CN118039166B - Health monitoring management method based on medical big data - Google Patents
Health monitoring management method based on medical big data Download PDFInfo
- Publication number
- CN118039166B CN118039166B CN202410431135.7A CN202410431135A CN118039166B CN 118039166 B CN118039166 B CN 118039166B CN 202410431135 A CN202410431135 A CN 202410431135A CN 118039166 B CN118039166 B CN 118039166B
- Authority
- CN
- China
- Prior art keywords
- health monitoring
- sequence
- dynamic information
- trend
- monitoring interval
- 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.)
- Active
Links
- 230000036541 health Effects 0.000 title claims abstract description 325
- 238000012544 monitoring process Methods 0.000 title claims abstract description 298
- 238000007726 management method Methods 0.000 title claims abstract description 24
- 238000004458 analytical method Methods 0.000 claims abstract description 52
- 230000035772 mutation Effects 0.000 claims abstract description 46
- 230000008859 change Effects 0.000 claims abstract description 25
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 8
- 230000000737 periodic effect Effects 0.000 claims description 72
- 238000000034 method Methods 0.000 claims description 43
- 230000008569 process Effects 0.000 claims description 12
- 230000036772 blood pressure Effects 0.000 claims description 4
- 239000008280 blood Substances 0.000 claims description 3
- 210000004369 blood Anatomy 0.000 claims description 3
- 238000012417 linear regression Methods 0.000 claims description 3
- WQZGKKKJIJFFOK-GASJEMHNSA-N Glucose Natural products OC[C@H]1OC(O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-GASJEMHNSA-N 0.000 claims description 2
- 239000008103 glucose Substances 0.000 claims description 2
- 230000003862 health status Effects 0.000 claims 1
- 238000001514 detection method Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 7
- 238000005457 optimization Methods 0.000 description 7
- 230000037396 body weight Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 230000002354 daily effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000017531 blood circulation Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000005802 health problem Effects 0.000 description 1
- 230000008376 long-term health Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 230000037323 metabolic rate Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Landscapes
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention relates to the technical field of medical information, in particular to a health monitoring management method based on medical big data, which comprises the following steps: obtaining each health monitoring dynamic information sequence through relevant medical health data of a target individual, and constructing a fluctuation trend influence factor and a fluctuation period influence index according to a time sequence decomposition result; dividing a medical analysis period, comparing the trend and the period intensity characteristic of the health monitoring dynamic information sequence of each interval in the period, and constructing trend distribution consistency and period intensity characteristic factors; further obtaining fluctuation characteristic values and constructing an information characteristic sequence of each health monitoring dynamic information; and constructing characteristic mutation consistency of the health monitoring dynamic information according to the relevance among different health monitoring dynamic information, and completing judgment of the health condition of the target individual. Therefore, health monitoring management based on medical big data is realized, the change of the health condition of a target individual is effectively reflected, and the individual pertinence and the accuracy of the health condition detection result are improved.
Description
Technical Field
The application relates to the technical field of medical information, in particular to a health monitoring management method based on medical big data.
Background
The effective health management can convert passive disease treatment into active self-health monitoring, the Internet of things and artificial intelligence technology are widely fused and applied to life, and the conversion from medical guessing and finding to medical practice can be accelerated through a big data technology; with the continuous growth of private and public medical data, big data technology provides new ways for people to store and manage medical big data, and digs valuable information from massive and complex data. The trend can promote more innovative technologies and products in the medical field, realize data acquisition and monitoring throughout the whole life cycle of a user, and comprehensively and intelligently analyze various data indexes, and serve the health management of the user, so that the health intervention and management capability is improved.
However, although the current health management medical service system can collect and evaluate health data of individuals, the trend of data of physical indexes is different due to different physical conditions and life styles of different individuals, and the occurrence of different health conditions is also different. Thus, using only a single health criterion for health assessment and providing health warnings for all individuals is not sufficiently personalized and accurate.
Disclosure of Invention
In order to solve the technical problems, the invention provides a health monitoring management method based on medical big data, which aims to solve the existing problems.
