KR20220106714A - IT device-based system for providing and managing personal biological rhythm customized drug intake information - Google Patents

Abstract

The present invention relates to a provision and management system of personal biological rhythm customized drug intake information based on personal terminal measurement information, in which a main server receives the individual's activity data and biometric information from the personal terminal (IT device), analyzes patterns of received individual activity data and biometric information to calculate circadian biorhythm, produces drug intake information (type of drug taken, timing of taking, duration of taking, dosage, etc.) that must be taken for the purpose of preventing or curing disease or health management and promotion for each individual by using the calculated circadian rhythm and the individual's gender, age, height, and weight, transmits the generated drug intake information to the personal terminal, and the personal terminal issues an alarm at the time of taking, and confirms and manages whether or not to take through a message confirming whether or not to take. The provision and management system of personal biological rhythm customized drug intake information comprises: the personal terminal that transmits the motion signal and the illuminance signal as activity information to the main server; and the main server that detects the biological rhythm using activity information received from the personal terminal, inputs personal information to a pre-learned artificial intelligence network, and transmits drug intake information output from the artificial intelligence network to the personal terminal.

Description

The system for providing and managing personal biological rhythm customized drug intake information based on personal terminal measurement information {IT device-based system for providing and managing personal biological rhythm customized drug intake information}

According to the present invention, the main server receives personal activity data and biometric information from a personal terminal (IT device), analyzes the received personal activity data and the pattern of biometric information to calculate the circadian biorhythm, and calculates the calculated work Information on drug intake that should be taken for the purpose of preventing, curing, or promoting health for each individual using the cycle biorhythm and the individual's gender, age, height, and weight (type of drug taken, time of taking, duration of use , dose, etc.), transmits the generated drug intake information to the personal terminal, the personal terminal alarms the time of taking, and checks and manages whether or not taking the drug is taken through a message confirming whether to take it. It relates to a system for providing and managing individual biorhythm customized drug intake information.

Most living things adapt to the Earth's rotation and have a 24-hour periodicity and precisely regulate almost all physiology, metabolism, and behavior at the hormonal level, which is called circadian rhythm.

In other words, the circadian rhythm refers to an intrinsic rhythm of about 24 hours created by signals such as external light, temperature, and food intake as a rhythm of various lengths having a periodic pattern of our body. When it occurs, various symptoms such as drowsiness, insomnia, fatigue, headache, and poor concentration occur in daily life.

All living things on Earth have evolved a circadian rhythm system to adapt to the cycle of about 24 hours caused by the Earth's rotation, and various body functions such as metabolism, physiology, behavior, sleep patterns, and hormone production work in our body. It is regulated by the periodic rhythm. From behaviors such as sleeping and eating to hormone secretion, blood pressure and body temperature regulation, the circadian rhythm is closely related to the behavior of almost all living things. In a nutshell, the reasons why you feel sleepy at night, have poor digestion if you eat late-night meals, and feel hungry at meals are all related to your circadian rhythm. Modern life is far from normal biological rhythms. Not only do they not sleep until late at night under excessive artificial lighting, but they also often break their circadian rhythm by traveling abroad, working at night, and eating late at night. Therefore, the difference between individual circadian rhythms at the present time is very large, and this should be considered as individual circadian rhythms.

Sleep, wakefulness, hormones, heart rate, blood pressure, body temperature, etc. have a repetitive pattern of a certain cycle for each individual, and accordingly, the circadian rhythm also has a repetitive pattern with a regular cycle. Biorhythms are closely related to many physiological and biochemical mechanisms in our body, and problems may arise in the body depending on the biorhythms.

Among many things related to the circadian rhythm, in the case of drugs, the effect of the same drug varies depending on when it is taken. can

The present invention calculates a 24-hour circadian rhythm (circadian circadian rhythm) for each individual, and uses the calculated circadian circadian rhythm and personal information, that is, gender, age, height, and weight, for each individual, prevention of disease, Creates drug intake information (type of drug to be taken, time of taking, duration, dose, etc.) that must be taken for the purpose of healing or health care and promotion, and transmits the generated drug intake information to a personal terminal, and the personal terminal We propose a system for providing and managing drug intake information tailored to personal biorhythms based on personal terminal measurement information, which provides an alarm at the time and checks and manages whether to take it through a message confirming whether to take it.

As a prior art, Korean Patent Application Laid-Open No. 10-2015-0101179 relates to a biorhythm management system, a motion detection sensor for detecting the movement of a subject, an illuminance sensor for detecting the light exposure information of the subject, and a change in the subject's body temperature Calculating the amount of activity from the temperature sensor and the detected movement of the subject, tracking the change in the calculated amount of activity to derive the subject's sleep state to generate activity information and sleep information, the generated activity information, Comparing the circadian circadian rhythm derived from sleep information and light exposure information with the circadian circadian rhythm in a normal state of the subject, comprising a processing unit detecting a change in the subject's circadian circadian rhythm, By detecting changes, it is possible to determine whether the subject is abnormal and share the information. The present invention detects the circadian rhythm from the activity amount information, the sleep information, and the light exposure information, but it is difficult to detect the exact circadian rhythm of the subject by only this.

