System for stratification of invasive fungal infection secondary to sepsis based on test data

Abstract

This invention relates to the field of biomedical data processing technology, specifically to a stratification system for sepsis-related secondary invasive fungal infections based on laboratory data. The invention obtains the degree of static intestinal baseline damage by considering serum lactate concentrations at different historical testing times, temporal characteristics, and a preset maximum backtracking period; it obtains the cumulative dosage of antimicrobial drugs and the hourly cumulative distribution value of drug use based on the drug dosage distribution at different medication times within a neighborhood of the real-time trigger time; it obtains the overall degree of inflammation reduction and the hourly cumulative distribution value of inflammation regression based on the procalcitonin concentration distribution at testing times within a neighborhood; and it further obtains the net drug-induced intervention level. Based on the degree of static intestinal baseline damage and the net drug-induced intervention level, the fungal infection characteristics are stratified. This invention, by considering non-drug-induced interference, accurately obtains the comprehensive risk level of fungal infection, improving the effectiveness of fungal infection stratification.

Description

Technical Field

[0001] This invention relates to the field of biomedical data processing technology, specifically to a stratification system for sepsis secondary invasive fungal infections based on test data. Background Technology

[0002] Sepsis is a systemic inflammatory response syndrome caused by severe infection. In routine anti-infective treatment in the intensive care unit, patients generally need to receive intravenous infusions of high-dose, broad-spectrum antibiotics, which kill pathogenic bacteria while also destroying a large amount of the normal symbiotic flora in the patient's intestines. The reduction of the normal symbiotic flora in the intestines allows previously suppressed intestinal colonizing fungi to gain a living space and translocate, thereby entering the bloodstream through the damaged intestinal mucosal barrier and causing secondary invasive fungal infections with extremely high mortality rates. Therefore, it is necessary to analyze the characteristics of fungal infections.

[0003] In existing technologies, fungal infection analysis is performed by comparing absolute laboratory values ​​or directly linking them to prescription instructions. However, this simple numerical correlation of data does not take into account the non-drug-induced interventions caused by the underlying dialysis equipment. This leads to the erroneous equating of the decrease in inflammatory indicators caused by dialysis with the effective macroscopic bactericidal effect of antibacterial drugs, resulting in misdiagnosis of fungal infection and poor stratification. Summary of the Invention

[0004] To address the technical problem of misdiagnosis and poor stratification results caused by simply performing numerical correlations on data without considering non-pharmacological interventions from underlying dialysis equipment, this invention aims to provide a stratification system for sepsis-related secondary invasive fungal infections based on test data. The specific technical solution adopted is as follows: This invention proposes a stratification system for secondary invasive fungal infections in sepsis based on test data, the system comprising: The data acquisition module is used to acquire the serum lactate concentration value at the time of serum lactate detection, the drug dosage at the time of medication, and the procalcitonin concentration at the time of testing within the historical range of the real-time trigger time. The data processing module is used to obtain the degree of static intestinal baseline damage based on the serum lactate concentration values, time characteristics, and preset maximum backtracking days at different serum lactate detection times within a historical range; to obtain the cumulative dosage of antimicrobial drugs and the cumulative distribution value of medication per hour based on the drug dosage distribution at different medication times within the neighborhood range of the real-time trigger time; and to obtain the overall degree of inflammation reduction and the cumulative distribution value of inflammation reduction per hour based on the procalcitonin test concentration distribution at the test time within the neighborhood range. The fungal infection quantification module is used to obtain the drug efficacy response time-series offset based on the differences in the cumulative distribution values ​​of drug use and the cumulative distribution values ​​of remission within different hours in the neighborhood; to obtain the net drug-induced intervention degree based on the cumulative dose of antimicrobial drugs, the overall degree of inflammation reduction, and the drug efficacy response time-series offset; and to obtain the comprehensive risk of fungal infection based on the degree of damage to the static intestinal baseline and the net drug-induced intervention degree. The fungal infection stratification module is used to stratify fungal infection characteristics based on the overall risk distribution of fungal infections.

[0005] Furthermore, the method for obtaining the static baseline degree of intestinal damage includes: If there are serum lactate detection times within the historical range, the serum lactate concentration value corresponding to the adjacent serum lactate detection time of the real-time trigger time is selected as the historical concentration of the intestinal marker; if there are no serum lactate detection times within the historical range, a preset serum lactate reference value is obtained as the historical concentration of the intestinal marker; and a preset serum lactate time is obtained as the time corresponding to the historical concentration of the intestinal marker. Obtain the time difference between the real-time trigger time and the corresponding time of the historical concentration of the intestinal marker. Calculate the time difference divided by the number of hours in the previous day as the difference in days. If the difference in days is greater than the preset maximum effective backtracking days, the preset maximum effective backtracking days are used as the test data lag days. If the difference in days is less than or equal to the preset maximum effective backtracking days, the difference in days are used as the test data lag days. If the historical concentration of the intestinal marker is greater than the preset serum lactate reference value, the degree of static intestinal baseline damage is obtained based on the preset lag compensation coefficient, the number of days of test data lag, and the concentration difference between the historical concentration of the intestinal marker and the preset serum lactate reference value. The preset lag compensation coefficient, the number of days of test data lag, and the concentration difference are all positively correlated with the degree of static intestinal baseline damage. If the historical concentration of the intestinal marker is less than or equal to the preset serum lactate reference value, the historical concentration of the intestinal marker is used as the degree of static intestinal baseline damage.

