KR20210098731A - The Method and Apparatus for Determining Dementia Risk Factors Using Deep Learning - Google Patents

Abstract

The present specification provides a method for allowing a server to determine dementia risk factors by using deep learning. According to the present invention, the method comprises the following steps: acquiring biometric information from each of subjects corresponding to a first comparative group through a wearable device; acquiring measurement information for each of the subjects corresponding to the first comparative group; extracting a first dementia risk factor on the basis of the biometric information and measurement information for each of the subjects; and extracting a second dementia risk factor related to the first dementia risk factor through deep learning performed on the basis of the biometric information related to the first dementia risk factor and comparative group information.

Description

The Method and Apparatus for Determining Dementia Risk Factors Using Deep Learning

The present specification relates to a method and apparatus for determining dementia risk factors using deep learning. Specifically, it relates to a method of deriving dementia risk factors using hospital measurement information and biometric information measured by a wearable device.

As social interest in dementia increases, research to identify the cause of dementia and prove dementia-related risk factors is increasing. Dementia may be a brain disease that accompanies aging. At this time, dementia may gradually cause memory impairment. In addition, depending on dementia, individual personality may change or cognitive ability may decrease.

Dementia can make daily life difficult, and even a normal person can develop dementia due to persistent damage to brain function. In order to prevent such dementia in advance, it may be necessary to analyze the causes of dementia and dementia-related risk factors.

In general, whether dementia progresses can be confirmed by a medical method such as a brain image or a biomarker. Also, as an example, symptoms of dementia may be alleviated or worsened according to physical activity such as regular exercise or regular sleep.

Considering the above, not only a medical method for confirming the symptoms of dementia, but also a method for preventing and treating dementia by identifying risk factors related to dementia may be needed. describe

An object of the present specification is to provide a method and apparatus for determining dementia risk factors using deep learning.

An object of the present specification is to provide a method of deriving dementia risk factors using hospital measurement information and biometric information measured by a wearable device.

An object of the present specification is to provide a method of deriving a new dementia risk factor using the dementia risk factor.

According to an embodiment of the present specification, it is possible to provide a method for the server to determine dementia risk factors using deep learning. In this case, the method of determining the dementia risk factors includes: obtaining biometric information from each of the subjects corresponding to the first comparison group through a wearable device; obtaining measurement information for each of the subjects corresponding to the first comparison group; , a step of deriving a first dementia risk factor based on biometric information and measurement information for each subject, a first dementia risk through deep learning performed based on biometric information related to the first dementia risk factor and information on a comparison group It may include deriving a second dementia risk factor related to the factor.

In addition, according to an embodiment of the present specification, it is possible to provide a server for determining dementia risk factors using deep learning. In this case, the server may include a transceiver for transmitting and receiving signals, a deep learning learning unit for performing deep learning learning, and a controller for controlling the transceiver and the deep learning learning unit. In this case, the controller obtains biometric information from each of the subjects corresponding to the first comparison group through the wearable device, obtains measurement information for each of the subjects corresponding to the first comparison group, and biometric information for each of the subjects and a first dementia risk factor based on the measurement information, and a second dementia risk factor related to the first dementia risk factor through deep learning performed based on biometric information and control group information related to the first dementia risk factor. can be derived

In addition, according to an embodiment of the present specification, it is possible to provide a system for determining dementia risk factors. In this case, the system may include a server that determines dementia risk factors through deep learning, a wearable device that measures biometric information, an IoT device that measures the wearable device and biometric information together, or an IoT device that transmits the measured biometric information to the server. . In this case, the server obtains biometric information from wearable devices of each of the subjects corresponding to the first comparison group, the server obtains measurement information for each of the subjects corresponding to the first comparison group, and the server obtains measurement information for each of the subjects corresponding to the first comparison group Derives a first dementia risk factor based on biometric information and measurement information for A second dementia risk factor can be derived.

In addition, the following matters may be commonly applied to the method, apparatus, server and system for deriving dementia risk factors.

