Method and computer program for determining level of consciousness

Abstract

A method for determining a patient's level of consciousness according to an embodiment of the present invention may include: an extraction step of extracting multiple components of one or more frequency bands from a first interval of brain waves; a first index calculation step of calculating a first index of each component of the one or more frequency bands, and calculating the first index based on the extent to which the size of each component of the one or more frequency bands exceeds a specified critical value relative to a specified reference component size in the first interval; a probability value calculation step of calculating the probability value of each of one or more patient states based on the first index of each component of one or more frequency bands using a learned artificial neural network; and a consciousness level determination step of determining the patient's level of consciousness based on the calculated probability value of each of the one or more patient states.

Description

Technical Field

[0001] Embodiments of the present invention relate to methods and computer programs for determining a subject's level of consciousness using a learned artificial neural network. Background Art

[0002] Generally, when a patient undergoes surgery or treatment, the patient will feel pain in the area where the surgery is being performed. In this case, anesthesia is used to block nerve conduction, thereby alleviating or eliminating the pain. Depending on the patient's symptoms or the area of ​​surgery, general anesthesia is sometimes used, and local anesthesia is sometimes used. In the case of general anesthesia, the patient cannot express himself, so the patient needs to be monitored more carefully.

[0003] For this reason, the depth of anesthesia needs to be continuously measured during surgery, and methods for measuring the depth of anesthesia generally include a method of observing clinical manifestations and a method of analyzing bioelectric signals.

[0004] Among the methods of analyzing bioelectric signals, there is a method of measuring and analyzing brain waves to evaluate the effects of anesthetics on the central nervous system. The prior art measures and analyzes brain waves in various ways.

[0005] Currently, the most widely used method for measuring the depth of anesthesia is the bispectral index (hereinafter referred to as "BIS") analysis method. The BIS analysis method is characterized by numerically processing the depth of anesthesia between 0 and 100.

[0006] It is reported that anesthesia depth measurement devices using the BIS analysis method have many problems in the accuracy of measuring the patient's anesthesia depth. Since the specific content of the device's analysis algorithm is not disclosed, it is difficult to prove that the algorithm is wrong.

[0007] Furthermore, a consciousness level measuring device using the BIS analysis method has a problem in that the tracking speed for a rapid change in the anesthesia state is slow due to the characteristics of this method, and therefore the anesthesia state of the patient cannot be detected accurately and quickly. Summary of the invention

[0008] The purpose of the present invention is to solve the above-mentioned problems, to provide accurate consciousness level measurement values ​​even if the anesthesia conditions change, and to provide consciousness level information in a timely manner even if the anesthesia and consciousness status change drastically.

[0009] Furthermore, an object of the present invention is to provide information related to the emotional state of a subject.

[0010] Furthermore, an object of the present invention is to provide anesthesia depth or consciousness level of various subjects (such as animals other than humans).

[0011] A method for determining a patient's level of consciousness according to an embodiment of the present invention may include: an extraction step of extracting multiple components of one or more frequency bands from a first interval of brain waves; a first index calculation step of calculating a first index of each component of the one or more frequency bands, and calculating the first index based on the extent to which the size of each component of the one or more frequency bands exceeds a specified critical value relative to a specified reference component size in the first interval; a probability value calculation step of calculating the probability value of each of one or more patient states based on the first index of each component of one or more frequency bands using a learned artificial neural network; and a consciousness level determination step of determining the patient's level of consciousness based on the calculated probability value of each of the one or more patient states.

[0012] In a method for determining a patient's level of consciousness according to an embodiment of the present invention, before the extraction step, the method may further include: an electroencephalogram acquisition step of acquiring the electroencephalogram of the patient; an electroencephalogram first interval generation step of generating a first interval of the electroencephalogram, wherein the first interval of the electroencephalogram includes at least a portion of the acquired electroencephalogram; and a noise removal step of removing noise from the first interval of the electroencephalogram. In this case, the extraction step may extract multiple components of one or more frequency bands from the first interval of the electroencephalogram from which the noise has been removed.

[0013] In the above-mentioned brain wave acquisition step, the brain waves of the above-mentioned patient sampled at a prescribed sampling frequency can be acquired.

[0014] In the step of generating the first electroencephalogram interval, the first electroencephalogram interval may be generated so as to include electroencephalograms within a predetermined time interval from a time point when the level of consciousness is determined.

[0015] The above-mentioned noise removal step includes the following replacement step: within the first interval of the above-mentioned brain wave, the first partial interval is replaced with a second partial interval different from the above-mentioned first partial interval, the above-mentioned first partial interval includes the time point when the size of the brain wave exceeds the specified critical size, and the above-mentioned first partial interval and the above-mentioned second partial interval can be at least a part of the above-mentioned first interval.

[0016] The above-mentioned noise removal step includes the following replacement step: when the pattern in which the size of the brain wave in the first interval of the above-mentioned brain wave exceeds the specified critical size corresponds to a preset pattern, the first interval of the above-mentioned brain wave is replaced with the second interval of the above-mentioned brain wave, and the above-mentioned second interval is different from the above-mentioned first interval and can be at least a part of the above-mentioned brain wave.

[0017] In the above-mentioned extraction step, a first component of 0.5 Hz to 4 Hz, a second component of 4 Hz to 8 Hz, a third component of 8 Hz to 16 Hz, a fourth component of 16 Hz to 25 Hz, a fifth component of 25 Hz to 30 Hz, a sixth component of 30 Hz to 48 Hz and a baseline component of 0.5 Hz to 55 Hz can be extracted from the first interval of the above-mentioned brain wave.

[0018] The multiple components of the one or more frequency bands include the size of the components of each frequency band at one or more time points belonging to the first interval, and the first index calculation step may include the following steps: calculating the size of the multiple components of the one or more frequency bands relative to the size of the reference component for each of the one or more time points; determining the time point at which the calculated size exceeds the specified critical value as the exceeding time point; and calculating the first index based on the ratio of the number of overall time points belonging to the first interval to the exceeding time point. In this case, the specified critical value can be determined based on the absolute size of the reference component in the first interval.

[0019] In the consciousness level determination method of one embodiment of the present invention, an input data generating step may be included before the above-mentioned probability value calculating step. In the above-mentioned input data generating step, the input data of the above-mentioned artificial neural network is generated by utilizing a combination of the first exponents of each component of the above-mentioned one or more frequency bands.

[0020] In a method for determining the level of consciousness of an embodiment of the present invention, the first index of each of N (N is a natural number) frequency bands is calculated from the first interval of the first brain wave of the patient obtained through the first channel, and the first index of each of the N frequency bands is calculated from the first interval of the second brain wave of the patient obtained through the second channel, and the second channel is different from the first channel. The input data generation step may include the following steps: generating N squared first input data based on the combination of the N first indexes of the first channel and the N first indexes of the second channel; generating N second input data corresponding to the N first indexes of the first channel; generating N third input data corresponding to the N first indexes of the second channel; generating M (M is a natural number) fourth input data based on the electromyography signal; and generating the input data including the first input data, the second input data, the third input data and the fourth input data. In this case, the N is 6 and the M can be 1.

[0021] The above-mentioned artificial neural network is a neural network that learns the relationship between brain wave characteristics, electromyographic characteristics and the patient's state based on learning data. The above-mentioned learning data includes data reflecting the characteristics of brain waves and data reflecting the electromyographic characteristics, and is marked with patient state data corresponding to the above-mentioned brain wave characteristics and the above-mentioned electromyographic characteristics. The above-mentioned data reflecting the characteristics of brain waves include: N squared first data generated based on a combination (Combination) of N (N is a natural number) first exponents of brain waves obtained through the first channel and N first exponents of brain waves obtained through the second channel; N second data corresponding to the N first exponents of the above-mentioned first channel; N third data corresponding to the N first exponents of the above-mentioned second channel; and M (M is a natural number) fourth data based on electromyographic signals. The above-mentioned patient's state data may include K (K is a natural number) probability values ​​corresponding to each of the patient's states.

[0022] The above-mentioned consciousness level determination step may include: a normalized probability value calculation step, normalizing the probability values ​​of each of the above-mentioned one or more patient states to calculate a normalized probability value; an application step, applying a weight set corresponding to the patient state to the above-mentioned normalized probability value, wherein the above-mentioned patient state has the largest probability value among the above-mentioned normalized probability values; and a step of determining the above-mentioned consciousness level based on the sum of the above-mentioned normalized probability values ​​to which weighted values ​​are applied.

[0023] The above-mentioned one or more patient states include awake (Awake) state, sedation (Sedation) state, general anesthesia (General Anesthesia) state, deep anesthesia (Deep Anesthesia) state and brain death (Brain Death) state. In the above-mentioned application step, one of the weight sets of each of the above-mentioned five patient states can be applied to the above-mentioned normalized probability value.

[0024] In the step of determining the above-mentioned consciousness level based on the sum of the above-mentioned multiple normalized probability values ​​with weighted values, the above-mentioned consciousness level can be determined based on the first consciousness level and the consciousness level with the second weighted value applied to the second consciousness level, the above-mentioned first consciousness level applies the specified first weighted value to the sum of the above-mentioned multiple normalized probability values ​​with weighted values, and the above-mentioned second consciousness level is determined according to the second method.

[0025] According to the present invention, the anesthesia depth or consciousness state of a patient can be measured more accurately.

