A method and system for automatic regulation of a cell culture process

Abstract

The application provides a cell culture process automatic regulation method and system, and relates to the technical field of biosensors. A non-contact optical trap is formed inside a culture container by focusing a laser beam, and extracellular vesicles suspended in the culture solution are captured by the optical trap. Excitation light is applied to the captured extracellular vesicles, and Raman scattering spectra are collected. The lipid phase transition temperature and nucleic acid characteristic peak intensity of the extracellular vesicles are analyzed from the characteristic peaks of the Raman scattering spectra. The arrival frequency of the captured extracellular vesicles is counted as the frequency of vesicle release events. The lipid phase transition temperature and nucleic acid characteristic peak intensity of the extracellular vesicles within a preset time window are used to determine a vesicle maturity index. The cell culture state is determined and automatically regulated according to the vesicle maturity index, the frequency of vesicle release events, and the culture environment parameters monitored by the biosensor. The application can realize in-situ dynamic tracking of extracellular vesicles, thereby providing precise markers for cell state regulation.

Description

Technical Field

[0001] This application relates to the field of biosensor technology, and more specifically, to an automatic control method and system for cell culture processes. Background Technology

[0002] With the rapid development of biopharmaceuticals, cell therapy, and other fields, biosensor technology, as a core detection technology, is becoming increasingly important. This technology boasts advantages such as high sensitivity, high specificity, and real-time response, directly capturing relevant biological signals and environmental parameters in cell culture systems without the need for complex offline processing. It is a key support for upgrading cell culture from traditional manual control to automation and precision, providing a reliable basis for accurate judgment of cell growth status. This has significant practical implications for ensuring the quality of cultured products, improving culture efficiency, and promoting the high-quality development of related industries.

[0003] The core challenge of cell culture lies in the precise control of cellular physiological state. Extracellular vesicles, as important indicators of cellular physiological state, are crucial biomarkers reflecting cellular physiological state and secretory function. The effectiveness of their monitoring directly impacts the scientific nature of culture regulation. Current technologies for extracellular vesicle monitoring have significant limitations, relying heavily on offline detection methods. These methods cannot achieve in-situ dynamic tracking during culture, and the detection process is complex and time-consuming, making it difficult to quickly capture dynamic changes in vesicle characteristics. Furthermore, current technologies struggle to establish a precise correlation between vesicle characteristics and the actual state of cells, resulting in the inability to obtain timely and effective state feedback during cell culture. This lack of targeted regulation hinders the application of automated cell culture technologies. Therefore, achieving in-situ dynamic tracking of extracellular vesicles to provide precise biomarkers for cell state regulation has become a major challenge for the industry. Summary of the Invention

[0004] This application provides an automated control method and system for cell culture, which enables in-situ dynamic tracking of extracellular vesicles, thereby providing precise markers for cell state regulation.

[0005] In a first aspect, this application provides an automatic control method for cell culture, the automatic control method comprising the following steps:

[0006] Step S1: Set up a biosensor inside the culture container and form a non-contact optical trap inside the culture container by focusing a laser beam to capture extracellular vesicles suspended in the culture medium.

[0007] Step S2: Apply excitation light to the captured extracellular vesicles and collect their Raman scattering spectra. Analyze the lipid phase transition temperature and nucleic acid characteristic peak intensity of the extracellular vesicles from the characteristic peaks of the Raman scattering spectra.

[0008] Step S3: Statistically determine the arrival frequency of captured extracellular vesicles within the preset time window as the vesicle release event frequency, and determine the vesicle maturity index based on the lipid phase transition temperature and nucleic acid characteristic peak intensity of extracellular vesicles within the preset time window.

[0009] Step S4: Determine the cell culture status and automatically adjust it based on the vesicle maturity index, the frequency of vesicle release events, and the culture environment parameters monitored by the biosensor.

[0010] In this embodiment, step S1 specifically includes:

[0011] The laser beam is highly focused onto a preset capture point inside the culture container using an objective lens to form a three-dimensional gradient force-optical trap.

[0012] The field of view near the preset capture point is monitored in real time. When extracellular vesicles are detected to diffuse past the preset capture point, the three-dimensional gradient force light trap captures and stably binds them to the center of the focal point, thus obtaining the captured extracellular vesicles.

[0013] In this embodiment, step S2 specifically includes:

[0014] Excitation light is focused onto the region where the captured extracellular vesicles are located, and Raman scattering spectra generated after vesicle molecules are scattered are collected.

[0015] The Raman scattering spectrum was baseline corrected and noise-reduced, and the peak position shift of lipid-related characteristic peaks was extracted and analyzed. The lipid phase transition temperature of extracellular vesicles was then determined from the peak position shift.

[0016] The peak areas of nucleic acid-related characteristic peaks were extracted and analyzed, and their values ​​were used as the intensity of nucleic acid characteristic peaks of extracellular vesicles.

[0017] In this embodiment, step S3 specifically includes:

[0018] Record the total number of vesicles successfully captured and spectral measurements completed within a preset time window, and determine the vesicle release event frequency based on the total number of vesicles and the duration of the preset time window;

[0019] The deviation of the lipid phase transition temperature of extracellular vesicles is determined based on the lipid phase transition temperature of extracellular vesicles within a preset time window.

