Helicobacter pylori drug sensitivity quantitative detection system

Abstract

The invention discloses a helicobacter pylori drug sensitivity quantitative detection system, and relates to the technical field of drug sensitivity quantitative detection.The helicobacter pylori drug sensitivity quantitative detection system comprises the steps that a drug sensitivity detection cabin module is constructed and used for providing and maintaining a specific gas phase environment and temperature for helicobacter pylori growth, and a reaction unit containing a microwell plate is constructed; constructing an optical detection module for scanning a reaction unit of the microwell plate at a preset time to generate a multi-time-sequence optical image of flora containing spatial information; a space game network construction module, which carries out virtual partition division on the flora based on the multi-time sequence optical image, constructs a game network among the flora based on the virtual partition, and analyzes the growth situation of the flora based on the game network to obtain drug sensitivity quantitative data of the helicobacter pylori; constructing a visual module for visually displaying the effect of the drug culture solution on the helicobacter pylori and drug sensitivity quantitative data; according to the invention, by analyzing the spatial game behavior in the helicobacter pylori flora, the early, sensitive and more insight quantitative detection of the drug effect is realized.

Description

Technical Field

[0001] This invention relates to the field of drug susceptibility quantitative detection technology, specifically to a Helicobacter pylori drug susceptibility quantitative detection system. Background Technology

[0002] Antimicrobial susceptibility testing of Helicobacter pylori is an important technical means to guide clinical eradication programs and address the increasingly serious problem of antibiotic resistance. Currently, mainstream standard methods such as agar dilution method, E-test method and turbidimetric method based on microplate culture all rely on the observation and judgment of the macroscopic phenotype of bacterial populations under the action of drugs.

[0003] The methods described above typically treat the entire bacterial colony as a homogeneous whole, determining the minimum inhibitory concentration (MIC) by measuring its growth inhibition at different drug concentrations. However, these methods have inherent limitations. First, their determination relies on macroscopic indicators after the bacterial population has reached a stationary phase, resulting in a long detection cycle and hindering early and rapid efficacy assessment. Second, these methods cannot analyze the spatial heterogeneity within the colony, ignoring the complex interactions and behavioral strategy evolution among individual microorganisms, thus failing to identify potential drug-resistant subpopulations and their dynamic competition processes. Furthermore, existing technologies primarily provide static and isolated endpoint data, severely lacking spatial and temporal information reflecting the dynamic development of drug efficacy, resulting in a single-dimensional information dimension that cannot comprehensively support accurate medication decisions. These limitations, to some extent, restrict the timeliness and depth of clinical drug susceptibility testing.

[0004] Therefore, developing an intelligent quantitative detection system for Helicobacter pylori drug sensitivity is of great significance. Summary of the Invention

[0005] The purpose of this invention is to provide a quantitative detection system for Helicobacter pylori drug susceptibility testing to address the shortcomings of the prior art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a Helicobacter pylori drug susceptibility quantitative detection system, comprising:

[0007] Antimicrobial susceptibility testing chamber module: used to provide and maintain a specific gaseous environment and temperature for the growth of Helicobacter pylori, and to construct a reaction unit that accommodates microplate-based reaction units;

[0008] Optical detection module: Connected to the drug sensitivity detection chamber module, it is used to scan the reaction unit of the microplate at preset times to generate multi-temporal optical images of the bacterial community containing spatial information;

[0009] The spatial game network module is connected with the optical detection module, and is used for virtually dividing the bacterial flora based on multi-time sequence optical images, constructing a game network among the bacterial flora based on the virtual division, and analyzing the growth trend of the bacterial flora based on the game network to obtain the drug sensitivity quantitative data of the H. pylori.

[0010] The visualization module is connected with the spatial game network module, and is used for visually displaying the effect of the drug culture solution on the H. pylori and the drug sensitivity quantitative data.

[0011] In a preferred embodiment, the drug sensitivity detection cabin module comprises:

[0012] The microenvironment control unit is used for providing a preset gas phase environment for the reaction unit;

[0013] The precision temperature control unit is used for providing a preset temperature suitable for the growth of the H. pylori for the reaction unit;

[0014] The reaction unit bearing unit is used for bearing the reaction unit for the reaction of the H. pylori and the drug.

[0015] In a preferred embodiment, the optical detection module comprises:

[0016] The high-frequency scanning imaging unit is used for performing global high-resolution scanning on a single reaction unit at a preset time interval to obtain a microscopic image of the bacterial flora of the reaction unit;

[0017] The focus control unit is used for maintaining a constant imaging focal plane during the scanning process to ensure the spatial consistency of the microscopic image;

[0018] The multi-group microscopic images of the reaction unit are used for generating multi-time sequence optical images of the bacterial flora.