According to an embodiment of the present specification, when deep learning for deriving a second dementia risk factor is performed, a second comparison group is further set, and biometric information related to the first dementia risk factor corresponds to the first comparison group includes biometric information obtained from each of the subjects and biometric information obtained from each of the subjects corresponding to the second comparison group, and deep learning compares the biometric information of the first comparison group and the biometric information of the second comparison group can be done through

Also, according to an exemplary embodiment of the present specification, the first comparison group may be a comparison group corresponding to a subjective cognitive impairment state, and the second comparison group may be a comparison group corresponding to a normal state.

In addition, according to an embodiment of the present specification, biometric information may be obtained by further using an IoT device.

Also, according to an embodiment of the present specification, measurement information may be derived based on at least one of brain image, CT, MRI, EEG, and clinical findings data.

In addition, according to an embodiment of the present specification, when the first dementia risk factor is derived based on the biometric information and the measurement information for each of the subjects, the biometric information and the measurement information are transmitted from the server to the medical institution, and the server The first dementia risk factor information determined by the medical institution may be received.

In addition, according to an embodiment of the present specification, when the first dementia risk factor is derived based on the biometric information and measurement information for each of the subjects, the server performs deep learning based on the biometric information and the measurement information, Based on deep learning, it can be obtained directly through the server.

The present specification may provide a method and apparatus for determining dementia risk factors using deep learning.

The present specification may provide a method of deriving dementia risk factors using hospital measurement information and biometric information measured by a wearable device.

The present specification may provide a method of deriving a new dementia risk factor using the dementia risk factor.

The effects obtainable in the present specification are not limited to the effects mentioned above, and other effects not mentioned may be clearly understood by those of ordinary skill in the art to which the present invention pertains from the description below. will be.

1 is a diagram illustrating a wearable device, a server, and an Internet of Things (IoT) device according to an embodiment of the present specification.
2 is a diagram illustrating a method of deriving dementia risk factors according to an embodiment of the present specification.
3 is a diagram illustrating a method for a wearable device to measure biometric information according to an embodiment of the present invention.
4 is a diagram illustrating a method of identifying dementia risk factors using biometric information according to an embodiment of the present invention.
5 is a diagram illustrating a method for an IoT device and a wearable device to measure biometric information according to an embodiment of the present invention.
6 is a diagram illustrating a method of performing deep learning according to an embodiment of the present invention.
7 is a diagram illustrating a method for performing deep learning according to an embodiment of the present invention.
8 is a diagram illustrating a method of performing deep learning according to an embodiment of the present invention.
9 is a diagram illustrating a method for performing deep learning according to an embodiment of the present invention.
10 is a flowchart of a method of deriving dementia risk factors according to an embodiment of the present invention.

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings. DETAILED DESCRIPTION The detailed description set forth below in conjunction with the appended drawings is intended to describe exemplary embodiments of the present invention and is not intended to represent the only embodiments in which the present invention may be practiced. The following detailed description includes specific details in order to provide a thorough understanding of the present invention. However, it will be apparent to one skilled in the art that the present invention may be practiced without these specific details.

The following embodiments combine elements and features of the present invention in a predetermined form. Each component or feature may be considered optional unless explicitly stated otherwise. Each component or feature may be implemented in a form that is not combined with other components or features. In addition, some components and/or features may be combined to constitute an embodiment of the present invention. The order of operations described in embodiments of the present invention may be changed. Some features or features of one embodiment may be included in another embodiment, or may be replaced with corresponding features or features of another embodiment.

Specific terms used in the following description are provided to help the understanding of the present invention, and the use of these specific terms may be changed to other forms without departing from the technical spirit of the present invention.

In some cases, well-known structures and devices are omitted or shown in block diagram form focusing on key functions of each structure and device in order to avoid obscuring the concept of the present invention. In addition, the same reference numerals are used to describe the same components throughout the present specification.

Also, in this specification, terms such as first and/or second may be used to describe various components, but the components should not be limited by the terms. The above terms are used only for the purpose of distinguishing one element from another element, for example, without departing from the scope of rights according to the concept of the present specification, a first element may be called a second element, and similarly , the second component may also be referred to as the first component.

