A precision drug-resistant bacteria administration assistance analysis method and system

Abstract

The application belongs to the technical field of medical information, and discloses a precision drug administration assisting analysis method and system for drug-resistant bacteria, which automatically collects and integrates clinical data of patients and pathogen drug sensitivity reports to generate a structured electronic tracking intervention record table in response to a microorganism submission event; based on preset distributed node rules, multi-role real-time collaborative tracking and permission-controlled interaction of pharmacists, clinicians and hospital infection officers are realized; by integrating a standardized auxiliary evaluation panel and a pharmacy knowledge base, the pharmacists are guided to perform structured and standardized evaluation on pathogenicity and rational drug use; with the aid of preset intelligent contract rules, batch analysis is performed on the archived cases, and a standardized report and a drug use warning are automatically generated, and a management closed loop is formed through directional pushing. The application realizes the full-process standardization and intelligentization from information acquisition, collaborative intervention to data analysis and utilization, and significantly improves the efficiency and quality of scientific management of antibacterial drugs.

Description

Technical Field

[0001] This invention belongs to the field of medical information technology and discloses a method and system for assisting in the precise administration of drugs to resistant bacteria. Background Technology

[0002] Currently, in medical institutions at all levels, pharmacists typically rely on manual or basic spreadsheet recording of relevant information, and collaborate with the laboratory, infection control and clinical departments for communication and management, in order to monitor the medication process, intervene in irrational use and prevent treatment-related adverse events.

[0003] However, existing practices face numerous technical challenges and bottlenecks. First, the recording tools relied upon for intervention tracking are often poorly designed and lack deep integration with hospital information systems. This necessitates manual entry of key information such as patient infection indicators and medication records, leading to inefficiency and a high risk of errors. The lack of standardized table design makes it difficult to clearly reflect dynamic changes in treatment and fails to effectively assist pharmacists in quickly identifying pathogens, assessing the rationality of medication use, and ensuring treatment safety. Second, the multidisciplinary collaboration process is loose, with drug-resistant bacteria information transmitted through non-real-time methods such as office systems, resulting in delayed pharmacist intervention. Unclear responsibilities and collaboration interfaces between departments, coupled with a lack of institutionalized coordination and supervision mechanisms, affect the timeliness and effectiveness of interventions. Finally, the analysis and utilization of accumulated intervention record data are in their early stages. Analysis reports are written haphazardly, medication warnings are issued through limited channels with low awareness, making it difficult to form an effective management loop to continuously improve clinical practice and enhance relevant quality control indicators.

[0004] Therefore, there is an urgent need to build a drug-resistant bacteria tracking and intervention platform that is deeply integrated with hospital information systems, has standardized processes, and can effectively support real-time multidisciplinary collaboration. This platform needs to address the problems of information fragmentation, inefficiency, and crude analysis in the existing model. Through technological means, it should achieve automatic data collection, intelligent assisted evaluation, standardized collaborative intervention, and in-depth data utilization, thereby systematically and systematically improving the quality and efficiency of pharmacists' tracking and intervention work, and truly ensuring the scientific management of antimicrobial drugs. Summary of the Invention

[0005] To achieve the above objectives, this application provides the following technical solution:

[0006] According to a first aspect of the present invention, the present invention claims protection for a method for assisting in the precise administration of drugs-resistant bacteria, the method comprising the following steps:

[0007] S1, in response to the submission of microbial specimens by hospitalized patients, obtains the patient's test request form identifier and patient identification identifier in real time;

[0008] S2, Based on the patient's identity identifier, simultaneously extract the patient's clinical dataset;

[0009] S3, when a pathogen drug susceptibility report containing drug resistance or special drug resistance markers is generated, an early warning event is triggered based on the test request form identifier, and the pathogen drug susceptibility report is associated and bound with the clinical dataset to generate an initial intervention case data package;

[0010] S4, call the standardized tracking intervention record form template, automatically fill each data item in the initial intervention case data package into the predefined fields of the standardized tracking intervention record form template, and generate an electronic tracking intervention record form;

[0011] S5, the electronic tracking intervention record form is assigned to the pharmacist's work queue, and based on the preset distributed node rules, the access and editing permissions of the electronic tracking intervention record form are simultaneously granted to the designated physician node of the clinical department to which the patient belongs and the designated hospital infection control specialist node of the hospital infection management department.

[0012] S6, after the electronic tracking intervention record form enters the editable state, the multi-role collaborative tracking stage is started;

[0013] S7. After the patient's infection treatment cycle ends, the pharmacist triggers a case closure operation in the electronic tracking intervention record table, marks the treatment outcome, changes the status of the electronic tracking intervention record table to a read-only archived status, and stores it in the tracking intervention case database.

[0014] S8, perform batch analysis on the archived record table in the tracking intervention case database, and extract key analysis dimension data from the archived record table through preset smart contract rules. The smart contract rules define the fields, formats and logical judgment conditions for data extraction.

[0015] Furthermore, the design and application of the standardized tracking intervention record form template described in step S4 includes:

[0016] The standardized tracking intervention record form template is in spreadsheet format, and the interface is divided into multiple interlocking logical areas;

[0017] The patient basic information data area includes fields for ward, hospital number, name, age, and diagnosis information;

[0018] The drug-resistant bacteria detection details data area includes fields such as submission date, specimen type, pathogen name, drug resistance mechanism, and key drug susceptibility results, and also has a drop-down selection field for preliminary pathogenicity assessment, with options for suspected pathogenic bacteria, colonizing bacteria, or contaminating bacteria;

[0019] The infection dynamic indicator data area dynamically links to the patient's time-series data through an interface, displaying the changes in body temperature, white blood cell count, C-reactive protein, and procalcitonin over time in the form of charts, and providing text fields for linking or summarizing infection-related imaging and symptom descriptions.

[0020] The details data area for antibiotic use is extracted from the medical order system. The list displays the name, dosage, frequency, route of administration, start and end time of antibiotics used currently and in the past. It also includes a preliminary judgment field for the rationality of drug use, with options of rational, needing adjustment, or unreasonable.

[0021] The pharmacist evaluation and suggestion data area contains several structured input components for recording pharmacists' analysis of the match between infection and medication, specific medication adjustment suggestions, adjustment basis, and key points of medication education;

[0022] The treatment process tracking log data area is a log panel arranged in reverse chronological order. It automatically records and displays key operations and textual notes made by pharmacists, physicians, and infection control specialists regarding the patient's condition and treatment, forming a complete audit trail of the treatment process.

[0023] Furthermore, step S6 also includes:

[0024] S61a. The pharmacist activates the pathogenicity auxiliary assessment panel, which is linked to the drug-resistant bacteria detection details data area and the infection dynamic indicator data area, in the interface of the electronic tracking intervention record form;

[0025] S61b. The pathogenicity auxiliary evaluation panel first calls the preset pathogen knowledge base, and automatically loads the background probability information of the microorganism as a common pathogen, conditional pathogen or common contaminant based on the microbial name in the pathogen drug sensitivity report, and displays it in the reference information bar of the panel.

[0026] S61c. The pathogenicity auxiliary assessment panel provides standardized assessment factor checkboxes for pharmacists to select. The assessment factors include at least the presence of clear symptoms or signs at the site of infection, the pathogen being consistent with the spectrum of common pathogens at the site of infection, inflammatory markers showing a significant increase before and after detection or changing synchronously with the course of infection, imaging results supporting active infection, and the patient not having used effective antimicrobial drugs against the bacteria before this detection.