[0019] In a preferred embodiment, the spatial game network module comprises:

[0020] The virtual division unit is used for dynamically dividing the bacterial flora image of each reaction unit into a plurality of virtual divisions that interact with each other based on the morphological characteristics and spatial distribution of the bacterial flora;

[0021] The dynamic feature extraction unit is used for generating initial morphological characteristics of the virtual division and extracting time sequence growth dynamics characteristics of the bacterial flora in the virtual division in real time;

[0022] The initial morphological characteristics include edge characteristics and initial density uniformity of the bacterial flora in the virtual division, and the time sequence growth dynamics characteristics include local biomass accumulation rate, metabolic activity change trend and morphological stability index;

[0023] The game network construction unit is used for constructing a game network among the bacterial flora based on the virtual division and the time sequence growth dynamics characteristics.

[0024] In one preferred embodiment, the game network construction unit comprises:

[0025] an instinct core subunit for generating a unique subcode of the virtual partition based on the initial morphological features and spatial position of the virtual partition, the unique subcode being used to represent the initial behavior preference of the virtual partition;

[0026] a strategy element generation subunit for simulating the behavior strategy of different virtual partitions based on the initial behavior preference and the time-series growth dynamics features and generating strategy elements;

[0027] a partition game subunit for analyzing the game results between the virtual partitions and generating the drug sensitivity quantitative test results.

[0028] In one preferred embodiment, the strategy element generation subunit is configured to simulate the behavior strategy of different virtual partitions based on the initial behavior preference and the time-series growth dynamics features and generate strategy elements by:

[0029] generating the initial strategy element concentration of the virtual partition based on the initial behavior preference, the initial strategy element concentration comprising an initial expansion element, an initial defense element, an initial cooperation element and an initial dormancy element;

[0030] real-time acquiring the strategy tendency of the microbial community in each virtual partition, the strategy tendency comprising the time-series growth dynamics features, external features and adjacent partition relationship data;

[0031] transmitting the strategy tendency to a strategy logic model to generate dynamic strategy elements;

[0032] wherein the strategy logic model is built-in with a system determination threshold generated by a control test without drugs, and the strategy logic model is configured to generate the dynamic strategy elements by:

[0033] generating a dynamic expansion element when the current growth rate of the virtual partition is significantly higher than the growth vitality threshold in the system determination threshold;

[0034] generating a dynamic defense element when the difference between the growth potential of the adjacent virtual partition and the growth potential of the virtual partition where the virtual partition is located is greater than a preset difference threshold;

[0035] generating a dynamic cooperation element when the difference between the growth potential of the virtual partition where the virtual partition is located and the growth potential of the adjacent virtual partition is less than the preset difference threshold, and the growth rate of both the virtual partition where the virtual partition is located and the adjacent virtual partition is higher than the system growth vitality threshold;

[0036] generating a dynamic dormancy element when the current growth rate of the virtual partition is significantly lower than the system growth vitality threshold.

[0037] In one preferred embodiment, the partition game subunit is configured to analyze the game results between the virtual partitions and generate the drug sensitivity quantitative test results by:

[0038] constructing a dynamic boundary channel on the boundary of multiple virtual partitions;

[0039] setting multiple game conduits in the dynamic boundary channel for mutual transmission of pheromones between virtual partitions;

[0040] in the dynamic boundary channel, based on the concentration of the imported strategy element, performing a predefined game decision rule for calculation;

[0041] The game decision rule includes antagonistic cancellation of expansion elements and defense elements, high concentration of cooperation elements to modulate the intensity of confrontation, and hibernation elements to suppress all interactive activities of the channel;

[0042] According to the game decision result, output the boundary moving instruction to drive the dynamic adjustment of the area of the adjacent virtual partition, and record the instantaneous game state of the channel;

[0043] Synchronize the output of all dynamic boundary channels, integrate to generate the real-time territory change map and game state vector of the entire bacterial community space as the time-series global game state vector;

[0044] Based on the comparison of the time-series global game state vector of the drug-containing reaction unit and the drug-free reaction unit, when the game state index representing expansion activity in the drug-containing reaction unit statistically significantly decays, and the game state index representing defense or hibernation becomes dominant, it is determined that the drug concentration is effective;

[0045] And the drug sensitivity quantitative data is the minimum drug concentration that effectively causes the dominant change of the game state.

[0046] In a preferred embodiment, the visualization module comprises:

[0047] A dynamic spatiotemporal evolution unit for receiving the real-time territory change map and game state vector output by the spatial game network module, and visualizing the territory range change and dominant strategy element type of different virtual partitions in the form of a dynamic map, wherein different strategy element types are encoded in different colors;

[0048] A pharmacokinetic curve unit for drawing and displaying a curve representing the change of the index representing the expansion activity of the entire bacterial community over time based on the time-series global game state vector, and synchronously displaying the corresponding curve of the drug-free reaction unit for comparison;

[0049] A drug sensitivity quantitative reporting unit for generating and presenting a final drug sensitivity quantitative detection report including the minimum inhibitory concentration value, the pharmacodynamic evaluation conclusion based on game state analysis, and the bacterial community space game state atlas at key time points.

