CN117653053A - Method for predicting health risk through intelligent watch - Google Patents
Method for predicting health risk through intelligent watch Download PDFInfo
- Publication number
- CN117653053A CN117653053A CN202311573445.4A CN202311573445A CN117653053A CN 117653053 A CN117653053 A CN 117653053A CN 202311573445 A CN202311573445 A CN 202311573445A CN 117653053 A CN117653053 A CN 117653053A
- Authority
- CN
- China
- Prior art keywords
- data
- unit
- user
- risk
- health
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000036541 health Effects 0.000 title claims abstract description 103
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 95
- 238000011156 evaluation Methods 0.000 claims abstract description 73
- 238000012502 risk assessment Methods 0.000 claims abstract description 60
- 238000001514 detection method Methods 0.000 claims abstract description 52
- 238000000605 extraction Methods 0.000 claims abstract description 39
- 230000000694 effects Effects 0.000 claims abstract description 37
- 238000004458 analytical method Methods 0.000 claims abstract description 34
- 238000012545 processing Methods 0.000 claims abstract description 15
- 238000011161 development Methods 0.000 claims abstract description 13
- 230000006872 improvement Effects 0.000 claims abstract description 12
- 238000012800 visualization Methods 0.000 claims abstract description 10
- 230000006870 function Effects 0.000 claims description 60
- 238000012795 verification Methods 0.000 claims description 34
- 238000007781 pre-processing Methods 0.000 claims description 27
- 238000012549 training Methods 0.000 claims description 25
- 238000007405 data analysis Methods 0.000 claims description 23
- 239000008280 blood Substances 0.000 claims description 21
- 210000004369 blood Anatomy 0.000 claims description 21
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 20
- 238000012544 monitoring process Methods 0.000 claims description 20
- 229910052760 oxygen Inorganic materials 0.000 claims description 20
- 239000001301 oxygen Substances 0.000 claims description 20
- 230000036772 blood pressure Effects 0.000 claims description 18
- 238000012360 testing method Methods 0.000 claims description 17
- 238000005457 optimization Methods 0.000 claims description 15
- 238000010801 machine learning Methods 0.000 claims description 12
- 230000009466 transformation Effects 0.000 claims description 12
- 230000033001 locomotion Effects 0.000 claims description 10
- 230000007958 sleep Effects 0.000 claims description 9
- 230000006399 behavior Effects 0.000 claims description 8
- 238000004891 communication Methods 0.000 claims description 8
- 238000004140 cleaning Methods 0.000 claims description 7
- 230000008859 change Effects 0.000 claims description 6
- 230000003860 sleep quality Effects 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 239000003814 drug Substances 0.000 claims description 5
- 239000000284 extract Substances 0.000 claims description 4
- 230000004617 sleep duration Effects 0.000 claims description 4
- 230000008901 benefit Effects 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000013501 data transformation Methods 0.000 claims description 3
- 238000011158 quantitative evaluation Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 230000001131 transforming effect Effects 0.000 claims description 3
- 230000000875 corresponding effect Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000036760 body temperature Effects 0.000 description 3
- 238000002790 cross-validation Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 206010020772 Hypertension Diseases 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000000241 respiratory effect Effects 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 208000031636 Body Temperature Changes Diseases 0.000 description 1
- 208000001953 Hypotension Diseases 0.000 description 1
- 206010037660 Pyrexia Diseases 0.000 description 1
- 208000006011 Stroke Diseases 0.000 description 1
- 206010042434 Sudden death Diseases 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000001154 acute effect Effects 0.000 description 1
- 230000005189 cardiac health Effects 0.000 description 1
- 230000036996 cardiovascular health Effects 0.000 description 1
- 210000000748 cardiovascular system Anatomy 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 235000006694 eating habits Nutrition 0.000 description 1
- 230000005548 health behavior Effects 0.000 description 1
- 208000019622 heart disease Diseases 0.000 description 1
- 230000036543 hypotension Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000003340 mental effect Effects 0.000 description 1
- 230000004630 mental health Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 208000024335 physical disease Diseases 0.000 description 1
- 238000000053 physical method Methods 0.000 description 1
- 230000035790 physiological processes and functions Effects 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 230000036632 reaction speed Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003938 response to stress Effects 0.000 description 1
- 230000033764 rhythmic process Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 208000011580 syndromic disease Diseases 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Abstract
The invention discloses a method for predicting health risk through an intelligent watch, which relates to the technical field of predicting health risk, and mainly solves the problems of low accuracy of health detection results, unstable risk prediction and unclear health development trend, and comprises the following steps: collecting and recording data information of user activities; processing and analyzing the acquired data information; comparing the user data information to perform health detection; predicting the health risk of the user according to the risk assessment; providing an improvement suggestion according to the risk assessment and prediction results, extracting characteristic values through a characteristic extraction algorithm to realize the utilization of effective information, and improving the accuracy of the health detection results; the analysis and evaluation of the physical condition of the user are realized through an analysis and evaluation algorithm, and the stability of risk prediction is improved; the development trend of the health level is displayed to the user through the visualization unit, so that the user's knowledge of the development of the health condition is improved.
Description
Technical Field
The invention relates to the technical field of predicting health risks, and more particularly relates to a method for predicting health risks through a smart watch.
