Simulation method for predicting future health based on human digital twin integrating omics data and biometric data

Abstract

A method of operating an electronic device is disclosed. The method of operation includes the steps of: generating omics data of a subject by integrating a plurality of genetic elements constituting a subject; obtaining biometric data of a subject from a wearable device worn by the subject; and inputting information about at least one subject event into a prediction model trained based on the omics data and biometric data to predict the biometric data of the subject at at least one future point in time when the subject event occurs.

Description

Technology Field

[0001] The present disclosure relates to an electronic device or system for predicting a user's future health, and more specifically, to a technology for constructing a human digital twin by collecting a user's biometric data and providing simulation results. Background Technology

[0002] Advancements in 21st-century medical technology have expanded the horizons of our understanding of human life phenomena. In particular, the convergence of digital technology and medicine has given rise to an innovative paradigm known as the Human Digital Twin (HDT), which embodies individual human biological information in a virtual space. This presents new possibilities for personalized healthcare and signifies a fundamental transformation of medical services. While existing medical systems focused on treatment after a disease occurred, super-intelligent HDTs emphasize prevention before disease develops and the optimization of an individual's health potential. This implies a fundamental shift in the medical paradigm from a reactive model to a predictive and preventive model.

[0003] Since the late 2010s, the rapid advancement of wearable technology has marked a new turning point for HDT. Devices such as smartwatches, fitness trackers, and continuous glucose monitors have enabled the continuous collection of health data during daily life. This signified a paradigm shift from incremental measurement to continuous monitoring. Wearable devices have made it possible to incorporate data such as heart rate and activity levels into HDT, allowing for technologies that track health status in real time. However, current HDT systems have limitations in dynamic prediction because their predictions are relatively static and they fail to adequately reflect the real-time changes in the human body and external environment. Furthermore, genome-based systems rely on genetic information that is inherently immutable, making it difficult to predict changes in health status in response to environmental shifts. There is also the issue that, due to the complex nature of diseases, disease progression is influenced by both genetic and environmental factors, which can vary from patient to patient and over time.

[0004] Therefore, there is an urgent need to develop technology that detects changes in external environmental factors as well as individual biometric data in real time, and dynamically reflects their impact on individual health to predict and simulate the future. Prior art literature

[0005] Published Patent Application No. 10-2024-0160084 The problem to be solved

[0006] The present disclosure provides a method of operation for an electronic device that integrally collects and analyzes real-time biometric data (heart rate, activity level, sleep pattern, blood glucose, skin temperature, blood oxygen saturation, etc.) collected from wearable devices and IoT sensors, and various environmental factors (ultraviolet rays, air pollutants, noise, atmospheric pressure, temperature, etc.), and predicts and dynamically simulates the multilayered effects of taking specific medication or changes in lifestyle habits (diet, exercise, sleep, stress management, etc.) on future health conditions based on an individual's integrated omics map and real-time changing biometric data and environmental factors.

[0007] Furthermore, the present disclosure provides a method of operation for an electronic device that generates optimal health management advice for a user based on simulation results, understands an individual's health based on the interaction between an individual's genetic predisposition and environmental factors (Gene × Environment Interaction, G×E), and reflects this in a simulation model.

[0008] The purposes of the present disclosure are not limited to those mentioned above, and other purposes and advantages of the present disclosure not mentioned may be understood from the following description and will be more clearly understood from the embodiments of the present disclosure. Furthermore, it will be readily apparent that the purposes and advantages of the present disclosure can be realized by the means and combinations thereof set forth in the claims. means of solving the problem

[0009] A method of operating an electronic device according to one embodiment of the present disclosure comprises: a step of generating omics data of a subject by integrating a plurality of genetic elements constituting the subject; a step of obtaining biometric data of the subject from a wearable device worn by the subject; and a step of inputting information regarding at least one subject event into a prediction model trained based on a digital twin of the subject including the omics data and the biometric data, thereby predicting the biometric data of the subject at at least one future point in time when the subject event occurs.

[0010] The method of operating the electronic device may include the step of constructing a basic model for predicting sequential biometric data based on omics data, based on omics data and biometric data of each of a plurality of subjects, and the step of obtaining the prediction model by performing fine-tuning on the basic model based on the omics data of the subject and the biometric data of the subject confirmed at a plurality of time points.

[0011] In this case, the method of operation of the electronic device may further include the step of training the prediction model based on information regarding a plurality of events that occurred to the subject and the subject's biometric data confirmed before and after each of the plurality of events.

[0012] In addition, the method of operating the electronic device may further include the step of acquiring environmental data from an IoT (Internet of Things) sensor related to the environment in which the subject is located. At this time, the step of training the prediction model may train the prediction model based on information regarding a plurality of events that occurred to the subject, the subject's biometric data confirmed before and after each of the plurality of events, and environmental data confirmed at the time when each of the plurality of events occurred.

