Intelligent respiratory disease data analysis method and system

Abstract

The invention relates to the technical field of respiratory tract data analysis, in particular to a respiratory tract disease data intelligent analysis method and system. Based on multi-dimensional respiratory tract data of a patient at multiple time points, a reference dimension is selected, data distribution of the reference dimension at each moment is analyzed to calculate an abnormal degree, and abnormal data of each dimension is screened according to the abnormal degree. Dividing the data into two types according to the physiological state; calculating the compensation expression degree based on the overlap ratio and distribution of the first type of data at the abnormal moment; and evaluating the symptom severity by combining the abnormal coincidence degree of the second type of data, the distribution of the first type of data and the abnormal extreme value mean value. And fusing the physiological data fluctuation degree, the compensatory expression and the symptom severity to obtain a physical state score. Further integrating the multi-cycle PEF variation rate distribution and the body score distribution to evaluate the illness state stability degree, and finally realizing the intelligent comprehensive analysis of the respiratory diseases. According to the invention, the body of the patient can be accurately analyzed, so that related personnel can accurately identify the patient and intervene in advance.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of respiratory tract data analysis, and particularly relates to a respiratory tract disease data intelligent analysis method and system. BACKGROUND

[0002] Respiratory tract diseases refer to a class of diseases affecting the human respiratory system, mainly divided into upper respiratory tract infection and lower respiratory tract infection, which are mainly caused by infection, allergy, environmental pollution and other factors. Asthma, as a common chronic respiratory disease, has a greater impact on the human body, and at present it cannot be completely cured, and long-term monitoring and management are needed to effectively control the disease. Asthma is also one of the most common chronic respiratory diseases in childhood, and its pathophysiological mechanism is consistent with that of adults, but because the immune system of children is in a dynamic development stage, it presents significant particularity, such as high infection-induced proportion, usually more severe onset, and easy to appear persistent state, etc. How to efficiently monitor and manage children with asthma is a problem that needs to be solved in medicine at present.

[0003] At present, the monitoring and management of children with asthma basically depends on family treatment, for example, parents continuously monitor the relevant physiological data of children through peak flow meter, intelligent detector and other equipment, and judge the recovery state according to the data, and then take corresponding measures. However, in actual scenarios, most parents lack relevant knowledge, have difficulty in data interpretation, and have weak self-management ability, so it is difficult for them to adjust medication or seek medical treatment in time according to the data, and it is difficult for them to adhere to it for a long time. In addition, the existing method mainly focuses on the efficiency and real-time performance of analysis, and has great algorithm bias risk and fairness defects, poor flexibility, so that the development direction of the disease is difficult to predict and control, and early identification and early intervention cannot be carried out. SUMMARY

[0004] In order to solve the technical problems that the prior art mainly focuses on the efficiency and real-time of the analysis when monitoring and managing the respiratory tract data of children asthma patients, there is a large risk of algorithm bias and fairness defects, poor flexibility, it is difficult to predict and control the development direction of the disease, and early identification and early intervention cannot be carried out, the purpose of the present application is to provide a respiratory disease data intelligent analysis method and system, the technical scheme adopted is as follows: a respiratory disease data intelligent analysis method, the method comprises: obtaining respiratory tract data of each dimension at different preset sampling moments of a patient; optionally, the respiratory tract data of one dimension is taken as the respiratory tract data of the reference dimension; according to the reference dimension respiratory tract data distribution characteristics obtained at each preset sampling moment in each preset time window, the abnormal degree of the reference dimension respiratory tract data at each preset sampling moment is obtained; according to the abnormal degree of the respiratory tract data of each dimension, the abnormal data of each dimension is screened out; according to the physiological state of the patient, the respiratory tract data of each dimension is divided into first class data and second class data; according to the abnormal degree distribution characteristics of the respiratory tract data of each dimension and the abnormal data distribution, the physiological data fluctuation degree of the patient is obtained; according to the coincidence degree of the abnormal data of different dimensions corresponding to the preset sampling moment in the first class data of different dimensions, and the abnormal data distribution of each dimension in the first class data, the compensatory performance degree of the patient is obtained; according to the coincidence degree of the abnormal data of different dimensions corresponding to the preset sampling moment in each dimension of the second class data, the abnormal data distribution of each dimension in the first class data and the mean value of the respiratory tract data with the highest abnormal degree in each dimension, the symptom severity of the patient is obtained; according to the physiological data fluctuation degree, the compensatory performance degree and the symptom severity, the body state score of the patient is obtained; the PEF variation rate of the patient in each preset period is obtained; according to the PEF variation rate distribution and the body state score distribution of the patient in the preset number of preset periods, the disease stability degree of the patient is obtained; the respiratory disease data of the patient is intelligently analyzed according to the body state score and the disease stability degree of the patient.