[0006] Furthermore, the method for obtaining the cumulative dosage of the antibacterial drug includes: The drug dosage at the time of administration is normalized and used as the single-dose equivalent dose at the time of administration. The cumulative value of the single-dose equivalent dose at all times within the neighborhood is obtained as the cumulative dose of the antibacterial drug.

[0007] Furthermore, the method for obtaining the cumulative distribution value of medication use includes: The difference between each medication time and the starting time within the neighborhood is obtained as the relative medication time; The ratio of the single dose at the relative time of medication administration to the sum of the cumulative dose of antimicrobial drugs and the preset adjustment coefficient is used as the percentage of a single dose at the relative time of medication administration. The cumulative value of the percentage of single-dose administration at any given time within the neighborhood is obtained as the cumulative distribution value of medication administration per hour.

[0008] Furthermore, the method for obtaining the overall degree of inflammation reduction includes: The maximum value of the procalcitonin test concentration at all test times within the neighborhood is obtained as the highest test concentration; the procalcitonin test concentration at the test time adjacent to the real-time trigger time is obtained as the last test concentration. Obtain the concentration difference between the highest test concentration and the last test concentration. If the concentration difference is greater than the preset difference threshold, the concentration difference is used as the overall fading concentration; if the concentration difference is less than or equal to the preset difference threshold, the overall fading concentration is set to 0. The overall reduction in inflammation is calculated as the sum of the overall remission concentration and the highest test concentration plus the preset adjustment coefficient.

[0009] Furthermore, the method for obtaining the fading cumulative distribution value includes: The difference between each test time and the starting time in the neighborhood is obtained as the relative fading time corresponding to each test time. If the relative time of decline is greater than the time corresponding to the highest test concentration, and the difference between the previous relative time of decline and the corresponding relative time of decline and the corresponding procalcitonin test concentration is greater than 0, the difference is calculated as the sum of the overall decline concentration and the preset adjustment coefficient, and is used as the single decline percentage of the corresponding relative time of decline; otherwise, the single decline percentage is set to 0. Obtain the cumulative value of the percentage of single regressions at all relative times within the neighborhood per hour, and use it as the cumulative regression distribution value for each hour.

[0010] Furthermore, the method for obtaining the drug efficacy response time offset includes: Obtain the absolute value of the difference between the cumulative distribution value of drug use and the cumulative distribution value of drug extinction over all hours within the neighborhood, and sum them up as the drug efficacy response time offset.

[0011] Furthermore, the method for obtaining the net drug-induced intervention level includes: The product of the cumulative dose of antimicrobial drugs and the overall decrease in inflammation is obtained as the theoretical efficacy response intensity; the theoretical efficacy response intensity is calculated as the sum of the efficacy response time offset and the preset adjustment coefficient, which is the net drug-induced intervention level.

[0012] Furthermore, the method for obtaining the overall risk level of fungal infection includes: The product of the degree of damage to the static intestinal baseline and the degree of net drug-induced intervention was obtained as the comprehensive risk of fungal infection.

[0013] Furthermore, the stratification of fungal infection characteristics includes: If the overall risk of fungal infection is less than or equal to the preset low-risk threshold for fungal infection, the fungal infection characteristics are determined to be at the first level. If the overall risk of fungal infection is greater than the preset low-risk threshold for fungal infection, but less than or equal to the preset high-risk threshold for fungal infection, the fungal infection characteristic is determined to be at the second level. If the overall risk of fungal infection is greater than the preset high-risk threshold for fungal infection, the fungal infection characteristics are judged to be at level three.

[0014] The present invention has the following beneficial effects: This invention obtains the static baseline degree of intestinal damage based on serum lactate concentration values, temporal characteristics, and a preset maximum backtracking number at different serum lactate detection times within a historical range. After time linear compensation, it quantifies the baseline data for intestinal damage assessment. Based on the drug dose distribution at different medication times within the neighborhood of the real-time trigger time, it obtains the cumulative dosage of antimicrobial drugs and the hourly cumulative distribution value of medication, reflecting the centroidal clustering characteristics of medication administration on the time axis. Based on the procalcitonin test concentration distribution at the neighboring testing times, it obtains the overall degree of inflammation reduction and the hourly cumulative distribution value of inflammation regression, reflecting the time axis of medication administration and the reduction of inflammatory markers. The invention utilizes the following methods: 1) The centroidal clustering characteristics of the drug; 2) The differences in the cumulative distribution values ​​of drug use and regression over different hours within a neighborhood are used to obtain the time-series offset of the drug response, reflecting the degree of temporal deviation between the inflammatory marker decline event and the drug use event; 3) The net drug-induced intervention degree is obtained based on the cumulative dosage of antibacterial drugs, the overall degree of inflammation reduction, and the time-series offset of the drug response, reflecting the dynamic characteristics of the degree to which normal intestinal symbiotic bacteria are eliminated purely by drug-induced sterilization; 4) The comprehensive risk of fungal infection is obtained based on the degree of damage to the static intestinal baseline and the net drug-induced intervention degree, quantifying the infection risk under conditions of intestinal damage and the elimination of a large number of symbiotic bacteria; 5) The fungal infection characteristics are stratified. This invention improves the effectiveness of fungal infection stratification by considering non-drug-induced interference and accurately obtaining the comprehensive risk of fungal infection. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a structural block diagram of a stratification system for secondary invasive fungal infections in sepsis based on test data, provided in one embodiment of the present invention. Figure 2 This is a flowchart illustrating a method for obtaining the degree of static intestinal baseline damage, as provided in one embodiment of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a stratification system for sepsis secondary invasive fungal infections based on test data proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] 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 this invention pertains.

