Patient behavior monitoring method and system based on big data analysis

Abstract

The invention discloses a patient behavior monitoring method and system based on big data analysis. The big data analysis-based patient behavior monitoring method comprises the following steps: S110, collecting physiological feature data and behavior feature data of a patient; s120, preprocessing the collected physiological feature data and behavior feature data; s130, calculating a physiological feature abnormal value according to the preprocessed physiological feature data; s140, performing behavior pattern analysis according to the preprocessed behavior feature data and the physiological feature abnormal value; through real-time and comprehensive monitoring and analysis, a patient can better manage own health, and the quality and efficiency of medical services are greatly improved on the whole.

Description

Technical Field

[0001] The present invention relates to big data analysis technology, and in particular to a method and system for monitoring patient behavior based on big data analysis. Background Art

[0002] With the acceleration of digitalization in the field of medical and health care, a large amount of patient-related behavioral and physiological data can be collected in real time through wearable devices, health applications and medical equipment. Traditional medical systems usually rely on regular physical examinations and patient self-reporting, and cannot continuously monitor patients' daily behaviors, making it difficult to detect potential health problems in a timely manner. In recent years, the application of big data analysis technology has brought innovation to the field of medical and health care. By combining behavioral data with physiological data for analysis, health risks can be discovered earlier, personalized health management plans can be provided, and doctors can be helped to optimize treatment strategies and improve the efficiency of medical resource utilization. This behavior monitoring system based on big data is becoming an important part of modern medicine, promoting the development of disease prevention, chronic disease management and precision medicine. Summary of the invention

[0003] The present invention provides a patient behavior monitoring method based on big data analysis, comprising:

[0004] S110, collecting physiological characteristic data and behavioral characteristic data of the patient;

[0005] S120, preprocessing the collected physiological characteristic data and behavioral characteristic data;

[0006] S130, calculating physiological characteristic abnormality values ​​according to the preprocessed physiological characteristic data;

[0007] S140, performing behavior pattern analysis based on the preprocessed behavior characteristic data and physiological characteristic abnormal values.

[0008] As described above, a patient behavior monitoring method based on big data analysis, wherein collecting the patient's physiological characteristic data and behavioral characteristic data is the basis for constructing an effective patient behavior monitoring method, specifically comprises the following sub-steps:

[0009] Collect the patient's physiological characteristic data and establish a physiological characteristic data set;

[0010] Collect patients' behavioral characteristic data and establish a behavioral characteristic data set.

[0011] The patient behavior monitoring method based on big data analysis as described above, wherein the collected physiological characteristic data and behavioral characteristic data are preprocessed, specifically comprises the following sub-steps:

[0012] Fill missing values ​​in the collected physiological and behavioral characteristic data;

[0013] The collected physiological characteristic data and behavioral characteristic data are input with outlier processing.

[0014] The patient behavior monitoring method based on big data analysis as described above, wherein the abnormal value of physiological characteristics is calculated according to the preprocessed physiological characteristic data, specifically comprises the following sub-steps:

[0015] Calculate the outlier value of each data point based on the preprocessed physiological characteristic data set;

[0016] Calculate the abnormal confidence of each data point based on its abnormal value;

[0017] The physiological characteristic outlier value is calculated based on the outlier value of each data point and the outlier confidence of the data point.

[0018] The patient behavior monitoring method based on big data analysis as described above, wherein behavior pattern analysis is performed based on preprocessed behavior characteristic data and physiological characteristic abnormal values, specifically comprising the following sub-steps:

[0019] Calculate the correlation between behavioral and physiological characteristics;

[0020] Calculate the behavioral pattern risk index based on correlation and physiological characteristic abnormalities;

[0021] Conduct behavioral pattern analysis based on the behavioral pattern risk index.

[0022] The present invention also provides a patient behavior monitoring system based on big data analysis, including: a data collection module, a preprocessing module, a physiological characteristic abnormal value calculation module, and a behavior pattern analysis module.

[0023] Data collection module 21: collects physiological characteristic data and behavioral characteristic data of patients;

[0024] Preprocessing module 22: preprocessing the collected physiological characteristic data and behavioral characteristic data;

[0025] Physiological characteristic abnormal value calculation module 23: calculates the physiological characteristic abnormal value according to the preprocessed physiological characteristic data;

[0026] Behavior pattern analysis module 24: performs behavior pattern analysis based on the pre-processed behavior feature data and physiological feature abnormal values.