In addition, throughout the specification, when a part "includes" a certain component, it means that other components may be further included, rather than excluding other components, unless otherwise stated. And terms such as “unit” and “unit” described in the specification mean a unit that processes at least one function or operation, which may be implemented by a combination of hardware and/or software.

1 is a diagram illustrating a wearable device, a server, and an Internet of Things (IoT) device according to an embodiment of the present specification. Referring to FIG. 1A , a wearable device 100 may include a transceiver 110 , a control unit 120 , and an input unit 130 . Also, as an example, the wearable device 100 may further include other components, and is not limited to the above-described embodiment.

For example, the transceiver 110 may be configured to transmit/receive a signal to/from another device. As an example, the other device may be a server (or center, 200 ). Also, the other device may be another device such as an IoT device or other smart phone. That is, the device with which the wearable device 100 can exchange signals may be another device, and is not limited to a specific device.

For example, the control unit 120 may be configured to control the transceiver 110 and the input unit 130 . In addition, the controller 120 may further control other components, and is not limited to the above-described embodiment.

Also, the input unit 130 may be configured to acquire a biosignal. For example, the input unit 130 may acquire the biosignal through at least one of a position sensor, an acceleration sensor, a motion sensor, and other sensors. That is, the input unit 130 may be configured to acquire or sense an external signal, and is not limited to the above-described embodiment.

In addition, the server (or center, 200 ) may include a transceiver 210 and a control unit 220 . In addition, the server 200 may further include a deep learning learning unit 230, and may further include other configurations.

For example, the transceiver 210 may be configured to transmit/receive a signal to/from another device. As an example, the other device may be the wearable device 100 . Also, the other device may be another device such as an IoT device or other smart phone. That is, the device with which the server 200 can exchange signals may be other devices, and is not limited to a specific device.

As an example, the controller 220 may be configured to control the transceiver 210 and the deep learning learner 230 . In addition, the controller 220 may further control other components, and is not limited to the above-described embodiment.

The deep learning learning unit 230 may perform deep learning through biometric information and other information obtained from the wearable device 100 and derive dementia risk factors based on the learning model. For example, dementia risk factors may be derived through biometric information and measurement information obtained from the wearable device 100 . In this case, as an example, the measurement information may be information measured through brain imaging, CT, MRI, EEG, etc., which will be described later.

Also, referring to FIG. 1B , the wearable device 100 may exchange signals with the IoT device 300 . That is, the wearable device 100 may communicate with the IoT device 300 . In this case, as an example, the IoT device 300 may include a configuration for performing communication. As an example, the IoT device 300 may include a transceiver 310 and a controller 320 . In this case, as an example, the transceiver 310 may communicate with the wearable device 100 or the server 200 . Also, as an example, the transceiver 310 may communicate with other devices, and the device type is not limited. Also, the control unit 320 may be a component that controls components in the IoT device 200 . As another example, the IoT device 200 may be a low-power device. The IoT device 200 may be a device that transmits only a specific signal to a low-power device, and is not limited to the above-described embodiment. Also, as an example, the IoT device 200 may be a device operating based on the IoT cloud. As an example, the IoT device 200 may be a device fixedly installed at a specific point. As a more specific example, the IoT device 200 may receive a signal from the wearable device 100 and transmit the data to a server or other device through the IoT cloud, but is not limited to the above-described embodiment. Also, as an example, the IoT device 200 may provide necessary information when the wearable device 100 measures a biosignal, which will be described later.

2 is a diagram illustrating a method of deriving dementia risk factors according to an embodiment of the present specification. Referring to FIG. 2 , dementia-related information may be obtained through various measurements.

For example, the measurement may be a brain image, CT, MRI, or EEG measurement. Also, as an example, measurement may mean measurement performed through amyloid PET. As another example, measurement may refer to measurement performed through a blood biomarker. As another example, measurement may refer to measurement performed by a doctor, such as a screening questionnaire or clinical findings.