[0013] The method of operation of the electronic device may include the steps of inputting information regarding each of a plurality of target events into the prediction model to predict the subject's biometric data according to each of the plurality of target events, and selecting at least one recommended event among the plurality of target events that is helpful for the subject's health management based on the biometric data of the subject predicted for each of the plurality of target events.

[0014] The above target events may include at least one of taking specific medication, changing diet, an exercise program, changing sleep duration, and stress management.

[0015] The method of operation of the electronic device described above may input the layer-specific output value of the prediction model, identified during the process of predicting biometric data according to the prediction model composed of multiple layers, into an explanatory model to obtain explanatory information regarding at least one major item among the multiple items constituting the omics data that has an influence of a certain level or more on the predicted biometric data. Effects of the invention

[0016] The method of operation of an electronic device according to the present disclosure has the advantage of providing optimal health management advice to a user based on simulation results and enabling preventive intervention before the onset of disease, thereby accelerating the transition to a predictive and preventive medical paradigm.

[0017] The method of operation of an electronic device according to the present disclosure integrates real-time biological data and environmental factors and performs dynamic simulations based on an individual's integrated omics map, thereby enabling the prediction of the multilayered effects of a specific intervention on future health with significantly high accuracy.

[0018] The method of operation of an electronic device according to the present disclosure can deeply understand the impact of complex interactions between genes and the environment (G×E) on individual health through real-time detection and reflection of environmental factors, and provide more comprehensive health management based thereon.

[0019] The method of operation of an electronic device according to the present disclosure helps patients clearly understand their health condition and actively participate in health management by providing complex simulation results to the user in an easy and intuitive form.

[0020] The method of operation of an electronic device according to the present disclosure can improve the overall efficiency and quality of the medical system by supporting medical staff in understanding a patient's complex health condition in a multi-layered way and making more reliable decisions by utilizing simulation-based predictive information.

[0021] The method of operation of an electronic device according to the present disclosure is applicable to various medical and healthcare fields, such as early diagnosis of diseases, prediction of drug response in new drug development, management of chronic diseases, and design of personalized health insurance products, in addition to future health prediction. Brief explanation of the drawing

[0022] FIG. 1 is a diagram illustrating the operation of an electronic device that collects data based on a user's wearable device and IoT sensors and predicts the user's future health based on a user's HDT (Human Digital Twin) according to one embodiment of the present disclosure. FIG. 2 is a block diagram for explaining the configuration of an electronic device according to one embodiment of the present disclosure, FIG. 3 is a flowchart illustrating the process of constructing a prediction model for predicting changes in a subject's biometric data according to one embodiment of the present disclosure, FIG. 4 is a diagram illustrating the operation of an electronic device according to an embodiment of the present disclosure predicting changes in a digital twin including a subject's biometric data through a prediction model, and FIG. 5 is a diagram illustrating the process of an electronic device according to one embodiment of the present disclosure predicting future health while managing the HDT of each of a plurality of subjects. Specific details for implementing the invention

[0023] Before specifically describing the present disclosure, the method of description in the specification and drawings is described.

[0024] First, the terms used in this specification and claims have been selected based on general terms considering their functions in the various embodiments of this disclosure. However, these terms may vary depending on the intent of those skilled in the art, legal or technical interpretations, and the emergence of new technologies. Additionally, some terms have been arbitrarily selected by the applicant. Such terms may be interpreted according to the meanings defined in this specification; in the absence of specific definitions, they may be interpreted based on the overall content of this specification and common technical knowledge in the relevant field.

[0025] In addition, the same reference numbers or symbols described in each drawing attached to this specification represent parts or components that perform substantially the same function. For convenience of explanation and understanding, the same reference numbers or symbols are used to describe different embodiments. That is, even if components having the same reference number are all depicted in multiple drawings, the multiple drawings do not imply a single embodiment.

[0026] Additionally, in this specification and claims, terms including ordinal numbers, such as "first," "second," etc., may be used to distinguish between components. These ordinal numbers are used to distinguish identical or similar components from one another, and the meaning of the terms should not be limited by the use of such ordinal numbers. For example, the order of use or arrangement of components combined with such ordinal numbers should not be restricted by the number. If necessary, each ordinal number may be used interchangeably.

[0027] In this specification, singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "consisting of" are intended to specify the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0028] In the embodiments of the present disclosure, terms such as "module," "unit," "part," etc. are used to refer to a component that performs at least one function or operation, and such component may be implemented in hardware or software, or in a combination of hardware and software. Additionally, a plurality of "modules," "units," "parts," etc. may be integrated into at least one module or chip and implemented as at least one processor, except where each needs to be implemented in specific individual hardware.

