Method and apparatus for early prediction of septic shock through artificial intelligence-based biometric data analysis, and computer program

Abstract

Provided are a method and an apparatus for early prediction of septic shock through artificial intelligence-based biometric data analysis, and a computer program. The method for early prediction of septic shock through artificial intelligence-based biometric data analysis, according to various embodiments of the present invention, is performed by a computing apparatus and comprises the steps of: collecting, from a user, time-series-based biometric data measured in real time; and predicting, on the basis of an artificial intelligence model, the occurrence of septic shock for the user by analyzing the collected biometric data, wherein the step of predicting the occurrence of septic shock comprises a step of, when the occurrence of sepsis for the user is detected by analyzing the collected biometric data, analyzing biometric data at a future time point, which is predicted on the basis of the collected biometric data, for early prediction of the occurrence of septic shock at the future time point.

Description

Method, device, and computer program for early prediction of septic shock using artificial intelligence-based biometric data analysis

[0001] Various embodiments of the present invention relate to a method, device, and computer program for early prediction of septic shock through artificial intelligence-based bio-data analysis.

[0002] Sepsis is a condition in which bacteria or bacterial toxins are present in the blood and cause a body-wide infection.

[0003] When two or more of the following symptoms are present: fever (body temperature rising above 38 degrees Celsius) or hypothermia (body temperature falling below 36 degrees Celsius), respiratory rate increasing to 24 times per minute (tachypnea) or heart rate exceeding 90 beats per minute (tachycardia), and increased or significantly decreased white blood cell count in blood tests, this is called systemic inflammatory response syndrome (SIRS). When this systemic inflammatory response syndrome is caused by a microbial infection, it is called sepsis.

[0004] When bacteria or toxins produced by bacteria appear in the blood, our body responds with various defense mechanisms, and patients show the above symptoms during this response process.

[0005] The body's response to sepsis, triggered by various inflammatory mediators and the accompanying blood flow disturbances, can lead to damage to vital organs. Sepsis accompanied by vital organ damage is called severe sepsis, and if blood pressure cannot be maintained with standard fluid therapy, it is called septic shock. Mortality rates are naturally higher in severe sepsis than in severe sepsis, and in septic shock than in severe sepsis.

[0006] If sepsis is not treated promptly, the patient's physical response can escalate like a whirlwind, leading to death within days. Early and appropriate use of antibiotics and fluid therapy can alleviate the progression of sepsis. Therefore, early treatment is essential to improving the patient's survival rate. The challenge for medical professionals is to act within two hours of the onset of symptoms, and at the latest within a day. Once sepsis has progressed, even with the help of various devices and specialized intensive care, it is difficult to improve survival rates.

[0007] Diagnosing sepsis requires vital signs and blood tests. Vital signs must be measured regularly by healthcare professionals, while blood tests require infrastructure capable of collecting and analyzing blood. Consequently, hospitals have struggled to monitor sepsis in real time. Sepsis is particularly burdensome in areas with low sociodemographic indices, as these areas often lack medical resources for sepsis diagnosis and treatment.

[0008] The problem to be solved by the present invention is to provide a method, device and computer program for early prediction of septic shock through artificial intelligence-based vital data analysis, which can diagnose and predict sepsis and septic shock in real time by analyzing time-series-based vital signs collected in real time from a user to determine whether the user has developed sepsis or septic shock, and can diagnose and predict sepsis and septic shock more easily through sensor data collected not only through a professional measuring device but also through a wearable device such as a smart watch.

[0009] Another problem to be solved by the present invention is to provide a method, device and computer program for early prediction of septic shock through artificial intelligence-based bio-data analysis, which can predict future bio-data based on current bio-data when the occurrence of sepsis is detected by analyzing bio-data collected from a user in real time, and detect the occurrence of septic shock by analyzing future bio-data, thereby enabling early prediction of the occurrence of septic shock, thereby drastically reducing the mortality rate of sepsis patients.

[0010] The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

[0011] According to an embodiment of the present invention for solving the above-described problem, a method for early prediction of septic shock through analysis of artificial intelligence-based bio-data is provided, which is performed by a computing device, including the steps of collecting time-series bio-data measured in real time from a user, and predicting the occurrence of septic shock for the user by analyzing the collected bio-data based on an artificial intelligence model, wherein the step of predicting the occurrence of septic shock may include the step of analyzing bio-data at a future point in time predicted based on the collected bio-data when the occurrence of sepsis for the user is detected by analyzing the collected bio-data, thereby early predicting the occurrence of septic shock at the future point in time.

[0012] In various embodiments, the step of early predicting the occurrence of septic shock at a future point in time may include the step of analyzing biometric data collected from the user at a first point in time to determine whether sepsis has occurred for the user at the first point in time, the step of predicting biometric data at a second point in time when a predetermined amount of time has passed from the first point in time based on the biometric data collected at the first point in time when it is determined that sepsis has occurred, and the step of analyzing the predicted biometric data at the second point in time to determine whether septic shock has occurred for the user at the second point in time.

[0013] In various embodiments, the step of determining whether sepsis occurs in the user at the first time point includes the step of determining whether sepsis occurs in the user at the first time point by analyzing biometric data collected at the first time point using a first artificial intelligence model, wherein the first artificial intelligence model may be a machine learning or deep learning-based classification model (Classifier) ​​trained using a plurality of biometric data labeled with information regarding the occurrence of sepsis as training data.

[0014] In various embodiments, the step of predicting the bio-data at the second time point includes the step of predicting the bio-data at the second time point by analyzing the bio-data collected at the first time point using a second artificial intelligence model, wherein the second artificial intelligence model may be a deep learning-based prediction model (Forecaster) learned using a plurality of bio-data collected from sepsis patients as learning data.

[0015] In various embodiments, the step of determining whether septic shock occurs for the user at the second time point includes the step of analyzing the predicted biometric data at the second time point through a third artificial intelligence model to determine whether septic shock occurs for the user at the second time point, wherein the third artificial intelligence model may be a machine learning or deep learning-based classification model (Classifier) ​​trained using a plurality of biometric data labeled with information regarding the occurrence of septic shock as training data.