[0027] As described above, a system for patient behavior monitoring based on big data analysis, in which collecting the patient's physiological characteristic data and behavioral characteristic data is the basis for building an effective patient behavior monitoring method, specifically includes the following submodules:

[0028] Collect the patient's physiological characteristic data and establish a physiological characteristic data set;

[0029] Collect patients' behavioral characteristic data and establish a behavioral characteristic data set.

[0030] The patient behavior monitoring system based on big data analysis as described above, wherein the collected physiological characteristic data and behavioral characteristic data are preprocessed, specifically comprises the following submodules:

[0031] Fill missing values ​​in the collected physiological and behavioral characteristic data;

[0032] The collected physiological characteristic data and behavioral characteristic data are input with outlier processing.

[0033] As described above, a system for monitoring patient behavior based on big data analysis, in which abnormal physiological characteristic values ​​are calculated based on preprocessed physiological characteristic data, specifically comprises the following submodules:

[0034] Calculate the outlier value of each data point based on the preprocessed physiological characteristic data set;

[0035] Calculate the abnormal confidence of each data point based on its abnormal value;

[0036] The physiological characteristic outlier value is calculated based on the outlier value of each data point and the outlier confidence of the data point.

[0037] As described above, a system for monitoring patient behavior based on big data analysis, wherein behavior pattern analysis is performed based on preprocessed behavior feature data and physiological feature abnormal values, specifically comprises the following submodules:

[0038] Calculate the correlation between behavioral and physiological characteristics;

[0039] Calculate the behavioral pattern risk index based on correlation and physiological characteristic abnormalities;

[0040] Conduct behavioral pattern analysis based on the behavioral pattern risk index.

[0041] The beneficial effects achieved by the present invention are as follows: through real-time and comprehensive monitoring and analysis, patients can better manage their own health, which greatly improves the quality and efficiency of medical services as a whole. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0043] Figure 1 This is a flow chart of a method for patient behavior monitoring based on big data analysis provided in Example 1 of the present application;

[0044] Figure 2 This is a schematic diagram of a system for patient behavior monitoring based on big data analysis provided in Embodiment 2 of the present invention. DETAILED DESCRIPTION

[0045] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0046] Embodiment 1

[0047] like Figure 1 As shown, Embodiment 1 of the present application provides a patient behavior monitoring method based on big data analysis.

[0048] Step S110, collecting the patient's physiological characteristic data and behavioral characteristic data;

[0049] Collecting the patient's physiological and behavioral characteristics data is the basis for building an effective patient behavior monitoring method, which specifically includes the following sub-steps:

[0050] Step S111, collecting the patient's physiological characteristic data and establishing a physiological characteristic data set;

[0051] The data sources of physiological characteristic data include: electronic health records such as patient basic information, medical history, and imaging examination results stored in the database of the hospital or clinic; wearable devices such as smart watches and health trackers that monitor physiological indicators such as heart rate, blood pressure, body temperature, and blood oxygen saturation in real time; laboratory test results such as blood sugar and cholesterol levels of blood, urine and other samples, which are specifically divided into basic data sets: age, gender, weight, height; real-time monitoring data sets: heart rate, blood pressure, blood oxygen, body temperature; laboratory test result data sets: blood sugar, cholesterol, liver and kidney function, specifically, the formula is: D = {D basic ,D time ,D laboratory} represents a physiological characteristic data set, where D represents a physiological characteristic data set, D basic represents the basic data set, D time represents the real-time monitoring dataset, D laboratory Represents a laboratory test results dataset.

[0052] Step S112, collecting the patient's behavioral characteristic data and establishing a behavioral characteristic data set;

[0053] The data sources of behavioral characteristic data include: information on patients' daily activities, eating habits, medication, sleep quality, etc. collected through questionnaires; information on patients' exercise volume, steps, activity types, etc. recorded through mobile applications; by monitoring patients' health-related topics and interactions on social media, it is specifically divided into daily activity data sets: steps, exercise types, exercise duration; eating habit data sets: number of meals, calorie intake, food types; medication compliance data sets: on-time medication, missed medication records; psychological state data sets: anxiety, stress, depression levels, specifically, using the formula:

[0054] R={R activity ,R dietary ,R medication ,R psychological} represents a behavioral feature data set, where R represents a behavioral feature data set, R activity represents the daily activity dataset, R dietary represents the dietary habits dataset, R medication represents the medication compliance dataset, R psychological Represents a mental state dataset.