[0005] Further, the abnormal degree acquisition method comprises: acquiring the abnormal degree according to an abnormal degree calculation formula, the abnormal degree calculation formula is as follows: In the formula, indicates the abnormal degree of the respiratory tract data of the reference dimension at the i-th preset sampling moment; indicates the number of preset time windows; indicates the respiratory tract data of the reference dimension at the i-th preset sampling moment; indicates the respiratory tract data of the reference dimension at the i-th preset sampling moment; indicates the mean value of the respiratory tract data of the reference dimension in the i-th preset time window; indicates the mean value of the respiratory tract data of the reference dimension in the i-th preset time window; indicates the mean value of the respiratory tract data of the reference dimension in the i-th preset time window; indicates the mean value of the respiratory tract data of the reference dimension in the i-th preset time window; The standard deviation of respiratory data in the reference dimension within a preset time window.

[0006] Furthermore, the method for obtaining abnormal data includes: calculating the standard deviation of the abnormality degree of respiratory data in each dimension as the first standard deviation; and taking the respiratory data in each dimension whose absolute value of the abnormality degree is greater than 3 times the first standard deviation as abnormal data.

[0007] Furthermore, the method for obtaining the physiological data fluctuation degree includes: obtaining the physiological data fluctuation degree according to a physiological data fluctuation degree calculation formula, and the physiological data fluctuation degree calculation formula is as follows: Where, Indicates the degree of fluctuation of the patient's physiological data; The number of dimensions representing the respiratory data; Indicates the The mean of abnormal data in each dimension; Indicates the The number of abnormal data in each dimension; No. The number of respiratory data in each dimension; Represents the normalization function.

[0008] Furthermore, the method for obtaining the degree of compensation performance includes: in the respiratory data of each dimension of the first category of data, when two consecutive abnormal data appear, the preset sampling time corresponding to the first abnormal data of the two abnormal data is used as the compensation start time; after the two consecutive abnormal data, when two consecutive normal respiratory data appear, the preset sampling time corresponding to the second respiratory data of the two consecutive normal respiratory data is used as the compensation end time; all preset sampling times between the compensation start time and the compensation end time constitute each compensation window of the respiratory data of each dimension in the first category of data; and the compensation performance degree is obtained according to the compensation performance degree calculation formula, which is as follows: Where, Indicates the degree of patient compensation performance; Indicates the number of dimensions of respiratory data in the first category of data; Indicates the degree of overlap of abnormal data in all dimensions corresponding to the preset sampling moments in the first category of data, which can be directly obtained by existing technologies; Indicates that in the first category of data, Dimensional compensation window; Indicates that in the first category of data, The number of compensation windows in each dimension; Represents the normalization function.

[0009] Furthermore, the method for obtaining the symptom severity includes: in the respiratory data of each dimension of the second category of data, when two consecutive abnormal data appear, the preset sampling time corresponding to the first abnormal data of the two consecutive abnormal data is used as the decompensation start time; after the two consecutive abnormal data, when two consecutive normal respiratory data appear, the preset sampling time corresponding to the second respiratory data of the two consecutive normal respiratory data is used as the decompensation end time; all preset sampling times between the decompensation start time and the decompensation end time constitute the decompensation window of the respiratory data of each dimension in the second category of data; and obtaining the symptom severity according to the symptom severity calculation formula, which is as follows: Where, Indicates the severity of the patient's symptoms; Indicates the number of dimensions of respiratory data in the second category of data; Indicates the degree of overlap of abnormal data in all dimensions corresponding to the preset sampling moments in the second category of data, which can be directly obtained by existing technologies; In the second type of data, The number of decompensation windows in each dimension; Indicates the number of dimensions of respiratory data in the first category of data; Indicates that in the first category of data, The length of each compensation window in each dimension; Indicates the mean of the abnormality levels of the abnormal data with the largest abnormality level in each dimension; Represents the normalization function.

[0010] Furthermore, the method for obtaining the physical condition score includes: setting a preset first weight for the patient's physiological data fluctuation degree, setting a preset second weight for the patient's compensatory performance degree, and setting a preset third weight for the patient's symptom severity; taking the sum of the product of the preset first weight and the physiological data fluctuation degree, the product of the preset second weight and the compensatory performance degree, and the product of the preset third weight and the symptom severity as the patient's physical illness probability; and taking the difference between the value 1 and the patient's physical illness probability as the patient's physical condition score.

[0011] Furthermore, the method for obtaining the degree of stability of the disease includes: obtaining the degree of stability of the disease according to a calculation formula for the degree of stability of the disease, and the calculation formula for the degree of stability of the disease is as follows: Where, Indicates the patient's condition stability; Indicates the preset number of preset cycles; Indicates the The PEF variation rate within a preset period is a first-order difference between the PEF variability in the first preset period and the PEF variability in the second preset period; a first-order difference between the body state score in the first preset period and the body state score in the second preset period; a first-order difference between the body state score in the first preset period and the body state score in the second preset period; a first-order difference between the body state score in the first preset period and the body state score in the second preset period; a mean value of the PEF variability in all preset periods. a mean value of the body state score in all preset periods.