Background
At present, a large number of high-pressure industries such as public security, fire protection, rescue, medical care and other practitioners often face extreme physical load and high-strength mental pressure or excessive overdraft of physical strength in work, and health risks such as hypertension, heart disease, stroke, sudden death and the like are easily caused. In order to ensure the health of the patients, the current method of annual physical examination is generally adopted to perform health screening, however, the annual physical examination has long time intervals and cannot be effectively identified for some acute or sudden health risks, so that a method capable of performing health detection in real time is needed. The invention provides a method for predicting health risk of a user by wearing an intelligent watch, and the intelligent watch can collect various key physical signs of the user, such as: the heart rate, blood pressure, blood oxygen saturation and other body data are intelligently monitored and identified by a machine learning method, so that the user is at health risk. Because the intelligent watch can be worn for 7 x 24 hours, the timeliness of health monitoring can be effectively improved, and prompt is carried out when health risks are monitored.
Because the common intelligent watch detects the physiological parameters of the user inaccurately, errors are generated on the health detection result of the user; the common intelligent watch has insufficient evaluation logic when evaluating the health risk of the user, and can cause larger error to the risk prediction of the next step; the old intelligent watch has no complete summary chart for the development trend of the health condition of the user, which is unfavorable for scientifically and normally maintaining the health of the user.
Disclosure of Invention
Aiming at the defects of the technology, the invention discloses a method for predicting health risk through an intelligent watch, wherein the characteristic value is extracted through a characteristic extraction algorithm to realize the utilization of effective information, so that the accuracy of a health detection result is improved; the analysis and evaluation of the physical condition of the user are realized through an analysis and evaluation algorithm, and the stability of risk prediction is improved; the development trend of the health level is displayed to the user through the visualization unit, so that the user's knowledge of the development of the health condition is improved.
Accordingly, the present invention provides a method for predicting health risk by a smart watch, comprising the steps of:
step 1, acquiring and recording data information of user activities;
collecting and storing data information of user activities through a data acquisition module;
step 2, processing and analyzing the acquired data information;
processing and analyzing the acquired data information through a data analysis module; the data analysis module comprises a data preprocessing unit, a feature extraction unit and a data analysis unit, wherein the data preprocessing unit performs preprocessing operation on collected original data through cleaning, sorting, denoising and missing value filling so as to realize the integrity and the efficiency of the data; the feature extraction unit extracts feature values from the preprocessed data through a feature extraction algorithm so as to realize the utilization of effective information; the data analysis unit analyzes the association and rule between the physiological parameter information of the user through machine learning and big data, the output end of the data preprocessing unit is connected with the input end of the characteristic extraction unit, and the output end of the characteristic extraction unit is connected with the input end of the data analysis unit;
step 3, comparing user data information to perform health detection;
detecting the health condition of the user through a health detection module;
step 4, predicting the health risk of the user according to the risk assessment;
carrying out risk assessment on physical conditions of the user through a risk assessment module; the risk assessment module comprises a risk modeling unit, a risk verification unit, a basic data acquisition unit and a risk assessment unit, wherein the risk modeling unit is used for carrying out risk assessment on physical conditions of a user by constructing a risk assessment model; the risk verification unit verifies the working performance of the risk assessment model through the training data set and the test data set; the basic data acquisition unit actively sets personal information through a touch screen user to provide basic data information of the user; the risk assessment unit inputs processed data into the risk assessment model through Bluetooth wireless communication to achieve assessment of health risks of users, the output end of the risk modeling unit is connected with the input end of the risk verification unit, the output end of the risk verification unit is connected with the input end of the data acquisition unit, and the output end of the data acquisition unit is connected with the input end of the data assessment unit;
step 5, providing improvement suggestions according to the risk assessment and prediction results;
and providing expertise according to the risk assessment result through a result feedback module.
As a further description of the above technical solution, the data acquisition module includes a function sensing unit, an activity monitoring unit and a data recording unit, where the function sensing unit senses physiological parameter information of a user by using a sensor; the activity monitoring unit monitors the activity state of the user in real time through the gyroscope and the accelerometer so as to classify and identify the behavior mode of the user; the data recording unit adopts an EEPROM model memory to store data information of user activities, the output end of the function sensing unit is connected with the input end of the activity monitoring unit, and the output end of the activity monitoring unit is connected with the input end of the data recording unit.
As a further description of the above technical solution, the health detection module includes a temperature detection unit, a heart rate detection unit, a sleep detection unit, a blood oxygen detection unit, and a blood pressure detection unit, where the temperature detection unit monitors the skin temperature of the user in real time through a temperature sensor to detect the health condition of the user; the heart rate detection unit detects heart rate data of a user through a photoelectric sensor; the sleep detection unit detects the sleep quality and the sleep duration of the user through the movement amplitude of the user when sleeping; the blood oxygen detection unit senses the blood oxygen saturation of the user through the infrared sensor so as to detect blood oxygen data of the user; the blood pressure detection unit measures a pressure change between the user and the skin by a pressure sensor to detect a blood pressure condition of the user.
As a further description of the above technical solution, the feature extraction algorithm is an algorithm for extracting feature information valuable for solving problems from raw data by processing and analyzing the raw data, and converting complex raw data into a compact, representative and easy-to-understand feature representation form, and the implementation steps of the feature extraction algorithm are as follows:
1) Data preprocessing: processing the original data, including cleaning the data, removing abnormal values or noise and filling missing values;
2) And (3) data transformation: transforming the data according to a transformation function to better adapt to the feature extraction algorithm, wherein the transformation function has a formula:
in the formula (1), T represents a transformation function, λ represents a value of data to be transformed, μ represents a defined common standard value, ζ represents a transformation factor, K (λ, μ) represents a sum of the data to be transformed and the standard value, and L (λ, μ) represents a difference between the data to be transformed and the standard value;
3) Feature selection: selecting the most representative and distinguishing features according to the problem to be solved and the factors of the features;
4) And (3) feature construction: constructing new features by combining the original features;
5) Feature dimension reduction: when the feature dimension is too high and redundant features exist, the movement can reduce the dimension so as to reduce the feature quantity;
6) Feature evaluation and verification: the extracted features are evaluated and verified through a feature verification function, the contribution degree and effect of the extracted features on the problem are checked, and the formula expression of the feature verification function is as follows:
in the formula (2), P represents a feature verification function, τ represents an extracted feature coefficient value, θ represents a verification factor value of the feature verification function, and ω represents an effect ratio of a user physiological parameter in the feature verification function.