That is, measurement may mean acquiring information on whether or not dementia occurs or whether dementia is likely to occur through various methods, and the information obtained through this may be measurement information. For example, the measurement information may be performed through a hospital or other medical facility, and may be information obtained through this. In this case, as an example, the degree of dementia may be measured according to the measurement information.

For example, the degree of dementia may be measured as a mild cognitive impairment stage according to the measurement information. Also, Alzheimer's dementia may be measured according to the measurement information. As another example, as a state in which dementia is in progress, the degree of dementia may be measured as a stage of subjective cognitive impairment. That is, through the above-described measurement, each stage according to the dementia progression state may be determined as the occurrence of dementia or the possibility of dementia. As an example, below, for convenience of explanation, the dementia progression stage is expressed by each number. In this case, for example, the second stage may be a more advanced dementia than the first stage, and the third stage may be a more advanced dementia than the second stage. However, the above-described steps are merely examples for convenience of description and are not limited to the above-described steps. As an example, the subjective cognitive impairment stage may be the first stage, and the mild cognitive impairment may be the second stage. However, this is only one example and is not limited thereto.

3 is a diagram illustrating a method for a wearable device to measure biometric information according to an embodiment of the present invention. Meanwhile, referring to FIG. 3 , a stage according to the degree of dementia may be determined based on the above-described measurement information. In this case, as an example, biometric information about people who are subjects of measurement information may be measured. For example, the biometric information may be various kinds of body information. As a more specific example, the body information may be at least one of consumed calories, active time, sleep time, sleep type for each sleep time, heart rate, step number, step distance, and number of exercise. In addition, the body information may be various information measured through other wearable devices, and is not limited to the above-described embodiment. That is, information obtained from a target person through the wearable device may be biometric information, and is not limited to the above-described embodiment.

4 is a diagram illustrating a method of identifying dementia risk factors using biometric information according to an embodiment of the present invention. In this case, as an example, the server (or center) may predict dementia risk factors for the occurrence of dementia using the biometric information obtained through the wearable device and the above-described measurement information. In this case, as an example, the dementia risk factor may be predicted by a person (e.g. a medical staff) based on biometric information and measurement information.

As another example, dementia risk factors may be derived through statistical information generated through biometric information and measurement information obtained from the server. Also, as an example, dementia risk factors may be derived through deep learning through biometric information and measurement information obtained from the server, and are not limited to the above-described embodiment. In this case, as an example, a new dementia risk factor may be derived through the deep learning method through the dementia risk factor derived as described above, which will be described later.

5 is a diagram illustrating a method of measuring biometric information by an IoT device and a wearable device according to an embodiment of the present invention.

For example, as described above, biometric information may be measured through the wearable device. However, as an example, biometric information measured using only the wearable device may have limitations. As an example, when sleep information is measured through the wearable device, information about the surrounding sleep environment may be further required. As an example, a case in which the ambient temperature around the sleep is high or low among the sleeping environment of the wearable device wearer may be considered. As a specific example, sleep information measured by the wearable device in an environment in which a high temperature and a constant temperature are maintained may be utilized differently from sleep information measured by the wearable device in an environment in which a low temperature and an irregular temperature are maintained.

As another example, since biometric information measured through the wearable device is sensed through an acceleration sensor, a motion sensor, and other sensors that detect body changes, there may be a limit to the measured information. As an example, there may be a limitation in obtaining information on whether the wearable device wearer regularly eats or leads a regular life.

In addition, there may be a limitation in acquiring various other information of the wearer. In consideration of the above, biometric information may be measured further using an IoT device. For example, referring to FIG. 5 , the wearable device 510 communicates with the IoT devices 520-1, 520-2, 520-3, 520-4, and 520-5 through a communication network, and from information can be obtained. In this case, as an example, the IoT devices 520-1, 520-2, 520-3, 520-4, and 520-5 may be devices in which different environmental conditions exist.

