Patient follow-up information management method and system

Abstract

The present invention relates to the field of medical data processing technology, and specifically to a patient follow-up information management method and system. The present invention analyzes the changing trend of the initial physiological data at different follow-up moments for any dimension to obtain the corrected physiological data at each follow-up moment; according to the changing trend of the corrected physiological data within the neighborhood range of each follow-up moment in different dimensions for each patient, the follow-up effectiveness of each patient at each follow-up moment is obtained; combined with the degree of concern about the condition of each patient at the real-time follow-up moment, all patients are clustered to obtain patient clusters; according to the degree of concern about the condition of all patients in each patient cluster at the real-time follow-up moment, the follow-up frequency increase coefficient of each patient cluster is obtained; combined with the initial follow-up number, the adjusted follow-up number of each patient is obtained. The present invention determines the accurate follow-up number by analyzing the patient's actual condition manifestations, thereby improving the effect of patient follow-up nursing management.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data processing, and in particular to a patient follow-up information management method and system. Background Art

[0002] Symptom management of a patient's condition is a continuous process, and its development is closely related to factors such as the patient's physiological characteristics and personal living habits. Therefore, it is necessary to monitor changes in the patient's physiological condition, which will help reduce the possibility of complications and improve the patient's quality of life.

[0003] In the existing technology, the physiological data of discharged patients are collected according to fixed follow-up time interval rules, and then the patient's condition is analyzed; however, in real life, the physical fitness of different patients varies, and the fixed follow-up time cannot adapt to patients with different disease progression. In addition, due to the lack of constraints on patients after discharge, the physiological data monitored during follow-up may be missing or distorted, resulting in a lack of authenticity of some data, errors in the analysis of the patient's condition, and poor effect of patient follow-up management. Summary of the Invention

[0004] In order to solve the technical problems that fixed follow-up time cannot adapt to patients with different disease progression, and the physiological data monitored during follow-up lacks authenticity, resulting in poor patient follow-up management results, the purpose of the present invention is to provide a patient follow-up information management method and system. The technical solutions adopted are as follows:

[0005] The present invention proposes a patient follow-up information management method, the method comprising:

[0006] Obtain the patient's multi-dimensional initial physiological data at each follow-up moment;

[0007] For any dimension of each patient, the abnormal probability of each follow-up moment is obtained based on the changing trend of the initial physiological data at different follow-up moments, and abnormal follow-up moments are screened out; based on the abnormal probability of each abnormal follow-up moment and the distribution characteristics of the initial physiological data in the neighborhood, the corrected physiological data at each follow-up moment is obtained;

[0008] Based on the number of abnormal follow-up moments in the neighborhood of each follow-up moment in different dimensions for each patient and the changing trend of the corrected physiological data, the follow-up effectiveness of each patient at each follow-up moment is obtained; based on the distribution of follow-up effectiveness of each patient at different follow-up moments, the degree of attention paid to the condition of each patient at the real-time follow-up moment is obtained; based on the follow-up effectiveness and degree of attention paid to the condition of each patient at the real-time follow-up moment, all patients are clustered to obtain patient clusters;

[0009] According to the disease attention level of all patients in each patient cluster at the time of real-time follow-up, the follow-up frequency increase coefficient of each patient cluster is obtained; according to the follow-up frequency increase coefficient of the patient cluster to which each patient belongs and the initial follow-up number, the adjusted follow-up number of each patient is obtained.

[0010] Furthermore, the method for obtaining the abnormal possibility includes:

[0011] For any dimension, if there is no initial physiological data at the follow-up time, the mean of the initial physiological data between the follow-up times within the historical range is obtained as the initial physiological data at the corresponding follow-up time;

[0012] The mean of the initial physiological data at all follow-up moments was obtained as the initial average physiological data; the difference between the initial physiological data at each follow-up moment and the initial average physiological data was obtained as the data fluctuation coefficient at each follow-up moment;

[0013] The mean of the data fluctuation coefficients at all follow-up moments was obtained as the average data fluctuation coefficient; the coefficient difference between the data fluctuation coefficient at each follow-up moment and the average data fluctuation coefficient was obtained, and the data abnormality coefficient at each follow-up moment was obtained based on the deviation of the coefficient difference from the average data fluctuation coefficient;

[0014] The abnormality probability of each follow-up moment was obtained based on the data abnormality coefficient at each follow-up moment and the difference in data fluctuation coefficient between each follow-up moment and the next follow-up moment. The difference in data abnormality coefficient and data fluctuation coefficient was positively correlated with the abnormality probability.

[0015] Furthermore, the method for obtaining the abnormal follow-up time includes:

[0016] If the abnormal possibility of the follow-up time is greater than the preset abnormal threshold, the corresponding follow-up time will be regarded as the abnormal follow-up time.

[0017] Furthermore, the method for obtaining the corrected physiological data includes:

[0018] For any dimension, for each abnormal follow-up moment, the data correction coefficient of each abnormal follow-up moment is obtained according to the distribution characteristics of the initial physiological data within the neighborhood; the mean of the initial physiological data of all follow-up moments within the neighborhood is obtained as the local data level;

[0019] Performing gain adjustment on the initial physiological data according to the data correction coefficient at each abnormal follow-up moment and the difference between the local data level and the initial physiological data to obtain the corrected physiological data at each abnormal follow-up moment;

[0020] For follow-up moments that do not correspond to abnormal follow-up moments, the corresponding initial physiological data are used as corrected physiological data.