[0029] Furthermore, in the embodiments of the present disclosure, when a part is described as being connected to another part, this includes not only a direct connection but also an indirect connection through another medium. Additionally, the meaning that a part includes a certain component implies that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0030] FIG. 1 is a diagram illustrating the operation of an electronic device that collects data based on a user's wearable device and IoT sensors according to one embodiment of the present disclosure and predicts the user's future health based on a user's HDT (Human Digital Twin).

[0031] Referring to FIG. 1, the electronic device (100) can create and manage a digital twin (10) of a subject (1).

[0032] The subject (1) may be a human (user) of various age groups, and the digital twin (10) is a digital entity that mimics the health status of the subject (1) in real time, constructed based on the subject's (1) omics data or biometric data. The digital twin (10) can be used as a tool to predict the subject's (1) future health or growth.

[0033] Referring to FIG. 1, the electronic device (100) can communicate with at least one wearable device (200) worn by a subject (1) to measure the subject (1)'s biometric data, at least one IoT sensor (300) to collect environmental data related to the environment where the subject (1) is located, and the subject (1)'s user terminal (400).

[0034] The wearable device (200) can be implemented in various forms such as a smart watch, smart ring, blood glucose meter (e.g., CGM), fitness tracker, HMD (Head Mounted Device), smart glasses, etc., and may include various biosensors (e.g., PPG sensor, ECG sensor, infrared sensor, blood glucose sensor, camera, etc.) for identifying heart rate, electrocardiogram, oxygen saturation, body temperature, blood glucose, etc. Additionally, the wearable device (200) may include a sleep analysis module for analyzing sleep patterns based on sensing data from the biosensors. Additionally, the wearable device (200) may identify and analyze the activity level of the subject (1) through a motion sensor. Additionally, the wearable device (200) may include a microphone sensor for measuring ambient noise and may calculate a stress level based on ambient noise and sleep patterns.

[0035] The IoT sensor (300) may be a temperature sensor, humidity sensor, air quality meter, UV sensor, noise sensor, barometer, etc., for measuring various environmental factors such as air quality (ozone, nitrogen dioxide, etc.), temperature, humidity, ultraviolet rays, noise, and pressure, but is not limited thereto.

[0036] The user terminal (400) may be a variety of terminal devices such as a smartphone, tablet PC, wearable device, or desktop PC, and it goes without saying that the user terminal of not only the subject but also a guardian (e.g., the subject's parents, medical personnel, guardians, attending physician, etc.) can be connected to the electronic device (100).

[0037] The electronic device (100) can model the growth or aging of the digital twin (10) by updating the digital twin (10) at regular intervals based on biometric data or environmental data confirmed in real time through a wearable device (200) or an IoT sensor (300).

[0038] Additionally, the electronic device (100) can train at least one prediction model based on the update history of the digital twin (10). The prediction model corresponds to a model for predicting the biometric data of the subject (1) at a future point in time of the digital twin (10).

[0039] Specifically, the electronic device (100) can obtain biometric data at a future point in time of the digital twin (10) predicted as the target event occurs by inputting information about at least one target event into a prediction model.

[0040] The event may include, but is not limited to, at least one of medication use, dietary changes, exercise programs, changes in sleep duration, and stress management.

[0041] For example, when a subject (1) inputs a plan for at least one event, such as exercise, game, study, or sleep, through a user terminal (400), the electronic device (100) can present a prediction result of biometric data according to the planned event in advance.

[0042] Alternatively, the electronic device (100) may recommend information about at least one event that is beneficial to the health of the subject (1) through the user terminal (400) by predicting biometric data based on the digital twin (10) based on information about various events (that have not yet occurred).

[0043] This will be explained in more detail through the drawings below.

[0044] FIG. 2 is a block diagram for explaining the configuration of an electronic device according to one embodiment of the present disclosure.

[0045] Referring to FIG. 2, the electronic device (100) may include a memory (110), a communication interface (120), a processor (130), etc. The electronic device (100) may be implemented as a service server that communicates with a guardian's terminal matched to each of a plurality of subjects to provide services (such as predicting the subject's growth / aging, predicting health status based on events, etc.), and may be implemented as a database management server, an AI server, etc. composed of at least one computer, but is not limited thereto.

[0046] The memory (110) is configured to store at least one instruction or data related to an operating system (OS) for controlling the overall operation of the components of the electronic device (100) and the components of the electronic device (100).

[0047] The memory (110) may include non-volatile memory such as ROM or flash memory, and may include volatile memory such as DRAM. Additionally, the memory may include auxiliary storage devices such as a hard disk or SSD (Solid State Drive).

[0048] Referring to FIG. 2, the memory (110) may include a digital twin (10) and a prediction model (111) that are matched to at least one subject.