[0026] In particular, the problem of existing consciousness level measurement devices based on the BIS analysis method, namely, the slow tracking speed that leads to slow reaction speed when the anesthesia level changes rapidly, is improved. Compared with the existing anesthesia depth analysis device, it can react more quickly when changing from the awake state to the anesthesia state (hypnosis), so that the patient's condition can be accurately and timely grasped.

[0027] Furthermore, due to the simplicity of the algorithm, the present invention is easy to process in real time, thereby being able to more accurately capture changes in the state of anesthesia.

[0028] Furthermore, the present invention can provide information related to the emotional state of the subject.

[0029] Furthermore, the present invention can provide anesthesia depth for various subjects (such as animals other than humans). BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A diagram schematically showing the structure of a consciousness level measuring system according to an embodiment of the present invention.

[0031] Figure 2 1 is a diagram for explaining a method in which the control unit 120 generates the first section of the electroencephalogram 400 according to an embodiment of the present invention.

[0032] Figure 3 1 is a diagram for explaining a method in which the control unit 120 generates a first segment 440FLT of an electroencephalogram from which noise has been removed by removing noise in an exemplary first segment 440 of an electroencephalogram according to an embodiment of the present invention.

[0033] Figure 4 4 is a diagram for explaining a process in which the control unit 120 extracts a plurality of components 411 , 412 , 413 , 414 , 415 , 416 , 417 , and 418 of one or more frequency bands from a first interval 410FLT of an electroencephalogram from which noise has been removed according to an embodiment of the present invention.

[0034] Figure 5 FIG. 1 is a diagram for explaining a method of calculating a first index by the control unit 120 according to an embodiment of the present invention.

[0035] Figure 6 FIG. 1 is a diagram for explaining a method of calculating the second index by the control unit 120 according to an embodiment of the present invention.

[0036] Figure 7 FIG. 1 is a diagram for explaining a process in which the control unit 120 causes the artificial neural network 520 to learn using a plurality of learning data 510 according to an embodiment of the present invention.

[0037] Figure 81 is a diagram for explaining patient status data 540 as output data of input data 531 , 532 , 533 , and 534 of an artificial neural network 520 according to an embodiment of the present invention.

[0038] Fig. 9 6 is a diagram showing one or more exemplary weight sets 610, 620, 630, 640, 650 in graph form.

[0039] Fig.10 1 is a diagram for explaining a process in which the control unit 120 determines the final consciousness level of the patient 300 according to an alternative embodiment of the present invention.

[0040] Fig.11 FIG. 1 is a flowchart for illustrating a method for determining a consciousness level executed by the user terminal 100 according to an embodiment of the present invention.

[0041] Description of Reference Numerals

[0042] 100: User terminal

[0043] 110: Ministry of Communications

[0044] 120: Control Department

[0045] 130: Memory

[0046] 140: Display unit

[0047] 200: Consciousness level measuring device

[0048] 210: Inspection Department

[0049] 220: Signal Processing Department

[0050] 211 to 214: Electrodes

[0051] 300: Patient DETAILED DESCRIPTION

[0052] The present invention is susceptible to various modifications and may have various embodiments, so specific embodiments will be illustrated in the drawings and described in detail in the detailed description. The effects, features and methods of achieving the same will become more apparent with reference to the drawings and the embodiments described in detail below. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various forms.

[0053] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. When describing with reference to the accompanying drawings, the same or corresponding structural elements will be given the same reference numerals and repeated description thereof will be omitted.

[0054] In the following embodiments, terms such as "first" and "second" are used to distinguish one structural element from other structural elements and do not have a restrictive meaning. In the following embodiments, expressions in the singular include expressions in the plural, unless otherwise clearly stated in the context. In the following embodiments, terms such as "including" or "having" mean that the features or structural elements described in the specification are present, without excluding in advance the possibility of the addition of more than one other features or structural elements. In the accompanying drawings, the size of the structural elements may be enlarged or reduced for ease of explanation. For example, for ease of explanation, the size and shape of each structure shown in the accompanying drawings are arbitrarily shown, so this description is not necessarily limited to what is shown.

[0055] Figure 1 A diagram schematically showing the structure of a consciousness level measuring system according to an embodiment of the present invention.

[0056] The consciousness level measurement system of one embodiment of the present invention can measure the consciousness level of the patient based on the vital signals of the patient 300. In this case, the "vital signals" may refer to various forms of signals directly or indirectly measured from the body of the patient 300, such as the brain waves of the patient 300, the electromyography signals of the patient 300, the electrooculogram (EOG) signals, etc. In addition, the "consciousness level" may refer to the degree to which the patient 300 can normally perceive and distinguish himself and the surrounding environment around him, and the degree to which he can wake up when receiving specific stimulation. In the present invention, "depth of anesthesia" is sometimes used as an indicator of the consciousness level, but this is illustrative and the concept of the present invention is not limited to this. In addition to the depth of anesthesia, the emotional state of the patient 300, the patient's sleep state, etc. can also be used as indicators to represent the consciousness level.

[0057] On the other hand, the patient 300 is used as an example of an object. Figure 1 However, the concept of the present invention is not limited thereto. Figure 1 As shown, the subject may be a patient 300, that is, a human being, or may be an animal as a subject.

[0058] like Figure 1 As shown, a consciousness level measuring system according to an embodiment of the present invention may include a user terminal 100 and a consciousness level measuring device 200 .

[0059] The consciousness level measuring device 200 of one embodiment of the present invention may include: a detection unit 210 attached to the body of the patient 300 to obtain the patient's living body signal; and a signal processing unit 220, processing the living body signal of the patient 300 obtained by the detection unit 210 and transmitting it to the user terminal 100.

[0060] As described above, the detection unit 210 of one embodiment of the present invention may refer to a unit that acquires a living body signal of the patient 300 by being attached to the body of the patient. Figure 1 As shown, the detection unit 210 may include: a reference electrode 211 for setting a reference potential; a ground electrode 212 for setting a ground potential; a first channel electrode 213 for measuring brain waves and electromyographic signals; and a second channel electrode 214 for measuring brain waves.

[0061] When the detection unit 210 is worn on the head of the patient 300 , the plurality of electrodes 211 , 212 , 213 , and 214 may be attached to the scalp of the patient 300 non-invasively or invasively to acquire a living body signal.

[0062] on the other hand, Figure 1 The shape of the detection unit 210, the number of the plurality of electrodes 211, 212, 213, 214 included in the detection unit 210, and the arrangement of the plurality of electrodes 211, 212, 213, 214 are illustrative only, and the concept of the present invention is not limited thereto. Therefore, any unit that can acquire a living body signal of the patient 300 by being attached to the body of the patient 300 can correspond to the detection unit 210 of the present invention.

[0063] The detection unit 210 of another embodiment of the present invention may omit one of the reference electrode 211 and the ground electrode 212. In other words, the detection unit 210 of another embodiment of the present invention may be configured to include the reference electrode 211, the first channel electrode 213 and the second channel electrode 214, or may be configured to include the ground electrode 212, the first channel electrode 213 and the second channel electrode 214. In this case, the reference potential and the ground potential are equivalent to the same potential, which can be set by the reference electrode 211 or the ground electrode 212.

[0064] The signal processing unit 220 of one embodiment of the present invention may refer to a unit that processes the living body signal obtained by the detection unit 210 and transmits it to the user terminal 100. In this case, "processing" the signal may refer to the user terminal 100 processing the signal in a calculable form, for example, sampling the signal or amplifying the signal.

[0065] The signal processing unit 220 of one embodiment of the present invention may amplify the living body signal acquired by the detection unit 210. Furthermore, the signal processing unit 220 of one embodiment of the present invention may sample the living body signal acquired by the detection unit 210 at a predetermined sampling frequency.

[0066] For example, the signal processing unit 220 may amplify the brain waves acquired by the first channel electrode 213 and the second channel electrode 214 of the detection unit 210 at a predetermined ratio and then perform sampling at a sampling frequency of 250 Hz.

[0067] Similarly, the signal processing unit 220 may amplify the electromyographic signal acquired by the first channel electrode 213 of the detection unit 210 at a predetermined ratio and then perform sampling at a sampling frequency of 250 Hz.

[0068] The signal processing unit 220 of an embodiment of the present invention can transmit the living body signal amplified and sampled through the above process to the user terminal 100 through various communication methods. For example, the signal processing unit 220 can transmit the living body signal (amplified and sampled) to the user terminal 100 through Bluetooth communication, or can transmit the living body signal to the user terminal 100 through Wi-Fi communication. Of course, the signal processing unit 220 can transmit the living body signal to the user terminal 100 using various well-known wired communication methods.

[0069] The user terminal 100 of one embodiment of the present invention can measure the anesthesia depth of the patient 300 based on the living body signal (amplified and sampled signal, for example, brain wave, electromyogram signal and electrooculogram signal obtained in two channels respectively) transmitted by the signal processing unit 220.

[0070] In this case, the user terminal 100 may be a general electronic device provided with an application for measuring the level of consciousness. For example, the user terminal 100 may be a mobile phone (or tablet computer) provided with an application for measuring the level of consciousness. Furthermore, the user terminal 100 may also be a dedicated electronic device that only drives the application for measuring the level of consciousness.

[0071] like Figure 1 As shown, the user terminal 100 of an embodiment of the present invention may include a communication unit 110, a control unit 120, a memory 130 and a display unit 140. In addition, although not shown, the user terminal 100 of this embodiment may also include an input / output unit and a program storage unit.