[0019] The following description, in conjunction with the accompanying drawings, details a specific scheme for a stratification system for sepsis secondary invasive fungal infections based on test data, provided by the present invention.

[0020] Please see Figure 1 The diagram illustrates a structural block diagram of a stratification system for secondary invasive fungal infections in sepsis based on test data, according to an embodiment of the present invention. The system specifically includes: a data acquisition module 101, a data processing module 102, a fungal infection quantification module 103, and a fungal infection stratification module 104. The data acquisition module 101 is used to acquire the serum lactate concentration value at the time of serum lactate detection, the drug dosage at the time of medication, and the procalcitonin test concentration at the time of testing within the historical range of the real-time trigger time.

[0021] In the embodiments of the present invention, considering that life support equipment in intensive care units usually operates as independent lower-level hardware, and the power-on operation status logs of independent equipment do not establish underlying real-time data interoperability authorization with the existing laboratory information system or hospital information system of medical institutions, the temporal distribution characteristics of structured medical order execution records and test feedback data are extracted and analyzed.

[0022] First, when the preset trigger time of the timed trigger script is reached, the system retrieves historical test data from the laboratory information system, searches for serum lactate test records reflecting the patient's intestinal damage level, and procalcitonin test records; the system issues a medical order data reading instruction to the hospital information system to read all intravenous infusion records of the patient's antibiotics; thus, it obtains the serum lactate concentration value at the time of serum lactate testing, the drug dosage at the time of medication administration, and the procalcitonin concentration at the time of testing within the historical range of the real-time trigger time.

[0023] It should be noted that, in the embodiments of the present invention, the real-time trigger time is the current time obtained by the system upon reaching the trigger time preset by the timed trigger script; the historical range is the range formed by the real-time trigger time and previous historical times.

[0024] It should be noted that the unit of time can be set by the implementers according to the specific circumstances. In the embodiments of the present invention, the unit is hours.

[0025] The data processing module 102 is used to obtain the degree of static intestinal baseline damage based on the serum lactate concentration values, time characteristics, and preset maximum backtracking days at different serum lactate detection times within the historical range; to obtain the cumulative dosage of antibacterial drugs and the cumulative distribution value of medication per hour based on the drug dosage distribution at different medication times within the neighborhood range of the real-time trigger time; and to obtain the overall degree of inflammation reduction and the cumulative distribution value of inflammation reduction per hour based on the procalcitonin test concentration distribution at the test time within the neighborhood range.

[0026] Serum lactate is a direct marker of intestinal mucosal damage, and serum lactate concentration reflects the extent of intestinal mucosal damage. The preset maximum retrospective days provide an objective baseline of intestinal barrier damage risk that has been time-corrected. Based on the serum lactate concentration, time characteristics, and preset maximum retrospective days at different serum lactate detection times within a historical range, the static baseline degree of intestinal damage is obtained.

[0027] Preferably, in one embodiment of the present invention, the method for obtaining the static intestinal baseline degree of damage is described in [reference needed]. Figure 2 It illustrates a flowchart of a method for obtaining the baseline level of damage to the intestinal tract at rest, including: Step S201: If there is a serum lactate detection time within the historical range, select the serum lactate concentration value corresponding to the adjacent serum lactate detection time of the real-time trigger time as the historical concentration of the intestinal marker; if there is no serum lactate detection time within the historical range, obtain the preset serum lactate reference value as the historical concentration of the intestinal marker, and obtain the preset serum lactate time as the time corresponding to the historical concentration of the intestinal marker.

[0028] It should be noted that, in the embodiments of the present invention, the preset serum lactate reference value can be extracted from a preset physiological reference database as the upper limit of normal serum lactate concentration in healthy individuals, and set according to relevant historical experience. To complete the fault-tolerant filling of initial data missing, the preset serum lactate time is the patient's admission registration time, that is, the physical time starting point for establishing medical data tracking; in other embodiments of the present invention, the preset serum lactate reference value and preset serum lactate time can be set according to specific circumstances, and are not limited or elaborated here.

[0029] Step S202: Obtain the time difference between the real-time trigger time and the corresponding time of the historical concentration of the intestinal marker, calculate the time difference divided by the number of all hours in the previous day, and use this as the difference in days; if the difference in days is greater than the preset maximum effective backtracking days, use the preset maximum effective backtracking days as the test data lag days; if the difference in days is less than or equal to the preset maximum effective backtracking days, use the difference in days as the test data lag days.

[0030] Considering that medication is usually administered on a daily cycle, the time difference data, which is in the unit of hours, is converted into days for subsequent analysis. The time difference is divided by the total number of hours of the previous day, i.e., the time difference is divided by 24, and the resulting value is used as the difference in days.

[0031] It should be noted that, in one embodiment of the present invention, based on clinical knowledge, the recovery period for intestinal injury or bacterial translocation is approximately two weeks. Data beyond two weeks has reduced reference value, so the maximum effective retrospective period is preset to 14 days. In other embodiments of the present invention, the maximum effective retrospective period can be set according to specific circumstances, and is not limited or elaborated here.