[0055] Step S120, pre-processing the collected physiological characteristic data and behavioral characteristic data;

[0056] Preprocessing the collected physiological characteristic data and behavioral characteristic data includes the following sub-steps:

[0057] Step S121, filling missing values ​​in the collected physiological characteristic data and behavioral characteristic data;

[0058] When processing physiological characteristic data and behavioral characteristic data, a series of preprocessing steps are required to ensure the quality and availability of the data. If there are missing values ​​in physiological or behavioral characteristic data, for numerical features, the mean of the feature is used to fill the missing values. Specifically, the formula is used:

[0059] Indicates missing value filling, where Filled(D) represents the missing items of the physiological feature data set D, n represents the total number of missing items in the physiological feature data set, i represents the coefficient of the missing items, and its value is 1-n. Where Filled(R) represents the missing items of the behavioral feature data set R, m represents the total number of missing items in the behavioral feature data set, j represents the coefficient of the missing items, and its value is 1-m.

[0060] Step S122, performing input abnormal value processing on the collected physiological characteristic data and behavioral characteristic data;

[0061] Input outliers in physiological or behavioral characteristic data need to be identified and processed. If the z-score of a data point exceeds the set threshold, it is considered an input outlier and the data is deleted. The z-score is used to measure the degree of deviation of a single data point from the average value of the data set. Specifically, the formula is used: Indicates input outlier processing, where del() represents a function that deletes data from a data set. represents the z-score of the ith physiological characteristic data, Z D Indicates the set threshold of physiological characteristic data, represents the z-score of the j-th behavior feature data, Z R Indicates the set threshold of the behavior characteristic data.

[0062] Step S130, calculating the physiological characteristic abnormality value according to the preprocessed physiological characteristic data;

[0063] Calculating physiological characteristic abnormal values ​​according to the preprocessed physiological characteristic data specifically includes the following sub-steps:

[0064] Step S131, calculating the outlier value of each data point according to the preprocessed physiological characteristic data set;

[0065] The outlier value of each data point is calculated based on the preprocessed physiological characteristic data set. The determination of outliers is defined by setting upper and lower boundaries. If the value of a data point is less than the lower bound or greater than the upper bound, the data point is considered an outlier. Specifically, the formula is: Determine outliers, where D unusual represents the set of abnormal values ​​of physiological characteristic data, D basic represents the basic data set, D time represents the real-time monitoring dataset, D laboratory represents a laboratory test result dataset, represents the anomaly lower bound of the underlying dataset, represents the abnormal upper bound of the underlying data set, represents the lower bound of the anomaly of the real-time monitoring data set, represents the abnormal upper bound of the real-time monitoring data set, represents the lower bound of anomaly for the laboratory test result dataset, Represents the upper bound of abnormality for a dataset of laboratory test results.

[0066] Step S132, calculating the outlier confidence of each data point according to the outlier value of the data point;

[0067] For multi-dimensional physiological characteristic data sets, after performing anomaly detection on each data point, the abnormal results of each feature are comprehensively evaluated. If multiple features of a data point are judged as abnormal at the same time, the confidence that the data point is abnormal can be enhanced. The features include basic data, real-time monitoring data, and laboratory test result data. Specifically, the formula is used: Calculate the outlier confidence of a data point, where represents the data point of the outlier of the ith physiological characteristic data, represents the abnormal confidence of the data point with the outlier value of the i-th physiological characteristic data, represents the indicator function, when the feature If it is judged as an abnormal value, it takes 1, otherwise it takes 0. k represents the subscript of the physiological characteristic data subset. The value of k is 1-n, and the maximum value of n is 3.

[0068] Step S133, calculating the physiological characteristic abnormal value according to the abnormal value of each data point and the abnormal confidence of the data point;

[0069] The physiological characteristic abnormal value is calculated according to the abnormal value of each data point and the abnormal confidence of the data point. The physiological characteristic abnormal value is the product sum of the abnormal value of each data point and the abnormal confidence of the data point. Specifically, the formula is used: Calculate the abnormal value of physiological characteristics, where d represents the abnormal value of physiological characteristics, represents the outlier value of the i-th data point, Represents the abnormal confidence of the i-th data point.