As another prior art, Korean Patent Application Laid-Open No. 10-2015-0081048 discloses a state detecting unit for detecting a plurality of state information including motion state information, and a state comparing the detected plurality of state information with a preset mode switching condition. The present invention relates to a wearable device for detecting biorhythms including a comparison unit and a mode switching unit for automatically switching between a sleep mode and an active mode according to a result of the comparison. The present invention is configured to switch between sleep mode and active mode by comparing motion, illuminance, time, GPS coordinate information, etc. with status information and preset mode switching conditions, so it is difficult to accurately detect individual circadian rhythms.

As another prior art, Korean Patent Registration No. 10-2124519 'Method, system and computer-readable medium for providing nutritional information' provides nutrients needed by the user based on the input information entered by the user for the input information category. It is made to calculate, but it is difficult to provide more suitable drug intake information to the user only by this.

The problem to be solved by the present invention is to determine the time when the drug responds most effectively in consideration of the individual 24-hour biorhythm, and at that point, to maximize the effect of the drug by allowing the user to administer the drug intensively, It is to provide a system for providing and managing personal biorhythm customized drug intake information based on personal terminal measurement information.

Another problem to be solved by the present invention is that the main server receives personal activity data, biometric information, etc. from a personal terminal (IT device), and analyzes the patterns of the received individual activity data and biometric information to analyze the circadian biorhythm Drugs that must be taken for each individual for the purpose of preventing, curing, or promoting health, using the calculated circadian rhythm and personal information (individual gender, age, height, weight) Creates intake information (type of medication, time of taking, period, dosage, etc.), transmits the generated medication intake information to a personal terminal It is to provide and manage a system for providing and managing personal biorhythm customized drug intake information based on personal terminal measurement information, which checks and manages whether or not there is.

In order to solve the above problems, the system for providing and managing personal biorhythm customized drug intake information of the present invention includes a motion detection sensor, a heart rate detection sensor, and an illuminance sensor to detect a motion signal, a heart rate signal, and an illuminance signal to transmit to the main server, personal terminal; The biorhythm is detected using the motion signal received from the personal terminal, the heart rate signal, and the illuminance signal, and the detected biorhythm and personal information consisting of the user's gender, age, height, and weight are input to the pre-learned artificial intelligence network. and a main server that transmits drug intake information, output from the artificial intelligence network, to a personal terminal.

The main server obtains activity amount information, sleep amount information, light exposure amount and light exposure time period using the movement signal, the heart rate signal, and the illuminance signal. If the amount of activity is less than the preset activity threshold, it is determined as a sleep state, and if it is greater than the activity threshold, it is determined as an active state, and the sleep time section and activity time section are detected during 24 hours. include

The main server detects the heart rate variability signal from the heart rate signal during the sleep time interval, and from the heart rate variability signal, the average heart rate, standard deviation of the average normal-to-normal intervals (SDANN), and LF/HF (low frequency power to high frequency power) and rRR (autocorrelation of R-R intervals) are detected, and using the detected heart rate variability parameter, a sleep phase, a waking phase, and a REM sleep phase are detected. (deep sleep stage) or non-REM sleep (Non REM) stage (shallow sleep stage) is judged and output as sleep quality, and the amount of sleep information includes sleep time and sleep quality information.

The personal terminal further includes a body temperature sensor, the detected body temperature signal is transmitted to the main server, and the main server detects the average body temperature signal per time section or body temperature variation information per time section as body temperature change information, and the illuminance signal, A time point having an illuminance signal exceeding the set illuminance threshold is determined as a light exposure time, counting successive light exposure time points to detect a light exposure time interval or light exposure time, and the detected light exposure time interval or light exposure time The average value of the illuminance signal during the period is detected as the amount of light exposure.

Biorhythm includes activity amount information, sleep amount information, body temperature change information, light exposure amount and light exposure time interval (or light exposure time) information, and the main server analyzes the 24 hour biorhythm through time series analysis to determine the trend. , the cycle (cycle), and the analyzed information on the irregularity (irregular) is taken as the circadian circadian rhythm.

The main server uses EM clustering using Gaussian mixture model (Expectation - Maximization clustering using Gaussian mixture model) to detect the degree of separation from the circadian rhythm and the circadian rhythm of the reference state as a vector value, It is determined whether the detected vector value is greater than the circadian rhythm threshold of a preset reference state, and the time point at which the detected vector value is located within the circadian rhythm threshold of the preset reference state is output as the effective response time of the drug and transmits the effective reaction time of the drug to the personal terminal as the time of taking.

The personal terminal receives the drug intake information from the main server, generates and outputs a dose alarm signal when the time of taking the drug is reached according to the drug intake information, outputs a dose confirmation message, and the user's input information according to the dose check message Accordingly, the dose confirmation information including the dose confirmation signal and the actual time taken is generated, and the dose confirmation information is transmitted to the main server.

The personal terminal includes a motion sensor, a heart rate sensor, an illuminance sensor, and a body temperature sensor to detect a biosignal, or the personal terminal includes a motion sensor, a heart rate sensor, an illuminance sensor, and a body temperature sensor to detect a biosignal and wired or wireless connection.

The main server stores the dosage confirmation information in the database to manage the dosage period.