[0050] In the above technical solution, the technical effects and advantages provided by the present application are:

[0051] 1、The present application can extremely sensitively capture the initial transition of the behavior pattern of the bacterial community under the action of the drug by constructing virtual partitions and simulating the strategy secretion and game interaction of the virtual partitions; when the effective drug exists, the spatial expansion strategy of the bacterial community will be early changed into the defense or dormancy strategy, and the macroscopic change of the game situation is far earlier than the change of the colony size visible to the naked eye; therefore, the system can give a reliable drug efficacy judgment earlier, and saves valuable time for clinical treatment; meanwhile, the system can clearly reveal the heterogeneity inside the bacterial community, for example, even under the action of a high concentration of drug, if a small number of partitions still maintain the expansion strategy, it is a strong indication that there may be a drug-resistant subpopulation, which is deep biological information that cannot be provided by the traditional homogeneous detection method;

[0052] 2、The present application provides rich decision support information superior to the traditional method by introducing a dynamic visualization and quantitative decision-making scheme, the system output is no longer a single minimum inhibitory concentration value, but a comprehensive visualization report covering the spatial structure evolution of the bacterial community, the strategy behavior transition and the drug efficacy kinetic process, the system directly observes how the drug step by step affects and reshapes the spatial structure of the microbial society, understands the action mode and efficacy, greatly assists the clinician in the depth evaluation and accurate prediction of the drug efficacy, in addition, the system can be applied to the evaluation scene of different types of antibacterial drugs and even new drugs that may appear in the future; it not only tells you whether the drug is effective, but also reveals the dynamic process of how the bacterial community becomes resistant step by step, provides an extremely valuable research tool and insight window for understanding the mechanism of drug resistance and developing new antibacterial strategies, and has far-reaching basic research and clinical application value. BRIEF DESCRIPTION OF DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art based on these drawings.

[0054] Figure 1 The system flowchart of the present application.

[0055] Figure 2 The logic block diagram of the present application. DETAILED DESCRIPTION

[0056] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0057] Embodiment 1, please refer to Figure 1 and Figure 2 The helicobacter pylori drug sensitivity quantitative detection system described in the embodiment includes:

[0058] The drug sensitivity detection cabin body module is used for providing and maintaining a specific gas phase environment and temperature for growth of the helicobacter pylori and constructing a reaction unit containing a microplate;

[0059] The optical detection module is connected with the drug sensitivity detection cabin body module and is used for scanning the reaction unit of the microplate at a preset time to generate a multi-time optical image containing spatial information of the bacterial population;

[0060] The spatial game network module is connected with the optical detection module, virtually partitions the bacterial population based on the multi-time optical image, constructs a game network between the bacterial populations based on the virtual partition, analyzes the growth trend of the bacterial population based on the game network, and obtains the drug sensitivity quantitative data of the helicobacter pylori;

[0061] The visualization module is connected with the spatial game network module and is used for visually displaying the effect of the drug culture solution on the helicobacter pylori and the drug sensitivity quantitative data;

[0062] Further, the drug sensitivity detection of the helicobacter pylori is an important technical means for guiding the clinical radical solution and coping with the increasingly serious problem of antibiotic resistance. At present, mainstream standard methods such as the agar dilution method, the E-test method and the turbidimetry method based on microplate culture all rely on observation and judgment of the macro phenotype of the bacterial population under the action of the drug;

[0063] The above method generally regards the whole colony as a homogeneous whole, determines the minimum inhibitory concentration by measuring the growth inhibition of the colony in different concentrations of drugs. However, such method has inherent limitations; first, its judgment relies on the terminal macroscopic index after the bacterial population growth reaches the stationary phase, the detection period is long, and it is difficult to realize early and rapid drug efficacy evaluation. Secondly, these methods cannot analyze the spatial heterogeneity inside the colony, ignore the complex interaction and behavior strategy evolution between microbial individuals, and thus cannot understand the possible existence of drug-resistant subpopulation and its dynamic competition process; in addition, the static and isolated end point data provided by the existing technology seriously lack spatial and time sequence information reflecting the dynamic development process of drug efficacy, resulting in single information dimension, which is difficult to fully support accurate drug use decision. These limitations restrict the timeliness and depth of clinical drug sensitivity detection to some extent;

[0064] The present application can extremely sensitively capture the initial change of the behavior pattern of the bacterial population under the action of the drug by constructing virtual partitions and simulating the strategy element secretion and game interaction; when the effective drug exists, the spatial expansion strategy of the bacterial population will be changed to the defense or dormancy strategy at an early stage. The macroscopic change of this game situation is far earlier than the change of the colony size visible to the naked eye; therefore, the system can give a reliable drug efficacy judgment earlier, which saves valuable time for clinical treatment; at the same time, the system can clearly reveal the heterogeneity inside the bacterial population, for example, even under the action of a high concentration of drug, if a small number of partitions still maintain the expansion strategy, it strongly suggests that there may be a drug-resistant subpopulation, which is deep biological information that cannot be provided by the traditional homogeneous detection method;