[0016] In various embodiments, the learning data includes a plurality of biometric data labeled with the occurrence of septic shock at a location corresponding to the time at which septic shock occurred, wherein the time at which septic shock occurred may be the time at which a vasopressor is administered.

[0017] In various embodiments, the artificial intelligence model includes a plurality of artificial intelligence models generated by classifying biometric data collected from a plurality of measuring devices having different properties by measuring device and learning each of the classified biometric data by measuring device as learning data, and the plurality of artificial intelligence models are models that are independently fine-tuned based on weights corresponding to each of the plurality of measuring devices, and the weights corresponding to each of the plurality of measuring devices may be determined based on the accuracy of sensor data measured through each of the plurality of measuring devices.

[0018] In various embodiments, the collected biometric data includes a plurality of biometric data of different types, and the step of predicting the occurrence of septic shock may further include a step of selecting at least one biometric data among the plurality of biometric data based on a type of a measuring device that measured the plurality of biometric data, and analyzing only the selected at least one biometric data through the artificial intelligence model to predict the occurrence of septic shock for the user.

[0019] In various embodiments, the collected biometric data includes a plurality of biometric data of different types, and the step of predicting the occurrence of septic shock may further include the step of extracting a plurality of result data by individually analyzing the plurality of biometric data through the artificial intelligence model, and predicting the occurrence of septic shock for the user using the plurality of extracted result data, and assigning a weight to each of the plurality of extracted result data based on the type of the measuring device that measured the plurality of biometric data, and using the plurality of weighted result data to predict the occurrence of septic shock for the user.

[0020] In various embodiments, the artificial intelligence model includes a plurality of artificial intelligence models generated by classifying a plurality of biometric data collected in different environments by situation and learning each of the plurality of biometric data classified by situation as learning data, and the step of predicting the occurrence of septic shock may further include a step of predicting the occurrence of septic shock for the user by analyzing the biometric data collected in the specific situation by adopting an artificial intelligence model learned by using the biometric data collected in the specific situation as learning data among the plurality of artificial intelligence models when biometric data in a specific situation is collected from the user.

[0021] In various embodiments, the artificial intelligence model may further include an explainable artificial intelligence model (XAI) and a generative artificial intelligence model, and may further include a step of deriving basis data for a result of early prediction of the occurrence of septic shock at the future time point using the explainable artificial intelligence model, a step of deriving response plan data for a result of early prediction of the occurrence of septic shock at the future time point using the generative artificial intelligence model, and a step of generating a result report including the result of early prediction of the occurrence of septic shock at the future time point, the derived basis data, and the derived response plan data, and providing the generated result report.

[0022] A computing device for performing a method for early prediction of septic shock through analysis of artificial intelligence-based bio-data according to another embodiment of the present invention for solving the above-described problem comprises a processor, a network interface, a memory, and a computer program loaded into the memory and executed by the processor, wherein the computer program includes an instruction for collecting time-series-based bio-data measured in real time from a user, an instruction for predicting the occurrence of septic shock for the user by analyzing the collected bio-data based on an artificial intelligence model, and the instruction for predicting the occurrence of septic shock may include an instruction for early predicting the occurrence of septic shock at a future time by analyzing bio-data at a future time predicted based on the collected bio-data when the occurrence of sepsis for the user is detected by analyzing the collected bio-data.

[0023] According to another embodiment of the present invention for solving the above-described problem, a method for early prediction of septic shock through artificial intelligence-based bio-data analysis is provided, which comprises a step of collecting time-series-based bio-data measured in real time from a user, and a step of predicting the occurrence of septic shock for the user by analyzing the collected bio-data based on an artificial intelligence model, wherein the step of predicting the occurrence of septic shock comprises a step of analyzing the bio-data at a future point in time predicted based on the collected bio-data, when the occurrence of sepsis for the user is detected by analyzing the collected bio-data, and early predicting the occurrence of septic shock at the future point in time, and the method may be stored in a recording medium readable by a computing device to execute the method.

[0024] Other specific details of the present invention are included in the detailed description and drawings.

[0025] According to various embodiments of the present invention, by analyzing time series-based vital signs collected in real time from a user to determine whether the user has developed sepsis or septic shock, sepsis and septic shock can be diagnosed and predicted in real time, and sepsis and septic shock can be diagnosed and predicted more easily through sensor data collected not only through a professional measuring device but also through a wearable device such as a smart watch.

[0026] In addition, when the occurrence of sepsis is detected by analyzing biometric data collected from the user in real time, future biometric data can be predicted based on the current biometric data, and the occurrence of septic shock can be detected by analyzing the future biometric data, thereby enabling early prediction of the occurrence of septic shock, which can drastically reduce the mortality rate of sepsis patients.

[0027] The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

[0028] FIG. 1 is a diagram illustrating a system for early prediction of septic shock through artificial intelligence-based bio-data analysis according to one embodiment of the present invention.

[0029] FIG. 2 is a diagram illustrating a hardware configuration of a computing device that performs a method for early prediction of septic shock through artificial intelligence-based bio-data analysis according to another embodiment of the present invention.

[0030] Figure 3 is a flowchart of a method for early prediction of septic shock through artificial intelligence-based bio-data analysis according to another embodiment of the present invention.

[0031] FIG. 4 is a diagram illustrating biometric data applicable to various embodiments.

[0032] FIG. 5 is a flowchart illustrating a method for early prediction of septic shock onset based on an artificial intelligence model in various embodiments.

[0033] FIG. 6 and FIG. 7 are diagrams illustrating a process of analyzing biometric data through an artificial intelligence model and deriving result data in various embodiments.

[0034] The advantages and features of the present invention, and the methods for achieving them, will become clearer with reference to the embodiments described in detail below together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms. These embodiments are provided solely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the present invention, and the present invention is defined solely by the scope of the claims.

[0035] The terminology used herein is for the purpose of describing embodiments only and is not intended to limit the present invention. In this specification, the singular also includes the plural unless specifically stated otherwise. As used herein, the terms "comprises" and / or "comprising" do not exclude the presence or addition of one or more other components in addition to the mentioned components. Like reference numerals refer to like components throughout the specification, and "and / or" includes each and any combination of one or more of the mentioned components. Although "first", "second", etc. are used to describe various components, these components are not limited by these terms. These terms are only used to distinguish one component from another. Therefore, it should be understood that a first component mentioned below may also be a second component within the technical spirit of the present invention.