As an example, the first IoT device 520-1 may be a device installed indoors to obtain information on temperature, humidity, illuminance, and other indoor conditions in a room in which the user sleeps. As another example, the second IoT device 520 - 2 may be a device that obtains state information of a company or other space where the user works.

As an example, the state information may be air pollution degree, temperature, humidity, and other surrounding environment information. That is, the various IoT devices 520-1, 520-2, 520-3, 520-4, and 520-5 may each exist in different environments, and each device may measure different information based on different conditions. can As a specific example, the first IoT device 520-1 may be installed indoors and measure surrounding environment information while the user is sleeping. In this case, the wearable device 510 may acquire more specific biometric information by acquiring surrounding environment information from the first IoT device 520-1 while measuring information about the user's sleep. As another example, the second IoT device 520 - 2 may be installed at the user's work place to provide information on the user's work environment. For example, whether the user stays in the same position for a long time or does not move may be measured through the wearable device 510 . In this case, the second IoT device 520 - 2 may measure and provide information about the surrounding environment when information about the user is measured, and through this, more specific biometric information may be obtained.

As another example, each of the IoT devices 520-1, 520-2, 520-3, 520-4, and 520-5 may provide independent information. For example, an IoT device installed indoors may measure a user's work time and leave time through interworking with the wearable device 510 , and may provide the information to a server. That is, the IoT devices 520-1, 520-2, 520-3, 520-4, and 520-5 may measure information according to each condition and transmit it to the server.

Through the above bar, when deriving dementia risk factors, more detailed information can be obtained, and information for deriving new dementia risk factors can be secured.

6 is a diagram illustrating a method of performing deep learning according to an embodiment of the present invention. Deep learning may mean learning a deep neural network based on a learning model. As an example, referring to FIG. 6 , the deep neural network may include an input layer, multiple hidden layers, and an output layer. That is, the deep neural network may refer to an artificial neural network including multiple hidden layers.

In this case, the hidden layer may provide output information to the output layer based on information learned based on the input information. In addition, the hidden layer stores a plurality of pieces of information related to the input layer and the output layer, and matching data may be calculated based on the information.

In addition, the learning model can be continuously learned by storing the information of the input layer and the information of the output layer, and using this as data for learning. As a specific example, the learning method of deep learning may include supervised learning and unsupervised learning. In this case, as an example, supervised learning may be a model that is learned based on a predetermined output with respect to an input. In addition, unsupervised learning may be a method of determining an output layer matching an input layer based on various pattern information rather than a predetermined output.

For example, in the present invention, a learning model is configured based on at least one of supervised learning and unsupervised learning, and matching may be performed based on this.

Also, as an example, input information for deriving dementia risk factors may be biometric information and measurement information, which will be described later. Also, as an example, input information for deriving a new dementia risk factor may be information on existing dementia risk factors and related factor information, which will be described later.

Also, as an example, the hidden layer may include information for determining the similarity of input layer information. In this case, output information may be derived depending on whether there is a similarity, and a new dementia risk factor may be derived as described above.

7 is a diagram illustrating a method for performing deep learning according to an embodiment of the present invention. Referring to FIG. 7 , the server may include the above-described deep learning learning unit.

For example, the deep learning learning unit may operate based on a learning model, and the learning model may be a learning model using at least one of supervised learning and unsupervised learning as described above. In addition, the learning model may perform learning based on input information and continuously update through this. In this case, as an example, input information for deep learning may be measurement information and biometric information. For example, as described above, a person (e.g. a medical staff) may determine dementia risk factors based on the measurement information and biometric information and input them into the device.

As another example, dementia risk factors may be derived through deep learning based on measurement information and biometric information. For example, the measurement information may be at least any one or more of the above-described brain image, CT, MRI, EEG, and clinical findings data. That is, the measurement information may be measurement information through a hospital or other medical institution. In this case, as an example, the input information may be stage information on whether or not dementia occurs or the possibility of dementia based on the above-described measurement information. As another example, the measurement information may be input information as measured information itself. For example, the degree of dementia may be compared based on the measured information. Also, the input information may include biometric information. Thereafter, deep learning may be performed through the learning model based on the above-described input information.

