Nosocomial infection risk monitoring method and nosocomial infection risk monitoring system based on anti-infective drug use trajectory

Abstract

The invention discloses a hospital infection risk monitoring method and system based on an anti-infective drug use trajectory. The method comprises the following steps: acquiring diagnosis and treatment feature data of a patient during hospitalization; cleaning and standardizing drug names involved in the diagnosis and treatment feature data to standard terms, and retaining discharge drug-carrying identifiers to obtain a preprocessing result; generating a drug effect weight system of five types of anti-infection drugs according to the preprocessing result in combination with statistical regression analysis and an expert scoring method so as to determine drug weights; according to the medicine weight and the invasive operation factor, calculating a daily infection risk index of the patient in real time through a rule engine; grading according to the daily infection risk index of the patient, performing grading early warning, and starting corresponding intervention measures. By implementing the method provided by the invention, the hospital infection risk can be evaluated and warned in real time, the early infection recognition capability is effectively improved, and the antibacterial drug management strategy is optimized.

Description

TECHNICAL FIELD

[0001] The present application relates to a medical data method, in particular to an in-hospital infection risk monitoring method and system based on anti-infective drug use trajectory. BACKGROUND

[0002] Infections occurring during medical procedures not only significantly increase the incidence of complications and mortality in patients, but also accelerate the development of drug-resistant strains. Existing monitoring mechanisms mainly rely on retrospective pathogenic detection and biochemical index analysis. This mode usually has a delay of 48 to 72 hours, making it difficult to discover early signs of infection in a timely manner. At the same time, due to the widespread use of prophylactic broad-spectrum antibiotics and frequent use of invasive diagnostic and treatment equipment, drug-resistant bacteria are inadvertently provided with a growth environment. Unfortunately, current decision support systems lack dynamic adjustment capabilities when evaluating treatment plans, and cannot effectively warn of potential infection risks.

[0003] Although modern medical information systems have achieved digital management of patient data, there are still technical bottlenecks in intelligently analyzing antibiotic use patterns. Most current risk assessment models are based on fixed parameter systems and do not fully consider factors such as changes in antibiotic types, adjustments in drug administration strategies, and individual differences among patients, resulting in insufficient accuracy of prediction results. In particular, in high-risk environments such as intensive care units and hematology oncology departments, traditional assessment methods are difficult to accurately measure the risk of hospital infections caused by special level antibiotics such as carbapenems and glycopeptides.

[0004] Therefore, it is necessary to design a new method to realize real-time evaluation and early warning of in-hospital infection risks, effectively improve early infection identification capabilities, and optimize antibiotic management strategies; to solve the problem that the prior art lacks dynamic monitoring and analysis of patient anti-infective drug use trajectory, resulting in limited early infection identification capabilities and difficulty in effectively optimizing antibiotic management strategies and reducing the risk of drug-resistant bacteria transmission. SUMMARY

[0005] The present application aims to overcome the shortcomings of the prior art and provide an in-hospital infection risk monitoring method and system based on anti-infective drug use trajectory.

[0006] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0007] Further technical solutions are as follows:

[0008] Compared with the prior art, the present application has the beneficial effects that: by integrating the diagnosis and treatment feature data during the patient's hospitalization, the drug name is cleaned and standardized, and the discharge drug identification is retained, the statistical regression analysis and expert scoring method are combined to generate the efficacy weight system of five types of anti-infective drugs, the invasive operation factor is considered, the daily infection risk index of the patient is calculated in real time by using the rule engine, and the risk index is divided into grades for grading early warning and starting corresponding intervention measures, so that the real-time assessment and early warning of the risk of nosocomial infection are realized, and the early infection identification capability is significantly improved; this method overcomes the problem that the prior art lacks dynamic monitoring and analysis of the use track of anti-infective drugs, effectively optimizes the anti-infective drug management strategy, and reduces the risk of drug-resistant bacteria transmission. This method provides a data-driven, accurate and timely tool for medical institutions to improve the effectiveness and efficiency of hospital infection control.

[0009] The present application will be further described below in conjunction with the drawings and specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0010] 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.

[0011] Figure 1 The application scenario diagram of the nosocomial infection risk monitoring method based on the use track of anti-infective drugs provided by the embodiments of the present application is shown.

[0012] Figure 2 The flowchart of the nosocomial infection risk monitoring method based on the use track of anti-infective drugs provided by the embodiments of the present application is shown.

[0013] Figure 3 The schematic block diagram of the nosocomial infection risk monitoring system based on the use track of anti-infective drugs provided by the embodiments of the present application is shown.

[0014] Figure 4 The schematic block diagram of the computer device provided by the embodiments of the present application is shown. DETAILED DESCRIPTION

[0015] 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 part of the embodiments of the present application, not all the 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.

[0016] It should be understood that the terms "comprise" and "include" as used in the specification and the appended 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.

[0017] It should also be understood that the terms used in the present application specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the present application specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0018] It should be further understood that the term "and / or" used in the present application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations thereof, and includes these combinations.

[0019] Please refer to Figure 1 and Figure 2 , Figure 1 The application scenario schematic diagram of the hospital infection risk monitoring method based on anti-infective drug use trajectory provided by the embodiments of the present application. Figure 2 The schematic flowchart of the hospital infection risk monitoring method based on anti-infective drug use trajectory provided by the embodiments of the present application. The hospital infection risk monitoring method based on anti-infective drug use trajectory is applied in a server. The server interacts with a terminal for data, acquires and processes the diagnosis and treatment feature data of a patient, including anti-bacterial drug use records, invasive operation data, etc., generates a drug efficacy weight system by using statistical regression analysis and expert scoring method, and combines a rule engine to calculate a daily infection risk index in real time, and then dynamically warns and starts corresponding intervention measures according to the risk level. This method realizes comprehensive and dynamic monitoring and analysis of the anti-infective drug use trajectory, effectively improves the early infection identification capability, optimizes the antibacterial drug management strategy, and solves the problems of insufficient early infection identification and increased risk of drug-resistant bacteria transmission caused by the lack of dynamic evaluation of the patient's drug use trajectory in the prior art.

[0020] Figure 2 The flowchart of the hospital infection risk monitoring method based on anti-infective drug use trajectory provided by the embodiments of the present application. As shown in Figure 2 , the method comprises the following steps S110 to S150.

[0021] S110, acquiring diagnosis and treatment feature data of a patient during hospitalization.

[0022] In the present embodiment, the diagnosis and treatment feature data comprises basic information, anti-bacterial drug use records, invasive operation data, and pathogenic detection results.

[0023] This step aims to comprehensively collect all relevant information that may affect the patient's risk of infection during hospitalization, ensuring the accuracy of the infection risk assessment for each patient by integrating data from the hospital information system (HIS), laboratory information system (LIS), and electronic medical record (EMR). Specifically:

[0024] ETL (Extract, Transform, Load) tools are used to incrementally extract data from the above systems daily, using the patient's hospitalization number as the unique identifier, to achieve cross-system information correlation, including medication records, test results, and medical order information.

[0025] Basic information: including but not limited to hospitalization number, age, gender, admission diagnosis, admission time, and discharge time, etc. Basic information provides necessary background information for subsequent analysis.