[0072] The communication unit 110 of an embodiment of the present invention may include hardware and software required for the user terminal 100 to send and receive signals such as control signals or data signals through wired or wireless connections with other devices such as the signal processing unit 220. For example, the communication unit 110 may include hardware and software for sending and receiving signals with the signal processing unit 220 via Bluetooth. On the other hand, in the case where the user terminal 100 is a general electronic device, the communication unit 110 may also include a communication modem and software for sending and receiving data using a general communication network (e.g., an LTE communication network, a 3G communication network, a Wi-fi communication network, etc.).

[0073] The control unit 120 of one embodiment of the present invention may include all types of devices capable of processing data, such as a processor. For example, a "processor" may refer to a data processing device built into hardware, having a circuit physically structured for executing the functions represented by the code or instructions in the program. As described above, as an example of a data processing device built into hardware, it may include a microprocessor, a central processing unit (CPU), a processor core (Processor Core), a multiprocessor (Multiprocessor), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) and other processing devices, but the scope of the present invention is not limited to this.

[0074] The memory 130 of one embodiment of the present invention performs the function of temporarily or permanently storing the data processed by the user terminal 100. The memory may include a magnetic storage medium or a flash storage medium, but the scope of the present invention is not limited thereto. For example, the memory 130 may temporarily and / or permanently store the living body signal received from the signal processing unit 220. In addition, the memory 130 may temporarily and / or permanently store a plurality of weight values ​​constituting the learned artificial neural network.

[0075] The display unit 140 of one embodiment of the present invention can perform the function of providing the user with the measured anesthesia depth. The display unit 140 can be implemented as a variety of well-known display devices, such as a liquid crystal display (LCD), an organic light emitting semiconductor (OLED), and a micro light emitting semiconductor (Micro LED). However, this is exemplary, and a device that displays a screen according to an electronic control signal can be equivalent to the display unit 140 of the present invention.

[0076] Hereinafter, the description will be mainly focused on a method in which the control unit 120 of the user terminal 100 measures the anesthesia depth of the patient.

[0077] The control unit 120 according to an embodiment of the present invention can obtain a living body signal of the patient 300 .

[0078] The control unit 120 of one embodiment of the present invention can obtain the brain waves of the patient 300. For example, the control unit 120 can obtain the brain waves by receiving the brain waves from the signal processing unit 220. In this case, the obtained brain waves can be brain waves that have been amplified and sampled at a specified sampling frequency. In addition, the control unit 120 can obtain two brain waves obtained from the signal processing unit 220 through different channels.

[0079] On the other hand, the control unit 120 of an embodiment of the present invention can also obtain brain waves and electromyographic signals at the same time. Of course, the electromyographic signals obtained in this case can also be electromyographic signals that have been amplified and sampled at a specified sampling frequency.

[0080] The control unit 120 of an embodiment of the present invention can obtain the living body signal of the patient 300 in real time. In this case, the control unit 120 can temporarily and / or permanently store the past living body signal in the memory 130 for measuring the anesthesia depth of the patient.

[0081] The control unit 120 according to an embodiment of the present invention may generate a first interval of a brain wave, where the brain wave includes at least a portion of the acquired brain wave.

[0082] Figure 2 1 is a diagram for explaining a method in which the control unit 120 generates the first section of the electroencephalogram 400 according to an embodiment of the present invention.

[0083] In the following, for the sake of convenience, the electroencephalogram 400 of the patient 300 is shown in the figure, and the current time point is assumed to be 4 seconds, 5 seconds, and 6 seconds respectively according to the situation.

[0084] In the process of generating the first interval of the brain wave 400, the control unit 120 of one embodiment of the present invention can generate the first interval of the brain wave in a manner that includes past brain waves within a specified time interval starting from the time point when the anesthesia depth is determined (i.e., the current time point).

[0085] For example, assuming that the current time point for determining the depth of anesthesia is 4 seconds, the control unit 120 may generate a first interval 410 of brain waves including past brain waves within a predetermined time interval (assuming 4 seconds) from the current time point (4 seconds). Similarly, when the current time point is 5 seconds, the control unit 120 may generate a first interval 420 of brain waves, and when the current time point is 6 seconds, the control unit 120 may generate a first interval 430 of brain waves.

[0086] In this case, the "prescribed time interval" can be set in various ways according to the system characteristics. For example, in a system that requires a quick response, the prescribed time interval can be set to be relatively short. Also, in a system that requires an accurate response, the prescribed time interval can be set to be relatively long.

[0087] The control unit 120 of one embodiment of the present invention can generate the first interval through the above process for two brain waves obtained through different channels. In addition, the control unit 120 of one embodiment of the present invention can repeatedly generate the first interval with the current time point as time goes by (i.e., as the time point for determining the depth of anesthesia changes).

[0088] The control unit 120 according to an embodiment of the present invention may remove noise in the first interval of the brain wave generated according to the above process.

[0089] Figure 3 1 is a diagram for explaining a method in which the control unit 120 generates a first segment 440FLT of an electroencephalogram from which noise has been removed by removing noise in an exemplary first segment 440 of an electroencephalogram according to an embodiment of the present invention.

[0090] In the first section 440 of the brain wave, the control unit 120 according to one embodiment of the present invention may replace the first partial section 444A in which the magnitude of the brain wave is greater than the predetermined critical magnitude Ath with a second partial section 443A different from the first partial section.

[0091] At this time, when the size of the brain wave at any time point in the first partial interval 444A is greater than the predetermined critical size Ath, the control unit 120 may determine that the size of the brain wave in the corresponding interval 444A is greater than the predetermined critical size Ath. In addition, the control unit 120 may replace the corresponding interval 444A with the interval 443A adjacent to the corresponding interval 444A.

[0092] Thus, the control unit 120 can generate a first interval 440FLT of the brain wave from which noise has been removed, wherein the first partial interval 441B, the second partial interval 442B, and the third partial interval 443B of the first interval 440FLT are the same as the first partial interval 441A, the second partial interval 442A, and the third partial interval 443A of the first interval 440, and the fourth partial interval 444B is the same as the third partial interval 443A of the first interval 440.

[0093] However, the length of the partial interval, the critical size Ath, and the replacement method of the partial interval are merely illustrative, and the concept of the present invention is not limited thereto.

[0094] In an optional embodiment, when the pattern in which the brain wave size in the first interval 440 of the brain wave is greater than the specified critical size corresponds to the preset pattern, the control unit 120 may also replace the first interval 440 of the brain wave with the second interval of the brain wave. In this case, the second interval of the brain wave may be a brain wave of an interval different from the first interval 440 of the brain wave, such as a brain wave of an adjacent interval.

[0095] Furthermore, the "preset mode" can be set in a variety of ways according to the purpose of the system. For example, the preset mode can be a mode in which multiple partial intervals within the same interval generate brain waves with a size greater than a specified critical size, or a mode in which brain waves with a size greater than a specified critical size are generated in two or more consecutive partial intervals. However, the above modes are illustrative only, and the concept of the present invention is not limited thereto.

[0096] The control unit 120 according to an embodiment of the present invention may extract a plurality of components of one or more frequency bands from the first section of the brain wave generated by the above process (or the first section of the brain wave from which noise has been removed).

[0097] Figure 4 4 is a diagram for explaining a process in which the control unit 120 extracts a plurality of components 411 , 412 , 413 , 414 , 415 , 416 , 417 , and 418 of one or more frequency bands from a first interval 410FLT of an electroencephalogram from which noise has been removed according to an embodiment of the present invention.

[0098] The control unit 120 of one embodiment of the present invention can extract a first component 412 of 0.5 Hz to 4 Hz, a second component 413 of 4 Hz to 8 Hz, a third component 414 of 8 Hz to 16 Hz, a fourth component 415 of 16 Hz to 25 Hz, a fifth component 416 of 25 Hz to 30 Hz, a sixth component 417 of 30 Hz to 48 Hz, a baseline component 411 of 0.5 Hz to 55 Hz, and a burst suppression rate (BSR) component 418 of 0.5 Hz to 30 Hz from the first interval 410FLT of the brain wave from which noise has been removed.

[0099] The control unit 120 of one embodiment of the present invention can use frequency filters corresponding to one or more frequency bands to extract multiple components 411, 412, 413, 414, 415, 416, 417, and 418 of one or more frequency bands from the first interval 410FLT of the brain wave from which noise has been removed. For example, the control unit 120 can use a bandpass filter (Ban-PassFilter) with a passband of 0.5 Hz to 4 Hz to extract the first component 412.

[0100] The extracted multiple components 411, 412, 413, 414, 415, 416, 417, 418 of one or more frequency bands may have the same time length as the first interval 410FLT of the brain wave from which noise has been removed. In addition, the multiple components 411, 412, 413, 414, 415, 416, 417, 418 of one or more frequency bands may include the size of the component of each frequency band at one or more time points belonging to the first interval.

[0101] For example, when the sampling frequency of the signal processing unit 220 is 250 Hz and the length of the first interval 410FLT of the brain wave from which noise has been removed is 4 seconds, the first interval 410FLT of the brain wave from which noise has been removed may include 1000 time points (or 1000 sampling time points) and the size of the brain wave at each time point. Thus, the multiple components 411, 412, 413, 414, 415, 416, 417, 418 of one or more frequency bands may include the size of the component of each frequency band at each of the 1000 time points belonging to the first interval.