[0020] The deviation of nucleic acid peak intensity of extracellular vesicles is determined based on the intensity of the characteristic nucleic acid peaks of extracellular vesicles within a preset time window;

[0021] The lipid phase transition temperature deviation and the nucleic acid peak intensity deviation are fuzzified to obtain the lipid phase transition temperature membership value and the nucleic acid peak intensity membership value.

[0022] Based on a preset fuzzy reasoning rule base, fuzzy reasoning is performed on the membership values ​​of the lipid phase transition temperature and the nucleic acid peak intensity to obtain the vesicle maturity index.

[0023] In this embodiment, step S4 specifically includes:

[0024] Operation A: When the frequency of vesicle release events increases but the vesicle maturity index decreases, the cell is determined to be in a state of high stress secretion based on the culture environment parameters monitored by the biosensor and the antioxidant injection procedure is initiated.

[0025] Operation B: When the frequency of vesicle release events increases synchronously with the vesicle maturity index, the cells are determined to be in a state of normal functional enhancement based on the culture environment parameters monitored by the biosensor, and the current culture conditions are maintained.

[0026] In this embodiment, operation A specifically includes:

[0027] Operation A1: Obtain the frequency of vesicle release events and the vesicle maturity index for the current time window, and compare them with their respective normal fluctuation threshold ranges;

[0028] Operation A2: When it is determined that the frequency of the vesicle release event exceeds the upper limit of its normal fluctuation threshold range, and at the same time it is determined that the vesicle maturity index is lower than the lower limit of its normal fluctuation threshold range, the culture environment parameters monitored by the biosensor are read, and a stress hypersecretion state determination signal is generated.

[0029] Operation A3: Based on the stress hypersecretion state determination signal, turn on the culture medium micro-injection pump to extract and inject a preset volume of antioxidant from the antioxidant storage bottle into the culture container.

[0030] In this embodiment, operation A2 specifically includes:

[0031] When it is determined that the frequency of the vesicle release event exceeds the upper limit of its normal fluctuation threshold range, and at the same time it is determined that the vesicle maturity index is lower than the lower limit of its normal fluctuation threshold range, the concentration of reactive oxygen free radicals in the culture environment parameters monitored by the biosensor is read.

[0032] Contour extraction and surface texture analysis were performed on extracellular vesicle images to extract the in-situ morphological features of extracellular vesicles.

[0033] By detecting transient changes in membrane potential at extracellular vesicle release sites, the potential coupling phase characteristics of the extracellular secretion process can be extracted.

[0034] A stress-induced hypersecretion state determination signal is generated based on the concentration of reactive oxygen species, the in-situ morphological characteristics, and the potential coupling phase characteristics.

[0035] In this embodiment, operation A3 specifically includes:

[0036] Based on the stress hypersecretion state determination signal, the total volume of the culture medium in the current culture container is obtained, and the amount of antioxidant to be injected is calculated based on the preset final concentration of antioxidant.

[0037] A start command is sent to the culture medium micro-injection pump connected to the antioxidant storage bottle, causing the culture medium micro-injection pump to draw and deliver the antioxidant into the culture container at a constant flow rate;

[0038] After delivery, turn off the culture medium micro-injection pump and record the injection time and volume of the antioxidant.

[0039] In this embodiment, operation B specifically includes:

[0040] Obtain the frequency of vesicle release events and the vesicle maturity index for the current time window, and compare them with their respective normal fluctuation threshold ranges.

[0041] When it is determined that the frequency of the vesicle release event exceeds the upper limit of its normal fluctuation threshold range, and at the same time it is determined that the vesicle maturity index also exceeds the upper limit of its normal fluctuation threshold range, the culture environment parameters monitored by the biosensor are read, and a normal functional enhancement state determination signal is generated.

[0042] Based on the normal function enhancement state determination signal, the temperature, humidity, carbon dioxide concentration, and culture medium perfusion rate of the incubator are kept constant.

[0043] Secondly, this application provides an automatic control system for cell culture processes, used to execute an automatic control method for cell culture processes. The automatic control system includes:

[0044] An extracellular vesicle capture module is used to set up a biosensor inside a culture container. A non-contact optical trap is formed inside the culture container by a focused laser beam, and the optical trap is used to capture extracellular vesicles suspended in the culture medium.

[0045] An extracellular vesicle feature extraction module is used to apply excitation light to the captured extracellular vesicles and collect their Raman scattering spectra, and to analyze the lipid phase transition temperature and nucleic acid characteristic peak intensity of the extracellular vesicles from the characteristic peaks of the Raman scattering spectra.

[0046] The vesicle maturity analysis module is used to count the arrival frequency of captured extracellular vesicles within a preset time window as the vesicle release event frequency, and to determine the vesicle maturity index based on the lipid phase transition temperature and nucleic acid characteristic peak intensity of extracellular vesicles within the preset time window.

