Medical examination microorganism culture system

Abstract

This invention discloses a medical testing microbial culture system, specifically relating to the field of microbial culture, including a pretreatment inoculation module, an environmental control culture module, a fully automated processing module, a colony imaging and interpretation module, and an incubator self-cleaning module. This medical testing microbial culture system organically integrates standardized automated inoculation, multi-dimensional environmental parameter coordinated control, fully automated anaerobic environment establishment, intelligent colony imaging and feedback, and full-coverage self-cleaning function. This significantly shortens the entire microbial testing process cycle, improves the overall positive detection rate of fastidious and anaerobic bacteria, effectively eliminates human error, greatly improves the consistency and repeatability of test results between different operators and different batches, and reduces the risk of cross-contamination.

Description

Technical Field

[0001] This invention relates to the field of microbial culture technology, and more specifically, to a medical testing microbial culture system. Background Technology

[0002] With the development of clinical microbiology testing technology, efficient and accurate in vitro culture of pathogenic microorganisms has become a key step in the diagnosis of infectious diseases. In the field of medical testing, a microbial culture system refers to a device or combination of equipment that can provide a suitable temperature, humidity, and gas environment to promote the growth and reproduction of microorganisms in clinical specimens. Through the regulation of environmental parameters and standardized operating procedures, the microbial culture cycle is shortened and the positive detection rate is improved, thus laying a reliable foundation for subsequent bacterial identification and drug sensitivity testing. It is an indispensable basic equipment for clinical microbiology laboratories.

[0003] Currently, the electric thermostatic incubator is the most representative basic solution. It generates heat through electric heating elements, combined with a PID intelligent temperature control system and forced air circulation design, maintaining the temperature within the set range to provide a thermal environment sufficient for the basic growth of common bacteria. Building upon this, some laboratories have introduced automated production line systems, achieving partial automation from sample inoculation to result interpretation. This reduces the frequency of manual operations, improves testing efficiency and operational standardization to a certain extent, and meets the basic clinical needs for timely diagnosis.

[0004] However, in actual use, it still has some shortcomings. For example, the existing constant temperature culture device lacks the ability to comprehensively regulate key culture parameters such as humidity and carbon dioxide concentration, which makes it difficult for some fastidious bacteria or anaerobic bacteria that have strict requirements for the growth environment to obtain suitable growth conditions in a single device, directly affecting the positive detection rate and culture success rate of such pathogens. Existing automated culture systems have failed to achieve fully automated operation. Key steps such as sample pretreatment, establishment of anaerobic culture environment, colony selection and result interpretation still rely heavily on manual operation, resulting in significant efficiency bottlenecks and points of human error in the operation chain, which restricts the consistency and reproducibility of test results. Existing culture devices lack effective means for continuous sterilization and purification of the internal environment during continuous operation. With the switching of specimen batches and long-term use, the inside of the chamber is prone to the accumulation of microbial aerosols and cross-contamination, which affects the accuracy of test results and poses a potential risk of misjudgment in clinical diagnosis. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, the present invention provides a medical testing microbial culture system, which addresses the problems mentioned in the background art through the following solutions.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A medical testing microbial culture system, comprising: Pretreatment and inoculation module: Used to automatically complete the standardized streaking and inoculation of clinical specimens via a robotic arm, and to transfer the inoculated culture medium to the environmentally controlled culture module; Environmental control and cultivation module: It is equipped with a closed cultivation chamber and integrates a temperature control unit, a humidity control unit and a carbon dioxide control unit. The temperature control unit, humidity control unit and carbon dioxide control unit work together to regulate the temperature, humidity and carbon dioxide concentration in the closed cultivation chamber, and are used to monitor the cultivation environment parameters in the closed cultivation chamber in real time. Fully automated processing module: used to establish and maintain the colony culture environment in the closed culture chamber according to the culture environment parameters through a dual mechanism of automatic gas replacement and chemical catalytic deoxygenation; Colony imaging interpretation module: used to periodically acquire plate images of the culture medium during the culture process, perform colony counting and morphological analysis by comparing images at different time points, and send feedback signals to the environmental control culture module to trigger the review of the corresponding specimen culture conditions or specific supplementary tests; The incubator self-purification module is connected to the gas path of the environmental control culture module and the fully automatic processing module respectively. It is used to start the disinfection program in response to the trigger signal of the end of the culture cycle or the switching of sample batches, and to sterilize and purify the closed culture chamber and the connected gas path pipeline.

[0007] Preferably, the pretreatment inoculation module includes an automatic marking inoculation unit that stores multiple standardized marking programs; The robotic arm is configured to select the corresponding program from the various standardized streaking programs based on the specimen type information of the clinical specimen, so as to automatically complete the zonal streaking or quantitative coating operation. The end effector of the robotic arm is equipped with a force sensor to provide real-time feedback on the pressure of the inoculation line. And a speed controller, based on real-time feedback from force sensors, uses a PID algorithm to adjust the robotic arm's movement speed to keep it stable near a preset value.

[0008] Preferably, the environmental control culture module integrates a distributed sensor group, including at least multiple temperature and humidity sensors respectively configured in the upper, middle, and lower layers of the closed culture chamber: The output power of the humidifier is decoupled and controlled by a pre-stored fuzzy logic controller.

[0009] Preferably, the input variables of the fuzzy logic controller include humidity deviation and its rate of change, and its output variable is the output power increment of the humidification device; The humidity deviation refers to the difference between the current humidity and the set value, and the humidity change rate is the amount of change of the humidity deviation over time.

[0010] Preferably, the fully automated processing module includes a gas replacement unit and a chemical catalytic deoxygenation unit; The gas replacement unit is used to pump the closed culture chamber to a preset negative pressure and then fill it with replacement gas, and repeat the cycle at least once to reduce the oxygen concentration in the chamber to a first threshold. The chemical catalytic deoxygenation unit is automatically activated when the oxygen concentration is lower than the first threshold, driving the gas in the chamber to circulate through the catalytic reaction column, further reducing the oxygen concentration to below the second threshold, so as to maintain the colony culture environment.

[0011] Preferably, the preset negative pressure is -80 kPa, the flow rate of the replacement gas is 15 L / min, and the number of cycles is 3. The first threshold is 1.0% vol, and the second threshold is 50 ppm; The catalytic reaction column is filled with a desiccant to adsorb trace amounts of moisture generated during the catalytic deoxygenation process.

[0012] Preferably, the colony imaging interpretation module includes an image recognition unit, a comparison analysis unit, and a feedback signal generation unit; The image recognition unit is configured to run a colony segmentation and recognition model based on the U-Net architecture, which is used to perform colony segmentation and recognition on the acquired plate images; The comparison analysis unit performs differential comparison based on images of the same plate collected at different time points, identifies the same bacterial colony by centroid coordinate matching, and calculates the growth rate of its colony. When the growth rate exceeds the preset normal range, the feedback signal generation unit generates a culture condition verification signal and sends it to the environment control culture module.

