Method and device for detecting abnormal biological signals

Abstract

To generate a model for detecting biological signals and to detect abnormal biological signals using the model with a monitor of the biological signal.SOLUTION: A device 1 for detecting abnormal biological signals can communicate with a sensor 2 and one or more external control devices 31, 32, 33, 34. The device 1 receives input data collected by the sensor 2. The sensor 2 emits radio waves toward a human body being the monitoring object, receives radio waves reflected from the surface of the object, and thereby collects characteristics relating to the shape or movement of the object. Thus, the sensor 2 collects changes in the thorax caused by respiration or heartbeat of the human body being the monitoring object. The obtained data is input to a neural network learned on the basis of predetermined learning data in a neuromorphic device that implements the neural network for detecting abnormal biological signals, thereby detecting abnormal biological signals.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present invention relates to a method and apparatus for detecting abnormal biological signals, and more particularly to a method and apparatus for generating a model for detecting abnormal biological signals, and a method and apparatus for detecting abnormal biological signals by monitoring the biological signals using the generated model. [Background technology]

[0002] Artificial neural networks mimic biological neural networks, but they are trained by a large amount of input data and are used to estimate or approximate results that are difficult to derive using conventional techniques. Artificial neural networks contain layers of interconnected neurons that exchange signals, and synapses have weights that can be tuned based on learning or experience. To train an artificial neural network, good quality training data is required, and the performance of the artificial neural network can vary depending on the training data.

[0003] Meanwhile, heart disease accounts for a high percentage of the leading causes of death both domestically and internationally. In particular, unintentional sudden death due to suffocation while sleeping occurs most frequently in infants aged 2 to 4 months. In addition, as the rate of lonely deaths gradually increases, there is a growing need to reduce social costs through prevention and management of lonely deaths.

[0004] The above-mentioned background art is technical information that the inventor possessed in order to derive the present invention or that he acquired in the process of deriving the present invention, and is not necessarily publicly known art that was disclosed to the general public prior to the filing of the present invention. Summary of the Invention [Problem to be solved by the invention]

[0005] An object of the present disclosure is to provide a method and an apparatus for detecting abnormal biological signals. The problems to be solved by the present disclosure are not limited to the technical problems described above, and other technical problems not described will be clearly understood by those having ordinary knowledge in the art from the description of the present invention, and will be further clearly understood from the embodiments of the present disclosure. It will also be understood that the problems and advantages to be solved by the present disclosure can be realized by the means and combinations thereof shown in the claims.

Means for Solving the Problems

[0006] As means for solving the above-described technical problems, a first aspect of the present disclosure includes a step of randomly extracting a frame having a first data length from raw data including biological signal data to generate a plurality of sample data, a step of sequentially extracting a frame having a second data length from the sample data in chronological order to generate a plurality of unit data, a step of generating a learning data set based on the plurality of unit data, and a step of generating an abnormal biological signal detection model by training an artificial intelligence model with the learning data set. A method for detecting abnormal biological signals can be provided.

[0007] A second aspect of the present disclosure includes a memory storing at least one program, and a processor that performs calculations by executing the at least one program. The processor randomly extracts a frame having a first data length from raw data including biological signal data to generate a plurality of sample data, sequentially extracts a frame having a second data length from the sample data in chronological order to generate a plurality of unit data, generates a learning data set based on the plurality of unit data, and generates an abnormal biological signal detection model by training an artificial intelligence model with the learning data set. An apparatus for detecting abnormal biological signals can be provided.

[0008] A third aspect of the present disclosure can provide a computer-readable recording medium recording a program for executing the method of the first aspect by a computer.

[0009] A fourth aspect of the present disclosure provides a neuromorphic device that realizes a neural network for detecting abnormal biological signals, including at least one processor that drives the neural network, and an on-chip memory including a crossbar array circuit that receives instructions from the at least one processor and performs in-memory operations. The at least one processor receives input data collected by a sensor, inputs the input data into the neural network trained based on predetermined training data to detect the abnormal biological signal. The input data includes a biological signal and a noise signal. The neural network includes an input layer into which the input data is input, one or more hidden layers that extract features of the input data, and an output layer that outputs an output signal.

[0010] A fifth aspect of the present disclosure provides a method for detecting abnormal biological signals using a neural network, including receiving input data collected by a sensor, and inputting the input data into the neural network trained based on predetermined training data to detect the abnormal biological signal. The input data includes a biological signal and a noise signal. The neural network includes an input layer into which the input data is input, one or more hidden layers that extract features of the input data, and an output layer that outputs the abnormal biological signal.

[0011] A sixth aspect of the present disclosure can provide a computer-readable recording medium recording a program for executing the method of the fifth aspect by a computer.

[0012] In addition to those, other methods for implementing the present invention, other devices, and a computer-readable recording medium recording a program for executing the method can be further provided.

[0013] Other aspects, features, and advantages will become apparent from the accompanying drawings, the claims, and the following detailed description of the invention.

Advantages of the Invention

[0014] According to the means for solving the problems of the present disclosure described above, it is possible to prevent and manage unexpected sudden death or lonely death due to suffocation accidents during the sleep of infants, and to reduce social costs.

[0015] Further, according to the means for solving the problems of the present disclosure, misrecognition of abnormal biological signals that greatly increase social costs can be minimized, and a highly accurate machine learning model can be constructed.

[0016] The effects according to the embodiments are not limited to the effects described above, and other effects not described will be clearly understood by those having ordinary knowledge in the art from the description of the present invention.