In addition, the driving method of the system for providing and managing personal biorhythm customized drug intake information includes a motion detection sensor, a heart rate detection sensor, and an illuminance sensor of a personal terminal, including a motion signal, a heart rate signal, and an illuminance signal. Detecting and transmitting the information to the main server, information measurement step; a biorhythm detecting step of the main server detecting a biorhythm using a motion signal received from the personal terminal, a heart rate signal, and an illuminance signal in the information measurement step; The main server inputs the biorhythm detected in the biorhythm detection step and the personal information consisting of the user's gender, age, height, and weight into the pre-learned artificial intelligence network, and the drug intake information output from the artificial intelligence network to the personal terminal. and a step of detecting drug intake information.

In the biorhythm detection step, the main server obtains activity amount information, sleep amount information, light exposure amount and light exposure time interval using a movement signal, a heart rate signal, and an illuminance signal. The average of the motion signals is obtained as the amount of activity, and if the amount of activity is less than a preset activity threshold, it is determined as a sleep state, and if it is greater than the activity threshold value, it is determined as an active state. Including activity time period and activity amount.

In the biorhythm detection step, the main server detects the heart rate variability signal from the heart rate signal during the sleep time interval, and from the heart rate variability signal, the average heart rate, standard deviation of the average normal-to-normal intervals (SDANN) and , LF/HF (low frequency power to high frequency power), and rRR (autocorrelation of R-R intervals) are detected, and using the detected heart rate variability parameter, the sleep phase, the waking phase, It is judged as any one of REM sleep stage (deep sleep stage) and non-REM sleep stage (light sleep stage) and output as sleep quality, and sleep amount information includes sleep time and sleep quality information.

The personal terminal further includes a body temperature sensor, the detected body temperature signal is transmitted to the main server, and in the biorhythm detection step, the main server detects an average body temperature signal per time section or body temperature variation information per time section, as body temperature change information, , when the illuminance signal has an illuminance signal exceeding a preset illuminance threshold is determined as a light exposure time, counting successive light exposure time points to detect a light exposure time period or light exposure time, and detected light exposure The average value of the illuminance signal during the time period or the light exposure time is detected as the light exposure amount.

Biorhythm includes activity amount information, sleep amount information, body temperature change information, light exposure amount and light exposure time interval (or light exposure time) information, and the main server analyzes the 24 hour biorhythm through time series analysis to determine the trend. , the cycle (cycle), and the analyzed information on the irregularity (irregular) is taken as the circadian circadian rhythm.

Between the biorhythm detection step and the drug intake information detection step, the step of detecting an effective reaction time of the drug is further included, wherein the effective reaction time detection step of the drug is performed by the main server, EM clustering (Expectation - Maximization) using a Gaussian mixture model. clustering using Gaussian mixture model), the degree of separation of the circadian rhythm from the circadian circadian rhythm of the reference state is detected as a vector value, and the detected vector value is the circadian circadian rhythm of a preset reference state It is determined whether it is greater than the threshold, and a time point at which the detected vector value is located within the circadian rhythm threshold of a preset reference state is output as an effective response time of the drug.

After the drug intake information detection step, the personal terminal receives the drug intake information from the main server, and according to the drug intake information, when the taking time comes, a dose alarm signal is generated and output, a dose alarm signal output step; The personal terminal outputs a confirmation message of whether to take the drug, and according to the user's input information according to the confirmation message of whether or not to take, generates the confirmation information of whether to take or not, and the actual time taken, and transmits the confirmation information to the main server. It further includes; a dosage confirmation step.

The present invention features a recording medium for storing a computer program source for a method of providing and managing a system for providing and managing drug intake information tailored to individual biorhythms.

The system for providing and managing personal biorhythm customized drug intake information based on personal terminal measurement information of the present invention identifies a time when a drug responds most effectively in consideration of a 24-hour biorhythm for each user, and at that point, allows the user to , to maximize the effect of the drug by intensively administering it.

That is, according to the present invention, the main server receives personal activity data, biometric information, etc. from a personal terminal (IT device), and analyzes the received individual activity data and biometric information pattern to calculate a circadian biorhythm, Using the calculated circadian rhythm and personal information (person's gender, age, height, and weight), information on drug intake (taking type of drug, time of taking, period of taking, dose, etc.), sending the generated drug intake information to the personal terminal, and the personal terminal alerts the user when taking the drug Management, at a time when the drug responds most effectively to each user, allows the user to intensively administer the drug, thereby maximizing the effect of the drug.

The present invention detects the user's movement with a personal terminal (IT device) in order to understand the biorhythm, derives a section in which activity information falls below a reference value for more than a certain period of time as a sleep state, and uses the user's activity amount to Distinguish between sleep and activity. Alternatively, if there is a possibility of error in classifying the sleep state and the active state as the basis of the activity information, the accuracy of the biorhythm pattern analysis is increased by using the individual biometric information (pulse, etc.), the individual past activity pattern information, and the time information in combination.

In addition, the present invention derives a "drug taking model" to provide drug intake information (type of drug to be taken, time of taking, duration, dose, etc.) It is derived by considering the individual's gender, age, height, and weight in a complex way, so it is possible to derive a drug administration model that is more suitable for the user.

In addition, the present invention derives the optimal drug intake information for each individual in consideration of the type of drug that an individual takes for the purpose of preventing, curing, or promoting health according to the drug intake model.

In addition, in the present invention, the derived optimal drug intake information is It is provided through a personal terminal, that is, an IT device (mobile device, smart device, wearable device, or other activity information and digital health data measurement device), and additionally, by alarming the time of taking and managing whether or not to take the drug, At the time of the most effective response, the user continuously manages to take the drug according to the drug intake information for each user, and accordingly, the user can show the effect of continuous and systematic drug intake.