[0013] Preferably, the incubator self-cleaning module specifically includes: Based on the specimen type information corresponding to the trigger signal received from the main control unit, the corresponding mode is called from multiple disinfection modes to execute a disinfection procedure that matches the risk level of the current specimen batch. The disinfection procedure includes a sequentially executed ultraviolet irradiation stage, a hydrogen peroxide dry fog disinfection stage, and a residue removal stage; the disinfection modes include at least a standard mode, a rapid mode, and an enhanced mode, with different execution durations for the ultraviolet irradiation stage and / or the hydrogen peroxide dry fog disinfection stage in each mode.

[0014] The technical effects and advantages of this invention are as follows: 1. This invention uses a pretreatment inoculation module to install a force sensor and a speed controller on the end effector of a robotic arm, so that the inoculation force and trajectory speed of the streaking are kept highly consistent between different specimens, which effectively improves the repeatability of colony isolation. 2. This invention constructs a closed culture chamber through an environmental regulation culture module, which coordinates the temperature control unit, humidity control unit and carbon dioxide control unit. It also achieves decoupled regulation of temperature and humidity through a pre-stored fuzzy logic controller, avoiding the problem of relative humidity going out of control due to temperature changes. This allows fastidious bacteria and anaerobic bacteria to obtain a suitable multi-dimensional growth environment in the same device, which directly affects the positive detection rate and culture success rate of such pathogens. 3. This invention employs a fully automated processing module and a dual deoxygenation mechanism of physical displacement and chemical catalysis, proceeding step by step, so that the establishment of an anaerobic environment no longer depends on manually operated anaerobic tanks or gas bags; at the same time, by connecting a drying tube filled with color-changing silica gel desiccant in series in the gas path at the rear end of the catalytic reaction column, it is ensured that the humidity parameters in the closed culture chamber are not affected by the deoxygenation process, avoiding the introduction of new uncontrolled factors in the culture environment to solve the problem of establishing an anaerobic environment, and ensuring the coordinated stability of multi-dimensional environmental parameters; 4. This invention uses a colony imaging and interpretation module to automatically complete colony counting and morphological analysis. Based on whether the growth rate exceeds the preset normal range, it generates a culture condition verification signal and sends it to the environmental control culture module to trigger dynamic adjustment of culture conditions. This achieves proactive optimization of the culture strategy based on the actual growth status of the colonies, further improving the final detection rate of fastidious bacteria and ensuring the accuracy of the test results. 5. This invention achieves full-coverage purification of the closed culture chamber and connecting gas pipelines through the incubator's self-purification module, ensuring the sterility of the culture environment for each batch of specimens from a physical source. At the same time, based on the received specimen type information or operator instructions, it automatically calls the disinfection program that matches the risk level of the current specimen batch, ensuring that the purification effect meets the standards while taking into account the efficiency of equipment use, avoiding unnecessary over-disinfection, and saving operating time and disinfectant consumption. 6. The present invention obtains objective growth data of colonies by comparing time-series images through a colony imaging and interpretation module, generates feedback signals and sends them to an environmental control and culture module, which automatically adjusts culture parameters so that the culture environment can be dynamically optimized according to the actual growth performance of microorganisms, significantly improving the ability to capture slow-growing or atypical fastidious bacteria. 7. The present invention connects the disinfection pipeline of the incubator self-purification module to the internal gas pipeline of the fully automatic processing module through a three-way solenoid valve. During the disinfection stage, hydrogen peroxide dry fog is actively delivered to the pipeline for gas replacement and catalytic deoxygenation, so that all surfaces inside the entire system that may come into contact with microorganisms can be purified, thus forming a complete sterile environment protection system. 8. The present invention relies on the real-time oxygen concentration data provided by the online sensor group of the environmental regulation and cultivation module as the basis for triggering and feedback control through the anaerobic environment establishment function of the fully automatic processing module; the multi-parameter coordinated adjustment of the environmental regulation and cultivation module provides stable temperature conditions for the chemical catalytic deoxygenation reaction of the fully automatic processing module. Together, they ensure the rapid establishment and long-term stable maintenance of the anaerobic culture environment. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the structure of a medical testing microbial culture system provided according to an embodiment of this application.

[0016] Figure 2 This is a schematic diagram of the modular composition of a medical testing microbial culture system according to an embodiment of this application.

[0017] Figure 3 This is a schematic diagram of the pretreatment inoculation module provided according to an embodiment of this application.

[0018] Figure 4 This is a schematic diagram of the structure of the environmental regulation culture module provided according to an embodiment of this application.

[0019] Figure 5 This is a schematic diagram of the structure of the fully automatic processing module provided according to an embodiment of this application.

[0020] Figure 6 This is a schematic diagram of the colony imaging interpretation module provided according to an embodiment of this application.

[0021] Figure 7 This is a schematic diagram of the structure of the incubator self-cleaning module provided according to an embodiment of this application. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to and includes any or all possible combinations of one or more of the listed items. In the description of the embodiments of this application, unless otherwise stated, “a plurality” means two or more.

[0024] As attached Figure 1 The diagram illustrates the structure of a medical laboratory microbial culture system. The system includes a main control unit, a system database, a human-machine interface (HMI), and a CAN bus. The main control unit is implemented as a programmable logic controller (PLC) or an embedded industrial control motherboard. It connects via the CAN bus to the pretreatment inoculation module, the environmental control and culture module, the fully automated processing module, the colony imaging and interpretation module, and the incubator self-cleaning module, forming a star-shaped distributed control topology. This topology is used to execute preset system-level scheduling logic, inter-module collaborative control, and data management. The system database communicates with the main control unit and stores specimen information, culture program parameters, historical culture environment data, colony image sequences, and interpretation results, enabling traceable management of the entire process. The HMI is implemented as a touch screen, connected to the main control unit, and displays the system's operating status, culture progress, colony images, and interpretation results, while also receiving operator input. The specific material flow paths and signal coupling relationships between the modules will be further explained in subsequent figures.

[0025] As attached Figure 2 The medical testing microbial culture system shown includes: a pretreatment inoculation module, an environmental control culture module, a fully automated processing module, a colony imaging and interpretation module, and an incubator self-cleaning module.

[0026] See Figure 3 One embodiment of a medical testing microbial culture system of the present invention includes: a pretreatment inoculation module, which includes at least a specimen receiving unit, an automatic streak inoculation unit, and a culture medium delivery unit.

[0027] The specimen receiving unit is used for the standardized reception of pre-processed clinical specimens. Pre-processing refers to the specimen undergoing routine operations such as liquefaction, dilution, and vortexing to become a liquid specimen suitable for inoculation. Specifically, the specimen receiving unit is implemented as a rotary carrier capable of accommodating 24 standard specimen tubes. Each specimen tube on the rotary carrier integrates an RFID reading antenna for automatically reading the specimen type information ("sputum," "urine," "bronchoalveolar lavage fluid," etc.) stored in the RFID tag attached to the specimen tube and transmitting this information to the main control unit.