Brief Description of the Drawings

[0017] [Figure 1] It is a diagram for explaining the architecture of a fully-connected neural network (FCNN) according to an embodiment of the present invention. [Figure 2] It is an exemplary diagram for comparing the von Neumann structure and the CIM (Computing-In-Memory) structure according to an embodiment of the present invention. [Figure 3] It is an exemplary diagram for comparing the von Neumann structure and the CIM (Computing-In-Memory) structure according to an embodiment of the present invention. [Figure 4] It is a diagram showing a neural network according to an embodiment of the present invention. [Figure 5] It is an exemplary diagram of an environment for detecting abnormal biological signals according to an embodiment of the present invention. [Figure 6] It is an exemplary diagram of an environment for detecting abnormal biological signals according to an embodiment of the present invention. [Figure 7] This is a diagram for explaining load data according to an embodiment of the present invention. [Figure 8] This is a diagram for explaining a method of generating a plurality of sample data according to an embodiment of the present invention. [Figure 9] This is a diagram for explaining a method of generating a plurality of unit data according to an embodiment of the present invention. [Figure 10] This is a diagram for explaining a method of generating a plurality of unit data according to an embodiment of the present invention. [Figure 11] This is a diagram for explaining a method of generating a learning data set according to an embodiment of the present invention. [Figure 12] This is a diagram for explaining a method of generating a learning data set according to an embodiment of the present invention. [Figure 13] This is a diagram for explaining a method of generating a learning data set according to an embodiment of the present invention. [Figure 14] This is a graph showing the result of training an artificial intelligence model with a learning data set according to an embodiment of the present invention. [Figure 15] This is a block diagram of an apparatus for detecting an abnormal biological signal according to an embodiment of the present invention.

Embodiments for Carrying Out the Invention

[0018] In describing the present invention, if it is determined that a specific description of related known technologies will obscure the gist of the present invention, the detailed description thereof may be omitted, and unless otherwise specifically defined, all terms used in this specification have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention pertains.

[0019] Phrases such as "according to an embodiment", "relating to an embodiment", and "by the implementation of an embodiment" in this specification do not necessarily all indicate the same embodiment.

[0020] In the embodiments, various modifications can be made and there can be various forms. Therefore, some embodiments are illustrated in the drawings and described in detail. However, this does not limit the embodiments to a specific form of disclosure, and it should be understood to include all modifications, equivalents, and alternatives included in the spirit and technical scope of the embodiments. The terms used in the specification are merely for the purpose of explaining the embodiments and are not intended to limit the embodiments.

[0021] The terms used in the embodiments are selected to be as general as currently widely used in consideration of the functions in this embodiment. However, they may vary depending on the intentions of those skilled in the technical field to which the embodiments belong, precedents, the emergence of new technologies, etc. Also, in certain cases, there are terms arbitrarily selected by the applicant, and in such cases, the meaning will be described in detail in the corresponding part. Therefore, the terms used in the embodiments should be defined based not only on the name of the terms but also on the meaning of the terms and the content throughout the embodiments.

[0022] Some embodiments of the present disclosure can be shown in a functional block configuration and various processing steps. Some or all of such functional blocks can be realized by various numbers of hardware and / or software configurations that execute specific functions. For example, the functional blocks of the present disclosure can be realized by one or more microprocessors or by a circuit configuration for a predetermined function.

[0023] Also, for example, the functional blocks of the present disclosure can be realized in various programming or scripting languages. The functional blocks can be realized by an algorithm executed by one or more processors. Furthermore, the present disclosure can adopt conventional technologies for electronic environment setting, signal processing, and / or data processing, etc.

[0024] Terms such as "database," "element," "means," and "configuration" can be used broadly and are not limited to mechanical and physical configurations. Furthermore, terms such as "unit" and "module" used in the specification refer to a unit that processes at least one function or operation, and can be realized by hardware or software, or a combination of hardware and software.

[0025] Note that the connecting lines or connecting members between components shown in the drawings are merely exemplary of functional and / or physical connections or circuit connections, and in an actual device, connections between components may be represented by various alternative or additional functional, physical, or circuit connections.

[0026] Furthermore, terms including ordinal numbers such as "first" and "second" used in this specification may be used to describe various components, but the components should not be limited by the terms. The terms are used only to distinguish one component from another.

[0027] In addition, the size and proportions of some components in the drawings may be somewhat exaggerated. Furthermore, components shown in one drawing may not be shown in other drawings.

[0028] Throughout the specification, the term "embodiment" is an arbitrary category for easily describing the invention in the present disclosure, and the respective embodiments do not necessarily have to be mutually exclusive. For example, a configuration disclosed in one embodiment can be applied and / or realized in other embodiments, and can be applied and / or realized with modifications without departing from the scope of the present disclosure.

[0029] Furthermore, the terms used in this disclosure are for the purpose of describing the embodiments and are not intended to limit the embodiments. In this disclosure, the singular includes the plural unless otherwise specified.

[0030] Hereinafter, with reference to the accompanying drawings, embodiments of the present disclosure will be described in detail so that those having ordinary knowledge in the art can easily implement them. However, the embodiments of the present disclosure can be realized in various different forms and are not limited to the embodiments described in the present disclosure.

[0031] Hereinafter, based on this, the present invention will be described in detail with reference to the drawings.

[0032] A neuromorphic chip is hardware that generates a circuit mimicking the form of neurons to imitate human brain functions. That is, a neuromorphic chip means a computer chip that mimics the structure of the nervous system.