In this case, the information output through deep learning may be dementia risk factor information. In this case, the dementia risk factor information may refer to factors affecting the occurrence of dementia. As an example, biometric information of people determined to have a high dementia risk group through measurement information may be learned through deep learning.

As a more specific example, the above-described information is compared through deep learning, and sleep time, step distance, and number of steps data may be determined as a risk factor. That is, as output information, dementia risk factor information may be derived as sleep time. That is, based on the measurement information and biometric information, information affecting the occurrence of dementia may be derived through deep learning as a dementia risk factor.

8 is a diagram illustrating a method of performing deep learning according to an embodiment of the present invention.

Also, as an example, referring to FIG. 8A , a new dementia risk factor may be derived through deep learning through the dementia risk factor. For example, in deep learning, input information may be dementia risk factors and biometric information related to dementia risk factors. As a more specific example, when sleep time is derived as a risk factor for dementia in FIG. 7, sleep-related biometric information such as a sleep pattern (REM sleep, deep sleep, light sleep) included in the sleep time is set as input information. can

That is, the input information may be dementia risk factors and biometric information related to dementia risk factors. In this case, as an example, new dementia risk factors may be derived based on the above-described dementia risk factors through a learning model of deep learning. For example, when sleep time is derived as a risk factor for dementia, REM sleep may be derived as a new risk factor for dementia related to dementia risk factors through deep learning. That is, new dementia risk factors can be derived from existing dementia risk factors.

Meanwhile, as a specific example, referring to FIG. 8B , in a learning model for deriving new dementia risk factors from existing dementia risk factors, comparison group information may be set to be compared.

As a specific example, the learning model may set information on the comparison group 1 having a high probability of dementia and the comparison group 2 having a low probability of dementia. Also, as an example, the learning model may set the comparison group through another method, and is not limited to the above-described embodiment.

In this case, the learning model may use dementia risk factors and biometric information as input information as input information for each comparison group. In this case, as an example, the comparison group may be set to the above-described first stage in which people in a normal category with a very low probability of developing dementia and a possibility of developing dementia exist.

That is, the comparison group may be set as a comparison group corresponding to the normal category and a comparison group determined to be subjective cognitive impairment, and mutual comparison may be performed.

As another example, the comparison group may be set as the first stage and the second stage as the above-described dementia stage. That is, a group having a difference in the degree of dementia may be set as a comparison group, and is not limited to the above-described embodiment. Thereafter, the learning model compares the output information obtained through each comparison group, and through this, the final dementia risk factor can be derived.

In this case, for example, when the output information may indicate the possibility of dementia based on information between the comparison groups, a new dementia risk factor may be derived. In addition, the learning model can be used to derive new dementia risk factors by storing the above-mentioned information, continuously updating it to set various comparison groups, and updating specific information about the comparison group. That is, new dementia risk factors can be derived through deep learning through existing dementia risk factors and biometric information.

9 is a diagram illustrating a method for performing deep learning according to an embodiment of the present invention.

Referring to FIG. 9 , two stages of deep learning may be applied. For example, deep learning performed based on learning model 1 may be a dementia risk factor derived through measurement information and biometric information. That is, it may be a dementia risk factor derived through the method of FIG. 7 described above. In this case, as an example, the output information derived through the aforementioned FIG. 7 may be input information for the learning model 2 . That is, the derived dementia risk factor may be new input information. In this case, learning model 2 may derive dementia risk factors by further adding biometric information as input information. That is, new dementia risk factors can be derived from existing dementia risk factors as shown in FIG. 8, and through this, risk factors related to dementia can be derived. That is, when deriving dementia risk factors, two stages of deep learning can be applied, and more specific dementia risk factors can be derived through the above-mentioned bar.

Meanwhile, as an example, the biometric information used in FIGS. 7 to 9 may include information obtained through the IoT device in FIG. 5 . That is, information obtained by further considering the biometric information obtained through the wearable device and the IoT device may be input information utilized in FIGS. 7 to 9 , and is not limited to the above-described embodiment.