[0049] The digital twin (10) may be constructed including the subject's (1) omics data and biometric data.

[0050] Omics data integrates multiple genetic elements that constitute a subject and may include information on various items, such as the genome, epigenome, transcriptome, proteome, metabolome, and microbiome. In particular, the epigenome (DNA methylation, histone modification, non-coding RNA) serves as a key link connecting the genome and proteome and reflecting environmental factors, which can help track and model in real time the dynamic effects of various environmental factors—such as an individual's lifestyle, diet, stress, exposure to toxic substances, infections, and age—on gene expression and protein function.

[0051] Biometric data may include height, body weight, and specific gravity of body components, as well as blood sugar, heart rate, electrocardiogram, oxygen saturation, body temperature, sleep patterns, stress levels, etc. Biometric data may be measured for a certain period of time (a certain number of times) or more by a medical institution such as a hospital or by a wearable device worn by the subject (1).

[0052] The electronic device (100) can generate a digital twin (10) of the subject (1) by collecting and integrating the aforementioned omics data, biometric data, etc. from a wearable device (200), an IoT sensor (300), a hospital server, etc.

[0053] The prediction model (111) is a model for predicting the growth / aging and health of the digital twin (10).

[0054] The prediction model (111) can be customized to predict changes in the biometric data of the subject (1) or the subject (1) based on the biometric data and environmental data of the subject (1) confirmed by the wearable device (200) and IoT sensor (300) at multiple points in time.

[0055] The prediction model (111) can predict the item-specific values ​​of the subject (1)'s biometric data for at least one future point in time based on the self-data constituting the digital twin (10), as well as predict changes in the subject (1)'s biometric data that occur as at least one event occurs. To this end, the prediction model (111) can be trained according to the history of occurrence of various events and the history of changes in the biometric data resulting therefrom.

[0056] The process of building or learning the above-described prediction model (111) can be performed on an electronic device (100), but it is also possible to perform it on an AI server provided separately from the electronic device (100).

[0057] The basic model or prediction model (111) can be implemented as a time-series-based prediction-capable RNN (Recurrent Neural Network), LSTM (Long Short-Term Memory), TCN (Temporal Convolutional Networks), a Transformer-based model, VAE (Variational Autoencoder), a Graph Neural Network (Graph Neural Networks, Graph Attention Networks, Graph Convolutional Networks), etc. Alternatively, the prediction model (111) can be implemented as a multimodal model that predicts by integrating the outputs of each of different types of models, but is not limited thereto.

[0058] As an additional embodiment, the prediction model (111) may learn the interaction between omics data and biometric data through ensemble learning techniques (e.g., a combination of Random Forest and LSTM) or dynamically select optimal recommendation events through reinforcement learning. Additionally, simulation results may be provided to the user through 3D visualization or an interactive dashboard.

[0059] The communication interface (120) can be connected to an external server and / or terminal device through one or more networks, and can exchange data through various wired and wireless communication methods.

[0060] Wireless communication may include at least one of the following communication methods: LTE (long-term evolution), LTE-A (LTE Advance), 5G (5th Generation) mobile communication, CDMA (code division multiple access), WCDMA (wideband CDMA), UMTS (universal mobile telecommunications system), WiBro (Wireless Broadband), GSM (Global System for Mobile Communications), DMA (Time Division Multiple Access), WiFi (Wi-Fi), WiFi Direct, Bluetooth, NFC (near field communication), Zigbee, etc.

[0061] Wired communication may include at least one of communication methods such as Ethernet, optical network, USB (Universal Serial Bus), Thunderbolt, and HDMI (High Definition Multimedia Interface).

[0062] Meanwhile, communication methods are not limited to the examples described above and may include new communication methods that emerge with technological advancements.

[0063] The electronic device (100) can periodically collect bio / environmental data of a subject and provide services by communicating with a wearable device (200), IoT sensor (300), and user terminal (400) of various subjects (1) through a communication interface (120).

[0064] Meanwhile, unlike FIG. 2, at least one of the digital twin (10) and the prediction model (111) may be stored in a separate database or a separate AI server rather than in the electronic device (100). In this case, the electronic device (100) may communicate with the aforementioned separate database or AI server through a communication interface (120).

[0065] The processor (130) is connected to the memory (110) and can control the electronic device (100) by executing at least one instruction stored in the memory (110).

[0066] To this end, the processor (130) may be implemented as a general-purpose processor such as a CPU (Central Processing Unit) or AP (Application Processor), a graphics-dedicated processor such as a GPU (Graphic Processing Unit) or VPU (Vision Processing Unit), or an artificial intelligence-dedicated processor such as an NPU (Neural Processing Unit). The processor may include volatile memory such as SRAM.