1 is an explanatory diagram schematically illustrating the configuration of a system for providing health functional food intake information based on analysis of health information and medication information of the present invention.
2 is a schematic diagram schematically illustrating a system for providing health functional food intake information based on analysis of health information and medication information of the present invention.
3 is a flowchart schematically illustrating a driving method of a system for providing health functional food intake information based on analysis of health information and medication information of the present invention.
4 is an explanatory diagram for explaining a degree of deviation from a current circadian rhythm from a circadian circadian rhythm threshold in a reference state.
5 is an explanatory diagram illustrating a clustering process of 42 clusters to which a vector weight is applied.
6 is an explanatory diagram for explaining a model in which the state of the subject calculated with personal information is clustered.

Hereinafter, the system for providing and managing personal biorhythm customized drug intake information based on personal terminal measurement information of the present invention will be described in detail with reference to the accompanying drawings.

1 is an explanatory diagram schematically illustrating the configuration of a system for providing health functional food intake information based on analysis of health information and medication information of the present invention, and FIG. 2 is a health function based on analysis of health information and medication information of the present invention. It is a schematic diagram schematically explaining a system that provides food intake information.

As shown in FIG. 1 , the system for providing health functional food intake information includes a personal terminal (IT device) 100 and a main server 200 .

The personal terminal 100 is a terminal in which a predetermined application program (application program) is installed, measures personal activity data, biometric information, and the like, and transmits it to the main server 200 .

The personal terminal 100 includes a motion sensor for detecting the movement of the subject, and an illuminance sensor for detecting the light exposure information of the subject. In addition, the personal terminal 100 may include a biosignal detection means for detecting an electrocardiogram signal, a photoplethysmogram signal, a heart rate signal, a pulse signal, an oxygen saturation signal, a blood pressure signal, or the like.

The individual activity data is a motion signal measured from a motion sensor and an illuminance signal measured from the illuminance sensor, and the biometric information may be at least one of an electrocardiogram signal, a photoplethysmogram signal, a heart rate signal, a pulse signal, and an oxygen saturation signal. .

In some cases, the personal terminal 100 receives biometric information (biometric signal), such as an electrocardiogram signal, a photoplethysmogram signal, a heart rate signal (pulse signal), or an oxygen saturation signal, from a biosignal measuring device that is linked to it via wire or wireless, and receives the received data. The biometric information may be transmitted to the main server 200 .

Here, the personal terminal 100 is an IT device, a mobile device (eg, a mobile phone, a laptop computer, etc.), a smart device (eg, a smart phone, a smart watch, etc.), a wearable device (eg, a smart watch, a necklace type smartphone, etc.) ), or other activity information and digital health data measurement device (eg, a device for the elderly with dementia including a GPS transceiver). In some cases, the wearable device may transmit data to the main server 200 via the mobile device.

Here, the personal terminal 100 and the main server 200 are means in which a microprocessor, a microcontroller, a computer, and the like are embedded.

In addition, the drug intake information includes the type of drug to be taken, the time of taking it, the period of taking it, the dosage, and the like. That is, the drug intake information includes the time when the drug most effectively acts (the time of taking), the type of drug to be taken, the period of taking it, the dosage, and the like.

1 and 2, the main server 200 receives personal activity data, biometric information, etc. from the personal terminal 100, and analyzes the received personal activity data and biometric information pattern to analyze the circadian biometrics. Calculating the rhythm, using the calculated circadian rhythm and pre-stored personal information, for example, the individual's gender, age, height, and weight, for the purpose of preventing, curing, or health management and promotion of diseases for each individual Creates drug intake information that should be taken with

That is, the main server 200 detects the sum (or average) of the motion signals for a predetermined time from the motion signals.

The main server 200 determines the sleep state when the sum (or average) of the motion signals is less than or equal to a preset activity amount reference value, and determines the active state if it is greater than the activity amount reference value. Alternatively, the main server 200 determines the sleep state if the sum (or average) of the motion signals is less than or equal to a preset activity threshold, and a predetermined biosignal (eg, heart rate) is less than or equal to, a preset, predetermined biological signal, reference value (eg, heart rate). , otherwise it is judged as an active state. In this way, the middle sleep time section and the active time section are detected for 24 hours, and the circadian rhythm is obtained.

The biological rhythm includes activity amount information, sleep amount information, body temperature change information, light exposure amount, and information on a light exposure time interval (or light exposure time). In some cases, information analyzed for trends, cycles, and irregularities, etc. may be used as circadian circadian rhythms through time series analysis of these 24-hour information.

The main server 200 determines the sleep state if the average of the motion signals detected by the motion sensor (not shown) is less than or equal to the preset activity amount reference value, and determines the active state if it is greater than the activity amount reference value, and sleeps during 24 hours. and detecting an activity time interval.

The motion sensor (not shown) may include one or more of a gyro sensor, an acceleration sensor, an angular velocity sensor, and a geomagnetic sensor.

The activity amount information may include an activity time interval (or activity time) and an activity amount.

The amount of activity may be obtained as an average of motion signals per predetermined time.

The sleep amount information may include a sleep time interval (or sleep time) and sleep quality information.