10 is a flowchart of a method of deriving dementia risk factors according to an embodiment of the present invention.

Referring to FIG. 10 , the server may acquire biometric information from each of the objects corresponding to the first comparison group through the wearable device. (S1010) In this case, as an example, the first comparison group in FIG. 10 may be subjects corresponding to the subjective cognitive impairment state. As another example, the first comparison group may be another comparison group according to the degree of dementia, and is not limited to the above-described embodiment. In this case, biometric information for each of the objects corresponding to the first comparison group may be obtained from the wearable device. (S1020) Also, as an example, the server may acquire measurement information for each of the subjects corresponding to the first comparison group. In this case, the measurement information may be at least any one or more of brain image, CT, MRI, EEG, and clinical findings data information. In addition, as an example, the measurement information may mean a measurement for determining whether or not dementia occurs and the degree of dementia in a medical institution, and is not limited to the above-described embodiment. In this case, as an example, the server may acquire measurement information for each of the objects corresponding to the above-described first comparison group. Thereafter, the server may derive the first dementia risk factor based on biometric information and measurement information for each of the subjects. (S1030) In this case, as an example, the server may transmit biometric information and measurement information from a medical institution or medical staff, and may derive the first dementia risk factor based on the information obtained therefrom. That is, the first dementia risk factor may be determined through a medical institution or a medical staff. As an example, dementia risk factors include calories burned based on measurement information corresponding to the comparison group, activity time, sleep time, type of sleep by sleep time, heart rate, number of steps, step distance, and number of exercise as described above. It may mean a derived factor that can affect dementia. That is, dementia risk factors may be factors affecting the occurrence of dementia, and may be derived by medical institutions or medical staff.

As another example, the first dementia risk factor may be derived through deep learning of the server. That is, the first dementia risk factor information can be directly derived through deep learning based on the information acquired by the server without transmitting biometric information and measurement information to a medical institution or medical staff. As an example, measurement information and biometric information for the comparison group may be matched through deep learning, and the first dementia risk factor information may be derived through this.

Next, a second dementia risk factor related to the first dementia risk factor may be derived through deep learning performed based on biometric information related to the first dementia risk factor and information on the comparison group. (S1040) In this case, as an example, as the comparison group, the second comparison group may be set together with the above-described first comparison group. For example, in FIG. 10 , the second comparison group may be a comparison group corresponding to a normal state. That is, the first comparison group may be a comparison group corresponding to the subjective cognitive impairment state, and the second comparison group may be a comparison group corresponding to the normal state. In addition, as an example, it may be possible to set the comparison group in various ways, and the above-described method may be one example. In this case, as an example, biometric information about the subjects corresponding to the first comparison group and the subjects corresponding to the second comparison group may be measured through the wearable device. The server may perform the above-described deep learning based on the above-described measurement information and comparison group characteristic information. Thereafter, the second dementia risk factor may be derived from the biometric information related to the first dementia risk factor. That is, the second dementia risk factor may be a new dementia risk factor derived from the first dementia risk factor. That is, as described above, the server can derive new dementia risk factors through deep learning based on existing dementia risk factors, biometric information, and comparative group information.

The above-described embodiments of the present invention may be implemented through various means. For example, embodiments of the present invention may be implemented by hardware, firmware, software, or a combination thereof.

In case of implementation by hardware, the method according to embodiments of the present invention may include one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs). , FPGAs (Field Programmable Gate Arrays), processors, controllers, microcontrollers, microprocessors, and the like.

In the case of implementation by firmware or software, the method according to the embodiments of the present invention may be implemented in the form of a module, procedure, or function that performs the functions or operations described above. The software code may be stored in the memory unit and driven by the processor. The memory unit may be located inside or outside the processor, and may transmit and receive data to and from the processor by various known means.