[0027] S61d. The pathogenicity auxiliary assessment panel has a built-in rule-based logic judgment engine. The engine runs a predefined set of judgment rules according to the combination of judgment factors selected by the pharmacist. The set of judgment rules is defined in the form of judgment tendency if the factor combination is selected.

[0028] S61e. The logic judgment engine outputs the running results in a non-mandatory suggestion form, which is displayed as a system auxiliary evaluation prompt;

[0029] S61f. The pharmacist, referring to the system's auxiliary evaluation prompts and combining their own professional judgment, selects one of the following from the preliminary pathogenicity assessment drop-down field: suspected pathogenic bacteria, colonizing bacteria, or contaminating bacteria, and completes manual confirmation and entry.

[0030] Furthermore, step S6 also includes:

[0031] S62a, the pharmacist activates the auxiliary evaluation panel for rational drug use, which is linked to the details data area for antimicrobial drug use and the drug sensitivity results, in the interface of the electronic tracking intervention record form;

[0032] S62b, the auxiliary evaluation panel for rational drug use automatically compares the currently used antimicrobial drug with the drug susceptibility results in the pathogen drug susceptibility report, generates an initial drug susceptibility matching report, and identifies sensitive, intermediate, resistant or untested.

[0033] S62c, the medication rationality auxiliary evaluation panel accesses the integrated pharmaceutical knowledge base to obtain the standard dosing regimen information of the currently used drug, including the recommended dose range, dosing interval, infusion requirements and adjustment strategies when major organ dysfunction occurs, and displays the adjustment strategies in parallel with the extracted actual medical order data;

[0034] S62d, the medication rationality assessment panel guides pharmacists to conduct step-by-step assessments: First, for drug selection, it provides options to evaluate the matching degree between the current medication and the drug sensitivity results; second, for the dosing regimen, it provides options to evaluate the appropriateness of the dosage, route, and frequency; third, for comprehensive evaluation, it provides options to evaluate the rationality of the overall treatment plan.

[0035] S62e, the auxiliary evaluation panel for rational drug use automatically generates an evaluation draft based on the pharmacist's selection in the aforementioned step-by-step evaluation and the recent inflammatory index trend read from the infection dynamic indicator data area. The draft includes: drug selection based on / deviating from drug sensitivity results, the current dosing regimen compared to the standard, and recent infection indicators showing that the treatment response is effective / ineffective / uncertain.

[0036] S62f, the pharmacist reviews and modifies the evaluation draft to form the final pharmacist evaluation text, fills it into the pharmacist evaluation and suggestion data area, and simultaneously selects the corresponding overall evaluation level in the preliminary judgment field of drug rationality.

[0037] Furthermore, in step S5, the specific implementation is as follows:

[0038] In the access control module of the medical information system, an independent application role matrix is ​​defined for the drug-resistant bacteria tracking and intervention platform;

[0039] The application role matrix includes at least three roles: pharmacist, attending physician, and infection control specialist. Each role is configured with read and write permissions for different data areas in the electronic tracking intervention record table.

[0040] When the initial intervention case data package is generated, the system automatically instantiates the application role matrix based on the relationship between the patient and the physician and the rules of the infection control specialist's assigned area, and binds the electronic tracking intervention record form to the instantiated role permissions.

[0041] Furthermore, the method ensures privacy and security during data flow, specifically in the following ways:

[0042] In steps S1 to S3, the obtained patient identification is transmitted in the hospital intranet through a secure data bus, and the secure data bus adopts a transport layer encryption protocol.

[0043] The electronic tracking intervention record table generated in step S4 is stored in a logically independent database partition. Access to this partition is controlled by the application role matrix and is physically or logically isolated from the production database.

[0044] In the multi-role collaborative tracking stage described in step S6, all add, delete, and modify operations on the record table are recorded in real time to an immutable operation audit log, which records the operator's identity, time, and the state before and after the modification.

[0045] Furthermore, step S8 specifically includes:

[0046] The smart contract rules are defined in the form of executable scripts and deployed on the server side of the tracking intervention case database;

[0047] Rule Script 1: Define the data fields and aggregation calculation methods required to extract pathogen distribution, specimen source composition, and drug resistance rate statistics within a specific period from the drug resistance bacteria detection details data area of ​​the archived record table;

[0048] Rule Script 2: Define the logic for extracting and analyzing the treatment outcome data of irrational drug use cases from the initial judgment field of drug rationality and the treatment outcome field;

[0049] Rule Script 3: Define a fixed template for generating report documents, including the format and content filling rules for the cover, abstract, data analysis charts, typical case analysis, key drug warnings, and improvement suggestions;

[0050] When a periodic analysis task is triggered, the system executes rule scripts one, two, and three in sequence, extracts the standardized dataset processed by the rules from the database, and populates it into the fixed template, ultimately generating a composite document containing text, tables, and charts.

[0051] Furthermore, the method achieves standardization and real-time sharing of pharmacist medication guidance through system collaboration and data integration, specifically manifested in:

[0052] The electronic tracking intervention record form generated in step S4 serves as a unique and synchronized information carrier for all participating nodes, ensuring that pharmacists, physicians, and infection control specialists make decisions and communicate based on real-time updated data.

[0053] The standardized evaluation options and structured text input in the pharmacist evaluation and suggestion data area, as well as the suggestions generated through the auxiliary evaluation panel, form a unified evaluation discourse system.

[0054] In the multi-role collaboration stage of step S6, the pharmacist's evaluation and suggestions are presented to the attending physician node in real time, and the physician's feedback and subsequent medical order adjustments are also recorded in real time.

[0055] All archived records form a searchable case knowledge base, allowing other pharmacists to refer to the evaluations and recommendations in historical cases when dealing with the same drug-resistant bacteria or clinical situations.

[0056] Furthermore, the comparative evaluation of system performance includes the implementation steps for the following evaluation dimensions:

[0057] Operational evaluation steps: After design and deployment, a structured questionnaire is distributed to pharmacists and physicians who use the system. The questionnaire questions focus on the clarity of the system interface logic, the convenience of data acquisition, the efficiency of record filling, and the smoothness of multi-role collaborative processes. Feedback is collected and the overall usability score is calculated.

[0058] Effectiveness evaluation steps: Extract treatment outcome data of all closed cases within a specific period after the system application from the tracking intervention case database, compare it with historical case data tracked by traditional methods in the same period before the system application, and statistically compare the difference in the proportion of cases marked as cured or effective in infection treatment between the two groups of data.

[0059] Safety evaluation steps: From the treatment process tracking log of the electronic tracking intervention record form, text mining technology is used to identify and count the number of times recorded adverse drug reaction events or treatment-related adverse events are mentioned; the incidence rates of the above events are compared before and after the system application in the same period.

[0060] Efficiency evaluation steps: Obtain comparable case groups of patients with drug-resistant bacterial infections before and after system application, and compare and analyze the median changes in key efficiency indicators such as average length of hospital stay and average duration of antibiotic use.

[0061] Evaluation steps for quality control indicators: retrieve hospital pharmaceutical quality control data, compare the numerical trends of the intensity of antimicrobial drug use, microbial testing rate, and antimicrobial drug prescription qualification rate in multiple consecutive statistical periods before and after the system application, and analyze the direction and magnitude of their changes.