[0102] In other words, the plurality of components 411 , 412 , 413 , 414 , 415 , 416 , 417 , 418 of one or more frequency bands may include the magnitude of the component of each frequency band at 1000 time points belonging to the first interval.

[0103] The control unit 120 according to an embodiment of the present invention may calculate the first index of each of the plurality of components 411 , 412 , 413 , 414 , 415 , 416 , 417 , and 418 of one or more frequency bands.

[0104] Figure 5 1 is a diagram for explaining a method for calculating a first index by the control unit 120 according to an embodiment of the present invention. Figure 5 The graph shown relates to the first component, and the description is given on the assumption that the first interval of the first component includes 15 time points.

[0105] The control unit 120 of one embodiment of the present invention can calculate the first index based on the extent to which the size of each component of one or more frequency bands of the predetermined reference component size in the first interval exceeds the predetermined threshold value Rth. For example, the control unit 120 of one embodiment of the present invention can calculate the sizes of multiple components of one or more frequency bands of the reference component for one or more time points (i.e., 15 time points) belonging to the first interval.

[0106] In this case, the benchmark composition may refer to Figure 4 The process of extracting the reference component 411 is described. The multiple components of more than one frequency band may be referred to as Figure 4 The control unit 120 can extract the multiple components 412 to 417 by the process described above. For example, the control unit 120 can calculate the ratio R1 of the size of the reference component 411 at the first time point of the first interval to the size of the first component 412 at the first time point of the first interval. Similarly, the control unit can calculate the ratios R2 to R15 of the size of the reference component 411 to the size of the first component 412 at the remaining time points. Of course, for the multiple remaining components 413 to 417, the control unit 120 can also calculate the respective ratios at multiple time points through the above process.

[0107] The control unit 120 of one embodiment of the present invention may determine the time point when the calculated ratio exceeds the predetermined threshold value Rth as the exceeding time point. Figure 5 The time points corresponding to R4, R12, R13, and R14 are determined as the exceeding time points. Of course, for the remaining components 413 to 417, the control unit can also determine multiple exceeding time points through the above process.

[0108] The control unit 120 of one embodiment of the present invention can calculate the first index based on the ratio of the number of overall time points belonging to the first interval to the number of exceeded time points. For example, for the first component 412, the control unit 120 can use the number of overall time points 15 and the number of exceeded time points 4 to calculate the first index as 4 / 15. In this case, the control unit 120 can normalize the first index by multiplying the calculated first index by a specified value. Of course, for the remaining multiple components 413 to 417, the control unit can also calculate the first index through the above process respectively.

[0109] On the other hand, the control unit 120 of one embodiment of the present invention can determine the prescribed threshold value Rth based on the absolute size of the reference component in the first interval. In the case of the reference component, the reference component is extracted in a wide frequency band, so when the size of the brain wave in the first interval is large, the prescribed threshold value Rth can be determined to be relatively high, and when the size of the brain wave is small, the prescribed threshold value Rth can be determined to be relatively low.

[0110] For two brain waves acquired through different channels, the control unit 120 of one embodiment of the present invention can respectively calculate the first index based on the above process. In addition, the control unit 120 of one embodiment of the present invention can repeatedly calculate the updated first index of the first interval according to the update of the first interval.

[0111] The control unit 120 of one embodiment of the present invention may calculate a second index, which reflects the degree to which the brain waves in the first interval conform to a prescribed pattern.

[0112] Figure 6 FIG. 1 is a diagram for explaining a method of calculating the second index by the control unit 120 according to an embodiment of the present invention.

[0113] The control unit 120 of one embodiment of the present invention may calculate the second index according to the burst suppression rate component 418. Figure 4 The process described above is to extract the burst suppression rate component 418 from the first segment 410FLT of the brain wave from which the noise has been removed. In this case, the burst suppression rate component 418 may be a component of 0.5 Hz to 30 Hz extracted from the first segment 410FLT of the brain wave from which the noise has been removed.

[0114] The control unit 120 of one embodiment of the present invention may confirm the number of repetitions of the first type of signal (or burst type signal) and the second type of signal (or suppression type signal) in the burst suppression rate component 418. The first type of signal is a type of brain wave that generates a size (i.e., the absolute value of the amplitude of the signal) greater than a first critical size within a first duration, and the second type of signal may be a type of brain wave that generates a size less than a second critical size within a second duration. In this case, the first duration may be shorter than the second duration.

[0115] exist Figure 6 In the embodiment, the first type of signal is a type of brain wave having a size greater than a first critical size Ath_1_P or Ath_1_N generated in a first duration shorter than a second duration, and the second type of signal is a type of brain wave having a size less than a second critical size Ath_2_P or Ath_2_N generated in a second duration longer than the first duration. In this case, the first critical size Ath_1_P or Ath_1_N and the second critical size Ath_2_P or Ath_2_N may be preset to appropriate sizes.

[0116] therefore, Figure 6 In the figure, B1, B2, and B3 may be signals of the first type, and S1, S2, and S3 may be signals of the second type, respectively.

[0117] on the other hand, Figure 6 In one embodiment of the present invention, the control unit 120 may confirm (or determine) the number of repetitions of the first type of signal and the second type of signal to be three.

[0118] The control unit 120 of one embodiment of the present invention can calculate the second index based on the ratio of the value corresponding to the length of the first interval 410FLT of the brain wave to the number of repetitions of the signal. In this case, for example, the value corresponding to the length of the first interval 410FLT of the brain wave may refer to a value proportional to the number of time points (e.g., 1000) included in the first interval 410FLT of the brain wave. Therefore, the control unit 120 can determine the second index as a value proportional to 3 / 1000 or 3 / 1000 based on the burst suppression rate component 418.

[0119] on the other hand, Figure 6 The method described is an exemplary method for determining the extent to which the brain waves in the first interval 410FLT conform to a prescribed pattern, but the concept of the present invention is not limited thereto. Any method that can determine whether a signal of a specific unit pattern of the brain waves in the first interval 410FLT is repeated and the number of repetitions can be used without restriction in the present invention to calculate the second index.

[0120] The control unit 120 of one embodiment of the present invention can calculate the third index based on at least one living body signal corresponding to the patient's state. For example, the control unit 120 of one embodiment of the present invention can generate the third index based on the electromyography signal measured by the first channel electrode 213 of the detection unit 210.

[0121] However, as described above, the electromyographic signal is used as an example as an element for generating the third index, and the calculation method of the third index may be different depending on the configuration of the consciousness level measurement system. For example, when the consciousness level measurement system further includes a component for measuring other biological signals of the patient 300, the control unit 120 may also generate the third index based on the signal measured by the corresponding component.

[0122] In an alternative embodiment, the control unit 120 may generate the third index based on the electrooculogram signal of the patient.

[0123] The control unit 120 according to an embodiment of the present invention can determine the anesthesia depth of the patient 300 based on at least one of the first index, the second index, and the third index calculated through the above process.

[0124] The control unit 120 of an embodiment of the present invention may determine the anesthesia depth of the patient 300 by using a learned artificial neural network and a first index of each component of one or more frequency bands and a third index based on an electromyography signal.

[0125] In the present invention, "artificial neural network" may refer to a neural network that learns the relationship between brain wave characteristics, electromyographic characteristics and the patient's condition based on learning data, wherein the learning data includes data reflecting the brain wave characteristics and data reflecting the electromyographic characteristics, and is marked with patient condition data corresponding to the brain wave characteristics and the electromyographic characteristics.

[0126] In other words, the artificial neural network may refer to a neural network that learns to output patient status data in response to input of data reflecting characteristics of brain waves and data reflecting characteristics of electromyograms.

[0127] In the present invention, the artificial neural network can be implemented as a neural network model of various structures. For example, the artificial neural network can be implemented as a convolutional neural network (CNN) model, a recursive neural network (RNN) model, and a long short-term memory (LSTM) model. However, the above neural network is exemplary, and a unit that can learn the relationship between input and output based on learning data can be used as the artificial neural network of the present invention.

[0128] Figure 7 1 is a diagram for explaining a process in which the control unit 120 causes the artificial neural network 520 to learn using a plurality of learning data 510 according to an embodiment of the present invention.

[0129] As described above, the artificial neural network 520 may refer to a neural network that learns the relationship between the characteristics of brain waves, the characteristics of electromyography, and the state of the patient based on the plurality of learning data 510 .

[0130] In this case, as described above, the plurality of learning data 510 may be data including data reflecting the characteristics of brain waves and data reflecting the characteristics of electromyograms, and data of the patient's status corresponding to the characteristics of brain waves and electromyograms may be marked.

[0131] For example, in the case of the first learning data 511, the data reflecting the characteristics of the brain wave may include: N data 511A corresponding to the first index of each of N (N is a natural number) frequency bands calculated from the first interval of the first brain wave of the patient 300 (obtained through the first channel), N data 511B corresponding to the first index of each of the N frequency bands calculated from the first interval of the second brain wave of the patient 300 (obtained through the second channel), and N squared data 511D generated based on the combination of the N first indices 511A of the first channel and the N first indices 511B of the second channel.

[0132] Furthermore, in the case of the first learning data 511, the data reflecting the electromyographic characteristics may include M (M is a natural number) data 511C based on the electromyographic signal.

[0133] Furthermore, the first learning data 511 may be marked with patient status data 511E, and the patient status data 511E includes the probabilities that the corresponding patient 300 corresponds to K patient statuses. In this case, for example, N may be 6, M may be 1, and K may be 5.