[0033] Since a neuromorphic chip is composed only of the circuits necessary for neural network operations, gains of several hundred times or more can be obtained in terms of power, area, and speed. Different from conventional computers, the human brain does not consume much power even when processing a large amount of data. Therefore, a neuromorphic chip mimics the operation mode of such a brain, constructs a structure connecting neurons and synapses in parallel, and can save energy by disconnecting the connection when data is not being processed.

[0034] For example, in the von Neumann architecture, which is a conventional computer, when data is input, the data is processed sequentially. Therefore, it is excellent for executing a precisely created program, but there are problems such as limitations in power consumption and low efficiency in pattern recognition, real-time recognition, etc.

[0035] In contrast, a neuromorphic chip uses an analog operation in which data gradually changes in various states instead of being digital such as 0 and 1. That is, the artificial neurons configured in parallel operate in an event-driven manner without a clock operation. Therefore, it is possible to efficiently process non-regular characters, voices, images, etc. that are difficult for conventional computers to intuitively recognize. Specifically, by dispersing neurons, which are nerve cells, and synapses, which are connection lines, with silicon transistor circuits and memory elements, data can be processed in parallel.

[0036] In one embodiment, when input data such as an image, voice, or electromagnetic wave is input to a neuromorphic chip, predetermined output data can be output by performing an operation inside the neuromorphic chip on the input data. The predetermined output data may include the recognition result of voice / image / electromagnetic wave by feature classification of the input data. For example, when an electromagnetic wave signal is input as input data, the predetermined output data may include the recognition result of the electromagnetic wave that classifies whether the input data contains an abnormal signal learned in advance.

[0037] On the other hand, the data input to the neuromorphic chip is not limited to the aforementioned image, voice, or electromagnetic wave, and may include various forms of data such as video and text.

[0038] The neuromorphic device according to one embodiment can be implemented on an edge AI chip. Edge AI refers to a technology that executes an AI algorithm on a hardware device using edge computing based on data generated in the system. Since AI processing is mainly performed in a cloud-based data center that requires a huge computing capacity, the dependence on the server is high. In contrast, when edge AI is used, the execution of the AI algorithm operation is performed locally, so the dependence on the cloud (server) is reduced, the communication cost is thereby reduced, and sensitive personal information is not transmitted to the cloud, so privacy is protected. Therefore, by configuring the neuromorphic device with an edge AI chip, not only can the cost be reduced and the security be improved, but also a highly responsive system can be realized because the execution of the operation is immediately processed within the same hardware.

[0039] In the following description, the device for detecting abnormal biological signals according to one embodiment of the present invention may be the aforementioned neuromorphic device. That is, the aforementioned neuromorphic device can function as a device for detecting abnormal biological signals according to one embodiment of the present invention.

[0040] FIG. 1 is a diagram for explaining the architecture of a fully-connected neural network (FCNN) according to one embodiment of the present invention.

[0041] Referring to FIG. 1, the FCNN, which is a fully-connected layer, is in a state where all neurons in one layer are connected to all neurons in the next layer, is included in a convolutional neural network, and is a layer used to classify data by a matrix flattened in the form of a one-dimensional array.

[0042] In an FCNN, there is a value for each node, and there are weights and biases for each neuron. When moving from one layer to another, the value obtained by multiplying the weight of each node and adding the bias becomes the node value of the next layer. Here, if the calculated value satisfies a specific condition, the output value passed through the activation function is input as the node value of the next layer for activation, and if the calculated value does not satisfy the specific condition, an activation function that deactivates the node after passing through the activation function intervenes.

[0043] Therefore, since the output value differs depending on the type of activation function, it is important to use an appropriate activation function as needed. Typically, there are the ReLU function and the Softmax function.

[0044] Since the input of the FCNN can only be data in a one-dimensional array, when the input data is an image in a three-dimensional array consisting of vertical, horizontal, and channels (colors), there is a drawback that it must be flattened into one-dimensional data and input to the FCNN. That is, there is a drawback that the spatial information of the image is ignored and the features incorporated in the shape cannot be extracted. Therefore, in the case of the FCNN, it is highly useful when inputting data that can be realized with one-dimensional array data such as electromagnetic wave data. According to one embodiment, the neural network described later may be an FCNN.

[0045] FIG. 2 and FIG. 3 are exemplary diagrams for comparing the von Neumann structure and the CIM (Computing-In-Memory) structure according to an embodiment of the present invention.

[0046] Referring to FIG. 2, the von Neumann structure is a computer structure presented by John von Neumann, and is a program-embedded computer structure consisting of a typical three-stage structure of a main memory device, a central processing unit, and an input / output device.

[0047] The von Neumann architecture has the advantage that when changing to other operations in a computing device, there is no need to rearrange the hardware (such as wires), and only the software (program) needs to be changed, thus greatly improving the versatility. However, since the arranged instructions are sequentially executed and the instructions are composed of operations that change the values in predetermined storage locations, it causes serious problems in the design of high-speed computers. This is called the von-Neumann bottleneck phenomenon.

[0048] To solve the von-Neumann bottleneck phenomenon, alternatives such as the Harvard architecture that divides the memory into a part where instruction words are stored and a part where data is stored, or the CIM architecture that not only stores data in the memory but also performs data operations, and a neuromorphic computing that mimics the brain structure of higher animals, which consists of many units integrating computing and storage functions connected in a network in parallel and then operating each unit in an event-driven manner, have been proposed.

[0049] Referring to Figure 3, it can be seen that the CIM structure is composed of a processor and a memory with computing functions.