[0062] According to a second aspect of the present invention, the present invention claims protection for a precision drug delivery auxiliary analysis system for drug-resistant bacteria, comprising:

[0063] One or more processors;

[0064] A memory storing one or more programs, which, when executed by one or more processors, enable the one or more processors to implement the aforementioned method for precise drug delivery and auxiliary analysis of drug-resistant bacteria.

[0065] This invention discloses a method and system for precise drug administration analysis of drug-resistant bacteria. By responding to microbial testing events, it automatically collects and integrates patient clinical data and pathogen drug susceptibility reports to generate a structured electronic tracking and intervention record form. Based on preset distributed node rules, it enables real-time collaborative tracking and access-controlled interaction among pharmacists, clinicians, and infection control specialists. By integrating a standardized auxiliary evaluation panel and a pharmaceutical knowledge base, it guides pharmacists to conduct structured and standardized assessments of pathogenicity and the rationality of drug use. Using pre-set smart contract rules, it performs batch analysis of archived cases, automatically generating standardized reports and medication warnings, and forming a management closed loop through targeted push notifications. This invention achieves standardization and intelligence throughout the entire process from information acquisition and collaborative intervention to data analysis and utilization, significantly improving the efficiency and quality of scientific management of antimicrobial drugs. Attached Figure Description

[0066] Figure 1 This is a flowchart illustrating the workflow of a precision drug delivery auxiliary analysis method for drug-resistant bacteria, as claimed in an embodiment of the present invention.

[0067] Figure 2 This is a second workflow diagram of an auxiliary analysis method for precise drug administration of drug-resistant bacteria, as claimed in an embodiment of the present invention.

[0068] Figure 3 This is a third workflow diagram of an auxiliary analysis method for precise drug administration of drug-resistant bacteria, as claimed in an embodiment of the present invention.

[0069] Figure 4 The fourth flowchart is shown in the embodiment of the present invention for a method for assisting in the precise administration of drugs to resistant bacteria. Detailed Implementation

[0070] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0071] The terms "first," "second," and "third" in this application are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of those features. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. All directional indications in the embodiments of this application, such as up, down, left, right, front, back, etc., are only used to explain the relative positional relationships and movements between components in a specific orientation as shown in the accompanying drawings. If the specific orientation changes, the directional indications will change accordingly. Furthermore, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0072] References to embodiments herein mean that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0073] According to the first embodiment of the present invention, referring to Figure 1 This invention claims protection for a method for precise drug administration analysis of drug-resistant bacteria, the method comprising the following steps:

[0074] S1, in response to the submission of microbial specimens by hospitalized patients, obtains the patient's test request form identifier and patient identification identifier in real time;

[0075] S2, Based on the patient's identity identifier, simultaneously extract the patient's clinical dataset;

[0076] S3, when a pathogen drug susceptibility report containing drug resistance or special drug resistance markers is generated, an early warning event is triggered based on the test request form identifier, and the pathogen drug susceptibility report is associated and bound with the clinical dataset to generate an initial intervention case data package;

[0077] S4, call the standardized tracking intervention record form template, automatically fill each data item in the initial intervention case data package into the predefined fields of the standardized tracking intervention record form template, and generate an electronic tracking intervention record form;

[0078] S5, the electronic tracking intervention record form is assigned to the pharmacist's work queue, and based on the preset distributed node rules, the access and editing permissions of the electronic tracking intervention record form are simultaneously granted to the designated physician node of the clinical department to which the patient belongs and the designated hospital infection control specialist node of the hospital infection management department.

[0079] S6, after the electronic tracking intervention record form enters the editable state, the multi-role collaborative tracking stage is started;

[0080] S7. After the patient's infection treatment cycle ends, the pharmacist triggers a case closure operation in the electronic tracking intervention record table, marks the treatment outcome, changes the status of the electronic tracking intervention record table to a read-only archived status, and stores it in the tracking intervention case database.

[0081] S8, perform batch analysis on the archived record table in the tracking intervention case database, and extract key analysis dimension data from the archived record table through preset smart contract rules. The smart contract rules define the fields, formats and logical judgment conditions for data extraction.

[0082] This embodiment is presented as a comprehensive overview of its application in the infectious disease department of a tertiary-level Class A hospital.

[0083] A patient, Mr. Chen, with cirrhosis and ascites, was admitted to the hospital due to fever and abdominal pain, with spontaneous bacterial peritonitis clinically suspected. The attending physician issued a request for ascites culture and drug sensitivity testing through the hospital information system (HIS). After receiving the specimen, the laboratory information system (LIS) generated a unique test request form identifier, which was linked to the patient's hospital number and patient identification.

[0084] When the LIS completes the culture and identification of Escherichia coli 3GCR-EC resistant to third-generation cephalosporins, and the drug susceptibility report automatically marks it as resistant, the system's built-in early warning rule is triggered. The HIS-LIS hospital information system-laboratory information system data interface accurately captures this early warning event based on the test request form identifier. The system then automatically extracts Chen's complete clinical dataset since admission from the HIS, using the patient's hospital number as an index. This dataset includes: demographic information, diagnosis of cirrhosis, daily temperature records, complete blood count (white blood cell count), neutrophil percentage, inflammatory markers (C-reactive protein), procalcitonin sequence, abdominal ultrasound report summary, and medical records regarding abdominal pain and distension.

[0085] The system intelligently associates and binds the 3GCR-EC drug susceptibility reports with this clinical dataset, packaging them into a structured initial intervention case data package. This data package is then sent to the drug-resistant bacteria tracking and intervention platform. The platform calls a predefined standardized electronic template for tracking and intervention record forms, automatically filling in the corresponding preset fields of the template with the various contents of the data package—such as patient information, drug susceptibility results, and inflammatory marker data points—to form an initialized electronic tracking and intervention record form.

[0086] According to the preset rules, the system automatically assigned this record sheet to the work queue of Zhang, the clinical pharmacist in charge of the infectious disease department that day. At the same time, based on the distributed node rules, the system automatically identified Chen's attending physician as Li and the specialist in charge of the area of ​​the hospital infection control department as Wang, and granted the restricted access and editing permissions of the record sheet to these two nodes simultaneously.

[0087] Pharmacist Zhang opened the record sheet and entered the multi-role collaborative tracking phase. She reviewed the integrated data and entered a preliminary evaluation in the record sheet. At the same time, Physician Li received a pop-up notification on his computer and could immediately view the pharmacist's evaluation and provide feedback. Infection control specialist Wang could also view the entire process simultaneously.

[0088] After a week of treatment and follow-up, Chen's infection was brought under control. Pharmacist Zhang marked the treatment outcome as cured in the record sheet and performed the case closure operation. The record sheet was then switched to read-only status and archived in a dedicated follow-up intervention case database.

[0089] Every quarter, the database backend automatically starts an analysis task, and the pre-set smart contract script begins to run. From all archived record tables, according to the rules defined in the script, it extracts fields and calculation logic such as the detection department of all Escherichia coli, specimen source, and resistance rate to ceftriaxone, and extracts key analysis dimension data in batches to prepare for the subsequent generation of management reports.