[0036] Unless otherwise defined, all terms (including technical and scientific terms) used herein may be used in their common sense to those skilled in the art to which the present invention pertains. Furthermore, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise.

[0037] The term "part" or "module" as used herein refers to a software or hardware component such as an FPGA or ASIC, and the "part" or "module" performs certain functions. However, the "part" or "module" is not limited to software or hardware. The "part" or "module" may be configured to reside on an addressable storage medium and may be configured to execute one or more processors. Thus, by way of example, the "part" or "module" includes components such as software components, object-oriented software components, class components, and task components, as well as processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays, and variables. The functionality provided within the components and "parts" or "modules" may be combined into fewer components and "parts" or "modules" or further separated into additional components and "parts" or "modules."

[0038] Spatially relative terms such as "below," "beneath," "lower," "above," and "upper" can be used to easily describe the relationship between one component and other components as depicted in the drawings. Spatially relative terms should be understood to include different orientations of the components during use or operation in addition to the orientations depicted in the drawings. For example, if a component depicted in the drawings were flipped over, a component described as "below" or "beneath" another component could end up "above" the other component. Thus, the exemplary term "below" can include both the above and below orientations. Components can also be oriented in other directions, and thus spatially relative terms can be interpreted accordingly.

[0039] In this specification, the term "computer" refers to any type of hardware device including at least one processor, and may also be understood to encompass software components operating on the hardware device, depending on the embodiment. For example, the term "computer" may be understood to encompass, but is not limited to, smartphones, tablet PCs, desktops, laptops, and all user clients and applications running on each device.

[0040] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings.

[0041] Although each step described in this specification is described as being performed by a computer, the subject of each step is not limited thereto, and at least some of each step may be performed by different devices depending on the embodiment.

[0042]

[0043] FIG. 1 is a diagram illustrating a system for early prediction of septic shock through artificial intelligence-based bio-data analysis according to one embodiment of the present invention.

[0044] Referring to FIG. 1, a system for early prediction of septic shock through artificial intelligence-based bio-data analysis according to one embodiment of the present invention may include a computing device (100), a user terminal (200), an external server (300), and a network (400).

[0045] Here, the early prediction system for septic shock through artificial intelligence-based bio-data analysis illustrated in FIG. 1 is according to one embodiment, and its components are not limited to the embodiment illustrated in FIG. 1, and may be added, changed, or deleted as needed.

[0046] In one embodiment, the computing device (100) can perform an early prediction process of septic shock through artificial intelligence-based bio-data analysis.

[0047] In various embodiments, the computing device (100) can collect time-series-based biometric data measured in real time from a user and predict the occurrence of septic shock in the user by analyzing the biometric data through an artificial intelligence model.

[0048] Here, the artificial intelligence model is a model trained using biometric data as learning data, and can be a model that uses biometric data as input data to predict whether there is sepsis or septic shock as output data.

[0049] An artificial intelligence model (e.g., a neural network) consists of one or more network functions, which may be comprised of a set of interconnected computational units, generally referred to as "nodes." These "nodes" may also be referred to as "neurons." One or more network functions comprise at least one node. The nodes (or neurons) comprising one or more network functions may be interconnected by one or more "links."

[0050] Within an AI model, one or more nodes connected via links can form a relationship between input nodes and output nodes. The concepts of input nodes and output nodes are relative, meaning that any node in an output node relationship with one node can also be in an input node relationship with another node, and vice versa. As described above, input node-to-output node relationships can be created around links. One input node can be connected to one or more output nodes via links, and vice versa.

[0051] In a relationship between input nodes and output nodes connected through a single link, the value of the output node can be determined based on the data input to the input node. Here, the node interconnecting the input nodes and output nodes can have a weight. The weight can be variable and can be varied by a user or an algorithm so that the artificial intelligence model can perform a desired function. For example, when one or more input nodes are interconnected to one output node through each link, the output node can determine the output node value based on the values ​​input to the input nodes connected to the output node and the weight set for the link corresponding to each input node.

[0052] As described above, an AI model consists of one or more nodes interconnected through one or more links, forming input and output node relationships within the AI ​​model. The characteristics of an AI model can be determined based on the number of nodes and links, the relationships between nodes and links, and the weights assigned to each link. For example, if two AI models exist with the same number of nodes and links but different weight values ​​between the links, the two AI models may be perceived as different from each other.

[0053] Some of the nodes that make up an AI model can form a layer based on their distances from the initial input node. For example, a set of nodes that are n distances from the initial input node can form n layers. The distance from the initial input node can be defined by the minimum number of links that must be passed to reach the node from the initial input node. However, this definition of a layer is arbitrary for illustrative purposes, and the order of layers within an AI model can be defined in a different way than described above. For example, the layer of nodes can also be defined by their distance from the final output node.

[0054] The initial input node may refer to one or more nodes in an AI model into which data is directly input without going through a link in its relationship with other nodes. Alternatively, in an AI model network, in terms of the relationship between nodes based on links, it may refer to nodes that do not have other input nodes connected by links. Similarly, the final output node may refer to one or more nodes in an AI model that do not have an output node in its relationship with other nodes. In addition, a hidden node may refer to nodes that constitute an AI model other than the initial input node and the final output node. An AI model according to one embodiment of the present invention may be an AI model in which the number of nodes in the input layer may be greater than that in the hidden layer closer to the output layer, and the number of nodes decreases as it progresses from the input layer to the hidden layer.

[0055] An AI model may include one or more hidden layers. Hidden nodes in a hidden layer can receive the output of the previous layer and the output of surrounding hidden nodes as input. The number of hidden nodes in each hidden layer may be the same or different. The number of nodes in the input layer may be determined based on the number of data fields in the input data and may be the same as or different from the number of hidden nodes. Input data entered into the input layer may be computed by the hidden nodes in the hidden layer and output by the output layer, a fully connected layer (FCL).