[0067] When a method according to one embodiment of the present disclosure includes a plurality of operations, the plurality of operations may be performed by a single processor or by a plurality of processors included in an electronic device (100). For example, when a first operation, a second operation, and a third operation are performed by a method according to one embodiment, the first operation, the second operation, and the third operation may all be performed by a first processor, or the first operation and the second operation may be performed by a first processor (e.g., a general-purpose processor) and the third operation may be performed by a second processor (e.g., an artificial intelligence dedicated processor).

[0068] One or more processors may be implemented as a single-core processor comprising one core, or as one or more multicore processors comprising multiple cores (e.g., homogeneous multicore or heterogeneous multicore). When one or more processors are implemented as multicore processors, each of the multiple cores included in the multicore processor may include internal processor memory such as on-chip memory, and a common cache shared by multiple cores may be included in the multicore processor. Additionally, each of the multiple cores included in the multicore processor (or some of the multiple cores) may independently read and execute program instructions for implementing a method according to one embodiment of the present disclosure, or all (or some) of the multiple cores may be linked together to read and execute program instructions for implementing a method according to one embodiment of the present disclosure.

[0069] When a method according to one embodiment of the present disclosure includes a plurality of operations, the plurality of operations may be performed by one of the plurality of cores included in a multi-core processor, or may be performed by a plurality of cores. For example, when a first operation, a second operation, and a third operation are performed by a method according to one embodiment, the first operation, the second operation, and the third operation may all be performed by a first core included in a multi-core processor, or the first operation and the second operation may be performed by a first core included in a multi-core processor and the third operation may be performed by a second core included in a multi-core processor.

[0070] In the embodiments of the present disclosure, a processor may mean a system-on-chip (SoC) in which one or more processors and other electronic components are integrated, a single-core processor, a multi-core processor, or a core included in a single-core processor or a multi-core processor, wherein the core may be implemented as a CPU, GPU, APU, MIC, DSP, NPU, hardware accelerator, or machine learning accelerator, but the embodiments of the present disclosure are not limited thereto.

[0071] Referring to FIG. 2, the processor (120) may include an omics construction module (131), a data collection module (132), a model learning module (133), a simulation module (134), an advice generation module (135), etc.

[0072] The omics construction module (131) is a module for constructing an omics map of a subject by collecting information on various items such as the subject's genome, epigenome, transcriptome, proteome, metabolome, and microbial genome. The omics construction module (131) can construct an omics map by collecting genetic information from a server of an institution, such as a hospital, related to the subject.

[0073] The data collection module (132) is configured to collect the subject's biometric data. The data collection module (132) can collect biometric data or environmental data at regular intervals from the wearable device (200).

[0074] As a result, a digital twin (10) of the subject with integrated omics data and biometric data can be constructed.

[0075] Additionally, the data collection module (132) can collect environmental data related to the subject through the IoT device (300).

[0076] In this regard, the electronic device (100) can update the digital twin (10) at regular intervals (e.g., 24 hours, 7 days, 30 days, 3 months, 6 months, 1 year, etc.) based on the aforementioned omics data and periodically received biometric data.

[0077] The model training module (133) is a module for training the prediction model (111). Specifically, the model training module (133) can train the prediction model based on omics data in which multiple genetic elements constituting the subject are integrated, and the subject's biological data.

[0078] In this regard, FIG. 3 is a flowchart for explaining the process of constructing a prediction model to predict changes in a subject's biological data according to one embodiment of the present disclosure.

[0079] Referring to FIG. 3, the model learning module (133) can build a basic model for predicting sequential biological data based on omics data and biological data of each of a plurality of subjects (S310). That is, the basic model can be trained based on statistical analysis of the influence of genetic elements constituting omics data on individual items constituting biological data, and can predict changes in biological data over time.

[0080] Additionally, the model learning module (133) can obtain a prediction model (111) customized for the subject (1) by performing fine-tuning on the basic model based on the subject (1)'s omics data and the subject (1)'s biometric data confirmed at multiple times. Of course, environmental data related to the subject (1) confirmed at multiple times may be additionally input during the fine-tuning process. That is, the prediction model (111) corresponds to a customized model capable of performing professional predictions only on the subject (1)'s digital twin (10).

[0081] Additionally, the model learning module (133) can train a prediction model (111) based on information about multiple events that occurred to the subject (1) and the subject's (1) biometric data confirmed before and after each of the multiple events. As a result, the prediction model (111) can be trained to predict changes in biometric data following the occurrence of events.

[0082] Additionally, the model learning module (133) can train a prediction model (111) based on information about multiple events that occurred to the subject (1), the subject's (1) biometric data confirmed before and after each of the multiple events, and environmental data confirmed at the time each of the multiple events occurred. As a result, the prediction model (111) can be trained to predict changes in biometric data according to the occurrence of events and environmental data.