[0090] Furthermore, the design and application of the standardized tracking intervention record form template described in step S4 includes:

[0091] The standardized tracking intervention record form template is in spreadsheet format, and the interface is divided into multiple interlocking logical areas;

[0092] The patient basic information data area includes fields for ward, hospital number, name, age, and diagnosis information;

[0093] The drug-resistant bacteria detection details data area includes fields such as submission date, specimen type, pathogen name, drug resistance mechanism, and key drug susceptibility results, and also has a drop-down selection field for preliminary pathogenicity assessment, with options for suspected pathogenic bacteria, colonizing bacteria, or contaminating bacteria;

[0094] The infection dynamic indicator data area dynamically links to the patient's time-series data through an interface, displaying the changes in body temperature, white blood cell count, C-reactive protein, and procalcitonin over time in the form of charts, and providing text fields for linking or summarizing infection-related imaging and symptom descriptions.

[0095] The details data area for antibiotic use is extracted from the medical order system. The list displays the name, dosage, frequency, route of administration, start and end time of antibiotics used currently and in the past. It also includes a preliminary judgment field for the rationality of drug use, with options of rational, needing adjustment, or unreasonable.

[0096] The pharmacist evaluation and suggestion data area contains several structured input components for recording pharmacists' analysis of the match between infection and medication, specific medication adjustment suggestions, adjustment basis, and key points of medication education;

[0097] The treatment process tracking log data area is a log panel arranged in reverse chronological order. It automatically records and displays key operations and textual notes made by pharmacists, physicians, and infection control specialists regarding the patient's condition and treatment, forming a complete audit trail of the treatment process.

[0098] In this embodiment, the template is presented in the form of an interactive web form, with clear logic and interconnected data in each area.

[0099] Patient Basic Information Data Area: Located at the top of the form, this is a read-only area that is automatically filled in. When the data package is loaded, this area automatically displays information such as ward: infectious disease department, hospital number, name: Chen, and primary diagnosis: decompensated cirrhosis, spontaneous bacterial peritonitis, providing background anchors for the entire record.

[0100] The detailed data area for drug-resistant bacteria detection includes: LIS data with automatic entry of submission date; specimen type: ascites; pathogen: Escherichia coli; specific resistance mechanism: ESBL positive; key drug susceptibility: ceftriaxone resistant, piperacillin-tazobactam sensitive, ertapenem sensitive. The core of this area is a drop-down menu for preliminary pathogenicity assessment, with options for suspected pathogens, colonizing bacteria, and contaminating bacteria, awaiting manual evaluation and selection by the pharmacist.

[0101] Infection Dynamic Indication Data Area: This area is designed as a dynamic visualization panel. On the left, time-series data from the patient's HIS is retrieved in real time via API interface, automatically plotting body temperature change curves and C-reactive protein change trend graphs. The charts support mouse hover to view specific values. On the right, there is a text field that automatically links to the latest imaging examination report, such as abdominal ultrasound: moderate amount of ascites, and a summary of key symptom descriptions such as abdominal pain and positive rebound tenderness.

[0102] The antibiotic use details section dynamically lists all antibiotic prescriptions from the date of submission for testing in tabular form, including drug name (e.g., ceftriaxone), dosage, frequency of administration, route of administration (intravenous drip), and start and stop times. Next to the table is a drop-down menu for preliminary assessment of medication rationality, allowing pharmacists to select whether the use is rational, requires adjustment, or is inappropriate.

[0103] Pharmacist Evaluation and Recommendation Data Area: This is the core workspace for pharmacists, containing several structured components: a multi-line text input box for infection and medication matching analysis; a set of checkboxes for specific intervention recommendations such as escalation therapy, de-escalation therapy, dose adjustment, and consultation recommendations; a text input box for adjustment criteria such as drug sensitivity, guidelines, and consensus; and a text box for recording key points of medication education for patients.

[0104] Treatment Process Tracking Log Data Area: Located at the bottom of the form, this is a log window that automatically refreshes in reverse chronological order. Each record includes a timestamp, the operator's role, and specific content. For example, the system automatically records the system date and time, and automatically loads the patient's admission temperature and CRP data. Pharmacist Zhang: Evaluates that the current ceftriaxone treatment may be ineffective and recommends a change; Physician Li: Adopted the suggestion, discontinued ceftriaxone, and switched to piperacillin-tazobactam. This log constitutes a complete and tamper-proof intervention audit trail.

[0105] Furthermore, referring to Figure 2 Step S6 also includes:

[0106] S61a. In the interface of the electronic tracking intervention record form, the pharmacist activates the pathogenicity auxiliary assessment panel, which is linked to the drug-resistant bacteria detection details data area and the infection dynamic indicator data area;

[0107] S61b. The panel first calls the preset pathogen knowledge base, and automatically loads the background probability information of the microorganism as a common pathogen, conditional pathogen or common contaminant based on the microbial name in the pathogen drug susceptibility report, and displays it in the reference information bar of the panel.

[0108] S61c. The panel provides standardized evaluation factor checkboxes for pharmacists to select. Evaluation factors include at least the presence of clear symptoms or signs at the site of infection, the pathogen being consistent with the spectrum of common pathogens at the site of infection, significant increases in inflammatory markers before and after detection or changes synchronously with the course of infection, imaging results supporting active infection, and the patient not having used effective antimicrobial drugs against the bacteria prior to this detection.

[0109] S61d. The panel has a built-in rule-based logic judgment engine. This engine runs a predefined set of judgment rules based on the combination of judgment factors selected by the pharmacist. The set of judgment rules is defined in the form of judgment tendency if the factor combination is selected.

[0110] S61e. The logic judgment engine outputs the results in a non-mandatory suggestion form, which is displayed as a system auxiliary evaluation prompt;

[0111] S61f. Pharmacist Reference System Assistance Assessment Prompts: Based on your own professional judgment, select one of the following from the preliminary pathogenicity assessment drop-down field: suspected pathogen, colonizing bacteria, or contaminating bacteria, and complete manual confirmation and entry.

[0112] This embodiment demonstrates how pharmacist Zhang, after receiving a report from Chen that 3GCR-EC had been cultured from ascites, used the system to assess pathogenicity.

[0113] Next to the details area for drug-resistant bacteria detection on the electronic record form, Zhang clicked the pathogenicity auxiliary assessment button to activate the dedicated panel.

[0114] Knowledge base information loading: The panel sidebar first displays background information retrieved from the integrated knowledge base: Escherichia coli: normal intestinal flora, and also a common pathogen in community and hospital-acquired infections, especially abdominal and urinary tract infections.

[0115] Standardization Factor Selection: The main panel lists the checkboxes for the evaluation factors. Zhang selected the appropriate ones based on Chen's situation.

[0116] ☑ The patient has clear symptoms or signs of infection, such as abdominal pain, fever, and signs of peritonitis.

[0117] ☑ This pathogen matches the spectrum of common pathogens found at the site of infection. Escherichia coli is the most common pathogen in spontaneous bacterial peritonitis.

[0118] ☑ Inflammatory markers show significant increases before and after detection or change synchronously with the course of infection. CRP is significantly elevated upon admission.

[0119] ☑ Imaging findings support active infection and ascites suggests an underlying infection.

[0120] Prior to this detection, the patient had not used any antibiotics effective against this bacterium. The patient had been using ceftriaxone after admission, and this factor was not met.

[0121] Rule Engine Reasoning and Hints: After selecting the items, Zhang clicks "Analyze." The built-in rule engine in the panel runs predefined logic. For example, a rule is triggered: If factors A, B, C, and D are selected, and the infection site is a sterile site such as ascites, then even if factor E does not meet the criteria of having used ineffective drugs, it is still highly suggested to be a pathogen. The engine outputs a non-mandatory suggestion: System Assisted Judgment Hint: Based on the selected clinical indications, the possibility of this 'Escherichia coli' being the 'pathogen' of this abdominal infection is extremely high.