[0047] The culture status analysis and automatic control module determines the cell culture status and performs automatic control based on the vesicle maturity index, the frequency of vesicle release events, and the culture environment parameters monitored by the biosensor.

[0048] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects:

[0049] Step S1: A biosensor is installed inside the culture container. A non-contact optical trap is formed inside the culture container using a focused laser beam to capture extracellular vesicles suspended in the culture medium. Step S2: Excitation light is applied to the captured extracellular vesicles, and their Raman scattering spectra are collected. The lipid phase transition temperature and nucleic acid characteristic peak intensity of the extracellular vesicles are analyzed from the characteristic peaks of the Raman scattering spectra. Step S3: The arrival frequency of captured extracellular vesicles within a preset time window is counted as the vesicle release event frequency. The vesicle maturity index is determined based on the lipid phase transition temperature and nucleic acid characteristic peak intensity of the extracellular vesicles within the preset time window. Step S4: The cell culture status is determined and automatically adjusted based on the vesicle maturity index, the vesicle release event frequency, and the culture environment parameters monitored by the biosensor.

[0050] Therefore, this application demonstrates that the cell culture status can be determined and automatically controlled based on the vesicle maturity index, the vesicle release event frequency, and the culture environment parameters monitored by the biosensor. Firstly, a biosensor is installed inside the culture container. A laser beam is highly focused onto a preset capture point within the culture container using an objective lens to form a three-dimensional gradient force trap. The field of view is monitored in real time, and vesicles are captured and stably bound to the focal point as they diffuse. This allows for in-situ, non-contact dynamic capture of extracellular vesicles without disrupting the culture environment or interfering with normal cell culture, overcoming the limitations of offline detection methods that require intervention in the culture process. Secondly, excitation light is focused onto the area where the captured vesicles are located, and Raman scattering spectra are collected. After baseline correction and noise reduction, the lipid phase transition temperature and nucleic acid characteristic peak intensity are analyzed. This enables real-time, non-destructive, and rapid analysis of the core physicochemical characteristics of vesicles, effectively addressing the technical pain points of existing detection methods, which are complex, time-consuming, and difficult to rapidly capture dynamic changes in vesicle characteristics. Furthermore, the total number of vesicles within a preset time window is counted to obtain the vesicle release event frequency, which is then combined with the lipid phase transition temperature deviation and nucleic acid... By employing fuzzy processing and fuzzy inference to quantify peak intensity deviation, the dynamic release behavior and intrinsic maturity characteristics of vesicles are transformed into quantifiable vesicle maturity indicators. This establishes a precise quantitative correlation system between vesicle characteristics and cell physiological state, overcoming the bottleneck of existing technologies that cannot accurately bind vesicle characteristics to actual cell state. Furthermore, when the frequency of vesicle release events increases but the maturity indicator decreases, a stress-induced hypersecretion state judgment signal is generated through biosensor-monitored culture environment parameters and multi-dimensional feature verification, and antioxidants are precisely injected. When both indicators increase simultaneously, a normal function enhancement state judgment signal is generated and culture conditions are maintained, achieving precise cell state judgment and targeted automatic regulation based on in-situ dynamic markers of vesicles. Finally, through the entire process of in-situ capture, real-time analysis, quantitative characterization, and closed-loop regulation, a complete technical process for in-situ dynamic tracking of extracellular vesicles and precise regulation of cell state is constructed. This makes extracellular vesicles truly precise markers reflecting cell physiological state and secretory function, providing real-time and reliable state feedback for the precise control of automated cell culture, and significantly improving the scientific rigor and targeting of cell culture regulation.

[0051] In summary, the technical solution adopted in this application can realize in-situ dynamic tracking of extracellular vesicles, thereby providing precise markers for cell state regulation. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this embodiment of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 This is an exemplary flowchart of an automatic control method for cell culture process provided in this application;

[0054] Figure 2 This is a schematic diagram of the structure for extracellular vesicle capture and monitoring provided in this application;

[0055] Figure 3 This is a flowchart illustrating the process for determining vesicle maturity indicators provided in this application;

[0056] Figure 4 This is a modular structure diagram of an automatic control system for cell culture process provided in this application. Detailed Implementation

[0057] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0058] This application provides an automatic control method and system for cell culture, the core of which is step S1: forming a non-contact optical trap inside the culture container using a focused laser beam, and capturing extracellular vesicles suspended in the culture medium using the optical trap; step S2: applying excitation light to the captured extracellular vesicles and collecting their Raman scattering spectra, and analyzing the lipid phase transition temperature and nucleic acid characteristic peak intensity of the extracellular vesicles from the characteristic peaks of the Raman scattering spectra; step S3: statistically analyzing the arrival frequency of captured extracellular vesicles within a preset time window as the vesicle release event frequency, and determining the vesicle maturity index based on the lipid phase transition temperature and nucleic acid characteristic peak intensity of the extracellular vesicles within the preset time window; step S4: determining the cell culture state based on the vesicle maturity index and the vesicle release event frequency and performing automatic control.

[0059] Example 1: To better understand the above technical solution, the following will provide a detailed description of the technical solution in conjunction with the accompanying drawings and specific implementation methods. (Refer to...) Figure 1 As shown in the figure, this is an exemplary flowchart of an automatic control method for a cell culture process according to this embodiment of the present application. The automatic control method includes the following steps:

[0060] In step S1, a biosensor is placed inside the culture container, and a non-contact optical trap is formed inside the culture container by focusing a laser beam to capture extracellular vesicles suspended in the culture medium.

[0061] It should be noted that the biosensor described in this application is an integrated biosensor device for real-time monitoring of cell culture environment parameters (including but not limited to reactive oxygen free radical concentration, culture medium pH, temperature, and dissolved oxygen concentration). In specific implementation, the probe end of the biosensor is inserted into the container through a pre-set sealed interface on the side wall of the culture container. The insertion depth is controlled to be 2-3 cm below the surface of the culture medium, and the probe end avoids the pre-set capture point of the subsequent optical trap to avoid interference from the laser beam on the sensor's detection accuracy. Then, a medical-grade silicone sealing ring is used to seal the interface to prevent leakage of the culture medium.

[0062] It should also be noted that in this application, references Figure 2 As shown in the figure, this is a schematic diagram of extracellular vesicle capture and monitoring provided in an embodiment of this application. In the figure, the cell culture dish contains culture medium and multiple extracellular vesicles are suspended in the medium. A biosensor is set at the bottom of the culture dish to collect the culture environment parameters of the microenvironment in which the extracellular vesicles are located in real time. The optical device above emits a vertical laser beam, which is focused to form a laser focal spot. The laser focal spot can capture extracellular vesicles at the target location in the culture medium in a non-contact manner, achieving non-damaging in-situ fixation and avoiding the damage to the vesicle structure caused by traditional capture methods.

[0063] In this embodiment, the non-contact optical trap formed inside the culture vessel by focusing a laser beam, and the capture of extracellular vesicles suspended in the culture medium using the optical trap, can be achieved through the following steps:

[0064] The laser beam is highly focused onto a preset capture point inside the culture container using an objective lens to form a three-dimensional gradient force-optical trap.

[0065] The field of view near the preset capture point is monitored in real time. When extracellular vesicles are detected to diffuse past the preset capture point, the three-dimensional gradient force light trap captures and stably binds them to the center of the focal point, thus obtaining the captured extracellular vesicles.

[0066] It should be noted that the three-dimensional gradient force optical trap described in this application refers to a non-contact optical trap structure with three-dimensional spatial optical confinement capability formed in the culture medium by highly focusing a laser beam.

[0067] It should also be noted that the preset capture point mentioned in this application refers to the target position pre-calibrated inside the culture container for laser focusing and extracellular vesicle capture. The capture point can be preset in the following way: First, the culture medium inside the culture container is pre-scanned in its entirety by a microscopic imaging module. The image segmentation algorithm is used to identify the cell aggregation area and the extracellular vesicle diffusion area. In order to avoid interfering with normal cell growth and to ensure capture efficiency, the densely aggregated cell area is avoided. The area with the highest extracellular vesicle diffusion frequency and a moderate distance from the cell secretion site is selected as the calibration area. The geometric center of this area is calibrated by a microscopic coordinate system. The position corresponding to the calibrated coordinates is used as the preset capture point. The preset is completed and the coordinates are recorded for subsequent laser focusing positioning.

[0068] In practice, firstly, a high numerical aperture microscope objective is used to highly focus the laser beam so that the laser focus falls precisely on a pre-marked capture point inside the culture container. The three-dimensional optical gradient force confinement structure formed at this focal point is used as a three-dimensional gradient force optical trap. Secondly, the culture medium field of view near the pre-marked capture point is acquired and monitored in real time by a microscopic imaging module. A target recognition algorithm is used to detect the extracellular vesicles diffusing in the field of view. When an extracellular vesicle is detected moving past the pre-marked capture point, the optical confinement effect of the three-dimensional gradient force optical trap captures and stably confines the extracellular vesicle to the center of the laser focus. The stably confined extracellular vesicle is used as the captured extracellular vesicle.

[0069] In step S2, excitation light is applied to the captured extracellular vesicles and their Raman scattering spectra are collected. The lipid phase transition temperature and nucleic acid characteristic peak intensity of the extracellular vesicles are analyzed from the characteristic peaks of the Raman scattering spectra.

[0070] In this embodiment, the capture of extracellular vesicles is subjected to excitation light and their Raman scattering spectra are collected. The lipid phase transition temperature and nucleic acid characteristic peak intensity of the extracellular vesicles can be determined from the characteristic peaks of the Raman scattering spectra using the following steps:

[0071] Excitation light is focused onto the region where the captured extracellular vesicles are located, and Raman scattering spectra generated after vesicle molecules are scattered are collected.

[0072] The Raman scattering spectrum was baseline corrected and noise-reduced, and the peak position shift of lipid-related characteristic peaks was extracted and analyzed. The lipid phase transition temperature of extracellular vesicles was then determined from the peak position shift.

[0073] The peak areas of nucleic acid-related characteristic peaks were extracted and analyzed, and their values ​​were used as the intensity of nucleic acid characteristic peaks of extracellular vesicles.

[0074] It should be noted that the Raman scattering spectrum mentioned in this application refers to the characteristic spectrum carrying chemical composition information generated by molecular vibration scattering after the captured extracellular vesicles are irradiated with excitation light; the lipid phase transition temperature refers to the characteristic temperature parameter reflecting the change in the arrangement state of lipid molecules in extracellular vesicles; and the nucleic acid characteristic peak intensity refers to the characteristic parameter of the nucleic acid content in extracellular vesicles.

[0075] In practice, firstly, the excitation light is precisely focused onto the region where the captured extracellular vesicles are located, and the optical signal generated by the scattering of vesicle molecules is collected. The characteristic spectrum formed by converting this optical signal is used as the Raman scattering spectrum. Secondly, a polynomial fitting algorithm is used to perform baseline correction on the Raman scattering spectrum, and a wavelet denoising algorithm is used to complete the spectral noise reduction. In the processed spectrum, the characteristic peak corresponding to the lipids of the extracellular vesicles is located. The actual peak position of the characteristic peak is detected using a peak position detection algorithm. The difference between the actual peak position and the standard lipid peak position is calculated, and this difference is used as the peak position offset of the lipid-related characteristic peak. The peak position offset is divided by the calibrated lipid Raman peak position temperature coefficient to obtain the temperature change of the extracellular vesicle relative to the reference temperature. The temperature change is added to the reference temperature, and the sum is used as the lipid phase transition temperature of the extracellular vesicle. Finally, the nucleic acid-related characteristic peak is located in the processed Raman scattering spectrum, and the peak area value of the characteristic peak is calculated using a peak area integration algorithm. This peak area value is used as the nucleic acid characteristic peak intensity of the extracellular vesicle.

[0076] It should be noted that the lipid Raman peak position temperature coefficient described in this application is obtained through a gradient temperature control calibration experiment using standard lipid samples. It represents the linear change coefficient of the peak position of lipid Raman characteristic peaks with temperature and can be obtained through the following experimental calibration procedure: Select a standard lipid sample with the same lipid composition as extracellular vesicles, prepare a suspension with the same system as the culture medium, place the suspension on a microscopic temperature control platform, set a gradient temperature in 1℃ increments covering the entire lipid phase transition range, use the same excitation light parameters, focusing conditions, and spectral acquisition parameters as the system under test at each temperature point, collect the Raman scattering spectrum of the standard lipid, perform baseline correction and noise reduction on the spectra at each temperature point, use the Lorentz function fitting algorithm to locate the lipid characteristic peaks and determine the peak positions, establish data points with temperature as the abscissa and peak position as the ordinate, perform linear fitting using the least squares method, and the slope of the fitted line is the lipid Raman peak position temperature coefficient. Repeat the independent calibration three times and take the average value as the final value.

[0077] In step S3, the arrival frequency of captured extracellular vesicles within a preset time window is counted as the vesicle release event frequency, and the vesicle maturity index is determined based on the lipid phase transition temperature and nucleic acid characteristic peak intensity of extracellular vesicles within the preset time window.

[0078] Preferably, in this embodiment, the arrival frequency of captured extracellular vesicles within a preset time window is used as the vesicle release event frequency. The vesicle maturity index is determined based on the lipid phase transition temperature and nucleic acid characteristic peak intensity of the extracellular vesicles within the preset time window, with reference to... Figure 3 As shown in the figure, this figure illustrates the determination of vesicle maturity indicators in some embodiments of this application. In this embodiment, the determination of vesicle maturity indicators can be achieved using the following steps:

[0079] In step S31, the total number of vesicles successfully captured and completed spectral measurement within a preset time window is recorded, and the frequency of vesicle release events is determined based on the total number of vesicles and the duration of the preset time window.

[0080] In step S32, the deviation of the lipid phase transition temperature of extracellular vesicles is determined based on the lipid phase transition temperature of extracellular vesicles within a preset time window.

[0081] In step S33, the deviation of the nucleic acid peak intensity of extracellular vesicles is determined based on the intensity of the characteristic nucleic acid peak of extracellular vesicles within a preset time window;

[0082] In step S34, the lipid phase transition temperature deviation and the nucleic acid peak intensity deviation are fuzzified to obtain the lipid phase transition temperature membership value and the nucleic acid peak intensity membership value.

[0083] In step S35, fuzzy reasoning is performed on the lipid phase transition temperature membership value and nucleic acid peak intensity membership value according to the preset fuzzy reasoning rule base to obtain the vesicle maturity index.

[0084] It should be noted that the preset time window mentioned in this application refers to a continuous fixed time interval used to statistically analyze the frequency of vesicle release events and calculate vesicle maturity indicators. The preset time window is set as follows: based on the vesicle secretion cycle of the target cell, the time consumed by single vesicle spectral detection, and the stability of statistical data, the basic secretion cycle of the cell is first determined through preliminary experiments. Combined with the average time consumed by optical trap capture and Raman detection, a time interval covering more than 3 secretion cycles and ensuring a sufficient number of statistical samples is selected. After optimization through more than 5 repeated experiments, the optimal fixed duration is determined as the preset time window.