[0056] In various embodiments, the artificial intelligence model may be a deep learning model.

[0057] A deep learning model (e.g., a deep neural network (DNN)) can refer to an artificial intelligence model that includes multiple hidden layers in addition to input and output layers. Using a deep neural network, it is possible to identify latent structures in data. That is, it is possible to identify latent structures in photos, text, videos, voices, and music (e.g., what objects are in the photo, what the content and emotion of the text are, what the content and emotion of the voice are, etc.).

[0058] Deep neural networks may include, but are not limited to, convolutional neural networks (CNNs), recurrent neural networks (RNNs), autoencoders, generative adversarial networks (GANs), restricted boltzmann machines (RBMs), deep belief networks (DBNs), Q networks, U networks, Siamese networks, etc.

[0059] In various embodiments, the septic shock early prediction and notification service through the septic shock early prediction process through artificial intelligence-based bio-data analysis provided from the computing device (100) can be implemented in the form of an application, and users can use the septic shock early prediction and notification service provided by the computing device (100) through devices capable of running the application (e.g., smartphones, wearable devices, etc.) by executing the application through the devices.

[0060] In various embodiments, the computing device (100) may be connected to a user terminal (200) via a network (400), and may collect various information (e.g., user information, user's bio-data, etc.) necessary for performing an early prediction method for septic shock through artificial intelligence-based bio-data analysis from the user terminal (200), or may provide various information (e.g., result reports) derived by performing an early prediction method for septic shock through artificial intelligence-based bio-data analysis.

[0061] Here, the user terminal (200) may refer to any type of entity(ies) in a system having a mechanism for communicating with the computing device (100). For example, the user terminal (200) may include a personal computer (PC), a notebook, a mobile terminal, a smart phone, a tablet PC, a wearable device, etc., and may include all types of terminals capable of connecting to a wired / wireless network. In addition, the user terminal (200) may include any computing device implemented by at least one of an agent, an Application Programming Interface (API), and a plug-in. In addition, the user terminal (200) may include an application source and / or a client application. In addition, the user terminal (200) may include, but is not limited to, an infotainment system installed in a vehicle.

[0062] In addition, here, the network (400) may refer to a connection structure that enables information exchange between each node, such as a plurality of terminals and servers. For example, the network (400) may include a local area network (LAN), a wide area network (WAN), the Internet (WWW), a wired and wireless data communication network, a telephone network, a wired and wireless television communication network, a controller area network (CAN), and Ethernet.

[0063] Wireless data communication networks may include, but are not limited to, 3G, 4G, 5G, 3GPP (3rd Generation Partnership Project), 5GPP (5th Generation Partnership Project), LTE (Long Term Evolution), WIMAX (World Interoperability for Microwave Access), Wi-Fi, the Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network), RF (Radio Frequency), Bluetooth network, NFC (Near-Field Communication) network, satellite broadcasting network, analog broadcasting network, DMB (Digital Multimedia Broadcasting) network, etc.

[0064] In one embodiment, the external server (300) may be connected to the computing device (100) via a network (400), and may store and manage various information and data required for the computing device (100) to perform an early prediction method for septic shock through artificial intelligence-based bio-data analysis, or may collect, store, and manage various information and data derived as the computing device (100) performs an early prediction method for septic shock through artificial intelligence-based bio-data analysis. For example, the external server (300) may be a storage server separately provided outside the computing device (100), but is not limited thereto. Hereinafter, the hardware configuration of the computing device (100) that performs an early prediction method for septic shock through artificial intelligence-based bio-data analysis will be described with reference to FIG. 2.

[0065]

[0066] FIG. 2 is a diagram illustrating a hardware configuration of a computing device that performs a method for early prediction of septic shock through artificial intelligence-based bio-data analysis according to another embodiment of the present invention.

[0067] Referring to FIG. 2, in various embodiments, a computing device (100) may include one or more processors (110), a memory (120) for loading a computer program (151) executed by the processor (110), a bus (130), a communication interface (140), and storage (150) for storing the computer program (151). Here, only components related to the embodiment of the present invention are illustrated in FIG. 2. Therefore, a person skilled in the art to which the present invention pertains may understand that other general components may be included in addition to the components illustrated in FIG. 2.

[0068] The processor (110) controls the overall operation of each component of the computing device (100). The processor (110) may be configured to include a CPU (Central Processing Unit), an MPU (Micro Processor Unit), an MCU (Micro Controller Unit), a GPU (Graphics Processing Unit), or any other type of processor well known in the art of the present invention.

[0069] Additionally, the processor (110) may perform operations for at least one application or program for executing a method according to embodiments of the present invention, and the computing device (100) may have one or more processors.

[0070] In various embodiments, the processor (110) may further include a Random Access Memory (RAM, not shown) and a Read-Only Memory (ROM, not shown) that temporarily and / or permanently store signals (or data) processed within the processor (110). In addition, the processor (110) may be implemented in the form of a system on chip (SoC) that includes at least one of a graphics processing unit, RAM, and ROM.

[0071] The memory (120) stores various data, commands, and / or information. The memory (120) can load a computer program (151) from the storage (150) to execute methods / operations according to various embodiments of the present invention. When the computer program (151) is loaded into the memory (120), the processor (110) can perform the method / operation by executing one or more instructions constituting the computer program (151). The memory (120) may be implemented as a volatile memory such as RAM, but the technical scope of the present disclosure is not limited thereto.

[0072] The bus (130) provides a communication function between components of the computing device (100). The bus (130) may be implemented as various types of buses, such as an address bus, a data bus, and a control bus.

[0073] The communication interface (140) supports wired and wireless Internet communication of the computing device (100). Furthermore, the communication interface (140) may support various communication methods other than Internet communication. To this end, the communication interface (140) may be configured to include a communication module well known in the technical field of the present invention. In some embodiments, the communication interface (140) may be omitted.

[0074] The storage (150) can non-temporarily store a computer program (151). When performing an early prediction process for septic shock through artificial intelligence-based bio-data analysis via a computing device (100), the storage (150) can store various types of information necessary to provide an early prediction process for septic shock through artificial intelligence-based bio-data analysis.