[0026] Clinical operation records: covering invasive operations (such as central venous catheterization, urinary catheterization duration, ventilator use time, surgical incision grade, and postoperative nursing records, etc.) and drug-related operations (such as antimicrobial skin test records and their time). These data are crucial for understanding the patient's clinical process, as they directly or indirectly affect the risk of infection.

[0027] Medical order data: Special attention is paid to the use of anti-infective drugs, including drug generic name, medication start and end time, medication purpose (treatment / prevention / discharge medication), and retention of medical order effective information (such as prescribing physician, medical order time). This helps to understand drug use patterns and their potential impact.

[0028] Test and microbiological data: including pathogenic detection results (such as sampling time and results of blood culture, sputum culture, and urine culture, including bacterial species name and drug sensitivity report), inflammation indicators (such as C-reactive protein (CRP), procalcitonin (PCT), white blood cell count and classification dynamic change value), and imaging examination results. These data are particularly crucial for monitoring and assessing changes in infection status.

[0029] Historical infection data: involving patients with a history of nosocomial infection, infection status at different stages, and corresponding antimicrobial use regimen. This is very important for identifying high-risk groups and developing targeted interventions.

[0030] By integrating the above data, the above steps not only provide a detailed view of the patient profile, but also lay a solid foundation for subsequent data analysis, model establishment, and risk assessment. In particular, by retaining the "discharge medication" label, unnecessary interference factors can be effectively excluded, allowing the rule engine to more accurately calculate the patient's infection risk index.

[0031] S120, clean and standardize the drug names involved in the diagnosis and treatment feature data to standard terms, and retain the out-of-hospital drug identification to obtain a preprocessing result.

[0032] In this embodiment, the preprocessing result refers to a structured and uniform format data set obtained from the original diagnosis and treatment feature data after a series of standardization and cleaning steps. Specifically, in this embodiment, the preprocessing result contains the following characteristics:

[0033] Drug name standardization: All drug names have been converted to their generic names, ensuring that data from different sources can be consistently compared and analyzed. This includes mapping trade names or aliases to their corresponding generic names (e.g., "Tienam" - imipenem cilastatin sodium), and handling special cases of drug name mapping (e.g., mandatory mapping of "Compound sulfamethoxazole" and "Compound Sulfamethoxazole").

[0034] Dose specification information removal: The dose and specification information in the drug description has been removed, leaving only the active ingredient name (e.g., "Amoxicillin capsules 0.5g" is simplified to "Amoxicillin"). This helps reduce variable complexity and facilitates subsequent data analysis.

[0035] Label incomplete data: For cases where key information cannot be identified or is missing, it has been labeled as "data incomplete". These records are only for statistical purposes and do not affect the final data collection process.

[0036] Retain out-of-hospital drug identification: While standardizing drug information, the relevant identification of drugs taken by patients upon discharge is also retained, which is very important for understanding the discharge medication of patients.

[0037] In summary, the preprocessing result is a clear, accurate and easy-to-analyze data set. Through systematic cleaning and standardization processes, it improves the quality and usability of data, providing a solid foundation for further data analysis. This result can be directly used in subsequent risk assessment, decision support and other application scenarios.

[0038] In an embodiment, the above step S120 can include steps S121-S124.

[0039] S121, extract and retain drug names from the medical order text involved in the diagnosis and treatment feature data, and label the cases where the drug names cannot be identified as "data incomplete".

[0040] In this embodiment, drug names are identified and extracted from medical order text to ensure that each mentioned drug is recorded.

[0041] If there are cases that cannot be identified (e.g., missing key information or unclear expression), it is marked as "data incomplete". Such records are only for statistical purposes and will not affect the final data collection process.

[0042] S122, the drug trade name or alias in the medical order text involved in the diagnosis and treatment feature data is converted into the corresponding generic name to standardize the representation.

[0043] In this embodiment, when a specific drug combination is mentioned (such as "0.9% NaCl 100ml + Ceftazidime 2g"), the main drug name (Ceftazidime) is automatically extracted. If the completion cannot be completed, it is also marked as "data incomplete".

[0044] S123, remove the dosage and specification information in the drug description involved in the diagnosis and treatment feature data, and only keep the active ingredient name.

[0045] In this embodiment, the trade name or alias of all drugs involved in the diagnosis and treatment feature data is converted into its generic name. For example, "Tian Neng" should be mapped to "Imipenem / Cilastatin Sodium". This is to ensure that data from different sources can be consistently compared and analyzed.

[0046] S124, the drug name in special cases in the diagnosis and treatment feature data is mapped through the manually maintained exception word table processing to obtain the pre-processing result.

[0047] In this embodiment, the trade name or alias appearing in the medical order is mapped to the corresponding generic name using the hospital's drug catalog database, further supporting the standardization process.

[0048] Remove the dosage and specification information in the drug description, and only keep the active ingredient name. For example, "Amoxicillin Capsules 0.5g" is simplified to "Amoxicillin". This helps to reduce the complexity of variables and facilitate subsequent analysis.

[0049] Focus on extracting the active ingredient name of the drug, ignoring its dosage information, so as to facilitate clearer data analysis.

[0050] Use the manually maintained exception word table to handle the drug name mapping problem in special cases. For example, the forced mapping of "Compound Sulfamethoxazole" and "Compound Sulfamethoxazole" ensures that even in the face of these special cases, the correct naming standardization can be achieved.

[0051] Through the above steps S121 to S124 and the corresponding supplementary explanations, the overall cleaning and standardization of the drug names involved in the diagnosis and treatment feature data are achieved, while the discharge medication identification is retained, thereby obtaining the preprocessing results that can be used for further analysis and processing. This process not only improves the consistency and accuracy of the data, but also provides a reliable data basis for subsequent risk assessment and decision support.

[0052] S130, generating a drug efficacy weight system of five categories of anti-infective drugs according to the preprocessing results combined with statistical regression analysis and expert scoring method, to determine the drug weight.

[0053] In this embodiment, the drug weight refers to a value obtained by adjusting the influence degree of the drug on the patient's health state transition (quantified by the risk ratio) combined with expert opinions.

[0054] In this embodiment, the step S130 aims to create a drug efficacy weight system that can quantify the influence of each anti-infective drug on the patient's health state transition according to the preprocessing results combined with statistical regression analysis and expert scoring method.

[0055] In an embodiment, the above step S130 can include steps S131-S133.

[0056] S131, creating a health condition-drug time sequence during hospitalization based on the patient's clinical symptoms, microbiological culture results and inflammation indicators in the preprocessing results.

[0057] In this embodiment, the health condition-drug time sequence during hospitalization refers to time sequence data recording the patient's daily health state and the use of anti-infective drugs.

[0058] In an embodiment, the above step S131 can include steps S1311-S1312.

[0059] S1311, dividing the patient's hospitalization state in the preprocessing results into five levels: healthy, suspected, infected, improved, and recovered.

[0060] In this embodiment, first, the patient's hospitalization state in the preprocessing results is divided into five levels: healthy, suspected, infected, improved, and recovered. These state definitions are based on clinical symptoms, microbiological culture results and inflammation indicators (such as CRP / PCT), ensuring the consistency and accuracy of the data.

[0061] S1312, dividing the patient's hospitalization record in the preprocessing results by day, and marking the daily health state and the use of anti-infective drugs and their doses, combined with the divided levels, to form the health condition-drug time sequence during hospitalization.