The health monitoring management method based on the medical big data adopts the following technical scheme:
One embodiment of the invention provides a health monitoring management method based on medical big data, which comprises the following steps:
collecting various health monitoring dynamic information; taking the preset time length as a health monitoring interval;
For various health monitoring dynamic information, taking a sequence formed by all data of the health monitoring dynamic information as a health monitoring dynamic information sequence in a health monitoring interval; decomposing the health monitoring dynamic information sequence through an STL sequence decomposition algorithm to obtain a periodic term sequence, a trend term sequence and a residual term sequence of the health monitoring dynamic information sequence; obtaining fluctuation trend influence factors of the health monitoring interval according to the data change in the trend item sequence and the residual item sequence; obtaining a fluctuation period influence index of the health monitoring interval according to the data change in the period item sequence; taking the time length occupied by the continuously preset number of health monitoring intervals as a medical analysis period; in a medical analysis period, obtaining trend distribution consistency of each health monitoring interval according to the trend item sequence and the fluctuation trend influence factor difference; obtaining the periodic intensity characteristic factors of each health monitoring interval according to the number of the health monitoring intervals and the fluctuation period influence index in the strong and weak periodic sets; obtaining fluctuation characteristic values of each health monitoring interval according to the data change, trend distribution consistency and periodic intensity characteristic factors in the health monitoring dynamic information sequence;
taking a sequence formed by fluctuation characteristic values of all the health monitoring intervals as an information characteristic sequence of the health monitoring dynamic information; obtaining the feature mutation consistency of each health monitoring dynamic information according to the data mutation condition in the information feature sequence of each health monitoring dynamic information; and judging the health state of the target individual according to the feature mutation consistency of the dynamic information of each health monitoring.
Preferably, the health monitoring dynamic information comprises: body mass index, blood pressure, blood glucose, heart rate data.
The optimization method, the fluctuation trend influence factor of the health monitoring interval is obtained according to the data change in the trend item sequence and the residual item sequence, specifically comprises the following steps:
Calculating variances of all data in the residual error item sequence, and marking the variances as first variances; adding elements at the same time in the residual error item sequence and the trend item sequence, obtaining a new sequence composed of elements at all times after addition, calculating variances of all elements in the new sequence, and marking the variances as second variances; calculating the ratio of the first variance to the second variance; taking the absolute value of the difference between the natural number 1 and the ratio as the trend intensity of the health monitoring interval;
Calculating the difference value between the maximum value and the minimum value of all data in the trend item sequence; and taking the product of the difference value between the maximum value and the minimum value and the trend intensity as a fluctuation trend influence factor of the health monitoring interval.
The optimization method is characterized by obtaining a fluctuation period influence index of a health monitoring interval according to the data change in the period item sequence, and specifically comprising the following steps:
Obtaining the periodic intensity of the ith health monitoring interval by acquiring the trend intensity of the health monitoring interval ; The expression for calculating the fluctuation period influence index of the health monitoring interval is as follows:
；
In the method, in the process of the invention, Indicating the fluctuation period influence index of the i-th health monitoring interval,Representing the sequence corresponding to the x-th period of the periodic item sequence obtained by decomposing the health monitoring dynamic information sequence of the i-th health monitoring interval,A periodic item sequence obtained by decomposing the health monitoring dynamic information sequence of the ith health monitoring interval is represented,Indicating an i-th health monitoring interval,、Respectively represent a strong periodic set and a weak periodic set,、Respectively representing a maximum function and a minimum function.
The preference method is that the strong periodic set and the weak periodic set are specifically as follows:
Taking a set formed by health monitoring intervals with the periodic intensity being greater than or equal to a preset periodic intensity threshold value as a strong periodic set; and taking a set formed by the health monitoring intervals with the cycle intensity smaller than the preset cycle intensity threshold as a weak periodic set.
The optimization method, according to the trend item sequence and the fluctuation trend influence factor, obtains the trend distribution consistency of each health monitoring interval, specifically includes:
Calculating a pearson correlation coefficient between an ith health monitoring interval and a trend item sequence of the health monitoring dynamic information sequence of each health monitoring interval; calculating standard deviation of fluctuation trend influence factors of all health monitoring intervals; calculating the absolute value of the difference between the ith health monitoring interval and the fluctuation trend influence factors of each health monitoring interval; calculating the product of the standard deviation and the absolute value of the difference; calculating the ratio of the pearson correlation coefficient to the product; and taking the sum value of all the ratios as the trend distribution consistency of the ith health monitoring interval.
And the optimization method is characterized in that the periodic intensity characteristic factors of each health monitoring interval are obtained according to the number of the health monitoring intervals and the fluctuation period influence index in the strong and weak periodic sets, and the specific expression is as follows:
；
In the method, in the process of the invention, A cycle intensity characteristic factor representing an ith health monitoring interval in a t-th medical analysis cycle, i representing a serial number value of the health monitoring interval,An exponential function based on natural constants is represented,Representing the number of elements in the acquisition set,、Respectively represent a strong periodic set and a weak periodic set in the t-th medical analysis period,、Respectively representing the number of health monitoring intervals in the strong and weak periodic sets in the t-th medical analysis period,Indicating the number of health monitoring intervals in the t-th medical analysis period,And the fluctuation period influence index of the ith health monitoring interval in the t-th medical analysis period is represented.
The optimization method, the fluctuation characteristic value of each health monitoring interval is obtained according to the data change, trend distribution consistency and periodic intensity characteristic factors in the health monitoring dynamic information sequence, specifically comprises the following steps:
Acquiring the mean value and standard deviation of all elements in the health monitoring dynamic information sequence of each health monitoring interval; calculating the ratio of the mean value to the standard deviation; calculating the product of trend distribution consistency and periodic intensity characteristic factors of each health monitoring interval; and taking the product of the ratio and the product as a fluctuation characteristic value of each health monitoring interval.