As a further description of the above technical solution, the risk assessment model includes a parameter receiving unit, a parameter assessment unit and a risk prediction unit, where the parameter receiving unit receives parameter information input by the risk assessment unit through bluetooth wireless communication; the parameter evaluation unit analyzes the user physical parameter data through an analysis and evaluation algorithm to evaluate the physical condition of the user; the risk prediction unit predicts the health risk condition of the user through big data analysis and machine learning, the output end of the parameter receiving unit is connected with the input end of the parameter evaluation unit, and the output end of the parameter evaluation unit is connected with the input end of the risk prediction unit.
As a further description of the above technical solution, the analysis and evaluation algorithm refers to quantitatively evaluating the model and the algorithm through a series of steps including selecting an index, preprocessing data, dividing a data set, training the model, evaluating the performance of the model, and analyzing the result, so as to select the best algorithm and model when solving the actual problem, where the implementation steps of the analysis and evaluation algorithm are as follows:
1) Determining an evaluation target and an index: firstly, determining a target to be evaluated and an index of interest;
2) Data preprocessing: the data preprocessing is to clean and prepare the data for subsequent evaluation;
3) Dividing a training set and a testing set: in order to evaluate the generalization ability of the algorithm, the data set needs to be divided into a training set and a testing set;
4) Algorithm implementation and training: training the model by the training set according to a data optimization function, wherein the formula expression of the data optimization function is as follows:
in formula (3), U represents a data optimization function, α i Representing the value of the data to be optimized, beta i Represents a standard value of the definition,an optimization factor representing a data optimization function;
5) Model evaluation: calculating the performance of a model on a test set according to a model evaluation function, wherein the formula expression of the model evaluation function is as follows:
in the formula (4), P (g) represents a model evaluation function, Φ i Representing model evaluation coefficient values, m representing model evaluation factors, and n representing performance indexes in the test set;
6) Analysis and interpretation of results: and analyzing according to the evaluation result, comparing the performance differences of different algorithms or parameter settings, and finding out the advantages and disadvantages of the model and the room for improvement.
As a further description of the above technical solution, the feedback module includes a warning unit, a suggestion unit, a reminding unit and a visualization unit, where the warning unit alerts the user to the physical health condition at the moment in the form of vibration, sound and popup; the suggestion unit provides improvement comments to a user in the form of words, images and audio based on the risk assessment module; the reminding unit prompts the user to exercise, take medicine and exercise through message pushing so as to realize supervision on the health of the user; the visualization unit displays the development trend of the health level to the user in a graph and curve mode based on the electronic screen.
The invention has the beneficial technical effects that compared with the prior art: the invention discloses a method for predicting health risk through an intelligent watch, wherein a characteristic value is extracted through a characteristic extraction algorithm to realize the utilization of effective information, so that the accuracy of a health detection result is improved; the analysis and evaluation of the physical condition of the user are realized through an analysis and evaluation algorithm, and the stability of risk prediction is improved; the development trend of the health level is displayed to the user through the visualization unit, so that the user's knowledge of the development of the health condition is improved.
Drawings
In order to more intuitively and clearly understand and understand the technical solution, when describing the embodiment of the present invention or the prior art, the drawings are often used for supplementing and describing, it should be noted that the drawing is only an expression mode of the embodiment of the present invention or the prior art, and in fact, the technical solution may also have other implementation modes and changes, which are all within the scope of protection of the present invention, so that a skilled person can design other drawings as needed to implement the technical solution of the present invention, where,
FIG. 1 is a schematic diagram of the overall architecture of the present invention;
FIG. 2 is a schematic diagram of a data acquisition module according to the present invention;
FIG. 3 is a schematic diagram of a data analysis module according to the present invention;
FIG. 4 is a schematic diagram of a risk assessment module according to the present invention;
fig. 5 is a schematic diagram of a data management module according to the present invention.
Detailed Description
Technical solutions in the embodiments herein will be clearly and completely omitted from descriptions of well-known structures and techniques by way of the drawings in the embodiments herein, so as to avoid unnecessarily obscuring the concepts of the present invention. It will be apparent that the embodiments described are only some, but not all, of the embodiments herein. Meanwhile, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present invention.
As shown in fig. 1-5, a method for predicting health risk by a smart watch, comprising the steps of,
step 1, acquiring and recording data information of user activities;
collecting and storing data information of user activities through a data acquisition module;
step 2, processing and analyzing the acquired data information;
processing and analyzing the acquired data information through a data analysis module; the data analysis module comprises a data preprocessing unit, a feature extraction unit and a data analysis unit, wherein the data preprocessing unit performs preprocessing operation on collected original data through cleaning, sorting, denoising and missing value filling so as to realize the integrity and the efficiency of the data; the feature extraction unit extracts feature values from the preprocessed data through a feature extraction algorithm so as to realize the utilization of effective information; the data analysis unit analyzes the association and rule between the physiological parameter information of the user through machine learning and big data, the output end of the data preprocessing unit is connected with the input end of the characteristic extraction unit, and the output end of the characteristic extraction unit is connected with the input end of the data analysis unit;
step 3, comparing user data information to perform health detection;
detecting the health condition of the user through a health detection module;
step 4, predicting the health risk of the user according to the risk assessment;
carrying out risk assessment on physical conditions of the user through a risk assessment module; the risk assessment module comprises a risk modeling unit, a risk verification unit, a basic data acquisition unit and a risk assessment unit, wherein the risk modeling unit is used for carrying out risk assessment on physical conditions of a user by constructing a risk assessment model; the risk verification unit verifies the working performance of the risk assessment model through the training data set and the test data set; the basic data acquisition unit actively sets personal information through a touch screen user to provide basic data information of the user; the risk assessment unit inputs processed data into the risk assessment model through Bluetooth wireless communication to achieve assessment of health risks of users, the output end of the risk modeling unit is connected with the input end of the risk verification unit, the output end of the risk verification unit is connected with the input end of the data acquisition unit, and the output end of the data acquisition unit is connected with the input end of the data assessment unit;
step 5, providing improvement suggestions according to the risk assessment and prediction results;
and providing expertise according to the risk assessment result through a result feedback module.
