A method for detecting microbial drug sensitivity

Abstract

The application discloses a microbial drug sensitivity detection method. The application comprises: loading a microbial database, establishing a corresponding relationship between secondary microbial names and primary categories; loading a drug breakpoint database based on the "antimicrobial drug sensitivity experiment execution standard"; placing a standard culture dish in an image acquisition device to acquire an image, and calibrating an image analysis system with a vernier caliper; inputting sample information to be detected, acquiring and uploading the image; an image analysis system extracts the position of a drug sensitivity paper sheet, identifies the edge of a bacterial inhibition zone and calculates the diameter; the system extracts corresponding drug breakpoint thresholds based on the microbial category, compares the measured diameter and determines the sensitivity classification (S / I / R); and finally, the result is uploaded to a LIS system. The method can realize the automation, standardization and high efficiency of drug sensitivity detection.

Description

Technical Field

[0001] This application relates to the field of medical testing technology, and in particular to a method for detecting microbial drug susceptibility. Background Technology

[0002] Antimicrobial susceptibility testing (AST) is a crucial step in clinical microbiology testing, widely used to determine the susceptibility of pathogenic microorganisms to antimicrobial drugs, thereby guiding rational drug use in clinical practice and preventing the emergence and spread of drug-resistant bacteria. Traditional susceptibility testing methods primarily employ disk diffusion methods (such as the KB method), determining the susceptibility level of microorganisms to corresponding drugs by measuring the size of the bacterial growth inhibition zone around the drug-containing disk on a petri dish. While this method is simple to operate and relatively inexpensive, the interpretation of results relies on manual visual inspection and measurement, making it susceptible to the influence of operator experience and subjective factors, leading to certain errors and poor repeatability.

[0003] With the development of computer vision and artificial intelligence technologies, microbial antimicrobial susceptibility testing systems based on image acquisition and analysis have emerged. This system acquires images of bacterial growth on petri dishes using a high-definition camera, automatically identifies the edges of inhibition zones and calculates their diameter using image processing technology, and combines this with authoritative standard antimicrobial susceptibility breakpoints from a database to achieve automatic, rapid, and accurate interpretation of antimicrobial susceptibility results. Compared to traditional manual measurements, image analysis systems can significantly improve detection efficiency and accuracy, reduce human error, and meet the clinical demand for high-throughput, high-precision microbial antimicrobial susceptibility testing.

[0004] However, existing technologies suffer from problems such as imperfect image acquisition system structure, insufficient image analysis accuracy, and inconvenient database updates and management, affecting the stability and applicability of the detection system. Therefore, there is an urgent need for a microbial drug susceptibility testing method that integrates a microbial database, a drug breakpoint database, and a high-efficiency image acquisition and analysis system to achieve automatic correlation between microbial classification and drug susceptibility interpretation, thereby improving the scientific rigor and practicality of drug susceptibility testing.

[0005] This invention addresses the aforementioned technical problems by providing a microbial drug susceptibility testing method based on a microbial database and a drug breakpoint database, combined with an image acquisition and analysis system. This method enables automated and high-precision interpretation of drug susceptibility results, meeting the practical needs of clinical laboratories for rapid and accurate drug susceptibility testing. Summary of the Invention

[0006] To address the aforementioned technical problems, this application provides a method for detecting microbial drug susceptibility.

[0007] The technical solution provided in this application is described below:

[0008] This application provides a method for detecting microbial antimicrobial susceptibility, including: Load the microbial database, which includes a primary microbial list and a secondary microbial list. The secondary microbial name is associated with its primary microbial category by entering it in the input interface. Load the microbial drug breakpoint database, which is established according to the "Standards for the Execution of Antimicrobial Drug Susceptibility Tests" and includes multiple antimicrobial drugs against different microorganisms and their inhibition zone diameter breakpoint judgment criteria; A standard culture dish with antimicrobial susceptibility testing discs is placed in an image acquisition device, and bacterial growth images are acquired through the image acquisition device. The bacterial growth image is input into the image analysis system to obtain the diameter measurement result, and the image analysis system is calibrated according to the diameter of the inhibition zone measured by the vernier caliper to make the diameter measurement result consistent with the actual value; Place the culture dish containing the sample to be tested in the image acquisition device, input the sample number and the name of the secondary microorganism, and transmit the acquired sample image to the image analysis system; The image analysis system extracts the location of the drug sensitivity test strip and identifies the edge of the inhibition zone and calculates its diameter using either diameter or radius mode. Based on the primary microbial category associated with the secondary microbial name, the diameter breakpoint threshold of the corresponding drug is extracted from the drug breakpoint database, and the diameter measured by the image analysis system is compared with the diameter breakpoint threshold. Based on the comparison result, the sensitivity classification of the corresponding microorganism to each drug is determined, wherein the sensitivity classification includes: sensitive (S), intermediate (I), and resistant (R). The diameter and sensitivity classification obtained from the tests are transmitted to the Laboratory Information System (LIS).

[0009] Optionally, the primary microbial list includes: Enterobacteriaceae, Pseudomonas aeruginosa, Acinetobacter spp., Burkholderia cepacia complex, Stenotrophomonas maltophilia, other non-Enterobacteriaceae, Staphylococcus spp., Enterococcus spp., Haemophilus influenzae and Haemophilus parainfluenzae, Neisseria gonorrhoeae, Streptococcus pneumoniae, β-hemolytic streptococci, viridans streptococci, Neisseria meningitidis, and anaerobic bacteria.

[0010] Optionally, the image analysis system extracts the edge of the inhibition zone using an algorithm that combines edge detection and morphological filtering, and calculates the diameter of the inhibition zone using a circular fitting algorithm.

[0011] Optionally, the image acquisition device includes a high-definition camera unit with a directional light source and a calibration scale.

[0012] Optionally, the results of the sensitivity classification are displayed in a color chart through a graphical interface, where "Sensitive S" is green, "Intermediate I" is yellow, and "Drug-resistant R" is red.

[0013] Optionally, the high-definition camera unit is used to capture images of bacterial growth on the culture dish, and adjusts the height and performs preliminary focusing via a connected telescopic rod.

[0014] Optionally, the petri dish placement area is located directly below the high-definition camera unit, and the bottom of the petri dish placement area is made of black material.

[0015] Optionally, before identifying the inhibition zone, the image analysis system performs distortion correction, brightness equalization, and background noise suppression processing on the bacterial growth image.

[0016] Optionally, when extracting the edge of the inhibition zone, the image analysis system identifies overlapping inhibition zones or inhibition zones with incomplete boundaries, and corrects the edge of the inhibition zone using a compensation algorithm based on contour continuity analysis before calculating the corresponding diameter.

[0017] Optionally, the method further includes: performing a logical consistency check on the diameter and sensitivity classification results before transmitting the test results to the Laboratory Information System (LIS), and generating a review prompt message when the check result exceeds a preset reasonable range.

[0018] As can be seen from the above technical solutions, this application has the following beneficial effects: The microbial drug susceptibility testing method described in this invention achieves systematization and precision in microbial classification by loading a microbial database containing primary and secondary microbial lists; and ensures the scientific rigor and authority of the drug susceptibility determination criteria by utilizing a microbial drug breakpoint database established based on authoritative standards.

[0019] The system employs an image acquisition device to capture bacterial growth images, combined with an image analysis system for automatic measurement and calibration of the inhibition zone diameter. This avoids subjective errors inherent in traditional manual measurements, improving the accuracy and repeatability of test results. Furthermore, the technology for automatically identifying the location of the susceptibility testing strip and the edge of the inhibition zone significantly enhances the automation level and work efficiency of the testing process.

[0020] By comparing the measured diameter of the inhibition zone with the corresponding drug susceptibility breakpoint threshold in the database, the sensitivity classification of microorganisms to various antimicrobial drugs (sensitive S, intermediate I, resistant R) can be determined quickly and accurately, thereby providing scientific and rational drug use guidance for clinical practice and reducing the risk of drug resistance.