[0012] The system comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the intelligent analysis method of respiratory disease data.

[0013] The present application has the following advantages: the present application obtains respiratory tract data of each dimension at different preset sampling times of a patient; when the patient has an asthma attack, a compensatory window period occurs, and the respiratory tract data changes dramatically at this time, so the abnormality degree of respiratory tract data of different dimensions is analyzed; after the patient has an asthma attack, the body usually compensates to maintain the stability of physiological indicators, and after compensation, the body begins to fail, causing dramatic changes in respiratory tract data of each dimension, so the respiratory tract data of each dimension is divided into first-type data and second-type data according to the physiological state of the patient; the physiological data fluctuation degree, compensatory performance degree, and symptom severity of the patient are analyzed by analyzing the body state performance of the patient in a period of time through the two types of data; the body state score of the patient is obtained through the physiological data fluctuation degree, compensatory performance degree, and symptom severity of the patient; since the daily variability of PEF is often used in medicine to determine the degree of disease development, the patient's PEF variability distribution and body state score distribution in a preset number of preset periods are combined to obtain the patient's disease stability degree; the respiratory disease data of the patient is intelligently analyzed according to the body state score and the disease stability degree of the patient. The present application can accurately analyze the patient's body, so that relevant personnel can accurately identify and intervene in advance. BRIEF DESCRIPTION OF DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, a brief introduction will be given to the drawings needed in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor based on these drawings.

[0015] Figure 1A flow chart of a respiratory disease data intelligent analysis method provided by an embodiment of the present application; Figure 2 A block diagram of a respiratory disease data intelligent analysis system provided by an embodiment of the present application. DETAILED DESCRIPTION

[0016] In order to further illustrate the technical means and effects taken by the present application to achieve the predetermined object, the specific implementation, structure, features and effects of the respiratory disease data intelligent analysis method and system according to the present application are described in detail below in combination with the drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0017] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0018] The specific scheme of the respiratory disease data intelligent analysis method and system provided by the present application is specifically described below in combination with the drawings.

[0019] Please refer to Figure 1 which shows a respiratory disease data intelligent analysis method provided by an embodiment of the present application, which comprises the following steps: step S1: acquiring respiratory tract data of each dimension at different preset sampling time points of a patient.

[0020] The present application is mainly applied to the data monitoring scene of respiratory diseases of patients. Since the real-time physiological data of respiratory diseases is an important basis for evaluating the condition, and respiratory diseases can cause changes in many physiological indicators of patients, such as heart rate, blood oxygen saturation, respiratory rate, etc., the respiratory tract data of each dimension at different preset sampling time points of the patient is first acquired.

[0021] In an embodiment of the present application, the condition monitoring of children with asthma is mainly studied, therefore, the asthma-related physiological data of children is first acquired by using a wearable device. In an embodiment of the present application, the dimensions are set as heart rate, blood oxygen saturation, respiratory rate, blood pressure, and carbon dioxide level. The acquisition device can be a portable device such as a wearable stethoscope, a patch sensor, a bracelet, etc., which is not limited here. Thus, the respiratory tract data of each dimension required in the subsequent steps is obtained.

[0022] In one embodiment of the present application, since the collection frequency of respiratory tract data of each dimension usually has certain difference, in order to compare and analyze the respiratory tract data of different dimensions at the same time, the least square method is used to fit the discrete real-time data to obtain the continuous respiratory tract data curve of each dimension. At this time, the preset sampling time is set to 12 hours, that is, respiratory tract data is collected every 12 hours on the respiratory tract data curve of different dimensions. It should be noted that the preset sampling time can be set by itself and is not limited here.

[0023] Step S2: optionally, respiratory tract data of one dimension is taken as reference dimension respiratory tract data; according to the reference dimension respiratory tract data distribution characteristics obtained in each preset sampling time in each preset time window, the abnormality degree of the reference dimension respiratory tract data in each preset sampling time is obtained; according to the abnormality degree of the respiratory tract data of each dimension, the abnormal data of each dimension is screened out; according to the physiological state of the patient, the respiratory tract data of each dimension is divided into first type data and second type data; according to the abnormality degree distribution characteristics of the respiratory tract data of each dimension and the abnormal data distribution, the physiological data fluctuation degree of the patient is obtained; according to the coincidence degree of the abnormal data of different dimensions corresponding to the preset sampling time in the first type data of different dimensions and the abnormal data distribution of each dimension in the first type data, the compensatory performance degree of the patient is obtained; according to the coincidence degree of the abnormal data of different dimensions corresponding to the preset sampling time in each dimension in the second type data, the abnormal data distribution of each dimension in the first type data and the mean value of the respiratory tract data with the highest abnormality degree in each dimension, the symptom severity of the patient is obtained; according to the physiological data fluctuation degree, the compensatory performance degree and the symptom severity, the physical state score of the patient is obtained.