[0062] Then, the patient's hospitalization records are divided by day, and the health status of each day and the anti-infective drugs used and their dosages are marked. This step provides basic data support for subsequent state transition correlation analysis by constructing a detailed health status-drug time series during hospitalization.

[0063] S132, according to the health status-drug time series during hospitalization, calculate the state transition probability and risk ratio of using a specific drug in different states, evaluate the influence of the drug on the patient's health state transition, and obtain the risk ratio of the drug.

[0064] In this embodiment, the risk ratio of the drug refers to the ratio of the probability of turning from health to infection under the condition of using a specific drug to the corresponding probability under the condition of not using the drug, which is used to evaluate the influence of the drug on state transition.

[0065] In an embodiment, the above step S132 can include steps S1321-S1322.

[0066] S1321, according to the health status-drug time series during hospitalization, statistically calculate the probability of all patients turning from one health state to another state under the condition of using a specific drug, and form a multi-dimensional transition probability matrix.

[0067] In this embodiment, the multi-dimensional transition probability matrix refers to the matrix formed by statistically calculating the probability of all patients turning from one health state to another state under the condition of using a specific drug, which is used to analyze the influence of different drugs on health state changes.

[0068] In this embodiment, according to the time series data obtained in S131, the probability of all patients turning from one health state to another state under the condition of using a specific drug is statistically calculated to form a multi-dimensional transition probability matrix. This matrix helps to quantify the state transition probability of using a specific drug in different states.

[0069] S1322, according to the multi-dimensional transition probability matrix, calculate the risk ratio of each drug causing the transition from health to infection state, evaluate the influence of the drug on state transition, and obtain the risk ratio of the drug.

[0070] In this embodiment, the risk ratio (HR) of each drug causing the transition from health to infection state is calculated using the above transition probability matrix. HR is an important evaluation index that measures the degree of influence of the drug on the patient's health state transition, thereby providing a basis for determining the drug efficacy weight.

[0071] S133, based on the risk ratio of the drug and combined with expert opinions, adjust to determine the basic classification drug efficacy weight of each antibacterial drug, and determine the drug weight.

[0072] In an embodiment, the step S133 described above can include steps S1331-S1332.

[0073] S1331, according to the risk ratio of the drug mapping to the preset weight range, the basic weight of each drug is determined.

[0074] In the present embodiment, based on the calculated risk ratio, it is mapped to the preset weight range (for example, HR=1 corresponds to W=0, HR=8 corresponds to W=300) to determine the basic weight of each drug. This process takes into account the actual degree of influence of the drug on the patient's state transition.

[0075] S1332, according to the drug efficacy, prevention effect, combined drug effect and drug resistance, the basic weight of each drug is adjusted to determine the efficacy weight of each type of antibacterial drug.

[0076] In the present embodiment, finally, the basic weight is further adjusted by the expert committee of the infectious disease department according to factors such as drug efficacy, prevention effect, combined drug effect and drug resistance. For example, for drugs with significant efficacy in treating confirmed infections (such as vancomycin for treating MRSA), their weights may be increased; while for drugs mainly used for preventing infections (such as cefazolin before surgery), their weights are correspondingly reduced. In addition, the synergistic effect needs to be considered in the case of combined drug use, and the weight of the drug with drug resistance report also needs to be appropriately reduced.

[0077] Through the above steps, a detailed efficacy weight system for five types of anti-infective drugs is finally formed, which not only quantifies the specific influence of the drug on the patient's health state transition, but also incorporates expert opinions, making the evaluation more scientific and reasonable.

[0078] In the present embodiment, the state-drug time series during the patient's hospitalization is established to provide basic data for subsequent weight calculation.

[0079] The state is defined as follows:

[0080] Healthy: no symptoms of infection, negative microbial culture results, and normal inflammation indicators (CRP / PCT).

[0081] Suspected: symptoms of infection but not yet confirmed (e.g. fever plus elevated white blood cells, but negative microbial culture results).

[0082] Infection: positive microbial culture or clinically confirmed infection (based on imaging or symptoms support).

[0083] Improved: symptoms are relieved, antibiotic treatment regimen is reduced, and microbial culture results turn negative.

[0084] Recovery: no recurrence or discharge after stopping anti-infective treatment for at least 72 hours.

[0085] The patient's hospitalization records are divided by day, and a daily health status label is generated. At the same time, the anti-infective drugs used on the day and their doses (such as imipenem, vancomycin, etc.) are labeled.

[0086] Quantify the strength of the impact of drugs on state transitions (such as from a healthy state to an infected state).

[0087] Statistical all patients from one health state to another state, build a 5x5 transition probability matrix. In the case of using a specific drug, calculate the conditional transition probability

[0088] Calculate the risk ratio of each drug D to promote the transition of a healthy state to an infected state

[0089] Map the risk ratio to a weight range, with the formula W base = 100 x log2(HRD). For example, HR = 1 corresponds to W = 0 (indicating no effect), and HR = 8 corresponds to W = 300 (indicating high risk).

[0090] Adjust the weight value by the expert committee of the infectious disease department according to the specific situation. For example, for drugs used to treat confirmed infections and have significant efficacy (such as vancomycin for MRSA), the weight is multiplied by 1.5; for drugs used to prevent infection (such as cefazolin before surgery), the weight is multiplied by 0.3; if it is a combination drug, the weight is adjusted according to the synergistic effect (for example, β-lactam + aminoglycoside, total weight = 1.2 x (W1 + W2)). If the drug sensitivity report indicates drug resistance, the weight of the relevant drug is multiplied by 0.5.

[0091] The final result of the classification of the efficacy weight of the antibacterial drug is shown in Table 1, and it should be noted that Table 1 only presents part of the weight.

[0092] Table 1. Drug weight

[0093]

[0094]

[0095] S140, according to the drug weight, invasive operation factor, through the rule engine to calculate the daily infection risk index of the patient.

[0096] In this embodiment, the daily infection risk index of the patient refers to a quantitative index reflecting the daily infection risk level of the patient, which is calculated by the rule engine by comprehensively considering the use of antibacterial drugs, pathogen detection, surgery and invasive operation, etc.

[0097] In an embodiment, the step S140 described above can include steps S141-S146.

[0098] S141, associate the pre-processing results by hospitalization ID to obtain an association result.

[0099] In this embodiment, the association result refers to integrating the multi-source data of the patient's basic information, medical order information, and antibacterial drug use into a comprehensive data set through the hospitalization ID.

[0100] Specifically, the information from different data sources (such as patient basic information A, medical order effective information B, and antibacterial drug use information C) is associated through the hospitalization ID.

[0101] The integrated data set contains all relevant information of the patient from admission to discharge, including but not limited to personal basic information, treatment plan, medication, and time points.

[0102] The data collected from S1 includes patient basic information A (hospitalization ID, admission and discharge time), medical order effective information B (information meeting "doctor ≠ null" and "order time ∈ A.t"), antibacterial drug use information C (including use start and end time), drug-related operation information D, antibacterial drug type and weight information E, pathogenic detection information F, and surgery / invasive operation information G.

[0103] S142, according to the association result, determine whether the patient's hospitalization time and antibacterial drug use time have intersection, to determine whether antibacterial drugs are used on the same day, to obtain a determination result.

[0104] In this embodiment, the determination result refers to determining whether the patient actually used antibacterial drugs on the same day based on the intersection of the patient's hospitalization time period and antibacterial drug use time.