The sleep quality information is information indicating which stage the sleep stage is in a REM sleep stage or a non-REM sleep stage.

The sleep quality information indicates which stage of the sleep stage is the awakening stage, the REM sleep stage, and the non-REM sleep stage by using the heart rate variability signal and the motion signal detected during the sleep time period.

A heart rate (heartbeat) signal is detected from a heart rate detection sensor (not shown) and transmitted to the main server 200 , and a motion signal is detected from a motion signal detector (not shown) and transmitted to the main server 200 .

The heart rate detection sensor (not illustrated) may include a photoplethysmography sensor (not illustrated) or an electrocardiogram sensor (not illustrated).

The main server 200 detects a heart rate variability signal from a heart rate signal received from a heart rate detection sensor (not shown), and detects a plurality of heart rate variability parameters from the heart rate variability signal. Here, the plurality of heart rate variability parameters include an average heart rate, a time-domain heart rate variability (eg, standard deviation of the average normal-to-normal intervals), and a frequency-domain heart rate variability (eg, LF/HF ( low frequency power to high frequency power), alpha, and correlation coefficient heartbeat intervals (eg, autocorrelation of R-R intervals (rRR)).

In addition, the main server 200 detects motion information from the motion signal, and the motion information may be a motion average value for a predetermined time.

In some cases, the motion information may include an average motion reference value, a maximum motion reference value, and a heart rate-based motion reference value. The average motion reference value, the maximum motion reference value, and the heart rate-based motion reference value are known in Korean Patent Application Laid-Open No. 10-2014-0120513, and thus a detailed description thereof will be omitted.

The main server 200 uses a plurality of heart rate variability parameters and movement information to determine sleep phases, waking phases, REM sleep phases (deep sleep phases), and non-REM sleep phases (light sleep). step) and output as sleep quality information.

The main server 200 inputs a plurality of heart rate variability parameters and movement information to a pre-learned sleep detection AI network model, and outputs the REM sleep phase (deep sleep phase) from the sleep detection AI network model. Alternatively, a value of any one of non-REM sleep stages (shallow sleep stages) is output as sleep quality information.

Alternatively, the main server 200 applies a method known in Korean Patent Publication No. 10-2014-0120513 to the sleep phase by using a plurality of heart rate variability parameters and motion information, and the REM sleep phase ( Deep sleep stage) or non-REM sleep stage (light sleep stage).

The body temperature change information may be an average body temperature signal per time section or body temperature variation information.

The main server 200 detects the average body temperature signal per time section from the body temperature signal received from the body temperature sensor (not shown), and subtracts the currently detected average body temperature signal per time section with the average body temperature signal per consecutive time section One value is detected as body temperature variation information.

The main server 200 receives the illuminance signal received from the illuminance sensor (not shown), and determines a time point at which the received illuminance signal has an illuminance signal exceeding a preset illuminance threshold as a light exposure time, and successive light By counting the exposure time, the light exposure time period or the light exposure time is detected.

The detected light exposure time period or the average value of the illuminance signal during the light exposure time is detected as the light exposure amount.

Here, the illuminance threshold may be a value set at the time of shipment from the factory or a value set by the user at the initial stage of use.

The biosignal measuring unit of the personal terminal 100 includes at least a motion sensor, a heart rate sensor, a body temperature sensor (not shown), and an illuminance sensor (not shown).

In some cases, the main server 200 may additionally adjust the current biorhythm according to the previous dose confirmation information (a dose confirmation signal and actual dose time).

The circadian circadian rhythm is obtained for each subject for one week, and these are averaged and designated as the circadian circadian rhythm of the subject's reference state. In other words, one week's worth of each of the biological rhythms, that is, activity amount information, sleep amount information, body temperature change information, light exposure and light exposure time period (or light exposure time) is detected and averaged, and the circadian body in the reference state do it in rhythm

The main server 200 detects, as a vector value, the degree to which the current circadian rhythm is separated from the circadian rhythm of the reference state, and the detected vector value is a preset circadian rhythm threshold (explanation) For convenience, it is determined whether it is greater than the 'standard deviation threshold'). Here, the reference deviation threshold may be automatically set to a predetermined multiple (eg, one or two times, etc.) of the standard deviation of the circadian rhythm of the reference state. That is, the degree of deviation of the current circadian rhythm from the circadian circadian rhythm threshold of the reference state is derived as a vector value.

In this case, the clustering method applies the mean-shift algorithm. The mean-shift algorithm finds the center of gravity when there is a distribution of any data, and performs clustering in the form of movement. It is a model in which data moves in the most dense form around specific data and forms a cluster while moving to the center of the data distribution step by step.

4 is an explanatory diagram for explaining a degree of deviation from a current circadian rhythm from a circadian circadian rhythm threshold in a reference state.

In the case of gender, there are two classification criteria for male and female, and for age, there are 7 classification criteria for teenagers, 20s, 30s, 40s, 50s, 60s, and 70s. Including those in their 70s and those in their 70s or older. In addition, according to height and age, three classification criteria are divided into underweight, standard, and overweight. According to these criteria, users (subjects) may be classified into a total of 42 clusters. A vector value, which is the degree of deviation from the current circadian rhythm from the circadian circadian rhythm threshold of the reference state, is applied as a weight to the classified cluster.