[0065] By introducing the scheme of dynamic visualization and quantitative decision, rich decision support information superior to the traditional method is provided, and the system output is no longer an isolated minimum inhibitory concentration value, but a comprehensive visualization report covering the spatial structure evolution of the bacterial population, the strategy behavior change and the drug efficacy kinetics process. The system directly observes how the drug step by step affects and reshapes the spatial structure of the microbial society, understands its action mode and efficacy, greatly assists the clinician in the depth evaluation and accurate prediction of drug efficacy. In addition, the system can be applied to the evaluation of different types of antibacterial drugs and even new drugs that may appear in the future; it not only tells you whether the drug is effective, but also reveals the dynamic process of how the bacterial population becomes resistant step by step, provides an extremely valuable research tool and insight window for understanding the mechanism of drug resistance and developing new antibacterial strategies, and has far-reaching basic research and clinical application value.

[0066] In one embodiment, the drug sensitivity detection cabin module comprises:

[0067] A microenvironment control unit for providing a preset gas phase environment for the reaction unit;

[0068] A precise temperature control unit is configured to provide a preset temperature suitable for growth of H. pylori for the reaction unit;

[0069] A reaction unit carrying unit is configured to carry the reaction unit for reaction of H. pylori and drugs;

[0070] Further, the microenvironment control unit of the drug sensitivity detection cabin module adopts a high-precision gas mixing and conveying system, accurately mixes carbon dioxide, nitrogen and oxygen by a mass flow controller, and continuously introduces the mixed gas into the sealed cabin after humidification by a humidification chamber, and integrates real-time oxygen concentration and humidity sensors to form a closed-loop feedback to maintain a stable micro-oxygen and high-humidity gas phase environment required for bacterial growth. The precise temperature control unit adopts a semiconductor thermoelectric temperature control device combined with a PID algorithm, the hot end surface of which is closely attached to the bottom of the reaction unit, and the hot and cold ends are connected to an efficient heat dissipation system, and the temperature is monitored and adjusted in real time by a plurality of high-precision temperature sensors to ensure that the temperature fluctuation of the culture solution in the reaction unit is stable within 5 degrees Celsius. The reaction unit carrying platform adopts a conventional design scheme, and a standardized microplate holder is used as the carrying main body. The holder is injection molded from engineering plastic with good corrosion resistance and thermal stability, and its structure is completely matched with the size of the international standard microplate. The microplate is precisely positioned and stably placed by four corner positioning pins and a bottom anti-skid rubber pad.

[0071] In one embodiment, the optical detection module comprises:

[0072] A high-frequency scanning imaging unit is configured to perform full-domain high-resolution scanning on a single reaction unit at a preset time interval to obtain micro images of the bacterial population of the reaction unit;

[0073] A focus control unit is configured to maintain a constant imaging focal plane during scanning to ensure spatial consistency of the micro images;

[0074] Multiple sets of micro images of the reaction unit are used to generate multiple time-series optical images of the bacterial population based on time sequence;

[0075] Further, the high-frequency scanning imaging unit adopts a scientific-grade CMOS camera combined with a long working distance objective lens to form a microscopic imaging system. The reaction unit is driven by a precision motorized translation stage to perform two-dimensional plane scanning motion. The camera is automatically triggered by the control system at a preset time interval to complete high-resolution image acquisition of each field of view. The acquired high-resolution images are synthesized into complete global micro images of the bacterial population based on an intelligent stitching algorithm. The focus control unit can adopt a laser focusing system, which projects an auxiliary laser beam onto the culture solution surface and receives the reflected signal to monitor the focal plane position change in real time. The axial position of the objective lens is adjusted by a piezoelectric ceramic driver to achieve sub-micron level focusing stability compensation, ensuring spatial consistency between time-series images. The acquired multiple sets of micro images are transmitted to the image processor in time sequence, and multiple time-series optical images containing spatial structure evolution of the bacterial population are generated by time domain registration and feature extraction algorithm.

[0076] In one embodiment, the spatial game network module comprises:

[0077] a virtual partition unit, configured to dynamically divide the colony image of each reaction unit into a plurality of interacting virtual partitions based on the morphological characteristics and spatial distribution of the colony;

[0078] a dynamic feature extraction unit, configured to generate initial morphological features of the virtual partitions and to extract time-series growth dynamics of the colony in the virtual partitions in real time;

[0079] wherein the initial morphological features include edge features and initial density uniformity of the colony in the virtual partitions, and the time-series growth dynamics include local biomass accumulation rate, metabolic activity change trend and morphological stability index;

[0080] a game network construction unit, configured to construct a game network between the colonies based on the virtual partitions and the time-series growth dynamics;