[0075] Storage (150) may be configured to include non-volatile memory such as ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), flash memory, a hard disk, a removable disk, or any type of computer-readable recording medium well known in the art to which the present invention pertains.

[0076] The computer program (151) may include one or more instructions that, when loaded into the memory (120), cause the processor (110) to perform a method / operation according to various embodiments of the present invention. That is, the processor (110) may perform the method / operation according to various embodiments of the present invention by executing the one or more instructions.

[0077] In one embodiment, the computer program (151) may include one or more instructions for performing a method for early prediction of septic shock through artificial intelligence-based bio-data analysis, including a step of collecting time-series-based bio-data measured in real time from a user and a step of predicting the occurrence of septic shock in the user by analyzing the collected bio-data based on an artificial intelligence model.

[0078] The steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or implemented by a combination thereof. The software module may reside in a random access memory (RAM), a read only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a flash memory, a hard disk, a removable disk, a CD-ROM, or any other form of computer-readable recording medium well known in the art to which the present invention pertains.

[0079] The components of the present invention may be implemented as a program (or application) to be executed in combination with a computer as hardware and stored on a medium. The components of the present invention may be implemented as software programming or software elements, and similarly, the embodiments may be implemented in a programming or scripting language such as C, C++, Java, assembler, etc., including various algorithms implemented as a combination of data structures, processes, routines, or other programming components. Functional aspects may be implemented as an algorithm executed on one or more processors. Hereinafter, with reference to FIG. 3, a method for early prediction of septic shock through artificial intelligence-based bio-data analysis performed by a computing device (100) will be described in more detail.

[0080]

[0081] Figure 3 is a flowchart of a method for early prediction of septic shock through artificial intelligence-based bio-data analysis according to another embodiment of the present invention.

[0082] Referring to FIG. 3, at step S110, the computing device (100) can collect biometric data measured from the user.

[0083] In various embodiments, the computing device (100) may be connected to a measurement device via a network (400) and may collect biometric data of the user measured via the measurement device.

[0084] Here, the user's biometric data may be time-series-based biometric data (e.g., FIG. 4) measured in real time from the user through a measuring device. For example, the user's biometric data may include at least one of the user's Shock Index (SI), Respiration (Rr), Saturation of percutaneous oxygen (SpO2), Temperature (Temp), Heart rate (Hr), and Mean arterial pressure (MAP), but is not limited thereto, and various possible biometric data that can determine the user's status may be applied.

[0085] In addition, here, the measuring device for measuring the user's biometric data may be a medical device including a plurality of sensors for measuring different types of biometric data or a plurality of medical devices including each of the plurality of sensors, but is not limited thereto, and the measuring device may be applied to any component capable of measuring the user's biometric data, such as a user terminal (200) (e.g., a user's mobile, wearable device, or an infotainment system of a vehicle in which the user is riding).

[0086] In various embodiments, the computing device (100) may directly receive biometric data about the user from the medical staff responsible for the user who wishes to predict the onset of septic shock. However, this is not limited to this, and may be applied to various methods of collecting biometric data about the user.

[0087] In various embodiments, the computing device (100) may collect the user's biometric data at preset intervals (e.g., every hour). Here, the preset interval may be determined based on the user's history of sepsis and / or septic shock, but is not limited thereto.

[0088]

[0089] At step S120, the computing device (100) can analyze the biometric data collected through step S110 to predict the occurrence of septic shock in the user.

[0090] In general, septic shock occurs as blood pressure drops rapidly during the development and progression of sepsis, that is, sepsis occurs and progresses before septic shock occurs. Taking this into account, the computing device (100) can analyze the user's biometric data to preferentially predict whether sepsis has occurred, and if it is determined that sepsis has occurred, it can predict whether septic shock has occurred.

[0091] In various embodiments, the computing device (100) can detect the occurrence of sepsis in a user by analyzing biometric data at the present time using a pre-learned artificial intelligence model, and when the occurrence of sepsis in the user is detected, the computing device can predict the occurrence of septic shock at a future time in advance by analyzing biometric data at a future time predicted based on the biometric data at the present time. Hereinafter, a method for early prediction of septic shock performed using an artificial intelligence model will be described in more detail with reference to FIGS. 5 to 7.

[0092]

[0093] FIG. 5 is a flowchart illustrating a method for predicting the occurrence of septic shock early based on an artificial intelligence model in various embodiments, and FIGS. 6 and 7 are diagrams illustrating a process for analyzing biological data through an artificial intelligence model to derive result data in various embodiments.

[0094] Referring to FIGS. 5 to 7, in step S210, when biometric data is collected from a user at a first point in time (e.g., the current point in time), the computing device (100) can analyze the biometric data collected at the first point in time to determine whether sepsis occurred at the first point in time.

[0095] In various embodiments, the computing device (100) can determine whether sepsis has occurred at a first time point by analyzing biometric data collected at a first time point using a first artificial intelligence model (10).

[0096] Here, the first artificial intelligence model (10) may be a model trained using multiple bio-data labeled with information on the occurrence of sepsis as learning data, and may be a model that outputs whether or not sepsis occurred as result data using specific bio-data as input data.

[0097] In addition, here, the first artificial intelligence model (10) may be a machine learning or deep learning-based classification model (Classifier), and may be a model that analyzes biometric data to classify a user into one of a plurality of predefined groups (e.g., a normal group, a mild sepsis group, and a severe sepsis group), but is not limited thereto.

[0098] In various embodiments, when multiple types of sensor data are collected from a user, the computing device (100) can individually analyze each of the multiple sensor data through the first artificial intelligence model (10) to extract multiple result data, and can determine whether sepsis has occurred at a first point in time by synthesizing the extracted multiple result data. For example, the computing device (100) can obtain six result data by analyzing each of the shock index, respiratory rate, transcutaneous oxygen saturation, body temperature value, heart rate, and mean arterial pressure collected from the user through the first artificial intelligence model (10), and can collate the six results (e.g., hard voting or soft voting) to determine whether sepsis has occurred in the end. However, the present invention is not limited thereto.

[0099] At step S220, if it is determined that sepsis has occurred through step S210, the computing device (100) can predict the bio-data at the second point in time.