[0122] The pharmacist ultimately confirmed that, based on this suggestion and considering key points such as the ascites being a sterile bodily fluid and the detection of a single bacterium, Zhang manually selected and confirmed the suspected pathogen in the initial pathogenicity assessment drop-down menu on the main interface of the record form. The entire process made the pharmacist's implicit clinical thinking process explicit and structured.

[0123] Furthermore, referring to Figure 3 Step S6 also includes:

[0124] S62a, in the interface of the electronic tracking intervention record form, the pharmacist activates the auxiliary evaluation panel for rational drug use, which is linked to the details data area of ​​antimicrobial drug use and drug sensitivity results;

[0125] S62b: The panel automatically compares the currently used antimicrobial drug with the drug susceptibility results in the drug susceptibility report, generates an initial drug susceptibility matching report, and identifies sensitive, intermediate, resistant, or untested drugs.

[0126] S62c, the panel accesses the integrated pharmaceutical knowledge base to obtain information on the standard dosing regimen for the currently used drug, including the recommended dose range, dosing interval, infusion requirements, and adjustment strategies for major organ dysfunction, and displays the adjustment strategies in parallel with the extracted actual medical order data;

[0127] S62d, the panel guides pharmacists to make step-by-step evaluations: First, for drug selection, it provides options to evaluate the matching degree between the current medication and the drug sensitivity results; second, for the dosing regimen, it provides options to evaluate the appropriateness of the dosage, route, and frequency; third, for comprehensive evaluation, it provides options to evaluate the rationality of the overall treatment plan.

[0128] S62e, the panel automatically generates an evaluation draft based on the pharmacist's selection in the aforementioned step-by-step evaluation and the recent inflammatory indicator trends read from the infection dynamic indicator data area. The content includes: drug selection based on / deviating from drug sensitivity results, the current dosing regimen compared to the standard, and recent infection indicators showing that the treatment response is effective / ineffective / uncertain.

[0129] S62f, the pharmacist reviews and modifies the evaluation draft to form the final pharmacist evaluation text, fills it into the pharmacist evaluation and suggestion data area, and simultaneously selects the corresponding overall evaluation level in the preliminary judgment field of medication rationality.

[0130] In this embodiment, after completing the pathogenicity assessment, Ms. Zhang needs to assess the rationality of the current ceftriaxone treatment plan. She clicks the auxiliary assessment button for rationality of medication next to the medication details area.

[0131] The panel first displays the comparison results. The left column shows the drug ceftriaxone, and the right column shows the drug sensitivity results: drug resistance is usually marked in red, and other sensitive drug options are also listed.

[0132] The panel retrieves the standard treatment regimen for ceftriaxone in intra-abdominal infections from the pharmaceutical knowledge base, including dosage range and dosing interval, and displays it alongside the actual medical orders extracted from the HIS, indicating whether the dosage is within the normal range.

[0133] The panel guides you through a three-step evaluation process:

[0134] Step 1: Drug selection: For ceftriaxone, the 3GCR-EC drug susceptibility results did not match. She selected "not a match."

[0135] The second step of the dosing regimen: Assuming the dosage meets the standard, she chooses an appropriate one;

[0136] Third step comprehensive evaluation: Given that the pathogenic bacteria are clear and the drug sensitivity is mismatched, her choice is unreasonable and needs to be adjusted immediately;

[0137] Evaluation draft generation: Based on the above selections and the persistently high CRP trend in the infection indication area, the panel generates a draft: Drug selection deviates from the drug sensitivity results, indicating ceftriaxone resistance; the current dosing regimen is appropriate. Recent infection marker CRP has not decreased, indicating that the current treatment response is ineffective.

[0138] Pharmacist's Refinement and Confirmation: Ms. Zhang refined this draft into a final evaluation: ESBL-positive Escherichia coli was detected, showing resistance to the currently used ceftriaxone, which was the main reason for the initial treatment failure. She recommended switching to a sensitive drug such as piperacillin-tazobactam or ertapenem based on the drug sensitivity results. She entered this text into the pharmacist's evaluation and suggestion area of ​​the main record form, and the system simultaneously updated the initial judgment of medication rationality to "irrational."

[0139] Furthermore, referring to Figure 4 In step S5, the specific implementation is as follows:

[0140] In the access control module of the medical information system, an independent application role matrix is ​​defined for the drug-resistant bacteria tracking and intervention platform;

[0141] The application role matrix includes at least three roles: pharmacist, attending physician, and infection control specialist. Each role is configured with read and write permissions for different data areas in the electronic tracking intervention record table.

[0142] When the initial intervention case data package is generated, the system automatically instantiates the application role matrix based on the relationship between the patient and the physician and the rules of the infection control specialist's assigned area, and binds the electronic tracking intervention record form to the instantiated role permissions.

[0143] In this embodiment, the administrator predefines an application role matrix called "Drug-resistant Bacteria Tracking and Intervention" in the platform's permission management backend.

[0144] Role and permission configuration includes:

[0145] Pharmacist role: Has read and write permissions to all data areas of the record sheet, including basic information, detection details, infection indications, medication details, evaluation suggestions, and tracking logs, and has exclusive access to the case closure button.

[0146] Attending Physician Role: Access is restricted; read-only access is granted to the pharmacist evaluation and suggestion data area. However, read and write permissions are available for a dedicated physician feedback field for responding to suggestions. Tracking logs can be viewed.

[0147] Infection Control Specialist Role: Has read-only access to all data areas, as well as read and write access to a dedicated infection control tag field used to mark infection type, whether there is a cluster, etc.

[0148] Automatic instantiation and binding: When Chen's initial data packet is generated, the system automatically executes:

[0149] a) Based on Chen's attending physician information in the HIS, instantiate his physician account Li as the attending physician role.

[0150] b) According to the department-commissioner responsibility system rules set up by the Department of Infection Control, Wang, the commissioner account responsible for the Department of Infection Control, was instantiated as the Infection Control Commissioner role.

[0151] c) Instantiate pharmacist Zhang as a pharmacist role based on pharmacist schedules or professional groups.

[0152] The system then creates an access control record, associating Chen's electronic tracking and intervention record with these three specific role instances. Any subsequent access requests must be verified through this matrix to ensure data security and clear responsibilities.

[0153] Furthermore, the method ensures privacy and security during data flow, specifically in the following ways:

[0154] In steps S1 to S3, the obtained patient identification is transmitted in the hospital intranet through a secure data bus, and the secure data bus adopts a transport layer encryption protocol.

[0155] The electronic tracking intervention record table generated in step S4 is stored in a logically independent database partition. Access to this partition is controlled by the application role matrix and is physically or logically isolated from the production database.

[0156] In the multi-role collaborative tracking stage described in step S6, all add, delete, and modify operations on the record table are recorded in real time to an immutable operation audit log, which records the operator's identity, time, and the state before and after the modification.

[0157] In this embodiment, when transmitting patient identification information from the HIS and LIS systems to the tracking and intervention platform server, a dedicated medical information exchange bus within the hospital's internal network is used. This bus enforces the use of a transport layer encryption protocol to encrypt all transmitted data packets, preventing eavesdropping or interception during transmission.

