Hospital infection aggregation risk early warning method and system based on medication intensity of patient

Abstract

The embodiment of the invention discloses a hospital infection aggregation risk early warning method and system based on patient medication intensity. The method comprises the following steps: acquiring basic information, medication records and inspection results of a patient; cleaning and standardizing data; based on the medication conditions in the short-term and long-term windows in the processing result, calculating medication intensity, and marking high-risk patients to obtain target patient data; performing rolling accumulation statistics on the target patient data, and determining a risk level through quartile anomaly detection or an inter-day mutation threshold method according to a statistical result; different early warning channels are automatically triggered according to the risk levels, and differential intervention measures are implemented. By implementing the method provided by the embodiment of the invention, the accuracy and the response speed of hospital infection prevention and control can be effectively improved, so that the infection risk and the medical cost of a patient are remarkably reduced, and the problem that the prior art lacks an effective means for performing real-time monitoring and individualized evaluation on the hospital infection risk is solved; and the technical problems of delayed prevention and control measures and low resource utilization efficiency are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical big data processing, and more particularly to a hospital infection aggregation risk early warning method and system based on patient medication intensity. BACKGROUND

[0002] Hospital infection prevention and control is a key link to ensure medical quality and patient safety. Effective infection prevention and control measures not only can significantly reduce the risk of complications caused by infection during treatment, but also can reduce hospitalization time and medical costs, and improve the overall operational efficiency of medical institutions. In addition, with the emergence and spread of multi-drug resistant bacteria, strengthening infection prevention and control has become an important means to prevent these drug-resistant strains from further spreading and protect public health safety. Therefore, establishing a scientific and systematic infection prevention and control system is of great importance to improve medical service level and protect public health.

[0003] However, the current hospital infection prevention and control field mainly relies on macro-level data analysis methods, which to some extent limits its effectiveness. Traditional monitoring methods focus on ward-level rather than detailed data of individual patients, making it difficult to capture the specific response of each patient to antibiotic exposure and its potential risks. Early warning systems based on fixed thresholds perform poorly in the face of complex clinical situations, and are prone to false positives or false negatives, which cannot meet the needs of personalized medicine in the era of precision medicine. Furthermore, manually processing large amounts of monitoring data not only consumes time and effort, but also may miss the best intervention opportunity, increasing the risk of hospital infection outbreaks. These problems highlight the shortcomings of existing technologies in dealing with highly dynamic medical environments, emphasizing the urgent need to develop more advanced and intelligent infection prevention and control solutions.

[0004] Therefore, it is necessary to design a new method to effectively improve the precision and response speed of hospital infection prevention and control, thereby significantly reducing the risk of patient infection and medical costs, to solve the technical problems that the existing technology lacks effective means for real-time monitoring and individualized assessment of hospital infection risk, leading to lagging prevention and control measures and low resource utilization efficiency. SUMMARY

[0005] The purpose of the present application is to overcome the defects of the prior art and provide a hospital infection aggregation risk early warning method and system based on patient medication intensity.

[0006] To achieve the above-mentioned purpose, the present application adopts the following technical solution: a hospital infection aggregation risk early warning method based on patient medication intensity, comprising:

[0007] Obtaining patient basic information, medication records and test results to obtain initial data;

[0008] Cleaning and standardizing the initial data to obtain a processing result;

[0009] based on the medication in the short-term and long-term windows in the processing result, calculate the medication intensity, mark high-risk patients, and obtain target patient data;

[0010] Rolling cumulative statistics are performed on the target patient data, and potential clustering risks are identified through quartile anomaly detection or inter-daily mutation threshold method according to the statistical results to determine the risk level;

[0011] Different early warning channels are automatically triggered according to the risk level, and differentiated intervention measures are implemented.

[0012] Further technical solutions thereof are that the initial data is cleaned and standardized to obtain a processing result, including:

[0013] Key drug information is extracted from the initial data and the drug name is completed, and data that cannot be identified is marked as invalid data to obtain an information completion result;

[0014] The information completion result is subjected to drug name standardization to obtain a name standardization result;

[0015] The name standardization result is subjected to separation of drug dosage values and units from the order, and the units are converted into standard units for storage, while the original dosage description is retained for review, to obtain a processing result.

[0016] Further technical solutions thereof are that based on the medication in the short-term and long-term windows in the processing result, the medication intensity is calculated, high-risk patients are marked, and target patient data are obtained, including:

[0017] The dosages in the medication records of the processing result are standardized and converted into milligrams, and the data are dynamically updated based on 7-day and 30-day time windows to obtain a standardization result;

[0018] Static weights are set according to the categories of anti-infective drugs and the corresponding drug resistance risks, and a correction coefficient is adjusted according to the ratio of actual and recommended dosages to obtain a weight coefficient;

[0019] In a selected time window, the total medication intensity of a patient is calculated based on the daily dosage of each antibacterial drug in the standardization result and the weight coefficient;

[0020] High-risk patients are marked according to the total medication intensity in combination with individual medication baselines and ward group baselines to obtain target patient data.

[0021] Further technical solutions thereof are that static weights are set according to the categories of anti-infective drugs and the corresponding drug resistance risks, and a correction coefficient is adjusted according to the ratio of actual and recommended dosages to obtain a weight coefficient, including:

[0022] A database of anti-infective drug classification knowledge is created, a static weight is set based on drug categories and drug resistance risks, and a dose correction coefficient is adjusted according to the ratio of the actual dose to the recommended dose, so that the coefficient does not exceed 1, to obtain a weight coefficient.

[0023] Further technical solutions thereof are that the total drug intensity of the patient is calculated by combining the daily dose of each antibacterial drug in the standardized result and the weight coefficient in a selected time window, including:

[0024] The total drug intensity is calculated by adopting The cumulative dose in the time window is calculated, wherein (T) is the drug intensity, T is the number of days in the time window, D is the number of antibacterial drugs used on the day, W d is the weight coefficient of the dth drug, D d,t is the daily dose of the dth drug on the tth day, D d,t = the prescribed dose x the correction coefficient.

[0025] The short-term drug intensity is calculated by adopting

[0026] The long-term drug intensity is calculated by adopting

[0027] The total drug intensity of the patient includes the drug intensity, the short-term drug intensity, and the long-term drug intensity.

[0028] Further technical solutions thereof are that the target patient data is rolled and accumulated, potential clustering risks are identified according to the statistical result by quartile anomaly detection or inter-daily mutation threshold method, and the risk level is determined, including:

[0029] The target patient is screened every day, the related information of the target patient in the past 30 days is counted and updated, and an N30 value is obtained;

[0030] The first quartile, the third quartile, and the quartile range of the historical data are calculated based on the N30 values of the past several days;

[0031] It is judged whether the N30 value is not less than a set warning threshold;

[0032] If the N30 value is not less than the set warning threshold, it is determined that the risk level is high risk;

[0033] If the N30 value is less than the set warning threshold, when N30 is less than 10 or does not reach the quartile warning standard, the difference between the N30 value of the day and the N30 value of the previous day is calculated;

[0034] The risk level is determined according to the difference.