[0100] Here, the second point in time may refer to a point in the future, a predetermined amount of time having elapsed from the current point in time, the first point in time. For example, the second point in time may be a point in time 3 to 4 hours after the current point in time, but is not limited thereto.

[0101] In various embodiments, the computing device (100) can predict biometric data at a second time point by analyzing biometric data collected at a first time point using a second artificial intelligence model (20).

[0102] Here, the second artificial intelligence model (20) may be a model trained using multiple bio-data collected from sepsis patients as learning data, and outputting bio-data at a specific point in time as input data and bio-data at a specific point in time after a predetermined period of time has elapsed from the specific point in time.

[0103] Additionally, the second artificial intelligence model (20) may be a deep learning-based prediction model (Forecaster), but is not limited thereto.

[0104] In various embodiments, when multiple types of sensor data are collected from a user, the computing device (100) can individually analyze each of the multiple sensor data through the second artificial intelligence model (20), thereby predicting multiple sensor data at a second time point corresponding to each of the multiple sensor data at a first time point. However, the present invention is not limited thereto.

[0105] Meanwhile, if the computing device (100) determines that sepsis has not occurred through step S210, the operation of the early prediction process of septic shock through artificial intelligence-based bio-data analysis can be terminated.

[0106] At step S230, the computing device (100) can analyze the biometric data predicted through step S220, i.e., the biometric data at the second point in time, to determine whether septic shock occurs in the user at the second point in time.

[0107] In various embodiments, the computing device (100) can determine whether septic shock occurs in the user at the second time point by analyzing the biometric data at the second time point through the third artificial intelligence model (30).

[0108] Here, the third artificial intelligence model (30) may be a model trained using multiple biometric data labeled with information regarding the occurrence of septic shock as learning data, and may be a model that outputs whether or not septic shock occurred as output data using specific biometric data as input data. For example, the third artificial intelligence model (30) may be a model trained using multiple biometric data labeled with the occurrence of septic shock at a location corresponding to the time at which septic shock occurred as learning data.

[0109] Here, the biometric data used as learning data for the third artificial intelligence model (30) requires labeling work regarding the time point at which septic shock occurred. However, since the time point at which septic shock occurred is a somewhat ambiguous time standard, there is a problem that it is difficult to accurately specify the time point, and there is a problem that the performance of the third artificial intelligence model (30) learned using such learning data may be poor.

[0110] Typically, septic shock occurs as blood pressure drops rapidly during the progression of sepsis. Considering that it is essential to immediately administer a vasopressor when septic shock occurs, by setting the time of septic shock occurrence as the time of vasopressor administration when generating learning data, a more accurate learning data set can be constructed, and through this, a more reliable third artificial intelligence model (30) can be constructed.

[0111] In addition, here, the third artificial intelligence model (30) may be a machine learning or deep learning-based classification model (Classifier), and may be a model that analyzes biometric data to classify a user into one of a plurality of predefined groups (e.g., normal group, caution group, and risk group), but is not limited thereto.

[0112] In various embodiments, when a plurality of sensor data corresponding to each of different types of sensor data is predicted through the second artificial intelligence model (20), the computing device (100) can individually analyze each of the plurality of sensor data through the third artificial intelligence model (30) to extract a plurality of result data, and can determine whether septic shock occurs at the second time point by synthesizing the extracted plurality of result data. For example, the computing device (100) can obtain six result data by analyzing each of the shock index, respiratory rate, transcutaneous oxygen saturation, body temperature value, heart rate, and mean arterial pressure predicted through the second artificial intelligence model (20), and can collate the six results (e.g., hard voting or soft voting) to determine whether septic shock occurs in the end. However, the present invention is not limited thereto.

[0113] In various embodiments, the computing device (100) selects an artificial intelligence model corresponding to a measurement device that measures biometric data from a user from among a plurality of pre-built artificial intelligence models, analyzes the selected artificial intelligence model to analyze the user's biometric data, and thereby determines whether septic shock occurs in the user.

[0114] Typically, professional medical devices that measure biometric data can obtain relatively reliable sensor data, whereas wearable devices can collect relatively unreliable sensor data compared to medical devices due to various factors such as the performance of the wearable device, the user's condition, situation, and posture.

[0115] Additionally, even if it is the same type of measuring device, the accuracy and reliability of each sensor may differ depending on the properties of the measuring device (e.g., in the case of a wearable device, the version, brand, etc. of the wearable device).

[0116] Additionally, in some cases, there are measuring devices that can only collect a limited portion of the biometric data required to determine whether sepsis or septic shock has occurred.

[0117] Taking these points into consideration, the computing device (100) can individually prepare multiple artificial intelligence models corresponding to each of multiple measuring devices having different properties in advance, and adopt the artificial intelligence model corresponding to the measuring device that collects biometric data from the user to determine whether sepsis or septic shock occurs in the user.

[0118] More specifically, first, the computing device (100) classifies biometric data collected from a plurality of measuring devices having different properties by measuring device, and trains different models using each of the biometric data classified by measuring device as learning data, thereby pre-constructing a plurality of artificial intelligence models corresponding to each of the plurality of measuring devices.

[0119] At this time, the multiple artificial intelligence models may be independently fine-tuned models based on weights corresponding to each of the multiple measurement devices.

[0120] Here, the weights corresponding to each of the plurality of measurement devices may be determined based on the accuracy of sensor data measured by each of the plurality of measurement devices. For example, the computing device (100) may assign a higher weight to biometric data with relatively high accuracy among different types of biometric data compared to biometric data with relatively low accuracy, but this is not limited thereto.

[0121] Thereafter, when biometric data is collected through a specific measuring device, the computing device (100) analyzes the biometric data by adopting an artificial intelligence model corresponding to a specific measuring device among multiple artificial intelligence models to determine whether sepsis or septic shock occurs in the user, thereby deriving more accurate results.

[0122] In various embodiments, when a plurality of different types of biometric data are acquired from a user, the computing device (100) analyzes the plurality of biometric data through an artificial intelligence model to determine whether the user has sepsis or septic shock, and selects at least one biometric data among the plurality of biometric data based on the type of a measuring device that measured the plurality of biometric data, and analyzes only the at least one biometric data selected through the artificial intelligence model to predict whether the user has septic shock.