[0032] Step S203: If the historical concentration of intestinal markers is greater than the preset serum lactate reference value, the degree of static intestinal baseline damage is obtained based on the preset lag compensation coefficient, the number of days of test data lag, and the concentration difference between the historical concentration of intestinal markers and the preset serum lactate reference value. The preset lag compensation coefficient, the number of days of test data lag, and the concentration difference are all positively correlated with the degree of static intestinal baseline damage. If the historical concentration of intestinal markers is less than or equal to the preset serum lactate reference value, the historical concentration of intestinal markers will be used as the static baseline degree of intestinal damage.

[0033] It should be noted that the larger the concentration difference, the greater the historical concentration of intestinal markers relative to the preset serum lactate reference value, the more likely they are to be damaged, and the greater the degree of damage to the static intestinal baseline. The larger the data lag period, the larger the preset lag compensation coefficient, and the longer the time since the last test, the higher the potential probability of the patient's intestinal barrier being continuously damaged or naturally deteriorating during this period, and the greater the degree of damage to the static intestinal baseline. Therefore, the preset lag compensation coefficient, the data lag period, and the concentration difference are all positively correlated with the degree of damage to the static intestinal baseline.

[0034] It should be noted that, in the embodiments of the present invention, the preset hysteresis compensation coefficient is used to quantify the daily average natural deterioration rate of intestinal damage over time. The method of obtaining the coefficient is as follows: collect the serum lactate concentration values ​​of critically ill patients in the first week after intestinal barrier damage, calculate the difference between serum lactate concentration values ​​between adjacent days, analyze the daily rate of change of serum lactate concentration, and extract the benchmark compensation hyperparameter by statistically fitting the rate of change using the least squares method; the coefficient is set to 0.1 based on relevant historical experience.

[0035] It should be noted that, in the embodiments of the present invention, if the historical concentration of intestinal markers is greater than the preset serum lactate reference value, the method for obtaining the degree of damage to the static intestinal baseline includes: obtaining the product of the preset lag compensation coefficient, the number of days of lag in the test data, and the concentration difference as the lag supplement concentration; and obtaining the sum of the lag supplement concentration and the historical concentration of intestinal markers as the degree of damage to the static intestinal baseline.

[0036] The formula is expressed as: ;in, Indicates the degree of damage to the static gut baseline; Indicates the historical concentration of gut biomarkers; This represents the concentration difference between the historical concentration of intestinal markers and the preset serum lactate reference value. This indicates the preset lag compensation coefficient; Indicates the number of days the test data is laging; This indicates the concentration replenished after a delay.

[0037] The dosage of drugs varies at different times of administration, and the weight of their impact on the patient's gut microbiota and condition is completely different. In order to describe the actual use of antimicrobial drugs on the time axis, the cumulative dosage of antimicrobial drugs and the cumulative distribution of drug dosage at each time of administration are obtained based on the drug dosage distribution at different times of administration within the neighborhood of the real-time trigger time.

[0038] Preferably, in one embodiment of the present invention, the method for obtaining the cumulative dosage of the antibacterial drug includes: The first step is to normalize the drug dosage at the time of administration, which will be used as the single-dose conversion at the time of administration. It should be noted that, in the embodiments of the present invention, based on the average daily maintenance dose benchmark for adults as defined by the World Health Organization for assessing the main indications of drugs, the standard defined daily dose corresponding to the drug is retrieved, and the ratio of the drug dose at the time of administration to the standard defined daily dose is obtained. That is, the drug dose at the time of administration is normalized as a single converted dose, which helps to eliminate the relative dose value of physical dimension differences.

[0039] The second step is to obtain the cumulative value of the single-dose conversion at all times of medication within the neighborhood, which is used as the cumulative dosage of the antibacterial drug.

[0040] It should be noted that, in the embodiments of the present invention, in order to provide a standardized input basis for the system to subsequently perform time alignment and time sequence overlap comparison, the analysis is performed by extracting a preset time window; the method for obtaining the neighborhood range is: based on the real-time trigger time, a range consisting of 72 historical hours is selected; in other embodiments of the present invention, the size of the neighborhood range can be set according to specific circumstances, and will not be limited or elaborated here.

[0041] Because conventional medical information systems lack a direct interface for reading the operating status of extracorporeal dialysis instruments, there is a natural misalignment in the input of clinical medication and laboratory tests on the absolute event axis. Therefore, it is necessary to quantify the cumulative distribution value of medication to reflect the cumulative dose density characteristics of medication actions on the time axis.

[0042] Preferably, in one embodiment of the present invention, the method for obtaining the cumulative drug distribution value includes: The difference between each medication time and the starting time within the neighborhood is obtained as the relative medication time; The ratio of the single dose at the relative time of medication administration to the sum of the cumulative dose of antimicrobial drugs and the preset adjustment coefficient is used as the percentage of a single dose at the relative time of medication administration. It should be noted that, in the embodiments of the present invention, when performing ratio calculations, in order to avoid the formula being meaningless when the cumulative dosage of antibacterial drugs is 0, the preset adjustment coefficient is to add a very small positive number with consistent dimensions to the denominator. Its value is specifically set according to the range of values ​​of the denominator, such as 0.01, which is not limited or elaborated here.

[0043] The cumulative value of the percentage of single-dose administration at any given time within the neighborhood is obtained as the cumulative distribution value of medication administration per hour.

[0044] Based on this, the cumulative distribution value of drug use characterizes the clustering skewness distribution of the frequency and dosage of antimicrobial drugs within the domain. The more clustered the skewness, the more concentrated the large dose of drug use is within the neighborhood.

[0045] Procalcitonin is an inflammatory protein. The concentration of procalcitonin reflects the temporal density and centroidal distribution of inflammatory marker regression events on the time axis, thus helping to quantify the overall degree of inflammation reduction and the cumulative distribution value of regression. Based on the distribution of procalcitonin concentration at each test time within the neighborhood, the overall degree of inflammation reduction and the hourly cumulative distribution value of regression can be obtained.

[0046] Preferably, in one embodiment of the present invention, the method for obtaining the overall degree of inflammation reduction includes: The first step is to obtain the maximum value of the procalcitonin test concentration among all test times within the neighborhood range, which is taken as the highest test concentration; and to obtain the procalcitonin test concentration of the adjacent test times at the real-time trigger time, which is taken as the last test concentration. The second step is to obtain the concentration difference between the highest test concentration and the last test concentration. If the concentration difference is greater than the preset difference threshold, the concentration difference is used as the overall fading concentration; if the concentration difference is less than or equal to the preset difference threshold, the overall fading concentration is set to 0. It should be noted that, in the embodiments of the present invention, the difference between the highest test concentration and the last test concentration is analyzed to characterize the decline of the procalcitonin test concentration during the process. The higher the highest test concentration is than the last test concentration, the larger the concentration difference and the greater the decline. The lower the highest test concentration is than or equal to the last test concentration, the smaller the concentration difference, which is negative or 0, indicating less decline. Therefore, the preset difference threshold is set to 0. In other embodiments of the present invention, the size of the preset difference threshold can be set according to specific circumstances, and is not limited or described here.

[0047] The third step is to obtain the overall reduction concentration ratio as the sum of the highest test concentration and the preset adjustment coefficient, which is used as the overall degree of inflammation reduction.

[0048] It should be noted that, in the embodiments of the present invention, when performing ratio calculations, in order to avoid the formula being meaningless when the highest test concentration is 0, the preset adjustment coefficient is to add a very small positive number with consistent dimensions to the denominator. Its value is specifically set according to the range of the denominator, such as 0.01, which is not limited or elaborated here.

[0049] Based on this, the higher the overall regression concentration, the greater the degree of inflammation relief; the higher the highest detection concentration, the smaller the overall regression concentration relative to the global change, and therefore the smaller the overall decrease in inflammation.

[0050] Preferably, in one embodiment of the present invention, the method for obtaining the fading cumulative distribution value includes: The first step is to obtain the difference between each test time and the starting time in the neighborhood, which is used as the relative fading time corresponding to each test time; The second step is to calculate the percentage of single-time decline at the corresponding relative decline time if the relative decline time is greater than the time corresponding to the highest test concentration, and the difference between the previous relative decline time and the corresponding relative decline time is greater than 0. If the difference is greater than 0, the difference is calculated as the sum of the overall decline concentration and the preset adjustment coefficient, and is used as the percentage of single-time decline at the corresponding relative decline time. Otherwise, the percentage of single-time decline is set to 0. It should be noted that before the highest test concentration, the concentration is constantly rising and has not yet begun to form a true macroscopic sterilization trend. Therefore, if the relative time of decline is less than or equal to the time corresponding to the highest test concentration, the percentage of single decline is set to 0. For the relative time of decline that is greater than the time corresponding to the highest test concentration, the larger the time, the smaller the concentration, that is, the difference is greater than 0, which shows the characteristics of decline. If the difference is less than or equal to 0, no decline characteristics are shown, and the percentage of single decline is set to 0.

[0051] The third step is to obtain the cumulative value of the percentage of single regressions at all relative times within the neighborhood per hour, which is used as the cumulative regression distribution value per hour.

[0052] Based on this, the fading cumulative distribution value reflects the actual density clustering skewness of the test index within its neighborhood.

[0053] The fungal infection quantification module 103 is used to obtain the drug efficacy response time shift based on the differences in the cumulative distribution values ​​of drug use and the cumulative distribution values ​​of remission within different hours in the neighborhood; to obtain the net drug-induced intervention degree based on the cumulative dose of antimicrobial drugs, the overall degree of inflammation reduction, and the drug efficacy response time shift; and to obtain the comprehensive risk of fungal infection based on the degree of damage to the static intestinal baseline and the net drug-induced intervention degree.

[0054] The cumulative distribution values ​​of drug use and the cumulative distribution values ​​of remission reflect the centroidal clustering characteristics of drug use and the decline of inflammatory markers over time. By analyzing the differences between the cumulative distribution values ​​of drug use and the cumulative distribution values ​​of remission, the degree of centroidal shift in the time distribution is quantified, reflecting the correlation between drugs and the decline in inflammation. Based on the differences between the cumulative distribution values ​​of drug use and the cumulative distribution values ​​of remission at different hours within the neighborhood, the temporal shift of the drug efficacy response is obtained.

[0055] Preferably, in one embodiment of the present invention, the method for obtaining the drug efficacy response time offset includes: Obtain the absolute value of the difference between the cumulative distribution value of drug use and the cumulative distribution value of drug extinction over all hours within the neighborhood, and sum them up as the drug efficacy response time offset.

[0056] It should be noted that obtaining the absolute value of the difference between the cumulative distribution value of medication and the cumulative distribution value of inflammation reduction within the neighborhood range for all hours and summing them up is equivalent to approximating the area of ​​the difference between medication and inflammation reduction within the neighborhood range. This quantifies the deviation between medication and reduction. The greater the deviation, the weaker the temporal causality between the inflammatory marker decrease event and the medication event, meaning that the decrease in procalcitonin is most likely caused by non-drug-induced physical clearance.

[0057] Because high-dose broad-spectrum antibiotics, while effectively killing pathogens (i.e., reducing overall inflammatory markers), inevitably kill normal symbiotic bacteria in the gut in a proportional manner, the net drug-induced intervention is greater. Furthermore, considering the potential for a false decrease in inflammation when interfered with by extracorporeal dialysis, a drug response time-series offset is used to reflect the time-series deviation between medication administration and the decrease in inflammation. This avoids misinterpreting a false decrease in inflammation as a bactericidal effect of the antibiotic, and more accurately quantifies the net drug-induced intervention purely caused by drug-induced bactericidal activity. The net drug-induced intervention is obtained based on the cumulative antibiotic dosage, the overall decrease in inflammation, and the drug response time-series offset.

[0058] Preferably, in one embodiment of the present invention, the method for obtaining the net drug-induced intervention level includes: The product of the cumulative dose of antimicrobial drugs and the overall decrease in inflammation is obtained as the theoretical efficacy response intensity; the theoretical efficacy response intensity is calculated as the sum of the efficacy response time offset and the preset adjustment coefficient, which is the net drug-induced intervention level.

[0059] The formula is expressed as: ;in, Indicates the degree of net drug-induced intervention; Indicates the cumulative dosage of antibacterial drugs; Indicates the overall degree of reduction in inflammation; Indicates the time shift of the drug effect response; This indicates the preset adjustment coefficient; This indicates the theoretical efficacy response intensity.

[0060] It should be noted that, in the embodiments of the present invention, when performing ratio calculations, in order to avoid the formula being meaningless when the drug effect response time offset is 0, the preset adjustment coefficient is to add a very small positive number with consistent dimensions to the denominator. Its value is specifically set according to the range of the denominator, such as 0.01, which is not limited or elaborated here.

[0061] It should be noted that if the cumulative dose of antibacterial drugs or the overall decrease in inflammation is equal to 0, it indicates that there is no macroscopic bactericidal effect dominated by drugs. That is, the patient did not receive sufficient antibacterial drugs, or the inflammatory markers did not show a substantial decrease, and the bactericidal effect of the drugs was poor. Therefore, the net drug-induced intervention caused purely by drug bactericidal effect is 0.

[0062] If both the cumulative dose of antimicrobial drugs and the overall decrease in inflammation are not zero, the higher the cumulative dose of antimicrobial drugs, the greater the overall decrease in inflammation, and the greater the bactericidal effect. Therefore, the product of the cumulative dose of antimicrobial drugs and the overall decrease in inflammation is calculated. When both are larger, the theoretical efficacy response intensity is greater, and therefore the net drug-induced intervention caused by drug bactericidal effect is greater. The temporal deviation of the efficacy response reflects the temporal causal deviation between the inflammatory marker decrease event and the medication event. The larger the deviation, the weaker the temporal causal relationship between the inflammatory marker decrease event and the medication event, and therefore the smaller the net drug-induced intervention caused by drug bactericidal effect.

[0063] The degree of damage to the static intestinal baseline reflects the quantitative baseline of the accumulated damage to the intestinal defenses over a historical period. The net degree of drug-induced intervention reflects the dynamic characteristics of the elimination of normal intestinal commensal flora caused by antibiotics in the surrounding area. A reduction in normal intestinal commensal flora allows previously suppressed intestinal colonizing fungi to gain space and translocate, subsequently crossing the damaged intestinal mucosal barrier into the bloodstream, leading to highly lethal secondary invasive fungal infections. Combining these two factors helps quantify the infection risk under conditions of intestinal damage and the elimination of a large number of commensal bacteria. A comprehensive risk level for fungal infection is obtained based on the degree of damage to the static intestinal baseline and the net degree of drug-induced intervention.

[0064] Preferably, in one embodiment of the present invention, the method for obtaining the overall risk level of fungal infection includes: The product of the degree of damage to the static intestinal baseline and the degree of net drug-induced intervention was obtained as the comprehensive risk of fungal infection.

[0065] It should be noted that the greater the degree of damage to the static intestinal baseline, the greater the degree of net drug-induced intervention. When the intestinal barrier is weak, a large number of normal probiotics are eliminated by drugs, resulting in a large amount of colonization space. Fungi can enter more easily, and therefore the overall risk of fungal infection is greater.

[0066] The fungal infection stratification module 104 is used to stratify fungal infection characteristics based on the overall risk distribution of fungal infections.

[0067] The overall risk of fungal infection reflects the dual superposition characteristics of the damaged intestinal barrier base and the open space of drug-induced space. The higher the overall risk of fungal infection, the more severe the historical damage to the intestinal mucosal barrier of the patient, and the larger the colonization space vacated after the large-scale elimination of the normal symbiotic flora in the intestine caused by the actual bactericidal effect of antimicrobial drugs. The more physical and ecological conditions are available for fungi to penetrate the mucosa and enter the blood, the greater the possibility of fungal infection, which helps to more accurately sort and stratify fungal infections.

[0068] Preferably, in one embodiment of the present invention, the fungal infection characteristics are stratified, including: If the overall risk of fungal infection is less than or equal to the preset low-risk threshold for fungal infection, the fungal infection characteristics are determined to be at the first level. If the overall risk of fungal infection is greater than the preset low-risk threshold for fungal infection, but less than or equal to the preset high-risk threshold for fungal infection, the fungal infection characteristic is determined to be at the second level. If the overall risk of fungal infection is greater than the preset high-risk threshold for fungal infection, the fungal infection characteristics are judged to be at level three.

[0069] It should be noted that the higher the overall risk of fungal infection, the higher the infection stratification level, i.e., the first level is lower than the second level, and the second level is lower than the third level. In the embodiments of the present invention, based on the mixed characteristic data set of confirmed invasive fungal infection cases and uninfected cases within the past year, the overall risk of fungal infection for each pathology in the set is calculated. In ascending order, the upper third position, i.e., the value at approximately 33.3%, is selected as the preset low-risk fungal risk cutoff threshold; the lower third position, i.e., the value at 66.7%, is selected as the preset high-risk fungal risk cutoff threshold. In other embodiments of the present invention, the preset low-risk fungal risk cutoff threshold and the preset high-risk fungal risk cutoff threshold can be set according to specific circumstances, and are not limited or elaborated here.

[0070] It should be noted that, in another embodiment of the present invention, the testing and adjustment are performed by obtaining the stratification of fungal infection: For the first level, the system maintains the original interface state of the nurse station screen and abandons issuing any sample modification instructions; for the second level, the system sends a text character stream suggesting attention to the hospital information system. After receiving the character stream, the hospital information system generates a text reminder pop-up window on the patient's electronic medical record homepage and abandons issuing lower-level test priority modification instructions; for the third level, the system sends an intervention warning label replacement instruction to the graphical user interface server of the hospital information system. Using this replacement instruction, the system forcibly removes the warning label. The routine green infection improvement indicator triggered by a simple decrease in procalcitonin levels within a historical range on the nurses' station screen was replaced with a red-background fungal infection pending confirmation warning indicator. A comprehensive search was performed on the pending blood sample data table in the laboratory information system, changing the key value of the test priority field from the routine level to the highest urgency level. This facilitates prioritizing and directly sending samples to the fungal-specific testing and culture channel. Based on this, and leveraging the stratified effect, the existing priority field modification function of the laboratory information system was directly invoked, automating the allocation of scarce testing resources to high-risk individuals without altering existing medical hardware connections.

[0071] In summary, this invention obtains the degree of static intestinal baseline damage based on serum lactate concentration values, temporal characteristics, and a preset maximum backtracking period at different serum lactate detection times within a historical range; it obtains the cumulative dosage of antimicrobial drugs and the hourly cumulative distribution value of drug use based on the drug dosage distribution at different medication times within the neighborhood of the real-time trigger time; it obtains the overall degree of inflammation reduction and the hourly cumulative distribution value of inflammation regression based on the procalcitonin test concentration distribution at the testing time within the neighborhood; thus, it obtains the net drug-induced intervention degree; and it stratifies fungal infection characteristics based on the degree of static intestinal baseline damage and the net drug-induced intervention degree. This invention, by considering non-drug-induced interference, accurately obtains the comprehensive risk level of fungal infection, improving the effectiveness of fungal infection stratification.

[0072] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0073] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A stratification system for secondary invasive fungal infections in sepsis based on test data, characterized in that, The system includes: The data acquisition module is used to acquire the serum lactate concentration value at the time of serum lactate detection, the drug dosage at the time of medication, and the procalcitonin concentration at the time of testing within the historical range of the real-time trigger time. The data processing module is used to obtain the degree of static intestinal baseline damage based on the serum lactate concentration values, time characteristics, and preset maximum backtracking days at different serum lactate detection times within a historical range; to obtain the cumulative dosage of antimicrobial drugs and the cumulative distribution value of medication per hour based on the drug dosage distribution at different medication times within the neighborhood range of the real-time trigger time; and to obtain the overall degree of inflammation reduction and the cumulative distribution value of inflammation reduction per hour based on the procalcitonin test concentration distribution at the test time within the neighborhood range. The fungal infection quantification module is used to obtain the drug efficacy response time-series offset based on the differences in the cumulative distribution values ​​of drug use and the cumulative distribution values ​​of remission within different hours in the neighborhood; to obtain the net drug-induced intervention degree based on the cumulative dose of antimicrobial drugs, the overall degree of inflammation reduction, and the drug efficacy response time-series offset; and to obtain the comprehensive risk of fungal infection based on the degree of damage to the static intestinal baseline and the net drug-induced intervention degree. The fungal infection stratification module is used to stratify fungal infection characteristics based on the overall risk distribution of fungal infections.