The detailed description of the preferred embodiments of the present invention disclosed as described above is provided to enable any person skilled in the art to make and practice the present invention. Although the above has been described with reference to the preferred embodiment of the present invention, those skilled in the art can variously modify and change the present invention within the scope without departing from the spirit and scope of the present invention as set forth in the claims below. You will understand that it can be done. Accordingly, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein. In addition, although preferred embodiments of the present specification have been illustrated and described above, the present specification is not limited to the specific embodiments described above, and the technical field to which the present invention belongs without departing from the gist of the present specification as claimed in the claims Various modifications may be made by those of ordinary skill in the art, and these modifications should not be individually understood from the technical spirit or perspective of the present specification.

In this specification, both product inventions and method inventions are described, and the descriptions of both inventions may be supplementally applied if necessary.

100: wearable device
110: transceiver of the wearable device
120: control unit of the wearable device
130: input unit of the wearable device
210: transceiver of the server
220: control unit of the server
230: deep learning part of the server
300: IoT device
310: Transceiver of IoT device
320: IoT device control unit

Claims (9)

In a method for a server to determine dementia risk factors using deep learning,
obtaining biometric information from each of the subjects corresponding to the first comparison group through a wearable device;
obtaining measurement information for each of the subjects corresponding to the first comparison group;
deriving a first dementia risk factor based on the biometric information and the measurement information for each of the subjects;
Deriving a second dementia risk factor related to the first dementia risk factor through deep learning performed based on the biometric information related to the first dementia risk factor and the comparison group information; Containing, dementia risk factor determination method .
The method of claim 1,
When deep learning for deriving the second dementia risk factor is performed, a second comparison group is further set, and biometric information related to the first dementia risk factor is obtained from each of the subjects corresponding to the first comparison group One biometric information and biometric information obtained from each of the subjects corresponding to the second comparison group,
The deep learning is performed by comparing the biometric information of the first comparison group and the biometric information of the second comparison group, dementia risk factor determination method.
3. The method of claim 2,
The first comparison group is a comparison group corresponding to the subjective cognitive impairment state,
The second comparison group is a comparison group corresponding to a normal state, the dementia risk factor determining method.
The method of claim 1,
The biometric information is obtained by further using an Internet of Things (IoT) device, dementia risk factor determination method.
The method of claim 1,
Wherein the measurement information is derived based on at least one or more of brain image, CT, MRI, EEG and clinical findings data data, dementia risk factor determination method.
The method of claim 1,
When a first dementia risk factor is derived based on the biometric information and the measurement information for each of the subjects, the biometric information and the measurement information are transmitted from the server to a medical institution, and the server is Receiving the determined first dementia risk factor information, dementia risk factor determination method.
The method of claim 1,
When a first dementia risk factor is derived based on the biometric information and the measurement information for each of the subjects, the server performs deep learning based on the biometric information and the measurement information, and based on the deep learning A method of determining dementia risk factors, which is obtained directly through the server.
In the server for determining dementia risk factors using deep learning,
a transceiver for transmitting and receiving a signal;
a deep learning learning unit that performs deep learning learning; and
Including a control unit for controlling the transceiver and the deep learning learning unit,
The control unit is
Obtaining biometric information through a wearable device from each of the subjects corresponding to the first comparison group,
Obtaining measurement information for each of the subjects corresponding to the first comparison group,
Derive a first dementia risk factor based on the biometric information and the measurement information for each of the subjects,
A server for determining a dementia risk factor, deriving a second dementia risk factor related to the first dementia risk factor through deep learning performed based on biometric information and comparison group information related to the first dementia risk factor.
In the system for determining dementia risk factors,
A server that determines dementia risk factors through deep learning; and
a wearable device for measuring biometric information;
An IoT device that measures the wearable device and the biometric information together or transmits the measured biometric information to a server; including,
The server obtains biometric information from the wearable devices of each of the subjects corresponding to the first comparison group,
The server obtains measurement information for each of the subjects corresponding to the first comparison group,
The server derives a first dementia risk factor based on the biometric information and the measurement information for each of the subjects,
The server is a system for determining a dementia risk factor, deriving a second dementia risk factor related to the first dementia risk factor through deep learning performed based on biometric information and comparison group information related to the first dementia risk factor .

Images