[0123] For example, the computing device (100) can define in advance a type of biometric data with high accuracy for each measuring device based on the biometric data collection performance of each measuring device in advance, and when a plurality of biometric data are collected from a specific measuring device, the computing device can selectively analyze only the type of biometric data with high accuracy among the plurality of biometric data collected from the specific measuring device, that is, the type of biometric data set in advance for the specific measuring device, to determine whether sepsis or septic shock occurs for the user, thereby deriving more accurate results.

[0124] In various embodiments, when a plurality of different types of biometric data are acquired from a user, the computing device (100) analyzes each of the plurality of biometric data through an artificial intelligence model to extract a plurality of result data, and collates the plurality of result data to determine whether sepsis or septic shock occurs in the user. However, based on the type of the measuring device that measured the plurality of biometric data, a weight is assigned to each of the plurality of result data, and using the plurality of weighted result data, the computing device (100) can determine whether sepsis or septic shock occurs in the user. For example, the computing device (100) can assign a greater weight to result data derived from biometric data with high accuracy than to result data derived from biometric data with low accuracy based on the type of the measuring device that measured the plurality of biometric data, thereby deriving a more accurate result.

[0125] In various embodiments, the computing device (100) selects an artificial intelligence model corresponding to the user's situation at the time of measuring biometric data from among a plurality of pre-built artificial intelligence models, and analyzes the selected artificial intelligence model to analyze the user's biometric data, thereby determining whether septic shock has occurred in the user.

[0126] More specifically, first, the computing device (100) classifies multiple biometric data collected in different environments by situation (e.g., while driving, during daily life, during exercise, during meditation (stationary posture), etc.), and trains different models using each of the multiple biometric data classified by situation as learning data, thereby enabling the pre-construction of multiple artificial intelligence models corresponding to each of the multiple different situations.

[0127] At this time, the multiple artificial intelligence models may be independently fine-tuned models based on weights corresponding to each of multiple different situations, and the weights here may be determined based on the accuracy of situational biometric data collection.

[0128] Thereafter, when biometric data is collected from a user in a specific situation, the computing device (100) analyzes the biometric data by adopting an artificial intelligence model corresponding to the specific situation among a plurality of artificial intelligence models to determine whether the user has developed sepsis or septic shock, thereby deriving more accurate results. For example, when biometric data is collected while the user is driving, the biometric data collected while the user is driving is used as learning data, and the artificial intelligence model learned to analyze the user's biometric data, thereby determining more accurately whether the user is driving and has developed sepsis or septic shock.

[0129] In various embodiments, the computing device (100) may provide a notification to the user corresponding to the occurrence of septic shock when the occurrence of septic shock is predicted for the user as a result of performing biometric data analysis on the user.

[0130] At this time, the properties of the notification (e.g., size, type, cycle, number of times, scope of provision, etc.) may be determined based on the predicted likelihood of occurrence of septic shock for the user.

[0131] For example, the computing device (100) may calculate the probability of occurrence of septic shock to the user based on the result data derived by analyzing biometric data through an artificial intelligence model, and may classify the user into a normal level, a caution level, or a risk level according to the calculated probability. If the user is a normal level, a separate notification may not be provided, and if the user is a caution level, a first size notification may be provided to the user every first cycle, and if the user is a risk level, a second size notification larger than the first size may be provided to the user as well as a guardian (e.g., a family member, a medical professional in charge, etc.) registered in advance for the user every second cycle, which is shorter than the first cycle.

[0132] As another example, the computing device (100) may not provide a separate notification if the user is classified as a normal level by analyzing the biometric data of the user while driving through an artificial intelligence model, may provide a notification to the user's mobile phone if the user is classified as a caution level, and may provide a notification to the user's mobile phone, the vehicle's infotainment system, and the mobile phones of passengers riding in the vehicle if the user is classified as a dangerous level.

[0133] In various embodiments, the computing device (100) can generate a result report corresponding to the results derived through the first artificial intelligence model (10), the second artificial intelligence model (20), and the third artificial intelligence model (30) through the fourth artificial intelligence model (40), and provide the same to the user.

[0134] For example, as illustrated in FIG. 7, the fourth artificial intelligence model (40) may include an explainable artificial intelligence model (XAI) (41) and a generative artificial intelligence model (42), and the computing device (100) may use the explainable artificial intelligence model (41) to derive basis data for the result of early prediction of the occurrence of septic shock at a future point in time, may use the generative artificial intelligence model (42) to derive response plan data for the result of early prediction of the occurrence of septic shock at a future point in time, and may generate and provide a result report including the result of early prediction of the occurrence of septic shock at a future point in time, basis data, and response plan data to the user.

[0135]

[0136] The aforementioned method for early prediction of septic shock through artificial intelligence-based biometric data analysis has been described with reference to the flowchart illustrated in the drawings. For simplicity, the method for early prediction of septic shock through artificial intelligence-based biometric data analysis has been illustrated and described as a series of blocks. However, the present invention is not limited to the order of the blocks, and some blocks may be performed in a different order or simultaneously than those illustrated and described herein. Furthermore, new blocks not described in the present specification and drawings may be added, or some blocks may be deleted or modified.

[0137]

[0138] While the embodiments of the present invention have been described above with reference to the attached drawings, those skilled in the art will appreciate that the present invention can be implemented in other specific forms without altering the technical concept or essential features thereof. Therefore, the embodiments described above should be understood to be illustrative in all respects and not restrictive.

Claims

1. In a method performed by a computing device, A step of collecting time series-based biometric data measured in real time from the user; and A step of predicting the occurrence of septic shock for the user by analyzing the collected biometric data based on an artificial intelligence model, The step of predicting the occurrence of the above septic shock is: A method for predicting the occurrence of sepsis in a user by analyzing the collected biometric data, comprising: analyzing biometric data at a future point in time predicted based on the collected biometric data; and predicting the occurrence of septic shock at the future point in time. An early prediction method for septic shock using artificial intelligence-based biometric data analysis.