[0070] Step S140, performing behavior pattern analysis based on the preprocessed behavior characteristic data and physiological characteristic abnormal values;

[0071] Certain behavioral characteristics may be associated with abnormal physiological characteristics. For example, too little exercise or too much high-intensity exercise may lead to abnormal heart rate or blood pressure. Bad eating habits (high calorie intake, nutrient-deficient diet) may be associated with abnormal blood sugar or cholesterol. Lack of sleep or excessive stress may affect body temperature, heart rate, etc. The present invention analyzes the patient's behavior pattern based on the patient's behavior characteristic data and physiological characteristic abnormal values, calculates the behavior pattern risk index by detecting the correlation between the behavior characteristics and the physiological characteristics and the comprehensive evaluation of the physiological characteristic abnormal values, and thus analyzes the patient's behavior pattern. Specifically, the present invention includes the following sub-steps:

[0072] Step S141, calculating the correlation between the behavioral characteristics and the physiological characteristics;

[0073] According to the behavioral feature data set and the physiological feature data set, the correlation between the behavioral feature and the physiological feature is calculated. Specifically, the formula is used: Calculate the correlation, where represents the correlation between the ith behavioral characteristic and the physiological characteristic, The value range of is [-1,1]. When , it means that the behavioral characteristics are negatively correlated with the physiological characteristics. When , it means that the behavioral characteristics are positively correlated with the physiological characteristics. When , it means that the behavioral characteristics have no correlation with the physiological characteristics, D i Represents the characteristic value of physiological data, R i Represents the characteristic value of behavioral data, represents the mean value of physiological characteristics, Represents the mean of behavioral characteristics.

[0074] Step S142, calculating the behavior pattern risk index according to the correlation and the abnormal value of the physiological characteristics;

[0075] The present invention proposes a behavior pattern risk index to assess the degree of influence of a person's behavior pattern on his or her physiological abnormal values. The index is based on a comprehensive assessment of correlation and abnormal values ​​to help identify whether certain behavior patterns will significantly affect the abnormal risk of physiological characteristics. Specifically, the formula is used: Calculate the behavior pattern risk index, where BPRI represents the behavior pattern risk index, λ i is the weight factor, which indicates the contribution of the ith behavior feature to the risk index. It represents the correlation between the ith behavioral characteristic and the physiological characteristic, and d represents the abnormal value of the physiological characteristic.

[0076] Step S143, performing behavior pattern analysis according to the behavior pattern risk index;

[0077] If a patient has a high BPRI value, it means that their behavior patterns, such as lack of exercise and unhealthy diet, are strongly associated with abnormal physiological characteristics. At this time, we can further analyze which specific behavioral characteristics, such as number of steps, exercise duration, and diet, cause higher risks. Individuals with lower BPRI have healthier behavior patterns and no obvious association with abnormal physiological characteristics. Such individuals have lower health risks.

[0078] By conducting statistical analysis on the BPRI of multiple individuals, we can identify which behavioral patterns are more common in the group and are highly correlated with health problems. We can divide individuals into high-risk, medium-risk, and low-risk groups, and analyze the behavioral patterns of these groups to find out which behavioral characteristics, such as lack of exercise and irregular diet, are more likely to lead to abnormal physiological characteristics.

[0079] For individuals with long-term monitoring data, time series analysis can be performed to observe how BPRI changes over time and whether BPRI decreases over time if individuals make positive behavioral adjustments, such as increasing exercise and improving diet, thereby reflecting improved health status.

[0080] Embodiment 2

[0081] like Figure 2 As shown, Embodiment 2 of the present application provides a patient behavior monitoring system based on big data analysis, including:

[0082] Data collection module 21: collects physiological characteristic data and behavioral characteristic data of patients;

[0083] Collecting the patient's physiological and behavioral characteristics data is the basis for building an effective patient behavior monitoring method, which specifically includes the following submodules:

[0084] Physiological characteristic data collection module: collects the patient's physiological characteristic data and establishes a physiological characteristic data set;

[0085] The data sources of physiological characteristic data include: electronic health records such as patient basic information, medical history, and imaging examination results stored in the database of the hospital or clinic; wearable devices such as smart watches and health trackers that monitor physiological indicators such as heart rate, blood pressure, body temperature, and blood oxygen saturation in real time; laboratory test results such as blood sugar and cholesterol levels of blood, urine and other samples, which are specifically divided into basic data sets: age, gender, weight, height; real-time monitoring data sets: heart rate, blood pressure, blood oxygen, body temperature; laboratory test result data sets: blood sugar, cholesterol, liver and kidney function, specifically, the formula is: D = {D basic ,D time ,D laboratory} represents a physiological characteristic data set, where D represents a physiological characteristic data set, D basic represents the basic data set, D time represents the real-time monitoring dataset, D laboratory Represents a laboratory test results dataset.