[0028] Meanwhile, the specimen receiving unit is equipped with an ultrasonic homogenization probe. It should be noted that the ultrasonic homogenization probe is designed for specimens with high viscosity (such as sputum). In a specific implementation scenario, when the main control unit determines that the current specimen is "sputum" based on the received specimen type information, before the robotic arm performs the smearing action, the ultrasonic homogenization probe will automatically descend and extend into the specimen tube, performing continuous ultrasonic processing on the specimen for 20 seconds at a frequency of 20kHz and a power of 100W.

[0029] The automated streaking and inoculation unit is used for unmanned streaking and inoculation of homogenized specimens. It is integrated into a six-axis linkage robotic arm, and its end effector is a gripping mechanism equipped with a force sensor, which is used to grasp disposable inoculation loops or precision micropipette tips.

[0030] The controller of the robotic arm pre-stores various standardized marking programs, including at least a four-zone marking method for sputum specimens, a quantitative smear method for urine specimens, and a spiral smear method for bronchoalveolar lavage fluid. During inoculation, the main control unit automatically matches and selects the corresponding marking program and issues a start command based on the specimen type information uploaded by the specimen receiving unit.

[0031] During the inoculation process performed by the robotic arm, adaptive control is achieved through a force sensor and a speed controller. In this embodiment, the force sensor provides real-time feedback on the contact pressure between the inoculation loop and the culture medium surface at a sampling frequency of 100Hz, and sends the feedback signal to the main control unit, which then runs a PID control algorithm. The speed controller, based on the force feedback signal, uses a PID algorithm to adjust the movement speed of the robotic arm's end effector in a closed loop on a millisecond time scale, ensuring that it remains stable near a preset value (60mm / s for the four-zone streaking method and 30mm / s for the quantitative coating method), with fluctuation deviation controlled within ±2mm / s.

[0032] In a specific line marking cycle, the robotic arm's execution process is as follows: Move to the disposable inoculation loop storage area and pick up a sterilized inoculation loop; Move the inoculation loop above the specimen receiving unit and dip it into the specimen tube to collect 10 μL of bacterial solution. Move to the culture medium plate working area and execute the selected streaking trajectory in the order of "first draw the first zone coarse expansion line, then draw the second to fourth zones fine subdivision lines".

[0033] The culture medium transfer unit is used to transport the inoculated culture medium plates to the next module. Specifically, it is implemented as a transfer track driven by a stepper motor, with barcode scanners integrated at specific locations along the transfer path. In this embodiment, the transfer track operates at a constant speed of 150 mm / s, on which the culture medium plates are transferred. An airtight door interface is located at the end of the transfer track for a sealed connection with the inlet of the environmentally controlled culture module. When the main control unit determines that the plate is in place and the environmentally controlled culture module is ready to receive it, the airtight door automatically opens, allowing the plate to enter the sealed culture chamber, achieving material transfer without environmental exposure.

[0034] As culture medium plates, already inoculated and with their lids automatically closed, pass through the barcode scanning station along the conveyor track, a fixed barcode scanner automatically reads and binds the barcode (i.e., its unique identification information) on the side of the culture medium plate. The main control unit then associates this unique identification information with the patient information, specimen type, inoculation procedure, inoculation timestamp, and planned culture procedure for the current specimen in the database, and uploads this association to the system database.

[0035] The key operating parameters of the pretreatment inoculation module in this embodiment and the basis for their values ​​are shown in Table 1 below: Table 1

[0036] See Figure 4 Another embodiment of a medical testing microbial culture system according to the present invention includes: an environmental control culture module receiving the inoculated culture medium and constructing a closed culture chamber.

[0037] The sealed culture chamber integrates a temperature control unit, a humidity control unit, and a carbon dioxide control unit. The temperature control unit maintains a constant temperature using a PID algorithm; the humidity control unit dynamically adjusts the humidification device based on temperature feedback to maintain a predetermined humidity level; and the carbon dioxide control unit precisely supplies CO2 via a gas mixer according to a set value. Simultaneously, the sensor array within monitors various parameters in real time and triggers mode switching in the fully automatic processing module. In this embodiment, the sealed culture chamber is specifically a 280L stainless steel airtight chamber with an inner wall that has been mirror-polished with a Ra≤0.4μm finish, used to accommodate inoculated culture medium plates from the pretreatment inoculation module. To achieve and maintain high airtightness of the culture environment, the chamber door employs a double-layer inflatable sealing ring design. Specifically, after the door is closed, compressed air at a pressure of 15kPa is injected into the cavity between the two sealing rings, causing the sealing rings to expand and tightly press against the door frame. After being pressurized to 15kPa, the static pressure drop within 24 hours does not exceed 5% of the initial pressure.

[0038] The temperature control unit is located in the side wall interlayer and back of the sealed culture chamber. It consists of a PTC heating element and a semiconductor cooling chip, and is equipped with a centrifugal circulating fan with an adjustable speed range of 500-2000 RPM. An incremental PID algorithm is used to maintain the culture temperature at the target temperature with a control accuracy of ±0.1°C. In this embodiment, the target temperature is set to 35°C.

[0039] The humidity control unit is located inside the sealed culture chamber and is specifically a piezoelectric ceramic ultrasonic atomizing humidifier with a resonant frequency of 1.7MHz. A signal linkage is established between the humidity control unit and the temperature control unit to achieve decoupled control of temperature and humidity.

[0040] The carbon dioxide concentration control unit is located at the gas inlet of the closed culture chamber and includes an external high-purity CO2 gas source with a purity of 99.999%, a high-precision mass flow controller with a control accuracy of ±0.1% of the set value, and a static gas mixer.

[0041] The sensor array adopts a distributed layout, with at least three platinum resistance temperature sensors (accuracy ±0.05°C), three capacitive humidity sensors (accuracy ±1.5%RH), one non-dispersive infrared CO2 sensor (accuracy ±0.1% vol), and one electrochemical oxygen sensor (accuracy ±1% reading) configured in the upper, middle, and lower layers of the closed culture chamber.

[0042] In one possible implementation, the target temperature is set to 35°C. When the platinum resistance temperature sensor detects that the current temperature is 34.8°C, the PID controller calculates the heating duty cycle deviation and outputs it to the PTC heating element, so that the temperature converges to a steady state of 35°C ± 0.1°C on a minute-scale.

[0043] This embodiment constructs a dual-input, single-output Mamdani-type fuzzy logic controller to achieve decoupled control of temperature and humidity. Its static model parameters and dynamic operating rules are as follows: The input variables are humidity deviation E and its rate of change EC. The basic universe of discourse for humidity deviation E is [-15%, +15%]RH, and its corresponding fuzzy universe of discourse is quantized as {-3, -2, -1, 0, +1, +2, +3}, with corresponding fuzzy subsets {NB (negative large), NM (negative medium), NS (negative small), ZO (zero), PS (positive small), PM (positive medium), PB (positive large)}. The quantization factor Ke = 3 / 15 = 0.2, obtained by dividing the upper limit of the fuzzy universe of discourse by the upper limit of the basic universe of discourse. The basic universe of discourse for the rate of change EC is [-3%, +3%]RH / min, and its fuzzy universe of discourse is also quantized into 7 levels, with the fuzzy subset division consistent with E. The quantization factor Kc = 3 / 3 = 1.0, obtained by dividing the upper limit of the fuzzy universe of discourse by the upper limit of the basic universe of discourse.