[0035] ​​A further technical solution is that the early warning threshold is equal to 1.5 times the interquartile range plus the third quartile.

[0036] A further technical solution is that the risk level is determined according to the difference value, including:

[0037] When the difference value is not less than 7, the risk level is high; when the difference value is not less than 5 and less than 7, the risk level is moderate; and when the difference value is less than 5 and not less than 3, the risk level is light.

[0038] A further technical solution is that different early warning channels are automatically triggered according to the risk level, and differential intervention measures are implemented, including:

[0039] According to the risk level, different early warning channels are automatically triggered, when the risk level is light, the doctor is reminded to review the medication scheme through the electronic medical record system, and the key area environmental monitoring is enhanced; when the risk level is moderate, the attending doctor and the nurse station are notified, the microbial culture application form is automatically generated, and the frequency of vital sign monitoring is increased; when the risk level is high, the early warning information is sent through all ports, the isolation medical order is generated for the patient and the single room is reserved, and the drug-resistant bacteria killing task is started.

[0040] The application also provides a hospital infection aggregation risk early warning system 300 based on patient medication intensity, including:

[0041] A data acquisition unit is configured to acquire patient basic information, medication records, and test results to obtain initial data.

[0042] A preprocessing unit is configured to clean and standardize the initial data to obtain a processing result.

[0043] A calculation unit is configured to calculate the medication intensity based on the medication in the short-term and long-term windows in the processing result, and mark high-risk patients to obtain target patient data.

[0044] A grade determination unit is configured to perform rolling cumulative statistics on the target patient data, and identify potential aggregation risk through quartile anomaly detection or interdiurnal mutation threshold method according to the statistical result to determine a risk level.

[0045] An early warning unit is configured to automatically trigger different early warning channels according to the risk level, and implement differential intervention measures.

[0046] Compared with the prior art, the present application has the beneficial effects that: the present application obtains initial data by integrating patient basic information, medication records and test results, and after data cleaning and standardization processing, the medication intensity is calculated by using short-term and long-term windows to identify and mark high-risk patients, and then the data of these target patients are statistically accumulated, the quartile anomaly detection or interdaily mutation threshold method is used to determine the potential aggregation risk level, and the corresponding early warning channel is automatically triggered according to the risk level to implement differentiated intervention measures. This method realizes real-time monitoring and individualized assessment of hospital infection risk, greatly improves the accuracy and response speed of prevention and control, effectively reduces the risk of patient infection and related medical costs, and solves the problem of lack of effective real-time monitoring means in the prior art, which leads to lagging prevention and control measures and low resource utilization efficiency.

[0047] The present application will be further described below in conjunction with the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0049] Figure 1 The application scenario schematic diagram of the hospital infection aggregation risk early warning method based on patient medication intensity provided by the embodiment of the present application is shown in the figure.

[0050] Figure 2 The flowchart of the hospital infection aggregation risk early warning method based on patient medication intensity provided by the embodiment of the present application is shown in the figure.

[0051] Figure 3 The schematic block diagram of the hospital infection aggregation risk early warning system based on patient medication intensity provided by the embodiment of the present application is shown in the figure.

[0052] Figure 4 The schematic block diagram of the computer device provided by the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0053] The technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0054] It should be understood that the terms "comprises" and "comprising," when used in this specification and the following claims, indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0055] It should also be understood that the terms used in the specification of the application herein are used for the purpose of describing particular embodiments only and are not intended to limit the application. As used in the specification and the appended claims of the application, the singular forms "a," "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

[0056] It should be further understood that the term "and / or" used in the specification of the application and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0057] Please refer to Figure 1 and Figure 2 , Figure 1 The application scenario schematic diagram of the hospital infection aggregation risk early warning method based on patient medication intensity provided by the embodiments of the application. Figure 2 The schematic flowchart of the hospital infection aggregation risk early warning method based on patient medication intensity provided by the embodiments of the application. The hospital infection aggregation risk early warning method based on patient medication intensity is applied in a server. By acquiring and processing the medication records, test results and other information of patients, the short-term and long-term medication intensities are calculated and high-risk patients are marked, and then the target patient data is statistically accumulated and rolled, the potential aggregation infection risk is identified by using the quartile anomaly detection or the inter-daily mutation threshold method, and the differentiated early warning channels and intervention measures are automatically triggered according to the risk level. This method realizes real-time monitoring and individualized evaluation of the hospital infection risk, improves the accuracy and response speed of the prevention and control measures, effectively reduces the patient infection risk and medical cost, and solves the problems of lag and low resource utilization efficiency of the prior art in hospital infection risk monitoring.

[0058] Figure 2 The flowchart of the hospital infection aggregation risk early warning method based on patient medication intensity provided by the embodiments of the application. As shown in Figure 2 The method comprises the following steps S110 to S150.

[0059] S110, acquiring patient basic information, medication records and test results to obtain initial data.

[0060] In this embodiment, the initial data mentioned above refers to a core information set collected in real-time from multiple sources such as hospital information systems (HIS), electronic medical records (EMR), and laboratory systems (LIS). These information specifically includes:

[0061] Patient basic information is mainly derived from HIS, including but not limited to patient's unique identifier (ID), age, gender, underlying disease history, admission date, inpatient department, and bed number, etc. In addition, it also includes transfer and discharge event records, providing complete inpatient trajectory for subsequent analysis.

[0062] Anti-infective drug use data is derived from the HIS system's medical orders and medication records, which detail standardized drug names, dosages, administration routes, treatment courses, and medication purposes (such as treatment or prevention). Through natural language processing (NLP) techniques, free-text medical orders are analyzed, medication purposes are associated, and all drug names are unified to a standard terminology library, facilitating subsequent calculations and comparisons.

[0063] Test results and microbiological data are derived from LIS, covering key information such as pathogen culture results, drug sensitivity reports, and inflammation markers. These data are crucial for assessing patients' responses to anti-infective drugs and are important for identifying potential high-risk patients.

[0064] Ward-bed data is derived from manually maintained ward topology information, including bed layout and adjacency relationships, and combined with transfer records in HIS to reconstruct patients' bed trajectories (in / out bed times) at different time periods, which is important for understanding potential cross-infection pathways between patients.

[0065] All the above-mentioned data is stored in a data lake after unified data cleaning and standardization processes to support subsequent risk analysis work. This process not only ensures data consistency and accuracy, but also lays a solid foundation for building anti-infective drug exposure intensity evaluation models based on individual patients. In this way, accurate monitoring and timely response to nosocomial infection risks are achieved, significantly improving the effectiveness of prevention and control measures and resource utilization efficiency.

[0066] S120, cleaning and standardizing the initial data to obtain a processing result.

[0067] In this embodiment, the processing result refers to the processing of initial data collected from multiple sources such as hospital information systems (HIS), electronic medical records (EMR), and laboratory systems (LIS) through a series of detailed data cleaning and standardization steps, thereby generating accurate, consistent, and structured data sets for subsequent analysis.

[0068] In an embodiment, the step S120 described above can include steps S121-S123.