[0021] Furthermore, the method for obtaining the data correction coefficient includes:

[0022] For any dimension, the mean of all initial physiological data of the patient during the hospitalization period was obtained as the reference physiological data;

[0023] Calculate the difference between the initial physiological data and the reference physiological data at each other follow-up time within the neighborhood as the first difference; obtain the cumulative value of the first differences corresponding to all other follow-up times within the neighborhood as the overall difference value;

[0024] The product of the abnormal probability and the overall difference value at each abnormal follow-up moment was obtained and normalized to serve as the data correction coefficient for each abnormal follow-up moment.

[0025] Furthermore, the method for obtaining the effectiveness of the follow-up includes:

[0026] For each dimension of each patient, the changing trend of the corrected physiological data is analyzed based on the abnormality probability acquisition method to obtain the abnormality correction probability at each follow-up moment;

[0027] The ratio between the number of all follow-up moments and the number of abnormal follow-up moments in the neighborhood of each follow-up moment is obtained as the local effective coefficient;

[0028] The follow-up effectiveness of each patient at each follow-up moment was obtained based on the abnormality correction possibility and local effectiveness coefficient of each patient in different dimensions at each follow-up moment. The abnormality correction possibility was negatively correlated with the follow-up effectiveness, while the local effectiveness coefficient was positively correlated with the follow-up effectiveness.

[0029] Furthermore, the method for obtaining the disease concern level includes:

[0030] Obtain the mean follow-up effectiveness of all patients at the real-time follow-up time as the average follow-up effectiveness;

[0031] For each patient, the ratio of the follow-up effectiveness between the earliest follow-up time and the real-time follow-up time was obtained as the relative attention level of the patient at the real-time follow-up time;

[0032] The difference between the patient's follow-up effectiveness at the real-time follow-up moment and the average follow-up effectiveness was calculated as the degree difference. A negative correlation mapping was performed on the patient's degree difference at the real-time follow-up moment, and the product of the negative correlation mapping result and the relative attention degree was calculated as the patient's disease attention degree at the real-time follow-up moment.

[0033] Furthermore, the method for obtaining the follow-up frequency increase coefficient includes:

[0034] The mean of the disease attention levels of all patients in each patient cluster at the time of real-time follow-up is obtained as the local disease attention level; the mean of the disease attention levels of all patients at the time of real-time follow-up is obtained as the overall disease attention level;

[0035] According to the number of patients in each patient cluster and the ratio of the degree of local attention to the disease and the degree of overall attention to the disease, the follow-up frequency increase coefficient of each patient cluster was obtained. The degree ratio was positively correlated with the follow-up frequency increase coefficient, and the number of patients was negatively correlated with the follow-up frequency increase coefficient.

[0036] Furthermore, the method for obtaining the adjusted number of follow-up visits includes:

[0037] The sum of the positive integer 1 and the follow-up frequency increase coefficient of each patient's cluster is calculated as the frequency adjustment weight. The product of the initial follow-up number and the frequency adjustment weight is calculated and rounded down to obtain the adjusted follow-up number.

[0038] The present invention also proposes a patient follow-up information management system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of any one of the patient follow-up information management methods are implemented.

[0039] The present invention has the following beneficial effects:

[0040] The present invention analyzes the changing trends of initial physiological data at different follow-up moments in any dimension to obtain revised physiological data at each follow-up moment, reflecting the patient's true physiological state and improving the accuracy and reliability of the data. Based on the number of abnormal follow-up moments in the neighborhood of each follow-up moment in different dimensions for each patient and the changing trends of the revised physiological data, the follow-up effectiveness of each patient at each follow-up moment is obtained, and the authenticity of the condition represented by the physiological data in each dimension is evaluated. Based on the distribution of follow-up effectiveness of each patient at different follow-up moments, the degree of concern for the condition of each patient at the real-time follow-up moment is obtained, which helps to more accurately grasp the patient's condition dynamics. Based on the follow-up effectiveness and degree of concern for each patient at the real-time follow-up moment, all patients are clustered to obtain patient clusters, and patients with similar follow-up effectiveness and degree of concern are grouped into the same cluster, which helps to more efficiently analyze the patient population. The degree of concern for the condition of all patients in each patient cluster at the real-time follow-up moment is analyzed to obtain the adjusted number of follow-up visits for each patient, which more accurately meets the patient's specific follow-up needs and helps to understand the changes in each patient's condition. The present invention determines the accurate number of follow-up visits by analyzing the patient's actual condition manifestations, thereby improving the effect of patient follow-up nursing management. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 A flowchart of a patient follow-up information management method provided by one embodiment of the present invention;

[0043] Figure 2 A flow chart of a method for obtaining abnormality probability provided by one embodiment of the present invention;

[0044] Figure 3 A flow chart of a method for obtaining a data correction coefficient provided by one embodiment of the present invention;

[0045] Figure 4 A flow chart of a method for obtaining the effectiveness of follow-up provided by one embodiment of the present invention;

[0046] Figure 5 This is a flow chart of a method for obtaining the degree of concern of a disease condition provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0047] To further illustrate the technical means and effectiveness of the present invention in achieving its intended objectives, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail a patient follow-up information management method and system according to the present invention, including its specific implementation, structure, features, and effectiveness. In the following description, references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0048] 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.

[0049] The following describes in detail a patient follow-up information management method and system provided by the present invention with reference to the accompanying drawings.

[0050] See also Figure 1 , which shows a method flow chart of a patient follow-up information management method provided by one embodiment of the present invention, specifically comprising:

[0051] Step S1: Acquire the patient's multi-dimensional initial physiological data at each follow-up time.