[0044] The output variable U represents the increment of the humidifier's output power, based on the humidifier's rated power. Its basic universe of discourse is [-30%, +30%], which is the range of power variation of the humidifier from a 30% decrease to a 30% increase from its rated power. The fuzzy universe of discourse and fuzzy subset partitioning method are the same as the input variable, and its scaling factor Ku is set to 10.

[0045] Based on the principle that "when the humidity deviation is negative (NB) and the rate of change is also negative (NB), the humidity is decreasing rapidly, therefore a positive (PB) power increment needs to be output for rapid compensation," a control rule table (Table 2) is constructed to correspond to the core decision-making logic, in order to achieve early prediction and compensation for temperature and humidity changes. Within each control cycle (e.g., 1 second), the accurate power increment output value ΔU is calculated based on the real-time calculated E and EC values ​​using a lookup table and a weighted average defuzzification method.

[0046] Table 2

[0047] Specifically, the system reads the measured value (35°C) from the platinum resistance temperature sensor and calculates the saturated vapor pressure (approximately 5.63 kPa) at that temperature using an internally stored vapor pressure formula table. It then reads the current measured humidity (65% RH) from the capacitive humidity sensor and compares it with the set value (70% RH), finding a deviation of -5% RH. The humidity deviation and saturated vapor pressure data are input into the fuzzy logic controller. The controller performs reasoning based on a pre-stored rule table and ultimately outputs the calculated power increment, dynamically adjusting the humidifier's output power to increase the atomization power from the current 60% to 75%.

[0048] By constructing the aforementioned fuzzy logic controller and pre-storing its rule base, the system can proactively adjust according to the trend of coupled temperature and humidity changes. Comparative verification shows that when the ambient temperature changes abruptly at a rate of 0.5℃ per minute, the maximum deviation of the relative humidity inside the closed culture chamber using this fuzzy decoupling control is only ±1.5%RH, and the recovery stabilization time is less than 5 minutes.

[0049] When the non-dispersive infrared CO2 sensor detects that the CO2 concentration in the chamber is lower than the set value, it instructs the mass flow controller to input CO2 gas into the gas mixer at a preset flow rate (50 sccm). After mixing with HEPA-filtered air, the mixture is evenly delivered to all parts of the closed culture chamber by a circulating fan at a speed of 1200 RPM, forming a uniform gas environment and restoring the CO2 concentration to 5.0% ± 0.1%.

[0050] An electrochemical oxygen sensor in the sensor array continuously monitors the oxygen concentration within the chamber. When an anaerobic culture command is issued for the specimen, the main control unit dynamically calls the corresponding oxygen concentration threshold according to the preset anaerobic level requirement: it detects the current oxygen concentration, and if it is higher than the preset threshold (where the trigger threshold is 5.0% vol when the culture program is "microaerobic culture" and 0.5% vol when it is "strict anaerobic culture"), it immediately sends a digital start command to the anaerobic control unit of the fully automatic anaerobic treatment module via the CAN bus, triggering it to execute the gas replacement and chemical catalytic deoxygenation program.

[0051] Range analysis is performed on the temperature data from three temperature sensors at the top, middle, and bottom layers. If, at a certain moment, the main control unit calculates that the temperature difference between the top layer (35.2°C) and the bottom layer (34.9°C) reaches 0.3°C, exceeding the preset uniformity threshold (0.2°C), the circulating fan speed is automatically increased from 1000 RPM to 1500 RPM to enhance gas convection within the cavity.

[0052] When a complete culture batch finishes running, the system automatically determines the "batch end" status based on the preset countdown zeroing signal, and then sends a start trigger signal to the incubator self-cleaning module to schedule it to execute the deep cleanup program.

[0053] In an alternative structural design for another possible implementation, a multi-cavity independent parallel culture architecture is constructed: The sealed culture chamber is physically divided into three independent airtight sub-chambers. Based on the type of specimen cultured in each sub-chamber, an independent culture program is run: Zone A runs an aerobic culture program of 35°C, 70%RH, and 5%CO2 for Haemophilus influenzae; Zone B runs a microaerobic culture program of 42°C, 80%RH, and 10%CO2 for Campylobacter jejuni; and Zone C runs a conventional aerobic culture program of 35°C.

[0054] The humidification control unit ensures the self-sterilization of the humidification water source. It uses a built-in 1L sterile distilled water bottle, with a 254nm wavelength, 8W flow-through UV sterilization lamp connected in series in the water outlet path to circulate and sterilize the flowing humidification water. When the water pump starts, the UV lamp illuminates simultaneously, ensuring that only sterilized water enters the atomizing device.

[0055] A multi-dimensional culture parameter combination table (as shown in Table 3) is pre-stored in the non-volatile memory of the main control unit, corresponding to the temperature, relative humidity, and CO2 concentration combinations for different pathogen species. In actual operation, based on the specimen type information preset in the system, or by receiving feedback signals from the colony imaging interpretation module, the corresponding set values ​​are automatically retrieved and executed, realizing adaptive dynamic adjustment of the culture environment.

[0056] Table 3

[0057] See Figure 5 Another embodiment of a medical testing microbial culture system in this invention includes: a fully automatic processing module receiving culture environment parameter instructions from the main control unit, and establishing and maintaining a strict anaerobic or microaerophilic environment within about 3 minutes through a dual mechanism of automatic gas replacement and chemical catalytic deoxygenation without human intervention. Its core hardware architecture includes at least a gas replacement unit, a chemical catalytic deoxygenation unit, and an anaerobic control unit.

[0058] The gas replacement unit is connected to the inlet and outlet of the closed culture chamber via an independent gas pipeline, forming a closed unidirectional replacement gas path. It is equipped with at least a vacuum pump, a solenoid valve assembly, a gas mass flow controller, and a high-purity gas source. In this embodiment, the gas source is configured with two lines: one line contains 99.999% pure nitrogen, and the other line contains a precisely prepared anaerobic mixture with a composition ratio of 85% nitrogen, 10% hydrogen, and 5% carbon dioxide.

[0059] When the anaerobic control unit receives the start command from the main control unit, the gas replacement unit immediately executes the first stage of physical deoxygenation as follows: The vacuum pump evacuates the sealed culture chamber to a predetermined negative pressure state, with the target pressure value set at -80 kPa; The intake solenoid valve opens, and the selected replacement gas is introduced into the cavity at a controlled flow rate of 15L / min until the pressure in the cavity returns to a slightly positive pressure of +5kPa.

[0060] In one specific embodiment, the vacuuming and gas filling replacement cycle was repeated three times. The number of cycles was set based on a 280L cavity volume and a replacement flow rate of 15L / min, with each cycle taking approximately 1 minute, and the total time for the three cycles being approximately 3 minutes. After the first cycle, the oxygen concentration in the cavity decreased from the initial 21.0% to approximately 4.2%; after the second cycle, it decreased to approximately 0.8%; and after the third cycle, the oxygen concentration in the cavity was reduced to below 0.3%, reaching a preset first threshold.