[0105] Specifically, it is determined whether antibacterial drugs are used on a certain day, and this is used as the basis for assessing the patient's infection risk.

[0106] If the patient's hospitalization time period intersects with the start and end time of antibacterial drug use, it is considered that the patient used antibacterial drugs on the same day; otherwise, it is not used. This result is an important basis for subsequent steps.

[0107] According to the patient's hospitalization ID, the patient's basic information A and medical order effective information B are associated to obtain medical order information B(a) that meets the conditions. If the patient has not been discharged, the statistical time is the discharge time.

[0108] S143, according to the determination result, filter out patient information using antibacterial drugs on the same day but excluding skin tests and non-discharged drug cases.

[0109] In this embodiment, cases of using antibiotics for skin test or non-discharge medication on the same day are excluded to ensure that only those drug use situations that truly affect the risk of infection are considered.

[0110] Check if A.t intersects with C.t, if yes, obtain the information of patients C(a) who have valid medical orders and used anti-infective drugs on the same day Y , and enter S144; if there is no intersection, it is considered that the patient did not use anti-infective drugs on the same day, and the infection risk index is 0.

[0111] S144, according to whether the time since the patient's admission is more than 48 hours and the case of re-admission within 48 hours after the last discharge, the patient group is subdivided.

[0112] In this embodiment, the patient is classified based on whether the time since the patient's admission is more than 48 hours and the case of re-admission within 48 hours after the last discharge. This subdivision helps to more accurately assess and adjust the infection risk index.

[0113] Further analyze C(a) Y , exclude the information of patients D(a) Y for skin test or non-discharge medication, and mark the remaining information as D(a) N . Select D(a) Y to enter the next step, and D(a) N its infection risk index is set to 0.

[0114] S145, analyze the changes of the patient's anti-infective drug use, and adjust the infection risk index according to the drug weight.

[0115] In this embodiment, the types of anti-infective drugs used by the patient each day and their corresponding weight changes are checked, and the differences between the current day and the previous day are compared, such as newly added drugs, reduced drugs, or replaced drugs with higher / lower weight, etc.

[0116] According to the above changes, the infection risk index of the patient is dynamically adjusted, for example, the addition of high-weight drugs increases the risk index, and vice versa.

[0117] Based on the patient's basic information A, judge whether the time from admission to the same day is ≤48 hours. The information A(a) Y of the patient who meets this condition enters the next stage, and A(a) N does not enter S146. For A(a) Y , further subdivides the patient information A(b) Y whose last discharge to this admission is not more than 48 hours, and marks the other information as A(b) NAnd set its infection risk index to 0.

[0118] S146, combined with etiology detection information, check if there is a matching pathogen detection with the use of anti-infective drugs, and adjust the infection risk index accordingly; for patients who cannot match the use of anti-infective drugs, classify according to the previous day's pathogen detection, and adjust the infection risk index accordingly; for other cases, based on the key pathogen detection of the day, make a further assessment and adjust the infection risk index; for patients whose risk index has not changed, consider whether they have surgery or invasive operation on the day, to decide whether to adjust the infection risk index, to get the daily infection risk index of the patient.

[0119] In this embodiment, check if there is a matching pathogen detection with the use of anti-infective drugs. If there is a match, the infection risk index is increased accordingly.

[0120] For patients without daily matching, review the previous day's pathogen detection to classify and adjust the risk index accordingly.

[0121] Further consider the key pathogen detection of the day, even if there is no actual detection, adjust the risk index according to the submission behavior.

[0122] Finally, for patients whose risk index has not changed, investigate whether they have surgery or invasive operation on the day, which may also affect the final risk index.

[0123] In summary, the S140 step aims to use comprehensive data correlation, detailed time intersection judgment, accurate patient classification, drug use change analysis and comprehensive etiology detection information to realize real-time calculation and adjustment of the daily infection risk index of patients through the rule engine. This method not only provides more personalized and accurate risk assessment, but also helps guide clinical decision-making and optimize treatment plans.

[0124] Merge A(a) N and A(b) Y Get A(c), according to the type and weight information E of anti-infective drug use, compare the change of anti-infective drug use weight between the day and the previous day, classify the patient information as E(a) Y or E(a) N , the former enters "merge all patient information that meets the conditions", the latter further judges whether there is new, reduced or replaced drug, respectively enters different sub-steps, and finally adjusts the infection risk index of the patient according to the specific situation.

[0125] Merge all eligible patient information E(d), and determine whether there is a pathogen detection matching the anti-infective drug used according to the etiology detection information F. If so, the patient information F(a) Y The infection risk index of the patient is +3 higher than that of the previous day; if not, proceed to the next step.

[0126] Refine the patient information based on the etiology detection information F and the surgical / invasive operation information G in turn, and adjust the infection risk index of the patient accordingly according to whether there is a key pathogen detection or submission matching the anti-infective drug used on the previous day or the current day, whether a surgical or invasive operation is performed, etc. Each step specifies in detail how to increase or keep the infection risk index unchanged according to specific conditions.

[0127] When there are multiple conditions, the priority processing path is followed to ensure that each step accurately reflects the infection risk status of the patient.

[0128] S150, according to the daily infection risk index of the patient, divide the level and grade the warning, and start the corresponding intervention measures.

[0129] In an embodiment, the above-mentioned step S150 can include steps S151-S152.

[0130] S151, according to the daily infection risk index of the patient, dynamically divide the infection risk into four levels of no risk, mild risk, moderate risk and high risk, and correspond to different clinical significance and monitoring needs;

[0131] S152, automatically calculate the daily infection risk index of the patient and compare the previous day's level, when the risk level is found to be higher or continuously high-risk, send a warning to medical staff through different channels, and take corresponding level of intervention measures.

[0132] According to the daily updated hospital infection value (0-10 points) of the patient, the infection risk level is dynamically divided as shown in Table 2.

[0133] Table 2. Infection risk level

[0134]

[0135]

[0136] The system automatically calculates the hospital infection value every morning, and if the level is higher than the previous day or continuously high-risk, it triggers a warning, as shown in Table 3.

[0137] Table 3. Intervention rules

[0138]

[0139] In this embodiment, the method realizes the whole-process closed-loop management from data collection to early warning response. Specifically, the patient's medication records, test results and operation records are extracted from the hospital information system in real time; the collected data is processed into a unified format to ensure data consistency and comparability; the historical infection data and expert scores are combined to generate the risk weight of the drug through linear regression analysis and Delphi method; the above weight is used to dynamically evaluate the risk value of each patient for nosocomial infection; and the corresponding early warning measures are triggered according to the risk level of the patient.

[0140] The method of this embodiment uses a mixed model of statistical analysis and expert knowledge to calculate the risk weight of the drug, which not only considers the drug use characteristics such as the number of days of medication and the number of combined medications, but also includes the professional scores of the infection department experts on the antibacterial spectrum and drug resistance; a differentiated early warning and intervention strategy for different risk levels is designed, the mild risk is reminded through the electronic medical record, the moderate risk sends a message to the doctor, and the high risk activates the full-port early warning, including mobile phone message warning and isolation of medical advice generation; a layered infection risk index calculation system is established based on the rule engine, which gives priority to drug use, followed by clinical operation and test indicators, to accurately assess the daily infection risk level of the patient.