The optimization method, the feature mutation consistency of each health monitoring dynamic information is obtained according to the data mutation condition in the information feature sequence of each health monitoring dynamic information, specifically comprises the following steps:
inputting information feature sequences of various health monitoring dynamic information into a piecewise linear regression algorithm to obtain mutation points of the information feature sequences; taking a sequence formed by all mutation points of each information characteristic sequence as a mutation index sequence; calculating the DTW distance between the mutation index sequences of the type a and other health monitoring dynamic information according to a dynamic time warping algorithm; and taking the normalized value of the average value of all the DTW distances as the characteristic mutation consistency of the type a health monitoring dynamic information.
The optimization method is characterized in that the health state of the target individual is judged according to the feature mutation consistency of the dynamic information of each health monitoring, and specifically comprises the following steps:
Taking all the health monitoring dynamic information with the feature mutation consistency larger than a preset mutation threshold value as potential risk factors; if the number of the potential risk factors is greater than the preset number, the target individual is in an unhealthy state; otherwise, the target individual is in a healthy state.
The invention has at least the following beneficial effects:
According to the method, the change trend and periodicity of the health monitoring dynamic information data of the target individual in the health monitoring interval are analyzed, the fluctuation trend influence factor and the fluctuation period influence index are constructed, and the trend and periodicity of the physical condition data change of the target individual in daily life are measured; dividing a medical analysis period, comparing trend characteristics of each health monitoring interval in the medical analysis period, constructing trend distribution consistency, measuring the consistency of data change of a target individual in a period of time, and reflecting the stability of the physical state of the target individual; constructing periodic intensity characteristic factors according to periodic characteristics of all health monitoring intervals in a medical analysis period, and correcting the periodicity of each health monitoring interval; constructing a fluctuation characteristic value, and effectively reflecting the condition of data fluctuation of a target individual; the feature mutation consistency is constructed, mutation features and correlations of different health monitoring dynamic information in a medical analysis period are described, the change of the health condition of a target individual is effectively reflected, and the individual pertinence and the accuracy of the health condition detection result are improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a health monitoring management method based on medical big data provided by the invention;
fig. 2 is a schematic diagram of steps of a health monitoring management method based on medical big data.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description refers to the specific implementation, structure, characteristics and effects of a health monitoring management method based on medical big data according to the invention by combining the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the health monitoring management method based on medical big data provided by the invention with reference to the accompanying drawings.
The embodiment of the invention provides a health monitoring management method based on medical big data.
Specifically, the following health monitoring management method based on medical big data is provided, please refer to fig. 1, the method includes the following steps:
step S001, acquiring relevant medical health data of a target individual, and preprocessing.
In this embodiment, the physiological signs of the target individual are mainly monitored, the health condition of the target individual is evaluated, potential health risk factors are identified, and possible health problems are predicted.
First, the relevant medical health data is collected. Health monitoring dynamic information of the target individual is obtained from the health medical management system, and the health monitoring dynamic information comprises, but is not limited to, body mass index, blood pressure, blood sugar and heart rate data. The health monitoring dynamic information is obtained through the movable medical wearable equipment, the sampling frequency of the health monitoring dynamic information is 10min once, and the health monitoring interval is 1 day. And the acquired data is subjected to data cleaning by adopting a regular expression technology, nonsensical words are removed, and the specific process is not repeated here because the regular expression technology is a well-known data cleaning technology. Due to interference of factors such as signals, the acquired data may have missing conditions, and the missing data is filled by adopting a mean interpolation method in the embodiment.
Thus, medical health data of the target individual is obtained.
Step S002, obtaining fluctuation trend influence factors and fluctuation period influence indexes according to the fluctuation change of each sequence in the time sequence decomposition results of each type of health monitoring dynamic information sequences; dividing a medical analysis period, and constructing a fluctuation characteristic value of each health monitoring interval by combining trend change characteristics and period intensity characteristics of the health monitoring dynamic information data of each health monitoring interval in the period.
Since the physiological characteristics of the target individual may fluctuate smoothly over a small range of health monitoring intervals, for example, the body mass index of a person is relatively low in the morning, slightly higher in noon, and higher in the evening. This is because people eat less at night, are metabolised slowly, resulting in a relatively high body weight, while eating and exercising more during the day, the metabolic rate is increased and the body weight is relatively light. In general, each physiological characteristic of a person every day has a fluctuation rule, so that fluctuation of each health monitoring dynamic information is analyzed.
Specifically, in this embodiment, a single health monitoring dynamic information is taken as an example to analyze, all data in each health monitoring interval of a health monitoring dynamic information of a target individual is taken as each health monitoring dynamic information sequence, firstly, trends of the health monitoring dynamic information sequences in each health monitoring interval are analyzed, the health monitoring dynamic information sequences in each health monitoring interval are taken as input of an STL sequence decomposition algorithm, and a periodic item sequence, a trend item sequence and a residual item sequence of the health monitoring dynamic information sequences in each health monitoring interval are obtained; according to the result of the sequence decomposition, the element distribution of the trend item sequence of the health monitoring dynamic information sequence decomposition is analyzed, as the trend item can reflect the overall increasing or decreasing trend of the sequence with the passage of time. Calculating fluctuation trend influence factors of each health monitoring interval, wherein the expression is as follows:
；
In the method, in the process of the invention, Representing the fluctuating trend impact factor of the ith health monitoring interval,Indicating the trend intensity of the ith health monitoring interval,、Respectively representing a residual error item sequence and a trend item sequence which are obtained by decomposing the health monitoring dynamic information sequence of the ith health monitoring interval,The variance function is represented as a function of the variance,、Respectively representing a maximum function and a minimum function. Wherein the method comprises the steps ofAs a first variance of the first set,And is the second variance. Wherein the method comprises the steps ofThe method is characterized in that elements at the same time in the residual term sequence and the trend term sequence are added, and then a new sequence consisting of elements and values at each time is taken as a result of adding the two sequences.
Acquiring trend intensity of the health monitoring dynamic information sequence of each health monitoring interval by means of the residual error item sequence and the trend item sequence acquired by sequence decomposition, wherein when the trend of the health monitoring dynamic information sequence is stronger, the variance of the sequence formed by each element and value of the residual error item sequence and the trend item sequence is larger, namelyThe larger and thenThe smaller the trend intensity of the health monitoring interval is, the closer to 1; when the extreme difference value of the trend item sequence is larger, the total trend of the health monitoring dynamic information sequence in the health monitoring interval is larger, and further the fluctuation trend influence factor of the health monitoring interval is larger.
Further, analyzing the periodicity of the health monitoring dynamic information sequence in each health monitoring interval, and acquiring the periodicity intensity of each health monitoring interval by means of the residual sequence by adopting a calculation method identical to the trend intensity of each health monitoring intervalSince the health monitoring dynamic information of the target individual has not only a trend change but also a periodic change in the health monitoring interval. Thus, there is a need to analyze the periodicity of the individual health monitoring dynamic information of the target individual. Setting a periodic intensity threshold, and dividing the health monitoring interval into a strong periodic set when the periodic intensity is greater than or equal to the periodic intensity thresholdIn (a) and (b); conversely, the health monitoring interval is divided into a weak periodic setIs a kind of medium. Calculating the fluctuation period influence index of each health monitoring interval, wherein the expression is as follows:
；
In the method, in the process of the invention, Indicating the fluctuation period influence index of the i-th health monitoring interval,The number of cycles contained in the sequence of cycle items representing the health monitoring dynamic information sequence of the i-th health monitoring interval,A sequence corresponding to the x-th period of the periodic item sequence representing the health monitoring dynamic information sequence of the i-th health monitoring interval,A periodic item sequence obtained by decomposing the health monitoring dynamic information sequence of the ith health monitoring interval is represented,Representing the cycle intensity of the ith health monitoring interval,Indicating an i-th health monitoring interval,、Respectively represent a strong periodic set and a weak periodic set,、Respectively representing a maximum function and a minimum function.
Considering the periodicity of the health monitoring dynamic information sequences of each health monitoring interval of a target individual, setting a periodic intensity threshold value, carrying out different processing on data with different periodicity, and when the health monitoring interval belongs to a strong periodic set, indicating that the periodicity of the health monitoring dynamic information sequences of the health monitoring interval is strong, so that the average value of the extreme values of each period of the periodic item sequence is used as the integral distribution of the periodic item sequence to obtain the fluctuation period influence index of the health monitoring interval; when the health monitoring interval belongs to a weak periodic set, the periodicity of the health monitoring dynamic information sequence of the health monitoring interval is weak, and at the moment, the probability that the health monitoring dynamic information sequence of the health monitoring interval has periodicity is small, the period intensity is closer to 0, so that the distribution of the period item sequence is used as the fluctuation influence index of the health monitoring interval. When the period intensity is greater and the degree of difference between elements of the periodic item sequence is greater, the fluctuation period influence index of the health monitoring interval is greater.
Because health management is a long-term trend, and needs to be analyzed by combining the data changes of each long-term health monitoring dynamic information of the target individual, the S health monitoring intervals before the current health monitoring interval include s+1 health monitoring intervals in total, and in this embodiment, the S value is 29, which is used as a medical analysis period. Analyzing fluctuation trend influence factors of all health monitoring intervals in a medical analysis period, and calculating trend distribution consistency of the health monitoring dynamic information sequences of all the health monitoring intervals and the health monitoring dynamic information sequences of other health monitoring intervals in the medical analysis period, wherein the expression is as follows:
；
In the method, in the process of the invention, Indicating the consistency of trend distribution in the ith health monitoring interval in the t-th medical analysis period,Indicating the number of health monitoring intervals in the t-th medical analysis period,、Respectively representing trend item sequences obtained by decomposing health monitoring dynamic information sequences of the ith and jth health monitoring intervals in the t-th medical analysis period,Representing the pearson correlation coefficient,Representing the standard deviation of the fluctuation trend impact factor for all health monitoring intervals during the t-th medical analysis period,、And respectively representing fluctuation trend influence factors of the ith and jth health monitoring intervals in the tth medical analysis period.
When the target individual is in a healthy state, the data fluctuation trend of the health monitoring dynamic information of all the health monitoring intervals is relatively consistent, so the fluctuation trend influence factors of the health monitoring intervals are extremely small in degree, namely、The smaller the value of (c) and thus the greater the trend distribution consistency; on the other hand, the trend correlation characteristics of the health monitoring interval are characterized according to the pearson correlation coefficients among the trend item sequences of the health monitoring dynamic information sequences at different sampling moments, namely, the greater the pearson correlation coefficient is, the greater the trend distribution consistency is.
Then analyzing the fluctuation period influence indexes among the health monitoring intervals in the medical analysis period to obtain the period intensity characteristic factors of each health monitoring interval in the medical analysis period, wherein the expression is as follows:
；
In the method, in the process of the invention, A cycle intensity characteristic factor representing an ith health monitoring interval in a t-th medical analysis cycle, i representing a serial number value of the health monitoring interval,An exponential function based on natural constants is represented,Representing the number of elements in the acquisition set,、Respectively represent a strong periodic set and a weak periodic set in the t-th medical analysis period,、Respectively representing the number of health monitoring intervals in the strong and weak periodic sets in the t-th medical analysis period,Indicating the number of health monitoring intervals in the t-th medical analysis period,And the fluctuation period influence index of the ith health monitoring interval in the t-th medical analysis period is represented.
Counting the probability of health monitoring intervals belonging to a strong periodic set and a weak periodic set in a medical analysis period, and if the number of the health monitoring intervals belonging to the strong periodicity in the medical analysis period is large, indicating that the overall periodicity in the medical analysis period is stronger, and further the cycle strength characteristic factor is stronger; conversely, the weaker.
In each health monitoring interval of the medical analysis period, according to element distribution characteristics of a health monitoring dynamic information sequence, combining trend distribution consistency of the health monitoring dynamic information and a period intensity characteristic factor to obtain a fluctuation characteristic value of the health monitoring dynamic information, wherein the expression is as follows:
；
In the method, in the process of the invention, Representing the fluctuation characteristic value of the ith health monitoring interval in the t-th medical analysis period,Representing the average value of all elements in the health monitoring dynamic information sequence of the ith health monitoring interval in the t-th medical analysis period,Representing standard deviations of all elements in the health monitoring dynamic information sequence of the ith health monitoring interval in the t-th medical analysis period,Representing the cycle intensity characteristic factor of the ith health monitoring interval in the t-th medical analysis cycle,And the trend distribution consistency of the ith health monitoring interval in the t-th medical analysis period is shown.
The ratio of the mean value to the standard deviation of the health monitoring dynamic information sequence in the health monitoring interval measures the discrete degree of the health monitoring dynamic information sequence in the health monitoring interval; the trend and the periodic characteristic of the health monitoring dynamic information sequence in the health monitoring interval are measured by multiplication with the trend distribution consistency and the periodic intensity characteristic factor, namely, the greater the discrete degree, the fluctuation trend influence factor and the fluctuation period influence index are, the more obvious the fluctuation characteristic of the health monitoring dynamic information in the health monitoring interval is, and the greater the fluctuation characteristic value is.
Step S003, an information feature sequence of each health monitoring dynamic information is constructed based on the fluctuation feature value; and constructing characteristic mutation consistency of the health monitoring dynamic information according to the relevance among different health monitoring dynamic information, and completing judgment of the health condition of the target individual.
The health monitoring dynamic information of the target individual has certain synergies, when the health condition of the target individual has a problem, the fluctuation condition of the corresponding health monitoring dynamic information can be intuitively represented, and the fluctuation condition has certain relativity. For example, when the blood pressure of a target individual increases or decreases, the heart rate may be adjusted accordingly to maintain balance of blood circulation within the body. Thus, the relevance of the fluctuation trend characteristics of each health monitoring dynamic information in the medical analysis period is analyzed.
Specifically, for each health monitoring dynamic information, a sequence formed by fluctuation characteristic values of all health monitoring intervals in a medical analysis period according to a time sequence is used as an information characteristic sequence of each health monitoring dynamic information. The information characteristic sequence is used as input of a piecewise linear regression algorithm, the output is a mutation point of the information characteristic sequence, the position serial numbers of the mutation point of the information characteristic sequence form a mutation index sequence, and the characteristic mutation consistency of each health monitoring dynamic information is calculated according to the mutation index sequence, wherein the expression is as follows:
；
In the method, in the process of the invention, Characteristic mutation consistency of the type a health monitoring dynamic information is represented,Representing the number of categories of health monitoring dynamic information,、Mutation index sequences respectively representing the a-th and b-th health monitoring dynamic information,Representing the calculation of the DTW distance between sequences by a dynamic time warping algorithm.
Setting a mutation threshold, and marking the health monitoring dynamic information as potential risk factors if the feature mutation consistency of the health monitoring dynamic information in the medical analysis period is greater than the mutation threshold; if it exceedsIf the feature mutation consistency of the potential risk factors in the individual medical analysis period is larger than a mutation threshold, judging that the target individual is unhealthy, and performing physical examination on the potential risk factors; otherwise, the target individual is judged to be in a healthy state. Mutation threshold value andCan be set by the practitioner, the embodiment sums the mutation threshold valuesThe values of (2) are set to 0.7 and 3, respectively. The steps of the method are schematically shown in fig. 2.
In summary, according to the embodiment of the invention, the change trend and periodicity of the health monitoring dynamic information data of the target individual in the health monitoring interval are analyzed, the fluctuation trend influence factor and the fluctuation period influence index are constructed, and the trend and periodicity of the physical condition data change of the target individual in daily life are measured; dividing a medical analysis period, comparing trend characteristics of each health monitoring interval in the medical analysis period, constructing trend distribution consistency, measuring the consistency of data change of a target individual in a period of time, and reflecting the stability of the physical state of the target individual; constructing periodic intensity characteristic factors according to periodic characteristics of all health monitoring intervals in a medical analysis period, and correcting the periodicity of each health monitoring interval; constructing a fluctuation characteristic value, and effectively reflecting the condition of data fluctuation of a target individual; the method comprises the steps of constructing feature mutation consistency, describing mutation features and relevance of different health monitoring dynamic information in a medical analysis period, effectively reflecting the change of the health condition of a target individual, paying attention to the physical health index with problems, and reminding the target individual to perform health examination on abnormality.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and the same or similar parts of each embodiment are referred to each other, and each embodiment mainly describes differences from other embodiments.
The above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; the technical solutions described in the foregoing embodiments are modified or some of the technical features are replaced equivalently, so that the essence of the corresponding technical solutions does not deviate from the scope of the technical solutions of the embodiments of the present application, and all the technical solutions are included in the protection scope of the present application.
Claims (4)
1. A health monitoring management method based on medical big data is characterized by comprising the following steps:
collecting various health monitoring dynamic information; taking the preset time length as a health monitoring interval;
For various health monitoring dynamic information, taking a sequence formed by all data of the health monitoring dynamic information as a health monitoring dynamic information sequence in a health monitoring interval; decomposing the health monitoring dynamic information sequence through an STL sequence decomposition algorithm to obtain a periodic term sequence, a trend term sequence and a residual term sequence of the health monitoring dynamic information sequence; obtaining fluctuation trend influence factors of the health monitoring interval according to the data change in the trend item sequence and the residual item sequence; obtaining a fluctuation period influence index of the health monitoring interval according to the data change in the period item sequence; taking the time length occupied by the continuously preset number of health monitoring intervals as a medical analysis period; in a medical analysis period, obtaining trend distribution consistency of each health monitoring interval according to the trend item sequence and the fluctuation trend influence factor difference; obtaining the periodic intensity characteristic factors of each health monitoring interval according to the number of the health monitoring intervals and the fluctuation period influence index in the strong and weak periodic sets; obtaining fluctuation characteristic values of each health monitoring interval according to the data change, trend distribution consistency and periodic intensity characteristic factors in the health monitoring dynamic information sequence;
Taking a sequence formed by fluctuation characteristic values of all the health monitoring intervals as an information characteristic sequence of the health monitoring dynamic information; obtaining the feature mutation consistency of each health monitoring dynamic information according to the data mutation condition in the information feature sequence of each health monitoring dynamic information; judging the health state of the target individual according to the feature mutation consistency of each health monitoring dynamic information;
the fluctuation trend influence factor of the health monitoring interval is obtained according to the data change in the trend item sequence and the residual item sequence, and the method specifically comprises the following steps:
Calculating variances of all data in the residual error item sequence, and marking the variances as first variances; adding elements at the same time in the residual error item sequence and the trend item sequence, obtaining a new sequence composed of elements at all times after addition, calculating variances of all elements in the new sequence, and marking the variances as second variances; calculating the ratio of the first variance to the second variance; taking the absolute value of the difference between the natural number 1 and the ratio as the trend intensity of the health monitoring interval;
Calculating the difference value between the maximum value and the minimum value of all data in the trend item sequence; taking the product of the difference value between the maximum value and the minimum value and the trend intensity as a fluctuation trend influence factor of the health monitoring interval;
the method for obtaining the fluctuation period influence index of the health monitoring interval according to the data change in the period item sequence specifically comprises the following steps:
Obtaining the periodic intensity of the ith health monitoring interval by acquiring the trend intensity of the health monitoring interval ; The expression for calculating the fluctuation period influence index of the health monitoring interval is as follows:
；
In the method, in the process of the invention, Indicating the fluctuation period influence index of the i-th health monitoring interval,The number of cycles contained in the sequence of cycle items representing the health monitoring dynamic information sequence of the i-th health monitoring interval,Representing the sequence corresponding to the x-th period of the periodic item sequence obtained by decomposing the health monitoring dynamic information sequence of the i-th health monitoring interval,A periodic item sequence obtained by decomposing the health monitoring dynamic information sequence of the ith health monitoring interval is represented,Indicating an i-th health monitoring interval,、Respectively represent a strong periodic set and a weak periodic set,、Respectively representing a maximum value function and a minimum value function;
The trend distribution consistency of each health monitoring interval is obtained according to the trend item sequence and the fluctuation trend influence factor, and the method specifically comprises the following steps:
Calculating a pearson correlation coefficient between an ith health monitoring interval and a trend item sequence of the health monitoring dynamic information sequence of each health monitoring interval; calculating standard deviation of fluctuation trend influence factors of all health monitoring intervals; calculating the absolute value of the difference between the ith health monitoring interval and the fluctuation trend influence factors of each health monitoring interval, and recording the absolute value of the difference as a first absolute value of the difference; calculating the product of the standard deviation and the absolute value of the first difference value, and recording the product as a first product; calculating the ratio of the pearson correlation coefficient to the first product, and recording the ratio as a first ratio; taking the sum of all the first ratios as the trend distribution consistency of the ith health monitoring interval;
the periodic intensity characteristic factors of each health monitoring interval are obtained according to the number of the health monitoring intervals and the fluctuation period influence index in the strong and weak periodic sets, and the specific expression is as follows:
；
In the method, in the process of the invention, A cycle intensity characteristic factor representing an ith health monitoring interval in a t-th medical analysis cycle, i representing a serial number value of the health monitoring interval,An exponential function based on natural constants is represented,Representing the number of elements in the acquisition set,、Respectively represent a strong periodic set and a weak periodic set in the t-th medical analysis period,、Respectively representing the number of health monitoring intervals in the strong and weak periodic sets in the t-th medical analysis period,Indicating the number of health monitoring intervals in the t-th medical analysis period,A fluctuation period influence index representing an ith health monitoring interval in a t-th medical analysis period;
The method for obtaining the fluctuation characteristic value of each health monitoring interval according to the data change, trend distribution consistency and periodic intensity characteristic factors in the health monitoring dynamic information sequence specifically comprises the following steps:
Acquiring the mean value and standard deviation of all elements in the health monitoring dynamic information sequence of each health monitoring interval, and respectively marking the mean value and the standard deviation as a first mean value and a first standard deviation; calculating the ratio of the first mean value to the first standard deviation, and recording the ratio as a second ratio; calculating the product of trend distribution consistency and periodic intensity characteristic factors of each health monitoring interval, and recording the product as a second product; taking the product of the second ratio and the second product as a fluctuation characteristic value of each health monitoring interval;
The method for obtaining the characteristic mutation consistency of the health monitoring dynamic information according to the data mutation situation in the information characteristic sequences of the health monitoring dynamic information comprises the following steps:
inputting information feature sequences of various health monitoring dynamic information into a piecewise linear regression algorithm to obtain mutation points of the information feature sequences; taking a sequence formed by all mutation points of each information characteristic sequence as a mutation index sequence; calculating the DTW distance between the mutation index sequences of the type a and other health monitoring dynamic information according to a dynamic time warping algorithm; and taking the normalized value of the average value of all the DTW distances as the characteristic mutation consistency of the type a health monitoring dynamic information.
2. The health monitoring management method based on medical big data according to claim 1, wherein the health monitoring dynamic information includes: body mass index, blood pressure, blood glucose, heart rate data.
3. The health monitoring management method based on medical big data according to claim 1, wherein the strong periodic set and the weak periodic set are specifically:
Taking a set formed by health monitoring intervals with the periodic intensity being greater than or equal to a preset periodic intensity threshold value as a strong periodic set; and taking a set formed by the health monitoring intervals with the cycle intensity smaller than the preset cycle intensity threshold as a weak periodic set.