[0083] The model learning module (133) can continuously learn and update the prediction model (111) by comparing the prediction result of the prediction model (111) with the actual biometric data of the subject (1) confirmed upon the arrival of the actual time.

[0084] The simulation module (134) is a module that performs a simulation to predict changes in the subject's (1) biological data through a prediction model (111). The simulation module (134) can model the subject's (1) digital twin according to the predicted biological data, and within the digital twin, complex physiological processes of the human body are represented by a mathematical model (e.g., Physiologically Based Pharmacokinetic (PBPK) model, glucose-insulin regulatory system model), and a sophisticated virtual entity can be constructed by reflecting the individual's genetic variations, metabolic characteristics, immune response patterns, etc.

[0085] In one embodiment, the simulation module (134) can perform real-time dynamic simulation by inputting the subject's (1) omics data, real-time biological data, real-time environmental data, etc., into the prediction model (111).

[0086] In addition, as an example, the simulation module (134) can perform a virtual intervention prediction to identify the multi-layered effects on an individual's health (e.g., weight change, improvement in blood sugar levels, change in blood pressure, change in cardiovascular disease risk, drug efficacy and side effects, etc.) when various virtual events occur, such as taking a specific drug, changing diet, an exercise program, optimizing sleep, and managing stress. In this case, the simulation module (134) can identify the effects of the event based on the bio-data predicted according to the output obtained by inputting information about the (virtual) event into the prediction model (111). For example, a scenario can be generated through the simulation in which "when a user inputs the event of '30 minutes of aerobic exercise every day,' the digital twin predicts through simulation that after 3 months, blood sugar levels will decrease by an average of 10 mg / dL and weight will decrease by 2 kg."

[0087] The advice generation module (135) is a module for recommending at least one event to the subject (1) or providing guidance information on whether at least one event is beneficial to health.

[0088] Specifically, information regarding each of the multiple target events is input into the prediction model (111) through the simulation module (134) to predict the biometric data of the subject (1) according to each of the multiple target events. At this time, the advice generation module (135) can select at least one recommended event among the multiple target events that is helpful for the subject's health management based on the biometric data of the subject predicted for each of the multiple target events. Specifically, a recommended event that causes the value of at least one item among the items constituting the biometric data to improve (e.g., high blood pressure -> normal range blood pressure) can be selected.

[0089] FIG. 4 is a diagram illustrating the operation of an electronic device according to one embodiment of the present disclosure predicting changes in a digital twin containing a subject's biometric data through a prediction model.

[0090] Referring to FIG. 4, the electronic device (100) can predict future biological changes based on a digital twin (10) at the present time that includes omics data (410) in which a plurality of genetic elements constituting the subject (1) are integrated, and biological data of the subject (1) measured in real time.

[0091] At this time, the omics construction module (131) generates omics data (410) and the data collection module (132) collects biometric data (420) to construct a digital twin (10), and the simulation module (134) inputs this into a prediction model (111) to predict the omics data (410') and biometric data (420') that constitute the future digital twin (10).

[0092] At this time, the electronic device (100) can predict the omics data (410') and / or biometric data (420') of the subject (1) that will change at a future time as the event occurs by inputting information about the current digital twin (10) and at least one event that may occur in the future into a prediction model (111). In the case of omics data, it is reflected that variable genetic data may also be included in addition to immutable genetic data.

[0093] Meanwhile, as the digital twin (10) is updated at a set interval (e.g., 24 hours, 7 days, 30 days, etc.), the electronic device (100) can compare the first biometric data predicted by the prediction model (111) according to the digital twin (10) prior to the update and the second biometric data predicted by the prediction model (111) according to the updated digital twin (10) for the same event.

[0094] In one embodiment, the electronic device (100) can identify at least one transitional event input into the prediction model (111) within a certain period (e.g., 1 hour, 24 hours, etc.) before the digital twin (10) is updated.

[0095] For example, each event may be input into the prediction model (111) according to the schedule of the event included in the plan confirmed through the user terminal (400), or may be input into the prediction model (111) according to the schedule of the event included in the inquiry received through the user terminal (400) (e.g., "How about running for 2 hours every day starting next week?"). In this case, if the time at which information about the event is input into the prediction model (111) in accordance with the specific schedule (date and time) of the event is within a certain period prior to the update, the event may be defined as a transitional event confirmed immediately before the update.

[0096] In this case, the electronic device (100) can acquire first biometric data predicted by the prediction model (111) according to the digital twin (10) prior to the update for the transitional event, and additionally generate second biometric data predicted through the updated digital twin (10).

[0097] As a result, whenever the digital twin (10) is updated, the electronic device (100) can obtain a transitional event, a first biometric data predicted according to the transitional event, and a second biometric data predicted according to the transitional event, respectively.