[0141] The method of this embodiment is particularly suitable for the infection control scene of general hospitals, and through dynamic tracking of the use trajectory of antibacterial drugs and risk coupling relationship analysis, it significantly improves the ability to identify nosocomial infections early. It not only optimizes the management strategy of antibacterial drugs, but also helps to reduce the probability of the spread of drug-resistant bacteria. In addition, this technical solution overcomes the problem of lag in traditional biomarker detection, providing a real-time risk stratification tool for the clinic, promoting the rational use of antibacterial drugs, and having important significance for preventing the spread of drug-resistant bacteria.

[0142] The above method automatically collects various medical treatment feature data of patients during hospitalization by using a data integration interface, including but not limited to basic hospitalization information, antibacterial drug use records (clearly distinguishing between treatment, prevention and discharge medication types), invasive medical treatment operation records and pathogenic detection results; a drug feature standardization model is developed for semantic mapping and normalization processing of drug names from different sources to ensure data consistency and accuracy; according to historical infection case data, statistical regression analysis combined with Delphi expert scoring method is used to determine the efficacy weight of various anti-infective drugs. This method not only considers the historical data of drug use, but also integrates the professional opinions of experts in the field; through the decision logic module, the drug use characteristics of the patient are analyzed in real time to generate a personalized infection risk index, thereby accurately reflecting the current infection risk level of each patient; according to the calculated risk level (no / low / medium / high risk), the corresponding multi-level warning mechanism is triggered. Warning information can be conveyed to medical staff through a medical workstation computer terminal, mobile device push or automatically generated isolation recommendations, etc., to achieve precise warning and timely intervention.

[0143] The above method can dynamically track and analyze the use trajectory of antibacterial drugs and its relationship with infection risk, thereby significantly improving the early identification capability of hospital-acquired infections. In addition, it also helps to optimize the management strategy of antibacterial drugs and reduce the risk of drug-resistant bacteria transmission, and is particularly suitable for application in the infection control environment of general hospitals. Through this comprehensive data-driven method, medical institutions can more effectively manage and reduce the incidence of nosocomial infections.

[0144] The above-mentioned hospital-acquired infection risk monitoring method based on the use trajectory of anti-infective drugs integrates the medical treatment feature data of patients during hospitalization, cleans and standardizes the drug names and retains the discharge medication identification, generates a drug efficacy weight system for five types of anti-infective drugs by combining statistical regression analysis and expert scoring method, considers the invasive operation factors, uses a rule engine to calculate the daily infection risk index of patients in real time, and according to the risk index classification, carries out graded warning and starts corresponding intervention measures, thereby realizing real-time assessment and early warning of nosocomial infection risk, and significantly improving the early infection identification capability; this method overcomes the problem of lack of dynamic monitoring and analysis of the use trajectory of anti-infective drugs in the prior art, effectively optimizes the management strategy of antibacterial drugs, and reduces the risk of drug-resistant bacteria transmission. This method provides a data-driven, accurate and timely tool for medical institutions to improve the effectiveness and efficiency of hospital infection control.

[0145] Figure 3 is a schematic block diagram of a hospital-acquired infection risk monitoring system 300 based on the use trajectory of anti-infective drugs provided by an embodiment of the present application. As shown in Figure 3Corresponding to the above hospital infection risk monitoring method based on anti-infective drug use trajectory, the present application also provides a hospital infection risk monitoring system 300 based on anti-infective drug use trajectory. The hospital infection risk monitoring system 300 based on anti-infective drug use trajectory includes units for executing the above hospital infection risk monitoring method based on anti-infective drug use trajectory, and the system can be configured in a server. Specifically, please refer to Figure 3 The hospital infection risk monitoring system 300 based on anti-infective drug use trajectory includes a data acquisition unit 301, a preprocessing unit 302, a weight determination unit 303, an index determination unit 304, and a warning unit 305.

[0146] The data acquisition unit 301 is used to acquire the diagnosis and treatment feature data of the patient during hospitalization; the preprocessing unit 302 is used to clean and standardize the drug names involved in the diagnosis and treatment feature data to standard terms, and retain the out-of-hospital drug identification to obtain a preprocessing result; the weight determination unit 303 is used to generate a drug efficacy weight system of five categories of anti-infective drugs according to the preprocessing result combined with statistical regression analysis and expert scoring method, to determine the drug weight; the index determination unit 304 is used to calculate the daily infection risk index of the patient in real time through a rule engine according to the drug weight and invasive operation factors; the warning unit 305 is used to divide the level according to the daily infection risk index of the patient, and grade the warning, and start the corresponding intervention measures.

[0147] In an embodiment, the preprocessing unit 302 includes:

[0148] The marking sub-unit is used to extract and retain the drug name from the medical order text involved in the diagnosis and treatment feature data, and mark it as incomplete data for cases that cannot be identified; the conversion sub-unit is used to convert the drug trade name or alias in the medical order text involved in the diagnosis and treatment feature data to the corresponding generic name for standardized representation; the removal sub-unit is used to remove the dosage and specification information in the drug description involved in the diagnosis and treatment feature data, and only retain the active ingredient name; the processing sub-unit is used to map the drug name in special cases in the diagnosis and treatment feature data through artificial maintenance of exception word table processing, to obtain a preprocessing result.

[0149] In an embodiment, the weight determination unit 303 includes:

[0150] The creating sub-unit is configured to create a health condition-drug time sequence during hospitalization based on the clinical symptoms, the microbial culture results and the inflammation indexes of the patients in the preprocessing results; the risk ratio calculating sub-unit is configured to calculate the state transition probability and the risk ratio of using a specific drug in different states according to the health condition-drug time sequence during hospitalization, evaluate the influence of the drug on the state transition of the patient, and obtain the risk ratio of the drug; and the drug weight determining sub-unit is configured to determine the basic classification drug efficacy weight of each antibacterial drug based on the risk ratio of the drug and combined with expert opinions adjustment, and determine the drug weight.

[0151] In an embodiment, the creating sub-unit comprises:

[0152] The state level dividing module is configured to divide the hospitalization states of the patients in the preprocessing results into five levels of health, suspected infection, infection, improvement and recovery; the state marking module is configured to divide the hospitalization records of the patients in the preprocessing results by day, mark the health state and the used anti-infective drugs and their dosages every day, and combine the divided levels to form the health condition-drug time sequence during hospitalization.

[0153] In an embodiment, the risk ratio calculating sub-unit comprises:

[0154] The probability matrix module is configured to calculate the probability of all patients in the health condition-drug time sequence during hospitalization from one health state to another state when using a specific drug, form a multi-dimensional transition probability matrix; and the evaluation module is configured to calculate the risk ratio of each drug leading to the transition from the health state to the infection state according to the multi-dimensional transition probability matrix, evaluate the influence of the drug on the state transition, and obtain the risk ratio of the drug.

[0155] In an embodiment, the drug weight determining sub-unit comprises:

[0156] The mapping module is configured to map the risk ratio of the drug to a preset weight range to determine the basic weight of each drug; and the adjustment module is configured to adjust the basic weight of each drug according to the drug efficacy, the preventive effect, the combined drug effect and the drug resistance to determine the efficacy weight of each antibacterial drug.