[0052] In the embodiment of the present invention, in order to monitor the patient's specific condition changes and physiological indicators, control the risk factors of the disease, and improve the patient's quality of life, it is necessary to provide follow-up care for the patient and timely understand the patient's physiological data; first, obtain the patient's initial physiological data in multiple dimensions at each follow-up time; wherein the multiple dimensions include heart rate , blood oxygen saturation , systolic blood pressure , diastolic blood pressure ,body temperature , Morisky medication compliance scale score , Seattle Angina Scale score And other physiological indicators that are easy for patients to measure themselves.

[0053] It should be noted that, in one embodiment of the present invention, when obtaining physiological data at the time of follow-up, if the follow-up time is within the first month of follow-up, the follow-up time interval is one week; if it is after one month of follow-up, the follow-up time interval is one month; in one embodiment of the present invention, the time interval can be set according to the specific situation, which is not limited or elaborated here.

[0054] It should be noted that, in the embodiments of the present invention, in order to facilitate subsequent processing of the data and avoid differences in units and numerical magnitudes between the data, the data is standardized to eliminate the influence of dimensions in data calculations.

[0055] Step S2: For any dimension of each patient, according to the changing trend of the initial physiological data at different follow-up moments, the abnormal possibility of each follow-up moment is obtained, and the abnormal follow-up moments are screened out; according to the abnormal possibility of each abnormal follow-up moment and the distribution characteristics of the initial physiological data in the neighborhood range, the corrected physiological data at each follow-up moment is obtained.

[0056] Due to differences in individual physical fitness, the course of the disease usually fluctuates to a certain extent. Data monitoring during the follow-up process can capture subtle changes in the data at different times. By analyzing the changing trends of physiological data at different follow-up times, it is helpful to analyze abnormal fluctuations or trends in the patient's physiological data. For any dimension of each patient, the possibility of abnormality at each follow-up time is obtained based on the changing trends of the initial physiological data at different follow-up times.

[0057] Preferably, in one embodiment of the present invention, the method for obtaining the abnormal possibility is as follows: Figure 2 , which shows a flow chart of a method for obtaining abnormality possibility, including:

[0058] Step S201: For any dimension, if there is no initial physiological data at the follow-up time, the mean of the initial physiological data between the follow-up times within the historical range is obtained as the initial physiological data at the corresponding follow-up time.

[0059] Since the degree of change in patients' conditions is relatively close in a short period of time, the mean is used to fill in these missing values ​​to ensure the integrity and continuity of the data.

[0060] It should be noted that, in one embodiment of the present invention, all follow-up moments are analyzed in chronological order, and the historical range is the range consisting of two adjacent historical follow-up moments for a certain follow-up moment; in other embodiments of the present invention, the size of the historical range can be set according to the specific circumstances, and is not limited or elaborated here.

[0061] Step S202: obtaining the mean of the initial physiological data at all follow-up moments as the initial average physiological data; obtaining the difference between the initial physiological data at each follow-up moment and the initial average physiological data as the data fluctuation coefficient at each follow-up moment.

[0062] Due to differences in individual physical fitness, there are certain fluctuations in the course of patients' diseases, and the corresponding physiological data also fluctuate accordingly. The initial average physiological data represents the baseline value of the overall level of initial physiological data at all follow-up moments. By analyzing the difference between the initial physiological data and the initial average physiological data at each follow-up moment, the degree of fluctuation of the data at the follow-up moment deviating from the normal level is quantified.

[0063] Step S203: Obtain the mean of the data fluctuation coefficients at all follow-up moments as the average data fluctuation coefficient; obtain the coefficient difference between the data fluctuation coefficient at each follow-up moment and the average data fluctuation coefficient, and obtain the data abnormality coefficient at each follow-up moment based on the deviation of the coefficient difference relative to the average data fluctuation coefficient.

[0064] The data fluctuation coefficient reflects the fluctuation of the data at each follow-up moment. The overall situation of the data fluctuation coefficients at all follow-up moments is quantified by averaging, reflecting the general level of data fluctuation during the entire follow-up process.

[0065] It should be noted that, in one embodiment of the present invention, by analyzing the coefficient difference between the data fluctuation coefficient at each follow-up moment and the average data fluctuation coefficient and the ratio of the average data fluctuation coefficient, the degree of deviation of the data fluctuation coefficient at each follow-up moment relative to the average data fluctuation coefficient can be reflected. The greater the degree of deviation, the greater the possibility of data abnormality. In other embodiments of the present invention, the coefficient difference can also be calculated. and the average data fluctuation coefficient The difference between them represents the deviation of the coefficient difference relative to the average data fluctuation coefficient. The correlation relationship is constructed that the larger the data fluctuation coefficient is relative to the average data fluctuation coefficient, the more likely there is an anomaly, and the greater the possibility of an anomaly. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0066] Step S204: Obtain the abnormality probability of each follow-up moment based on the data abnormality coefficient at each follow-up moment and the difference in the data fluctuation coefficient between each follow-up moment and the next follow-up moment. The data abnormality coefficient and the difference in the data fluctuation coefficient are both positively correlated with the abnormality probability.

[0067] Among them, the larger the data abnormality coefficient, the greater the difference in the data fluctuation coefficient, the larger the data fluctuation coefficient, the smaller the initial average physiological data, the larger the initial physiological data at each follow-up moment relative to the overall initial average physiological data, the more fluctuations there are, the greater the possibility of abnormality, and a positive correlation is shown.

[0068] In one embodiment of the present invention, the formula for abnormal probability is expressed as:

[0069] ;

[0070] ;

[0071] in, Indicates the The probability of abnormality at each follow-up time; Indicates the The data fluctuation coefficient at each follow-up time; It represents the mean of the data fluctuation coefficients at all follow-up moments, that is, the average data fluctuation coefficient; Indicates the The data fluctuation coefficient at each follow-up time; Indicates the Initial physiological data at the time of follow-up; represents the mean of the initial physiological data at all follow-up moments, that is, the initial average physiological data; represents the logistic function, Indicates taking the absolute value.