[0085] It should also be noted that, in this application, the vesicle release event frequency represents the average number of extracellular vesicles released by cells and successfully captured and detected within a preset time window, characterizing the release rate of extracellular vesicles; the lipid phase transition temperature deviation represents the degree of deviation between the average lipid phase transition temperature of all detected extracellular vesicles within the preset time window and the standard lipid phase transition temperature; the nucleic acid peak intensity deviation represents the degree of deviation between the average nucleic acid characteristic peak intensity of all detected extracellular vesicles within the preset time window and the standard nucleic acid characteristic peak intensity; the fuzzification process represents the process of converting the two precise values ​​of lipid phase transition temperature deviation and nucleic acid peak intensity deviation into membership values ​​in a fuzzy set; the lipid phase transition temperature membership value and the nucleic acid peak intensity membership value represent the degree to which the corresponding deviation belongs to a certain fuzzy level, with a value range of 0-1; the vesicle maturity index represents an indicator parameter characterizing the maturity of extracellular vesicles, used to determine the cell culture status.

[0086] Additionally, it should be noted that the fuzzy inference rule base described in this application refers to a set of pre-defined inference rules based on the correlation between extracellular vesicle maturity and lipid and nucleic acid characteristics. The pre-defined fuzzy inference rule base is set up as follows: it is constructed based on prior biological knowledge of extracellular vesicle maturity. First, the membership values ​​of lipid phase transition temperature and nucleic acid peak intensity are divided into three fuzzy subsets: low, medium, and high. Then, core inference rules are written (if both lipid and nucleic acid membership values ​​are high, then the vesicle maturity is high; if both lipid and nucleic acid membership values ​​are medium, then the vesicle maturity is medium; if both lipid and nucleic acid membership values ​​are low, then the vesicle maturity is low). After constructing the rule set, standard mature / immature vesicle samples are used for verification and iterative optimization, ultimately forming a fuzzy inference rule base adapted to this system.

[0087] In practice, firstly, the total number of extracellular vesicles successfully captured by the optical trap and subjected to Raman scattering spectroscopy within a preset time window is recorded in real time. This total number of vesicles is divided by the duration of the preset time window to obtain the number of vesicles captured per unit time, which is used as the vesicle release event frequency. Secondly, the arithmetic mean of the lipid phase transition temperatures of all detected extracellular vesicles within the preset time window is calculated. The difference between this average and the standard lipid phase transition temperature is calculated using the absolute value method, and this difference is used as the lipid phase transition temperature deviation of the extracellular vesicles. Next, the arithmetic mean of the nucleic acid characteristic peak intensities of all detected extracellular vesicles within the preset time window is calculated, and the absolute value method is used to calculate the average... The difference between the mean and the standard nucleic acid characteristic peak intensity is used as the nucleic acid peak intensity deviation of extracellular vesicles. Then, a triangular membership function is used to fuzzify the lipid phase transition temperature deviation and the nucleic acid peak intensity deviation, respectively. The corresponding membership values ​​are mapped according to the magnitude of the deviation, and these two values ​​are used as the lipid phase transition temperature membership value and the nucleic acid peak intensity membership value, respectively. Finally, a preset fuzzy inference rule base (containing the corresponding rules of different membership combinations and vesicle maturity) is called, and the Mamdani fuzzy inference algorithm is used to infer the lipid phase transition temperature membership value and the nucleic acid peak intensity membership value. The comprehensive output value obtained from the calculation is used as the vesicle maturity index.

[0088] In step S4, the cell culture status is determined and automatically adjusted based on the vesicle maturity index and the frequency of vesicle release events.

[0089] In this embodiment, determining the cell culture status and automatically regulating it based on the vesicle maturity index and the frequency of vesicle release events can be achieved through the following steps:

[0090] Operation A: When the frequency of vesicle release events increases but the vesicle maturity index decreases, the cell is determined to be in a state of high stress secretion based on the culture environment parameters monitored by the biosensor and the antioxidant injection procedure is initiated.

[0091] Operation B: When the frequency of vesicle release events increases synchronously with the vesicle maturity index, the cells are determined to be in a state of normal functional enhancement based on the culture environment parameters monitored by the biosensor, and the current culture conditions are maintained.

[0092] In this embodiment, reference Figure 2 As shown in the figure, this is a flowchart illustrating the process of determining the cell's stress-induced secretion state and initiating an antioxidant injection procedure according to this application. In this embodiment, operation A: when the frequency of vesicle release events increases but the vesicle maturity index decreases, determining that the cell is in a stress-induced hypersecretion state based on the culture environment parameters monitored by the biosensor and initiating the antioxidant injection procedure can be achieved through the following steps:

[0093] Operation A1: Obtain the frequency of vesicle release events and the vesicle maturity index for the current time window, and compare them with their respective normal fluctuation threshold ranges;

[0094] Operation A2: When it is determined that the frequency of the vesicle release event exceeds the upper limit of its normal fluctuation threshold range, and at the same time it is determined that the vesicle maturity index is lower than the lower limit of its normal fluctuation threshold range, the culture environment parameters monitored by the biosensor are read, and a stress hypersecretion state determination signal is generated.