In a further embodiment, the data acquisition module includes a function sensing unit, an activity monitoring unit and a data recording unit, wherein the function sensing unit senses physiological parameter information of a user by adopting a sensor; the activity monitoring unit monitors the activity state of the user in real time through the gyroscope and the accelerometer so as to classify and identify the behavior mode of the user; the data recording unit adopts an EEPROM model memory to store data information of user activities, the output end of the function sensing unit is connected with the input end of the activity monitoring unit, and the output end of the activity monitoring unit is connected with the input end of the data recording unit.
The working principle of the data acquisition module is as follows: the function sensing unit senses physiological parameter information of a user, such as heart rate, body temperature, blood pressure and the like, by adopting a sensor, so that the physical condition of the user can be objectively and accurately known; the activity monitoring unit monitors the activity state of the user in real time through the gyroscope and the accelerometer, wherein the activity state comprises the step number, the movement distance, the consumed heat and the like, and further analyzes the movement behavior and the living habit of the user. The data acquisition module can better know the behavior mode of the user, and even can discover the physical abnormality of the user in advance by identifying abnormal activities so as to intervene in time; the data recording unit is responsible for storing the acquired user data information, and the EEPROM type memory is used for storing, so that the safety of long-term storage and transmission of data is ensured. Meanwhile, in the data recording unit, user behavior data can be anonymized by the system, so that the privacy of a user is protected, and meanwhile, a more comprehensive basis can be provided for data statistics and analysis.
In a further embodiment, the feature extraction algorithm is an algorithm for extracting feature information valuable for solving a problem from raw data by processing and analyzing the raw data, and converting complex raw data into a simple, representative and easy-to-understand feature representation form, and the implementation steps of the feature extraction algorithm are as follows:
1) Data preprocessing: processing the original data, including cleaning the data, removing abnormal values or noise and filling missing values;
2) And (3) data transformation: transforming the data according to a transformation function to better adapt to the feature extraction algorithm, wherein the transformation function has a formula:
in the formula (1), T represents a transformation function, λ represents a value of data to be transformed, μ represents a defined common standard value, ζ represents a transformation factor, K (λ, μ) represents a sum of the data to be transformed and the standard value, and L (λ, μ) represents a difference between the data to be transformed and the standard value;
3) Feature selection: selecting the most representative and distinguishing features according to the problem to be solved and the factors of the features;
4) And (3) feature construction: constructing new features by combining the original features;
5) Feature dimension reduction: when the feature dimension is too high and redundant features exist, the movement can reduce the dimension so as to reduce the feature quantity;
6) Feature evaluation and verification: the extracted features are evaluated and verified through a feature verification function, the contribution degree and effect of the extracted features on the problem are checked, and the formula expression of the feature verification function is as follows:
in the formula (2), P represents a feature verification function, τ represents an extracted feature coefficient value, θ represents a verification factor value of the feature verification function, and ω represents an effect ratio of a user physiological parameter in the feature verification function.
The working principle of the feature extraction algorithm is as follows: the main function of the feature extraction algorithm is to process and analyze the original data, extract the algorithm of valuable feature information for solving the problem from the original data, select the most representative and distinguishing features according to the data property and the problem requirement by methods such as mathematics, statistics, machine learning and the like, so as to support the data analysis, modeling and decision making processes. The special featureThe problem solved by the syndrome extraction algorithm is how to extract the most representative and discriminative features from the raw data to support data analysis, modeling and decision making. By extracting valid features, a deeper understanding of the data and more accurate predictions can be achieved. By the difference between the sum K (lambda, mu) of the data to be converted and the standard value and the difference L (lambda, mu) of the data to be converted and the standard value, the difference between the data is calculated and byConverting the data; evaluating and verifying the extracted characteristic value under the calculation of tau characteristic coefficient and theta verification factor, and for +.>And integrating to obtain the contribution degree of the characteristic data to the problem. Specific data processing software and programming tools are utilized to realize specific operation of the algorithm, meanwhile, massive data sets can be processed by means of carriers such as cloud computing, a large data platform and the like, more powerful computing resources are provided, the auxiliary means can improve efficiency and expandability of the feature extraction algorithm, and meanwhile, differences exist among different methodologies of the feature extraction algorithm, as shown in table 1:
TABLE 1 differences between different feature extraction algorithms
In summary, the feature accuracy, the extraction efficiency and the data contribution degree of the feature extraction algorithm are 98.13%, 96.58% and 89.66%, respectively, and the reaction speed of the feature extraction algorithm is faster than that of the convolutional neural network algorithm and the word bag model algorithm, so that the working efficiency of the feature extraction algorithm is improved, and the feature extraction algorithm is the best choice of the invention.