[0072] In the formula for anomaly probability, Indicates calculation of The difference between the initial physiological data at the time of follow-up and the initial average physiological data, that is, the data fluctuation coefficient. The larger the data fluctuation coefficient, the greater the difference between the physiological data at the time of follow-up and the initial average physiological data; Add 0.01 to avoid the denominator of the formula being 0, which would make the formula meaningless. Indicates the calculation of the coefficient difference between the data fluctuation coefficient at each follow-up moment and the average data fluctuation coefficient, and the calculation of the coefficient difference and The ratio of the coefficient difference to the average data fluctuation coefficient is used to obtain the data anomaly coefficient. The larger the data anomaly coefficient, the larger the data fluctuation coefficient is relative to the average data fluctuation coefficient, the more likely it is that anomalies exist, and the greater the possibility of anomalies. It represents the difference in the data fluctuation coefficient between each follow-up moment and the next follow-up moment. The larger the difference, the more inconsistent the data changes between adjacent moments, the greater the data fluctuation, and the greater the possibility of abnormality.

[0073] The possibility of abnormality reflects the fluctuation of the patient's physiological data. The greater the possibility of abnormality, the greater the data fluctuation, the more likely it is to produce a large deviation, and the more likely the data will be distorted, affecting the accuracy of the patient's physiological information. Therefore, abnormal follow-up moments are screened out for analysis.

[0074] Preferably, in one embodiment of the present invention, the method for obtaining the abnormal follow-up time includes:

[0075] If the abnormal possibility of the follow-up time is greater than the preset abnormal threshold, the corresponding follow-up time will be regarded as the abnormal follow-up time.

[0076] It should be noted that, in one embodiment of the present invention, the preset abnormality threshold is 0.7; in other embodiments of the present invention, the preset abnormality threshold may be set according to specific circumstances, which is not limited or elaborated herein.

[0077] There is a certain correlation between the patient's physiological data at different times. The distribution characteristics of the initial physiological data within the neighborhood can reflect the patient's dynamic change trend. Correcting the physiological data based on the abnormal possibility can help reflect the patient's true condition or change trend. Based on the abnormal possibility of each abnormal follow-up moment and the distribution characteristics of the initial physiological data within the neighborhood, the corrected physiological data at each follow-up moment is obtained.

[0078] Preferably, in one embodiment of the present invention, the method for obtaining corrected physiological data includes:

[0079] For any dimension, for each abnormal follow-up moment, the data correction coefficient of each abnormal follow-up moment is obtained according to the distribution characteristics of the initial physiological data within the neighborhood; the mean of the initial physiological data of all follow-up moments within the neighborhood is obtained as the local data level;

[0080] Performing gain adjustment on the initial physiological data according to the data correction coefficient at each abnormal follow-up moment and the difference between the local data level and the initial physiological data to obtain the corrected physiological data at each abnormal follow-up moment;

[0081] In one embodiment of the present invention, the formula for correcting physiological data is expressed as:

[0082] ;

[0083] in, Indicates the Corrected physiological data at each abnormal follow-up moment; Indicates the Initial physiological data at the time of abnormal follow-up; Indicates the Data correction coefficient for each abnormal follow-up moment; Indicates the The mean of the initial physiological data of all follow-up moments within the neighborhood of the abnormal follow-up moment, that is, the local data level; Represents the normalization function.

[0084] In the formula for correcting physiological data, The function will Normalized to [-1, 1], Indicates calculation of The difference between the local data level at each abnormal follow-up moment and the initial physiological data, the larger the difference, the greater the local data level is than the initial physiological data, the more the initial physiological data is adjusted. Conversely, the smaller the difference, the more the initial physiological data is adjusted. The larger the data correction coefficient, the greater the adjustment range, and the smaller the data correction coefficient, the smaller the adjustment range. Indicates calculation of The product of the data correction coefficient and the normalized difference at each abnormal follow-up moment is used as the weight to adjust the gain of the initial physiological data, that is, ,The larger the data correction coefficient is, the larger the difference between the local data level and the initial physiological data is,the larger the initial physiological data is, the larger the corrected physiological data is.

[0085] It should be noted that, in one embodiment of the present invention, the neighborhood range of each abnormal follow-up moment is the range formed by each abnormal follow-up moment and all historical follow-up moments; in other embodiments of the present invention, the neighborhood range can be specifically set according to specific circumstances and is not limited or detailed here. For follow-up moments that do not correspond to abnormal follow-up moments, the corresponding initial physiological data is used as the revised physiological data.

[0086] Preferably, in one embodiment of the present invention, the method for obtaining the data correction coefficient is as follows: Figure 3 , which shows a flow chart of a method for obtaining a data correction coefficient, including:

[0087] Step S301: For any dimension, obtain the mean of all initial physiological data of the patient during the hospitalization period as reference physiological data.

[0088] Considering that the physiological data of each dimension during the hospitalization period were measured by professionals, the patient's condition was relatively obvious, and the data had a strong reference value, the level of physiological data during the entire hospitalization period was quantified by taking the mean value. The data was monitored once a day during the hospitalization period.

[0089] Step S302: Calculate the difference between the initial physiological data and the reference physiological data at each other follow-up time within the neighborhood as the first difference; obtain the cumulative value of the first differences corresponding to all other follow-up times within the neighborhood as the overall difference value.

[0090] The first difference reflects the degree of deviation between the data at each follow-up moment and the reference physiological data. By analyzing the cumulative value of the first difference corresponding to all other follow-up moments in the neighborhood, the overall change trend of the physiological data in the neighborhood is reflected. The larger the cumulative value, the greater the volatility of the data, and the more subsequent adjustments are needed.