[0069] S121, extracting key drug information from the initial data and completing the drug name, and marking the data that cannot be identified as invalid data, to obtain an information completion result.

[0070] In this embodiment, the information completion result refers to the extraction and completion of the drug name from the initial data, and the records that cannot be identified are marked as “invalid data”.

[0071] First, key drug information is extracted from the initial data, and the drug name is attempted to be completed. For example, a medical order record such as “5% GS 250ml + Ceftazidime 2g” will be parsed as “Ceftazidime”. This step aims to ensure that all drugs have a unified standard name to facilitate subsequent data processing and analysis. For those medical order records that cannot be identified or cannot be completed, they are marked as “invalid data”, and these data are only used for statistical reports and do not participate in model training.

[0072] S122, standardizing the drug name of the information completion result to obtain a name standardization result.

[0073] In this embodiment, the name standardization result refers to converting the completed drug name into a standard name according to the hospital drug list, ensuring the consistency of the drug name.

[0074] Next, the information completion result obtained in S121 is standardized for the drug name. This step involves maintaining a detailed hospital drug list that contains the mapping relationship between the original name and the standard name. For example, “Tienam” will be converted to “Imipenem / Cilastatin”, and “Zosyn” will be converted to “Cefoperazone / Sulbactam”. In this way, confusion caused by different sources using different terms to describe the same drug can be eliminated, ensuring the consistency and accuracy of the data.

[0075] S123, separating the drug dose value and unit from the name standardization result from the medical order, and converting the unit to a standard unit for storage, while retaining the original dose description for review, to obtain a processing result.

[0076] In this embodiment, the processing result refers to the completion of drug information extraction, drug name completion and standardization, and the unification of dose units, generating a structured and analyzable data set.

[0077] Finally, based on the name standardization, the dose value and unit in each medical order are further processed. The specific operation is to separate the dose value and unit from the original medical order, and convert all dose units into standard units such as milligrams (mg). For example, "ceftriaxone 2g" will be converted to 2000mg, and "levofloxacin 0.5g" will be converted to 500mg. At the same time, in order to facilitate manual review, the original dose description will also be retained. The purpose of this is to ensure that when calculating the drug intensity of patients, quantitative analysis can be based on unified standards to improve the accuracy and reliability of the evaluation.

[0078] Through the above three steps, the final processing result is a set of cleaned and standardized structured data, which not only has high consistency and accuracy, but also can be directly applied to the subsequent risk assessment model to realize effective early warning and management of hospital infection risk.

[0079] S130, based on the drug use in the short-term and long-term window in the processing result, calculate the drug intensity, mark the high-risk patients to obtain the target patient data.

[0080] In this embodiment, the target patient data refers to the list of high-risk patients identified by comparing the short-term and long-term drug intensity with the risk threshold.

[0081] In an embodiment, the above step S130 can include steps S131-S134.

[0082] S131, standardize and convert the dose in the drug record of the processing result to milligrams, and dynamically update the data based on the 7-day and 30-day time window to obtain the standardized result.

[0083] In this embodiment, the standardized result refers to the drug record after converting all drug doses to milligrams and dynamically updating according to the 7-day and 30-day time window.

[0084] Specifically, all drug doses are converted to milligrams (mg), and corresponding conversion is performed according to different units (such as g to mg, ml to mg through density, etc.). Then, for each patient, the drug record is dynamically updated based on the 7-day and 30-day time window to ensure the standardization and latest state of the dose information.

[0085] Obtain S2 the cleaned and standardized drug record (including drug name, dose, drug time, ward / bed information), ward-bed topology data (bed number, adjacent relationship);

[0086] Convert the dose in different units to milligrams (mg):

[0087] For example: "Ceftriaxone 2g" - 2000mg; "Levofloxacin 0.5g" - 500mg;

[0088] Time window division:

[0089] Short-term window: the past 7 days (including the current day);

[0090] Long-term window: the past 30 days (including the current day);

[0091] Sliding window update: automatically rolled over and calculated every morning (historical window data is retained for trend analysis).

[0092] S132, according to the anti-infective drug category and the corresponding drug resistance risk, set the static weight, and adjust the correction coefficient through the ratio of actual and recommended dose, to obtain the weight coefficient.

[0093] In this embodiment, the weight coefficient refers to the static weight set based on the anti-infective drug category and the drug resistance risk, and the correction coefficient adjusted according to the ratio of actual and recommended dose.

[0094] In an embodiment, an anti-infective drug classification knowledge base is created, and a static weight is set based on the drug category and the drug resistance risk, while a dose correction coefficient is adjusted according to the ratio of actual and recommended dose, and the coefficient is ensured not to exceed 1, to obtain the weight coefficient.

[0095] Specifically, an anti-infective drug classification knowledge base is created, in which a static weight is set for different types of anti-infective drugs according to the drug category and the drug resistance risk. At the same time, a dose correction coefficient is adjusted according to the ratio of actual and recommended dose, and the coefficient is ensured not to exceed 1, so as to accurately reflect the rationality of use of each drug and obtain an accurate weight coefficient.

[0096] A knowledge base is established to define five categories of anti-infective drugs, and a static weight and a recommended dose are set according to the drug category as shown in Table 1.

[0097] Table 1. Five categories of anti-infective drugs

[0098]

[0099] The drug weight is pre-set according to the drug category and the drug resistance risk (such as carbapenems = 1800, penicillins = 300);

[0100] The drug weight is mapped to 0-2;

[0101] Dose correction coefficient:

[0102] According to the drug category and the actual dose ≤ recommended dose, the correction coefficient = 1;

[0103] If the actual dose is greater than the recommended dose, then the correction factor = recommended dose / actual dose.

[0104] S133. Within the selected time window, the overall medication intensity for the patient is calculated by combining the daily dose of each antimicrobial drug in the standardized results with the weighting coefficient.

[0105] In this embodiment, the following is adopted: Calculate the cumulative dose within the time window, where (T) represents the drug intensity, T is the number of days in the time window, D is the number of antimicrobial drugs used on that day, and W is the cumulative dose within the time window. d D represents the weighting coefficient of the d-th drug; d,t D represents the daily dose of drug d on day t; d,t =Prescription dosage × Correction factor;

[0106] use Calculate the intensity of short-term medication;

[0107] use Calculate the intensity of long-term medication;

[0108] The patient’s overall medication intensity includes medication intensity, short-term medication intensity, and long-term medication intensity.

[0109] The patient’s overall medication intensity includes the medication intensity derived by taking into account the daily doses of all antimicrobial drugs within the time window and their weighting coefficients, the short-term medication intensity calculated specifically for 7 days, and the long-term medication intensity calculated specifically for 30 days.

[0110] Within a selected 7-day or 30-day time window, the overall medication intensity for each patient is calculated by combining the standardized daily dose and the corresponding weighting factor. This includes short-term medication intensity (obtained by averaging the daily dose of each antimicrobial agent over 7 days by its weighting factor) and long-term medication intensity (calculated using the same method but over 30 days).