[0157] In an embodiment, the index determining unit 304 comprises:

[0158] The association subunit is used for associating the preprocessing result with the hospitalization ID to obtain an association result; the judgment subunit is used for judging whether the hospitalization time of the patient intersects with the time of using the antibacterial drug according to the association result, so as to determine whether the antibacterial drug is used on the day to obtain a judgment result; the screening subunit is used for screening the patient information of using the antibacterial drug on the day but excluding the skin test and the non-discharge drug taking condition according to the judgment result; the subdivision subunit is used for subdividing the patient group according to whether the time since the patient is hospitalized is more than 48 hours and whether the patient is hospitalized again within 48 hours after the last discharge; the analysis subunit is used for analyzing the change of the patient in using the antibacterial drug, and adjusting the infection risk index according to the drug weight; the index adjustment subunit is used for checking whether the pathogen detection matches the used antibacterial drug on the day in combination with the pathogen detection information, and adjusting the infection risk index accordingly; for the patient who cannot be matched with the used antibacterial drug, the patient is classified according to the pathogen detection of the previous day, and the infection risk index is adjusted accordingly; for other conditions, the infection risk index is adjusted again based on the key pathogen detection on the day; for the patient whose infection risk index is still not changed, whether the patient has surgery or invasive operation on the day is considered, so as to determine whether the infection risk index is adjusted, so as to obtain the daily infection risk index of the patient.

[0159] In an embodiment, the early warning unit 305 comprises:

[0160] The risk division subunit is used for dynamically dividing the infection risk into four levels of no risk, low risk, medium risk and high risk according to the daily infection risk index of the patient, and corresponding different clinical significance and monitoring requirements; the intervention subunit is used for automatically calculating the daily infection risk index of the patient and comparing the level of the previous day, and when the risk level is found to be increased or continuously high, the medical staff is warned through different channels, and the intervention measures of the corresponding level are taken.

[0161] It should be noted that the specific implementation process of the above-mentioned hospital infection risk monitoring system 300 based on the antibacterial drug use track and each unit can be clearly understood by those skilled in the art, which can refer to the corresponding description in the foregoing method embodiments. For the convenience and brevity of description, it will not be repeated here.

[0162] The above-mentioned hospital infection risk monitoring system 300 based on the antibacterial drug use track can be realized in the form of a computer program, which can run on a computer device as shown in the accompanying drawings. Figure 4

[0163] 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.

[0164] 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.

[0165] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions that, when executed, cause the processor 502 to perform a hospital infection risk monitoring method based on anti-infective drug use trajectory.

[0166] The processor 502 is configured to provide computing and control capabilities to support the operation of the entire computer device 500.

[0167] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503, which, when executed by the processor 502, causes the processor 502 to perform a hospital infection risk monitoring method based on anti-infective drug use trajectory.

[0168] 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.

[0169] The processor 502 is configured to run the computer program 5032 stored in the memory to implement the following steps:

[0170] Obtain diagnosis and treatment feature data during the patient's hospitalization; clean and standardize the drug names involved in the diagnosis and treatment feature data to standard terms, and retain the out-of-hospital drug identification to obtain a preprocessing result; generate a drug efficacy weight system of five types of anti-infective drugs according to the preprocessing result combined with statistical regression analysis and expert scoring method to determine drug weights; calculate the patient's daily infection risk index in real time through a rule engine according to the drug weights and invasive operation factors; classify and grade the patient's daily infection risk index, and grade the warning to start corresponding intervention measures.

[0171] The diagnosis and treatment feature data includes basic information, antibacterial drug use records, invasive operation data, and pathogenic detection results.

[0172] In an embodiment, when implementing the step of preprocessing the diagnosis and treatment feature data, the processor 502 specifically implements the following steps:

[0173] Extract and retain drug names from the order text involved in the diagnosis and treatment feature data, and mark as incomplete data for cases that cannot be identified; convert the drug trade names or aliases in the order text involved in the diagnosis and treatment feature data into the corresponding generic names for standardized representation; remove the dosage and specification information in the drug description involved in the diagnosis and treatment feature data, and only retain the active ingredient name; map the drug names in special cases in the diagnosis and treatment feature data through artificial maintenance of an exception word table to obtain the preprocessing result.

[0174] In an embodiment, when implementing the step of determining the drug weight by generating a drug efficacy weight system of five types of anti-infective drugs according to the preprocessing result combined with statistical regression analysis and expert scoring method, the processor 502 specifically implements the following steps:

[0175] Based on the clinical symptoms, microbiological culture results, and inflammation indicators of the patient in the preprocessing result, a health status-drug use time series during hospitalization is created; according to the health status-drug use time series during hospitalization, the state transition probability and risk ratio of using a specific drug in different states are calculated, and the influence of the drug on the patient's health state transition is evaluated to obtain the risk ratio of the drug; based on the risk ratio of the drug and combined with expert opinion adjustment, the basic classification drug efficacy weight of each antibacterial drug is determined to determine the drug weight.

[0176] In an embodiment, when implementing the step of creating a health status-drug use time series during hospitalization based on the clinical symptoms, microbiological culture results, and inflammation indicators of the patient in the preprocessing result, the processor 502 specifically implements the following steps:

[0177] The hospitalization state of the patient in the preprocessing result is divided into five levels: healthy, suspected, infected, improved, and recovered; the patient's hospitalization record in the preprocessing result is divided by day, and the daily health status and the anti-infective drugs used and their dosages are marked, and combined with the divided levels, a health status-drug use time series during hospitalization is formed.

[0178] In an embodiment, the processor 502, in implementing the step of calculating the state transition probability and risk ratio of using a specific drug in different states according to the health state-drug time series during the hospitalization period, evaluates the influence of the drug on the state transition of the patient's health to obtain the risk ratio of the drug, specifically implements the following steps:

[0179] According to the health state-drug time series during the hospitalization period, the probability of all patients transferring from one health state to another state under the condition of using a specific drug is calculated and formed into a multi-dimensional transition probability matrix. The risk ratio of each drug leading to the transition from health to infection state is calculated according to the multi-dimensional transition probability matrix to evaluate the influence of the drug on the state transition to obtain the risk ratio of the drug.

[0180] In an embodiment, the processor 502, in implementing the step of determining the basic classification drug efficacy weight of each antibacterial drug based on the risk ratio of the drug and adjusting in combination with expert opinions to determine the drug weight, specifically implements the following steps:

[0181] According to the mapping of the risk ratio of the drug to a preset weight range, the basic weight of each drug is determined. According to the drug efficacy, preventive effect, combined drug effect, and drug resistance, the basic weight of each drug is adjusted to determine the efficacy weight of each antibacterial drug.

[0182] In an embodiment, the processor 502, in implementing the step of calculating the daily infection risk index of the patient in real time according to the drug weight and invasive operation factors through a rule engine, specifically implements the following steps:

[0183] The pre-processing results are associated by the hospital ID to obtain an association result. Whether the hospitalization time of the patient intersects with the antibacterial drug use time is judged according to the association result to determine whether the antibacterial drug is used on the day to obtain a judgment result. The patient information using the antibacterial drug on the day but excluding the skin test and non-discharge drug cases is screened according to the judgment result. The patient groups are subdivided according to whether the time since the patient's admission is more than 48 hours and whether the patient is admitted again within 48 hours after the last discharge. The change of the patient's anti-infective drug use is analyzed, and the infection risk index is adjusted according to the drug weight. Whether there is a pathogen detected matching the anti-infective drug used on the day is checked in combination with the pathogen detection information, and the infection risk index is adjusted accordingly. For patients who cannot be successfully matched with the anti-infective drug used, the previous day's pathogen detection is classified, and the infection risk index is adjusted accordingly. For other cases, the infection risk index is adjusted again based on the key pathogen submission on the day. For patients whose risk index has not changed, whether there is a surgery or invasive operation on the day is considered to determine whether to adjust the infection risk index to obtain the daily infection risk index of the patient.