[0098] In this case, the electronic device (100) can identify the update sensitivity of the transitional event based on the first biometric data and the second biometric data identified for each transitional event.

[0099] Update sensitivity refers to the effect of the update of the digital twin (10) on identifying changes in biometric data due to the occurrence of a transitional event.

[0100] In one embodiment, the electronic device (100) may train at least one artificial intelligence model to define an update sensitivity for each individual transitional event based on training data including data reflected in the digital twin (10) prior to the update (e.g., omics data, biometric data, environmental data, etc.), data reflected in the digital twin (10) after the update (e.g., omics data, biometric data, environmental data, etc.), predicted first biometric data, and predicted second biometric data.

[0101] For example, in the case of a transitional event with high update sensitivity, the difference between the first biometric data and the second biometric data will be relatively large, whereas in the case of a transitional event with low update sensitivity, the difference between the first biometric data and the second biometric data will be relatively small.

[0102] Here, the artificial intelligence model for defining update sensitivity may be a model trained based on algorithms such as Random Forest, Support Vector Machine (SVM), Gradient Boosting, Multi-Layer Perceptron (MLP), Graph Neural Network (GNN), and Recurrent Neural Network (RNN), but is not limited thereto.

[0103] At this time, the electronic device (100) can identify a transitional event in which the update sensitivity exceeds a preset value as a sensitive event.

[0104] In cases where a prediction is performed based on a sensitive event, the biometric data predicted by the prediction model (111) can be provided to the user terminal (400), just as with a general event.

[0105] However, the electronic device (100) can determine the time interval by comparing the point in time when a sensitive event is input into the prediction model (111) with the nearest expected update time of the digital twin (10). Here, the expected update time of the digital twin (10) corresponds to the time when a certain period has elapsed since the digital twin (10) was last updated.

[0106] Here, if the time interval is less than a preset threshold value (e.g., one week, one month, six months, etc.), the electronic device (100) can predict second biometric data by inputting the updated digital twin (10) and major events into a prediction model (111) within a set time (e.g., 3 hours, 6 hours, 12 hours, etc.) from the time the digital twin (10) is updated. That is, in addition to the biometric data predicted before the update, biometric data predicted after the update can be additionally obtained.

[0107] The aforementioned threshold value, which is preset in relation to the time interval, can be set according to the update sensitivity confirmed for sensitive events. Specifically, the electronic device (100) can set the threshold value larger as the sensitivity increases.

[0108] Additionally, the electronic device (100) can compare biometric data before and after an update confirmed for a major event. If the difference between the biometric data before and after the update (e.g., difference by item of biometric data) satisfies a preset condition, the electronic device (100) can additionally provide information about the biometric data predicted after the update to the user terminal (400) in addition to the biometric data predicted before the update provided earlier.

[0109] For example, the preset conditions may include, but are not limited to, a condition in which the average of the differences between items of biometric data before and after an update is greater than a certain value, or a condition in which the comparison result of biometric data before and after an update results in a change to an extent that exceeds the preset allowable range for at least one item.

[0110] As a result, users can receive more accurate and complex prediction results for sensitive events that are heavily affected by updates to the digital twin (10).

[0111] In addition, as an additional embodiment, the electronic device (100) can identify the ratio in which the difference between the biometric data before and after the update satisfies a preset condition based on the predicted history (e.g., identified at regular intervals) of the biometric data before and after the update of the digital twin (10) for each major event.

[0112] At this time, the electronic device (100) can periodically update the aforementioned threshold value, which is preset in relation to the time interval, according to the ratio identified for each major event. At this time, if the ratio is greater than a certain value, the threshold value is increased, and if the ratio is less than a certain value, the threshold value is decreased, and the magnitude of the increase / decrease may be greater the greater the difference from the certain value.

[0113] Meanwhile, although not illustrated, the electronic device (100) may further include at least one explanatory model for explaining the prediction results of the prediction model (111). Specifically, the electronic device (100) inputs the output values ​​of each layer of the prediction model, which are identified during the process of predicting biometric data according to a prediction model composed of multiple layers, into the explanatory model, thereby obtaining explanatory information regarding at least one major item that has an influence of a certain level or more on the predicted biometric data among the multiple items constituting the omics data.

[0114] Likewise, the electronic device (100) can input the output value of each layer of the prediction model identified during the process of predicting biometric data into the explanation model to obtain explanation information for at least one major item among the items of compensation constituting environmental data that has an influence of a certain level or more on the predicted biometric data.

[0115] The explanation model is an explainable (XAI) model and may be implemented with explainable AI technologies such as SHAP (SHapley Additive exPlanations), LIME (Local Interpretable Model-agnostic Explanations), or attention mechanisms to provide biological interpretations of the prediction results of the AI ​​model, but is not limited thereto.