[0050] In contrast to the conventional von Neumann architecture where all data inside the memory moves to the processor for calculation, in the CIM architecture, when the instruction words of the processor are transmitted, calculations are performed in the memory and only the result data is sent to the processor, so there is no movement of a large amount of data, and thus the aforementioned von-Neumann bottleneck phenomenon can be effectively solved. In addition, it has the advantage of significantly reducing power consumption.

[0051] A neural network according to an embodiment of the present invention can perform operations using only on-chip memory without using external memory. For example, the neural network can perform operations for each layer on a CIM basis using only on-chip memory without using external memory (e.g., off-chip memory), thereby performing operations without memory updates while processing input signals. Specifically, the neural network can perform CIM-based operations in which each memory cell is directly connected to a processor.

[0052] However, compared to general memory such as RAM (Random Access Memory), CIM may have limitations in memory capacity and power consumption of computing devices. Therefore, producing a high-performance CIM-structured chip may require a new hardware structure, which may increase manufacturing costs. Therefore, a CIM structure may be advantageous for performing relatively small or simple calculations. In contrast, a neural network according to an embodiment of the present invention may be configured with a memory capable of implementing multi-bit data to overcome the drawbacks of such a CIM structure. For example, a neural network may be configured with a memory capable of implementing 7 bits and 128 analog memory states. By configuring a neural network with a large capacity, it can process large amounts of data with low power consumption and high performance even over long periods of use, unlike general CIM chips that exhibit problems such as heat generation and performance degradation.

[0053] Meanwhile, on-chip memory can be realized by a crossbar array circuit. That is, the crossbar array can receive instructions from a processor and perform calculations, and CIM calculations can be achieved by integrating memory elements in a crossbar array structure into the on-chip memory. As an example, a processor can detect abnormal biological signals by receiving input data collected by a sensor, driving a neural network circuit trained based on predetermined training data, and receiving abnormal biological signals output by the neural network circuit. Specifically, the processor can detect abnormal biological signals by inputting input data into the neural network. Below, the configuration of the neural network and the training data for generating an abnormal biological signal detection model implemented by the neural network are described in detail.

[0054] FIG. 4 is a diagram illustrating a neural network according to one embodiment of the present invention.

[0055] 4, in one embodiment, the neural network may include an input layer 410 to which input data 400 is input, one or more hidden layers 420, 430, 440 that extract features of the input data 400, and an output layer 450 that outputs an output signal 460. Here, the input layer 410, the one or more hidden layers 420, 430, 440, and the output layer 450 may refer to each layer of the fully connected neural network (FCNN) described above.

[0056] In one embodiment, the neural network can classify the input data 400 into one of a normal biological signal, an abnormal biological signal, and noise. Specifically, the neural network may include an input layer 410 for classifying the input data 400, one or more hidden layers 420, 430, 440, and an output layer 450.

[0057] In one embodiment, input data 400 including a biological signal and a noise signal may be input to the input layer 410 of the neural network. The biological signal is a signal generated by the movement of the human body, and the noise signal may include simple movements of the monitoring target that are not the biological signals to be monitored (for example, breathing or heartbeat of the heart), such as turning over, or movements of targets other than the monitoring targets such as surrounding people or companion animals.

[0058] In one embodiment, the hidden layers 420, 430, 440 of the neural network can extract or classify the features of the input data by reflecting the weights on the input data 400. Therefore, the larger the number of hidden layers 420, 430, 440, the more precisely the input data 400 can be processed and the function of the algorithm can be improved.

[0059] In one embodiment, the output layer 450 of the neural network can output which of the normal biological signal, abnormal biological signal, and noise the input data 400 corresponds to from the features extracted by at least one of the hidden layers 420, 430, 440. That is, the output layer 450 can output, as an output signal, which of the normal biological signal, abnormal biological signal, and noise the neural network classifies the input data 400 into.

[0060] FIG. 5 and FIG. 6 are exemplary diagrams of an environment for detecting an abnormal biological signal according to an embodiment of the present invention.

[0061] Referring to FIG. 5, a device 1 for detecting an abnormal biological signal (hereinafter referred to as "device") can communicate with a sensor 2 and one or more external control devices 31, 32, 33, 34.

[0062] As an example, the device 1 can receive the input data collected by the sensor 2.

[0063] Further, when an abnormal biological signal is detected, the device 1 can transmit an emergency signal to an external control device including at least one of a control center 31, a government agency 32, an external terminal 33, and an external server 34.

[0064] In one embodiment, the biological signal and the noise signal included in the input data can be collected by the sensor 2.

[0065] The change in the chest due to human breathing is at the 4 - 12 mm level, and the change in the chest due to the heartbeat is at the 0.6 - 2 mm level. Thus, in one embodiment, the sensor 2 may be a millimeter - wave radar (mmWave Radar) whose radio - wave wavelength band is in the millimeter (mm) band or whose radio - wave frequency band corresponds to 24 GHz - 77 GHz. However, the radio - wave wavelength band and frequency band of the sensor 2 are not limited to the above - described embodiment, and the type of radar is not limited to the millimeter - wave radar.

[0066] Here, the sensor 2 can collect characteristics regarding the form or movement of the target by emitting radio waves to the human body to be monitored and receiving the radio waves reflected back from the surface of the target. Thus, the sensor 2 can collect the change in the chest due to the breathing or heartbeat of the human body to be monitored.