[0134] The remaining learning data including the second learning data 512 and the third learning data 513 also include the same data as the first learning data and can be marked with the same data. Figure 8 A specific method for generating individual data included in individual learning data will be described later.

[0135] As described above, the present invention can enable the artificial neural network 520 to learn based on the learning data, and the learning data includes data generated by the patient's biological signal and the probabilities corresponding to the multiple patient states. Thus, the artificial neural network 520 can output the probabilities corresponding to the corresponding patient and the multiple patient states according to the input of the data generated by the patient's state signal. Figure 7 The artificial neural network 520 described in the description is described on the premise that the learning based on the plurality of learning data 510 has been completed.

[0136] Figure 81 is a diagram for explaining patient status data 540 as output data of input data 531 , 532 , 533 , and 534 of an artificial neural network 520 according to an embodiment of the present invention.

[0137] The control unit 120 of one embodiment of the present invention can generate input data 531, 532, 533, 534 of the artificial neural network 520, so as to use the artificial neural network 520 learned through the above process to calculate the probability values ​​of the patient 300 belonging to more than one patient state. In this case, the artificial neural network 520 may include: an input layer 521, including at least one input code input with input data 531, 532, 533, 534; an intermediate layer 522 (or hidden layer), including multiple intermediate codes (or hidden codes); and an output layer 523, including at least one output code. As shown in the figure, the intermediate layer 522 may include more than one fully connected layer. In the case where the intermediate layer 522 includes multiple layers, the artificial neural network 520 may include a function for defining the relationship between each hidden layer.

[0138] The control unit 120 Figures 2 to 5 The described process calculates the first index of each of the N frequency bands from the first interval of the first brain wave of the patient 300 obtained through the first channel. The control unit 120 calculates the first index of each of the N frequency bands from the first interval of the second brain wave of the patient 300 obtained through the second channel through the same process. For the sake of convenience, the following description will be based on the above content.

[0139] Under the above premise, the control unit 120 of one embodiment of the present invention can generate N square first input data 533 based on the combination of N first indices of the first channel and N first indices of the second channel. The first input data 533 can be the learning data used for learning the artificial neural network 520. Figure 7 The item corresponding to the data 511D.

[0140] For example, the control unit 120 may generate the first input data 533 to include data obtained by multiplying the first index of the first channel by the N indexes of the second channel and data obtained by multiplying the second index of the first channel by the N indexes of the second channel. Of course, for the remaining indexes of the first channel, the control unit 120 may generate the first input data 533 in the same manner as described above.

[0141] The control unit 120 of one embodiment of the present invention may generate N second input data 532 corresponding to the N first indexes of the first channel. The second input data 532 may be the learning data used for learning the artificial neural network 520. Figure 7For example, the control unit 120 may generate N second input data 532 so that the N first indexes of the first channel and the N second input data 532 are respectively the same value.

[0142] The control unit 120 of one embodiment of the present invention may generate N third input data 534 corresponding to the N first indexes of the second channel. The third input data 534 may be the learning data used for learning the artificial neural network 520. Figure 7 For example, the control unit 120 may generate N third input data 534 so that the N first indexes of the second channel and the N third input data 534 are respectively the same value.

[0143] The control unit 120 of one embodiment of the present invention may generate M fourth input data 531 based on the electromyographic signal. The fourth input data 531 may be the learning data used for learning the artificial neural network 520. Figure 7 For example, the control unit 120 of one embodiment of the present invention may generate the fourth input data 531 using the third index generated by the above process.

[0144] In one embodiment of the present invention, N may be 6, and M may be 1. Therefore, the first input data 533 includes 36 data, the second input data 532 and the third input data 534 each include 6 data, and the fourth input data 531 may include 1 data. However, as described above, the number of data included in each of the data 531 to 534 is exemplary, and the concept of the present invention is not limited thereto.

[0145] The control unit 120 of one embodiment of the present invention can obtain the patient's status data 540 by inputting the input data 531, 532, 533, and 534 generated by the artificial neural network 520 according to the above process. The above patient's status data 540 includes the probability value that the patient 300 belongs to more than one patient status.

[0146] The patient status data 540 of one embodiment of the present invention may include probability values ​​of K (K is a natural number) patient statuses corresponding to the patient 300. For example, K is 5, and the one or more patient statuses may include awake status (Awake) Status 1, sedation status (Sedation) Status 2, general anesthesia status (General Anesthesia) Status 3, hyper anesthesia or deep anesthesia status (Hyper or Deep Anesthesia) Status 4, and brain death status (Status 5).

[0147] For example, the control unit 120 may obtain the patient's state data 540 from the artificial neural network 520, such as [0.81, 0.62, 0.34, 0.17, 0.01]. The above patient's state data 540 may mean that the probability that the patient 300 is in an awake state (Awake) is 81%, the probability that the patient 300 is in a sedation state (Sedation) is 62%, the probability that the patient 300 is in a general anesthesia state (General Anesthesia) is 34%, the probability that the patient 300 is in a hyperanesthesia or general anesthesia state (Hyper or Deep Anesthesia) is 17%, and the probability that the patient 300 is in a brain death state (Brain Death) is 1%.

[0148] In another embodiment of the present invention, the K states and / or the number may be set in a variety of ways. For example, when the present invention is used to determine the emotional state of a patient, the K states may be multiple states representing the patient's emotions. However, this is for illustrative purposes only, and the concept of the present invention is not limited thereto.

[0149] The control unit 120 according to an embodiment of the present invention may determine the anesthesia depth of the patient 300 based on the probability values ​​of each of more than one patient status.

[0150] More specifically, the control unit 120 of one embodiment of the present invention can calculate the normalized probability value by normalizing the probability values ​​of each of the more than one patient states included in the patient state data 540. For example, the control unit 120 can calculate the normalized probability value of each of the more than one patient states in such a way that the sum of the probability values ​​of each of the more than one patient states is 1.

[0151] Furthermore, the control unit 120 according to an embodiment of the present invention may apply a weight set corresponding to the patient state to the normalized probability value, wherein the weight set has the largest probability value among the normalized probability values.

[0152] Fig. 9The figure shows one or more exemplary weight sets 610, 620, 630, 640, 650 in the form of a curve graph. The first weight set 610 is a weight set corresponding to the awake state (Awake) Status 1, the second weight set 620 is a weight set corresponding to the sedation state (Sedation) Status 2, the third weight set 630 is a weight set corresponding to the general anesthesia state (General Anesthesia) Status 3, the fourth weight set 640 is a weight set corresponding to the hyperanesthesia or deep anesthesia state (Hyper or Deep Anesthesia) Status 4, and the fifth weight set 650 is a weight set corresponding to the brain death state (Brain Death) Status 5.

[0153] For example, when the normalized probability value calculated according to the above process is [0.45, 0.25, 0.15, 0.1, 0.05], the probability that the patient 300 is in the awake state (Awake) Status 1 is the highest. Therefore, the control unit 120 can calculate the probability such as [0.45, 0.2, 0.5, 0.01, 0] by applying the weight set 610 corresponding to the awake state (Awake) Status 1 to the normalized probability value.

[0154] The control unit 120 according to an embodiment of the present invention may determine the anesthesia depth of the patient 300 based on the sum of a plurality of normalized probability values ​​to which weighted values ​​are applied.

[0155] In an alternative embodiment, the control unit 120 of an embodiment of the present invention may determine the final anesthesia depth of the patient 300 by referring to the anesthesia depth of the patient 300 determined by other means.

[0156] Fig.10 1 is a diagram for explaining a process in which the control unit 120 determines the final anesthesia depth of the patient 300 according to an alternative embodiment of the present invention.

[0157] The control unit 120 of the optional embodiment of the present invention can determine the final anesthesia depth DOA_F of the patient 300 based on the sum of the first anesthesia depth and the second anesthesia depth. The first anesthesia depth specifies the weighted value W1 to be applied according to Figures 2 to 9 The second method for calculating the anesthesia depth DOA_2 may refer to at least a portion of the process and the second method for calculating the anesthesia depth DOA_2. Figures 2 to 9The described process is different from the method for calculating the depth of anesthesia. For example, the second depth of anesthesia can be determined based on a first index, a second index, and a third index, wherein the first index is calculated from the patient's brain wave, the second index reflects the degree to which the patient's brain wave conforms to a prescribed pattern, and the third index is based on at least one living body signal corresponding to the patient's state.

[0158] Fig.11 FIG. 1 is a flowchart for explaining a method for determining a level of consciousness performed by a user terminal 100 according to an embodiment of the present invention. Figures 1 to 10 The description of the content repeated in the description will be referred to at the same time Figures 1 to 10 To explain.

[0159] The user terminal 100 according to an embodiment of the present invention may enable the artificial neural network to learn based on a plurality of learning data (step S910 ).

[0160] In the present invention, "artificial neural network" may refer to a neural network that learns the relationship between brain wave characteristics, electromyographic characteristics and the patient's condition based on learning data, wherein the learning data includes data reflecting the brain wave characteristics and data reflecting the electromyographic characteristics, and is marked with patient condition data corresponding to the brain wave characteristics and the electromyographic characteristics.

[0161] In other words, the artificial neural network may refer to a neural network that learns to output patient status data in response to input of data reflecting characteristics of brain waves and data reflecting characteristics of electromyograms.