[0091] Step S303: Obtain the product of the abnormal probability and the overall difference value at each abnormal follow-up moment, and normalize it to serve as the data correction coefficient at each abnormal follow-up moment.

[0092] In one embodiment of the present invention, the formula for the data correction coefficient is expressed as:

[0093] ;

[0094] in, Indicates the Data correction coefficient for each abnormal follow-up moment; Indicates the The probability of abnormality correction at each follow-up time; Indicates the The number of follow-up moments within the neighborhood of an abnormal follow-up moment; Indicates the The first abnormal follow-up time in the neighborhood Physiological data at other follow-up times; It represents the mean value of all initial physiological data of the patient during the hospitalization period, that is, the reference physiological data; Represents the normalization function.

[0095] In the formula of the data correction coefficient, Indicates the neighborhood The difference between the initial physiological data and the reference physiological data at other follow-up moments, namely the first difference; It means calculating the mean of the first differences corresponding to all other follow-up moments within the neighborhood, that is, the overall difference value. The larger the overall difference value, the more the initial physiological data deviates from the reference physiological data, the greater the degree of fluctuation, and the more correction is needed; the greater the possibility of abnormality, the greater the fluctuation of the initial physiological data, the more adjustment is needed, and the larger the data correction coefficient.

[0096] It should be noted that in other embodiments of the present invention, a positive correlation can be constructed by adding the overall difference value and the abnormality possibility, where the larger the overall difference value, the greater the abnormality possibility, and the larger the data correction coefficient. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0097] Step S3: According to the number of abnormal follow-up moments in the neighborhood of each follow-up moment in different dimensions for each patient, and the changing trend of the corrected physiological data, the follow-up effectiveness of each patient at each follow-up moment is obtained; according to the distribution of follow-up effectiveness of each patient at different follow-up moments, the degree of attention paid to the condition of each patient at the real-time follow-up moment is obtained; according to the follow-up effectiveness and degree of attention paid to the condition of each patient at the real-time follow-up moment, all patients are clustered to obtain patient clusters.

[0098] The abnormal follow-up moment corresponds to the moment when the data becomes abnormal. The more times the abnormality occurs, the more difficult it is to analyze the patient's physiological state, and the worse the effectiveness of the data follow-up. The corrected physiological data is the patient's physiological data after adjustment, and observing the changing trend of the corrected physiological data more accurately reflects the patient's true physiological condition. Therefore, according to the number of abnormal follow-up moments in the neighborhood range of each follow-up moment in different dimensions of each patient, and the changing trend of the corrected physiological data, the follow-up effectiveness of each patient at each follow-up moment is obtained.

[0099] Preferably, in one embodiment of the present invention, the method for obtaining the effectiveness of follow-up is as follows: Figure 4 , which shows a flow chart of a method for obtaining the effectiveness of follow-up, including:

[0100] Step S401: For each dimension of each patient, the changing trend of the corrected physiological data is analyzed based on the abnormality probability acquisition method to obtain the abnormality correction probability at each follow-up moment.

[0101] It should be noted that the possibility of abnormality is obtained by replacing the initial physiological data with the corrected physiological data, and the possibility of abnormal correction is obtained; the changing trend of the corrected physiological data reflects the dynamic changes in the patient's physiological condition. By analyzing the possibility of abnormality of the corrected physiological data, the degree of fluctuation of the corrected physiological data can be reflected, and the authenticity and reliability of the abnormal data can be evaluated.

[0102] Step S402: Obtain the ratio between the number of all follow-up moments and the number of abnormal follow-up moments within the neighborhood of each follow-up moment as a local effective coefficient.

[0103] By calculating the ratio of the number of all follow-up moments in the neighborhood to the number of abnormal follow-up moments, the stability and reliability of the data within the range can be evaluated. The larger the ratio, the more stable the data in the neighborhood and the fewer abnormal values, which reflects better follow-up quality and data validity.

[0104] It should be noted that, in one embodiment of the present invention, the neighborhood range of each follow-up moment is a range based on each follow-up moment and all historical follow-up moments; in other embodiments of the present invention, the neighborhood range can be set according to specific circumstances, and is not limited or elaborated here.

[0105] Step S403: Obtain the follow-up effectiveness of each patient at each follow-up moment based on the abnormality correction possibility of each patient at each follow-up moment in different dimensions and the local effectiveness coefficient. The abnormality correction possibility is negatively correlated with the follow-up effectiveness, and the local effectiveness coefficient is positively correlated with the follow-up effectiveness.

[0106] Among them, the greater the possibility of abnormal correction, that is, the greater the degree of abnormal data performance, the lower the follow-up effectiveness; the larger the local effectiveness coefficient, the fewer the number of abnormal follow-up moments, and the greater the follow-up effectiveness, which is positively correlated.

[0107] In one embodiment of the present invention, for any patient, the formula for the effectiveness of follow-up is expressed as:

[0108] ;

[0109] in, Indicates the The effectiveness of follow-up at the time of follow-up; Indicates the The number of follow-up moments within the neighborhood of the follow-up moment; Indicates the Dimension The number of abnormal opponent moments within the neighborhood of the follow-up moment; Indicates the Dimension the likelihood of abnormality correction at the time of follow-up; Indicates the number of dimensions.

[0110] In the formula for follow-up effectiveness, It represents the ratio between the number of all follow-up moments and the number of abnormal follow-up moments within the neighborhood of each follow-up moment, that is, the local effectiveness coefficient. The larger the local effectiveness coefficient, the more the number of all follow-up moments, the fewer the number of abnormal follow-up moments, the greater the validity of the data, and the greater the effectiveness of the follow-up; the greater the possibility of abnormal correction, the greater the possibility of abnormal fluctuations in physiological data, the smaller the validity of the data, and the smaller the effectiveness of the follow-up.