[0067] Referring to FIG. 6, in one embodiment, the apparatus 1 may include a communication unit (not shown) that uses at least one communication protocol of Serial Peripheral Interface (SPI) 12, Digital Video Port (DVP), Parallel Serial Peripheral Interface (PSPI), Inter-Integrated Circuit (I2C), General Purpose Input / Output (GPIO), Pulse Width Modulation (PWM), Inter-IC Sound (I2S), and Universal Asynchronous Receiver / Transmitter (UART) 13.

[0068] The Serial Peripheral Interface 12 is a synchronous serial communication protocol for communication between a plurality of peripheral devices and a microcontroller or a microprocessor. The Serial Peripheral Interface 12 can be used to exchange data between a single apparatus 1 according to one embodiment and a plurality of slave devices such as the sensor 2 and the external memory 21. The Universal Asynchronous Receiver / Transmitter 13 is a form of serial communication protocol and is used to transmit data asynchronously. The Universal Asynchronous Receiver / Transmitter 13 can be used for serial communication between the apparatus 1 according to one embodiment and a peripheral device such as the external terminal 33.

[0069] As an example, the apparatus 1 can receive a biological signal and a noise signal related to the movement of the chest of the human body 600 collected by the sensor 2 by communicating with the sensor 2 using the Serial Peripheral Interface 12.

[0070] Also, the apparatus 1 can transmit and receive data with the external memory 21 by communicating with the external memory 21 using the Serial Peripheral Interface 12. Here, the external memory 21 can store at least one of a normal biological signal, an abnormal biological signal, and noise of the human body 600 detected by the apparatus 1.

[0071] However, when the device 1 communicates with the sensor 2 and / or the external memory 21, in addition to the serial peripheral interface 12, at least one communication protocol of DVP (Digital Video Port), PSPI (Parallel Serial Peripheral Interface), universal asynchronous receiver / transmitter (UART), I2C (Inter-Integrated Circuit), GPIO (General Purpose Input / Output), and PWM (Pulse Width Modulation), or a communication device can be used. That is, the device 1 may include a communication unit (not shown) that uses at least one communication protocol of DVP (Digital Video Port), PSPI (Parallel Serial Peripheral Interface), universal asynchronous receiver / transmitter (UART), I2C (Inter-Integrated Circuit), GPIO (General Purpose Input / Output), and PWM (Pulse Width Modulation).

[0072] In one embodiment, the device 1 can transmit an emergency signal to an external terminal 33 or the like using the universal asynchronous receiver / transmitter 13. Here, the external terminal 33 is only one embodiment, and it goes without saying that an emergency signal can be transmitted to the aforementioned control center 31, government office 32, and external server 34. The external terminal 33 may be a preset terminal of a guardian. The emergency signal can be transmitted to the external terminal 33 in the form of text, image, voice, notification, or the like.

[0073] However, when the device 1 communicates with the external terminal 33, in addition to the general-purpose asynchronous transceiver 13, at least one of the peripheral interfaces of GPIO (General Purpose Input / Output), I2C (Inter-Integrated Circuit), I2S (Inter-IC Sound), and PWM (Pulse Width Modulation) can also be used. That is, the device 1 may include a communication unit (not shown) that uses at least one communication protocol of GPIO, I2C, I2S, and PWM.

[0074] However, the communication protocols and peripheral interfaces described above are only one embodiment, and the way the device 1 communicates with the sensor 2, the external memory 21, the external terminal 33, etc. is not limited thereto.

[0075] In one embodiment, the device 1 is installed in a predetermined space to monitor the biological signals of the human body 600. When an abnormal biological signal is detected from the human body 600, an emergency signal is transmitted to a preset external terminal 33, so that an emergency situation can be immediately responded to. Thereby, unintentional asphyxiation accidents of infants, lonely deaths of the elderly, etc. can be prevented.

[0076] FIG. 7 is a diagram for explaining the low data according to an embodiment of the present invention.

[0077] The device according to an embodiment of the present invention is a neuromorphic device including a neural network, and may be an artificial intelligence model learned based on predetermined learning data. That is, the abnormal biological signal detection model may be realized by a neuromorphic device.

[0078] Here, the predetermined learning data may be generated based on the low data. By processing the low data according to the following description, the predetermined learning data can be generated.

[0079] In one embodiment, the load data may include normal biological signal data 710, abnormal biological signal data 720, and noise data 730.

[0080] Hereinafter, it will be described assuming that the biological signal is a respiration signal, but this is merely for convenience of explanation, and the biological signal can include various human body signals such as a heartbeat signal.

[0081] The normal biological signal data 710 is data obtained by measuring only normal respiration signals. For example, it may be a predetermined amount of data collected from a monitoring subject who is breathing normally by the sensor. Similarly, the abnormal biological signal data 720 may be data obtained by measuring unstable respiration such as when a respiratory disorder occurs. For example, it may be a predetermined amount of data collected from a monitoring subject who is breathing abnormally by the sensor. Here, the predetermined amount can be changed according to various embodiments such as 3 hours and 20 minutes, 10 hours, etc.

[0082] The noise data 730 may be data of a 3-hour and 20-minute amount obtained by measuring signals other than the biological signal of the monitoring subject, such as the movement of the monitoring subject or the movement of an observer or companion animal. By generating learning data not only for normal and abnormal biological signals but also for noise, higher performance can be achieved when the product is actually applied.

[0083] As an example, the amounts of the normal biological signal data 710, abnormal biological signal data 720, and noise data 730 may be the same. Thus, the model can be appropriately learned about normal / abnormal biological signals and noise, and high performance can be achieved. However, the amounts of each data may be different, and the amounts of each data are not limited to the above-described embodiments.