In a further embodiment, the health detection module comprises a temperature detection unit, a heart rate detection unit, a sleep detection unit, an blood oxygen detection unit and a blood pressure detection unit, wherein the temperature detection unit monitors the skin temperature of the user in real time through a temperature sensor so as to detect the health condition of the user; the heart rate detection unit detects heart rate data of a user through a photoelectric sensor; the sleep detection unit detects the sleep quality and the sleep duration of the user through the movement amplitude of the user when sleeping; the blood oxygen detection unit senses the blood oxygen saturation of the user through the infrared sensor so as to detect blood oxygen data of the user; the blood pressure detection unit measures a pressure change between the user and the skin by a pressure sensor to detect a blood pressure condition of the user.
The working principle of the health detection module is as follows: the temperature detection unit monitors the skin temperature of the user in real time by using the temperature sensor, and can know the body temperature change of the user by monitoring the skin temperature so as to judge the physical health condition of the user. For example, if the skin temperature of the user is abnormally elevated, it may be in fever or in a state of fatigue; the heart rate detection unit detects heart rate data of a user by using a photoelectric sensor, and the heart rate is one of important physiological indexes of a human body and reflects the functional state of a cardiovascular system. By monitoring the heart rate, we can determine the heart health and the degree of physical stress of the user, and in case of high frequency work or stress, the heart rate of the user may increase, which may be a stress response; the sleep detection unit detects a sleep quality and a sleep duration of a user using the motion amplitude. Research has shown that sleep is very important for human physical and mental health. By monitoring the movement amplitude of the user during sleep, the user can know the sleep quality of the user and judge whether the user has enough rest; the blood oxygen detection unit senses the blood oxygen saturation of the user by using the infrared sensor, the blood oxygen saturation is an important index for measuring the oxygen supply condition of the human body, and the health condition of the respiratory and circulatory system of the user can be judged by monitoring the blood oxygen saturation. For example, if the blood oxygen saturation is below the normal range, it may mean that the user has respiratory problems or circulatory system anomalies; the blood pressure detection unit measures a pressure change between a user and skin using a pressure sensor, thereby detecting a blood pressure condition of the user. Blood pressure is one of the important indexes for assessing cardiovascular health, and by monitoring blood pressure, a user can judge whether the user has the problems of hypertension or hypotension and the like.
In a further embodiment, the risk assessment model includes a parameter receiving unit, a parameter assessment unit and a risk prediction unit, where the parameter receiving unit receives parameter information input by the risk assessment unit through bluetooth wireless communication; the risk assessment unit analyzes the user physical parameter data through an analysis and assessment algorithm to assess the physical health condition of the user; the risk prediction unit predicts the health risk condition of the user through big data analysis and machine learning, the output end of the parameter receiving unit is connected with the input end of the parameter evaluation unit, and the output end of the parameter evaluation unit is connected with the input end of the risk prediction unit.
The working principle of the risk assessment model is as follows: in the risk assessment model, the parameter receiving unit plays a key role, and parameter information from the risk assessment unit, which may include physical measurement data of the user, such as heart rate, blood pressure, blood sugar, etc., can be received through bluetooth wireless communication, and these important parameter information are transmitted to the parameter assessment unit through the parameter receiving unit so as to start assessment of the physical health condition of the user. The parameter receiving unit adopts a Bluetooth wireless communication technology to transmit parameter information required by the risk assessment unit into the model, wherein the parameter information comprises physiological indexes, physical conditions, living habits and the like of a user, and the data quality and accuracy of the model are ensured through efficient transmission and real-time reception of data; the parameter evaluation unit utilizes an advanced analysis evaluation algorithm to comprehensively analyze and evaluate physical parameter data of the user, and evaluates the physical health condition of the user by fine interpretation of physiological indexes of the user and combining medical knowledge and professional experience, so that the scientific evaluation method has high reliability and accuracy, can help the user to know the physical condition of the user and timely take corresponding health management measures; the risk prediction unit predicts health risk conditions of users based on modeling and analysis of a large amount of user data by utilizing big data analysis and machine learning technology, and can accurately predict health risks possibly faced by users in the future by deep mining and comprehensive analysis of user historical data and take corresponding preventive measures in advance.
In a further embodiment, the analysis and evaluation algorithm refers to quantitative evaluation of the model and the algorithm through a series of steps including index selection, data preprocessing, data set partitioning, model training, model performance evaluation and result analysis, so as to select the best algorithm and model when solving the actual problem, wherein the implementation steps of the analysis and evaluation algorithm are as follows:
1) Determining an evaluation target and an index: firstly, determining a target to be evaluated and an index of interest;
2) Data preprocessing: the data preprocessing is to clean and prepare the data for subsequent evaluation;
3) Dividing a training set and a testing set: in order to evaluate the generalization ability of the algorithm, the data set needs to be divided into a training set and a testing set;
4) Algorithm implementation and training: training the model by the training set according to a data optimization function, wherein the formula expression of the data optimization function is as follows:
in formula (3), U represents a data optimization function, α i Representing the value of the data to be optimized, beta i Represents a standard value of the definition,an optimization factor representing a data optimization function;
5) Model evaluation: calculating the performance of a model on a test set according to a model evaluation function, wherein the formula expression of the model evaluation function is as follows:
in the formula (4), P (g) represents a model evaluation function, Φ i Representing model evaluation coefficient values, m representing model evaluation factors, and n representing performance indexes in the test set;
6) Analysis and interpretation of results: and analyzing according to the evaluation result, comparing the performance differences of different algorithms or parameter settings, and finding out the advantages and disadvantages of the model and the room for improvement.