[0021] Furthermore, this method can directly transmit test results to the Laboratory Information System (LIS), enabling efficient data management and tracking, and facilitating subsequent statistical analysis and quality control. Attached Figure Description

[0022] Figure 1 This is a schematic flowchart of an embodiment of the method provided in this application; Figure 2 This is a schematic diagram showing the growth of bacteria in a petri dish; Figure 3 This is a schematic diagram of image analysis; Figure 4 This is a diagram illustrating the display and editing of a microbial database; Figure 5 This is a schematic diagram illustrating the display and editing of the microbial drug breakpoint database; Figure 6 This is a schematic diagram showing the results of microbial antimicrobial susceptibility testing. Detailed Implementation

[0023] In this application, the terms "upper", "lower", "left", "right", "front", "rear", "top", "bottom", "inner", "outer", "middle", "vertical", "horizontal", "lateral", "longitudinal" and other terms indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only used to describe the relative positional relationship between the components or parts and do not specifically limit the specific installation orientation of each component or part.

[0024] Furthermore, in addition to indicating location or positional relationship, some of the aforementioned terms may also have other meanings. For example, the term "above" may also be used in some cases to indicate a certain dependency or connection relationship. Those skilled in the art can understand the specific meaning of these terms in this application based on the specific circumstances.

[0025] Furthermore, the terms "installation," "setup," "equipped with," "connection," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral structure; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium, or an internal connection between two devices, components, or parts. Those skilled in the art can understand the specific meaning of these terms in this application based on the specific circumstances.

[0026] Furthermore, the structures, proportions, sizes, etc., drawn in the accompanying drawings of this application are only used to complement the content disclosed in the specification for those skilled in the art to understand and read, and are not intended to limit the conditions under which this application can be implemented. Therefore, they have no substantial technical significance. Any modification to the structure, change in the proportional relationship, or adjustment of the size, without affecting the effects and purposes that this application can produce, should still fall within the scope of the technical content disclosed in this application.

[0027] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0028] See Figures 1 to 5 This application first provides an embodiment of a microbial antimicrobial susceptibility testing method, which includes: S101. Load the microbial database, which includes a primary microbial list and a secondary microbial list. The secondary microbial name is associated with its primary microbial category by entering it in the input interface. First, a pre-set microbial database is loaded into the computer system. This database includes a primary microbial list and a secondary microbial list. The primary microbial list is classified according to genus or family, such as "Staphylococcus aureus" and "Pseudomonas aeruginosa". The secondary microbial list includes specific species names, such as "Staphylococcus aureus ATCC 25923" and "Escherichia coli ATCC 25922".

[0029] Users can enter the name of the secondary microorganism of the sample to be tested through the drop-down menu or search box on the system interface. The system will automatically associate it with the corresponding primary microorganism category, which will be used to select the correct comparison standard from the drug susceptibility breakpoint database.

[0030] S102. Load the microbial drug breakpoint database, which is established according to the "Standard for the Execution of Antimicrobial Drug Sensitivity Tests" and includes multiple antimicrobial drugs against different microorganisms and their inhibition zone diameter breakpoint judgment criteria. Subsequently, the system loads a pre-established database of antimicrobial drug breakpoints. This database, constructed according to standards for antimicrobial susceptibility testing (such as CLSI or EUCAST standards), contains breakpoint values ​​for the inhibition zone diameters of multiple common microorganisms and antimicrobial drug combinations. For example, for the combination of Staphylococcus aureus and penicillin, the breakpoint might be S ≥ 29 mm and R < 29 mm. The database is continuously updated to ensure synchronization with clinical reference standards.

[0031] The antimicrobial drug breakpoint database should comply with the "Standards for the Implementation of Antimicrobial Susceptibility Testing" (CLSIM100). For example, when the primary microbial list is displayed and "Enterobacteria" is entered in the editing area, the corresponding antimicrobial drugs and inhibition zone diameter breakpoints are as follows: Ak (amikacin), inhibition zone diameter breakpoint ≥20mm indicates sensitive (S), 17-19mm indicates intermediate (I), and ≤16mm indicates resistant (R); AMP (ampicillin), inhibition zone diameter breakpoint ≥17mm indicates sensitive (S), and 14-16mm indicates intermediate. (I) ≤13mm indicates resistance (R); Cn (Sinofaxin), inhibition zone diameter breakpoint ≥19mm indicates sensitivity (S), 15-18mm indicates intermediate (I), ≤14mm indicates resistance (R); Kz (Cefazolin), inhibition zone diameter breakpoint ≥23mm indicates sensitivity (S), 20-22mm indicates intermediate (I), ≤19mm indicates resistance (R); ATM (Aztreonam), inhibition zone diameter breakpoint ≥21mm indicates sensitivity (S), 18-20mm indicates intermediate (I), ≤17mm indicates resistance (R), etc. Once verified correctly, it can be used long-term without modification. When the judgment criteria are modified, they should be modified and saved in the microbial drug breakpoint database display and editing interface.

[0032] S103. Place a standard culture dish with antimicrobial susceptibility test strips in an image acquisition device and acquire bacterial growth images through the image acquisition device. A pre-prepared culture dish containing a standard bacterial strain (e.g., using a standard strain such as ATCC25922) with antimicrobial susceptibility testing discs is placed in the image acquisition device. In one embodiment, the image acquisition device may include components such as an image acquisition system main body, a high-definition camera unit, and a culture dish placement area. The main body of the acquisition system is shielded with light-blocking material to reduce external light interference; the high-definition camera unit, in conjunction with an LED ring light source, provides uniform illumination and automatically adjusts the focus by driving the lens with a built-in motor; the culture dish placement area has a black background, which helps improve image contrast and facilitates the identification of the edge of the inhibition zone.

[0033] After the high-definition camera unit acquires images, the images are transmitted to the computer system for storage and used as a calibration benchmark for subsequent image analysis models.

[0034] S104. Input the bacterial growth image into the image analysis system to obtain the diameter measurement result, and calibrate the image analysis system according to the diameter of the inhibition zone measured by the vernier caliper so that the diameter measurement result is consistent with the actual value. The standard sample images are input into the image analysis system, which automatically identifies the edge of the inhibition zone and calculates its initial diameter. The diameter of the same inhibition zone is then manually measured (using calipers for precise measurement), and the actual diameter is compared with the image recognition result to generate a calibration factor. This factor is used to adjust the pixel diameter value output by the image system to match the actual physical value, thereby ensuring the accuracy of subsequent measurements.

[0035] In an optional embodiment, the image analysis system performs distortion correction, brightness equalization, and background noise suppression processing on the bacterial growth image before identifying the inhibition zone.

[0036] In an optional embodiment, the image analysis system preprocesses the acquired bacterial growth image before performing inhibition zone identification to improve the accuracy and stability of subsequent inhibition zone edge extraction.

[0037] Specifically, the distortion correction is used to eliminate geometric deformation caused by the perspective effect of the image acquisition device lens or the tilt of the culture dish. The image analysis system performs geometric mapping transformation on the bacterial growth image based on the pre-calibrated intrinsic parameters of the camera unit, so that the outline of the culture dish in the image remains a regular circle.

[0038] The brightness equalization is used to eliminate local overly bright or dark areas caused by uneven illumination from directional light sources or reflections from the culture dish. The image analysis system adjusts the grayscale distribution of the bacterial growth image through histogram equalization, adaptive brightness compensation, or regional brightness normalization to maintain a consistent contrast between the inhibition zone area and the colony growth area.

[0039] The background noise suppression is used to reduce the interference of culture medium texture, stray reflection or camera noise on the identification of inhibition zones. The image analysis system smooths non-target small-scale noise through median filtering, Gaussian filtering or morphological opening and closing operations, thereby preserving the true edge structure of the inhibition zone.

[0040] The processing described in this embodiment can standardize the bacterial growth images entering the inhibition zone recognition process in terms of geometric shape, brightness distribution, and background interference, thereby improving the accuracy and repeatability of subsequent edge detection and diameter calculation.

[0041] In an optional embodiment, when extracting the edge of the inhibition zone, the image analysis system identifies overlapping inhibition zones or inhibition zones with incomplete boundaries, and corrects the edge of the inhibition zone using a compensation algorithm based on contour continuity analysis before calculating the corresponding diameter.

[0042] In an optional embodiment, when extracting the edge of the inhibition zone, the image analysis system automatically identifies and compensates for overlapping inhibition zones or inhibition zones with incomplete boundaries in the culture dish, so as to avoid the influence of abnormal morphology on the diameter measurement results.

[0043] Specifically, the image analysis system first performs preliminary segmentation of the antibacterial zone outline based on the gray-scale gradient change and circular features of the antibacterial zone. When the edge spacing between adjacent antibacterial zones is less than a preset threshold, or when there is an obvious broken area in the antibacterial zone outline, it is determined to be an overlapping antibacterial zone or an antibacterial zone with incomplete boundaries.

[0044] For inhibition zones that are identified as abnormal, the image analysis system performs contour continuity analysis on the identified valid edge points, extracts the direction information of continuous edge segments, and compensates and fits the missing areas based on a circular constraint model to make the edge of the inhibition zone form a closed contour.

[0045] After the compensation correction is completed, the image analysis system performs a circular fitting operation on the corrected inhibition zone outline to calculate the corresponding inhibition zone diameter, so that stable and reliable measurement results can still be obtained even in the presence of local occlusion, overlap or uneven growth.

[0046] This embodiment enables drug sensitivity testing to maintain high recognition tolerance and measurement consistency even under complex culture conditions.

[0047] S105. Place the culture dish of the sample to be tested in the image acquisition device, input the sample number and the name of the secondary microorganism, and transmit the acquired sample image to the image analysis system; The actual clinical sample culture dish to be tested is placed in the image acquisition device. The operator enters the sample number and the presumed secondary microorganism name into the system and starts the image acquisition operation to obtain an image of the inhibition zone around the drug sensitivity test strip on the culture dish. After compression and standardization, the acquired image is transmitted to the image analysis system for analysis.

[0048] S106. Extract the position of the drug sensitivity paper using the image analysis system, and identify the edge of the inhibition zone and calculate its diameter using diameter mode or radius mode; Image analysis systems can use pre-trained deep learning algorithms or traditional edge detection algorithms (such as Canny and Hough transforms) to automatically extract the location of the drug-sensitive paper in the image and identify the boundary of the inhibition zone around the drug-sensitive paper as the center.

[0049] The system supports two detection modes: Diameter mode: Directly identifies the two ends of the antibacterial zone boundary and calculates its pixel diameter; Radius mode: Using the center of the drug sensitivity paper as the origin, rays are emitted in multiple directions. After identifying the boundary points, the average radius is calculated and the diameter is deduced.

[0050] Based on the aforementioned calibration coefficients, the system outputs the diameter value of the inhibition zone in actual units (e.g., mm).

[0051] The diameter measurement mode is primarily used to measure the inhibition zone diameter of drugs with complete inhibition zones, directly measuring the diameter of the inhibition zones for different drugs. The radius measurement mode is mainly used to measure the inhibition zones of two different drugs where their diameters cannot be measured due to overlap. By measuring the radius of the inhibition zones for different drugs, the corresponding diameter is calculated. Simultaneously, the image analysis system can extract the code from the drug sample. The effect is that the image analysis system extracts the code from the drug sample, measures the diameter of the inhibition zone using either the diameter or radius mode, further associates the secondary microbial name with the primary microbial list, finds the corresponding drug and inhibition zone diameter inflection point in the primary microbial list, and compares it with the standard inhibition zone diameter inflection point to calculate the microbial drug sensitivity.