[0024] In the prior art, when asthma patients have asthma attacks, their bodies will maintain physiological functions through a series of compensatory mechanisms within a period of time, and the patients will produce some spontaneous behaviors such as increasing respiratory frequency to ensure some important physiological functions such as normal blood oxygen level, so this period is called compensatory window period. Since the airway of children is very narrow and the respiratory system is very sensitive, the physiological data fluctuation of the child patient when the body is abnormal will be more violent and the compensatory window period will be shorter, therefore, in the embodiment of the present application, according to the reference dimension respiratory tract data distribution characteristics obtained in each preset sampling time in each preset time window, the abnormality degree of the reference dimension respiratory tract data in each preset sampling time is obtained.

[0025] Preferably, in one embodiment of the present application, the abnormality degree acquisition method comprises: obtaining the abnormality degree according to the abnormality degree calculation formula, and the abnormality degree calculation formula is as follows: In the formula, represents the reference dimension under the first an abnormality degree of the respiratory tract data at the preset sampling time point; indicates a preset time window number; indicates the respiratory tract data of the reference dimension in the first preset sampling time point; indicates the respiratory tract data of the reference dimension in the first preset time window; indicates the respiratory tract data of the reference dimension in the first preset time window;

[0026] In the abnormality degree calculation formula, the greater the difference between the respiratory tract data of the reference dimension in the first preset sampling time point and the mean value of the respiratory tract data in the preset time window in which the respiratory tract data of the reference dimension in the first preset sampling time point is, the greater the deviation from the normal value is, and the greater the abnormality degree of the first preset sampling time point is; the smaller the standard deviation of the respiratory tract data of the reference dimension in the first preset time window is, the greater the deviation from the normal value is, that is, the greater the abnormality degree of the first preset time window is; the deviation between the first preset sampling time point and the normal value in each preset time window is calculated, and finally the abnormality degree of the respiratory tract data of the reference dimension is obtained.

[0027] In an embodiment of the present application, according to the general mode of asthma occurrence, the preset time window is set as three windows of long, medium and short, the short window is 5 minutes, which is used to monitor the instant physiological state of the child and capture the acute change signal; the medium window is 1 hour, which is used to monitor the short-term stable state of the child and capture the change trend signal; and the long window is 24 hours, which is used to monitor the overall control level of the child and capture the disease progression state. The final abnormality factor is determined by the deviation of the respiratory tract data at different preset sampling time points in the three different time windows, so as to avoid the calculation error caused by a single time window.

[0028] Preferably, in an embodiment of the present application, the method for obtaining abnormal data comprises: calculating the abnormality degree standard deviation of the respiratory tract data of each dimension as a first standard deviation; and taking the respiratory tract data of each dimension with an absolute value of the abnormality degree greater than 3 times the first standard deviation as abnormal data. It should be noted that the judgment standard of abnormal data can be set by itself and is not limited herein. In the embodiment of the present application, the Z-score conventional method is used for judgment, which is a technical means familiar to those skilled in the art and is not limited herein.

[0029] ​​​In the prior art, after an asthma attack, the patient's body usually compensates to maintain the stability of physiological indicators. After the compensation, the body will begin to fail, causing drastic changes in respiratory data of various dimensions. Therefore, in an embodiment of the present invention, the respiratory data of each dimension are divided into the first category of data and the second category of data according to the patient's physiological state. Specifically: the first category of data is the data that changes during the compensation window period in order to maintain physiological stability, such as respiratory data of dimensions such as respiratory rate, heart rate, and blood pressure, which will accelerate sharply during the compensation window period; the second category of data is the data that changes after the body begins to fail after the compensation window period, such as blood oxygen saturation will drop sharply and carbon dioxide levels will rise sharply. It also explains that the data dimensions possessed by the first category of data and the second category of data are different, so as to facilitate understanding of subsequent steps. It should be noted that the judgment and classification methods of the first category of data and the second category of data are technical means well known to those skilled in the art and are not limited here.

[0030] In an embodiment of the present invention, two types of data are used to determine the patient's physical condition over a period of time, specifically analyzing the patient's physiological data fluctuation degree, compensatory performance degree, and symptom severity. The specific method includes: Preferably, in one embodiment of the present invention, the method for obtaining the physiological data fluctuation degree includes: obtaining the physiological data fluctuation degree according to a physiological data fluctuation degree calculation formula, and the physiological data fluctuation degree calculation formula is as follows: Where, Indicates the degree of fluctuation of the patient's physiological data; The number of dimensions representing the respiratory data; Indicates the The mean of abnormal data in each dimension; Indicates the The number of abnormal data in each dimension; No. The number of respiratory data in each dimension; Represents the normalization function.

