisting of predetermined feature data labeled with the results of a suicidal ideation test scale, and can be trained to output the suicide risk group classification results for the subject being measured. This is explained in detail below.

FIG. 2 is a block diagram showing the configuration of the computing device of FIG. 1 in more detail.

Referring to FIG. 2 , the computing device 200 may include at least one memory 210 and at least one processor 230.

At least one memory 210 may store at least one instruction. Additionally, the memory 210 may store data used in various tasks related to executing a method of determining a suicide risk group using an fNIRS signal in the computing device 200.

At least one processor 230 may execute a process corresponding to an instruction stored in the memory 210. At least one processor 230 may execute instructions stored in the memory 210 to learn a suicide risk group discrimination model using the training data set, and output a suicide risk group classification result of the measured person using the learned suicide risk group discrimination model. can do.

Suicide risk group discrimination models include machine learning models such as support vector machines, linear discriminant analysis, k-nearest neighbor classifier, and artificial neural network (ANN), and convolutional neural networks (CNN). It can be implemented with the same deep learning model.

The computing device 200 may extract fNIRS feature data from the fNIRS signal data as feature data to be input into a suicide risk group determination model.

fNIRS feature data may include one or more statistical values of mean, slope, median, maximum, variance, kurtosis, and skewness of the concentration of oxidized hemoglobin and reduced hemoglobin.

The computing device 200 may extract fNIRS feature data by dividing the fNIRS signal data into predetermined time intervals. For example, 0~n seconds, n~2n seconds, 2n~3n seconds, … Divide the time section at intervals of n seconds (n is a natural number), and calculate the average value, slope, median, maximum value, variance, kurtosis, and skewness of the concentration of oxidized hemoglobin and reduced hemoglobin in the corresponding time section from the fNIRS signal data. It can be extracted. When fNIRS signal data is acquired from a plurality of channels, fNIRS feature data can be extracted for each channel.

What statistical values are extracted from fNIRS feature data may vary depending on the embodiment, the channel or number of channels to be extracted may also vary depending on the embodiment, and the time interval dividing the time section or the time section in which the fNIRS feature data is extracted may vary depending on the embodiment. The number, etc. may also vary. Additionally, fNIRS feature data can be continuously extracted from all time intervals, or it can also be implemented to be extracted only from a predetermined time interval. Additionally, it can be implemented to extract fNIRS feature data only in the time section corresponding to the task performance period. It is also possible to extract fNIRS feature data from the task execution phase in a specific channel or time section, and extract fNIRS feature data from the rest phase in another channel or time section. The type of fNIRS feature data extracted, the channel through which fNIRS feature data is extracted, the number of time sections, etc. can be determined by considering the classification performance and system load of the suicide risk group determination model.

The suicide risk group discrimination model can be most simply learned with a training data set consisting of fNIRS feature data and the suicidal ideation test scale results labeled with it.

The suicide risk group discrimination model is trained with a training data set in which the Scale for Suicide Ideation (SSI) test results are labeled with fNIRS characteristic data, and when the fNIRS characteristic data of the measured person is input, the suicide risk group corresponding to the measured person is determined. It can be trained to output classification results. Beck's suicidal ideation test scale results can be used, but are not limited to this, and it is also possible to construct a training data set by labeling the suicidal ideation test scale results obtained by other predetermined criteria.

When using the results of Beck's Suicidal Ideation Test Scale, the training data set can be labeled by grading scores above a predetermined score (e.g., 25 points or more) into a high suicide risk group, and those below a predetermined score into a low suicide risk group. Of course, it is possible to label the results of the suicidal ideation test scale in more detail into high-risk, medium-risk, and low-risk groups, and it is also possible to create a training data set by labeling the suicidal ideation test scale scores themselves.

According to another embodiment of the present invention, the computing device 200 may analyze the functional connectivity of the brain's default mode network by analyzing fNIRS signal data measured for the subject during the resting period. . Additionally, the computing device 200 may output a suicide risk group classification result by inputting feature data extracted from the brain's default mode network analysis result into a suicide risk group determination model.

Here, the default mode network analysis results include clustering coefficient, characteristic path length, local efficiency, global efficiency, and whole-brain correlation analysis. It can be included, and among these, characteristic data to be input into the suicide risk group discrimination model can be extracted.

The suicide risk group discrimination model is a training data set constructed by labeling the suicidal ideation test scale results to the feature data extracted from the default mode network analysis results and can be trained to output the suicide risk group classification results for the measured subject.

Meanwhile, according to another embodiment of the present invention, the computing device 200 may be implemented to receive input from the user such as the task type, task difficulty, and task performance results of the cognitive task performed by the person being measured. In this case, the suicide risk group discrimination model can be trained to output a suicide risk group classification result by receiving one or more of the task type, task difficulty, and task performance results of the cognitive task performed by the subject along with the feature data extracted from the fNIRS signal data. there is.

Figure 3 is a flowchart of a method for determining a suicide risk group using fNIRS signals according to an embodiment of the present invention.

Referring to FIG. 3, first, the functional near-infrared spectroscopy device 100 can measure fNIRS signal data of the subject during the rest period before the subject performs a predetermined cognitive task (S300).

Next, the computing device 200 may analyze the default mode network of the subject's brain using the fNIRS signal data acquired during the resting period, and extract predetermined feature data from the default mode network analysis result of the subject's brain (S310 ).