에 대해 이 데이터가 어떠한 가우시안 분포(Gaussian distribution)에서 생성되었는지를 찾는 것이다. 이를 위해 레스폰서빌리티(responsibility, 클러스터에 속할 추정확률),

를 수학식1와 같이 정의한다. In the present invention, a Gaussian mixture model (GMM, Gaussian mixture model) that can be utilized for various types of clusters of subjects was utilized. Classification using GMM is

It is to find out what Gaussian distribution this data is generated from. For this, responsibility (estimated probability of belonging to a cluster),

is defined as in Equation 1.

에 대한 구성 요소 할당을 나타내며,

이 k번째 구성요소에서 추출되면 Znk = 1이고 나머지는 모두 0이다.Zn is a data point using a vector of size K × 1

represents the component assignment to

When extracted from this kth component, Znk = 1 and all others are 0.

EM (Expectation - Maximization) algorithm is applied to GMM to find the optimal state value of the subject. EM is divided into two steps, E-step and M-step, to find the optimal value while repeating the steps. Using the EM clustering using Gaussian mixture model, we define the probability that the subject's state value belongs to the Gaussian distribution, and apply the clustering process to find the optimal cluster based on the calculated conditional probability. At this time, the clustering is classified by applying the criteria according to the method of administration of the clinician.

Responsibility and EM are known in the paper "Technical Details about the Expectation Maximization (EM) Algorithm" by Dawen Liang published on February 25, 2015, and the like, and detailed descriptions thereof will be omitted. (https://dawenl.github.io/files/em.pdf)

5 is an explanatory diagram illustrating a clustering process of 42 clusters to which a vector weight is applied.

6 is an explanatory diagram for explaining a model in which the state of the subject calculated with personal information is clustered.

A model in which a subject's state calculated with personal information consisting of a biorhythm and a user's gender, age, height, and weight is clustered is as shown in the example of FIG. 6 .

The present invention provides information to determine when a drug responds most effectively in consideration of a 24-hour biorhythm for each user, and to allow the user to administer the drug intensively at that time.

That is, the main server 200 detects the effective reaction time of the drug by using the EM (Expectation - Maximization) clustering (EM clustering using Gaussian mixture model) using the Gaussian mixture model.

In other words, the main server 200 uses EM clustering using a Gaussian mixture model to detect the degree to which the current circadian rhythm is separated from the circadian rhythm in the reference state as a vector value, and the detected vector The effective response of the drug is determined by determining whether the value is greater than the circadian rhythm threshold of the preset reference state (that is, the reference deviation threshold), and the time point at which the detected vector value is located within the circadian rhythm threshold of the preset reference state. output as a point in time.

In addition, the main server 200 reads the user's personal information (gender, age, height, weight) pre-stored in the database, and reads personal information and circadian rhythm, that is, activity amount information, sleep amount information, body temperature change information, light The information of the exposure amount and the light exposure time interval (or light exposure time) is input to the artificial intelligence network of the pre-learned drug intake information, and the drug intake information is output from the artificial intelligence network.

For the artificial intelligence network that outputs drug intake information, it is described in a number of patents such as domestic patent application No. 10-2020-0152599 by the present applicant, and a more detailed description is omitted.

The personal terminal 100 receives the drug intake information from the main server 200, and generates and outputs a dosing alarm signal when the time of taking (taking time) comes according to the drug intake information, and generates an alarm by outputting an alarm, and a message to confirm whether to take or not. Outputs and according to the user's input, generates taking confirmation information consisting of a confirmation signal and actual time taken, and transmits the confirmation information to the main server 200, and the main server 200 receives confirmation information is stored in a database (not shown) to manage the time of taking.

The artificial intelligence network is an artificial neural network trained by multiple data sets consisting of personal information, circadian rhythm, and drug intake information determined by a medical professional.

3 is a flowchart schematically illustrating a driving method of a system for providing health functional food intake information based on analysis of health information and medication information of the present invention.

As an initial setting step, the personal terminal 100 transmits the personal information (gender, age, height, weight) pre-stored in the personal terminal 100 to the main server 200 when a predetermined application program is run (S110) .

In the information measurement step, the movement signal and the illuminance signal measured from the motion detection sensor and the illuminance sensor of the personal terminal 100 are transmitted to the main server 200 as activity information (S120), and the personal terminal 100 is The signal is measured and transmitted to the main server 200 .

In the biorhythm detection step, the personal terminal 100 and the main server 200 are linked, and the main server 200 detects the biorhythm (S130). Here, the circadian rhythm includes information on activity amount information, sleep amount information, body temperature change information, light exposure amount and light exposure time interval (or light exposure time), and in some cases, 24 hours of these information through time series analysis, trends Analyzed information on trends, cycles, and irregularities may be used as circadian circadian rhythms.

In some cases, it is possible to adjust the circadian rhythm according to the previous dose confirmation information stored in the database (taking confirmation signal and actual dose time). For example, if the actual dosing time is earlier than the dosing time, the start time of the active period can be moved forward in the circadian rhythm. You can set the time back.

As an effective reaction time detection step of the drug, the main server 200 uses EM clustering using a Gaussian mixture model to detect the degree of separation of the circadian circadian rhythm and the circadian circadian rhythm from the reference state as a vector value. , by determining whether the detected vector value is greater than the circadian rhythm threshold of the preset reference state (that is, the reference deviation threshold), and determining when the detected vector value is located within the circadian rhythm threshold of the preset reference state; It is output as the effective reaction time of the drug (S135).