[0031] In the calculation formula of physiological data fluctuation degree, The mean of abnormal data in each dimension The larger the The more unstable the respiratory data in the first dimension is, the greater the fluctuation of the patient's physiological data will be. The proportion of abnormal data in each dimension The larger the The more unstable the respiratory data in each dimension is, the greater the fluctuation of the patient's physiological data will be.

[0032] Preferably, in one embodiment of the present application, the method for obtaining the compensatory performance degree comprises: in the respiratory tract data of each dimension of the first type of data, when two continuous abnormal data appear, taking the preset sampling time corresponding to the former abnormal data in the two continuous abnormal data as a compensatory start time, after the two continuous abnormal data, when two continuous normal respiratory tract data appear, taking the preset sampling time corresponding to the latter respiratory tract data in the two continuous normal respiratory tract data as a compensatory end time; and taking all the preset sampling times during the compensatory start time to the compensatory end time as each compensatory window of the respiratory tract data of each dimension in the first type of data.

[0033] The compensatory performance degree is obtained according to a compensatory performance degree calculation formula as follows: In the formula, Compensatory performance degree represents the compensatory performance degree of the patient; n represents the number of dimensions of the respiratory tract data in the first type of data; Coincidence degree represents the coincidence degree of the abnormal data corresponding to the preset sampling times in all the dimensions in the first type of data, which can be directly obtained by the prior art; and Compensatory window number represents the number of compensatory windows of the first dimension in the first type of data.

[0034] It should be noted that the use of the within-group correlation coefficient to calculate the coincidence degree of the abnormal data of different dimensions is a technical means familiar to those skilled in the art, and will not be described here.

[0035] In the compensatory performance degree calculation formula, the higher the coincidence degree of the abnormal data in all the dimensions in the first type of data, the greater the possibility of the compensatory reaction of the patient, that is, the greater the compensatory performance degree of the patient; the more the number of compensatory windows of each dimension in the first type of data, the greater the compensatory reaction of the patient, and the greater the compensatory performance degree of the patient.

[0036] Preferably, in one embodiment of the present application, the method for obtaining the symptom severity degree comprises: in the respiratory tract data of each dimension of the second type of data, when two continuous abnormal data appear, taking the preset sampling time corresponding to the former abnormal data in the two continuous abnormal data as a decompensation start time, after the two continuous abnormal data, when two continuous normal respiratory tract data appear, taking the preset sampling time corresponding to the latter respiratory tract data in the two continuous normal respiratory tract data as a decompensation end time; and taking all the preset sampling times during the decompensation start time to the decompensation end time as a decompensation window of the respiratory tract data of each dimension in the second type of data.

[0037] ​​​​​​The symptom severity is obtained according to the symptom severity calculation formula. The symptom severity calculation formula is as follows: Where, Indicates the severity of the patient's symptoms; Indicates the number of dimensions of respiratory data in the second category of data; Indicates the degree of overlap of abnormal data in all dimensions corresponding to the preset sampling moments in the second category of data, which can be directly obtained by existing technologies; In the second type of data, The number of decompensation windows in each dimension; Indicates the number of dimensions of respiratory data in the first category of data; Indicates that in the first category of data, The length of each compensation window in each dimension; Indicates the mean of the abnormality levels of the abnormal data with the largest abnormality level in each dimension; Represents the normalization function.

[0038] In the symptom severity calculation formula, in the second type of data, the degree of overlap of abnormal data in different dimensions The larger the decompensation window, the larger the decompensation window in different dimensions. The more it is, the more severe the changes in the body's failure to overcome the onset of the disease; and in the first type of data, the average length of the compensation window The shorter it is, the shorter the duration of the compensation process, that is, the faster the body fails; the larger the mean between the abnormal data with the largest abnormality in each dimension, the more dangerous the patient's physiological indicators are, the more severe the symptoms are, and the greater the severity of the patient's symptoms.

[0039] The patient's physical condition score is obtained through the degree of fluctuation of the patient's physiological data, the degree of compensatory performance and the severity of symptoms.

[0040] Preferably, in one embodiment of the present invention, the method for obtaining the physical condition score includes: setting a preset first weight for the degree of fluctuation of the patient's physiological data, setting a preset second weight for the degree of compensation performance of the patient, and setting a preset third weight for the severity of the patient's symptoms; taking the sum of the product of the preset first weight and the degree of fluctuation of the physiological data, the product of the preset second weight and the degree of compensation performance, and the sum of the product of the preset third weight and the severity of the symptoms as the probability of the patient's physical illness.

[0041] The difference between the value 1 and the probability of the patient's physical illness is used as the patient's physical condition score. In an embodiment of the present invention, a physical condition score calculation formula is provided as follows: Where, Indicates the patient's physical status score; indicates the degree of physiological data fluctuation of the patient; indicates the degree of compensatory performance of the patient; indicates the severity of symptoms of the patient.