[0061] It should be noted that the first threshold is set to 1.0% vol. If the electrochemical oxygen sensor detects that the oxygen concentration in the chamber is lower than this threshold after any replacement cycle, the anaerobic control unit terminates the physical replacement stage and switches to the chemical catalytic deoxygenation stage. Therefore, the actual number of cycles can be dynamically adjusted according to the initial oxygen concentration and replacement efficiency. Three cycles are a typical value for a normal oxygen environment (21% O2).

[0062] The chemical catalytic deoxygenation unit is configured in the bypass of the circulating gas path of the closed culture chamber. Its core is a catalytic reaction column, which is filled with spherical palladium catalyst with a particle size of 3 mm at room temperature. In this embodiment, the effective volume of the closed culture chamber is 280 L, and the amount of catalyst is 50 g, which is determined based on engineering quantities.

[0063] When the electrochemical oxygen sensor in the online sensor group detects that the oxygen concentration in the chamber has dropped below the first threshold (1.0% vol), the anaerobic control unit automatically triggers the second chemical catalytic deoxygenation process: The gas circulation pump starts at a flow rate of 5L / min, driving the gas in the chamber to continuously flow through the catalytic reaction column; Under the active action of palladium catalyst, hydrogen in the mixed gas reacts with residual trace oxygen at room temperature (35°C) to produce water.

[0064] In one specific embodiment, after 30 minutes of cyclic catalytic treatment, the oxygen concentration in the cavity was further reduced from about 0.3% to below 50 ppm (i.e., 0.005% vol), which is the second threshold suitable for the growth of obligate anaerobic bacteria.

[0065] It should be noted that, in order to avoid the trace amounts of moisture generated by the catalytic reaction from interfering with the humidity parameters in the closed culture chamber, a drying tube is connected in series in the gas path at the rear end of the catalytic reaction column. The tube is filled with color-changing silica gel desiccant. When the gas flows through the drying tube, the trace amounts of moisture generated by the reaction are adsorbed and removed, so that the gas dew point is maintained below -40°C, thereby ensuring that the humidity parameters of the culture environment are not affected by the deoxygenation process.

[0066] The anaerobic control unit is connected to the main control unit via a CAN bus and is equipped with a closed-loop feedback control program: when the main control unit identifies that the current batch of specimens needs to be anaerobic culture based on the specimen information, or when the electrochemical oxygen sensor in the sensor group detects that the oxygen concentration in the chamber exceeds the preset threshold (0.5%) for anaerobic culture, the main control unit immediately sends a digital start command to the anaerobic control unit to start the fully automatic processing module immediately.

[0067] Specifically, during the anaerobic culture maintenance phase, the closed-loop feedback control program continuously reads the monitoring value of the electrochemical oxygen sensor at a frequency of 1Hz. If, during long-term culture, a slight leak in the chamber seal causes the oxygen concentration to rise and exceed the set maintenance lower limit (50ppm), the anaerobic control unit automatically triggers a small-flow gas replacement pulse lasting 30 seconds, or briefly reconnects the catalytic deoxygenation circuit for 10 minutes, thereby continuously suppressing the oxygen concentration below 50ppm throughout the entire culture cycle (up to 14 days).

[0068] When the anaerobic culture cycle ends, the main control unit sends a shutdown command to the anaerobic control unit. The anaerobic control unit then automatically switches the gas valve to external circulation mode, introducing fresh air treated by a HEPA filter to ventilate and replace the chamber at a flow rate of 10L / min, so that the gas environment inside the chamber returns to the normal aerobic or microaerophilic level within 5 minutes, ready for the next batch of specimens.

[0069] It should be noted that the key operating parameters of the fully automatic processing module in this embodiment and their values ​​are based on the following Table 4: Table 4

[0070] See Figure 6 Another embodiment of a medical testing microbial culture system in this invention includes: a colony imaging interpretation module for periodically acquiring images of culture medium plates during the culture process, completing colony counting and morphological analysis through time-series image comparison and analysis, and converting the analysis results into feedback control signals for culture conditions, triggering the verification of corresponding specimen culture conditions or specific supplementary tests.

[0071] The core hardware architecture of the colony imaging and interpretation module includes a high-resolution imaging device, an image recognition unit, a comparison and analysis unit, and a feedback signal generation unit. It should be noted that the signal connections between the image recognition unit, the comparison and analysis unit, and the feedback signal generation unit are achieved through data transmission via the PCIe bus inside the embedded image processing board. Specifically, after the image recognition unit completes inference for each image, the output feature vector group is transmitted to the comparison and analysis unit in JSON structured data format via a shared memory area. After completing the time-series comparison, the comparison and analysis unit stores the comparison results (colony number, growth rate, morphological anomaly markers) in binary structure form in a shared memory queue. The feedback signal generation unit reads the comparison results from this queue in a polling manner, matches the trigger rules, and sends the corresponding feedback signal encoding to the main control unit via the CAN bus. The units operate asynchronously in a producer-consumer model to ensure that image acquisition is not blocked by subsequent processing.

[0072] The high-resolution imaging device employs a 21-megapixel CMOS industrial camera, paired with a macro lens with a 1:1 magnification ratio and a ring-shaped LED white light source, achieving a physical resolution of 10 μm / pixel, capable of distinguishing pinpoint-sized colonies as small as 0.1 mm in diameter. It is fixedly positioned inside the upper part of a sealed culture chamber, corresponding to the placement of each culture plate. Simultaneously, it integrates a 365 nm ultraviolet excitation light source and a corresponding 460 nm bandpass fluorescence filter, allowing switching to fluorescence imaging mode. When early screening of pathogens producing specific fluorescence (such as Pseudomonas aeruginosa) is required, rapid preliminary identification is achieved by capturing characteristic fluorescence signals.

[0073] The image recognition unit, connected to the high-resolution imaging device, consists of an embedded image processing board based on NVIDIA Jetson Xavier NX. It runs a colony segmentation and recognition model trained on a U-Net convolutional neural network architecture. The training sample set includes at least 5000 labeled images from clinical sputum, urine, blood, and bronchoalveolar lavage fluid specimens.

[0074] In this embodiment, the hierarchical structure of the colony segmentation and recognition model is shown in Table 5 below. It follows a symmetrical encoder-decoder architecture. Each level of the encoder copies the feature map through skip connections and splices it with the upsampling result of the decoder in the channel dimension, so that the decoder can simultaneously utilize high-level semantic localization information and low-level detail texture information to achieve accurate segmentation of colony boundaries.

[0075] Table 5

[0076] The parameters at each level (number of channels 64-128-256-512) follow the engineering principle of doubling layer by layer to achieve a balance between model capacity and inference speed. The 512-dimensional feature vector of the bottleneck layer is the highest-level abstract representation of the entire image. The L1-Concat step of the decoder concatenates the abstract localization information output from the bottleneck layer with the edge detail information retained by the L3 encoder in terms of channel dimension to obtain a 768-dimensional fused feature, which is then compressed to 256 dimensions by a 3×3 convolution, and so on, layer by layer, to restore the original resolution.