[0081] Further, the spatial game module comprises the virtual partition unit, the dynamic feature extraction unit and the game network construction unit, wherein the virtual partition unit adopts a deep learning semantic segmentation model based on the U-Net architecture, which is trained by a large number of labeled colony images and can accurately identify and segment continuous bacterial plaques in microscopic images; the segmentation algorithm first extracts morphological features such as edge gradient and texture density of the bacterial colonies, and then dynamically divides the image into a plurality of virtual partitions with unique ID labels according to the spatial neighbor criterion and pixel connectivity; the dynamic feature extraction unit quantitatively analyzes the image sequence of each partition, which uses the optical flow method to calculate the pixel displacement of the colony edge in adjacent frames of images, thereby accurately quantifying the local expansion rate; at the same time, by analyzing the time series change of the average pixel gray value in the partition, the optical density derivative is calculated as an index of metabolic activity change trend; all these time-series growth dynamics are extracted in real time and stored in the database, and the game network construction unit abstracts each virtual partition as a network node, establishes a spatial connection relationship by calculating the edge distance of adjacent partitions, and constructs a game area at the edge of the virtual partition as the game space between adjacent virtual partitions.

[0082] In one embodiment, the game network construction unit comprises:

[0083] an instinct core subunit, configured to generate a unique code of each virtual partition based on the initial morphological features and spatial position of the virtual partition, the unique code being used to represent the initial behavior preference of the virtual partition;

[0084] a strategy element generation subunit, configured to simulate the behavior strategy of different virtual partitions and generate strategy elements based on the initial behavior preference and the time-series growth dynamics;

[0085] a partition game subunit configured to analyze the game results between the virtual partitions and generate the drug susceptibility quantitative test results;

[0086] Further, the behavior core subunit acquires the initial morphological characteristics of the flora in the virtual partitions, uses a multi-dimensional feature fusion coding technology to input the initial morphological characteristics and the spatial positions to which the virtual partitions belong into an improved iris hash algorithm, converts the seed code into a behavior preference vector including an initial expansion tendency coefficient, a defense sensitivity parameter, and the like through a pre-trained behavior mapping matrix, and stores the seed code as a virtual partition behavior preference basic parameter in a database; the strategy element generation subunit constructs a strategy state machine with a space-time attention mechanism by receiving real-time growth dynamics data streams of each virtual partition, synchronously analyzes time-series growth dynamics characteristics streams through a convolutional long short-term memory network, performs tensor fusion on the real-time biomass change gradient, metabolic oscillation frequency, and the like, and the behavior preference vector, and outputs normalized concentration values of four strategy elements including expansion elements, defense elements, cooperation elements, and dormancy elements as game points between the virtual partitions through a softmax multi-classifier, and the partition game subunit constructs game rules between the partitions, and multiple virtual partitions generate partition area change instructions based on the game rules, and finally generates a drug susceptibility quantitative test result report based on a macroscopic growth inhibition rate in a converged state of all virtual partitions.

[0087] In one embodiment, the strategy element generation subunit is configured to simulate the behavior strategies of different virtual partitions and generate strategy elements based on initial behavior preferences and time-series growth dynamics characteristics, and the steps are as follows:

[0088] Generating initial strategy element concentrations of the virtual partitions based on the initial behavior preferences includes generating initial expansion elements, initial defense elements, initial cooperation elements, and initial dormancy elements;

[0089] Real-time acquisition of the strategy tendencies of the flora in each virtual partition includes time-series growth dynamics characteristics, external characteristics, and adjacent partition relationship data;

[0090] Transmitting the strategy tendencies to a strategy logic model to generate dynamic strategy elements;

[0091] The strategy logic model is built-in with a system determination threshold generated by a control test without drugs, and the steps of generating dynamic strategy elements include:

[0092] Generating a dynamic expansion element when the current growth rate of the virtual partition is significantly higher than the growth vigor threshold in the system determination threshold;

[0093] Generating a dynamic defense element when the difference between the growth potential of the adjacent virtual partition and the growth potential of the virtual partition in which the virtual partition is located is greater than a preset difference threshold;

[0094] When the growth potential difference between the virtual partition where the self is located and the adjacent virtual partition is less than the preset difference threshold value, and the growth rates of both are higher than the system growth vigor threshold value, a dynamic cooperation pheromone is generated;

[0095] When the current growth rate of the virtual partition is significantly lower than the system growth vigor threshold value, a dynamic dormancy pheromone is generated;