[0162] In the present invention, the artificial neural network can be implemented as a neural network model of various structures. For example, the artificial neural network can be implemented as a convolutional neural network (CNN) model, a recursive neural network (RNN) model, and a long short-term memory (LSTM) model. However, the above neural network is exemplary, and a unit that can learn the relationship between input and output based on learning data can be used as the artificial neural network of the present invention.

[0163] Figure 7 1 is a diagram for explaining a process in which the user terminal 100 enables the artificial neural network 520 to learn using a plurality of learning data 510 according to an embodiment of the present invention.

[0164] As described above, the artificial neural network 520 may refer to a neural network that learns the relationship between the characteristics of brain waves, the characteristics of electromyography, and the state of the patient based on the plurality of learning data 510 .

[0165] In this case, as described above, the plurality of learning data 510 may be data including data reflecting the characteristics of brain waves and data reflecting the characteristics of electromyograms, and data of the patient's status corresponding to the characteristics of brain waves and electromyograms may be marked.

[0166] For example, in the case of the first learning data 511, the data reflecting the characteristics of the brain wave may include: N data 511A corresponding to the first index of each of N (N is a natural number) frequency bands calculated from the first interval of the first brain wave of the patient 300 (obtained through the first channel), N data 511B corresponding to the first index of each of the N frequency bands calculated from the first interval of the second brain wave of the patient 300 (obtained through the second channel), and N squared data 511D generated based on the combination of the N first indices 511A of the first channel and the N first indices 511B of the second channel.

[0167] Furthermore, in the case of the first learning data 511, the data reflecting the electromyographic characteristics may include M (M is a natural number) data 511C based on the electromyographic signal.

[0168] Furthermore, the first learning data 511 may be marked with patient status data 511E, and the patient status data 511E includes the probabilities that the corresponding patient 300 corresponds to K patient statuses. In this case, for example, N may be 6, M may be 1, and K may be 5.

[0169] The remaining learning data including the second learning data 512 and the third learning data 513 also include the same data as the first learning data and can be marked with the same data.

[0170] As described above, the present invention can enable the artificial neural network 520 to learn based on the learning data, and the learning data includes data generated by the patient's biological signal and the probabilities corresponding to the multiple patient states. Thus, the artificial neural network 520 can output the probabilities corresponding to the corresponding patient and the multiple patient states according to the input of the data generated by the patient's state signal. Figure 7 The artificial neural network 520 described in the description is described on the premise that the learning based on the plurality of learning data 510 has been completed.

[0171] The user terminal 100 according to an embodiment of the present invention may acquire a living body signal of the patient 300 (step S920 ).

[0172] The user terminal 100 of an embodiment of the present invention can obtain a living body signal of the patient 300. For example, the user terminal 100 can obtain brain waves by receiving brain waves from the above-mentioned signal processing unit 220. In this case, the obtained brain waves can be brain waves that have been amplified and sampled at a specified sampling frequency. In addition, the user terminal 100 can obtain two brain waves obtained from the signal processing unit 220 through different channels.

[0173] On the other hand, the user terminal 100 of an embodiment of the present invention can also obtain brain waves and electromyographic signals at the same time. Of course, the electromyographic signals obtained in this case can also be electromyographic signals that have been amplified and sampled at a specified sampling frequency.

[0174] The user terminal 100 of an embodiment of the present invention can obtain the living body signal of the patient 300 in real time. In this case, the user terminal 100 can be used to measure the anesthesia depth of the patient by temporarily and / or permanently storing the past living body signal in the memory 130.

[0175] The user terminal 100 according to an embodiment of the present invention may generate a first interval of brain waves, where the brain waves include at least a portion of the acquired brain waves (step S930 ).

[0176] Figure 2 4 is a diagram for explaining a method for the user terminal 100 to generate a first section of the electroencephalogram 400 according to an embodiment of the present invention.

[0177] In the following, for the sake of convenience, the electroencephalogram 400 of the patient 300 is shown in the figure, and the current time point is assumed to be 4 seconds, 5 seconds, and 6 seconds respectively according to the situation.

[0178] In the process of generating the first interval of the brain wave 400, the user terminal 100 of an embodiment of the present invention can generate the first interval of the brain wave in a manner that includes past brain waves within a specified time interval starting from the time point of determining the anesthesia depth (ie, the current time point).

[0179] For example, assuming that the current time point for determining the depth of anesthesia is 4 seconds, the user terminal 100 includes a first interval 410 of brain waves of past brain waves within a prescribed time interval (assuming 4 seconds) from the current time point (4 seconds). Similarly, when the current time point is 5 seconds, the user terminal 100 may generate a first interval 420 of brain waves, and when the current time point is 6 seconds, the user terminal 100 may generate a first interval 430 of brain waves.

[0180] In this case, the "prescribed time interval" can be set in various ways according to the system characteristics. For example, in a system that requires a quick response, the prescribed time interval can be set to be relatively short. Also, in a system that requires an accurate response, the prescribed time interval can be set to be relatively long.

[0181] The user terminal 100 of an embodiment of the present invention can generate the first interval through the above process for two brain waves obtained through different channels. In addition, the user terminal 100 of an embodiment of the present invention can repeatedly generate the first interval with the current time point as time goes by (that is, as the time point for determining the depth of anesthesia changes).

[0182] The user terminal 100 according to an embodiment of the present invention may remove noise in the first interval of the brain wave generated according to the above process (step S940 ).

[0183] Figure 3 1 is a diagram for explaining a method in which the user terminal 100 generates a first section 440FLT of an electroencephalogram from which noise has been removed by removing noise in an exemplary first section 440 of an electroencephalogram according to an embodiment of the present invention.

[0184] In the first section 440 of the brain wave, the user terminal 100 of an embodiment of the present invention may replace the first partial section 444A where the brain wave magnitude is greater than a predetermined critical magnitude Ath with a second partial section 443A different from the first partial section.

[0185] At this time, if the size of the brain wave at any time point in the first partial interval 444A is greater than the specified critical size Ath, the user terminal 100 may determine that the size of the brain wave in the corresponding interval 444A is greater than the specified critical size Ath. In addition, the user terminal 100 may replace the corresponding interval 444A with the interval 443A adjacent to the corresponding interval 444A.

[0186] Thus, the user terminal 100 can generate a first interval 440FLT of the brain wave from which noise has been removed, wherein the first partial interval 441B, the second partial interval 442B, and the third partial interval 443B of the first interval 440FLT are the same as the first partial interval 441A, the second partial interval 442A, and the third partial interval 443A of the first interval 440, and the fourth partial interval 444B is the same as the third partial interval 443A of the first interval 440.

[0187] However, the length of the partial interval, the critical size Ath, and the replacement method of the partial interval are merely illustrative, and the concept of the present invention is not limited thereto.

[0188] In an optional embodiment, when the pattern in which the size of the brainwave in the first interval 440 of the brainwave is greater than the specified critical size corresponds to the preset pattern, the user terminal 100 may also replace the first interval 440 of the brainwave with the second interval of the brainwave. In this case, the second interval of the brainwave may be a brainwave in an interval different from the first interval 440 of the brainwave, such as a brainwave in an adjacent interval.

[0189] Furthermore, the "preset mode" can be set in a variety of ways according to the purpose of the system. For example, the preset mode can be a mode in which multiple partial intervals within the same interval generate brain waves with a size greater than a specified critical size, or a mode in which brain waves with a size greater than a specified critical size are generated in two or more consecutive partial intervals. However, the above modes are illustrative only, and the concept of the present invention is not limited thereto.

[0190] The user terminal 100 according to an embodiment of the present invention may extract a plurality of components of one or more frequency bands from the first interval of the brain wave generated according to the above process (or the first interval of the brain wave from which noise has been removed) (step S950 ).

[0191] Figure 4 4 is a diagram for explaining a process in which the user terminal 100 extracts a plurality of components 411 , 412 , 413 , 414 , 415 , 416 , 417 , and 418 of one or more frequency bands from a first interval 410FLT of an electroencephalogram from which noise has been removed according to an embodiment of the present invention.

[0192] The user terminal 100 of one embodiment of the present invention can extract a first component 412 of 0.5 Hz to 4 Hz, a second component 413 of 4 Hz to 8 Hz, a third component 414 of 8 Hz to 16 Hz, a fourth component 415 of 16 Hz to 25 Hz, a fifth component 416 of 25 Hz to 30 Hz, a sixth component 417 of 30 Hz to 48 Hz, a baseline component 411 of 0.5 Hz to 55 Hz, and a burst suppression rate component 418 of 0.5 Hz to 30 Hz from the first interval 410FLT of the brain wave from which noise has been removed.

[0193] The user terminal 100 of one embodiment of the present invention can use frequency filters corresponding to more than one frequency band to extract multiple components 411, 412, 413, 414, 415, 416, 417, 418 of more than one frequency band from the first interval 410FLT of the brain wave from which noise has been removed. For example, the user terminal 100 can use a bandpass filter (Ban-PassFilter) with a passband of 0.5 Hz to 4 Hz to extract the first component 412.

[0194] The extracted multiple components 411, 412, 413, 414, 415, 416, 417, 418 of one or more frequency bands may have the same time length as the first interval 410FLT of the brain wave from which noise has been removed. In addition, the multiple components 411, 412, 413, 414, 415, 416, 417, 418 of one or more frequency bands may include the size of the component of each frequency band at one or more time points belonging to the first interval.