[0111] Cumulative dose calculation within the time window: drug intensity

[0112] T: Time window length (7 / 30 days);

[0113] D: Number of types of antimicrobial drugs used on that day;

[0114] W d Weighting coefficients for drug d (see Table 1);

[0115] D d,t : Daily dose (g) of drug d on day t;

[0116] Short-term strength (7 days):

[0117]

[0118] Capture the surge of medication (such as postoperative infection, septic shock).

[0119]

[0120] Identify the sustained high exposure (such as repeated infection, drug-resistant bacteria colonization).

[0121] Personal baseline (individualized reference): (k = 1, 2, …, K);

[0122] K: the number of effective observation periods in the past 90 days (30 days for one period) - when K≥3, it is enabled, otherwise the ward baseline is used Ward baseline (group reference): (p∈ same ward patients);

[0123] P: the number of same type patients in the same ward (grouped according to the type of infection drug use).

[0124] D d,t = Prescribed dose × Correction factor.

[0125] S134, according to the total drug intensity combined with the personal drug baseline and the ward group baseline, mark high-risk patients to obtain target patient data.

[0126] According to the calculated short-term and long-term drug intensity, compare the acute risk threshold (≥7.5 or > personal baseline × 1.8), the chronic risk threshold (≥6.0 or > ward baseline × 1.5), identify and mark those patients with abnormally high drug intensity as high-risk patients for further attention or intervention.

[0127] Through the above steps, the target patient group that needs special attention can be effectively screened from a large amount of patient medication data, and more accurate and personalized medical management can be promoted.

[0128] S140, roll up and count the target patient data, and identify potential clustering risks according to the statistical results through quartile anomaly detection or interdaily mutation threshold method to determine the risk level.

[0129] In an embodiment, the above step S140 can include steps S141-S146.

[0130] S141, screen the target patients every day, count and update the relevant information of the target patients in the past 30 days, and obtain N30 value.

[0131] In the present embodiment, the N30 value refers to the total number of all high-risk (nosocomial infection value ≥ 5) patients in the current day and the past 29 days, so as to reflect the recent infection pressure. For example, the N30 value on December 30 is equal to the cumulative total of high-risk patients from December 1 to December 30.

[0132] S142, calculate the first quartile, the third quartile and the interquartile range of the historical data based on the N30 values of the past several days.

[0133] In the present embodiment, the first quartile (Q1) refers to the data point at the 25th percentile after arranging the N30 values of the past 90 days (excluding the current day) in ascending order, which is used to measure the lower boundary of the data distribution.

[0134] The third quartile (Q3) refers to the data point at the 75th percentile after arranging the N30 values in the same way, which is used to measure the upper boundary of the data distribution. The interquartile range (IQR) refers to the difference between Q3 and Q1, i.e. IQR = Q3-Q1, which is a key indicator for measuring the stability of the data trend.

[0135] Specifically, the historical baseline is set: using the N30 values of the past 90 days (excluding the current day) as historical data to establish the baseline.

[0136] The first quartile (Q1) is the value at the 25th percentile after arranging the historical data in ascending order.

[0137] The third quartile (Q3) is the value at the 75th percentile after arranging the historical data in the same way.

[0138] The interquartile range (IQR) is determined: IQR = Q3-Q1, which is used to measure the dispersion of the data distribution.

[0139] The standard for abnormal surge is determined by the warning threshold = Q3+1.5×IQR. When the N30 value of a certain day exceeds this threshold, it triggers a high warning, indicating that there may be an aggregated infection event.

[0140] This mechanism is applicable to the case where N30 < 10 or the quartile warning is not triggered, mainly by monitoring the change amount (Δ) of the daily N30 value: Δ = N30 of the current day-Y30.

[0141] S143, determine whether the N30 value is not less than a set warning threshold.

[0142] In the present embodiment, the warning threshold is equal to 1.5 times the interquartile range plus the third quartile.

[0143] The early warning threshold is equal to 1.5 times the interquartile range plus the third quartile (early warning threshold = 1.5 x IQR + Q3). If the N30 value of the day is greater than this threshold, it is considered that an abnormal surge has occurred.

[0144] S144, if the N30 value is not less than the set early warning threshold, the risk level is determined to be high risk.

[0145] If the above conditions are met, it indicates that there is a significant increase in the risk of infection, and immediate measures should be taken to block the transmission chain and prioritize control of the source of infection.

[0146] S145, if the N30 value is less than the set early warning threshold, and N30 is less than 10 or does not reach the quartile early warning standard, calculate the difference between the N30 value of the day and the N30 value of the previous day.

[0147] In this embodiment, for those cases where the quartile early warning is not triggered but N30 is less than 10, or N30 is higher than 10 but lower than the early warning threshold, further analysis of short-term fluctuations is needed, and the change trend is evaluated by calculating Δ = N30 of the day - Y30 (Y30 is the N30 value of the previous day).

[0148] Y30 (N30 of the previous day) refers to the N30 value calculated the previous day, mainly used to monitor short-term trends. For example, if the N30 value on December 30 is 45, then the Y30 value on December 31 is also 45.

[0149] S146, determine the risk level according to the difference.

[0150] In this embodiment, when the difference is not less than 7, the risk level is high; when the difference is not less than 5 and less than 7, the risk level is moderate; when the difference is less than 5 and not less than 3, the risk level is mild. This classification helps to take targeted measures and strengthen basic infection control to prevent risk escalation.

[0151] The risk level classification rules are shown in Table 2.

[0152] Table 2. Risk level classification rules

[0153]

[0154] S150, automatically trigger different early warning channels according to the risk level and implement differentiated intervention measures.

[0155] In this embodiment, different early warning channels are automatically triggered according to the risk level. When the risk level is low, the doctor is reminded to review the medication plan through the electronic medical record system, and the monitoring of the key area environment is enhanced; when the risk level is medium, the attending physician and the nurse station are notified, a microbial culture application form is automatically generated, and the frequency of vital sign monitoring is increased; when the risk level is high, the early warning information is sent through all ports, an isolation order is generated for the patient, a single room is reserved, and the drug-resistant bacteria killing task is started.

[0156] For the aggregation early warning, the system automatically calculates N30 of the current day and further analyzes and statistics. The red early warning and the short-term fluctuation high early warning belong to the high early warning, and the specific early warning and intervention measures are shown in Table 3.

[0157] Table 3. Specific early warning and intervention measures

[0158]

[0159] The patient's antibacterial drug use record and ward transfer track are extracted from the hospital information system in real time, so as to establish a three-dimensional original data set containing time and space dimensions. This process ensures that the dynamic changes of the patient during treatment can be fully captured, providing a solid data foundation for subsequent analysis.

[0160] The 7-day short-term window and 30-day long-term window are used to dynamically process the data stream, and the antibacterial spectrum classification weight and recommended dose adjustment coefficient are combined to calculate the individualized medication intensity value. This design not only considers the acute medication pressure (through the short-term window), but also evaluates the chronic exposure risk (through the long-term window). In this way, the limitations of traditional ward mean statistics are broken through, and more accurate individualized medication intensity quantification is achieved.