[0111] The effectiveness of follow-up is used to evaluate the authenticity of the condition expressed by physiological data in each dimension. The higher the effectiveness of follow-up, the more accurate the physiological data. By analyzing the distribution of the effectiveness of follow-up at different follow-up times, we can understand the stability and variability of the patient's condition. Based on the distribution of the effectiveness of follow-up for each patient at different follow-up times, we can obtain the degree of attention paid to the condition of each patient at the real-time follow-up time.

[0112] Preferably, in one embodiment of the present invention, the method for obtaining the disease concern level can be found in Figure 5 , which shows a flow chart of a method for obtaining the degree of disease concern, including:

[0113] Step S501: Obtain the mean of the follow-up effectiveness of all patients at the real-time follow-up time as the average follow-up effectiveness.

[0114] By calculating the mean follow-up effectiveness of all patients at the real-time follow-up moment, we can obtain an overall follow-up effectiveness benchmark, which can reflect the follow-up behavior characteristics of most patients and provide a comparison standard for subsequent analysis.

[0115] Step S502: For each patient, a ratio of the follow-up effectiveness between the earliest follow-up time and the real-time follow-up time is obtained as the relative attention level of the patient at the real-time follow-up time.

[0116] By analyzing the ratio of the follow-up effectiveness between the earliest follow-up time and the real-time follow-up time, it is helpful to understand the continuity of patients' follow-up behavior and measure the changing trend of patients' attention level during the follow-up process. The smaller the ratio, the greater the follow-up effectiveness at the real-time follow-up time and the lower the relative attention level.

[0117] Step S503: Calculate the difference between the patient's follow-up effectiveness at the real-time follow-up time and the average follow-up effectiveness as the degree difference; perform negative correlation mapping on the patient's degree difference at the real-time follow-up time, and calculate the product of the negative correlation mapping result and the relative attention degree as the patient's condition attention degree at the real-time follow-up time.

[0118] The degree difference reflects the degree of deviation between the patient's real-time follow-up time and the overall situation. The greater the deviation, the greater the follow-up effectiveness at the real-time follow-up time and the less attention the patient pays to the disease; the smaller the relative attention, the less attention the patient pays to the disease.

[0119] In one embodiment of the present invention, the formula for the degree of concern of the disease is expressed as:

[0120] ;

[0121] in, Indicates the Patients at the real-time follow-up time The degree of concern about the condition; Indicates the The effectiveness of follow-up for each patient at the earliest follow-up time; express Patients at the real-time follow-up time The effectiveness of follow-up; Indicates that all patients are at the time of real-time follow-up The mean of the follow-up effectiveness is taken as the average follow-up effectiveness.

[0122] In the formula for the degree of concern of the disease, It indicates the ratio of the follow-up effectiveness between the earliest follow-up time and the real-time follow-up time, that is, the relative attention level of the patient at the real-time follow-up time. The larger the ratio, the lower the follow-up effectiveness at the real-time follow-up time, and the greater the possibility that the real-time follow-up time needs attention. It means calculating the difference between the effectiveness of follow-up of the patient at the real-time follow-up moment and the average effectiveness of follow-up, that is, the degree difference. The larger the degree difference, the greater the effectiveness of follow-up at the real-time follow-up moment, and the less attention is needed. Adding 0.01 to the formula avoids the denominator of the formula being 0, which makes the formula meaningless.

[0123] Through clustering, patients with similar follow-up effectiveness and disease attention levels are divided into one category, and patients with different characteristics are analyzed in a targeted manner. All patients are clustered according to the follow-up effectiveness and disease attention level of each patient at the real-time follow-up moment to obtain patient clusters.

[0124] It should be noted that, in one embodiment of the present invention, for each patient, the follow-up effectiveness is used as the horizontal axis and the disease attention level is used as the vertical axis to construct sample points in a two-dimensional sample space, and the K-means algorithm is used for clustering, wherein the K value is obtained as follows: ,in, Indicates the maximum level of concern about the condition among all patients. Indicates the minimum value of the degree of concern for the condition among all patients. Indicates the greatest difference in the degree of concern for the condition among all patients, represents the mean of the Euclidean distances between sample points in the sample space, Indicates the round-up symbol.

[0125] Step S4: According to the disease attention level of all patients in each patient cluster at the time of real-time follow-up, the follow-up frequency increase coefficient of each patient cluster is obtained; according to the follow-up frequency increase coefficient of the patient cluster to which each patient belongs and the initial follow-up number, the adjusted follow-up number of each patient is obtained.

[0126] There are significant differences in the degree of patients' concern about their condition in different patient clusters. By analyzing the degree of patients' concern about their condition, the follow-up of patients is adjusted. The stronger the degree of concern about the condition, the more frequent follow-up is needed to understand the changes in the patient's condition in a timely manner and improve the effect of follow-up. Therefore, based on the degree of concern of all patients in each patient cluster at the real-time follow-up moment, the follow-up frequency increase coefficient of each patient cluster is obtained.

[0127] Preferably, in one embodiment of the present invention, the method for obtaining the follow-up frequency increase coefficient includes:

[0128] The mean of the disease attention levels of all patients in each patient cluster at the time of real-time follow-up is obtained as the local disease attention level; the mean of the disease attention levels of all patients at the time of real-time follow-up is obtained as the overall disease attention level;

[0129] According to the number of patients in each patient cluster and the ratio of the degree of local attention to the disease and the degree of overall attention to the disease, the follow-up frequency increase coefficient of each patient cluster was obtained. The degree ratio was positively correlated with the follow-up frequency increase coefficient, and the number of patients was negatively correlated with the follow-up frequency increase coefficient.