The working principle of the analysis and evaluation algorithm is as follows: the analysis and evaluation algorithm refers to quantitative evaluation of the model and the algorithm through a series of steps, including index selection, data preprocessing, data set division, model training, model performance evaluation and result analysis, so as to select the best algorithm and model when solving the actual problem. The function of the analytical evaluation algorithm is to evaluate the performance of the model or algorithm in solving a particular problem, help select the optimal model or algorithm, and provide guidance for further improvement. The analysis and evaluation algorithm solves the problem of selecting the most appropriate one of the various models or algorithms and comparing the performance differences of the different models or algorithms over a particular problem. By passing throughThe model matrix is calculated to obtain a test training model, the data is evaluated according to the performance of the test training model, and meanwhile, the Python programming language and MATLAB software can more conveniently realize the algorithm and perform the data preprocessing.
In a further specific embodiment, the work-dependent hardware carrier of the analytical evaluation algorithm model mainly comprises the following aspects: CPU: the algorithm model requires a large amount of computation and processing and therefore requires a powerful CPU to support. Generally, the more cores the CPU, the faster the processor speed, and the higher the efficiency of the algorithm model operation. Gpu: GPUs are hardware dedicated to graphics processing, but are also widely used in the field of machine learning and deep learning due to their high parallelism and powerful computing power. The training speed and the reasoning speed of the algorithm model can be greatly improved by using the GPU. 3. Memory: the algorithm model requires a large amount of memory to store data and intermediate results. If the memory is insufficient, the algorithm model cannot normally operate or the operation efficiency is low. 4. A memory: the algorithm model requires reading data and model parameters from memory, and also requires saving the trained model to memory. Therefore, the speed and capacity of the memory is also an important factor on which the algorithm model operates. 5. Network: the speed and stability of the network is also important if the algorithm model needs to acquire data from or send results to a remote server. In summary, the hardware carrier on which the algorithm model depends needs to have the characteristics of strong computing power, high-capacity memory and storage, high-speed network connection, and the like. Meanwhile, in order to improve the running efficiency of the algorithm model, a proper hardware combination needs to be selected, for example, GPU is used for accelerating calculation, SSD is used for replacing a traditional mechanical hard disk, and the like.
It can be seen from the above algorithm that the analytical evaluation algorithm model can be realized by constructing a hardware carrier. The above is an exemplary scenario, and other algorithms operate in the same manner as the algorithm.
In the process of evaluating data performance, different algorithms have different performances, such as the differences between the analysis evaluation algorithm and the cross-validation algorithm are shown in table 2:
table 2 analysis of the differences between the evaluation algorithm and the cross-validation algorithm
In summary, the average evaluation rate, data integrity and evaluation accuracy of the analytical evaluation algorithm were 95.74%, 98.02% and 93.67%, respectively; meanwhile, the average evaluation rate, the data integrity and the evaluation accuracy of the cross-validation algorithm are 91.74%, 93.34% and 89.95%, respectively, so that the analysis evaluation algorithm is the best choice of the invention.
In a further embodiment, the feedback module includes a warning unit, a suggestion unit, a reminding unit and a visualization unit, where the warning unit alerts the user about the physical health condition at the moment in the form of vibration, sound and popup window; the suggestion unit provides improvement comments to a user in the form of words, images and audio based on the risk assessment module; the reminding unit prompts the user to exercise, take medicine and exercise through message pushing so as to realize supervision on the health of the user; the visualization unit displays the development trend of the health level to the user in a graph and curve mode based on the electronic screen.
The working principle of the feedback module is as follows: the warning unit can timely warn the user through vibration, sound and popup window to remind the user of paying attention to the current physical health condition, whether the blood pressure is high, the blood sugar is abnormal or other physical disorder is caused, and the user can realize the existing problems through the warning and take corresponding countermeasures; the suggestion unit provides the user with improvement suggestions in the form of words, images and audio based on the result of the risk assessment module, and the suggestion unit can customize personalized suggestions according to the individual situation of the user, whether the user is eating habits, life style or daily exercises. Through these suggestions, the user can understand his own health risks and take corresponding actions to improve the physical condition; the reminding unit can send a reminder to a user in a message pushing mode, for example, the user is reminded of regularly exercising, taking medicines on time or participating in sports, and the like, and the reminding function is beneficial to the user to establish health habits and keep a regular life rhythm. Meanwhile, the reminding unit can also monitor the health behaviors of users, so that the users can perform related activities on time to promote the development of health; the visual unit intuitively displays the development trend of the health level of the user through the chart and the curve displayed on the electronic screen. By means of the visualization mode, the user can know the health condition of the user more clearly and track the change of the health index. Such data display not only can motivate the user to pay more attention to his own health, but also can provide a reference for medical staff to perform diagnosis and treatment more accurately.
In a specific embodiment, firstly, the smart watch senses physiological function states of a user, such as heart rate, body temperature and the like, through the built-in sensor, and collects and records the data, and meanwhile, the smart watch records and stores the data collected from the function sensing unit and the activity monitoring unit for later analysis and use. Then, preprocessing the data collected from the intelligent watch, including cleaning, denoising, filling missing values and the like, so as to improve the quality of subsequent data analysis, extracting characteristic information with representativeness and distinguishing degree from the preprocessed data through the characteristic extraction algorithm, the statistical method and the machine learning technology, describing and quantifying the health state and the behavior characteristics of the user, and further analyzing, modeling and predicting the extracted characteristics by using various analysis methods and models, such as machine learning, statistical analysis and the like, so as to evaluate the health risk of the user. The smart watch detects the body temperature, heart rate, sleep quality, blood oxygen and blood pressure of the user by sensing the physiological parameters of the user. Then, based on the collected data and the existing health risk knowledge, the risk assessment model is established, the relation and the weight among a plurality of risk factors are considered, and the established risk assessment model is used for verifying and analyzing the health condition of the user to assess the potential risk of the user. Furthermore, related basic data are obtained from data sources such as medical records, personal files and the like and used as input of risk assessment, the basic data and risk factors are synthesized, and the health risk of a user is assessed and predicted through a risk assessment model; finally, according to the result of the risk assessment, the intelligent watch sends an alarm signal to the user to remind the user of possible health risks, and according to the risk assessment and the personal situation of the user, the intelligent watch provides specific health management suggestions and behavior change suggestions to help the user take corresponding measures and actions. Meanwhile, through the reminding function of the intelligent watch, reminding and notification of health management, such as regular medicine taking reminding, sports reminding and the like, are sent to a user regularly or in real time. The intelligent watch also visually displays the health data and the analysis result to the user in the form of a chart, a graph or an interface, so that the user can be helped to more intuitively understand and manage the health condition of the user.