[0052] For example: In the microbial drug susceptibility test result display interface, enter the sample number in the number display and editing area, and enter the microbial name "Escherichia coli" in the secondary microbial name display and editing area. At this time, the secondary microbial name is associated with the primary microbial list, and then the corresponding drugs "Ak (Amikacin), AMP (Ampicillin), Cn (Sinoxacin), Kz (Cefazolin), FEP (Cefepime), ATM (Aztreonam)" are found in the primary microbial list. Click "Start Detection," and the image analysis system extracts the code of the drug paper. At the same time, it measures the diameter of the inhibition zone of the drug through diameter mode or radius mode, and further compares it with the drug susceptibility diameter breakpoint of different drugs (Ak (Amikacin), inhibition zone diameter breakpoint ≥20mm is sensitive (S), 17-19mm). The following parameters were compared and calculated to further analyze the drug susceptibility of microorganisms: AMP (ampicillin), inhibition zone diameter breakpoint ≥17mm, sensitive (S), 14-16mm, intermediate (I), ≤13mm, resistant (R); Cn (sinoxacin), inhibition zone diameter breakpoint ≥19mm, sensitive (S), 15-18mm, intermediate (I), ≤14mm, resistant (R); Kz (cefazolin), inhibition zone diameter breakpoint ≥23mm, sensitive (S), 20-22mm, intermediate (I), ≤19mm, resistant (R); ATM (aztreonam), inhibition zone diameter breakpoint ≥21mm, sensitive (S), 18-20mm, intermediate (I), ≤17mm, resistant (R), etc.

[0053] S107: Extracting diameter breakpoint thresholds of corresponding drugs from a drug breakpoint database according to the first-level microorganism category associated with the input second-level microorganism name, comparing the diameter measured by the image analysis system with the diameter breakpoint threshold, and determining the sensitivity classification of the corresponding microorganism to each drug based on the comparison result, wherein the sensitivity classification comprises: sensitive S, intermediate I, resistant R; The system automatically matches the first-level classification according to the input second-level microorganism name, and extracts the corresponding antibacterial drug breakpoint standard from the drug breakpoint database. Comparing the inhibition zone diameter measured by the image analysis system with the breakpoint value of the corresponding drug in the database: If the diameter ≥ S threshold, it is determined as sensitive (S); If R < diameter < S, it is determined as intermediate (I); If the diameter ≤ R threshold, it is determined as resistant (R).

[0054] For example, if the inhibition zone diameter of a sample against amoxicillin measured by the system is 28mm, the S threshold of the microorganism is 29mm, and the R threshold is 20mm, the judgment result is intermediate I.

[0055] S108: Transmitting the detected diameter and sensitivity classification to a Laboratory Information System (LIS).

[0056] The finally detected inhibition zone diameter and the corresponding sensitivity classification result are transmitted to a hospital or laboratory information system (LIS) through an interface, for case recording, doctor viewing and statistical analysis.

[0057] The method can be integrated with automated laboratory workflows, is suitable for standardized operation procedures of daily clinical drug sensitivity tests, and can also be extended to application scenarios such as new drug screening and basic microorganism research.

[0058] In an alternative embodiment, the image analysis system uses a multi-model fusion algorithm to identify the edge of the inhibition zone. The system combines a fusion analysis strategy based on traditional image processing methods (such as gray threshold segmentation, edge detection, contour extraction) and a deep learning-based image segmentation model (such as U-Net, Mask R-CNN), so as to improve the recognition accuracy under the conditions of complex background, blurred edges or multiple interferences.

[0059] Specifically, the image is first subjected to noise removal, brightness equalization and enhancement by a traditional image preprocessing module, and then input into a deep neural network model for pixel-level edge segmentation to extract an accurate inhibition zone boundary. For the boundary result output by the model, the system can automatically evaluate the edge integrity and select the optimal recognition result, and meanwhile provide a high-confidence basis for the subsequent diameter calculation.

[0060] This implementation method further enhances the adaptability and robustness of the system of the present invention under different culture conditions and different colony morphologies.

[0061] To improve the clinical reliability of test results, the present invention may also include a result verification and manual intervention mechanism.

[0062] When the diameter of the inhibition zone identified by the image analysis system deviates significantly from the expected value (based on historical data of bacterial species and drugs), the system can automatically issue a prompt, mark it as "requiring manual verification," and display the image and measurement results for manual technicians to confirm and adjust.

[0063] In addition, the system provides a "manual dragging of boundary points" function, allowing technicians to fine-tune the recognized edges using a mouse or touchscreen, thereby correcting recognition errors in cases of blurred or overlapping boundaries. The modified results can also be recorded by the system and fed back into the training model for subsequent model optimization and continuous learning.