[0077] Furthermore, the training process of the colony segmentation and recognition model is as follows: Training dataset: A total of 5,000 plate images from clinical sputum, urine, blood and bronchoalveolar lavage fluid specimens. After manually annotating the colony boundaries to form a mask ground truth, the images were randomly divided into a training set (4,000 images), a validation set (500 images) and a test set (500 images) in a ratio of 8:1:1. Loss function: A weighted combination of DiceLoss and binary cross-entropy loss (BCELoss) is used as the training objective function, specifically expressed as: ; in, This is a weighting coefficient, with a value of 0.7 (based on the overlap of regions emphasized by Dice loss). The Dice loss measures the degree of overlap between the predicted segmentation region and the ground truth region. The segmentation region predicted by the model. This is the manually labeled truth region. Indicates the number of pixels within the area; The binary cross-entropy loss is used to evaluate classification accuracy pixel-by-pixel, where... This is the truth label (0 or 1) for the j-th pixel. To predict the probability that this pixel belongs to a bacterial colony region for the model, This represents the total number of pixels in the image.

[0078] Training configuration: The optimizer uses Adam, with an initial learning rate of 1×10⁻ 4The batch size is 8; the total number of training rounds is 100 epochs; an early stopping strategy is adopted, and training is terminated if the validation set loss does not decrease for 10 consecutive epochs.

[0079] Data augmentation: During training, horizontal flipping (probability 0.5), vertical flipping (probability 0.5), random rotation (±15°, probability 0.5), and brightness and contrast fine-tuning (amplitude ±10%, probability 0.3) are randomly applied to each batch of images.

[0080] Furthermore, the colony segmentation and recognition model is used to perform the following process: The raw images acquired by the high-resolution imaging device are bilinearly interpolated and scaled to 512×512 pixels×3 channels, and white balance correction and background subtraction preprocessing based on Gaussian mixture model are performed. The colony region is segmented from the culture medium background using an encoder-decoder structure, and a segmentation probability map with a size of 512×512×1 is output. After binarization with a threshold of 0.5, the final segmentation result is obtained. For each independent colony, multiple morphological features were extracted, including colony number, centroid coordinates (x, y), area (mm²), roundness, edge roughness, color (mean of RGB three channels), and transparency (grayscale variance), forming a 1×7 feature vector. The physical area of ​​the colony was calculated from the pixel area using the following conversion relationship: Physical area (mm²) = Pixel area × (Physical resolution calibrated by the imaging device)² × 10⁻ 6 The centroid coordinates (x, y) are the coordinates in the image pixel coordinate system, with the origin located at the upper left corner of the image, the x-axis to the right, and the y-axis downward. They can be converted to the physical coordinates of the flat plate through the camera calibration matrix. Based on the above morphological characteristics, the colonies were initially classified into at least three categories: lactose-fermenting, non-lactose-fermenting, and hemolytic.

[0081] The comparison and analysis unit is connected to the image recognition unit by signal. Internally, it maintains an image database arranged in time series, and its working logic is based on differential comparison of time-series images. When it receives the recognition result at the current time point (T=12h), it automatically performs a pixel-by-pixel differential comparison with the historical image data of the same plate at the previous time point (T=10h) to mark the area of ​​newly formed colonies.

[0082] Specifically, the comparison analysis unit performs colony matching by minimizing the Euclidean distance between the centroid coordinates. The matching threshold is set to a centroid distance of less than 50 μm. After a successful match, the difference in colony area and the time interval are calculated to determine the growth rate. Using the change in the equivalent circle diameter of the colony as an indicator, if a colony has a diameter of 0.2 mm at T=10 h and a diameter of 0.5 mm at T=12 h, its growth rate is calculated to be 150 μm / h. The comparison analysis unit uses whether the growth rate exceeds a preset normal range (e.g., the linear growth rate range for common Enterobacteriaceae is set to 50-200 μm / h) as one of the core criteria for generating feedback signals.

[0083] It should be noted that the preset normal growth rate range of 50-200 μm / h is based on statistically analyzed data from the linear growth phase of common Enterobacteriaceae pathogens (Escherichia coli, Klebsiella pneumoniae, etc.) cultured on blood agar plates at 35°C. Specifically, standard strains of the above-mentioned bacteria were subjected to 10 independent culture experiments. During the linear growth phase of 4 to 12 hours, the equivalent circle diameter of the colonies was measured every 2 hours, and the growth rate was calculated. The results followed a normal distribution N(125, 37.5²). The normal range of 50-200 μm / h was determined using μ ± 2σ (i.e., 125 ± 75). Growth rates exceeding this range (<10 μm / h or >300 μm / h) indicate unsuitable culture conditions or suspected contamination, requiring a culture condition verification procedure.

[0084] The feedback signal generation unit is connected to the comparison analysis unit and the main control unit. It has multiple preset interpretation rules and trigger logic, as shown in Table 6. Table 6

[0085] The trigger thresholds in Table 6 are all preset based on the clinical microbiology expert knowledge base. Roundness is defined as 4π × area / perimeter², with a standard circle having a roundness of 1.0. The roundness threshold (0.7) is determined by subtracting approximately 3.7 standard deviations (σ=0.06) from the statistical mean (0.92) of roundness from 3200 normal colony samples in the training set. That is, when the roundness of a colony is below 0.7, its morphology is considered to significantly deviate from the normal range. The trigger threshold for edge roughness (0.27) is determined based on the statistical distribution of normal colony samples: using the statistical values ​​of edge roughness from 3200 normal colony samples in the training set as a basis, with a distribution mean of 0.15 and a standard deviation σ=0.06, a 2σ threshold is set, meaning that when the edge roughness of a colony is >0.27 (0.15+2×0.06), it is considered morphologically abnormal. Edge roughness is defined as the standard deviation of the distance from the colony contour pixel to the boundary of its fitted ellipse.

[0086] When the output of the comparison analysis unit matches any rule, a corresponding feedback signal is generated and sent to the main control unit. In response to this feedback signal, the main control unit automatically performs at least one of the following operations: extending the culture period of the current specimen, fine-tuning the temperature or gas concentration settings within the sealed culture chamber, or triggering a supplementary biochemical identification procedure for the corresponding specimen.

[0087] In one specific embodiment, the interpretation process of the colony imaging interpretation module is as follows: The high-resolution imaging device automatically captures a 21-megapixel color image of each culture medium plate in the closed culture chamber at a time interval (every 2 hours) preset by the main control unit. Taking a clinical sputum specimen as an example, one plate image is acquired at each of the following time points during the culture cycle: 0h, 2h, 4h, 6h, 8h, 10h, 12h, and 24h, for a total of eight time points.

[0088] After receiving each image, the recognition unit immediately initiates the analysis process: scaling and white balance correction of the image, and background subtraction based on a Gaussian mixture model; using a trained colony segmentation and recognition model, the colony region is segmented from the culture medium background; morphological features are extracted from each segmented independent colony, including colony number, centroid coordinates, area (mm²), roundness, edge roughness, RGB color mean, and transparency. In this embodiment, if a plate identifies 23 independent colonies at T=12h, colony number 1 has an area of ​​0.78mm², roundness of 0.92, edge roughness of 0.12, and color mean RGB (245, 235, 220), and is preliminarily classified as a lactose fermentation colony.