[0161] Based on the received medication intensity value, three fusion rules are used for judgment, including absolute threshold comparison, personal historical baseline comparison, and ward group reference analysis. This method first directly marks high-risk cases through the set absolute threshold, then further screens out potential risk cases by comparing the personal 90-day moving average intensity with the personal historical baseline, and finally compares with the average level of patients of the same type in the same ward to output a real-time updated high-risk patient list.

[0162] Using the rolling calculation method, the 30-day cumulative number (N30) and the daily increment index (Δ value) are calculated based on the high-risk patient list, and the early warning is triggered through two mode detection mechanisms. One is the historical data analysis based on the quartile method, which is used to identify continuous abnormalities; the other is the stepwise mutation threshold method, which is specially designed to capture single-day dramatic increase in risk. Through the combination of these two strategies, three-level early warning information can be generated, effectively monitoring and warning possible aggregation infection events.

[0163] This method captures the acute medication pressure and chronic exposure risk of patients through two different time windows: a 7-day short-term window and a 30-day long-term window. Specifically:

[0164] Short-term window (7 days): mainly used to assess the concentration and intensity of drug use in the short term to identify possible acute medication pressure.

[0165] Long-term window (30 days): used to assess drug use over a longer period of time, particularly the risk of chronic exposure.

[0166] In addition, the model incorporates the drug weight coefficients set by the antibacterial drug classification system and introduces a knowledge graph-based anti-infective drug recommended dose adjustment mechanism. Through these information, a function calculation model is established to accurately quantify the medication intensity for each patient. This method not only breaks through the limitations of traditional ward mean statistics, but also provides more accurate and personalized treatment recommendations.

[0167] To more accurately identify high-risk patients, a hierarchical and progressive risk determination method is constructed, which adopts the following three levels of fusion rules:

[0168] First-level absolute threshold trigger: when the short-term intensity reaches or exceeds 7.5, or the long-term intensity reaches or exceeds 6.0, it is directly marked as a high-risk patient.

[0169] Second-level dynamic baseline comparison: if a patient's medication intensity exceeds the moving average of the past 90 days, it is considered that the patient has a higher individual risk.

[0170] Third-level group reference determination: compare the patient's medication intensity with the data of patients of the same type in the same ward, if it exceeds the average level, it is further confirmed as high-risk.

[0171] In the above three determination methods, the priority order is absolute threshold > individual baseline > ward baseline, ensuring that both obvious risk signals can be quickly discovered and true high-risk patients can be accurately distinguished, thereby improving the sensitivity and specificity of risk identification.

[0172] To effectively monitor and warn possible group infection events, a high-risk patient aggregation analysis rule is proposed, including the following aspects:

[0173] Sliding window statistical mechanism: use a ring buffer to efficiently calculate the 30-day rolling cumulative number of high-risk patients (N30), and generate an incremental change Δ value with the previous day's cumulative value (Y30).

[0174] Dual-mode warning strategy:

[0175] Long-term trend warning: Adopt quartile method (Q3+1.5IQR) to identify abnormal baseline fluctuation, and help to find the persistent risk growth trend.

[0176] Short-term mutation warning: Capture the sharp rise of local risk according to the Δ value classification (3 / 5 / 7), and timely issue an alarm.

[0177] In addition, the rule also innovatively integrates spatio-temporal clustering analysis technology, independently calculates N30 according to the disease area, and correlates pathogen homology data (such as genetic sequencing drug resistance spectrum), so as to more accurately locate the hotspot area of infection transmission. This method not only improves the accuracy of early warning, but also provides strong support for hospital management and public health decision-making.

[0178] The above-mentioned hospital infection aggregation risk early warning method based on patient drug intensity obtains initial data by integrating patient basic information, drug records and test results, and after data cleaning and standardization processing, calculates the drug intensity by using short-term and long-term windows to identify and mark high-risk patients, and then performs rolling cumulative statistics on the data of these target patients, uses quartile anomaly detection or interdaily mutation threshold method to determine the potential aggregation risk level, and automatically triggers the corresponding early warning channel to implement differentiated intervention measures according to the risk level. This method realizes real-time monitoring and individualized evaluation of hospital infection risk, greatly improves the accuracy and response speed of prevention and control, effectively reduces the risk of patient infection and related medical costs, and solves the problem of lack of effective real-time monitoring means in the prior art, which leads to lagging prevention and control measures and low resource utilization efficiency.

[0179] Figure 3 is a schematic block diagram of a hospital infection aggregation risk early warning system 300 based on patient drug intensity provided by an embodiment of the present application. As Figure 3 shown, corresponding to the above hospital infection aggregation risk early warning method based on patient drug intensity, the present application also provides a hospital infection aggregation risk early warning system 300 based on patient drug intensity. The hospital infection aggregation risk early warning system 300 based on patient drug intensity includes a unit for executing the above hospital infection aggregation risk early warning method based on patient drug intensity, and the system can be configured in a server. Specifically, please refer to Figure 3 , the hospital infection aggregation risk early warning system 300 based on patient drug intensity includes a data acquisition unit 301, a preprocessing unit 302, a calculation unit 303, a level determination unit 405, and a warning unit 305.

[0180] The data acquisition unit 301 is configured to acquire patient basic information, medication records and test results to obtain initial data; the preprocessing unit 302 is configured to clean and standardize the initial data to obtain processing results; the calculation unit 303 is configured to calculate medication intensity based on medication in a short-term and long-term window in the processing results, and mark high-risk patients to obtain target patient data; the grade determination unit 405 is configured to perform rolling cumulative statistics on the target patient data, and identify potential aggregation risks by quartile anomaly detection or inter-daily mutation threshold method according to a statistical result to determine a risk grade; and the early warning unit 305 is configured to automatically trigger different early warning channels according to the risk grade, and implement differentiated intervention measures.

[0181] In an embodiment, the preprocessing unit 302 comprises:

[0182] The completion sub-unit is configured to extract key drug information from the initial data and complete drug names, and mark data that cannot be identified as invalid data to obtain information completion results; the name standardization sub-unit is configured to perform drug name standardization on the information completion results to obtain name standardization results; and the unit conversion sub-unit is configured to separate drug dose values and units from the name standardization results, and store the units converted into standard units while retaining original dose descriptions for review to obtain processing results.

[0183] In an embodiment, the calculation unit 303 comprises:

[0184] The standardization sub-unit is configured to standardize and convert doses in medication records of the processing results to milligrams, and dynamically update data based on a 7-day and 30-day time window to obtain standardization results; the coefficient correction sub-unit is configured to set static weights according to anti-infective drug categories and corresponding drug resistance risks, and adjust correction coefficients by a ratio of actual to recommended doses to obtain weight coefficients; the intensity calculation sub-unit is configured to calculate overall medication intensity of a patient in a selected time window by combining daily doses of each anti-infective drug in the standardization results and the weight coefficients; and the marking sub-unit is configured to mark high-risk patients according to the overall medication intensity in combination with individual medication baselines and ward group baselines to obtain target patient data.

[0185] In an embodiment, the coefficient correction sub-unit is configured to create an anti-infective drug classification knowledge base, set static weights based on drug categories and drug resistance risks, and adjust dose correction coefficients according to a ratio of actual to recommended doses, and ensure that the coefficients do not exceed 1 to obtain weight coefficients.