[0042] In one embodiment of the present application, the first weight is set to 0.3, the second weight is set to 0.3, and the third weight is set to 0.4.

[0043] Wherein, the higher the physical state score, the better the disease state, and the lower the physical state score, the worse the disease state.

[0044] Step S3: obtaining the PEF variability of the patient in each preset period; obtaining the disease stability degree of the patient according to the PEF variability distribution and the physical state score distribution of the patient in the preset number of preset periods; and intelligently analyzing the respiratory tract disease data of the patient according to the physical state score and the disease stability degree of the patient.

[0045] In the prior art, monitoring the peak expiratory flow (PEF) is of great significance for evaluating the development trend of the disease of an asthma patient. In medicine, the day-to-day variability of PEF is often used to judge the degree of disease development, that is, the relative change rate of the peak expiratory flow of the patient in the day and night. However, the PEF of children is lower than that of adults, and has stronger variability. The airway can be significantly contracted and expanded at multiple times in a day, and the time span of the day-to-day variability is too long, which can cause some symptoms to be ignored. Therefore, in an embodiment of the present application, an adaptive monitoring period is used to evaluate the disease.

[0046] In one embodiment of the present application, when the physical state score is greater than 0.8, it is considered that the physical state of the patient is good, and the preset period is set to 1 day; when the physical state score is less than 0.8, it is considered that the physical state of the patient is poor, and the preset period is set to 12 hours; and when the physical state score is less than 0.6, it is considered that the physical state of the patient is poor, and the preset period is set to 8 hours. It should be noted that the preset period can be set by the user, and is not limited herein.

[0047] The PEF variability of the patient in each preset period is calculated, and the specific calculation method is a technical means known to those skilled in the art, which is not described herein. The disease stability degree of the patient is obtained according to the PEF variability distribution and the physical state score distribution of the patient in the preset number of preset periods.

[0048] Preferably, in one embodiment of the present application, the method for obtaining the disease stability degree comprises: obtaining the disease stability degree according to the disease stability degree calculation formula, and the disease stability degree calculation formula is as follows: In the formula, indicates the disease stability degree of the patient; indicates the preset number of the preset period; Indicates the The PEF variation rate within a preset period is The first-order difference between the PEF variation rates within a preset period; Indicates the The physical condition score within the preset period is The first-order difference between the physical condition scores within a preset period; Indicates the mean value of PEF variation rate within all preset cycles; Indicates the average physical condition score within all preset periods.

[0049] In one embodiment of the present invention, the preset number is set to 7, that is, the PEF variation rate and physical condition score of the patient within 7 preset periods are analyzed.

[0050] In the calculation formula for disease stability, the smaller the sum of the first-order difference of the PEF variation rate and the first-order difference of the physical condition score within all preset periods, the smaller the fluctuation of the patient's physical condition and the greater the stability of the patient's disease. The smaller it is, the more consistent the patient's physical condition is with the peak expiratory flow rate index. At this time, the possibility that the patient has a hidden trend of worsening of the condition is less, and the patient's condition is more stable.

[0051] In one embodiment of the present invention, using Reflect the real-time condition status, thus providing indicative assistance to relevant personnel, such as deciding the time and dosage of medication. Ability to assess the development trend of the disease over a period of time in the future to provide early warning of the child's condition, when the stability shows a continuous downward trend or falls below the dangerous threshold When the patient has asthma symptoms, an alert will be issued in time to remind the patient to seek medical treatment, or the remote nursing staff will contact the patient's family for treatment, so as to achieve early identification and intervention of the patient's asthma symptoms and complete intelligent analysis of the data.

[0052] In summary, the respiratory tract data of each dimension at different preset sampling moments of the patient is obtained; respiratory tract data of an optional dimension is taken as the reference dimension respiratory tract data; the abnormality degree of the reference dimension respiratory tract data at each preset sampling moment is obtained according to the distribution characteristics of the reference dimension respiratory tract data obtained at each preset sampling moment in each preset time window; the abnormal data of each dimension is screened out according to the abnormality degree of the respiratory tract data of each dimension; the respiratory tract data of each dimension is divided into first-class data and second-class data according to the physiological state of the patient; the physiological data fluctuation degree of the patient is obtained according to the abnormality degree distribution characteristics of the respiratory tract data of each dimension and the abnormal data distribution; the compensatory performance degree of the patient is obtained according to the coincidence degree of the abnormal data of different dimensions in different dimensions in the first-class data corresponding to the preset sampling moment and the abnormal data distribution of each dimension in the first-class data; the symptom severity of the patient is obtained according to the coincidence degree of the abnormal data of different dimensions in different dimensions in the second-class data corresponding to the preset sampling moment, the abnormal data distribution of each dimension in the first-class data and the mean value of the respiratory tract data with the highest abnormality degree in each dimension; the physical state score of the patient is obtained according to the physiological data fluctuation degree, the compensatory performance degree and the symptom severity; the PEF variation rate of the patient in each preset period is obtained; the disease stability degree of the patient is obtained according to the PEF variation rate distribution and the physical state score distribution of the patient in the preset number of preset periods; and the respiratory tract disease data of the patient is intelligently analyzed according to the physical state score and the disease stability degree of the patient.