[0064] This mechanism ensures the accuracy and controllability of image recognition under high-throughput operation.

[0065] In another preferred embodiment, the petri dish placement area of ​​the image acquisition device is designed as an adjustable structure, which can be compatible with petri dishes of various sizes (such as 90mm, 100mm, 150mm in diameter, etc.) and supports the simultaneous placement of multiple petri dishes for batch image acquisition.

[0066] The specific design includes: multiple grooved positioning frames with sliding guides and elastic limiting devices to ensure that culture dishes of different sizes can stably fit the acquisition surface and prevent tilting or displacement. The system automatically identifies the boundary area of ​​each culture dish and performs batch image partitioning processing through image segmentation, improving image acquisition efficiency.

[0067] This implementation method is applicable to diverse laboratory testing scenarios and has good versatility and scalability.

[0068] In one optional implementation, the primary microbial catalog in the microbial database specifically includes the following categories: Enterobacteriaceae, Pseudomonas aeruginosa, Acinetobacter spp., Burkholderia cepacia complex, Stenotrophomonas maltophilia, other non-Enterobacteriaceae, Staphylococcus spp., Enterococcus spp., Haemophilus influenzae and Haemophilus parainfluenzae, Neisseria gonorrhoeae, Streptococcus pneumoniae, β-hemolytic streptococci, viridans streptococci, Neisseria meningitidis, anaerobic bacteria.

[0069] After receiving the input secondary microbial name, the image analysis system can automatically call the primary microbial category corresponding to the secondary name for subsequent inflection point extraction and sensitivity assessment.

[0070] This approach improves database retrieval efficiency and result matching accuracy through structured classification, making it particularly suitable for large-scale clinical microbial sample testing scenarios.

[0071] In another optional embodiment, the image acquisition device includes a high-definition camera unit with a resolution of not less than 8 megapixels and is equipped with a directional LED ring light source to provide uniform illumination and avoid uneven image exposure due to differences in ambient light.

[0072] The camera is equipped with a laser calibration scale or embedded size scale, which can synchronously record the calibration length in the image. This is used by the subsequent image analysis system to automatically convert the pixel scale to the actual size, thereby calibrating and correcting the diameter of the antibacterial zone.

[0073] This structure can significantly improve image quality, reduce human intervention, and enhance the objectivity and consistency of measurements.

[0074] In this implementation, the high-definition camera unit is integrated with a fixed bracket by connecting a retractable mechanical rod, and has adjustable height, angle and focus functions.

[0075] Operators can adjust the camera distance up and down according to the size of the petri dish and the characteristics of bacterial growth to ensure that the inhibition zone is in the optimal imaging focal plane; at the same time, they can also perform preliminary focusing based on the real-time preview image provided by the system to improve image clarity.

[0076] The above structure effectively enhances the flexibility and accuracy of image acquisition under different culture conditions.

[0077] In another specific embodiment, the petri dish placement area of ​​the image acquisition device is located directly below the high-definition camera unit, and an embedded groove structure is used to ensure stable positioning of the petri dish.

[0078] To enhance image contrast, the bottom of the culture dish placement area is preferably made of matte black ABS or painted metal material, which is used to absorb background stray light and form a clear brightness difference between the colony and the background, thereby improving the image analysis system's accuracy in recognizing the boundary of the inhibition zone.

[0079] This setting reduces light reflection interference, improves imaging quality, and is compatible with various culture dish diameters.

[0080] In one user interface optimization implementation, the image analysis system visually displays the diameter of the inhibition zone and sensitivity classification of each detected antimicrobial drug through a color graphical interface.

[0081] Specifically: Sensitive (S) is displayed as a green bar, Intermediate (I) as a yellow bar, and Resistant (R) as a red bar; at the same time, the corresponding drug, measured diameter, judgment criteria and conclusion can be displayed in the list on the right side of the interface.

[0082] This display method helps clinicians quickly identify important information, improves efficiency, and is compatible with the style of hospital information systems.