[0186] In an embodiment, the intensity calculation sub-unit is configured to adopt Cumulative dose in a time window is calculated, wherein (T) is the medication intensity, T is the number of days in the time window; D is the number of the types of antibacterial drugs used on the day; W d is the weight coefficient of the dth drug; D d,t is the daily dose of the dth drug on the tth day; D d,t = the prescribed dose x the correction coefficient;

[0187] The short-term medication intensity is calculated. The short-term medication intensity is calculated.

[0188] The long-term medication intensity is calculated. The long-term medication intensity is calculated.

[0189] The total medication intensity of the patient includes the medication intensity, the short-term medication intensity and the long-term medication intensity.

[0190] In an embodiment, the grade determination unit 405 includes:

[0191] The screening statistical subunit is configured to screen the target patient on a daily basis, to statistically and update the relevant information of the target patient in the past 30 days to obtain an N30 value; the value calculation subunit is configured to calculate the first quartile, the third quartile and the quartile range of the historical data based on the N30 values of the past several days; the judgment subunit is configured to judge whether the N30 value is not less than a set early warning threshold; the determination subunit is configured to determine that the risk grade is high if the N30 value is not less than the set early warning threshold; the difference calculation subunit is configured to calculate the difference between the N30 value of the current day and the N30 value of the previous day if the N30 value is less than the set early warning threshold, when the N30 is less than 10 or does not reach the quartile early warning standard; and the grade determination subunit is configured to determine the risk grade according to the difference.

[0192] In an embodiment, the difference calculation subunit is configured to determine that the risk grade is high when the difference is not less than 7, to determine that the risk grade is moderate when the difference is not less than 5 and less than 7, and to determine that the risk grade is light when the difference is less than 5 and not less than 3.

[0193] In an embodiment, the early warning unit 305 is configured to automatically trigger different early warning channels according to the risk grade, to remind the doctor to review the medication scheme through the electronic medical record system and to enhance the environmental monitoring of the key area when the risk grade is light, to notify the attending doctor and the nurse station, to automatically generate a microbial culture application form and to increase the frequency of vital sign monitoring when the risk grade is moderate, and to send early warning information through all ports, to generate an isolation medical order for the patient and to make an appointment for a single room, and to start the task of killing drug-resistant bacteria when the risk grade is high.

[0194] It should be noted that the specific implementation process of the hospital infection aggregation risk early warning system 300 based on the medication strength of the patient and each unit can be clearly understood by those skilled in the art, and can refer to the corresponding description in the foregoing method embodiments. For the convenience and brevity of description, it will not be repeated here.

[0195] The hospital infection aggregation risk early warning system 300 based on the medication strength of the patient can be implemented in the form of a computer program, which can run on a computer device as shown in the figure. Figure 4

[0196] Please refer to Figure 4 , Figure 4 is a schematic block diagram of a computer device provided by an embodiment of the present application. The computer device 500 can be a server, wherein the server can be a stand-alone server or a server cluster composed of multiple servers.

[0197] Referring to Figure 4 , the computer device 500 includes a processor 502, a memory, and a network interface 505 connected through a system bus 501, wherein the memory can include a non-volatile storage medium 503 and an internal memory 504.

[0198] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions which, when executed, can cause the processor 502 to perform a hospital infection aggregation risk early warning method based on the medication strength of the patient.

[0199] The processor 502 is configured to provide computing and control capabilities to support the operation of the entire computer device 500.

[0200] The internal memory 504 provides an environment for the running of the computer program 5032 in the non-volatile storage medium 503, which, when executed by the processor 502, can cause the processor 502 to perform a hospital infection aggregation risk early warning method based on the medication strength of the patient.

[0201] The network interface 505 is configured to perform network communication with other devices. Those skilled in the art can understand that Figure 4 the structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device 500 to which the scheme of the present application is applied. The specific computer device 500 can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0202] The processor 502 is configured to run the computer program 5032 stored in the memory to implement the following steps: ​

[0203] Obtaining patient basic information, medication records and test results to obtain initial data; cleaning and standardizing the initial data to obtain processed results; calculating medication intensity based on medication within short-term and long-term windows in the processed results, marking high-risk patients to obtain target patient data; rolling cumulative statistics on the target patient data, identifying potential clustering risks through quartile anomaly detection or inter-daily mutation threshold method according to the statistical results to determine risk levels; automatically triggering different early warning channels according to the risk levels and implementing differentiated intervention measures.

[0204] In an embodiment, the processor 502, when implementing the cleaning and standardizing the initial data to obtain processed results step, specifically implements the following steps:

[0205] Extracting key drug information from the initial data and completing drug names, marking unrecognizable data as invalid data to obtain information completion results; standardizing drug names of the information completion results to obtain name standardization results; separating drug dose values and units from the name standardization results and storing the units converted into standard units while retaining the original dose descriptions for review to obtain processed results.

[0206] In an embodiment, the processor 502, when implementing the calculating medication intensity based on medication within short-term and long-term windows in the processed results, marking high-risk patients to obtain target patient data step, specifically implements the following steps:

[0207] Standardizing and converting doses in the medication records of the processed results to milligrams, and dynamically updating data based on 7-day and 30-day time windows to obtain standardized results; setting static weights according to anti-infective drug categories and corresponding drug resistance risks, and adjusting correction coefficients through actual-to-recommended dose ratios to obtain weight coefficients; calculating the overall medication intensity of the patient in the selected time window by combining the daily dose of each antibacterial drug in the standardized results and the weight coefficients; marking high-risk patients according to the overall medication intensity combined with individual medication baselines and ward group baselines to obtain target patient data.

[0208] In an embodiment, the processor 502, when implementing the setting static weights according to anti-infective drug categories and corresponding drug resistance risks, and adjusting correction coefficients through actual-to-recommended dose ratios to obtain weight coefficients step, specifically implements the following steps:

[0209] Creating an anti-infective drug classification knowledge base, setting static weights based on drug categories and drug resistance risks, and adjusting dose correction coefficients according to actual-to-recommended dose ratios to ensure that the coefficients do not exceed 1 to obtain weight coefficients.

[0210] In an embodiment, the processor 502, when implementing the step of calculating the overall drug intensity of the patient in combination with the daily dose of each antibacterial drug in the standardized result and the weight coefficient within the selected time window, specifically implements the following steps:

[0211] Using Calculate the cumulative dose within the time window, where (T) is the drug intensity, T is the number of days in the time window; D is the number of antibacterial drugs used on the day; W d is the weight coefficient of the dth drug; D d,t is the daily dose of the dth drug on the tth day; D d,t = Prescribed dose x Correction coefficient; Using Calculate the short-term drug intensity; Using Calculate the long-term drug intensity; Where the overall drug intensity of the patient includes the drug intensity, the short-term drug intensity, and the long-term drug intensity.