[0158] The generated electronic tracking intervention record form is not stored in the core business database of HIS or LIS, but rather in a physically independent or logically strictly partitioned clinical decision support database. This database has a single access point and is subject to mandatory verification via the application role matrix described in the embodiments. This effectively isolates the data from production business systems, reducing the risk of unauthorized access to core data.

[0159] The platform records all user actions in full. When pharmacist Zhang selects "suspected pathogen" as the initial pathogenicity assessment, or physician Li enters text in the physician feedback field, the system's backend operation audit log immediately adds a new record. This log entry includes: a precise timestamp, the operator's unique login ID, the ID of the record table being operated on, the name of the modified field, the value before modification, and the value after modification. This log uses an append-only design, making it impossible for anyone to delete or modify, forming an immutable chain of evidence for post-event traceability, quality review, and dispute resolution.

[0160] Furthermore, step S8 specifically includes:

[0161] The smart contract rules are defined in the form of executable scripts and deployed on the server side of the tracking intervention case database;

[0162] Rule Script 1: Define the data fields and aggregation calculation methods required to extract pathogen distribution, specimen source composition, and drug resistance rate statistics within a specific period from the drug resistance bacteria detection details data area of ​​the archived record table;

[0163] Rule Script 2: Define the logic for extracting and analyzing the treatment outcome data of irrational drug use cases from the initial judgment field of drug rationality and the treatment outcome field;

[0164] Rule Script 3: Define a fixed template for generating report documents, including the format and content filling rules for the cover, abstract, data analysis charts, typical case analysis, key drug warnings, and improvement suggestions;

[0165] When a periodic analysis task is triggered, the system executes rule scripts one, two, and three in sequence, extracts the standardized dataset processed by the rules from the database, and populates it into the fixed template, ultimately generating a composite document containing text, tables, and charts.

[0166] In this embodiment, several executable scripts, which function as smart contracts, are deployed on the server of the tracking intervention platform.

[0167] Rule Script 1: Data Extraction and Aggregation: This script defines how to extract data from the previous quarter from the drug-resistant bacteria detection details data area of ​​the archived record table. For example, it instructs the database to: find all records whose pathogen names contain Escherichia coli and whose drug resistance mechanism contains ESBL; group and count by ward field to obtain the detection volume of each department; calculate the drug resistance rate to drugs such as ceftriaxone, piperacillin-tazobactam, and carbapenems, and output the results as a standardized data object.

[0168] Rule Script Two: Association Analysis. This script performs in-depth analysis, linking the initial assessment of medication rationality with the treatment outcome fields. For example, it identifies all cases initially deemed irrational, analyzes the distribution of these cases' eventual cure, ineffectiveness, or death, and calculates the clinical consequences of irrational medication use. The script may also analyze the physician adoption rate of pharmacist recommendations, and the analysis results are incorporated into the data object.

[0169] Rule Script 3: Report Synthesis: This script connects to a pre-made Word report template. The template contains fixed-format sections and placeholders, such as {esbl_ecoli_distribution_by_ward}, {antibiotic_resistance_trend_chart}, and {typical case summary}. The script reads the aforementioned data objects, transforms the data into corresponding charts and tables, selects 1-2 educational, desensitized cases, and writes brief analyses. Finally, the script replaces all placeholders in the template with this content, automatically synthesizing a well-formatted, illustrated quarterly ESBL (Escherichia coli) monitoring and clinical intervention analysis report.

[0170] Furthermore, the method achieves standardization and real-time sharing of pharmacist medication guidance through system collaboration and data integration, specifically manifested in:

[0171] The electronic tracking intervention record form generated in step S4 serves as a unique and synchronized information carrier for all participating nodes, ensuring that pharmacists, physicians, and infection control specialists make decisions and communicate based on real-time updated data.

[0172] The standardized evaluation options and structured text input in the pharmacist evaluation and suggestion data area, as well as the suggestions generated through the medication rationality auxiliary evaluation panel, form a unified evaluation discourse system.

[0173] In the multi-role collaboration stage of step S6, the pharmacist's evaluation and suggestions are presented to the attending physician node in real time, and the physician's feedback and subsequent medical order adjustments are also recorded in real time.

[0174] All archived records form a searchable case knowledge base, allowing other pharmacists to refer to the evaluations and recommendations in historical cases when dealing with the same drug-resistant bacteria or clinical situations.

[0175] In this embodiment, during Chen's treatment, pharmacist Zhang, physician Li, and infection control specialist Wang consistently reviewed and discussed the same real-time updated electronic record sheet. Suggestions entered by Zhang were immediately visible on Li's computer; Li's feedback and changes to medical orders were also reflected in the record sheet's log in real time. This eliminated delays and ambiguities in information transmission.

[0176] Through auxiliary evaluation panels and structured input fields such as reasonable / needs adjustment / unreasonable, pharmacists are guided to use standardized clinical thinking paths and conclusive language. Regardless of which pharmacist handles Chen's case, the analysis process, guided by the system, will cover core aspects such as pathogenicity assessment, drug sensitivity matching, and efficacy evaluation. The structure and key wording of the output evaluation tend to be consistent, reducing fluctuations in work quality caused by differences in personal habits and seniority.

[0177] Traditional pharmaceutical intervention models can be delayed. In this system, after Zhang submits his evaluation suggestions, physician Li receives a real-time notification on his HIS interface, enabling instant communication between pharmacists and physicians. This close collaboration allows pharmacists' suggestions to directly and quickly influence physicians' immediate decisions, moving the intervention window forward and improving the timeliness and effectiveness of the intervention.

[0178] All the case records, like Chen's, that have been closed and archived constitute a valuable real-world case database. When new or rotating pharmacists encounter cases of cirrhotic ascites with 3GCR-EC infection, they can search for similar historical cases in the system. They can intuitively see how senior pharmacists analyzed and judged the cases, what suggestions they made, how the physicians responded, and what the final treatment outcomes were. This learning based on real, structured cases greatly promotes the efficient and standardized transmission of knowledge and experience.

[0179] Furthermore, the comparative evaluation of system performance includes the implementation steps for the following evaluation dimensions:

[0180] Operational evaluation steps: After design and deployment, a structured questionnaire is distributed to pharmacists and physicians who use the system. The questionnaire questions focus on the clarity of the system interface logic, the convenience of data acquisition, the efficiency of record filling, and the smoothness of multi-role collaborative processes. Feedback is collected and the overall usability score is calculated.

[0181] Effectiveness evaluation steps: Extract treatment outcome data of all closed cases within a specific period after the system application from the tracking intervention case database, compare it with historical case data tracked by traditional methods in the same period before the system application, and statistically compare the difference in the proportion of cases marked as cured or effective in infection treatment between the two groups of data.

[0182] Safety evaluation steps: From the treatment process tracking log of the electronic tracking intervention record form, text mining technology is used to identify and count the number of times recorded adverse drug reaction events or treatment-related adverse events are mentioned; the incidence rates of the above events are compared before and after the system application in the same period.

[0183] Efficiency evaluation steps: Obtain comparable case groups of patients with drug-resistant bacterial infections before and after system application, and compare and analyze the median changes in key efficiency indicators such as average length of hospital stay and average duration of antibiotic use.

[0184] Evaluation steps for quality control indicators: retrieve hospital pharmaceutical quality control data, compare the numerical trends of the intensity of antimicrobial drug use, microbial testing rate, and antimicrobial drug prescription qualification rate in multiple consecutive statistical periods before and after the system application, and analyze the direction and magnitude of their changes.