While the invention has been described in terms of the above specific embodiments, it will be appreciated by those skilled in the art that these embodiments are provided by way of example only and do not limit the scope and application of the invention. Various omissions, substitutions and changes in the form and details of the invention may be made by those skilled in the art to achieve substantially similar results without departing from the spirit and scope of the invention. Accordingly, the scope of the invention is limited only by the following claims.
Claims (7)
1. A method for predicting health risk by a smart watch, characterized by: the method comprises the following steps:
step 1, acquiring and recording data information of user activities;
collecting and storing data information of user activities through a data acquisition module;
step 2, processing and analyzing the acquired data information;
processing and analyzing the acquired data information through a data analysis module; the data analysis module comprises a data preprocessing unit, a feature extraction unit and a data analysis unit, wherein the data preprocessing unit performs preprocessing operation on collected original data through cleaning, sorting, denoising and missing value filling so as to realize the integrity and the efficiency of the data; the feature extraction unit extracts feature values from the preprocessed data through a feature extraction algorithm so as to realize the utilization of effective information; the data analysis unit analyzes the association and rule between the physiological parameter information of the user through machine learning and big data, the output end of the data preprocessing unit is connected with the input end of the characteristic extraction unit, and the output end of the characteristic extraction unit is connected with the input end of the data analysis unit;
step 3, comparing user data information to perform health detection;
detecting the health condition of the user through a health detection module;
step 4, predicting the health risk of the user according to the risk assessment;
carrying out risk assessment on physical conditions of the user through a risk assessment module; the risk assessment module comprises a risk modeling unit, a risk verification unit, a basic data acquisition unit and a risk assessment unit, wherein the risk modeling unit is used for carrying out risk assessment on physical conditions of a user by constructing a risk assessment model; the risk verification unit verifies the working performance of the risk assessment model through the training data set and the test data set; the basic data acquisition unit actively sets personal information through a touch screen user to provide basic data information of the user; the risk assessment unit inputs processed data into the risk assessment model through Bluetooth wireless communication to achieve assessment of health risks of users, the output end of the risk modeling unit is connected with the input end of the risk verification unit, the output end of the risk verification unit is connected with the input end of the data acquisition unit, and the output end of the data acquisition unit is connected with the input end of the data assessment unit;
step 5, providing improvement suggestions according to the risk assessment and prediction results;
and providing expertise according to the risk assessment result through a result feedback module.
2. A method of predicting health risk with a smart watch as recited in claim 1, wherein: the data acquisition module comprises a function sensing unit, an activity monitoring unit and a data recording unit, wherein the function sensing unit senses physiological parameter information of a user by adopting a sensor; the activity monitoring unit monitors the activity state of the user in real time through the gyroscope and the accelerometer so as to classify and identify the behavior mode of the user; the data recording unit adopts an EEPROM model memory to store data information of user activities, the output end of the function sensing unit is connected with the input end of the activity monitoring unit, and the output end of the activity monitoring unit is connected with the input end of the data recording unit.
3. A method of predicting health risk with a smart watch as recited in claim 1, wherein: the feature extraction algorithm is an algorithm for extracting valuable feature information for solving problems from original data by processing and analyzing the original data, and converting complex original data into a simple, representative and easy-to-understand feature representation form, and the implementation steps of the feature extraction algorithm are as follows:
1) Data preprocessing: processing the original data, including cleaning the data, removing abnormal values or noise and filling missing values;
2) And (3) data transformation: transforming the data according to a transformation function to better adapt to the feature extraction algorithm, wherein the transformation function has a formula:
in the formula (1), T represents a transformation function, λ represents a value of data to be transformed, μ represents a defined common standard value, ζ represents a transformation factor, K (λ, μ) represents a sum of the data to be transformed and the standard value, and L (λ, μ) represents a difference between the data to be transformed and the standard value;
3) Feature selection: selecting the most representative and distinguishing features according to the problem to be solved and the factors of the features;
4) And (3) feature construction: constructing new features by combining the original features;
5) Feature dimension reduction: when the feature dimension is too high and redundant features exist, the movement can reduce the dimension so as to reduce the feature quantity;
6) Feature evaluation and verification: the extracted features are evaluated and verified through a feature verification function, the contribution degree and effect of the extracted features on the problem are checked, and the formula expression of the feature verification function is as follows:
in the formula (2), P represents a feature verification function, τ represents an extracted feature coefficient value, θ represents a verification factor value of the feature verification function, and ω represents an effect ratio of a user physiological parameter in the feature verification function.
4. A method of predicting health risk with a smart watch as recited in claim 1, wherein: the health detection module comprises a temperature detection unit, a heart rate detection unit, a sleep detection unit, a blood oxygen detection unit and a blood pressure detection unit, wherein the temperature detection unit monitors the skin temperature of a user in real time through a temperature sensor so as to detect the health condition of the user; the heart rate detection unit detects heart rate data of a user through a photoelectric sensor; the sleep detection unit detects the sleep quality and the sleep duration of the user through the movement amplitude of the user when sleeping; the blood oxygen detection unit senses the blood oxygen saturation of the user through the infrared sensor so as to detect blood oxygen data of the user; the blood pressure detection unit measures a pressure change between the user and the skin by a pressure sensor to detect a blood pressure condition of the user.