4. The health monitoring management method based on medical big data according to claim 1, wherein the judging of the health status of the target individual according to the feature mutation consistency of each health monitoring dynamic information is specifically:
Taking all the health monitoring dynamic information with the feature mutation consistency larger than a preset mutation threshold value as potential risk factors; if the number of the potential risk factors is greater than the preset number, the target individual is in an unhealthy state; otherwise, the target individual is in a healthy state.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410431135.7A CN118039166B (en) | 2024-04-11 | 2024-04-11 | Health monitoring management method based on medical big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410431135.7A CN118039166B (en) | 2024-04-11 | 2024-04-11 | Health monitoring management method based on medical big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118039166A CN118039166A (en) | 2024-05-14 |
| CN118039166B true CN118039166B (en) | 2024-07-02 |
Family
ID=90984370
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410431135.7A Active CN118039166B (en) | 2024-04-11 | 2024-04-11 | Health monitoring management method based on medical big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118039166B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2957705A1 (en) * | 2011-07-05 | 2013-01-10 | Saudi Arabian Oil Company | Floor mat system and associated, computer medium and computer-implemented methods for monitoring and improving health and productivity of employees |
| CN116250834A (en) * | 2022-09-08 | 2023-06-13 | 湖北智奥物联网科技有限公司 | Noninvasive personalized dynamic blood glucose trend monitoring and early warning method and device |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021044594A1 (en) * | 2019-09-05 | 2021-03-11 | Hitachi, Ltd. | Method, system, and apparatus for health status prediction |
| CN116936104B (en) * | 2023-09-15 | 2023-12-08 | 广东恒腾科技有限公司 | Health detector data analysis system and method based on artificial intelligence |
| CN117558449A (en) * | 2023-11-24 | 2024-02-13 | 启康保(北京)健康科技有限公司 | Medical health data management system based on big data |
| CN117457147B (en) * | 2023-12-25 | 2024-03-19 | 深圳市尼罗河移动互联科技有限公司 | Personalized nursing planning method and system for rehabilitation patient |
-
2024
- 2024-04-11 CN CN202410431135.7A patent/CN118039166B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2957705A1 (en) * | 2011-07-05 | 2013-01-10 | Saudi Arabian Oil Company | Floor mat system and associated, computer medium and computer-implemented methods for monitoring and improving health and productivity of employees |
| CN116250834A (en) * | 2022-09-08 | 2023-06-13 | 湖北智奥物联网科技有限公司 | Noninvasive personalized dynamic blood glucose trend monitoring and early warning method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118039166A (en) | 2024-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rong et al. | A multi-type features fusion neural network for blood pressure prediction based on photoplethysmography | |
| Baker et al. | A hybrid neural network for continuous and non-invasive estimation of blood pressure from raw electrocardiogram and photoplethysmogram waveforms | |
| Sideris et al. | Building continuous arterial blood pressure prediction models using recurrent networks | |
| Zaitcev et al. | A deep neural network application for improved prediction of $\text {HbA} _ {\text {1c}} $ in type 1 diabetes | |
| CN107203700B (en) | Method and device based on continuous blood glucose monitoring | |
| CN112043252B (en) | Emotion recognition system and method based on respiratory component in pulse signal | |
| Di Martino et al. | High-resolution physiological stress prediction models based on ensemble learning and recurrent neural networks | |
| CN111685748A (en) | Quantile-based blood pressure early warning method, quantile-based blood pressure early warning device, quantile-based blood pressure early warning equipment and storage medium | |
| van der Westhuizen et al. | Bayesian LSTMs in medicine | |
| CN114512239A (en) | Cerebral apoplexy risk prediction method and system based on transfer learning | |
| Chen et al. | Edge2Analysis: a novel AIoT platform for atrial fibrillation recognition and detection | |
| WO2022237162A1 (en) | Blood glucose prediction method and application thereof, and blood glucose prediction system | |
| Roy et al. | BePCon: a photoplethysmography-based quality-aware continuous beat-to-beat blood pressure measurement technique using deep learning | |
| CN114758786A (en) | Dynamic early warning system for post-traumatic hemorrhagic shock based on noninvasive parameters | |
| CN118039166B (en) | Health monitoring management method based on medical big data | |
| Abrar et al. | A multi-agent approach for personalized hypertension risk prediction | |
| CN112768074A (en) | Artificial intelligence-based serious disease risk prediction method and system | |
| CN117095828A (en) | Blood glucose value prediction and alarm method based on follow-up record of diabetics | |
| CN116999063A (en) | Method for realizing electrocardiographic atrial fibrillation detection based on signal decomposition and convolution network | |
| CN115376638A (en) | Physiological characteristic data analysis method based on multi-source health perception data fusion | |
| CN110786847A (en) | Electrocardiogram signal library building method and analysis method | |
| Gu et al. | Detecting epileptic seizures via non-uniform multivariate embedding of EEG signals | |
| CN114504298A (en) | Physiological feature distinguishing method and system based on multi-source health perception data fusion | |
| Lakshminarayan et al. | Feature extraction and automated classification of heartbeats by machine learning | |
| Prabhu et al. | A multi-layered LSTM for predicting physician stress during an ED shift |
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 | ||
| GR01 | Patent grant |