[0185] In this embodiment, how can the hospital management department conduct a scientific and multi-dimensional performance evaluation of the system after it has been fully online and running for a full year?

[0186] The online questionnaire was jointly designed by the hospital's information technology department and pharmacy department. It was distributed to all clinical pharmacists who had used the system and clinicians who frequently interacted with it. The questionnaire focused on user experience, including statements such as that obtaining information on drug-resistant patients through the system was more convenient than before, the auto-fill function of the record forms significantly reduced my paperwork time, and the system improved efficiency in collaborating with pharmacists / physicians. A consent scale was used to collect feedback. Statistical analysis was performed on all returned questionnaires to produce a subjective evaluation report on the system's usability and efficiency improvements.

[0187] The research team established clear inclusion and exclusion criteria. All eligible and closed cases of drug-resistant bacterial infections within one year of the system's implementation were extracted from the intervention case database. Simultaneously, cases meeting the same criteria within one year prior to the system's implementation were manually and retrospectively collected from historical medical records as a control group. The core efficacy endpoint was the clinical treatment success rate, calculated using the cure and effective outcomes marked in the record table. Standard statistical methods were used to compare whether there was a significant difference in treatment success rates between the two groups, objectively assessing whether the system-assisted intervention improved patient prognosis.

[0188] Evaluation personnel used natural language processing tools to perform text mining on treatment process tracking logs from all archived records within one year of the system's launch. They screened for records containing keywords related to adverse drug reactions such as rash, abnormal liver function, kidney damage, and allergic reactions. These records were manually verified, and the exact number of adverse drug events was counted. The same method was used to retrospectively analyze the medical records and nursing records of the control group. The incidence of adverse drug events was compared between the two groups to assess the impact of enhanced end-to-end monitoring on medication safety.

[0189] Hospitalization information of patients with drug-resistant bacterial infections was obtained from the hospital's medical record system, representing two control groups before and after system application. Two key efficiency indicators were extracted: total days of antibiotic use and total days of hospitalization. Since this type of data typically does not conform to a normal distribution, descriptive statistics were performed using the median and interquartile range. Appropriate nonparametric tests were then used to compare whether these two medians changed significantly before and after system application, thereby analyzing whether the system saved medical resources and time costs by optimizing treatment pathways.

[0190] From the annual quality control report of the Hospital Pharmacy Management and Therapeutics Committee, core indicators for hospital-wide antimicrobial drug management were extracted for several consecutive quarters before and after the system's implementation, such as the four quarters of the year before and the four quarters of the year after implementation. These indicators mainly included: antimicrobial drug usage intensity, inpatient antimicrobial drug usage rate, and the rate of pathogen testing before inpatients' therapeutic use of antimicrobial drugs. These indicators were plotted as a continuous trend chart over time to visually demonstrate the changes and fluctuations before and after the system's introduction, and to comprehensively evaluate the system's long-term contribution to improving the hospital's overall scientific management of antimicrobial drugs.

[0191] According to a second embodiment of the present invention, the present invention claims protection for a precision drug delivery auxiliary analysis system for drug-resistant bacteria, comprising:

[0192] One or more processors;

[0193] A memory storing one or more programs, which, when executed by one or more processors, enable the one or more processors to implement the aforementioned method for precise drug delivery and auxiliary analysis of drug-resistant bacteria.

[0194] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units. The above are merely embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

[0195] The specific embodiments of the invention have been described in detail above, but they are only examples, and this application is not limited to the specific embodiments described above. For those skilled in the art, any equivalent modifications or substitutions to the invention are also within the scope of this application. Therefore, all equivalent changes, modifications, and improvements made without departing from the spirit and principles of this application should be covered within the scope of this application.

Claims

1. A precision drug-resistant bacteria administration assistance analysis method, characterized in that, The method comprises the following steps: S1, in response to the submission operation of the inpatient microbial specimen, the test application form identification and patient identification of the patient are acquired in real time; S2, based on the patient identification, the clinical data set of the patient is synchronously extracted; S3, when a pathogen drug sensitivity report containing drug resistance or special drug resistance markers is generated, a warning event is triggered according to the test application form identification, the pathogen drug sensitivity report is associated and bound with the clinical data set, and an initial intervention case data package is generated; S4, a standardized tracking intervention record table template is called, each data item in the initial intervention case data package is automatically filled into the pre-defined field of the standardized tracking intervention record table template, and an electronic tracking intervention record table is generated; S5, the electronic tracking intervention record table is distributed to the work queue of the pharmacist, and based on the preset distributed node rule, the access and editing permission of the electronic tracking intervention record table is synchronously granted to the designated doctor node of the clinical department to which the patient belongs and the designated hospital infection management department specialist node; S6, after the electronic tracking intervention record table enters the editable state, the multi-role collaborative tracking stage is started; S7, after the infection treatment cycle of the patient ends, the pharmacist triggers the closing operation in the electronic tracking intervention record table, marks the treatment outcome, changes the state of the electronic tracking intervention record table to the read-only archiving state, and stores it into the tracking intervention case database; S8, the archived record table in the tracking intervention case database is analyzed in batches, and the key analysis dimension data is extracted from the archived record table through the preset smart contract rule, and the smart contract rule defines the field, format and logical judgment condition of data extraction.

2. The method of claim 1, wherein, The design and application of the standardized tracking intervention record table template in step S4 comprises: The standardized tracking intervention record table template adopts an electronic form, and the interface is divided into a plurality of interlocked and associated logical areas; The patient basic information data area includes ward, hospitalization number, name, age, and diagnosis information fields; The drug-resistant bacteria detection detail data area includes submission date, specimen type, pathogen name, drug resistance mechanism, and key drug sensitivity result fields, and is provided with a pathogenicity preliminary judgment drop-down selection field with options of suspected pathogenic bacteria, colonized bacteria or contaminated bacteria; The infection dynamic indication data area dynamically links the patient time series data through an interface, displays the temperature, white blood cell count, C-reactive protein and procalcitonin change curves with time in a chart form, and provides a text field for linking or summarizing infection related imaging and symptom description; The antibacterial drug use detail data area is extracted from the medical order system, and the current and historical antibacterial drug names, doses, frequencies, administration routes, start and end times are displayed in a list, and a drug rationality preliminary judgment field is provided, with options of reasonable, need to adjust or unreasonable; The pharmacist evaluation and suggestion data area comprises a plurality of structured input components for recording the pharmacist's analysis of the matching degree of infection and drug use, specific drug adjustment suggestions, adjustment basis and drug education points. The treatment process tracking log data area is a log panel arranged in reverse chronological order, which automatically records and displays the key operations and text notes on the patient's condition and treatment by pharmacists, doctors, and infection control officers, forming a complete audit tracking of the treatment process.

3. The method of claim 2, wherein, Step S6 further comprises: S61a. The pharmacist activates the pathogenicity auxiliary judgment panel linked to the drug-resistant bacteria detection details data area and the infection dynamic indication data area in the interface of the electronic tracking intervention record table; S61b. The pathogenicity auxiliary judgment panel first calls the preset pathogen knowledge base, automatically loads the microorganism as the background probability information of common pathogenic bacteria, conditional pathogenic bacteria or common contaminating bacteria according to the microorganism name in the pathogen drug sensitivity report, and displays it in the reference information column of the panel; S61c. The pathogenicity auxiliary judgment panel provides a standardized evaluation factor checkbox for the pharmacist to check, and the evaluation factor at least includes that the patient has clear infection site symptoms or signs, the pathogen is consistent with the common pathogenic bacteria spectrum of the infection site, the inflammatory markers show an increase before and after detection or change synchronously with the infection process, the imaging examination result supports active infection, and the patient has not used effective antibiotics against the bacteria before this detection; S61d. The pathogenicity auxiliary judgment panel has a rule-based logical judgment engine built-in, which runs a predefined set of judgment rules according to the combination of evaluation factors selected by the pharmacist, and the set of judgment rules is defined in the form of if factor combination, then evaluation tendency; S61e. The logical judgment engine outputs the running result in the form of non-mandatory suggestion, which is displayed as a system auxiliary judgment prompt; S61f. The pharmacist refers to the system auxiliary judgment prompt, combines his own professional judgment, selects one of the suspected pathogenic bacteria, colonizing bacteria or contaminating bacteria in the pathogenicity preliminary judgment drop-down selection field, and completes manual confirmation and entry.

4. The method of claim 2, wherein, Step S6 further comprises: S62a. The pharmacist activates the medication rationality auxiliary judgment panel linked to the antibacterial drug use details data area and the drug sensitivity result in the interface of the electronic tracking intervention record table; S62b. The medication rationality auxiliary judgment panel automatically compares the variety of the currently used antibacterial drug with the drug sensitivity result in the pathogen drug sensitivity report, generates an initial drug sensitivity matching degree report, and identifies sensitive, intermediate, resistant or not tested; S62c. The medication rationality auxiliary judgment panel accesses the integrated pharmaceutical knowledge base to obtain the standard drug administration scheme information of the currently used drug, including the recommended dose range, drug administration interval, infusion requirements and adjustment strategies for major organ dysfunction, and displays the adjustment strategies and the extracted actual medical order data side by side; S62d. The medication rationality auxiliary judgment panel guides the pharmacist to perform step-by-step evaluation: first, for drug selection, it provides selection items to evaluate the matching degree of the current medication and the drug sensitivity result; second, for drug administration scheme, it provides selection items to evaluate the suitability of dose, route and frequency; third, comprehensive evaluation, it provides selection items to evaluate the rationality of the overall treatment scheme; S62e, the drug rationality auxiliary evaluation panel automatically generates an evaluation draft based on the choices of the pharmacist in the previous step-by-step evaluation and the recent inflammation indicator trend read from the infection dynamic indicator data area, including: drug selection based on / deviating from drug sensitivity results, current drug regimen compared to standards, and recent infection indicators showing effective / ineffective / uncertain treatment response; S62f, the pharmacist reviews and modifies the evaluation draft to form a final pharmacist evaluation text, fills in the pharmacist evaluation and suggestion data area, and synchronously selects the corresponding overall evaluation level in the drug rationality preliminary judgment field.

5. The method of claim 1, wherein, In step S5, the specific implementation is: In the permission management module of the medical information system, an independent application role matrix is defined for the drug-resistant bacteria tracking intervention platform; The application role matrix at least includes three roles of pharmacist, supervising physician, and infection control officer, and each role is configured with read and write permissions for different data areas in the electronic tracking intervention record table; When the initial intervention case data package is generated, the system automatically instantiates the application role matrix based on the patient-physician association and the rules for the division of the infection control officer's area of responsibility, and binds the electronic tracking intervention record table with the instantiated role permissions.

6. The method of claim 5, wherein, The method guarantees privacy and security during data flow, which is specifically embodied in: In steps S1 to S3, the patient's identity is transmitted through a secure data bus within the hospital network, and the secure data bus uses a transport layer encryption protocol; The electronic tracking intervention record table generated in step S4 is stored in a logically independent database partition, and the access to this partition is controlled by the application role matrix and is physically or logically isolated from the production database; In the multi-role collaborative tracking phase in step S6, all operations of adding, deleting, and modifying the record table are recorded in real time in the operation audit log, which records the operator's identity, time, and the state before and after the modification.

7. The method of claim 1, wherein, In step S8, specifically includes: The smart contract rules are defined in the form of executable scripts and deployed on the server side of the tracking intervention case database; Rule script one: define the data fields and aggregation calculation methods required for pathogen distribution, specimen source composition, and drug resistance rate statistics within a specific period from the drug-resistant bacteria detection details data area of the archived record table; Rule script two: define the logic for extracting and correlatively analyzing the treatment outcome data of unreasonable drug use cases from the drug rationality preliminary judgment field and the treatment outcome field; Rule script three: define the fixed template for generating report documents, including cover, abstract, data analysis charts, typical case analysis, core drug warning points, and improvement suggestions in terms of format and content filling rules; When the periodic analysis task is triggered, the system executes the rule scripts one, two, and three in sequence to extract the standardized data set after rule processing from the database, and fills it into the fixed template to finally generate a composite document containing text, tables, and charts.

8. The method of claim 4, wherein, The method realizes the standardization and real-time sharing of pharmacist drug guidance through system collaboration and data integration, which is specifically embodied in: The electronic tracking intervention record table generated in step S4 is a unique and synchronized information carrier for all participating nodes, ensuring that pharmacists, physicians, and infection control officers make decisions and communicate based on real-time updated data; The standardized evaluation options and structured text input in the pharmacist evaluation and suggestion data area, as well as the suggestions generated by the drug use rationality auxiliary judgment panel, form a unified evaluation discourse system; In the multi-role collaboration phase of step S6, the pharmacist's evaluation and suggestion is presented to the supervising physician node in real time, and the physician's feedback and subsequent order adjustment is also recorded in real time; All archived record tables constitute a searchable case knowledge base, which can be referred to by other pharmacists when dealing with the same drug-resistant bacteria or clinical situation.

9. The method of claim 1, wherein, For comparative evaluation of system operation effect, the following evaluation steps are included: Operational evaluation step: After design and deployment, a structured questionnaire is sent to the pharmacist and physician groups using the system, focusing on the clarity of the system interface logic, the convenience of data acquisition, the efficiency of record filling, and the smoothness of the multi-role collaboration process. Collect feedback and calculate the overall ease of use score; Effectiveness evaluation step: From the tracking intervention case database, extract the treatment outcome data of all closed records in a specific period after the system is applied, and compare it with the historical data of cases tracked by traditional methods in the same period before the system is applied. Calculate and compare the proportion of cases marked as cured or effective in the two groups of data; Safety evaluation step: From the treatment process tracking log of the electronic tracking intervention record table, identify and count the number of adverse drug reactions or treatment-related adverse events recorded in the text mining technology; Compare the incidence of the above events in the same period before and after the application of the system; Efficiency evaluation step: Obtain groups of drug-resistant bacteria infection patient cases before and after the application of the system, and compare the median changes of key efficiency indicators such as average hospitalization days and average antibiotic use course; Quality control index evaluation step: Retrieve hospital pharmacy quality control data and compare the numerical trends of hospital-wide antibiotic use intensity, microbiological submission rate, and antibiotic prescription pass rate in consecutive statistical periods before and after the application of the system. Analyze the change direction and amplitude.

10. A precision drug-resistant bacteria administration assistance analysis system, characterized in that, Comprise: One or more processors; A memory having one or more programs stored thereon, when the one or more programs are executed by the one or more processors, so that the one or more processors implement a drug-resistant bacteria precise drug administration auxiliary analysis method according to any one of claims 1 to 9.

Images