Here, the circadian circadian rhythm of the reference state is preset by the main server 200 as the circadian circadian rhythm of the reference state by obtaining the circadian circadian rhythm for each subject and averaging them.

In the drug intake information detection step, the main server 200 reads the user's personal information (gender, age, height, weight) pre-stored in the database, personal information and circadian rhythm, that is, activity information, sleep information, Input body temperature change information, light exposure amount, and light exposure time interval (or light exposure time) into the pre-learned artificial intelligence network, and store the output from the artificial intelligence network as drug intake information including the time of administration in the database, , and also transmitted to the personal terminal 100 (S150).

The artificial intelligence network is an artificial neural network learned by a plurality of data sets consisting of personal information, circadian rhythm, and drug intake information determined by a medical professional.

In the dose alarm signal output step, the personal terminal 100 generates a dose alarm signal when the dose time (taking time) comes according to the effective reaction time of the drug and the drug intake information received from the main server 200 (S160) , by outputting a dose alarm signal to the personal terminal (S120) to perform an alarm.

In the dose confirmation step, the personal terminal 100 outputs a dose confirmation message, etc., and according to the user's input, generates dose confirmation information including a dose confirmation signal and actual time taken, and sends the dose confirmation information to the main server. 200, and the main server 200 stores the dose confirmation information in a database (not shown) to manage the dose timing.

In the foregoing, the present invention has been shown and described with respect to certain preferred embodiments. However, the present invention is not limited to the above-described embodiments, and those of ordinary skill in the art to which the present invention pertains can make various changes without departing from the spirit of the present invention described in the claims below. will be able to carry out

100: personal terminal
200: main server

Claims (18)