Thereafter, while the subject is performing a cognitive task, the functional near-infrared spectroscopy device 100 can measure the fNIRS signal data of the subject (S320).

Next, the computing device 200 may extract fNIRS feature data from the fNIRS signal data acquired during the cognitive task performance period (S330).

Thereafter, the computing device 200 inputs the brain's default mode network analysis feature data extracted in step S310 and/or the fNIRS feature data extracted in step S330 into a suicide risk group discrimination model to determine the suicide risk group for the subject. Classification results can be output (S340).

Depending on the embodiment, only the steps (S300, S310, S320, S350) for outputting a suicide risk group classification result using the default mode network analysis and the result may be performed. Of course, depending on the embodiment, it is also possible to omit steps S300, S310, and S320 and perform only steps S330, S340, and S350.

The embodiments described above may be implemented with 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, and a field programmable gate (FPGA). It may be implemented using one or more general-purpose computing devices or special-purpose computing devices, such as an array, a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. A processing device may execute an operating system (OS) and one or more software applications that run on the operating system. Additionally, a processing device may access, store, manipulate, process, and generate data in response to execution of software. For ease of understanding, a single processing device may be described as being used; however, those skilled in the art will understand that a processing device includes multiple processing elements and/or multiple types of processing elements. It can be seen that it may include. For example, a processing device may include a plurality of processors or one processor and one controller. Additionally, other processing configurations, such as parallel processors, are possible.

Software may include a computer program, code, instructions, or a combination of one or more of these, which configures a processing unit to operate as desired or to process independently or collectively. You can command the device. Software and/or data may be used on any type of machine, component, physical device, virtual equipment, computer storage medium or device to be interpreted by or to provide instructions or data to a processing device. It can be embodied permanently or temporarily. Software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording 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. The computer-readable medium may include program instructions, data files, data structures, etc., singly or in combination. Program instructions recorded on the medium may be specially designed and configured for the embodiment or may be known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -Includes optical media (magneto-optical media) and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, etc. Examples of program instructions include machine language code, such as that produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. 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.

Although the embodiments have been described with limited drawings as described above, those skilled in the art can apply various technical modifications and variations based on the above. For example, the described techniques are performed in a different order than the described method, and/or components of the described system, structure, device, circuit, etc. are combined or combined in a different form than the described method, or other components are used. Alternatively, appropriate results may be achieved even if substituted or substituted by an equivalent.

Claims (9)

In a method for determining suicide risk groups using fNIRS signals using a computing device,
Obtaining fNIRS signal data measuring changes in hemoglobin concentration according to changes in blood flow in the brain of the subject using a functional near-infrared spectroscopy (fNIRS) device,
Extracting feature data to be input into a suicide risk group discrimination model based on the fNIRS signal data, and
Inputting the extracted feature data into the suicide risk group determination model to output a suicide risk group classification result for the measured person.
How to include .
In paragraph 1,
The suicide risk group discrimination model is,
How the suicidal ideation test scale test results are learned with a training data set labeled with feature data extracted from fNIRS signal data.
In paragraph 2,
The fNIRS signal data is time series data including at least one of the concentration of oxidized hemoglobin and the concentration of reduced hemoglobin in the cerebral blood flow of the subject.
In paragraph 3,
When the fNIRS signal data is acquired during a rest period before the subject performs a cognitive task,
A method of extracting the feature data from a brain default mode network analysis result of the subject obtained by analyzing the fNIRS signal data.
In paragraph 3,
When the fNIRS signal data is acquired during a task performance period in which the subject performs a predetermined cognitive task,
A method of dividing the fNIRS signal data into predetermined time intervals and extracting the feature data with one or more statistical values of mean value, slope, median value, maximum value, variance, kurtosis, and skewness.
In paragraph 2,
The training data set further includes one or more of the task type, task difficulty, and task performance result of the cognitive task performed by the subject,
One or more of the task type, task difficulty, and task performance results of the cognitive task performed by the measured person are input into the suicide risk group discrimination model along with the characteristic data extracted for the measured person, resulting in a suicide risk group classification for the measured person. How to print .
A computer-readable recording medium recording a program for executing the method according to any one of claims 1 to 6.
As a computing device,
a memory storing at least one instruction; and
at least one processor; Including,
The at least one processor,
Acquiring fNIRS signal data measuring changes in hemoglobin concentration according to changes in blood flow in the brain of the subject using a functional near-infrared spectroscopy device,
Extracting feature data to be input into a suicide risk group discrimination model based on the fNIRS signal data, and
Inputting the extracted feature data into the suicide risk group determination model to output a suicide risk group classification result for the measured person.
Run ,
The suicide risk group discrimination model is,
A computing device in which the Scale for Suicide Ideation (SSI) test results are learned using a training data set labeled with feature data extracted from fNIRS signal data.
In paragraph 8:
The fNIRS signal data is time series data including at least one of the concentration of oxidized hemoglobin and the concentration of reduced hemoglobin in the cerebral blood flow of the subject,
When the fNIRS signal data is acquired during a rest period before the subject performs a cognitive task, the feature data is extracted from a brain default mode network analysis result of the subject obtained by analyzing the fNIRS signal data,
When the fNIRS signal data is acquired during a task performance period in which the subject performs a predetermined cognitive task, the fNIRS signal data is divided into predetermined time intervals to select one of the mean value, slope, median, maximum, variance, kurtosis, and skewness. A computing device that extracts the feature data using the above statistical values.