[0212] In an embodiment, the processor 502, when implementing the step of rolling cumulative statistics on the target patient data and identifying potential clustering risks through quartile anomaly detection or inter-daily mutation threshold method based on the statistical result to determine the risk level, specifically implements the following steps:

[0213] Filter the target patient every day, count and update the relevant information of the target patient in the past 30 days to obtain the N30 value; Calculate the first quartile, the third quartile, and the interquartile range of the historical data based on the N30 values of the past several days; Determine whether the N30 value is not less than the set warning threshold; If the N30 value is not less than the set warning threshold, determine that the risk level is high risk; If the N30 value is less than the set warning threshold, when N30 is less than 10 or does not reach the quartile warning standard, calculate the difference between the N30 value of the current day and the N30 value of the previous day; Determine the risk level according to the difference.

[0214] Wherein the warning threshold is equal to 1.5 times the interquartile range plus the third quartile.

[0215] In an embodiment, the processor 502, when implementing the step of determining the risk level according to the difference, specifically implements the following steps:

[0216] When the difference is not less than 7, the risk level is high; When the difference is not less than 5 and less than 7, the risk level is moderate; When the difference is less than 5 and not less than 3, the risk level is mild.

[0217] In an embodiment, the processor 502, when implementing the step of automatically triggering different warning channels according to the risk level and implementing differentiated intervention measures, specifically implements the following steps:

[0218] According to the risk level, different early warning channels are automatically triggered, when the risk level is light, the doctor is reminded to review the medication scheme through the electronic medical record system, and the environment monitoring of the key area is enhanced; when the risk level is medium, the attending doctor and the nurse station are notified, the microbial culture application form is automatically generated, and the frequency of vital sign monitoring is increased; when the risk level is high, the early warning information is sent through all ports, the isolation order is generated for the patient, and the single room is reserved, and the drug-resistant bacteria killing task is started.

[0219] It should be understood that, in the embodiments of the present application, the processor 502 can be a central processing unit (CPU), and the processor 502 can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0220] It can be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware. The computer program includes program instructions, and the computer program can be stored in a storage medium, which is a computer readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the above-mentioned embodiments.

[0221] Therefore, the present application also provides a storage medium. The storage medium can be a computer readable storage medium. The storage medium stores a computer program, wherein the computer program is executed by a processor to make the processor execute the following steps:

[0222] Obtaining patient basic information, medication records and test results to obtain initial data; cleaning and standardizing the initial data to obtain processing results; calculating the medication intensity based on the medication in the short-term and long-term windows in the processing results, and marking high-risk patients to obtain target patient data; rolling cumulative statistics are performed on the target patient data, and potential aggregation risks are identified according to the statistical results through quartile anomaly detection or interdaily mutation threshold method to determine the risk level; different early warning channels are automatically triggered according to the risk level, and differentiated intervention measures are implemented.

[0223] In an embodiment, the processor, when executing the computer program to implement the cleaning and standardizing the initial data to obtain a processing result, implements the following steps:

[0224] extracting key drug information from the initial data and completing drug names, and marking unrecognizable data as invalid data to obtain an information completion result; performing drug name standardization on the information completion result to obtain a name standardization result; separating drug dose values and units from the name standardization result, and converting the units into standard units while retaining the original dose descriptions for review to obtain a processing result.

[0225] In an embodiment, the processor, when executing the computer program to implement the calculating the medication intensity based on the medication situation in the short-term and long-term windows in the processing result, and marking high-risk patients to obtain target patient data, implements the following steps:

[0226] standardizing and converting the dose in the medication record of the processing result into milligrams, and dynamically updating the data based on a 7-day and 30-day time window to obtain a standardization result; setting a static weight according to the anti-infective drug category and the corresponding drug resistance risk, and adjusting a correction coefficient by the ratio of the actual dose to the recommended dose to obtain a weight coefficient; in a selected time window, combining the daily dose of each antibacterial drug in the standardization result and the weight coefficient to calculate the overall medication intensity of the patient; marking high-risk patients according to the overall medication intensity in combination with the individual medication baseline and the ward group baseline to obtain target patient data.

[0227] In an embodiment, the processor, when executing the computer program to implement the setting a static weight according to the anti-infective drug category and the corresponding drug resistance risk, and adjusting a correction coefficient by the ratio of the actual dose to the recommended dose to obtain a weight coefficient step, implements the following steps:

[0228] creating an anti-infective drug classification knowledge base, setting a static weight based on the drug category and the drug resistance risk, and adjusting a dose correction coefficient according to the ratio of the actual dose to the recommended dose, ensuring that the coefficient does not exceed 1, to obtain a weight coefficient.

[0229] In an embodiment, the processor, when executing the computer program to implement the calculating the overall medication intensity of the patient in a selected time window, combining the daily dose of each antibacterial drug in the standardization result and the weight coefficient step, implements the following steps:

[0230] adopting calculating the cumulative dose in the time window, where (T) is the medication intensity, T is the number of days in the time window, and D is the number of antibacterial drugs used on the day.d is a weight coefficient of the dth drug; D d,t is a daily dose of the dth drug on the tth day; D d,t = Prescribed dose x Correction coefficient

[0231] The short-term medication intensity is calculated by using The short-term medication intensity is calculated by using

[0232] The long-term medication intensity is calculated by using The long-term medication intensity is calculated by using

[0233] The total medication intensity of the patient includes the medication intensity, the short-term medication intensity and the long-term medication intensity.

[0234] In an embodiment, when the processor executes the computer program to implement the rolling cumulative statistics of the target patient data, and according to the statistical result, identifies the potential aggregation risk by the quartile anomaly detection or the inter-daily mutation threshold method to determine the risk level, the following steps are implemented:

[0235] The target patient of each day is screened, the related information of the target patient in the past 30 days is counted and updated to obtain an N30 value; the first quartile, the third quartile and the interquartile range of the historical data are calculated based on the N30 values of the past several days; it is judged whether the N30 value is not less than a set early warning threshold; if the N30 value is not less than the set early warning threshold, it is determined that the risk level is high risk; if the N30 value is less than the set early warning threshold, when the N30 is less than 10 or the quartile early warning standard is not reached, the difference between the N30 value of the current day and the N30 value of the previous day is calculated; the risk level is determined according to the difference.

[0236] The early warning threshold is equal to 1.5 times the interquartile range plus the third quartile.

[0237] In an embodiment, when the processor executes the computer program to implement the step of determining the risk level according to the difference, the following steps are implemented:

[0238] When the difference is not less than 7, the risk level is high; when the difference is not less than 5 and less than 7, the risk level is moderate; when the difference is less than 5 and not less than 3, the risk level is mild.

[0239] In an embodiment, when the processor executes the computer program to implement the step of automatically triggering different early warning channels according to the risk level and implementing differentiated intervention measures, the following steps are implemented:

[0240] According to the risk level, different early warning channels are automatically triggered, when the risk level is light, the doctor is reminded to review the medication scheme through the electronic medical record system, and the environment monitoring of the key area is enhanced; when the risk level is medium, the attending doctor and the nurse station are informed, the microbial culture application form is automatically generated, and the frequency of vital sign monitoring is increased; when the risk level is high, the early warning information is sent through all ports, the isolation order is generated for the patient, and the single room is reserved, and the drug-resistant bacteria killing task is started.