[0095] Operation A3: Based on the stress hypersecretion state determination signal, turn on the culture medium micro-injection pump to extract and inject a preset volume of antioxidant from the antioxidant storage bottle into the culture container.

[0096] It should be noted that the normal fluctuation threshold range mentioned in this application refers to the reasonable fluctuation range of the vesicle release event frequency and vesicle maturity index when the cells are in normal culture state, which is obtained through repeated experiments under normal culture state; the stress-induced hypersecretion state determination signal refers to the electrical signal used to trigger the antioxidant injection program, which is a high-level effective signal; the culture medium micro-injection pump refers to the precision equipment used to accurately deliver a preset volume of antioxidant, which is connected to the antioxidant storage bottle and culture container through pipelines.

[0097] It should also be noted that the preset volume mentioned in this application refers to the fixed injection volume of antioxidants adapted to the cell stress state; the preset volume is set as follows: through cell stress relief pre-experiment, the cell stress state is classified into three levels: mild, moderate and severe. The injection volume of antioxidants that can effectively relieve the corresponding level of stress without damaging the cells is tested. The volume values ​​of each group that are stably effective are recorded. A one-to-one correspondence table between stress level and injection volume is established. The table is pre-stored in the system control unit. After the system identifies the stress level, it automatically retrieves the corresponding volume as the preset volume for this injection.

[0098] In specific implementation, firstly, the calculated vesicle release event frequency and vesicle maturity index within the current time window are retrieved. Simultaneously, the pre-stored normal fluctuation threshold ranges corresponding to each are called. A threshold comparison algorithm is used to compare the vesicle release event frequency with the corresponding normal fluctuation threshold range, and the vesicle maturity index with the corresponding normal fluctuation threshold range, and the comparison results of the two parameters are recorded respectively. Secondly, when it is determined that the vesicle release event frequency exceeds the upper limit of its normal fluctuation threshold range, and at the same time it is determined that the vesicle maturity index is lower than the lower limit of its normal fluctuation threshold range, a stress-induced hypersecretion state determination signal is generated. Finally, based on the stress-induced hypersecretion state determination signal, the culture medium micro-injection pump is turned on, and a preset volume of antioxidant is drawn from the antioxidant storage bottle and injected into the culture container.

[0099] In this embodiment, operation A2: When it is determined that the frequency of the vesicle release event exceeds the upper limit of its normal fluctuation threshold range, and at the same time it is determined that the vesicle maturity index is lower than the lower limit of its normal fluctuation threshold range, reading the culture environment parameters monitored by the biosensor and generating a stress-induced hypersecretion state determination signal can be achieved by the following steps:

[0100] When it is determined that the frequency of the vesicle release event exceeds the upper limit of its normal fluctuation threshold range, and at the same time it is determined that the vesicle maturity index is lower than the lower limit of its normal fluctuation threshold range, the concentration of reactive oxygen free radicals in the culture environment parameters monitored by the biosensor is read.

[0101] Contour extraction and surface texture analysis were performed on extracellular vesicle images to extract the in-situ morphological features of extracellular vesicles.

[0102] By detecting transient changes in membrane potential at extracellular vesicle release sites, the potential coupling phase characteristics of the extracellular secretion process can be extracted.

[0103] A stress-induced hypersecretion state determination signal is generated based on the concentration of reactive oxygen species, the in-situ morphological characteristics, and the potential coupling phase characteristics.

[0104] It should be noted that the in-situ morphological features described in this application refer to the morphological features of the regularity of the outline and the roughness of the surface texture of extracellular vesicles; the potential coupling phase features refer to the temporal correlation features between changes in cell membrane potential and secretion during the release of extracellular vesicles.

[0105] In specific implementation, firstly, when the frequency of the vesicle release event exceeds the upper limit of its normal fluctuation threshold range, and simultaneously the vesicle maturity index is below the lower limit of its normal fluctuation threshold range, the concentration of reactive oxygen species in the culture environment parameters monitored by the biosensor is read; secondly, vesicle images of extracellular vesicles are acquired, and the contours of the vesicle images are extracted using the Canny edge detection algorithm. Surface texture analysis of the images is performed using the gray-level co-occurrence matrix algorithm, and the obtained contour regularity and surface roughness data are used as the in-situ morphological features of the extracellular vesicles; then, a microelectrode membrane potential detection sensor is applied close to the extracellular vesicles. At the vesicle release site, transient waveforms of cell membrane potential changes are collected. The temporal correlation features between secretion action and potential changes are extracted using a phase analysis algorithm. These temporal correlation features are used as the potential-coupled phase features of the extracellular secretion process. Finally, the concentration of reactive oxygen species, in-situ morphological features, and potential-coupled phase features are input into a pre-trained support vector machine (SVM) classification model. The SVM classification model automatically classifies and identifies the physiological state of the cell. When the SVM classification model determines that the cell is in a state of stress-induced hypersecretion, a stable high-level control signal is generated. This high-level control signal is used as the determination signal for the state of stress-induced hypersecretion.