2. In paragraph 1, The step of early prediction of the occurrence of septic shock at the above future point in time is: A step of analyzing biometric data collected from the user at a first point in time to determine whether sepsis has occurred in the user at the first point in time; When it is determined that the above sepsis has occurred, a step of predicting the bio-data at a second time point after a predetermined period of time has passed from the first time point based on the bio-data collected at the first time point; and A step of analyzing the biometric data at the predicted second time point to determine whether septic shock occurs in the user at the second time point, An early prediction method for septic shock using artificial intelligence-based biometric data analysis.

3. In paragraph 2, The step of determining whether sepsis has occurred in the user at the first point in time is as follows: A step of determining whether sepsis occurs in the user at the first time point by analyzing the biometric data collected at the first time point using the first artificial intelligence model, The above first artificial intelligence model is, A classification model (Classifier) ​​based on machine learning or deep learning that is trained using multiple labeled biometric data with information on the occurrence of sepsis as learning data. An early prediction method for septic shock using artificial intelligence-based biometric data analysis.

4. In paragraph 2, The step of predicting the biometric data at the second point in time is: A step of predicting biometric data at the second point in time by analyzing biometric data collected at the first point in time using a second artificial intelligence model, The above second artificial intelligence model is, It is characterized by a deep learning-based prediction model (Forecaster) trained using multiple bio-data collected from sepsis patients as learning data. An early prediction method for septic shock using artificial intelligence-based biometric data analysis.

5. In paragraph 2, The step of determining whether septic shock occurs in the user at the second point in time is as follows: A step of analyzing the predicted biometric data at the second time point using a third artificial intelligence model to determine whether septic shock occurs in the user at the second time point, The above third artificial intelligence model is, A classification model (Classifier) ​​based on machine learning or deep learning that is trained using multiple labeled bio-data with information on the occurrence of septic shock as learning data. An early prediction method for septic shock using artificial intelligence-based biometric data analysis.

6. In paragraph 5, The above learning data is, A device comprising a plurality of bio-data labeled with the occurrence of septic shock at a location corresponding to the time at which septic shock occurred, characterized in that the time at which septic shock occurred is the time at which a vasopressor is administered. An early prediction method for septic shock using artificial intelligence-based biometric data analysis.

7. In paragraph 1, The above artificial intelligence model is, It includes a plurality of artificial intelligence models generated by classifying biometric data collected from a plurality of measuring devices with different properties by measuring device and learning each of the biometric data classified by measuring device as learning data. The above multiple artificial intelligence models are: A model that is independently fine-tuned based on weights corresponding to each of the above plurality of measuring devices, The weight corresponding to each of the above plurality of measuring devices is, It is determined based on the accuracy of sensor data measured through each of the above multiple measuring devices. An early prediction method for septic shock using artificial intelligence-based biometric data analysis.

8. In paragraph 1, The above collected biometric data is, Contains multiple types of biometric data, The step of predicting the occurrence of the above septic shock is: A method for predicting the occurrence of septic shock for a user further comprises selecting at least one biometric data among the plurality of biometric data based on the type of the measuring device that measured the plurality of biometric data, and analyzing only the selected at least one biometric data through the artificial intelligence model. An early prediction method for septic shock using artificial intelligence-based biometric data analysis.

9. In paragraph 1, The above collected biometric data is, Contains multiple types of biometric data, The step of predicting the occurrence of the above septic shock is: A method for predicting the occurrence of septic shock for the user, the method comprising: extracting a plurality of result data by individually analyzing the plurality of bio-data through the artificial intelligence model, and using the plurality of extracted result data to predict the occurrence of septic shock for the user, wherein a weight is assigned to each of the plurality of extracted result data based on the type of the measuring device that measured the plurality of bio-data, and using the plurality of weighted result data to predict the occurrence of septic shock for the user. An early prediction method for septic shock using artificial intelligence-based biometric data analysis.

10. In paragraph 1, The above artificial intelligence model is, It includes multiple artificial intelligence models generated by classifying multiple biometric data collected in different environments by situation and learning each of the multiple biometric data classified by situation as learning data. The step of predicting the occurrence of the above septic shock is: In the case where biometric data is collected from the user in a specific situation, a step of predicting the occurrence of septic shock for the user by analyzing the biometric data collected in the specific situation by adopting an artificial intelligence model trained using the biometric data collected in the specific situation as learning data among the plurality of artificial intelligence models is further included. An early prediction method for septic shock using artificial intelligence-based biometric data analysis.

11. In paragraph 1, The above artificial intelligence model is, Includes explainable AI (XAI) models and generative AI models. A step of deriving basis data for the results of early prediction of the occurrence of septic shock at the future point in time using the above explainable artificial intelligence model; A step of deriving response plan data for the results of early prediction of the occurrence of septic shock at a future point in time using the generative artificial intelligence model; and As a result of predicting the occurrence of septic shock at the future point in time, a result report including the derived basis data and the derived response plan data is generated, and a step of providing the generated result report is further included. An early prediction method for septic shock using artificial intelligence-based biometric data analysis.

12. Processor; network interface; memory; and A computer program loaded into the above memory and executed by the above processor, The above computer program, Instructions to collect time-series based biometric data measured in real time from the user; An instruction for predicting the occurrence of septic shock in the user by analyzing the collected biometric data based on an artificial intelligence model, The instructions for predicting the occurrence of the above septic shock are: An instruction for analyzing the biometric data collected above and detecting the occurrence of sepsis for the user, and predicting the occurrence of septic shock at a future time by analyzing the biometric data at a future time predicted based on the collected biometric data. A computing device that performs an early prediction method for septic shock through artificial intelligence-based bio-data analysis.

13. Combined with a computing device, A step of collecting time series-based biometric data measured in real time from the user; and A step of predicting the occurrence of septic shock for the user by analyzing the collected biometric data based on an artificial intelligence model, The step of predicting the occurrence of the above septic shock is: A computer program stored in a recording medium readable by a computing device to execute a method for early prediction of septic shock through artificial intelligence-based bio-data analysis, including a step of analyzing bio-data at a future point in time predicted based on the collected bio-data and early predicting the occurrence of septic shock at the future point in time when the occurrence of sepsis is detected for the user by analyzing the collected bio-data.

Images