[0241] The storage medium can be a U disk, a mobile hard disk, a read-only memory (ROM), a magnetic disk or an optical disk, and various computer readable storage media that can store program codes.

[0242] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been described in the above description in general terms. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0243] In several embodiments provided by the present application, it should be understood that the disclosed system and method can be implemented in other ways. For example, the system embodiments described above are only illustrative. For example, the division of each unit is only a logical functional division, and actual implementation can have another division method. For example, a plurality of units or components can be combined or integrated into another system, or some features can be omitted or not executed.

[0244] The steps in the method of the embodiments of the present application can be adjusted, combined and deleted in sequence according to actual needs. The units in the system of the embodiments of the present application can be combined, divided and deleted according to actual needs. In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit.

[0245] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a storage medium. Based on such an understanding, the technical solutions of the present application essentially or say the part that contributes to the prior art, or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a terminal, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application.

[0246] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical range disclosed by the present application, and these modifications or replacements should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for early warning of hospital-acquired infection clustering risk based on patient medication intensity, characterized in that, include: Obtain basic patient information, medication records, and test results to obtain initial data; The initial data is cleaned and standardized to obtain the processing result; Based on the medication usage within the short-term and long-term windows in the processing results, the medication intensity is calculated, and high-risk patients are marked to obtain target patient data; The target patient data is statistically analyzed on a rolling basis, and potential clustering risks are identified based on the statistical results using quartile anomaly detection or diurnal mutation threshold method to determine the risk level. Different early warning channels are automatically triggered based on the risk level, and differentiated intervention measures are implemented.

2. The method for early warning of hospital-acquired infection clustering risk based on patient medication intensity according to claim 1, characterized in that, The process of cleaning and standardizing the initial data to obtain the processing result includes: Key drug information is extracted from the initial data and the drug name is completed. Unidentifiable data is marked as invalid data to obtain the information completion result. The information completion results are then standardized using drug names to obtain standardized names. The standardized name results are obtained by separating the drug dosage value and unit from the medical order, converting the unit into a standard unit for storage, and retaining the original dosage description for review, so as to obtain the processing result.

3. The method for early warning of hospital-acquired infection clustering risk based on patient medication intensity according to claim 1, characterized in that, The process involves calculating medication intensity based on short-term and long-term medication usage within the processing results, identifying high-risk patients, and obtaining target patient data, including: The dosage in the medication records of the processed results is standardized and converted to milligrams, while the data is dynamically updated based on 7-day and 30-day time windows to obtain the standardized results; Static weights are set according to the types of anti-infective drugs and the corresponding drug resistance risks, and correction coefficients are adjusted by the ratio of actual to recommended doses to obtain the weight coefficients. Within a selected time window, the overall medication intensity for the patient is calculated by combining the daily dose of each antimicrobial drug in the standardized results with the weighting coefficient. High-risk patients were identified by combining the overall medication intensity with individual medication baselines and ward population baselines to obtain target patient data.

4. The method for early warning of hospital-acquired infection clustering risk based on patient medication intensity according to claim 3, characterized in that, The process involves setting static weights based on the type of anti-infective drug and the corresponding drug resistance risk, and adjusting correction coefficients by the ratio of actual to recommended doses to obtain weighting coefficients, including: A knowledge base for classifying anti-infective drugs is created, and static weights are set based on drug categories and drug resistance risks. At the same time, the dosage correction coefficient is adjusted according to the ratio of actual to recommended dosage, ensuring that the coefficient does not exceed 1, so as to obtain the weight coefficient.

5. The method for early warning of hospital-acquired infection clustering risk based on patient medication intensity according to claim 3, characterized in that, Within a selected time window, the overall medication intensity for the patient is calculated by combining the daily dose of each antimicrobial drug in the standardized results with the weighting coefficients, including: use Calculate the cumulative dose within the time window, where (T) represents the drug intensity, T is the number of days in the time window, D is the number of antimicrobial drugs used on that day, and W is the cumulative dose within the time window. d D represents the weighting coefficient of the d-th drug; d,t D represents the daily dose of drug d on day t. d,t =Prescription dosage × Correction factor; use Calculate the intensity of short-term medication; use Calculate the intensity of long-term medication; The patient’s overall medication intensity includes medication intensity, short-term medication intensity, and long-term medication intensity.

6. The method for early warning of hospital-acquired infection clustering risk based on patient medication intensity according to claim 1, characterized in that, The rolling cumulative statistical analysis of the target patient data, and the identification of potential clustering risks based on the statistical results using quartile anomaly detection or diurnal mutation thresholding to determine the risk level, includes: Target patients are screened daily, and relevant information of the target patients over the past 30 days is collected and updated to obtain the N30 value; Calculate the first quartile, third quartile, and interquartile range of historical data based on the N30 value over the past several days; Determine whether the N30 value is not less than the set warning threshold; If the N30 value is not less than the set warning threshold, then the risk level is determined to be high risk; If the N30 value is less than the set warning threshold, when N30 is less than 10 or does not reach the quartile warning standard, the difference between the N30 value of the day and the N30 value of the previous day is calculated. The risk level is determined based on the difference.

7. The method for early warning of hospital-acquired infection clustering risk based on patient medication intensity according to claim 6, characterized in that, The warning threshold is equal to 1.5 multiplied by the interquartile range plus the third quartile.

8. The method for early warning of hospital-acquired infection clustering risk based on patient medication intensity according to claim 7, characterized in that, The step of determining the risk level based on the difference includes: When the difference is not less than 7, the risk level is high; when the difference is not less than 5 and less than 7, the risk level is moderate; when the difference is less than 5 and not less than 3, the risk level is mild.

9. The method for early warning of hospital-acquired infection clustering risk based on patient medication intensity according to claim 1, characterized in that, The automatic triggering of different early warning channels based on the risk level, and the implementation of differentiated intervention measures, include: Based on the risk level, different early warning channels are automatically triggered. At a low risk level, the electronic medical record system reminds doctors to review medication plans and enhances environmental monitoring in key areas. At a medium risk level, the attending physician and nurses' station are notified, a microbial culture request form is automatically generated, and the frequency of vital sign monitoring is increased. At a high risk level, early warning information is sent to all terminals, an isolation order is generated for the patient, a single ward is reserved, and a drug-resistant bacteria disinfection task is initiated.

10. A hospital infection cluster risk early warning system 300 based on patient medication intensity, characterized in that, include: The data acquisition unit is used to acquire basic patient information, medication records, and test results to obtain initial data; A preprocessing unit is used to clean and standardize the initial data to obtain the processing result; The calculation unit is used to calculate the medication intensity based on the medication information within the short-term and long-term windows in the processing results, and to mark high-risk patients in order to obtain target patient data; The risk level determination unit is used to perform rolling cumulative statistics on the target patient data, and identify potential clustering risks based on the statistical results using quartile anomaly detection or diurnal mutation threshold method to determine the risk level. The early warning unit is used to automatically trigger different early warning channels based on the risk level and implement differentiated intervention measures.

Images