[0053] An embodiment of the present application provides a respiratory disease data intelligent analysis system, which comprises a memory, a processor and a computer program, wherein the memory is used for storing the corresponding computer program, the processor is used for running the corresponding computer program, and the computer program can realize the method described in steps S1-S3 when running in the processor, and specifically comprises: a data acquisition module 101, which is used for acquiring respiratory tract data of each dimension at different preset sampling moments of a patient; a data analysis module 102, which is used for selecting respiratory tract data of an optional dimension as reference dimension respiratory tract data; obtaining an abnormality degree of the reference dimension respiratory tract data at each preset sampling moment according to a reference dimension respiratory tract data distribution feature obtained at each preset sampling moment in each preset time window; screening out abnormal data of each dimension according to the abnormality degree of the respiratory tract data of each dimension; dividing the respiratory tract data of each dimension into first type data and second type data according to a physiological state of the patient; obtaining a physiological data fluctuation degree of the patient according to an abnormality degree distribution feature of the respiratory tract data of each dimension and an abnormal data distribution; obtaining a compensatory performance degree of the patient according to a coincidence degree of abnormal data of different dimensions in the first type data corresponding to preset sampling moments and the abnormal data distribution of each dimension in the first type data; obtaining a symptom severity degree of the patient according to a coincidence degree of abnormal data of different dimensions in the second type data corresponding to preset sampling moments, the abnormal data distribution of each dimension in the first type data and a mean value of respiratory tract data with the highest abnormality degree in each dimension; obtaining a physical state score of the patient according to the physiological data fluctuation degree, the compensatory performance degree and the symptom severity degree; a disease condition analysis module 103, which is used for acquiring a PEF variation rate of the patient in each preset period; obtaining a disease condition stability degree of the patient according to a PEF variation rate distribution and a physical state score distribution of the patient in a preset number of preset periods; and intelligently analyzing the respiratory disease data of the patient according to the physical state score and the disease condition stability degree of the patient.

[0054] An embodiment of the present application provides a computer device, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, and the processor realizes the method described in steps S1-S3 when executing the computer program.

[0055] An embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program realizes the method described in steps S1-S3 when executed by a processor.

[0056] It is to be noted that the sequential order of the above-described embodiments of the present application only for the purpose of description, but not the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0057] Each of the embodiments in the specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the difference from other embodiments.

Claims

1. A method for intelligent analysis of respiratory disease data, characterized in that: The method includes: obtaining respiratory data of each dimension at different preset sampling moments of the patient; selecting respiratory data of any one dimension as respiratory data of a reference dimension; obtaining the abnormality of respiratory data of the reference dimension at each preset sampling moment according to the distribution characteristics of respiratory data of the reference dimension obtained at each preset sampling moment in each preset time window; screening out abnormal data of each dimension according to the abnormality of respiratory data of each dimension; dividing respiratory data of each dimension into first category data and second category data according to the physiological state of the patient; obtaining the fluctuation degree of physiological data of the patient according to the abnormality distribution characteristics of respiratory data of each dimension and the distribution of abnormal data; obtaining the overlap degree of abnormal data of different dimensions corresponding to the preset sampling moments in the first category data, and the first category data. The method comprises the following steps: first, the method comprises the following steps: first, the method comprises the following steps: first, the method comprises the following steps: a) the distribution of abnormal data in each dimension of the first category data, and b) the distribution of abnormal data in each dimension of the second category data to obtain the degree of compensatory performance of the patient; c) the distribution of abnormal data in each dimension of the first category data corresponding to the preset sampling time, and c) the distribution of abnormal data in each dimension of the first category data and the mean of the respiratory data with the highest degree of abnormality in each dimension to obtain the severity of the patient's symptoms; e) the method comprises the following steps: f) the physical condition score of the patient according to the degree of fluctuation of the physiological data, the degree of compensatory performance and the severity of the symptoms; f) the PEF variation rate of the patient in each preset cycle; f) the distribution of the PEF variation rate and the physical condition score of the patient within a preset number of preset cycles to obtain the stability of the patient's condition; and f) the method comprises the following steps: ... an intelligent analysis of the patient's respiratory disease data according to the patient's physical condition score and the stability of the condition.

2. A respiratory disease data intelligent analysis method according to claim 1, characterized in that: The method for obtaining the abnormality degree includes: obtaining the abnormality degree according to an abnormality degree calculation formula, and the abnormality degree calculation formula is as follows: Where, Indicates the reference dimension The abnormality of respiratory data at a preset sampling time; Indicates the number of preset time windows; Indicates the Respiratory data of the reference dimension at a preset sampling moment; Indicates the The mean value of respiratory data in the reference dimension in a preset time window; Indicates the The standard deviation of respiratory data in the reference dimension within a preset time window.