[0184] In an embodiment, the processor 502 implements the following steps when implementing the step of classifying the levels according to the daily infection risk index of the patient and grading the early warning and starting the corresponding intervention measures:

[0185] According to the daily infection risk index of the patient, the infection risk is dynamically classified into four levels of no risk, mild risk, moderate risk and high risk, which correspond to different clinical significance and monitoring requirements; the daily infection risk index of the patient is automatically calculated and compared with the previous day's level, when the risk level is found to be increased or continuously high, the medical staff is warned through different channels and corresponding level intervention measures are taken.

[0186] 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. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0187] It can be understood by those skilled in the art that all or part of the processes in the method of 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 of the method.

[0188] 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 perform the following steps:

[0189] Obtaining diagnosis and treatment feature data of a patient during hospitalization; cleaning and standardizing drug names involved in the diagnosis and treatment feature data to standard terms, and retaining discharge medication identification to obtain a preprocessing result; generating a drug efficacy weight system of five types of anti-infective drugs according to the preprocessing result combined with statistical regression analysis and expert scoring method to determine drug weights; calculating a daily infection risk index of the patient in real time through a rule engine according to the drug weights and invasive operation factors; grading and early warning according to the daily infection risk index of the patient, and starting corresponding intervention measures.

[0190] The diagnosis and treatment feature data includes basic information, anti-bacterial drug use records, invasive operation data, and pathogenic detection results.

[0191] In an embodiment, when the processor executes the computer program to implement the step of cleaning and standardizing drug names involved in the diagnosis and treatment feature data to standard terms, and retaining discharge medication identification to obtain a preprocessing result, the following steps are specifically implemented:

[0192] Extracting and retaining drug names from the order text involved in the diagnosis and treatment feature data, and marking as incomplete data for cases that cannot be identified; converting drug trade names or aliases in the order text involved in the diagnosis and treatment feature data to corresponding generic names for standardized representation; removing dose and specification information in drug descriptions involved in the diagnosis and treatment feature data, and retaining only active ingredient names; mapping drug names in special cases in the diagnosis and treatment feature data through manually maintained exception word table processing to obtain a preprocessing result.

[0193] In an embodiment, when the processor executes the computer program to implement the step of generating a drug efficacy weight system of five types of anti-infective drugs according to the preprocessing result combined with statistical regression analysis and expert scoring method to determine drug weights, the following steps are specifically implemented:

[0194] Based on the clinical symptoms, microbiological culture results, and inflammation indicators of the patient in the preprocessing result, a health status-drug use time series during hospitalization is created; state transition probabilities and risk ratios of using specific drugs in different states are calculated according to the health status-drug use time series during hospitalization, and the influence of drugs on the health state transition of the patient is evaluated to obtain the risk ratio of the drug; based on the risk ratio of the drug and combined with expert opinion adjustment, the basic classification drug efficacy weight of each anti-bacterial drug is determined to determine the drug weight.

[0195] In an embodiment, when the processor executes the computer program to implement the step of creating a health status-drug use time series during hospitalization based on the clinical symptoms, microbiological culture results, and inflammation indicators of the patient in the preprocessing result, the following steps are specifically implemented:

[0196] The hospitalization state of the patient in the preprocessed result is divided into five levels of health, suspected, infection, improvement, and recovery; the hospitalization record of the patient in the preprocessed result is segmented in days, and the health state and the anti-infective drug and its dose used each day are marked, and combined with the divided levels, a health state-drug time series during hospitalization is formed.

[0197] In an embodiment, when the processor executes the computer program to realize the calculation of the state transition probability and risk ratio of using a specific drug in different states according to the health state-drug time series during hospitalization, and evaluates the influence of the drug on the health state transition of the patient to obtain the risk ratio of the drug, the following steps are specifically implemented:

[0198] According to the health state-drug time series during hospitalization, the probability of all patients transferring from one health state to another health state when using a specific drug is counted and calculated to form a multi-dimensional transition probability matrix; according to the multi-dimensional transition probability matrix, the risk ratio of each drug leading to the transition from a health state to an infection state is calculated to evaluate the influence of the drug on the state transition to obtain the risk ratio of the drug.

[0199] In an embodiment, when the processor executes the computer program to realize the determination of the basic classification efficacy weight of each antibacterial drug based on the risk ratio of the drug and combined with expert opinions adjustment to determine the drug weight step, the following steps are specifically implemented:

[0200] According to the risk ratio of the drug mapped to a preset weight range, the basic weight of each drug is determined; according to the efficacy, preventive effect, combined drug effect, and drug resistance of the drug, the basic weight of each drug is adjusted to determine the efficacy weight of each antibacterial drug.

[0201] In an embodiment, when the processor executes the computer program to realize the calculation of the daily infection risk index of the patient according to the drug weight and the invasive operation factor through a rule engine in real time, the following steps are specifically implemented:

[0202] The pre-processing result is associated by hospitalization ID to obtain an association result; whether the hospitalization time of the patient intersects with the time of use of the antibacterial drug is judged according to the association result, so as to determine whether the antibacterial drug is used on the day, to obtain a judgment result; patient information using the antibacterial drug on the day is screened out according to the judgment result, excluding skin test and non-discharge drug cases; patient groups are subdivided according to whether the time since the patient is hospitalized is more than 48 hours and whether the patient is hospitalized again within 48 hours after the last discharge; the change of the patient's use of the anti-infective drug is analyzed, and the infection risk index is adjusted according to the drug weight; combined with the pathogenic detection information, whether a pathogen matching the used anti-infective drug is detected on the day is checked, and the infection risk index is adjusted accordingly; for patients who cannot be successfully matched with the used anti-infective drug, classification is made according to the pathogen detection of the previous day, and the infection risk index is adjusted accordingly; for other cases, a re-evaluation is made based on the key pathogen submission on the day, and the infection risk index is adjusted; for patients whose risk index has not changed, whether the patient has surgery or invasive operation on the day is considered to determine whether to adjust the infection risk index, to obtain the daily infection risk index of the patient.

[0203] In an embodiment, when the processor executes the computer program to implement the step of grading according to the daily infection risk index of the patient and grading early warning and starting corresponding intervention measures, the following steps are specifically implemented:

[0204] According to the daily infection risk index of the patient, the infection risk is dynamically divided into four grades of no risk, mild risk, moderate risk and high risk, and different clinical significance and monitoring requirements are corresponded; the daily infection risk index of the patient is automatically calculated and compared with the grade of the previous day, and when it is found that the risk grade is raised or continuously high, the medical staff is warned through different channels and corresponding level intervention measures are taken.

[0205] 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.

[0206] Those skilled in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by 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 the examples have been described in the above description in general. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0207] In several embodiments provided by the present application, it should be understood that the disclosed system and method can be implemented in other manners. For example, the system embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division, and there can be another division manner in actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0208] The steps in the method embodiments of the present application can be adjusted, combined and deleted according to actual needs. The units in the system 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.

[0209] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a storage medium. Based on such understanding, the technical solutions of the present application essentially or 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 several 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 methods described in the embodiments of the present application.

[0210] The above describes only specific embodiments 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 in 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 monitoring hospital-acquired infection risk based on the trajectory of anti-infective drug use, characterized in that: include: Obtain the patient's diagnosis and treatment characteristics data during hospitalization; Cleaning and standardizing the drug names involved in the diagnosis and treatment characteristic data to standard terms, and retaining the discharge medication identification to obtain pretreatment results; Based on the preprocessing results, combined with statistical regression analysis and expert scoring method, a drug efficacy weight system of five categories of anti-infective drugs is generated to determine drug weights; Calculate the patient's daily infection risk index in real time through a rule engine based on the drug weights and invasive procedure factors; The patients are divided into levels according to their daily infection risk index, and graded warnings are issued to initiate corresponding intervention measures.

2. The method for monitoring hospital infection risk based on anti-infective drug usage trajectory according to claim 1, characterized in that: The diagnosis and treatment characteristic data include basic information, antimicrobial drug usage records, invasive operation data and pathogen detection results.

3. The method for monitoring hospital infection risk based on anti-infective drug usage trajectory according to claim 2, characterized in that: The cleaning and standardization of the drug names involved in the diagnosis and treatment characteristic data to standard terms, and retaining the discharge medication identification to obtain the pre-processing results, include: Extracting and retaining drug names from the medical order text involved in the diagnosis and treatment feature data, and marking unrecognizable drug names as incomplete data; Convert the drug trade name or alias in the medical order text involved in the diagnosis and treatment characteristic data into the corresponding common name for standardized representation; Remove the dosage and strength information from the drug description involved in the diagnosis and treatment characteristic data, and only retain the name of the active ingredient; The names of medicines in special cases in the diagnosis and treatment feature data are mapped through a manually maintained exception word list to obtain a preprocessing result.

4. The method for monitoring hospital infection risk based on anti-infective drug usage trajectory according to claim 1, characterized in that: The pre-processing results are combined with statistical regression analysis and expert scoring method to generate a drug efficacy weight system for five categories of anti-infective drugs to determine drug weights, including: Creating a health status-drug use time series during hospitalization based on the patient's clinical symptoms, microbial culture results, and inflammatory indicators in the pretreatment results; Calculating the state transition probability and hazard ratio when using a specific drug in different states based on the health status-drug use time series during the hospitalization period, and evaluating the impact of the drug on the patient's health status transition to obtain the drug's hazard ratio; Based on the risk ratio of the drugs and adjusted with expert opinions, the basic classification efficacy weight of each antimicrobial drug was determined to determine the drug weight.

5. The method for monitoring hospital infection risk based on anti-infective drug usage trajectory according to claim 4, characterized in that: The method of creating a health status-medication time series during hospitalization based on the patient's clinical symptoms, microbial culture results, and inflammatory indicators in the pretreatment results includes: The hospitalization status of the patients in the pre-treatment results is divided into five levels: healthy, suspected, infected, improved, and recovered; The patient hospitalization records in the preprocessing results are divided into days, and the daily health status and the anti-infective drugs used and their dosages are marked. Combined with the divided levels, a health status-drug use time series during hospitalization is formed.

6. The method for monitoring hospital infection risk based on anti-infective drug usage trajectory according to claim 4, characterized in that: The state transition probability and risk ratio of using a specific drug under different states are calculated based on the health status-drug use time series during the hospitalization period, and the impact of the drug on the patient's health status transition is evaluated to obtain the drug risk ratio, including: According to the health status-drug use time series statistics during the hospitalization, the probability of all patients transitioning from one health status to another when using a specific drug is calculated to form a multi-dimensional transition probability matrix; The risk ratio of each drug causing the transition from a healthy state to an infected state is calculated based on the multidimensional transition probability matrix, and the effect of the drug on the state transition is evaluated to obtain the risk ratio of the drug.

7. The method for monitoring hospital infection risk based on anti-infective drug usage trajectory according to claim 4, characterized in that: The risk ratio of the drug is adjusted based on expert opinions to determine the basic classification efficacy weight of each antimicrobial drug to determine the drug weight, including: Determining a base weight for each drug based on mapping the drug's hazard ratio to a preset weight range; The basic weight of each drug is adjusted according to its efficacy, preventive effect, combination effect and drug resistance to determine the efficacy weight of each type of antimicrobial drug.

8. The method for monitoring hospital infection risk based on anti-infective drug usage trajectory according to claim 1, characterized in that: The daily infection risk index of the patient is calculated in real time by a rule engine based on the drug weight and invasive operation factors, including: Associating the preprocessing results with the hospitalization ID to obtain an associated result; Judging whether the patient's hospitalization time and the time of antibiotic use have an intersection based on the association result, so as to determine whether antibiotics were used on that day, and thus obtain a judgment result; According to the determination results, the information of patients who used antimicrobial drugs on the same day but excluded skin tests and those who took the drugs before discharge was screened; Patient groups were subdivided based on whether the patient had been hospitalized for more than 48 hours and whether they had been readmitted within 48 hours of their last discharge; Analyze changes in patients' use of anti-infective drugs and adjust the infection risk index based on drug weights; Combined with the information from pathogen detection, check whether any pathogens matching the anti-infective drugs used are detected on that day, and adjust the infection risk index accordingly; for patients whose anti-infective drugs cannot be successfully matched, classify them according to the pathogen detection results of the previous day, and adjust the infection risk index accordingly; for other situations, make another assessment based on the key pathogens sent for inspection on that day, and adjust the infection risk index; for patients whose risk index has not changed, consider whether they have undergone surgery or invasive procedures on that day, and decide whether to adjust the infection risk index to obtain the patient's daily infection risk index.

9. The method for monitoring hospital infection risk based on anti-infective drug usage trajectory according to claim 1, characterized in that: The patient's daily infection risk index is divided into levels, and graded warnings are issued to initiate corresponding intervention measures, including: Based on the patient's daily infection risk index, the infection risk is dynamically divided into four levels: no risk, mild risk, moderate risk, and high risk, corresponding to different clinical significance and monitoring needs; The daily infection risk index of the patient is automatically calculated and compared with the level of the previous day. When it is found that the risk level increases or remains high, an early warning is issued to medical staff through different channels, and intervention measures of corresponding levels are taken.

10. A hospital infection risk monitoring system based on the use trajectory of anti-infective drugs is characterized by: include: A data acquisition unit, used to acquire the patient's diagnosis and treatment characteristic data during hospitalization; A pre-processing unit, for cleaning and standardizing the drug names involved in the diagnosis and treatment characteristic data to standard terms, and retaining the discharge medication identification to obtain a pre-processing result; A weight determination unit, configured to generate a drug efficacy weight system for five categories of anti-infective drugs based on the preprocessing results in combination with statistical regression analysis and expert scoring method to determine drug weights; an index determination unit, configured to calculate the patient's daily infection risk index in real time through a rule engine based on the drug weights and invasive procedure factors; The early warning unit is used to classify the patients according to their daily infection risk index, issue graded early warnings, and initiate corresponding intervention measures.

Images