[0130] In one embodiment of the present invention, the formula for the follow-up frequency increase coefficient is expressed as:

[0131] ;

[0132] in, represents the increase coefficient of follow-up frequency of the jth patient cluster; represents the mean of the disease attention level of all patients in the jth patient cluster at the time of real-time follow-up, that is, the local disease attention level; It represents the mean of the degree of concern about the condition of all patients at the time of real-time follow-up, that is, the overall degree of concern about the condition; represents the number of patients in the patient cluster; Represents the normalization function.

[0133] In the formula for the increase coefficient of follow-up frequency, It means calculating the ratio of the local attention level to the disease condition in the j-th patient cluster to the overall attention level, that is, the degree ratio. The larger the degree ratio, the greater the local attention level to the disease condition in the j-th patient cluster, and the more it is necessary to increase the follow-up frequency for attention; the smaller the number of patients in the patient cluster, the greater the difference in the patient's follow-up situation from other patients, the more attention is needed, and the greater the coefficient of increasing the follow-up frequency.

[0134] It should be noted that, in one embodiment of the present invention, by To construct a correlation relationship in which the larger the degree ratio, the smaller the number of patients, and the larger the increase coefficient of follow-up frequency, in other embodiments of the present invention, the addition method can also be used, that is, To construct a correlation relationship in which the larger the degree ratio, the smaller the number of patients, and the greater the increase coefficient of follow-up frequency, the specific means are technical means well known to those skilled in the art and will not be described here.

[0135] The follow-up frequency increase coefficient can reflect the changes in the patient's specific follow-up needs. By adjusting the number of follow-up visits, the patient's specific follow-up needs can be more accurately met, which helps to understand the changes in each patient's condition. Therefore, the adjusted follow-up number of each patient is obtained based on the follow-up frequency increase coefficient and the initial follow-up number of the patient cluster to which each patient belongs.

[0136] Preferably, in one embodiment of the present invention, the method for adjusting the number of follow-up visits includes:

[0137] The sum of the positive integer 1 and the follow-up frequency increase coefficient of each patient's cluster is calculated as the frequency adjustment weight. The product of the initial follow-up number and the frequency adjustment weight is calculated and rounded down to obtain the adjusted follow-up number.

[0138] In one embodiment of the present invention, the formula for adjusting the number of follow-up visits is expressed as:

[0139]

[0140] Where, Indicates the Adjusted number of follow-up visits per patient; Indicates the The initial follow-up number of patients; Indicates the The increase coefficient of follow-up frequency of each patient cluster; Indicates the floor symbol.

[0141] In the formula for adjusting the number of follow-up visits, It means calculating the sum of the positive integer 1 and the follow-up frequency increase coefficient of each patient's patient cluster as the frequency adjustment weight. The larger the follow-up frequency increase coefficient, the more attention is needed. The larger the frequency adjustment weight, the larger the initial follow-up frequency is adjusted, and the larger the adjusted follow-up frequency is.

[0142] It should be noted that, in one embodiment of the present invention, the initial number of follow-up visits is the number of follow-up visits set by the implementer in the month corresponding to the real-time follow-up time.

[0143] After obtaining the adjusted number of follow-up visits, the patient follow-up situation can be managed, including: selecting the patient with the largest number of adjusted follow-up visits and sending them a follow-up notice; for patients with an adjusted number of follow-up visits of 0, removing them from the follow-up list; and then optimizing the patient's follow-up plan, regularly updating the follow-up list, and improving the efficiency of patient follow-up management.

[0144] In summary, the present invention analyzes the changing trend of the initial physiological data at different follow-up moments for any dimension, obtains the abnormal possibility of each follow-up moment, and filters out the abnormal follow-up moments; according to the abnormal possibility of each abnormal follow-up moment and the distribution of initial physiological data within the neighborhood range, obtains the corrected physiological data at each follow-up moment; according to the number of abnormal follow-up moments within the neighborhood range of each follow-up moment on different dimensions for each patient and the changing trend of the corrected physiological data, obtains the follow-up effectiveness of each patient at each follow-up moment; obtains the degree of concern about the condition of each patient at the real-time follow-up moment and clusters all patients to obtain patient clustering clusters; according to the degree of concern about the condition of all patients in each patient cluster at the real-time follow-up moment, obtains the follow-up frequency increase coefficient of each patient cluster; according to the follow-up frequency increase coefficient and the initial follow-up number of the patient clustering cluster to which each patient belongs, obtains the adjusted follow-up number of each patient. The present invention determines the accurate follow-up number by analyzing the patient's true condition manifestations, thereby improving the effect of patient follow-up nursing management.

[0145] The present invention also proposes a patient follow-up information management system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any one of the steps of a patient follow-up information management method.