[0086] Behavior characteristic data collection module: collects the patient's behavior characteristic data and establishes a behavior characteristic data set;

[0087] The data sources of behavioral characteristic data include: information on patients' daily activities, eating habits, medication, sleep quality, etc. collected through questionnaires; information on patients' exercise volume, steps, activity types, etc. recorded through mobile applications; by monitoring patients' health-related topics and interactions on social media, it is specifically divided into daily activity data sets: steps, exercise types, exercise duration; eating habit data sets: number of meals, calorie intake, food types; medication compliance data sets: on-time medication, missed medication records; psychological state data sets: anxiety, stress, depression levels, specifically, using the formula:

[0088] R={R activity ,R dietary ,R medication ,R psychological} represents a behavioral feature data set, where R represents a behavioral feature data set, R activity represents the daily activity dataset, R dietary represents the dietary habits dataset, R medication represents the medication compliance dataset, R psychological Represents a mental state dataset.

[0089] Preprocessing module 22: preprocessing the collected physiological characteristic data and behavioral characteristic data;

[0090] Preprocess the collected physiological and behavioral characteristic data, which specifically includes the following submodules:

[0091] Missing value supplement module: fill in missing values ​​for the collected physiological characteristic data and behavioral characteristic data;

[0092] When processing physiological characteristic data and behavioral characteristic data, a series of preprocessing steps are required to ensure the quality and availability of the data. If there are missing values ​​in physiological or behavioral characteristic data, for numerical features, the mean of the feature is used to fill the missing values. Specifically, the formula is used:

[0093] Indicates missing value filling, where Filled(D) represents the missing items of the physiological feature data set D, n represents the total number of missing items in the physiological feature data set, i represents the coefficient of the missing items, and its value is 1-n. Where Filled(R) represents the missing items of the behavioral feature data set R, m represents the total number of missing items in the behavioral feature data set, j represents the coefficient of the missing items, and its value is 1-m.

[0094] Input abnormal value processing module: performs input abnormal value processing on the collected physiological characteristic data and behavioral characteristic data;

[0095] Input outliers in physiological or behavioral characteristic data need to be identified and processed. If the z-score of a data point exceeds the set threshold, it is considered an input outlier and the data is deleted. The z-score is used to measure the degree of deviation of a single data point from the average value of the data set. Specifically, the formula is used: Indicates input outlier processing, where del() represents a function that deletes data from a data set. represents the z-score of the ith physiological characteristic data, Z D Indicates the set threshold of physiological characteristic data, represents the z-score of the j-th behavior feature data, Z R Indicates the set threshold of the behavior characteristic data.

[0096] Physiological characteristic abnormal value calculation module 23: calculates the physiological characteristic abnormal value according to the preprocessed physiological characteristic data;

[0097] Calculate the physiological characteristic abnormal value based on the preprocessed physiological characteristic data, which specifically includes the following submodules:

[0098] Outlier calculation module: calculates the outlier value of each data point based on the preprocessed physiological characteristic data set;

[0099] The outlier value of each data point is calculated based on the preprocessed physiological characteristic data set. The determination of outliers is defined by setting upper and lower boundaries. If the value of a data point is less than the lower bound or greater than the upper bound, the data point is considered an outlier. Specifically, the formula is: Determine outliers, where D unusual represents the set of abnormal values ​​of physiological characteristic data, D basic represents the basic data set, D time represents the real-time monitoring dataset, D laboratory represents a laboratory test result dataset, represents the anomaly lower bound of the underlying dataset, represents the abnormal upper bound of the underlying data set, represents the lower bound of the anomaly of the real-time monitoring data set, represents the abnormal upper bound of the real-time monitoring data set, represents the lower bound of anomaly for the laboratory test result dataset, Represents the upper bound of abnormality for a dataset of laboratory test results.