[0084] In one embodiment, each raw data may include in-phase raw data ("I-data" in FIG. 7) and phase-shifted raw data ("Q-data" in FIG. 7). As an example, the phase-shifted raw data may be quadrature data. The in-phase raw data can mean the real part of the raw data represented by a complex number, and the phase-shifted raw data can mean the imaginary part of the raw data represented by a complex number. The in-phase raw data and the phase-shifted raw data can be used together to indicate other dimensions of a complex number signal.

[0085] FIG. 8 is a diagram for explaining a method of generating a plurality of sample data according to an embodiment of the present invention.

[0086] The device described in FIGS. 1 to 7 means a neuromorphic device that detects an abnormal biological signal, but the device that generates an abnormal biological signal detection model realized by the neuromorphic device (hereinafter, "model generation device") can mean a device that generates or processes learning data for learning the abnormal biological signal detection model.

[0087] In one embodiment, the model generation device can extract a plurality of frames from the raw data to generate a plurality of sample data 800. Here, the model generation device can extract a plurality of frames having a first data length 801 from the raw data.

[0088] As an example, the first data length 801 can be 30 seconds. Assuming that the length 802 of the raw data is 3 hours and 20 minutes and the raw data is sampled at 20 Hz, the number of data points corresponding to the length 802 of the raw data can be 240,000, and the number of data points of the first data length 801 corresponding to 30 seconds can be 600.

[0089] In one embodiment, the model generation device can randomly extract a plurality of frames from the load data. That is, instead of sequentially extracting by data points, it can randomly extract a plurality of frames having a first data length 801. Also, the model generation device can extract a preset number (for example, 2000) of the plurality of frames and generate a plurality of sample data 800 of the preset number.

[0090] On the other hand, when the load data includes in-phase load data and moving-phase load data, the model generation device can randomly extract a frame having a first data length 801 from the in-phase load data and generate a plurality of in-phase sample data 800. Also, the model generation device can extract a frame having a first data length 801 from the moving-phase load data so as to correspond to each signal of the plurality of in-phase sample data 800 and generate a plurality of moving-phase sample data 810. That is, the in-phase sample data 800 can be randomly extracted, and the moving-phase sample data 810 can extract a signal at a time point corresponding to the in-phase sample data 800. As an example, the model generation device can extract a frame obtained by moving the in-phase sample data 800 by the time difference between the moving-phase load data and the in-phase load data from the moving-phase load data and generate the moving-phase sample data 810 so as to correspond to the signal of the in-phase sample data 800.

[0091] FIGS. 9 and 10 are diagrams for explaining a method of generating a plurality of unit data according to an embodiment of the present invention.

[0092] Referring to FIG. 9, a plurality of unit data 900 to 990 generated from sample data having a first data length 801 are shown.

[0093] In one embodiment, the model generation device can extract frames having a second data length 901 from sample data in chronological order to generate a plurality of unit data 900 to 990. As an example, the second data length 901 can be 7.5 seconds. If the first data length 801 of the sample data is 30 seconds and the number of data points of the first data length 801 is 600 with a data sampling period of 20 Hz, the number of data points of the second data length 901 can be 150.

[0094] In one embodiment, the model generation device can sequentially extract frames having a second data length 901 so that they overlap by a third data length 902 among the plurality of unit data 900 to 990. Specifically, the model generation device can extract the first frame 900 including the first data point, and then extract the second frame 910 so that it overlaps with the first frame 900 by the third data length 902. Similarly, the model generation device can extract the third frame (not shown) so that it overlaps with the second frame 910 by the third data length 902.

[0095] As an example, the third data length 902 can be 5 seconds. When the second data length 901 is 7.5 seconds, the model generation device can move by a length of 2.5 seconds in extracting each frame to extract the next frame. Here, when the number of data points of the second data length 901 is 150, the third data length 902 can be 100, and a total of 10 frames are extracted, so a plurality of unit data 900 to 990 can be generated 10. Also, since a plurality of unit data 900 to 990 are generated for each of the sample data, the number of unit data corresponding to one load data can be generated by multiplying the number of a plurality of sample data (for example, 2000) by the number of unit data generated from each sample data (for example, 10). In FIG. 9, it shows that 10 unit data 900 to 990 are generated, but the number of a plurality of unit data 900 to 990 can be variously changed depending on the third data length 902.

[0096] In one embodiment, the plurality of sample data may include a plurality of in-phase sample data and a plurality of moving-phase sample data. Therefore, the model generation device can generate a plurality of in-phase unit data corresponding to the plurality of in-phase sample data and a plurality of moving-phase unit data corresponding to the plurality of moving-phase sample data. That is, the model generation device can generate a plurality of in-phase unit data for each of the in-phase sample data and generate a plurality of moving-phase unit data for each of the moving-phase sample data.

[0097] In one embodiment, the model generation device can convert the generated plurality of unit data into frequency components. For example, the model generation device can perform a Fast Fourier Transform (FFT) on the generated plurality of unit data.

[0098] Referring to FIG. 10, a plurality of unit data 1010 to 1100 converted into frequency components are shown.

[0099] In one embodiment, the model generation device can convert the plurality of in-phase unit data into frequency components and convert the plurality of moving-phase unit data into frequency components.

[0100] FIGS. 11 to 13 are diagrams for explaining a method of generating a learning data set according to an embodiment of the present invention.

[0101] Referring to FIG. 11, a plurality of unit data 1100 to 1190 are shown.