[0146] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0147] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for intelligent management of patient follow-up nursing information, characterized in that: The method comprises: Obtain the patient's multi-dimensional initial physiological data at each follow-up moment; For any dimension of each patient, the abnormal probability of each follow-up moment is obtained based on the changing trend of the initial physiological data at different follow-up moments, and abnormal follow-up moments are screened out; based on the abnormal probability of each abnormal follow-up moment and the distribution characteristics of the initial physiological data in the neighborhood, the corrected physiological data at each follow-up moment is obtained; Based on the number of abnormal follow-up moments in the neighborhood of each follow-up moment in different dimensions for each patient and the changing trend of the corrected physiological data, the follow-up effectiveness of each patient at each follow-up moment is obtained; based on the distribution of follow-up effectiveness of each patient at different follow-up moments, the degree of attention paid to the condition of each patient at the real-time follow-up moment is obtained; based on the follow-up effectiveness and degree of attention paid to the condition of each patient at the real-time follow-up moment, all patients are clustered to obtain patient clusters; According to the disease attention level of all patients in each patient cluster at the time of real-time follow-up, the follow-up frequency increase coefficient of each patient cluster is obtained; according to the follow-up frequency increase coefficient of each patient cluster and the initial follow-up number, the adjusted follow-up number of each patient is obtained; The method for obtaining the corrected physiological data includes: For any dimension, for each abnormal follow-up moment, the data correction coefficient of each abnormal follow-up moment is obtained according to the distribution characteristics of the initial physiological data within the neighborhood; the mean of the initial physiological data of all follow-up moments within the neighborhood is obtained as the local data level; Performing gain adjustment on the initial physiological data according to the data correction coefficient at each abnormal follow-up moment and the difference between the local data level and the initial physiological data to obtain the corrected physiological data at each abnormal follow-up moment; For follow-up moments that do not correspond to abnormal follow-up moments, the corresponding initial physiological data will be used as the revised physiological data; The method for obtaining the data correction coefficient includes: For any dimension, the mean of all initial physiological data of the patient during the hospitalization period was obtained as the reference physiological data; Calculate the difference between the initial physiological data and the reference physiological data at each other follow-up time within the neighborhood as the first difference; obtain the cumulative value of the first differences corresponding to all other follow-up times within the neighborhood as the overall difference value; The product of the abnormal probability and the overall difference value at each abnormal follow-up moment was obtained and normalized to serve as the data correction coefficient at each abnormal follow-up moment; The method for obtaining the effectiveness of the follow-up includes: For each dimension of each patient, the changing trend of the corrected physiological data is analyzed based on the abnormality probability acquisition method to obtain the abnormality correction probability at each follow-up moment; The ratio between the number of all follow-up moments and the number of abnormal follow-up moments in the neighborhood of each follow-up moment is obtained as the local effective coefficient; The follow-up effectiveness of each patient at each follow-up moment was obtained based on the abnormality correction possibility and local effectiveness coefficient of each patient in different dimensions at each follow-up moment. The abnormality correction possibility was negatively correlated with the follow-up effectiveness, while the local effectiveness coefficient was positively correlated with the follow-up effectiveness. The method for obtaining the disease condition attention level includes: The mean follow-up effectiveness of all patients at the real-time follow-up time was obtained as the average follow-up effectiveness; For each patient, the ratio of the follow-up effectiveness between the earliest follow-up time and the real-time follow-up time was obtained as the relative attention level of the patient at the real-time follow-up time; The difference between the patient's follow-up effectiveness at the real-time follow-up moment and the average follow-up effectiveness was calculated as the degree difference. A negative correlation mapping was performed on the patient's degree difference at the real-time follow-up moment, and the product of the negative correlation mapping result and the relative attention degree was calculated as the patient's disease attention degree at the real-time follow-up moment.

2. A patient follow-up nursing information intelligent management method according to claim 1, characterized in that: The method for obtaining the abnormal possibility includes: For any dimension, if there is no initial physiological data at the follow-up time, the mean of the initial physiological data between the follow-up times within the historical range is obtained as the initial physiological data at the corresponding follow-up time; The mean of the initial physiological data at all follow-up moments was obtained as the initial average physiological data; the difference between the initial physiological data at each follow-up moment and the initial average physiological data was obtained as the data fluctuation coefficient at each follow-up moment; The mean of the data fluctuation coefficients at all follow-up moments was obtained as the average data fluctuation coefficient; the coefficient difference between the data fluctuation coefficient at each follow-up moment and the average data fluctuation coefficient was obtained, and the data abnormality coefficient at each follow-up moment was obtained based on the deviation of the coefficient difference from the average data fluctuation coefficient; The abnormality probability of each follow-up moment was obtained based on the data abnormality coefficient at each follow-up moment and the difference in data fluctuation coefficient between each follow-up moment and the next follow-up moment. The difference in data abnormality coefficient and data fluctuation coefficient was positively correlated with the abnormality probability.

3. The method for intelligent management of patient follow-up nursing information according to claim 1, characterized in that: The method for obtaining the abnormal follow-up time includes: If the abnormal possibility of the follow-up time is greater than the preset abnormal threshold, the corresponding follow-up time will be regarded as the abnormal follow-up time.

4. The method for intelligent management of patient follow-up nursing information according to claim 1, characterized in that: The method for obtaining the follow-up frequency increase coefficient includes: The mean of the disease attention levels of all patients in each patient cluster at the time of real-time follow-up is obtained as the local disease attention level; the mean of the disease attention levels of all patients at the time of real-time follow-up is obtained as the overall disease attention level; According to the number of patients in each patient cluster and the ratio of the degree of local attention to the disease and the degree of overall attention to the disease, the follow-up frequency increase coefficient of each patient cluster was obtained. The degree ratio was positively correlated with the follow-up frequency increase coefficient, and the number of patients was negatively correlated with the follow-up frequency increase coefficient.

5. The method for intelligent management of patient follow-up nursing information according to claim 1, characterized in that: The method for obtaining the adjusted number of follow-up visits includes: The sum of the positive integer 1 and the follow-up frequency increase coefficient of each patient's cluster is calculated as the frequency adjustment weight. The product of the initial follow-up number and the frequency adjustment weight is calculated and rounded down to obtain the adjusted follow-up number.

6. An intelligent management system for patient follow-up nursing information, the 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 patient follow-up nursing information intelligent management method according to any one of claims 1 to 5 are implemented.

Images