[0195] For example, when the sampling frequency of the signal processing unit 220 is 250 Hz and the length of the first interval 410FLT of the brain wave from which noise has been removed is 4 seconds, the first interval 410FLT of the brain wave from which noise has been removed may include 1000 time points (or 1000 sampling time points) and the size of the brain wave at each time point. Thus, the multiple components 411, 412, 413, 414, 415, 416, 417, 418 of one or more frequency bands may include the size of the component of each frequency band at each of the 1000 time points belonging to the first interval.

[0196] In other words, the plurality of components 411 , 412 , 413 , 414 , 415 , 416 , 417 , 418 of one or more frequency bands may include the magnitude of the component of each frequency band at 1000 time points belonging to the first interval.

[0197] The user terminal 100 of one embodiment of the present invention may calculate the first index of each of the plurality of components 411 , 412 , 413 , 414 , 415 , 416 , 417 , and 418 of more than one frequency band (step S960 ).

[0198] Figure 5 1 is a diagram for illustrating a method for calculating a first index by a user terminal 100 according to an embodiment of the present invention. Figure 5 The graph shown relates to the first component, and the description is given on the assumption that the first interval of the first component includes 15 time points.

[0199] The user terminal 100 of an embodiment of the present invention can calculate the first index based on the extent to which the size of each component of one or more frequency bands of the size of the predetermined reference component in the first interval exceeds the predetermined threshold value Rth. For example, the user terminal 100 of an embodiment of the present invention can calculate the sizes of multiple components of one or more frequency bands of the reference component for one or more time points (i.e., 15 time points) belonging to the first interval.

[0200] In this case, the benchmark composition may refer to Figure 4 The process of extracting the reference component 411 is described. The multiple components of more than one frequency band may be referred to as Figure 4The plurality of components 412 to 417 are extracted by the process described above. For example, the user terminal 100 may calculate the ratio R1 of the size of the reference component 411 at the first time point of the first interval to the size of the first component 412 at the first time point of the first interval. Similarly, the control unit calculates the ratios R2 to R15 of the size of the reference component 411 to the size of the first component 412 at the remaining time points. Of course, for the plurality of remaining components 413 to 417, the user terminal 100 may also calculate the respective ratios at the plurality of time points through the above process.

[0201] The user terminal 100 of an embodiment of the present invention may determine the time point at which the calculated ratio exceeds the specified threshold value Rth as the exceeding time point. Figure 5 The time points corresponding to R4, R12, R13, and R14 are determined as the exceeding time points. Of course, for the remaining components 413 to 417, the control unit can also determine multiple exceeding time points through the above process.

[0202] The user terminal 100 of one embodiment of the present invention can calculate the first index based on the ratio of the number of overall time points belonging to the first interval to the number of exceeded time points. For example, for the first component 412, the user terminal 100 can use the number of overall time points 15 and the number of exceeded time points 4 to calculate the first index as 4 / 15. In this case, the user terminal 100 can normalize the first index by multiplying the calculated first index by the prescribed value. Of course, for the remaining multiple components 413 to 417, the control unit can also calculate the first index through the above process respectively.

[0203] On the other hand, the user terminal 100 of an embodiment of the present invention can determine the prescribed critical value Rth based on the absolute size of the reference component in the first interval. In the case of the reference component, the reference component is extracted in a wide frequency band, so when the size of the brain wave in the first interval is large, the prescribed critical value Rth can be determined to be relatively high, and when the size of the brain wave is small, the prescribed critical value Rth can be determined to be relatively low.

[0204] For two brain waves acquired through different channels, the user terminal 100 of an embodiment of the present invention can respectively calculate the first index based on the above process. In addition, the user terminal 100 of an embodiment of the present invention can repeatedly calculate the updated first index of the first interval according to the update of the first interval.

[0205] The user terminal 100 of one embodiment of the present invention can generate input data 531, 532, 533, 534 of the artificial neural network 520, so as to use the artificial neural network 520 learned through the above process to calculate the probability value that the patient 300 belongs to more than one patient state (step S970).

[0206] Figure 8 1 is a diagram for explaining patient status data 540 as output data of input data 531 , 532 , 533 , and 534 of an artificial neural network 520 according to an embodiment of the present invention.

[0207] The user terminal 100 passes Figures 2 to 5 The described process calculates the first index of each of the N frequency bands from the first interval of the first brain wave of the patient 300 obtained through the first channel. The user terminal 100 calculates the first index of each of the N frequency bands from the first interval of the second brain wave of the patient 300 obtained through the second channel through the same process. For the sake of convenience, the following description will be based on the above content.

[0208] Under the above premise, the user terminal 100 of an embodiment of the present invention can generate N square first input data 533 based on the combination of N first indices of the first channel and N first indices of the second channel. The above first input data 533 can be the learning data used for learning the artificial neural network 520. Figure 7 The item corresponding to the data 511D.

[0209] For example, the user terminal 100 may generate the first input data 533 to include data obtained by multiplying the first index of the first channel by the N indexes of the second channel and data obtained by multiplying the second index of the first channel by the N indexes of the second channel. Of course, for the remaining indexes of the first channel, the user terminal 100 may generate the first input data 533 in the same manner as described above.

[0210] The user terminal 100 of an embodiment of the present invention may generate N second input data 532 corresponding to the N first indexes of the first channel. The second input data 532 may be the learning data used for learning the artificial neural network 520. Figure 7 For example, the user terminal 100 may generate N second input data 532 so that the N first indexes of the first channel and the N second input data 532 are respectively the same value.

[0211] The user terminal 100 of an embodiment of the present invention may generate N third input data 534 corresponding to the N first indexes of the second channel. The third input data 534 may be the learning data used for learning the artificial neural network 520. Figure 7 For example, the user terminal 100 may generate N third input data 534 so that the N first indexes of the second channel and the N third input data 534 are respectively the same value.

[0212] The user terminal 100 of one embodiment of the present invention may generate M fourth input data 531 based on the electromyography signal. The fourth input data 531 may be the learning data used for learning the artificial neural network 520. Figure 7 For example, the user terminal 100 of one embodiment of the present invention may generate the fourth input data 531 using the third index generated according to the above process.

[0213] In one embodiment of the present invention, N may be 6, and M may be 1. Therefore, the first input data 533 includes 36 data, the second input data 532 and the third input data 534 each include 6 data, and the fourth input data 531 may include 1 data. However, as described above, the number of data included in each of the data 531 to 534 is exemplary, and the concept of the present invention is not limited thereto.

[0214] The user terminal 100 of one embodiment of the present invention can obtain the patient's status data 540 by inputting the input data 531, 532, 533, and 534 generated by the artificial neural network 520 according to the above process. The above patient's status data 540 includes the probability value that the patient 300 belongs to more than one patient status (step S980).

[0215] The patient status data 540 of one embodiment of the present invention may include probability values ​​of K (K is a natural number) patient statuses corresponding to the patient 300. For example, K is 5, and the one or more patient statuses may include awake status (Awake) Status 1, sedation status (Sedation) Status 2, general anesthesia status (General Anesthesia) Status 3, hyper anesthesia or deep anesthesia status (Hyper or Deep Anesthesia) Status 4, and brain death status (Brain Death) Status 5.

[0216] For example, the user terminal 100 may obtain the patient's state data 540 from the artificial neural network 520, such as [0.81, 0.62, 0.34, 0.17, 0.01]. The above patient's state data 540 may mean that the probability that the patient 300 is in an awake state (Awake) is 81%, the probability that the patient 300 is in a sedation state (Sedation) is 62%, the probability that the patient 300 is in a general anesthesia state (General Anesthesia) is 34%, the probability that the patient 300 is in a hyperanesthesia or general anesthesia state (Hyper or Deep Anesthesia) is 17%, and the probability that the patient 300 is in a brain death state (Brain Death) is 1%.

[0217] In another embodiment of the present invention, the K states and / or the number may be set in a variety of ways. For example, when the present invention is used to determine the emotional state of a patient, the K states may be multiple states representing the patient's emotions. However, this is for illustrative purposes only, and the concept of the present invention is not limited thereto.

[0218] The user terminal 100 according to an embodiment of the present invention may determine the anesthesia depth of the patient 300 based on the probability values ​​of each of more than one patient status (step S990 ).

[0219] More specifically, the user terminal 100 of an embodiment of the present invention can calculate the normalized probability value by normalizing the probability values ​​of each of the more than one patient states included in the patient state data 540. For example, the user terminal 100 can calculate the normalized probability value of each of the more than one patient states in such a way that the sum of the probability values ​​of each of the more than one patient states is 1.

[0220] Furthermore, the user terminal 100 according to an embodiment of the present invention may apply a weight set corresponding to the patient state to the normalized probability value, wherein the weight set has the largest probability value among the normalized probability values.

[0221] Fig. 9 The figure shows one or more exemplary weight sets 610, 620, 630, 640, 650 in the form of a curve graph. The first weight set 610 is a weight set corresponding to the awake state (Awake) Status 1, the second weight set 620 is a weight set corresponding to the sedation state (Sedation) Status 2, the third weight set 630 is a weight set corresponding to the general anesthesia state (General Anesthesia) Status 3, the fourth weight set 640 is a weight set corresponding to the hyperanesthesia or deep anesthesia state (Hyper or Deep Anesthesia) Status 4, and the fifth weight set 650 is a weight set corresponding to the brain death state (Brain Death) Status 5.