[0096] Further, the strategy pheromone generation subunit realizes high-speed strategy calculation by using a parallel processing architecture, and its internal integration of a multi-channel data acquisition interface realizes real-time acquisition of time series dynamics characteristics such as the growth rate of each partition, metabolic activity, and the potential difference of adjacent regions. A strategy logic model is realized by using an embedded microprocessor architecture, and the model takes a plurality of predefined system decision thresholds as the core decision benchmark. The preset growth vigor threshold value is generally set to be 70% of the average expansion rate of the bacterial population index in the drug-free control group in the exponential growth phase, and the preset competition difference threshold value is generally set to be 90% of the growth potential difference of the adjacent regions in the control group. The model realizes real-time acquisition of time series data streams such as the growth rate of each virtual partition, the change rate of metabolic activity, and the growth potential difference with all adjacent partitions. The execution process of the strategy logic model includes first comparing the current partition growth rate with the preset growth vigor threshold value. If the growth rate exceeds the threshold value by more than 15% for three consecutive sampling periods, a dynamic expansion pheromone is generated. At the same time, the growth potential difference between the partition and each adjacent partition is calculated. When any difference is greater than the preset competition difference threshold value, a dynamic defense pheromone is immediately generated. If the potential difference with all adjacent partitions is less than the competition difference threshold value, and the growth rate of each party is higher than the growth vigor threshold value, a dynamic cooperation pheromone is generated. When the partition growth rate continues to be lower than the growth vigor threshold value by more than 25%, a dynamic dormancy pheromone is generated and other strategy pheromones are inhibited. The concentration of all strategy pheromones is output in 12-bit precision numerical value, and is transmitted to the partition game subunit through a digital isolator. At the same time, the timestamp and intensity parameter of each strategy pheromone generation event are recorded in real time to a circular buffer for subsequent game situation analysis.

[0097] In one embodiment, the partition game subunit is used to analyze the game results between virtual partitions and generate drug sensitivity quantitative test results, and the steps are as follows:

[0098] A dynamic boundary channel is constructed on the boundary of a plurality of virtual partitions;

[0099] A plurality of game pipes are arranged in the dynamic boundary channel, which are used to transmit pheromones between virtual partitions;

[0100] In the dynamic boundary channel, based on the concentration of the imported strategy pheromone, a predefined game arbitration rule is executed for calculation;

[0101] The game arbitration rule includes antagonistic cancellation of expansion pheromone and defense pheromone, modulation of antagonistic intensity by high-concentration cooperation pheromone, and suppression of all interactive activities in the channel by dormancy pheromone.

[0102] According to the game decision result, output boundary movement instructions, drive the dynamic adjustment of adjacent virtual partition area, and record the instantaneous game state of the channel;

[0103] Synchronize the output of all dynamic boundary channels, integrate the real-time territory change map and game state vector of the entire microbial community space to generate a time-series global game state vector;

[0104] Based on the comparison of the time-series global game state vectors of the drug-containing reaction unit and the drug-free reaction unit, when the game state index representing expansion vitality in the drug-containing reaction unit statistically significantly decays, and the game state index representing defense or dormancy becomes dominant, it is determined that the drug concentration is effective;

[0105] And the drug sensitivity quantitative data is the minimum drug concentration that effectively causes the dominant change of the game state;

[0106] Further, the partition game subunit adopts a biological excitation type computing architecture, constructs a dynamic boundary channel with bionic characteristics at the virtual partition boundary, and sets multiple game pipes in the dynamic boundary channel for the interaction of strategy elements. In order to better meet the game of strategy elements, a multi-modal signal coupling transmission mechanism is introduced, and three characteristic signals including strategy element concentration value, metabolic oscillation frequency of partition edge cells and pH gradient change value between partitions are transmitted in parallel in each game pipe, wherein the strategy element concentration value is used as the main data stream; the metabolic oscillation frequency of the partition edge cells is used as the carrier signal for synchronization timing; and the pH gradient change between partitions is used as the environmental reference signal for transmission attenuation calibration. Based on the mechanism, the interaction process between different bacterial groups can be more effectively simulated, and more accurate data can be provided for subsequent game adjudication. At the same time, an autonomous game coordinator is embedded in the dynamic boundary channel, which contains a double feedback loop structure. The inner loop dynamically adjusts the channel bandwidth allocation by monitoring the strategy element transmission rate, and the outer loop establishes a prediction model based on historical game data, predicts the strategy tendency of adjacent virtual partitions through machine learning algorithm, and adjusts the channel gain parameter in advance. After the game pipe is constructed, the bacterial groups in the virtual partition transmit the generated strategy elements to the game pipe based on the game adjudication rules. The game adjudication engine processes the strategy element concentration through a special arbitrator. First, the concentration difference between expansion elements and defense elements is calculated. When the concentration of expansion elements exceeds the concentration of defense elements and reaches a preset adjudication threshold, a boundary movement instruction is generated. The preset adjudication threshold is usually set to 20% of the total concentration range. When the concentration of cooperation elements participates in the calculation, the antagonistic intensity coefficient is reduced in proportion, and its adjustment coefficient is generally set to 5% reduction of antagonistic intensity per unit concentration. When the concentration of dormant elements exceeds the activation threshold, all interaction calculations of the channel are immediately interrupted, wherein the activation threshold is usually set to 30% of the maximum possible concentration. Further, the adjudication threshold and the activation threshold, the boundary movement instruction is converted into the partition area adjustment amount by the PID controller, drives the virtual boundary coordinate update and records the game state marker of the channel. The output data of all channels is synchronized and integrated by the central processor, and the global territory change map is generated through the topology reconstruction algorithm, and the state of each channel is coded into a 128-dimensional game state vector. The system determines the efficacy by comparing the standard deviation of the time series game state vectors of the drug-containing group and the control group. When the expansion activity index of the drug-containing group is continuously lower than the mean value of the control group by two standard deviations and the proportion of defense / dormant index exceeds the preset proportion, it is determined that the drug is effective, wherein the preset proportion is usually set to 75%. Finally, the minimum inhibitory concentration is determined by fitting the game state change curve under different concentrations.