[0100] Anomaly confidence calculation module: calculates the anomaly confidence of each data point based on its anomaly value;

[0101] For multi-dimensional physiological characteristic data sets, after performing anomaly detection on each data point, the abnormal results of each feature are comprehensively evaluated. If multiple features of a data point are judged as abnormal at the same time, the confidence that the data point is abnormal can be enhanced. The features include basic data, real-time monitoring data, and laboratory test result data. Specifically, the formula is used: Calculate the outlier confidence of a data point, where represents the data point of the outlier of the ith physiological characteristic data, represents the abnormal confidence of the data point with the outlier value of the i-th physiological characteristic data, represents the indicator function, when the feature If it is judged as an abnormal value, it takes 1, otherwise it takes 0. k represents the subscript of the physiological characteristic data subset. The value of k is 1-n, and the maximum value of n is 3.

[0102] Physiological characteristic outlier module: calculates the physiological characteristic outlier value based on the outlier value of each data point and the outlier confidence of the data point;

[0103] The physiological characteristic abnormal value is calculated according to the abnormal value of each data point and the abnormal confidence of the data point. The physiological characteristic abnormal value is the product sum of the abnormal value of each data point and the abnormal confidence of the data point. Specifically, the formula is used: Calculate the abnormal value of physiological characteristics, where d represents the abnormal value of physiological characteristics, represents the outlier value of the i-th data point, Represents the abnormal confidence of the i-th data point.

[0104] Behavior pattern analysis module 24: performs behavior pattern analysis based on pre-processed behavior characteristic data and physiological characteristic abnormal values;

[0105] Certain behavioral characteristics may be associated with abnormal physiological characteristics. For example, too little exercise or too much high-intensity exercise may lead to abnormal heart rate or blood pressure. Bad eating habits (high calorie intake, nutrient-deficient diet) may be associated with abnormal blood sugar or cholesterol. Lack of sleep or excessive stress may affect body temperature, heart rate, etc. The present invention analyzes the patient's behavior pattern through the patient's behavior characteristic data and physiological characteristic abnormal values. By detecting the correlation between the behavior characteristics and the physiological characteristics and the comprehensive evaluation of the physiological characteristic abnormal values, the behavior pattern risk index is calculated, thereby analyzing the patient's behavior pattern. Specifically, it includes the following submodules:

[0106] Correlation calculation module: calculates the correlation between behavioral characteristics and physiological characteristics;

[0107] According to the behavioral feature data set and the physiological feature data set, the correlation between the behavioral feature and the physiological feature is calculated. Specifically, the formula is used: Calculate the correlation, where represents the correlation between the ith behavioral characteristic and the physiological characteristic, The value range of is [-1,1]. When , it means that the behavioral characteristics are negatively correlated with the physiological characteristics. When , it means that the behavioral characteristics are positively correlated with the physiological characteristics. When , it means that the behavioral characteristics have no correlation with the physiological characteristics, D i Represents the characteristic value of physiological data, R i Represents the characteristic value of behavioral data, represents the mean value of physiological characteristics, Represents the mean of behavioral characteristics.

[0108] Behavior pattern risk index calculation module: calculates the behavior pattern risk index based on correlation and physiological characteristic abnormal values;

[0109] The present invention proposes a behavior pattern risk index to assess the degree of influence of a person's behavior pattern on his or her physiological abnormal values. The index is based on a comprehensive assessment of correlation and abnormal values ​​to help identify whether certain behavior patterns will significantly affect the abnormal risk of physiological characteristics. Specifically, the formula is used: Calculate the behavior pattern risk index, where BPRI represents the behavior pattern risk index, λ i is the weight factor, which indicates the contribution of the ith behavior feature to the risk index. It represents the correlation between the ith behavioral characteristic and the physiological characteristic, and d represents the abnormal value of the physiological characteristic.

[0110] Analysis module: conduct behavior pattern analysis based on behavior pattern risk index;

[0111] If a patient has a high BPRI value, it means that their behavior patterns, such as lack of exercise and unhealthy diet, are strongly associated with abnormal physiological characteristics. At this time, we can further analyze which specific behavioral characteristics, such as number of steps, exercise duration, and diet, cause higher risks. Individuals with lower BPRI have healthier behavior patterns and no obvious association with abnormal physiological characteristics. Such individuals have lower health risks.

[0112] By conducting statistical analysis on the BPRI of multiple individuals, we can identify which behavioral patterns are more common in the group and are highly correlated with health problems. We can divide individuals into high-risk, medium-risk, and low-risk groups, and analyze the behavioral patterns of these groups to find out which behavioral characteristics, such as lack of exercise and irregular diet, are more likely to lead to abnormal physiological characteristics.