5. A method of predicting health risk with a smart watch as recited in claim 1, wherein: the risk assessment model comprises a parameter receiving unit, a parameter assessment unit and a risk prediction unit, wherein the parameter receiving unit receives parameter information input by the risk assessment unit through Bluetooth wireless communication; the parameter evaluation unit analyzes the user physical parameter data through an analysis and evaluation algorithm to evaluate the physical condition of the user; the risk prediction unit predicts the health risk condition of the user through big data analysis and machine learning, the output end of the parameter receiving unit is connected with the input end of the parameter evaluation unit, and the output end of the parameter evaluation unit is connected with the input end of the risk prediction unit.
6. A method of predicting health risk with a smart watch as recited in claim 1, wherein: the analysis and evaluation algorithm refers to quantitative evaluation of the model and the algorithm through a series of steps, including index selection, data preprocessing, data set division, model training, model performance evaluation and result analysis, so as to select the optimal algorithm and model when solving the actual problem, and the implementation steps of the analysis and evaluation algorithm are as follows:
1) Determining an evaluation target and an index: firstly, determining a target to be evaluated and an index of interest;
2) Data preprocessing: the data preprocessing is to clean and prepare the data for subsequent evaluation;
3) Dividing a training set and a testing set: in order to evaluate the generalization ability of the algorithm, the data set needs to be divided into a training set and a testing set;
4) Algorithm implementation and training: training the model by the training set according to a data optimization function, wherein the formula expression of the data optimization function is as follows:
in formula (3), U represents a data optimization function, α i Representing the value of the data to be optimized, beta i Represents a standard value of the definition,an optimization factor representing a data optimization function;
5) Model evaluation: calculating the performance of a model on a test set according to a model evaluation function, wherein the formula expression of the model evaluation function is as follows:
in the formula (4), P (g) represents a model evaluation function, Φ i Representing model evaluation coefficient values, m representing model evaluation factors, and n representing performance indexes in the test set;
6) Analysis and interpretation of results: and analyzing according to the evaluation result, comparing the performance differences of different algorithms or parameter settings, and finding out the advantages and disadvantages of the model and the room for improvement.
7. A method of predicting health risk with a smart watch as recited in claim 1, wherein: the feedback module comprises a warning unit, a suggestion unit, a reminding unit and a visualization unit, wherein the warning unit warns a user of the physical health condition at the moment in the forms of vibration, sound and popup window; the suggestion unit provides improvement comments to a user in the form of words, images and audio based on the risk assessment module; the reminding unit prompts the user to exercise, take medicine and exercise through message pushing so as to realize supervision on the health of the user; the visualization unit displays the development trend of the health level to the user in a graph and curve mode based on the electronic screen.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311573445.4A CN117653053A (en) | 2023-11-23 | 2023-11-23 | Method for predicting health risk through intelligent watch |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311573445.4A CN117653053A (en) | 2023-11-23 | 2023-11-23 | Method for predicting health risk through intelligent watch |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117653053A true CN117653053A (en) | 2024-03-08 |
Family
ID=90079991
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311573445.4A Pending CN117653053A (en) | 2023-11-23 | 2023-11-23 | Method for predicting health risk through intelligent watch |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117653053A (en) |
-
2023
- 2023-11-23 CN CN202311573445.4A patent/CN117653053A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10039485B2 (en) | Method and system for assessing mental state | |
| JP5841196B2 (en) | Residue-based management of human health | |
| US9861308B2 (en) | Method and system for monitoring stress conditions | |
| US20200205709A1 (en) | Mental state indicator | |
| US20210345897A1 (en) | Heart Rate Variability Composite Scoring and Analysis | |
| CN106714682A (en) | Apparatus, system, method and computer program for assessing the risk of an exacerbation and/or hospitalization | |
| KR102053604B1 (en) | Method for sleeping analysis and device for sleeping analysis using the same | |
| WO2019075522A1 (en) | Risk indicator | |
| US20210345896A1 (en) | Heart Rate Variability Monitoring and Analysis | |
| CN115721284A (en) | Health monitoring system, method, electronic device and storage medium | |
| US20230248320A1 (en) | Detection of User Temperature and Assessment of Physiological Symptoms with Respiratory Diseases | |
| Srinivasan et al. | A human-in-the-loop segmented mixed-effects modeling method for analyzing wearables data | |
| US20210345915A1 (en) | Methods Circuits Devices Systems and Machine Executable Code for Glucose Monitoring Analysis and Remedy | |
| CN117653053A (en) | Method for predicting health risk through intelligent watch | |
| JP2024513618A (en) | Methods and systems for personalized prediction of infections and sepsis | |
| Tolentino et al. | CAREdio: Health screening and heart disease prediction system for rural communities in the Philippines | |
| Shirwaikar et al. | Design framework for a data mart in the neonatal intensive care unit | |
| CN116098595B (en) | System and method for monitoring and preventing sudden cardiac death and sudden cerebral death | |
| EP4224373A1 (en) | System for forecasting a mental state of a subject and method | |
| Ji et al. | Measurement Scheduling for ICU Patients with Offline Reinforcement Learning | |
| Alharbi et al. | Complexity Analysis of Pulse Oximeter Signals for Predicting Changes in Cardiorespiratory control system with Aging | |
| GB2608766A (en) | Latent bio-signal estimation using bio-signal detectors |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |