Visual evaluation system for antibacterial spectrum radar map and presentation method using visual evaluation system

Abstract

The invention belongs to the technical field of medical information, and particularly relates to an antibacterial spectrum radar map visual evaluation system and a presentation method using the same, and the system comprises a structured antibacterial drug sensitivity database, a multi-dimensional radar map generation engine, and a weighted calculation and interactive display module. The structured antibacterial drug sensitivity database comprises a basic numerical database and a graph and weighted numerical database; the multi-dimensional radar map generation engine maps the sensitivity rates of drugs to six types of pathogens into graphic coordinates; the weighted calculation and interactive display module is used for calculating the overall sensitivity rate of antibacterial drugs to target pathogens based on infection parts and characteristics, the information reading and comprehensive decision-making efficiency is remarkably improved, the quantitative evaluation display decision-making time is shortened by 50% or above, complementary visual evaluation of a drug combination scheme is achieved, and the method is suitable for large-scale popularization and application. Reasonable use of antibacterial drugs is promoted, and extension of professional decision support tools from a fixed workstation to mobility and scenario is realized.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of medical information technology, and particularly relates to an antibacterial spectrum radar chart visualization evaluation system and a presentation method using the same. BACKGROUND

[0002] Currently, when choosing antibacterial drugs, doctors mainly rely on static text or table data such as drug instruction manuals, clinical guidelines, and antibacterial spectrum classification tables for judgment. Such traditional display methods have the following problems: 1. Low information extraction efficiency: doctors need to find the sensitivity information of the target drug to different bacteria in a large amount of text or table, and it is difficult to quickly compare the differences in antibacterial spectrum of multiple drugs; 2. Difficulty in evaluating combined drug use: traditional methods cannot directly display the overlap and complementary relationship of the antibacterial spectrum of two or more drugs, and it is difficult to determine whether the combined drug use covers the target bacterial population; 3. Inconvenient for mobile applications: existing tools are mostly computer-based or paper-based, and cannot be used for real-time queries in mobile scenarios such as bedside and emergency; 4. Data update lag: Most antibacterial spectrum data is in a fixed version, and it is difficult to dynamically adjust with guideline updates and new drug / bacterial species.

[0003] As can be seen, although there are antibacterial drug query software such as Micromedex and UpToDate in the prior art, they still mainly use text and tables, lack graphical and multi-dimensional intuitive comparison functions, and are high in cost and complex in deployment, making it difficult to promote in primary hospitals. Therefore, there is an urgent need for a visualization method that can intuitively, dynamically, and mobile display antibacterial spectrum. SUMMARY

[0004] The present application provides an antibacterial spectrum radar chart visualization evaluation system and a presentation method using the same to solve the above problems.

[0005] To achieve the above purpose, the present application adopts the following technical solutions: An antibacterial spectrum radar chart visualization evaluation system, comprising a structured antibacterial drug sensitivity database, a multi-dimensional radar chart generation engine, and a weighted calculation and interactive display module; The structured antibacterial drug sensitivity database includes a basic numerical database and a graphical and weighted numerical database; The multi-dimensional radar chart generation engine maps the sensitivity rate of the drug to six types of pathogens into a graphical coordinate; The weighted calculation and interactive display module calculates the overall sensitivity rate of the antibacterial drug to the target pathogen based on the infection site and characteristics.

[0006] Further, the basic numerical database is composed of an excel table.

[0007] Further, the multi-dimensional radar chart generation engine supports multi-drug radar chart overlay comparison and uses color light three primary color mixing method to display the overlapping area.

[0008] Further, the basic numerical database contains the corresponding relationship of at least 45 kinds of antibacterial drugs and 89 kinds of pathogenic bacteria in 6 categories.

[0009] Further, the 6 categories of pathogenic bacteria include gram-positive cocci, gram-positive bacilli, aerobic non-fermenting gram-negative bacilli, aerobic gram-negative bacilli-enteric bacilli, and aerobic gram-negative bacilli-non-enteric bacilli and anaerobes.

[0010] An antibacterial spectrum radar chart visualization presentation method, comprising the following steps: Step 1: query condition input and analysis, the user selects the following key parameters through the interface: Mandatory: pre-use antibacterial drugs: the user can fill in the pre-use antibacterial drugs one by one, up to three at a time, at this time, through data processing, the "drug-bacteria" sensitivity data is converted into radar chart coordinate points through coordinate mapping algorithm, and six-dimensional radar chart of each drug is generated.

[0011] Optional: infection site and characteristics: infection sites include joints, abdominal cavity, lungs, blood flow, etc. Characteristics include brain abscess of primary or adjacent site infection, brain abscess after surgery or trauma, age:> 50 years old, alcoholics, people with serious underlying diseases or impaired cellular immune function, etc.; used to obtain suspected pathogen categories: the system supports fast screening based on "graph and weighted numerical database" preset scenarios, the user can obtain suspected pathogen categories through infection site and characteristics, and reference weight values of different pathogens are given.

[0012] If only the mandatory part is filled in, only the basic numerical database is entered to retrieve data and generate graphs; if the mandatory and optional parts are filled in, the basic numerical database is entered to retrieve data, and then the graph and weighted numerical database are entered to retrieve weighted data and calculate the weighted graph; Step 2: data retrieval and weighted integration After the system receives the query conditions, the data processing engine is started, and first data positioning is performed: according to the antibacterial drugs selected by the user, the corresponding data column set is accurately positioned in the basic numerical database, which is the basis for subsequent analysis, and the "drug-bacteria" sensitivity data is converted into radar chart coordinate points through coordinate mapping method, and six-dimensional radar chart of each drug is generated. In special cases, multi-pathogen weighted calculations are performed: If the user selects the infection site and characteristics, the system does not simply list them, but executes a crucial weighted average algorithm. This algorithm combines the sensitivity rates of different pathogen categories to the same antimicrobial drug. By default, it uses an arithmetic mean, but the system allows the introduction of weighting coefficients to calculate an overall sensitivity rate representing "coverage of this group of pathogens." The formula is used to calculate the weighted average algorithm for the overall sensitivity rate of antimicrobial drugs. Formula illustration: For antimicrobial drug i, its overall sensitivity against the selected set of pathogens: Where n is the number of selected pathogen categories, wj is the weight of the j-th pathogen, and rij is the sensitivity rate of the j-th pathogen to drug i in the database; Step 3: Based on the antibacterial drug selected by the user, the system converts the "sensitivity rate" or "coverage probability" dataset calculated in the previous step into a multi-dimensional radar chart; Step 4: Interaction and Decision Support The generated radar charts are highly interactive, featuring hover tooltips: hovering the mouse over a polygonal area or axis displays the specific drug name and precise sensitivity values. Dynamic adjustments are also included: users can add or delete drugs in real-time, and the radar chart will immediately recalculate and render, helping to explore optimal personalized treatment plans.

[0013] Compared with the prior art, the present invention has the following advantages: 1. This invention employs multi-dimensional data organization, arranging data in a matrix according to pathogen taxonomy and antimicrobial drug categories. It utilizes graphical and weighted numerical databases to define the weights or default selections of common pathogens at different infection sites, serving as initial input parameters for subsequent analysis. This not only allows for visual understanding of the antimicrobial spectrum of a particular antimicrobial drug but also determines the range of antimicrobial spectrum coverage when using two or three drugs in combination. Furthermore, during weighted calculations, it reveals the probability that a particular antimicrobial drug can cover suspected pathogens at certain infection sites. Finally, based on the Kingsoft Docs sharing platform, it enables real-time querying and graphical display on mobile devices and computers, significantly improving information retrieval and comprehensive decision-making efficiency.

[0014] 2, The application realizes the complementary visualization evaluation of the combined medication scheme, and promotes the reasonable use of antibacterial drugs: when evaluating the combined medication, the existing method is difficult to directly judge whether the antibacterial spectrum of the drugs is complementary or overlapping. The application allows the radar chart of different medication schemes, such as A drug alone, B drug alone, and A+B combination, to be superimposed and displayed, and by comparing the morphological changes of the superimposed graph, the doctor can clearly identify: if the polygon is extended in each dimension after combination, and the area is significantly increased, it indicates that the antibacterial spectrum of the two drugs is complementary and covers a wider range; if the graph is only expanded in part after combination, and the overall area increases limitedly, it indicates that there is significant overlap, and it may be unnecessary combination. This intuitive comparison based on graphics provides a powerful visual decision tool for “avoiding unreasonable combination of antibacterial drugs”, and helps to change from empirical combination to precise combination based on spectrum complementarity.

[0015] 3, The application adopts a loosely coupled architecture of “database excel and general front-end display”, and the core data file can be stored in a hospital internal server or a safe cloud document platform, such as Tencent document and Jinshan document. Doctors can access and operate the system in real time through a lightweight APP such as WeChat on office computers, ward rounds, personal mobile phones or tablet computers through authorization. This feature makes it particularly suitable for emergency room rapid decision-making, bedside multidisciplinary consultation, and off-site remote consultation, which have high requirements for timeliness and flexibility, and realizes the extension of professional decision support tools from fixed workstations to mobility and scene.

[0016] 4, The basic numerical database updating mechanism of the application is flexible and efficient, and guarantees the timeliness of the system knowledge. Unlike traditional professional software that requires complex programming interfaces and long upgrade cycles, the knowledge core of the application, the antibacterial drug sensitivity database, is maintained in a structured general table format, such as Excel. Professional personnel in the hospital pharmacy department only need to add new pathogens in the form of rows and new antibacterial drugs or modify existing sensitivity rate values in the form of columns in the table according to the established format to complete the annual update or real-time correction of the data. This update can be realized by replacing the cloud file or internal network file synchronization. This design ensures that the system can quickly follow the hospital drug sensitivity monitoring data, the national bacterial drug resistance monitoring network report and the latest international guidelines, and always maintains its advanced nature and applicability.

[0017] 5, The application has extremely low deployment and maintenance cost, and is beneficial to be widely promoted in medical institutions at all levels. It does not need to purchase expensive special commercial software, nor does it need to deeply modify or develop interfaces for the existing hospital information system of the hospital. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 It is a six-dimensional radar chart coordinate model schematic diagram of the application; Figure 2 It is a radar chart schematic diagram of a single antibacterial drug. Figure 3 This is a radar schematic diagram of the two antibacterial drugs of the present invention; Among them, aerobic Gram-positive cocci 1, aerobic Gram-positive bacilli 2, aerobic Gram-negative bacilli-Enterobacteria 3, aerobic Gram-negative bacilli-non-Enterobacteria 4, aerobic non-fermenting Gram-negative bacilli 5, and anaerobic bacteria 6; Figure 2 Taking "ceftriaxone" as an example, the coverage of ceftriaxone in aerobic Gram-positive cocci is 7, the coverage of ceftriaxone in aerobic Gram-positive bacilli is 8, the coverage of ceftriaxone in aerobic Gram-negative bacilli-Enterobacteria is 9, the coverage of ceftriaxone in aerobic Gram-negative bacilli-non-Enterobacteria is 10, the coverage of ceftriaxone in aerobic non-fermenting Gram-negative bacilli is 11, and the coverage of ceftriaxone in anaerobic bacteria is 12. Figure 3 Taking "ceftriaxone" and "levofloxacin" as examples, the antibacterial spectrum of ceftriaxone is visualized in the six-dimensional capability radar chart 13, the antibacterial spectrum of levofloxacin is visualized in the six-dimensional capability radar chart 14, the overlapping part of the antibacterial spectra of ceftriaxone and levofloxacin is visualized in the six-dimensional capability radar chart 15, and the part of the antibacterial spectrum of levofloxacin that extends beyond that of ceftriaxone is visualized in the six-dimensional capability radar chart 16. Detailed Implementation

[0019] To further illustrate the technical solution of the present invention, the present invention will be further described below through embodiments.

[0020] like Figures 1-3 As shown, an antibacterial spectrum radar chart visualization evaluation system includes a structured antibacterial drug sensitivity database, a multidimensional radar chart generation engine, and a weighted calculation and interactive display module. The structured antimicrobial susceptibility database includes a basic numerical database and a graphical and weighted numerical database; The multidimensional radar chart generation engine maps the drug's sensitivity rate to six types of pathogens into graphical coordinates; The weighted calculation and interactive display module calculates the overall sensitivity rate of antimicrobial drugs to the target pathogen based on the infection site and characteristics.

[0021] The basic numerical database is based on Excel spreadsheets.

[0022] The multi-dimensional radar image generation engine supports the overlay and comparison of multiple drug radar images and uses the three primary colors of light mixing method to display overlapping areas.

[0023] The basic numerical database contains the corresponding relationship of at least 45 kinds of antibacterial drugs and 89 kinds of pathogenic bacteria of 6 categories, including gram-positive cocci, gram-positive bacilli, aerobic non-fermenting gram-negative bacilli, aerobic gram-negative bacilli-enteric bacilli and aerobic gram-negative bacilli-non-enteric bacilli and anaerobic bacteria.

[0024] A radar chart visualization presentation method of antibacterial spectrum, comprising the following steps: Step 1: query condition input and analysis, the user selects the following key parameters through the interface: Mandatory: pre-use antibacterial drugs: the user can fill in the pre-use antibacterial drugs one by one, up to three at a time, at this time through data processing, the "drug-bacteria" sensitivity data is converted into radar chart coordinate points through coordinate mapping algorithm, and six-dimensional radar chart of each drug is generated.

[0025] Optional: infection site and characteristics: infection sites include joints, abdominal cavity, lungs, blood flow, etc. Characteristics include brain abscess of primary or adjacent site infection, brain abscess after surgery or trauma, age:> 50 years old, alcoholics, people with serious underlying diseases or people with impaired cellular immune function, etc.; used to obtain suspected pathogen categories: the system supports quick filtering based on the "graph and weighted numerical database" preset scenario, the user can obtain the suspected pathogen categories through the infection site and characteristics, and the reference weight value of different pathogens is given.

[0026] If only the mandatory part is filled, only the basic numerical database is entered to retrieve data and generate a graph; if the mandatory and optional parts are filled, the basic numerical database is entered to retrieve data first, and then the graph and weighted numerical database are entered to retrieve weighted data and calculate the weighted graph; Step 2: data retrieval and weighted integration After the system receives the query conditions, the data processing engine is started, and first data positioning is performed: according to the antibacterial drugs selected by the user, the corresponding data column set is accurately positioned in the basic numerical database, which is the basis for subsequent analysis, and the "drug-bacteria" sensitivity data is converted into radar chart coordinate points through a specific algorithm, and six-dimensional radar chart of each drug is generated. All filled in will perform multi-pathogen weighted calculation: if the user selects the infection site and characteristics, the system will not simply list, but will perform a key weighted average algorithm. The algorithm integrates the sensitivity rates of different pathogen categories to the same antibacterial drug, and by default, an arithmetic average is used, but the system allows the introduction of a weighted coefficient to calculate an overall sensitivity rate representing "covering the group of pathogens". The formula is used to calculate the weighted average algorithm of the overall sensitivity rate of the antibacterial drug. Formula: for antibacterial drug i, the overall sensitivity rate of the selected pathogen set n is the number of selected pathogen categories, wj is the weight of the jth pathogen, and rij is the susceptibility rate of the jth pathogen to drug i in the database; Step 3: According to the user's selected antibacterial drug, the system converts the "susceptibility rate" or "coverage probability" data set calculated in the previous step into a multi-dimensional radar chart. The main points are as follows: Dimension mapping: The six common pathogenic bacteria are the main grid lines of the radar axes of the radar chart. The six dimensions are arranged in clockwise order from the "12 o'clock direction" as Gram-positive cocci, Gram-positive bacilli, aerobic Gram-negative bacilli-enteric bacilli, aerobic Gram-negative bacilli-non-enteric bacilli, aerobic non-fermenting Gram-negative bacilli, and anaerobic bacteria.

[0027] Numerical mapping: The calculated "susceptibility rate" values are mapped to the coordinates of the corresponding dimensions. The higher the value, the farther the point on that axis from the center.

[0028] Basic graph drawing: Connect all the points on the dimensions to form a closed polygon. The area and shape of the polygon intuitively represent the antibacterial spectrum of the selected antibacterial drug. The larger the area, the broader the antibacterial spectrum of the drug. If the user selects multiple drug combinations, separate closed polygons are formed. The overlapping areas of the polygons are displayed using the color light three primary color mixing method, such as red plus green, which generates yellow in additive color mixing.

[0029] Weighted graph drawing: In the case of complete filling, the user has selected an antibacterial drug and suspected infection site and characteristics. The system will generate a weighted graph below the basic graph. At this time, the data used is the graph and weighted value database. The system uses the aforementioned weighting formula for multi-pathogen weighting calculation to obtain the overall susceptibility rate. At this time, the value of the overall susceptibility rate is mapped to the coordinates of the six dimensions to form a weighted graph in the form of a parallelepiped. Users can observe the area size or coordinate axis value of the weighted graph to determine the sensitivity probability of the antibacterial drug in a specific infection site and characteristics. The larger the area or the larger the value on the axis, the higher the sensitivity, and the more recommended it is used.

[0030] Multi-scheme comparison: Users can select different antibacterial drug combination schemes, and the system can generate radar charts for each scheme and display them superimposed or side by side. By directly comparing the size, shape, and extension of the polygons on each dimension, users can instantly determine which scheme has a better antibacterial spectrum coverage, which specific drug category has advantages or weaknesses.

[0031] Step 4: Interaction and decision support The generated radar chart is highly interactive, such as hover tips: mouse hover over a polygon area or an axis can show the specific drug name, the exact sensitivity value. Such as dynamic adjustment: users can add or delete a certain drug in real time in the interface, the radar chart will immediately recalculate and render, helping to explore the optimal individualized scheme.

[0032] The foregoing description and drawings merely explain and illustrate the invention and the only limit to the scope of the invention is the claims.

[0033] Furthermore, it should be understood that although the description herein is made on the basis of the embodiments, not every embodiment contains only one independent technical solution, and the description herein is only for the sake of clarity, and those skilled in the art should consider the description as a whole, and the technical solutions in each embodiment can be properly combined to form other embodiments which can be understood by those skilled in the art.

Claims

1. A visualization and evaluation system for antibacterial spectrum radar charts, characterized in that: It includes a structured antimicrobial susceptibility database, a multidimensional radar chart generation engine, and a weighted calculation and interactive display module; The structured antimicrobial susceptibility database includes a basic numerical database and a graphical and weighted numerical database; The multidimensional radar chart generation engine maps the drug's sensitivity rate to six types of pathogens into graphical coordinates; The weighted calculation and interactive display module calculates the overall sensitivity rate of antimicrobial drugs to the target pathogen based on the infection site and characteristics.

2. The antibacterial spectrum radar chart visualization evaluation system according to claim 1, characterized in that: The basic numerical database is based on Excel spreadsheets.

3. The antibacterial spectrum radar chart visualization evaluation system according to claim 1, characterized in that: The multi-dimensional radar image generation engine supports the overlay and comparison of multiple drug radar images and uses the three primary colors of light mixing method to display overlapping areas.

4. The antibacterial spectrum radar chart visualization evaluation system according to claim 1, characterized in that: The basic numerical database contains the correspondence between at least 45 antimicrobial drugs and 89 pathogenic bacteria in 6 categories.

5. The antibacterial spectrum radar chart visualization evaluation system according to claim 3, characterized in that: The six categories of pathogenic bacteria include Gram-positive cocci, Gram-positive bacilli, aerobic non-fermenting Gram-negative bacilli, aerobic Gram-negative bacilli-Enterobacteria, aerobic Gram-negative bacilli-non-Enterobacteria, and anaerobic bacteria.

6. A method for visualizing an antibacterial spectrum radar chart according to any one of claims 1-5, characterized in that, Includes the following steps: Step 1: Input and parsing query conditions. Users select the following key parameters through the interface: Step 2: Data retrieval and weighted integration; Step 3: Based on the antibacterial drug selected by the user, the system converts the "sensitivity rate" or "coverage probability" dataset calculated in the previous step into a multi-dimensional radar chart; Step 4: Interaction and Decision Support The generated radar chart is highly interactive, with features such as hover tooltips: hovering the mouse over a polygonal area or axis displays the specific drug name, precise sensitivity values, and dynamic adjustments. Users can add or delete drugs in real time, and the radar chart will be immediately recalculated and rendered to help explore the optimal personalized approach.

7. The method for visualizing an antibacterial spectrum radar chart according to claim 6, characterized in that, Step one includes the following two cases: If only the required fields are filled in, it is considered the default case, and only the basic numerical database is entered to retrieve data and generate a graph; if both required and optional fields are filled in, it is considered a special case, and the basic numerical database is entered first to retrieve data, and then the graph and weighted numerical database is entered to retrieve weighted data, calculate and generate a weighted graph.

8. The method for visualizing an antibacterial spectrum radar chart according to claim 6, characterized in that, Step two includes the following steps: after receiving the query conditions, the system starts the data processing engine and first performs data positioning: based on the antibacterial drug selected by the user, the corresponding data column set is accurately located in the basic numerical database. These data are the basis for subsequent analysis. The "drug-bacteria" sensitivity data is converted into radar chart coordinate points through the coordinate mapping method to generate a six-dimensional radar chart for each drug. In special cases, multi-pathogen weighted calculations are performed: If the user selects the infection site and characteristics, the system does not simply list them, but executes a crucial weighted average algorithm. This algorithm combines the sensitivity rates of different pathogen categories to the same antimicrobial drug. By default, it uses an arithmetic mean, but the system allows the introduction of weighting coefficients to calculate an overall sensitivity rate representing "coverage of this group of pathogens." The formula used to calculate the weighted average algorithm for the overall antimicrobial drug sensitivity rate is as follows: Formula illustration: For antimicrobial drug i, its overall sensitivity against the selected set of pathogens: Where n is the number of selected pathogen categories, wj is the weight of the j-th pathogen, and rij is the sensitivity rate of the j-th pathogen to drug i in the database.

Images