2. The stratification system for sepsis secondary invasive fungal infections based on test data according to claim 1, characterized in that, The method for obtaining the degree of damage to the static intestinal baseline includes: If there are serum lactate detection times within the historical range, the serum lactate concentration value corresponding to the adjacent serum lactate detection time of the real-time trigger time is selected as the historical concentration of the intestinal marker; if there are no serum lactate detection times within the historical range, a preset serum lactate reference value is obtained as the historical concentration of the intestinal marker; and a preset serum lactate time is obtained as the time corresponding to the historical concentration of the intestinal marker. Obtain the time difference between the real-time trigger time and the corresponding time of the historical concentration of the intestinal marker. Calculate the time difference divided by the number of hours in the previous day as the difference in days. If the difference in days is greater than the preset maximum effective backtracking days, the preset maximum effective backtracking days are used as the test data lag days. If the difference in days is less than or equal to the preset maximum effective backtracking days, the difference in days are used as the test data lag days. If the historical concentration of the intestinal marker is greater than the preset serum lactate reference value, the degree of static intestinal baseline damage is obtained based on the preset lag compensation coefficient, the number of days of test data lag, and the concentration difference between the historical concentration of the intestinal marker and the preset serum lactate reference value. The preset lag compensation coefficient, the number of days of test data lag, and the concentration difference are all positively correlated with the degree of static intestinal baseline damage. If the historical concentration of the intestinal marker is less than or equal to the preset serum lactate reference value, the historical concentration of the intestinal marker is used as the degree of static intestinal baseline damage.

3. The stratification system for sepsis secondary invasive fungal infections based on test data according to claim 1, characterized in that, The method for obtaining the cumulative dosage of the antibacterial drug includes: The drug dosage at the time of administration is normalized and used as the single-dose equivalent dose at the time of administration. The cumulative value of the single-dose equivalent dose at all times within the neighborhood is obtained as the cumulative dose of the antibacterial drug.

4. A stratification system for sepsis secondary to invasive fungal infections based on test data according to claim 3, characterized in that, The method for obtaining the cumulative distribution value of medication use includes: The difference between each medication time and the starting time within the neighborhood is obtained as the relative medication time; The ratio of the single dose at the relative time of medication administration to the sum of the cumulative dose of antimicrobial drugs and the preset adjustment coefficient is used as the percentage of a single dose at the relative time of medication administration. The cumulative value of the percentage of single-dose administration at any given time within the neighborhood is obtained as the cumulative distribution value of medication administration per hour.

5. A stratification system for sepsis secondary invasive fungal infections based on test data according to claim 1, characterized in that, The method for obtaining the overall reduction in inflammation includes: The maximum value of the procalcitonin test concentration at all test times within the neighborhood is obtained as the highest test concentration; the procalcitonin test concentration at the test time adjacent to the real-time trigger time is obtained as the last test concentration. Obtain the concentration difference between the highest test concentration and the last test concentration. If the concentration difference is greater than the preset difference threshold, the concentration difference is used as the overall fading concentration; if the concentration difference is less than or equal to the preset difference threshold, the overall fading concentration is set to 0. The overall reduction in inflammation is calculated as the sum of the overall remission concentration and the highest test concentration plus the preset adjustment coefficient.

6. A stratification system for sepsis secondary to invasive fungal infections based on test data according to claim 5, characterized in that, The method for obtaining the cumulative distribution value of the fading process includes: The difference between each test time and the starting time in the neighborhood is obtained as the relative fading time corresponding to each test time. If the relative time of decline is greater than the time corresponding to the highest test concentration, and the difference between the previous relative time of decline and the corresponding relative time of decline and the corresponding procalcitonin test concentration is greater than 0, the difference is calculated as the sum of the overall decline concentration and the preset adjustment coefficient, and is used as the single decline percentage of the corresponding relative time of decline; otherwise, the single decline percentage is set to 0. Obtain the cumulative value of the percentage of single regressions at all relative times within the neighborhood per hour, and use it as the cumulative regression distribution value for each hour.

7. A stratification system for sepsis secondary to invasive fungal infections based on test data according to claim 1, characterized in that, The method for obtaining the time offset of the drug efficacy response includes: Obtain the absolute value of the difference between the cumulative distribution value of drug use and the cumulative distribution value of drug extinction over all hours within the neighborhood, and sum them up as the drug efficacy response time offset.

8. A stratification system for sepsis secondary to invasive fungal infections based on test data according to claim 1, characterized in that, The methods for obtaining the net drug-induced intervention level include: The product of the cumulative dose of antimicrobial drugs and the overall decrease in inflammation is obtained as the theoretical efficacy response intensity; the theoretical efficacy response intensity is calculated as the sum of the efficacy response time offset and the preset adjustment coefficient, which is the net drug-induced intervention level.

9. A stratification system for sepsis secondary invasive fungal infections based on test data according to claim 1, characterized in that, The methods for obtaining the overall risk level of fungal infection include: The product of the degree of damage to the static intestinal baseline and the degree of net drug-induced intervention was obtained as the comprehensive risk of fungal infection.

10. A stratification system for sepsis secondary to invasive fungal infections based on test data according to claim 1, characterized in that, The stratification of fungal infection characteristics includes: If the overall risk of fungal infection is less than or equal to the preset low-risk threshold for fungal infection, the fungal infection characteristics are determined to be at the first level. If the overall risk of fungal infection is greater than the preset low-risk threshold for fungal infection, but less than or equal to the preset high-risk threshold for fungal infection, the fungal infection characteristic is determined to be at the second level. If the overall risk of fungal infection is greater than the preset high-risk threshold for fungal infection, the fungal infection characteristics are judged to be at level three.

Images