[0089] The comparison and analysis unit compares the digitized list at the current time point (T=12h) with the list of the same plate at the previous time point (T=10h): by minimizing the Euclidean distance of the centroid coordinates, it confirms that colony 1 at T=10h and colony 1 at T=12h are the same bacterial colony. In this embodiment, the area of ​​the colony increased from 0.52mm² to 0.78mm² within 2 hours, the equivalent circle diameter increased from 0.81mm to 1.00mm, and the growth rate was 95μm / h, which is within the normal growth range of 50-200μm / h; at the same time, the comparison and analysis unit identified 3 newly formed colonies at T=12h, which were not detected at T=10h.

[0090] The feedback signal generation unit analyzes and compares the results in real time (including colony counts and morphological analysis reports). When it is determined that the morphological parameters and growth rates of all identified colonies are within the normal range, a "culture process is normal" status signal is generated, and the program continues to run according to the preset procedure.

[0091] In another specific embodiment, when the culture reached T=24h, a new colony No. 2 was detected on the plate, with a roundness of 0.51 (a deviation of 0.41 from the normal morphology of 0.92, which is >0.3) and an edge roughness of 0.35 (>0.27). This comparison result triggered the rules in Table 6. The feedback signal generation unit immediately sent a "morphological anomaly warning" coded signal to the main control unit. After responding, the main control unit automatically extended the culture time of the current specimen by 12 hours and sent supplementary biochemical identification request information, which at least included the specimen number, current culture conditions, and a description of the abnormal colony morphology.

[0092] See Figure 7 Another embodiment of the medical testing microbial culture system of the present invention includes: the incubator self-purification module is connected to the gas path of the environmental control culture module and the fully automatic processing module respectively. In response to the trigger signal of the end of the culture cycle or the switching of specimen batches, it autonomously starts a set of combined disinfection programs to perform full-coverage sterilization and purification on the closed culture chamber and the connected gas path pipelines to ensure that each specimen is cultured in a sterile environment.

[0093] Specifically, the core hardware architecture of the incubator self-cleaning module is as follows: The trigger signal receiving unit is connected to the sensor group and main control unit of the environmental control culture module via a CAN bus to receive disinfection trigger commands: the "cycle end" signal sent by the main control unit after the countdown of the preset culture cycle is reset to zero; when the online sensor group detects that at least two of the parameters of temperature, humidity and gas concentration in the closed culture chamber deviate from the preset culture set value and tend to the background level after the chamber door is opened, the main control unit determines and sends a "sample batch switching" signal.

[0094] The ultraviolet irradiation unit serves as the primary physical disinfection component. It consists of six quartz glass mercury lamps with a main wavelength of 254 nm, distributed across the top, four walls, and above the conveyor track within the sealed culture chamber. The irradiance of each mercury lamp at a distance of 1 meter from its surface is no less than 90 μW / cm², ensuring that under continuous irradiation for 30 minutes, the ultraviolet radiation dose (irradiance × time) received by all exposed surfaces of the chamber is no less than 162 mJ / cm², meeting the effective killing dose requirements for common bacterial vegetative cells and spores, including Bacillus subtilis var. niger.

[0095] The hydrogen peroxide dry fog disinfection unit serves as the execution component for the second stage of chemical disinfection. It is connected via independent disinfection pipelines to the internal pipes of the sealed culture chamber and the gas replacement unit and chemical catalytic deoxygenation unit of the fully automated processing module. It includes a flash vaporization device for instantly vaporizing a 30%–35% hydrogen peroxide solution into dry fog particles with an average particle size of less than 1 μm; and a circulating air pump for driving the dry fog particles to evenly diffuse along the gas path pipes into all connected chambers and pipes, chemically sterilizing shaded areas, crevices, and the inner walls of slender gas path pipes that cannot be directly exposed to ultraviolet light.

[0096] The connection between the hydrogen peroxide dry fog disinfection unit and the gas pipeline of the fully automatic processing module is specifically achieved through a three-way solenoid valve. Its common end connects to the dry fog delivery pipeline, its normally open end connects to the internal gas pipeline of the fully automatic processing module, and its normally closed end is sealed. During the incubation cycle, this solenoid valve is in a de-energized and normally closed state, ensuring complete physical isolation between the dry fog disinfection pipeline and the incubation gas pipeline, preventing interference. When the second stage of the disinfection program is initiated, the disinfection control unit sends an energizing signal to the solenoid valve, causing it to switch to the normally open position, thereby introducing hydrogen peroxide dry fog into the gas pipeline of the fully automatic processing module. After the disinfection stage ends, the solenoid valve is de-energized and reset, re-isolating the two gas pipelines.

[0097] The disinfection control unit is connected to the trigger signal receiving unit, the ultraviolet irradiation unit, and the hydrogen peroxide dry fog disinfection unit. Its internal non-volatile memory stores the complete disinfection program logic. It is used to schedule the execution order, duration, and intensity of ultraviolet irradiation and dry fog disinfection based on the type of the received trigger signal and the risk level of the current sample batch, and ultimately control the start and stop of the residue removal process.

[0098] The incubator's self-purification module is normally in a dormant standby state. When the main control unit counts down to zero according to the preset incubation cycle (48 hours), or when the sensor group detects that the temperature, humidity, and CO2 concentration in the sealed incubation chamber deviate from the set values ​​and approach the environmental background level, and the main control unit determines that "the specimen has been removed and the batch has been switched," the trigger signal receiving unit is activated, immediately waking up the disinfection control unit. The purification program then starts automatically, requiring no manual intervention throughout the process. The specific disinfection procedure is as follows: The first stage is the ultraviolet (UV) irradiation stage. Upon receiving a trigger signal, the disinfection control unit activates the UV irradiation unit, simultaneously illuminating all UVC mercury lamps within the chamber. This provides continuous, direct, and high-intensity irradiation to all exposed solid surfaces, including the inner surface of the sealed culture chamber, the shelves, and the conveyor tracks. The 254nm wavelength UV radiation causes irreversible damage to the DNA / RNA structure of bacteria and viruses attached to the surfaces, killing most directly exposed microorganisms. This stage is complete when the timer inside the disinfection control unit records that the UV irradiation unit has been operating continuously for a preset 30 minutes. The disinfection control unit then automatically cuts off the power supply to the UV irradiation unit and, through its pre-stored program logic, sends a start command to the flash vaporization device and circulating air pump of the hydrogen peroxide dry fog disinfection unit after a 5-second delay, seamlessly transitioning to the second stage.