[0116] Meanwhile, FIG. 5 is a diagram illustrating the process of an electronic device according to one embodiment of the present disclosure managing the HDT of each of a plurality of subjects and predicting their future health.

[0117] Referring to FIG. 5, the electronic device (100) can construct individual digital twins (10-1, 2, 3, …) based on omics data and biometric data of each of the multiple subjects (501, 502, 503, …), and the health status of each subject can be predicted based on multiple prediction models fine-tuned for each individual subject.

[0118] Meanwhile, the various embodiments described above may be implemented by combining two or more embodiments, provided that they do not conflict or contradict each other.

[0119] Meanwhile, the various embodiments described above may be implemented in a recording medium readable by a computer or a similar device using software, hardware, or a combination thereof.

[0120] According to hardware implementation, the embodiments described in this disclosure may 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), processors, controllers, microcontrollers, microprocessors, and other electrical units for performing functions.

[0121] In some cases, the embodiments described herein may be implemented as the processor itself. In a software implementation, embodiments such as the procedures and functions described herein may be implemented as separate software modules. Each of the aforementioned software modules may perform one or more functions and operations described herein.

[0122] Meanwhile, computer instructions or computer programs for performing processing operations in electronic devices, etc., according to the various embodiments of the present disclosure described above may be stored in a non-transitory computer-readable medium. When such computer instructions or computer programs stored in the non-transitory computer-readable medium are executed by a processor of a specific device, the specific device described above performs processing operations in electronic devices, etc., according to the various embodiments described above.

[0123] A non-transient computer-readable medium refers to a medium that stores data semi-permanently and can be read by a device, unlike media that store data for a short period of time such as registers, caches, and memory. Specific examples of non-transient computer-readable media include CDs, DVDs, hard disks, Blu-ray discs, USBs, memory cards, and ROMs.

[0124] Although preferred embodiments of the present disclosure have been illustrated and described above, the present disclosure is not limited to the specific embodiments described above. It is understood that various modifications can be made by those skilled in the art without departing from the essence of the present disclosure as claimed in the claims, and such modifications should not be understood individually from the technical spirit or perspective of the present disclosure. Explanation of the symbols

[0125] 100: Electronic device 110: Memory 120: Communication interface 130: Processor 200: Wearable devices 300: IoT Sensor 400: User terminal

Claims

Claim 1 A method of operating an electronic device comprises: a step of generating omics data of a subject by integrating a plurality of genetic elements constituting the subject; a step of obtaining biometric data of the subject from a wearable device worn by the subject; and a step of inputting information regarding at least one subject event into a prediction model trained based on a digital twin of the subject including the omics data and the biometric data, thereby predicting the biometric data of the subject at at least one future point in time following the occurrence of the subject event; wherein the method of operating the electronic device comprises: a step of constructing a basic model for predicting sequential biometric data according to the omics data based on the omics data and biometric data of each of a plurality of subjects; and a step of obtaining the prediction model by performing fine-tuning on the basic model based on the omics data of the subject and the biometric data of the subject confirmed at a plurality of points in time. The method of operation of the electronic device further comprises the step of training the prediction model based on information regarding a plurality of events that occurred to the subject and biometric data of the subject confirmed before and after each of the plurality of events; and the method of operation of the electronic device further comprises the step of acquiring environmental data from an IoT (Internet of Things) sensor related to the environment in which the subject is located; and the step of training the prediction model comprises training the prediction model based on information regarding a plurality of events that occurred to the subject, biometric data of the subject confirmed before and after each of the plurality of events, and environmental data confirmed at the time when each of the plurality of events occurred; and the method of operation of the electronic device further comprises the step of updating the digital twin according to a certain period based on the biometric data and the environmental data. Claim 2 delete Claim 3 delete Claim 4 delete Claim 5 A method of operation of an electronic device according to claim 1, comprising: a step of inputting information for each of a plurality of target events into a prediction model to predict biometric data of the subject according to each of the plurality of target events; and a step of selecting at least one recommended event among the plurality of target events that is helpful for health management of the subject based on the biometric data of the subject predicted for each of the plurality of target events. Claim 6 A method of operation of an electronic device according to claim 1, wherein the target event comprises at least one of taking a specific drug, changing the diet, an exercise program, changing the sleep time, and stress management. Claim 7 In claim 1, the method of operation of the electronic device comprises inputting the layer-specific output value of the prediction model, identified during the process of predicting biometric data according to the prediction model composed of multiple layers, into an explanatory model to obtain explanatory information regarding at least one major item among the multiple items constituting the omics data that has an influence of a certain level or more on the predicted biometric data. Claim 8 An electronic device comprising: a memory storing at least one instruction; and a processor that executes the instruction to perform the method of operation of claim 1.

Images