3. The method for intelligent analysis of respiratory disease data according to claim 1, characterized in that: The method for acquiring abnormal data includes: calculating the standard deviation of the abnormality degree of respiratory data in each dimension as a first standard deviation; and taking the respiratory data in each dimension whose absolute value of the abnormality degree is greater than 3 times the first standard deviation as abnormal data.

4. The method for intelligent analysis of respiratory disease data according to claim 1, characterized in that: The method for obtaining the physiological data fluctuation degree includes: obtaining the physiological data fluctuation degree according to a physiological data fluctuation degree calculation formula, and the physiological data fluctuation degree calculation formula is as follows: Where, Indicates the degree of fluctuation of the patient's physiological data; The number of dimensions representing the respiratory data; Indicates the The mean of abnormal data in each dimension; Indicates the The number of abnormal data in each dimension; No. The number of respiratory data in each dimension; Represents the normalization function.

5. The method for intelligent analysis of respiratory disease data according to claim 1, characterized in that: The method for obtaining the degree of compensation performance includes: in the respiratory data of each dimension of the first category of data, when two consecutive abnormal data appear, using the preset sampling time corresponding to the first abnormal data of the two abnormal data as the compensation start time; after the two consecutive abnormal data, when two consecutive normal respiratory data appear, using the preset sampling time corresponding to the second respiratory data of the two consecutive normal respiratory data as the compensation end time; all preset sampling times between the compensation start time and the compensation end time constitute each compensation window of the respiratory data of each dimension in the first category of data; and obtaining the degree of compensation performance according to a compensation performance calculation formula, wherein the compensation performance calculation formula is as follows: Where, Indicates the degree of patient compensation performance; Indicates the number of dimensions of respiratory data in the first category of data; Indicates the degree of overlap of abnormal data in all dimensions corresponding to the preset sampling moments in the first category of data, which can be directly obtained by existing technologies; Indicates that in the first category of data, The number of compensation windows in each dimension; Represents the normalization function.

6. The method for intelligent analysis of respiratory disease data according to claim 1, characterized in that: The method for obtaining the symptom severity includes: in the respiratory data of each dimension of the second type of data, when two consecutive abnormal data appear, using the preset sampling time corresponding to the first abnormal data of the two consecutive abnormal data as the decompensation start time; after the two consecutive abnormal data, when two consecutive normal respiratory data appear, using the preset sampling time corresponding to the second respiratory data of the two consecutive normal respiratory data as the decompensation end time; all preset sampling times between the decompensation start time and the decompensation end time constitute the decompensation window of the respiratory data of each dimension in the second type of data; and obtaining the symptom severity according to a symptom severity calculation formula, the symptom severity calculation formula is as follows: Where, Indicates the severity of the patient's symptoms; Indicates the number of dimensions of respiratory data in the second category of data; Indicates the degree of overlap of abnormal data in all dimensions corresponding to the preset sampling moments in the second category of data, which can be directly obtained by existing technologies; In the second type of data, The number of decompensation windows in each dimension; Indicates the number of dimensions of respiratory data in the first category of data; Indicates that in the first category of data, The length of each compensation window in each dimension; Indicates the mean of the abnormality levels of the abnormal data with the largest abnormality level in each dimension; Represents the normalization function.

7. The method for intelligent analysis of respiratory disease data according to claim 1, characterized in that: The method for obtaining the physical condition score includes: setting a preset first weight for the patient's physiological data fluctuation degree, setting a preset second weight for the patient's compensatory performance degree, and setting a preset third weight for the patient's symptom severity; taking the sum of the product of the preset first weight and the physiological data fluctuation degree, the product of the preset second weight and the compensatory performance degree, and the product of the preset third weight and the symptom severity as the patient's physical illness probability; and taking the difference between the value 1 and the patient's physical illness probability as the patient's physical condition score.

8. The method for intelligent analysis of respiratory disease data according to claim 1, characterized in that: The method for obtaining the degree of stability of the disease includes: obtaining the degree of stability of the disease according to a calculation formula for the degree of stability of the disease, and the calculation formula for the degree of stability of the disease is as follows: Where, Indicates the patient's condition stability; Indicates the preset number of preset cycles; Indicates the The PEF variation rate within a preset period is The first-order difference between the PEF variation rates within a preset period; Indicates the The physical condition score within the preset period is The first-order difference between the physical condition scores within a preset period; Indicates the mean value of PEF variation rate within all preset cycles; Indicates the average physical condition score within all preset periods.

9. A respiratory disease data intelligent analysis system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for intelligent analysis of respiratory disease data as described in any one of claims 1 to 8 are implemented.

Images