[0099] The second stage is the hydrogen peroxide dry fog disinfection stage. A flash vaporization device vaporizes the hydrogen peroxide solution into micron-sized dry fog particles. A circulating air pump actively delivers these dry fog particles at a flow rate of 10 L / min through dedicated pipelines to every corner of the sealed culture chamber. The particles then enter the internal walls of the gas replacement unit and chemical catalytic deoxygenation unit of the fully automated processing module, where they fully diffuse and act for at least 45 minutes. Utilizing the small particle size (<1 μm) of hydrogen peroxide dry fog, it possesses excellent diffusivity, allowing it to penetrate shaded areas and crevices where ultraviolet light cannot directly reach. The third stage is the residue removal stage. The disinfection control unit instructs the gas circuit valves to switch to external circulation mode, purging the entire closed culture chamber and all connected gas circuit pipelines with ambient air filtered by the HEPA high-efficiency filter at a flow rate of 30L / min. Simultaneously, the hydrogen peroxide concentration sensor located at the exhaust port monitors the concentration of the exhaust gas in real time at a frequency of 1Hz. The disinfection control unit continuously reads this concentration value until it falls below the safe exposure limit of hydrogen peroxide (1ppm), and maintains this state for 30 seconds. After that, it determines that the residue removal is complete, and then sends a "system safe and ready" signal to the main control unit. The system then returns to a sterile, ready-to-operate state.

[0100] When the status register of the disinfection control unit indicates that it is currently in the "disinfection program running" state, or when the reading of the hydrogen peroxide concentration sensor located at the exhaust port is higher than 1 ppm at any given time, the main control unit continuously sends a locking signal to the chamber door electromagnetic lock through its digital output port, keeping it in the locked state to ensure the safety of the operator. The locking signal is only released when the disinfection control unit is in the "standby" state and the concentration sensor reading is consistently lower than 1 ppm.

[0101] It should be noted that the non-volatile memory of the disinfection control unit pre-stores multiple disinfection modes (as shown in Table 7), including a rapid mode suitable for routine batch processing and an enhanced mode suitable for processing highly infectious specimens. Based on specimen type information received from the main control unit or manual instructions, the disinfection control unit automatically calls the corresponding program from the mode library to perform ultraviolet irradiation and / or hydrogen peroxide dry fog disinfection of different intensities and durations.

[0102] Table 7

[0103] The execution parameters for each stage can be configured independently as shown in Table 7. Notably, the dry fog disinfection duration in the rapid mode is not set; that is, after the ultraviolet irradiation stage, the disinfection control unit skips the generation of hydrogen peroxide dry fog and directly executes the ventilation replacement stage for residual removal to quickly restore the cavity to its standby state.

[0104] When the disinfection control unit receives a trigger signal, a confirmation dialog box asking "Perform self-cleaning?" pops up on the system's touchscreen display. The disinfection program will only officially start if the operator clicks "confirm" within 30 seconds. If no confirmation is received within 30 seconds, the system will default to standby mode and log the information to prevent unnecessary purification operations due to system misjudgment.

[0105] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other. In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A medical testing microbial culture system, characterized in that, include: Pretreatment and inoculation module: Used to automatically complete the standardized streaking and inoculation of clinical specimens via a robotic arm, and to transfer the inoculated culture medium to the environmentally controlled culture module; Environmental control and cultivation module: It is equipped with a closed cultivation chamber and integrates a temperature control unit, a humidity control unit and a carbon dioxide control unit. The temperature control unit, humidity control unit and carbon dioxide control unit work together to regulate the temperature, humidity and carbon dioxide concentration in the closed cultivation chamber, and are used to monitor the cultivation environment parameters in the closed cultivation chamber in real time. Fully automated processing module: used to establish and maintain the colony culture environment within the closed culture chamber based on the culture environment parameters, through a dual mechanism of automatic gas replacement and chemical catalytic deoxygenation; specifically including: The sealed culture chamber is evacuated to a preset negative pressure and then filled with replacement gas. This process is repeated at least once to reduce the oxygen concentration in the chamber to a first threshold. When the oxygen concentration falls below the first threshold, the chamber automatically activates, driving the gas to circulate through the catalytic reaction column to further reduce the oxygen concentration to below a second threshold. The preset negative pressure is -80 kPa, the flow rate of the replacement gas is 15 L / min, and the number of cycles is 3. The first threshold is 1.0% vol, and the second threshold is 50 ppm. The catalytic reaction column is filled with a desiccant to adsorb trace amounts of moisture generated during the catalytic deoxygenation process. Colony imaging and interpretation module: used to periodically acquire plate images of the culture medium during the culture process, perform colony counting and morphological analysis by comparing images at different time points, and send feedback signals to the environmental control culture module to trigger the review of the corresponding specimen culture conditions or specific supplementary tests; including an image recognition unit, a comparison analysis unit and a feedback signal generation unit; The image recognition unit is configured to run a colony segmentation and recognition model based on the U-Net architecture, used to perform colony segmentation and recognition on the acquired plate images; the comparison and analysis unit performs differential comparison based on images acquired from the same plate at different time points, identifies the same bacterial cells by centroid coordinate matching, and calculates their colony growth rate; the feedback signal generation unit generates a culture condition verification signal and sends it to the environment control culture module when the growth rate exceeds the preset normal range. The incubator self-purification module is connected to the gas path of both the environmental control incubation module and the fully automatic processing module. It is used to initiate a sterilization program in response to trigger signals indicating the end of the incubation cycle or sample batch switching, sterilizing and purifying the sealed incubation chamber and its connected gas path pipelines. Specifically, this includes: Based on the specimen type information corresponding to the trigger signal received from the main control unit, the corresponding mode is called from multiple disinfection modes to execute a disinfection procedure that matches the risk level of the current specimen batch; wherein, the disinfection procedure includes a UV irradiation stage, a hydrogen peroxide dry fog disinfection stage, and a residue removal stage executed sequentially; the disinfection modes include at least a standard mode, a rapid mode, and an enhanced mode, and the execution duration of the UV irradiation stage and / or the hydrogen peroxide dry fog disinfection stage is different in each mode.

2. The medical testing microbial culture system according to claim 1, characterized in that: The pretreatment inoculation module includes an automatic marking inoculation unit, which stores a variety of standardized marking programs; The robotic arm is configured to select the corresponding program from the various standardized streaking programs based on the specimen type information of the clinical specimen, so as to automatically complete the zonal streaking or quantitative coating operation. The end effector of the robotic arm is equipped with a force sensor to provide real-time feedback on the pressure of the inoculation line. And a speed controller, based on real-time feedback from force sensors, uses a PID algorithm to adjust the robotic arm's moving speed to keep it stable near a preset value.

3. The medical testing microbial culture system according to claim 1, characterized in that: The environmental control culture module integrates a distributed sensor array, including at least multiple temperature and humidity sensors respectively configured in the upper, middle, and lower layers of the closed culture chamber: The output power of the humidifier is decoupled and controlled by a pre-stored fuzzy logic controller.

4. A medical testing microbial culture system according to claim 3, characterized in that: The input variables of the fuzzy logic controller include humidity deviation and its rate of change, and its output variable is the output power increment of the humidifier. The humidity deviation refers to the difference between the current humidity and the set value, and the humidity change rate is the amount of change of the humidity deviation over time.

Images