[0107] In one embodiment, the visualization module comprises:

[0108] a dynamic spatiotemporal evolution unit configured to receive the real-time territory change map and the game state vector output by the spatial game network module, and to visualize the changes in the territory range of different virtual partitions and the dominant strategy element type in the form of a dynamic map, wherein different strategy element types are encoded in different colors;

[0109] a pharmacodynamic curve unit configured to draw and display a curve representing the change over time of an index characterizing the overall expansion activity of the bacterial community based on the time-series global game state vector, and to display the corresponding curve of the drug-free reaction unit synchronously for comparison;

[0110] a drug sensitivity quantitative reporting unit configured to generate and present a final drug sensitivity quantitative detection report including the minimum inhibitory concentration value, the pharmacodynamic evaluation conclusion based on game state analysis, and the bacterial community spatial game state atlas at key time points;

[0111] Further, the dynamic spatiotemporal evolution unit receives the real-time territory change map and the game state vector output by the spatial game network module in real time through the WebSocket protocol, adopts the WebGL graphics engine to build a dynamic map rendering framework, dynamically updates the territory range of different virtual partitions in the form of polygon vector graphics, and simultaneously color codes the dominant strategy element type of each partition based on pre-set color coding rules, for example, red for expansion elements, blue for defense elements, green for cooperation elements, and gray for dormancy elements, and ensures the real-time and consistency of the display of territory range changes and strategy element types through frame synchronization technology. The pharmacodynamic curve unit extracts core indicators characterizing the overall expansion activity of the bacterial community from the time-series global game state vector, such as the average of the territory expansion area per unit time and the statistical average of the expansion element concentration, adopts the ECharts visualization library to build a two-axis curve model with time as the X-axis and the expansion activity index value as the Y-axis, draws the curves of the drug-containing reaction unit (presented in solid lines) and the drug-free reaction unit (presented in dashed lines) respectively, and realizes synchronous superimposed display. The drug sensitivity quantitative reporting unit obtains the minimum inhibitory concentration value (i.e., the lowest drug concentration that effectively causes a dominant change in the game state), the game state analysis data (such as the difference in expansion activity decay between the drug-containing group and the control group, and the change trend of the defense / dormancy index proportion), and the corresponding bacterial community spatial game state atlas at key time points (such as the time when the game state first undergoes a dominant change and the time when the index is stably maintained) from the spatial game network module through a data interface, adopts the ReportLab PDF generation tool to build a structured report template, which includes a detection basic information area (detection number, drug type, reaction conditions), a core data area (minimum inhibitory concentration value, pharmacodynamic evaluation conclusion), and a graphical display area (game state atlas at key time points, pharmacodynamic curve screenshot), and through data automatic filling and format standardization processing, generates a directly exportable PDF format drug sensitivity quantitative detection report.

[0112] The above merely provides the specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A Helicobacter pylori drug sensitivity quantitative detection system, characterized in that, a drug sensitivity detection cabin module: for providing and maintaining a specific gas phase environment and temperature for the growth of Helicobacter pylori, and constructing a reaction unit containing a microplate; an optical detection module: connected with the drug sensitivity detection cabin module, for scanning the reaction unit of the microplate at a preset time to generate a multi-time optical image containing spatial information of the bacterial population; a spatial game network module: connected with the optical detection module, for virtually partitioning the bacterial population based on the multi-time optical image, constructing a game network between the bacterial populations based on the virtual partition, and analyzing the growth trend of the bacterial population based on the game network to obtain the drug sensitivity quantitative data of Helicobacter pylori; a visualization module: connected with the spatial game network module, for visualizing and displaying the effect of the drug culture solution on Helicobacter pylori and the drug sensitivity quantitative data.

2. The Helicobacter pylori drug sensitivity quantitative detection system according to claim 1, characterized in that, The drug sensitivity detection cabin module comprises: a microenvironment control unit for providing a preset gas phase environment for the reaction unit; a precision temperature control unit for providing a preset temperature suitable for the growth of Helicobacter pylori for the reaction unit; a reaction unit bearing unit for bearing the reaction unit of the reaction between Helicobacter pylori and drugs.

3. The Helicobacter pylori drug sensitivity quantitative detection system according to claim 1, characterized in that, The optical detection module comprises: a high-frequency scanning imaging unit for scanning the entire reaction unit at a preset time interval to obtain a microscopic image of the bacterial population in the reaction unit; a focusing control unit for maintaining a constant imaging focal plane during scanning to ensure the spatial consistency of the microscopic image; a plurality of microscopic images of the reaction unit generate a multi-time optical image of the bacterial population.