[0102] In one embodiment, the model generation device can generate a learning data set based on the plurality of unit data 1100 to 1190. Specifically, the model generation device can synthesize the plurality of unit data 1100 to 1190 converted into frequency components in the order of extraction.

[0103] This allows one piece of learning data to be generated corresponding to one piece of sample data. The reason why the multiple unit data 1100 to 1190 that have already been generated by overlapping each other are synthesized is that if a signal such as a delta function is included in the signal, it can be confirmed whether it is noise or not depending on whether there is continuity with adjacent unit data and reflected in the learning data.

[0104] Referring to FIG. 12, a plurality of in-phase unit data 1100 to 1190 and a plurality of shifted phase unit data 1101 to 1191 are shown.

[0105] In one embodiment, the model generating device can cross-combine a plurality of in-phase unit data 1100-1190 and a plurality of moving phase unit data 1101-1191. Cross-combine can mean combining in order, such that the first moving phase unit data 1101 is positioned after the first in-phase unit data 1100, the second in-phase unit data 1110 is positioned after the first moving phase unit data 1101, and the second in-phase unit data 1111 is positioned after the second in-phase unit data 1110. Similarly, the third in-phase unit data 1120 and the third moving phase unit data 1121 can be positioned after the second moving phase unit data 1111, and depending on the number of unit data, the last in-phase unit data 1190 and the last moving phase unit data 1191 can be positioned at the end.

[0106] Referring to FIG. 13, the raw data includes normal biological signal data, abnormal biological signal data, and noise data, so the model generating device can generate normal biological signal training data 1310 from the normal biological signal data 710, generate abnormal biological signal training data 1320 from the abnormal biological signal data 720, and generate noise signal training data 1330 from the noise data 730.

[0107] A plurality of sample data and a plurality of unit data may be generated for normal biological signal data, abnormal biological signal data, and noise data respectively. For example, the model generation device can generate N (e.g., 2000) in-phase sample data 800 and N (e.g., 2000) moving-phase sample data 810 from the normal biological signal data 710.

[0108] In addition, since a plurality of unit data (not shown) generated from one sample data 800 are further synthesized into one learning data, N (e.g., 2000) normal biological signal learning data 1310, abnormal biological signal learning data 1320, and noise signal learning data 1330 can also be generated.

[0109] In one embodiment, the learning data set generated by the model generation device may include normal biological signal learning data 1310, abnormal biological signal learning data 1320, and noise signal learning data 1330.

[0110] On the other hand, in one embodiment, the model generation device can generate an abnormal biological signal detection model by training an artificial intelligence model with the learning data set.

[0111] FIG. 14 is a graph showing the result of training an artificial intelligence model with the learning data set according to one embodiment of the present invention.

[0112] In one embodiment, the model generation device can adjust the parameters of the artificial intelligence model with the learning data set. Here, the artificial intelligence model may be a machine learning model.

[0113] Referring to FIG. 14, it can be seen that as the number of times (number of epochs) learned in the training dataset increases, the training loss function and validation loss function of the machine learning model change as shown in FIG. 14. The training loss function (train_loss) is a value indicating the disappearance of the training dataset, and the training loss function decreases as the machine learning model makes more accurate predictions for the training dataset. The validation loss function (val_loss) is a value indicating the disappearance of new data (i.e., validation data) that the machine learning model has never seen before, and the validation loss function decreases as the machine learning model becomes more generalized.

[0114] In one embodiment, the model generation device can select the machine learning model with the lowest loss function and the highest model accuracy among them and determine it as the abnormal biological signal detection model.

[0115] The abnormal biological signal detection model generated in this way is realized by a neural network, and the processor of the neuromorphic device can input the input data collected by the sensor into the abnormal biological signal detection model and detect the abnormal biological signal based on the output of the abnormal biological signal detection model.

[0116] FIG. 15 is a block diagram of an apparatus for detecting an abnormal biological signal according to an embodiment of the present invention.

[0117] Referring to FIG. 15, the apparatus 1500 may include a communication unit 1510, a processor 1520, and a DB 1530. Only the components related to the embodiment are shown in the apparatus 1500 of FIG. 15. Therefore, those skilled in the art will understand that other general-purpose components may be further included in addition to the components shown in FIG. 15.

[0118] The communication unit 1510 may include one or more components that enable wired / wireless communication with an external server or device. For example, the communication unit 1510 may include at least one of a short-range communication unit (not shown), a mobile communication unit (not shown), and a broadcast receiving unit (not shown). In one embodiment, the communication unit 1510 can use at least one communication protocol of a serial peripheral interface (SPI) and a universal asynchronous receiver / transmitter (UART). In another embodiment, the communication unit 1510 can communicate with sensors, external memories, and external control devices.

[0119] The DB 1530 is hardware that stores various data processed within the device 1500, and can store programs for processing and controlling the processor 1520.

[0120] DB1530 may include random access memory (RAM), such as dynamic random access memory (DRAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), CD-ROM, Blu-ray or other optical disc storage, hard disk drive (HDD), solid state drive (SSD), or flash memory.

[0121] The processor 1520 controls the overall operation of the device 1500. For example, the processor 1520 may generally control an input unit (not shown), a display (not shown), a communication unit 1510, a DB 1530, etc. by executing a program stored in the DB 1530. The processor 1520 may generally control the operation of the device 1500 by executing a program stored in the DB 1530.

[0122] The processor 1520 can control at least a part of the operation of the device or the model generation device that detects the abnormal biological signals described above with reference to FIGS. 1 to 14.