Claims

1. A method for detecting antimicrobial susceptibility, characterized in that, include: Load the microbial database, which includes a primary microbial list and a secondary microbial list. The secondary microbial name is associated with its primary microbial category by entering it in the input interface. Load the microbial drug breakpoint database, which is established according to the "Standards for the Execution of Antimicrobial Drug Susceptibility Tests" and includes multiple antimicrobial drugs against different microorganisms and their inhibition zone diameter breakpoint judgment criteria; A standard culture dish with antimicrobial susceptibility testing discs is placed in an image acquisition device, and bacterial growth images are acquired through the image acquisition device. The bacterial growth image is input into the image analysis system to obtain the diameter measurement result, and the image analysis system is calibrated according to the diameter of the inhibition zone measured by the vernier caliper to make the diameter measurement result consistent with the actual value; Place the culture dish containing the sample to be tested in the image acquisition device, input the sample number and the name of the secondary microorganism, and transmit the acquired sample image to the image analysis system; The image analysis system extracts the location of the drug sensitivity test strip and identifies the edge of the inhibition zone and calculates its diameter using either diameter or radius mode. Based on the primary microbial category associated with the secondary microbial name, the diameter breakpoint threshold of the corresponding drug is extracted from the drug breakpoint database, and the diameter measured by the image analysis system is compared with the diameter breakpoint threshold. Based on the comparison result, the sensitivity classification of the corresponding microorganism to each drug is determined, wherein the sensitivity classification includes: sensitive (S), intermediate (I), and resistant (R). The diameter and sensitivity classification obtained from the tests are transmitted to the Laboratory Information System (LIS).

2. The microbial antimicrobial susceptibility testing method according to claim 1, characterized in that, The primary microbial list includes: Enterobacteriaceae, Pseudomonas aeruginosa, Acinetobacter spp., Burkholderia cepacia complex, Stenotrophomonas maltophilia, other non-Enterobacteriaceae, Staphylococcus spp., Enterococcus spp., Haemophilus influenzae and Haemophilus parainfluenzae, Neisseria gonorrhoeae, Streptococcus pneumoniae, β-hemolytic streptococci, viridans streptococci, Neisseria meningitidis, and anaerobic bacteria.

3. The microbial antimicrobial susceptibility testing method according to claim 1, characterized in that, The image analysis system extracts the edge of the inhibition zone using an algorithm that combines edge detection and morphological filtering, and calculates the diameter of the inhibition zone using a circular fitting algorithm.

4. The microbial antimicrobial susceptibility testing method according to claim 1, characterized in that, The image acquisition device includes a high-definition camera unit with a directional light source and a calibration scale.

5. The microbial antimicrobial susceptibility testing method according to claim 1, characterized in that, The results of the sensitivity classification are displayed in a color chart through a graphical interface, where "Sensitive S" is green, "Intermediate I" is yellow, and "Drug-resistant R" is red.

6. The microbial antimicrobial susceptibility testing method according to claim 4, characterized in that, The high-definition camera unit is used to capture images of bacterial growth on the culture dish, and the height is adjusted and preliminary focusing is performed via a connected telescopic rod.

7. The microbial antimicrobial susceptibility testing method according to claim 4, characterized in that, The petri dish placement area is located directly below the high-definition camera unit, and the bottom of the petri dish placement area is made of black material.

8. The microbial antimicrobial susceptibility testing method according to claim 1, characterized in that, Before identifying the inhibition zone, the image analysis system performs distortion correction, brightness equalization, and background noise suppression on the bacterial growth image.

9. The microbial antimicrobial susceptibility testing method according to claim 1, characterized in that, When extracting the edge of the inhibition zone, the image analysis system identifies overlapping or incomplete inhibition zones, corrects the edge of the inhibition zone using a compensation algorithm based on contour continuity analysis, and then calculates the corresponding diameter.

10. The microbial antimicrobial susceptibility testing method according to claim 1, characterized in that, The method further includes: performing a logical consistency check on the diameter and sensitivity classification results before transmitting the test results to the Laboratory Information System (LIS), and generating a review prompt message when the check result exceeds a preset reasonable range.

Images