4. The Helicobacter pylori drug sensitivity quantitative detection system according to claim 1, characterized in that, The spatial game network module comprises: a virtual partition unit for dynamically partitioning the bacterial population image of each reaction unit into multiple interacting virtual partitions based on the morphological characteristics and spatial distribution of the bacterial population; a dynamic feature extraction unit for generating initial morphological characteristics of the virtual partition and real-time extracting time-series growth dynamics characteristics of the bacterial population in the virtual partition; wherein the initial morphological characteristics include edge features and initial density uniformity of the bacterial population in the virtual partition, and the time-series growth dynamics characteristics include local biomass accumulation rate, metabolic activity change trend and morphological stability index; a game network construction unit for constructing a game network between the bacterial populations based on the virtual partition and the time-series growth dynamics characteristics.

5. The Helicobacter pylori drug sensitivity quantitative detection system according to claim 1, characterized in that, The game network construction unit comprises: an instinct core subunit for generating a unique subcode of the virtual partition based on the initial morphological characteristics and spatial position of the virtual partition, the unique subcode being used to represent the initial behavior preference of the virtual partition; a strategy element generation subunit for simulating the behavior strategy of different virtual partitions and generating strategy elements based on the initial behavior preference and the time-series growth dynamics characteristics; a partition game subunit for analyzing the game results between the virtual partitions and generating drug sensitivity quantitative test results.

6. The Helicobacter pylori drug sensitivity quantitative detection system according to claim 5, characterized in that, The strategy element generation subunit is used to simulate the behavior strategy of different virtual partitions and generate strategy elements based on the initial behavior preference and the time-series growth dynamics characteristics, and the steps are: generate the initial strategy element concentration of the virtual partition based on the initial behavior preference, including initial expansion element, initial defense element, initial cooperation element and initial dormancy element; The strategy tendency of each virtual partition includes time-series growth kinetics, external characteristics, and adjacent partition relationship data; The strategy tendency is transmitted to a strategy logic model to generate a dynamic strategy element; The strategy logic model has a system determination threshold value generated by a control test without drugs, and the steps for generating the dynamic strategy element include: When the current growth rate of the virtual partition is significantly higher than the growth vitality threshold value in the system determination threshold value, a dynamic expansion element is generated; When the difference between the growth potential of the adjacent virtual partition and the growth potential of the virtual partition where the virtual partition is located is greater than a preset difference threshold value, a dynamic defense element is generated; When the difference between the growth potential of the virtual partition where the virtual partition is located and the growth potential of the adjacent virtual partition is less than the preset difference threshold value, and the growth rates of both are higher than the system growth vitality threshold value, a dynamic cooperation element is generated; When the current growth rate of the virtual partition is significantly lower than the system growth vitality threshold value, a dynamic dormancy element is generated.

7. The Helicobacter pylori drug sensitivity quantitative detection system according to claim 5, characterized in that, The partition game subunit is configured to analyze the game results between the virtual partitions and generate a drug sensitivity quantitative test result. A dynamic boundary channel is constructed on the boundary of the multiple virtual partitions. Multiple game pipelines are arranged in the dynamic boundary channel to transmit pheromones between the virtual partitions. Based on the concentration of the imported strategy element in the dynamic boundary channel, a predefined game arbitration rule is executed for calculation. The game arbitration rule includes antagonistic cancellation of the expansion element and the defense element, high-concentration cooperation element modulation of the antagonistic strength, and dormancy element suppression of all interactive activities in the channel. According to the game arbitration result, a boundary movement instruction is output to drive dynamic adjustment of the area of the adjacent virtual partition, and the instantaneous game state of the channel is recorded. The outputs of all the dynamic boundary channels are processed synchronously to integrate and generate a real-time territory change map and a game state vector of the entire bacterial population space as a time-series global game state vector. Based on the comparison of the time-series global game state vectors of the drug-containing reaction unit and the drug-free reaction unit, when the game state indicators representing expansion activity in the drug-containing reaction unit statistically significantly attenuate, and the game state indicators representing defense or dormancy become dominant, it is determined that the drug concentration is effective. The drug sensitivity quantitative data is the lowest drug concentration that effectively causes the dominant change in the game state.

8. The Helicobacter pylori drug sensitivity quantitative detection system according to claim 1, characterized in that, The visualization module includes: A dynamic spatiotemporal evolution unit is configured to receive the real-time territory change map and the game state vector output by the spatial game network module, and visualize and display the territory range change and the dominant strategy element type of different virtual partitions in the form of a dynamic map, wherein different strategy element types are encoded in different colors. A pharmacokinetic curve unit is configured to draw and display a curve representing the change of an indicator representing the overall expansion activity of the bacterial population over time based on the time-series global game state vector, and synchronously display the corresponding curve of the drug-free reaction unit for comparison. A drug sensitivity quantitative report unit is configured to generate and present a final drug sensitivity quantitative test report including the minimum inhibitory concentration value, the pharmacodynamic evaluation conclusion based on the game state analysis, and the bacterial population space game state atlas at the key time points.

Images