[0123] The processor 1520 can be implemented using at least one of ASICs (application specific integrated circuits), DSPs (digital signal processors), DSPDs (digital signal processing devices), PLDs (programmable logic devices), FPGAs (field programmable gate arrays), controllers, micro-controllers, microprocessors, and other electrical units for performing functions.

[0124] In one embodiment, the device 1500 may be a server. The server can be implemented by a computer device or a plurality of computer devices that communicate via a network to provide instructions, codes, files, contents, services, etc. The server can receive data necessary for lightweighting the neural network model and lightweight the neural network model based on the received data.

[0125] On the other hand, embodiments according to the present invention can be realized in the form of a computer program executable by various components on a computer, and such a computer program can be recorded on a computer-readable medium. Here, the medium includes magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specially configured to store and execute program instructions such as ROMs, RAMs, and flash memories.

[0126] On the other hand, the computer program may be one specially designed and constructed for the present invention, or it may be one that is well known and available to those skilled in the art of computer software. Examples of computer programs include not only machine language code such as that produced by a compiler, but also high-level language code that is executed by a computer using an interpreter, etc.

[0127] According to one embodiment, methods according to various embodiments of the present disclosure can be provided in a computer program product. The computer program product can be traded as a commodity between a seller and a buyer. The computer program product can be distributed in the form of a machine-readable storage medium (e.g., a compact disc read only memory (CD-ROM)) or can be distributed through an application store (e.g., the Play Store). TM ) or directly between two user devices. In the case of online distribution, at least a portion of the computer program product may be at least temporarily stored on or temporarily generated by a machine-readable storage medium, such as the memory of a manufacturer's server, an application store server, or an intermediary server.

[0128] Unless explicitly stated or stated to the contrary, steps constituting a method according to the present invention may be performed in any suitable order. The present invention is not necessarily limited to the order of steps described above. The use of all examples or exemplary terms in the present invention is merely for the purpose of explaining the present invention in detail, and the scope of the present invention is not limited by the examples or exemplary terms unless limited by the claims. Furthermore, those skilled in the art will understand that various modifications, combinations, and variations can be made within the scope of the claims or their equivalents, depending on design conditions and factors.

[0129] Therefore, the concept of the present invention should not be limited to the above-described embodiments, and not only the scope of the appended claims, but also all scopes equivalent to or modified equivalently from the scope of the claims, can be said to fall within the scope of the concept of the present invention.

Claims

1. Randomly extracting frames with a first data length from raw data including biosignal data to generate a plurality of sample data; Extracting frames with a second data length from the sample data in chronological order to generate a plurality of unit data; Generating a training data set based on the plurality of unit data; Generating an abnormal biosignal detection model by training an artificial intelligence model with the training data set, the method for detecting an abnormal biosignal including the steps above.

2. The plurality of sample data and the plurality of unit data are Generated for each of the normal biosignal data, abnormal biosignal data, and noise data included in the raw data, The training data set is The method for detecting an abnormal biosignal according to claim 1, including normal biosignal training data generated from the normal biosignal data, abnormal biosignal training data generated from the abnormal biosignal data, and noise signal training data generated from the noise data.

3. The step of generating the plurality of unit data includes Sequentially extracting frames with the second data length so that they overlap by a third data length among the plurality of unit data, the method for detecting an abnormal biosignal according to claim 1.

4. The step of generating the plurality of unit data includes Converting the generated plurality of unit data into frequency components, the method for detecting an abnormal biosignal according to claim 1.

5. The step of generating the training data set includes Combining the converted plurality of unit data in the extraction order, the method for detecting an abnormal biosignal according to claim 4.

6. The raw data includes In-phase raw data and phase-shifted raw data, The step of generating the plurality of sample data includes Randomly extracting frames with the first data length from the in-phase raw data to generate a plurality of in-phase sample data; Extracting frames with the first data length from the phase-shifted raw data so as to correspond to the signals of the plurality of in-phase sample data respectively to generate a plurality of phase-shifted sample data, the method for detecting an abnormal biosignal according to claim 5.

7. The step of generating the training data set is The method for detecting an abnormal biological signal according to claim 6, comprising a step of cross-combining and synthesizing a plurality of in-phase unit data corresponding to the plurality of in-phase sample data and a plurality of moving-phase unit data corresponding to the plurality of moving-phase sample data.

8. The step of generating the abnormal biological signal detection model includes: adjusting parameters of the artificial intelligence model with the learning data set; and generating the abnormal biological signal detection model based on a loss function according to the adjusted number of times (number of epochs) and the accuracy of the artificial intelligence model. The method for detecting an abnormal biological signal according to claim 1.

9. The abnormal biological signal detection model is implemented by a neuromorphic device, wherein the neuromorphic device includes at least one processor for driving the abnormal biological signal detection model, and a crossbar array circuit for receiving instructions from the at least one processor and performing in-memory operations. The at least one processor inputs input data collected by a sensor into the abnormal biological signal detection model, and detects an abnormal biological signal based on an output of the abnormal biological signal detection model. The method for detecting an abnormal biological signal according to claim 1.

10. a memory storing at least one program, and a processor for performing operations by executing the at least one program, wherein the processor randomly extracts a frame having a first data length from raw data including biological signal data to generate a plurality of sample data, extracts a frame having a second data length from the sample data in chronological order to generate a plurality of unit data, generates a learning data set based on the plurality of unit data, and generates an abnormal biological signal detection model by training an artificial intelligence model with the learning data set. An apparatus for detecting an abnormal biological signal.

11. A computer-readable recording medium recording a program for executing the method according to claim 1 by a computer.

Images