[0222] For example, when the normalized probability value calculated according to the above process is [0.45, 0.25, 0.15, 0.1, 0.05], the probability that the patient 300 is in the awake state (Awake) Status 1 is the highest. Therefore, the user terminal 100 can calculate the probability such as [0.45, 0.2, 0.5, 0.01, 0] by applying the weight set 610 corresponding to the awake state (Awake) Status 1 to the normalized probability value.

[0223] The user terminal 100 according to an embodiment of the present invention may determine the anesthesia depth of the patient 300 based on the sum of a plurality of normalized probability values ​​to which weighted values ​​are applied.

[0224] In an optional embodiment, the user terminal 100 of an embodiment of the present invention may determine the final anesthesia depth of the patient 300 by referring to the anesthesia depth of the patient 300 determined by other means.

[0225] Fig.10 1 is a diagram for explaining a process in which the user terminal 100 determines a final anesthesia depth of the patient 300 according to an alternative embodiment of the present invention.

[0226] The user terminal 100 of the optional embodiment of the present invention can determine the final anesthesia depth DOA_F of the patient 300 based on the sum of the first anesthesia depth and the second anesthesia depth. The first anesthesia depth specifies the weighted value W1 to be applied according to Figures 2 to 9 The second method for calculating the anesthesia depth DOA_2 may refer to at least a portion of the process and the second method for calculating the anesthesia depth DOA_2. Figures 2 to 9 The described process is different from the method for calculating the depth of anesthesia. For example, the second depth of anesthesia can be determined based on a first index, a second index, and a third index, wherein the first index is calculated from the patient's brain wave, the second index reflects the degree to which the patient's brain wave conforms to a prescribed pattern, and the third index is based on at least one living body signal corresponding to the patient's state.

[0227] The embodiments of the present invention described above can be implemented in the form of a computer program that can be executed by a variety of structural elements on a computer, and the above-mentioned computer program can be recorded on a computer-readable medium. In this case, the medium can store a program that can be executed by a computer. As an example of a medium, it can be a medium configured to store program instructions, including magnetic media such as hard disks, floppy disks and tapes, optical recording media such as compact disk read-only memories (CD-ROMs) and digital versatile discs (DVDs), magneto-optical media such as floppy disks (magneto-optical mediums), read-only memories (ROMs), random access memories (RAMs), flash memories, etc.

[0228] On the other hand, the computer program may be specially designed and configured for the present invention, or may be a program known and usable by a person skilled in the art in the field of computer software. Examples of computer programs include not only machine language codes generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like.

[0229] The specific implementation described in the present invention is an example and does not limit the scope of the present invention in any way. For the sake of brevity of the specification, the description of existing electronic configurations, control systems, software, and other functional aspects of the above-mentioned systems may be omitted. In addition, the connection of lines or connecting parts between the structural elements shown in the drawings illustratively represents functional connections and / or physical connections or circuit connections, and in actual devices, various functional connections, physical connections, or circuit connections that can be replaced or added are represented. In addition, if there is no specific mention of "essential", "important", etc., it may not be a structural element necessary for the application of the present invention.

[0230] Therefore, the concept of the present invention is not limited to the above embodiments, and the attached invention claims and all scopes equivalent to or equivalently modified to the invention claims belong to the concept of the present invention.

Claims

1. A method for determining the level of consciousness of a patient, characterized in that: include: An extraction step of extracting a plurality of components of one or more frequency bands from a first interval of the brain wave; A first index calculation step of calculating a first index for each component of the one or more frequency bands, the first index being calculated based on the extent to which the size of each component of the one or more frequency bands exceeds a prescribed threshold value relative to a prescribed reference component size in the first interval; A probability value calculation step, using a learned artificial neural network to calculate the probability value of each of the more than one patient state according to the first index of each component of the more than one frequency band; as well as a consciousness level determination step, determining the consciousness level of the patient based on the calculated probability values ​​of each of the one or more patient states, The plurality of components of the one or more frequency bands include the magnitude of the component of each frequency band at one or more time points belonging to the first interval, The first index calculation step comprises the following steps: calculating, for each of the one or more time points, the magnitudes of the plurality of components of the one or more frequency bands relative to the magnitude of the reference component; Determining a time point at which the calculated size exceeds the prescribed critical value as a exceeding time point; as well as The first index is calculated based on the ratio of the number of the overall time points belonging to the first interval to the number of the exceeding time points.

2. The method for determining the level of consciousness according to claim 1, characterized in that: Before the extraction step, the method further comprises: A brain wave acquisition step, acquiring the brain waves of the patient; a step of generating a first interval of brain waves, wherein the first interval of brain waves includes at least a portion of the acquired brain waves; and A noise removal step of removing noise in the first interval of the brain wave; In the extraction step, a plurality of components of one or more frequency bands are extracted from a first section of the brain wave from which the noise has been removed.

3. The method for determining the level of consciousness according to claim 2, wherein: In the brain wave acquisition step, the brain waves of the patient sampled at a prescribed sampling frequency are acquired.

4. The method for determining the level of consciousness according to claim 2, wherein: In the step of generating the first electroencephalogram interval, the first electroencephalogram interval is generated so as to include electroencephalograms from the time point when the consciousness level is determined to within a predetermined time interval.

5. The method for determining the level of consciousness according to claim 2, wherein: The noise removal step includes the following replacement step: in a first interval of the brain wave, a first partial interval is replaced with a second partial interval different from the first partial interval, the first partial interval includes a time point when the size of the brain wave exceeds a specified critical size, The first partial interval and the second partial interval are at least a part of the first interval.

6. The method for determining the level of consciousness according to claim 2, wherein: The noise removal step includes the following replacement step: when the pattern in which the size of the brain wave in the first interval of the brain wave exceeds the specified critical size corresponds to a preset pattern, replacing the first interval of the brain wave with the second interval of the brain wave, The second interval is different from the first interval and is at least a part of the brain wave.

7. The method for determining the level of consciousness according to claim 1, characterized in that: In the extraction step, a first component of [0.5Hz, 4Hz), a second component of [4Hz, 8Hz), a third component of [8Hz, 16Hz), a fourth component of [16Hz, 25Hz), a fifth component of [25Hz, 30Hz), a sixth component of [30Hz, 48Hz] and a reference component of [0.5Hz, 55Hz] are extracted from the first interval of the brain wave.

8. The method for determining the level of consciousness according to claim 1, wherein: The prescribed threshold value is determined based on the absolute magnitude of the reference component within the first interval.

9. The method for determining the level of consciousness according to claim 1, wherein: In the consciousness level determination method, an input data generating step is further included before the probability value calculating step, in which input data of the artificial neural network is generated using a combination of first exponents of each component of the one or more frequency bands.

10. The method for determining the level of consciousness according to claim 9, characterized in that: In the method for determining the level of consciousness, From the first interval of the first brain wave of the patient obtained through the first channel, calculate the first index of each of N frequency bands, where N is a natural number, calculating the first index of each of the N frequency bands from a first interval of the second brain wave of the patient acquired through a second channel, the second channel being different from the first channel, The input data generation step comprises the following steps: Generate N squared first input data based on a combination of the N first indices of the first channel and the N first indices of the second channel; Generate N second input data corresponding to the N first indexes of the first channel; Generate N third input data corresponding to the N first indexes of the second channel; generating M fourth input data based on the electromyography signal, wherein M is a natural number; and The input data including the first input data, the second input data, the third input data, and the fourth input data is generated.

11. The method for determining the level of consciousness according to claim 10, characterized in that: The N is 6, and the M is 1.

12. The method for determining the level of consciousness according to claim 1, wherein: The artificial neural network is a neural network that learns the relationship between the characteristics of brain waves, electromyographic characteristics, and the patient's state based on learning data, wherein the learning data includes data reflecting the characteristics of brain waves and data reflecting the electromyographic characteristics, and data of the patient's state corresponding to the characteristics of the brain waves and the electromyographic characteristics are marked, The data reflecting the characteristics of brain waves include: N squared first data are generated based on a combination of N first indices of the brain wave obtained through the first channel and N first indices of the brain wave obtained through the second channel, where N is a natural number; N second data corresponding to the N first indexes of the first channel; N third data corresponding to the N first indexes of the second channel; and M fourth data based on the electromyography signal, where M is a natural number, The patient's status data includes probability values ​​corresponding to K patient states, where K is a natural number.

13. The method for determining the level of consciousness according to claim 1, characterized in that: The consciousness level determination step comprises: a normalized probability value calculation step of normalizing the probability values ​​of the more than one patient states to calculate a normalized probability value; an applying step of applying a weight set corresponding to a patient state to the normalized probability values, wherein the patient state has a maximum probability value among the normalized probability values; and The step of determining said level of consciousness based on a sum of a plurality of said normalized probability values ​​to which a weighting value is applied.

14. The method for determining the level of consciousness according to claim 13, wherein: The one or more patient states include awake state, sedation state, general anesthesia state, deep anesthesia state and brain death state, In the applying step, one of the weight sets for each of the five patient states is applied to the normalized probability value.

15. The method for determining the level of consciousness according to claim 13, characterized in that: In the step of determining the consciousness level based on the sum of multiple normalized probability values ​​with weighted values, the consciousness level is determined based on a first consciousness level and a consciousness level with a second weighted value applied to the second consciousness level, the first consciousness level applies a specified first weighted value to the sum of multiple normalized probability values ​​with weighted values, and the second consciousness level is determined in a second manner.

Images