[0113] For individuals with long-term monitoring data, time series analysis can be performed to observe how BPRI changes over time and whether BPRI decreases over time if individuals make positive behavioral adjustments, such as increasing exercise and improving diet, thereby reflecting improved health status.

[0114] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solution of the present invention should be included in the scope of protection of the present invention.

Claims

1. A patient behavior monitoring method based on big data analysis, characterized in that: include: S110, collecting physiological characteristic data and behavioral characteristic data of the patient; S120, preprocessing the collected physiological characteristic data and behavioral characteristic data; S130, calculating physiological characteristic abnormality values ​​according to the preprocessed physiological characteristic data; S140, performing behavior pattern analysis based on the preprocessed behavior characteristic data and physiological characteristic abnormal values.

2. A patient behavior monitoring method based on big data analysis as claimed in claim 1, characterized in that: Collecting the patient's physiological and behavioral characteristics data is the basis for building an effective patient behavior monitoring method, which specifically includes the following sub-steps: Collect the patient's physiological characteristic data and establish a physiological characteristic data set; Collect patients' behavioral characteristic data and establish a behavioral characteristic data set.

3. A patient behavior monitoring method based on big data analysis as claimed in claim 1, characterized in that: Preprocessing the collected physiological characteristic data and behavioral characteristic data includes the following sub-steps: Fill missing values ​​in the collected physiological and behavioral characteristic data; The collected physiological characteristic data and behavioral characteristic data are input with outlier processing.

4. A patient behavior monitoring method based on big data analysis as claimed in claim 1, characterized in that: Calculating physiological characteristic abnormal values ​​according to the preprocessed physiological characteristic data specifically includes the following sub-steps: Calculate the outlier value of each data point based on the preprocessed physiological characteristic data set; Calculate the abnormal confidence of each data point based on its abnormal value; The physiological characteristic outlier value is calculated based on the outlier value of each data point and the outlier confidence of the data point.

5. The patient behavior monitoring method based on big data analysis according to claim 1, characterized in that: Behavioral pattern analysis is performed based on the preprocessed behavioral feature data and physiological feature abnormal values, which specifically includes the following sub-steps: Calculate the correlation between behavioral and physiological characteristics; Calculate the behavioral pattern risk index based on correlation and physiological characteristic abnormalities; Conduct behavioral pattern analysis based on the behavioral pattern risk index.

6. A system for monitoring patient behavior based on big data analysis, characterized in that: include: Data collection module 21: collects physiological characteristic data and behavioral characteristic data of patients; Preprocessing module 22: preprocessing the collected physiological characteristic data and behavioral characteristic data; Physiological characteristic abnormal value calculation module 23: calculates the physiological characteristic abnormal value according to the preprocessed physiological characteristic data; Behavior pattern analysis module 24: performs behavior pattern analysis based on the pre-processed behavior feature data and physiological feature abnormal values.

7. A system for monitoring patient behavior based on big data analysis as claimed in claim 6, characterized in that: Collecting the patient's physiological and behavioral characteristics data is the basis for building an effective patient behavior monitoring method, which specifically includes the following submodules: Collect the patient's physiological characteristic data and establish a physiological characteristic data set; Collect patients' behavioral characteristic data and establish a behavioral characteristic data set.

8. The system for monitoring patient behavior based on big data analysis as claimed in claim 6, characterized in that: Preprocess the collected physiological and behavioral characteristic data, which specifically includes the following submodules: Fill missing values ​​in the collected physiological and behavioral characteristic data; The collected physiological characteristic data and behavioral characteristic data are input with outlier processing.

9. The system for monitoring patient behavior based on big data analysis according to claim 6, characterized in that: Calculate the physiological characteristic abnormal value based on the preprocessed physiological characteristic data, which specifically includes the following submodules: Calculate the outlier value of each data point based on the preprocessed physiological characteristic data set; Calculate the abnormal confidence of each data point based on its abnormal value; The physiological characteristic outlier value is calculated based on the outlier value of each data point and the outlier confidence of the data point.

10. A system for monitoring patient behavior based on big data analysis as claimed in claim 6, characterized in that: Behavioral pattern analysis is performed based on the preprocessed behavioral feature data and physiological feature abnormal values, which specifically includes the following submodules: Calculate the correlation between behavioral and physiological characteristics; Calculate the behavioral pattern risk index based on correlation and physiological characteristic abnormalities; Conduct behavioral pattern analysis based on the behavioral pattern risk index.

Images