a personal terminal including a motion detection sensor, a heart rate detection sensor, and an illuminance sensor to detect a motion signal, a heart rate signal, and an illuminance signal and transmit the detected motion signal to a main server;
The biorhythm is detected using the motion signal received from the personal terminal, the heart rate signal, and the illuminance signal, and the detected biorhythm and personal information consisting of the user's gender, age, height, and weight are input to the pre-learned artificial intelligence network. And, the main server for transmitting the drug intake information output from the artificial intelligence network to the personal terminal;
A system for providing and managing personal biorhythm customized drug intake information, comprising: The method of claim 1, wherein the main server,
Obtaining activity amount information, sleep amount information, light exposure amount and light exposure time interval using a motion signal, a heart rate signal, and an illuminance signal,
When detecting biorhythms, the average of the motion signals for a predetermined unit time is calculated as the amount of activity, and if the amount of activity is less than a preset activity threshold, it is determined as a sleep state, and if it is greater than the activity amount threshold, it is determined as an active state. Detects the activity time period,
The activity amount information provides and manages personal biorhythm customized drug intake information, characterized in that it includes an activity time interval and an activity amount. The method of claim 2, wherein the main server,
The heart rate variability signal is detected from the heart rate signal during the sleep time interval, and from the heart rate variability signal, the average heart rate, standard deviation of the average normal-to-normal intervals (SDANN), and low frequency power to high frequency (LF/HF) are detected. power) and HR parameters including autocorrelation of RR intervals (rRR) are detected, and using the detected HR parameters, sleep phase, waking phase, and REM sleep phase (deep sleep phase) are detected. ) and non-REM sleep (Non REM) stage (shallow sleep stage) and output as sleep quality,
Sleep amount information is characterized in that it includes sleep time and sleep quality information, a system for providing and managing personal biorhythm customized drug intake information. 3. The method of claim 2,
The personal terminal further includes a body temperature sensor, and the detected body temperature signal is transmitted to the main server,
the main server
Detecting the average body temperature signal per time section or body temperature variation information per time section as body temperature change information,
A time point at which the illuminance signal has an illuminance signal exceeding a preset illuminance threshold is determined as a light exposure time, and the consecutive light exposure time is counted to detect a light exposure time period or light exposure time,
A system for providing and managing individual biorhythm customized drug intake information, characterized in that the detected light exposure time interval or the average value of the illuminance signal during the light exposure time is detected as the light exposure amount. 5. The method of claim 4,
The biological rhythm includes information on activity amount information, sleep amount information, body temperature change information, light exposure amount and light exposure time period (or light exposure time),
Personal biorhythm customized drug, characterized in that the main server performs circadian biorhythms with information analyzed on trends, cycles, and irregularities through time series analysis of 24 hour biorhythms A system for provision and management of intake information. 6. The method of claim 5,
The main server uses EM clustering using Gaussian mixture model (Expectation - Maximization clustering using Gaussian mixture model) to detect the degree of separation from the circadian rhythm and the circadian rhythm of the reference state as a vector value, It is determined whether the detected vector value is greater than the circadian rhythm threshold of a preset reference state, and the time point at which the detected vector value is located within the circadian rhythm threshold of the preset reference state is output as the effective response time of the drug A system for providing and managing personal biorhythm customized drug intake information, characterized in that. 7. The method of claim 6,
The main server transmits the effective reaction time of the drug to the personal terminal as the time of taking, the personal biorhythm customized drug intake information providing and management system. 8. The method of claim 7,
The personal terminal receives the drug intake information from the main server, generates and outputs a dosage alarm signal when it is time to take the drug according to the drug intake information, and outputs a dosage confirmation message,
Personal biorhythm customized drug, characterized in that according to the user's input information according to the taking confirmation message, the taking confirmation signal and the actual taking time are generated, and the taking confirmation information is transmitted to the main server. A system for provision and management of intake information. According to claim 1,
A personal terminal is equipped with a motion sensor, a heart rate sensor, an illuminance sensor, and a body temperature sensor to detect a biosignal,
or a system for providing and managing personal biorhythm customized drug intake information, characterized in that the personal terminal is connected to a biosignal detecting means including a motion sensor, a heart rate sensor, an illuminance sensor, and a body temperature sensor through wired/wireless communication. 9. The method of claim 8,
The main server stores the dosage confirmation information in the database to manage the dosage period, the personal biorhythm customized drug intake information providing and management system. an information measuring step of detecting a motion signal, a heart rate signal, and an illuminance signal, including a motion detection sensor, a heart rate detection sensor, and an illuminance sensor of the personal terminal, and transmitting the detected motion signal to the main server;
a biorhythm detecting step of the main server detecting a biorhythm using a motion signal received from the personal terminal, a heart rate signal, and an illuminance signal in the information measurement step;
The main server inputs the biorhythm detected in the biorhythm detection step and the personal information consisting of the user's gender, age, height, and weight into the pre-learned artificial intelligence network, and the drug intake information output from the artificial intelligence network to the personal terminal. Transmitting, drug intake information detection step;
A method of operating a system for providing and managing personal biorhythm customized drug intake information, comprising: 12. The method of claim 11, wherein in the biorhythm detection step,
The main server obtains activity amount information, sleep amount information, light exposure amount and light exposure time interval using a motion signal, a heart rate signal, and an illuminance signal,
When detecting biorhythms, the average of the motion signals for a predetermined unit time is calculated as the amount of activity, and if the amount of activity is less than a preset activity threshold, it is determined as a sleep state, and if it is greater than the activity amount threshold, it is determined as an active state. Detects the activity time period,
The activity amount information is a method of providing and managing a system for providing and managing personal biorhythm customized drug intake information, characterized in that it includes an activity time interval and an activity amount. The method according to claim 12, wherein in the biorhythm detection step,
The main server detects the heart rate variability signal from the heart rate signal during the sleep time interval, and from the heart rate variability signal, the average heart rate, standard deviation of the average normal-to-normal intervals (SDANN), and LF/HF (low frequency power to high frequency power) and rRR (autocorrelation of RR intervals) are detected, and using the detected heart rate variability parameter, a sleep phase, a waking phase, and a REM sleep phase are detected. (deep sleep stage) or non-REM sleep (Non REM) stage (shallow sleep stage) is judged and output as sleep quality,
Sleep amount information, characterized in that it includes sleep time and sleep quality information, a method of driving a system for providing and managing personal biorhythm customized drug intake information. 14. The method of claim 13,
The personal terminal further includes a body temperature sensor, and the detected body temperature signal is transmitted to the main server,
In the biorhythm detection stage, the main server
Detecting the average body temperature signal per time section or body temperature variation information per time section as body temperature change information,
A time point at which the illuminance signal has an illuminance signal exceeding a preset illuminance threshold is determined as a light exposure time, and the consecutive light exposure time is counted to detect a light exposure time period or light exposure time,
A method of providing and managing a personal biorhythm customized drug intake information, characterized in that detecting the detected light exposure time interval or the average value of the illuminance signal during the light exposure time as the light exposure amount. 15. The method of claim 14,
The biological rhythm includes information on activity amount information, sleep amount information, body temperature change information, light exposure amount and light exposure time period (or light exposure time),
Personal biorhythm customized drug, characterized in that the main server performs circadian biorhythms with information analyzed on trends, cycles, and irregularities through time series analysis of 24 hour biorhythms A method of providing intake information and operating a management system. The method according to claim 5, further comprising a step of detecting an effective reaction point of the drug between the step of detecting the biological rhythm and the step of detecting the drug intake information, wherein the step of detecting the effective reaction point of the drug comprises:
The main server uses EM clustering using Gaussian mixture model (Expectation - Maximization clustering using Gaussian mixture model) to detect the degree of separation of the circadian rhythm from the circadian rhythm of the reference state as a vector value, It is determined whether the detected vector value is greater than the circadian rhythm threshold of a preset reference state, and the time point at which the detected vector value is located within the circadian rhythm threshold of the preset reference state is output as the effective response time of the drug A method of operating a system for providing and managing personal biorhythm customized drug intake information, characterized in that. 17. The method of claim 16,
After the drug intake information detection step, the personal terminal receives the drug intake information from the main server, and according to the drug intake information, when the taking time comes, a dose alarm signal is generated and output, a dose alarm signal output step;
The personal terminal outputs a confirmation message of whether to take the medicine, and according to the user's input information according to the confirmation message of whether or not to take it, it generates the confirmation information of whether to take the medicine and the actual time taken, and transmits the confirmation information to the main server. taking confirmation step;
A method of driving a system for providing and managing personal biorhythm customized drug intake information, characterized in that it further comprises a. 118. A recording medium for storing a computer program source for the method of providing and managing the personal biorhythm customized drug intake information according to any one of claims 11 to 117.

Images