[0106] It should be noted that the pre-trained support vector machine classification model can be implemented in the following way: sample data of normal cell state and stress-induced hypersecretion state can be collected, and the concentration of reactive oxygen free radicals, in-situ morphological features, and potential coupling phase features can be constructed into a three-dimensional input feature vector. The feature vector is standardized and preprocessed, and the classification labels of normal state and stress state are labeled respectively. A binary classification support vector machine model is constructed using radial basis kernel function, and the model parameters are optimized and cross-validated through sample data.

[0107] In this embodiment, operation A3: based on the stress-induced hypersecretion state determination signal, activating the culture medium micro-injection pump to extract and inject a preset volume of antioxidant from the antioxidant storage bottle into the culture container can be achieved through the following steps:

[0108] Based on the stress hypersecretion state determination signal, the total volume of the culture medium in the current culture container is obtained, and the amount of antioxidant to be injected is calculated based on the preset final concentration of antioxidant.

[0109] A start command is sent to the culture medium micro-injection pump connected to the antioxidant storage bottle, causing the culture medium micro-injection pump to draw and deliver the antioxidant into the culture container at a constant flow rate;

[0110] After delivery, turn off the culture medium micro-injection pump and record the injection time and volume of the antioxidant.

[0111] It should be noted that, in the present application, the total volume of the culture solution refers to the actual volume of the culture solution in the current culture container; the final concentration of the antioxidant refers to the target concentration that the antioxidant needs to reach in the culture container after the antioxidant is injected; the constant flow rate refers to the stable flow rate when the micro-infusion pump for culture solution delivers the antioxidant; the injection duration refers to the full duration from starting the infusion pump to turning off the infusion pump.

[0112] In specific implementation, firstly, after receiving the determination signal of the stress-induced hypersecretion state, the current liquid level data of the culture solution is collected by the liquid level detection sensor matched with the culture container, and the total volume of the culture solution in the current culture container is obtained through conversion in combination with the specification parameters of the culture container. Then, the amount of the antioxidant to be injected is calculated by a volume conversion method according to the preset final concentration of the antioxidant; secondly, a start instruction is sent to the micro-infusion pump for culture solution connected to the antioxidant reservoir, the instruction includes preset constant flow rate parameters and target injection volume, and the micro-infusion pump is controlled to adopt a stepping motor driving mode, extract a corresponding volume of antioxidant from the antioxidant reservoir at a constant flow rate, and smoothly inject the antioxidant into the culture container through a special delivery pipeline; then, when the cumulative flow value of the micro-infusion pump reaches the preset antioxidant injection volume, a delivery completion signal is sent to the control unit, the control unit immediately sends a shutdown instruction to the micro-infusion pump to stop the delivery of the antioxidant, and automatically records the injection duration and the actual injection volume of the antioxidant this time.

[0113] In this embodiment, with reference to Figure 3 , the figure is a schematic flow chart of determining the normal function enhancement state of cells and maintaining the current culture conditions provided by the present application. In this embodiment, operation B: when the frequency of vesicle release events and the vesicle maturity index increase synchronously, determining that cells are in a normal function enhancement state according to the culture environment parameters monitored by the biosensor and maintaining the current culture conditions can be implemented by the following steps:

[0114] obtaining the frequency of vesicle release events and the vesicle maturity index in the current time window, and comparing them with the corresponding normal fluctuation threshold ranges respectively;

[0115] when it is determined that the frequency of vesicle release events exceeds the upper limit of the normal fluctuation threshold range thereof, and it is simultaneously determined that the vesicle maturity index also exceeds the upper limit of the normal fluctuation threshold range thereof, reading the culture environment parameters monitored by the biosensor and generating a determination signal of a normal function enhancement state;

[0116] maintaining the temperature, humidity, carbon dioxide concentration and culture solution perfusion rate of the incubator unchanged according to the determination signal of the normal function enhancement state.

[0117] It should be noted that the normal function enhancement state determination signal mentioned in this application refers to an electrical signal used to trigger the instruction to maintain the current culture conditions, and is a high-level valid signal; the culture conditions (temperature, humidity, carbon dioxide concentration, culture medium perfusion rate) are the optimal parameter combination for normal cell growth, and maintaining them unchanged can ensure the continuous enhancement of cell function.

[0118] In practice, firstly, the calculated vesicle release event frequency and vesicle maturity index within the current time window are retrieved. A threshold comparison algorithm is used to compare the vesicle release event frequency with the corresponding threshold range and the vesicle maturity index with the corresponding threshold range, respectively. Secondly, when it is determined that the vesicle release event frequency exceeds the upper limit of its normal fluctuation threshold range, and simultaneously the vesicle maturity index also exceeds the upper limit of its normal fluctuation threshold range, the culture environment parameters monitored by the biosensor are read, and a normal functional enhancement state determination signal is generated. Next, the generated normal functional enhancement state determination signal is transmitted to the control unit of the incubator and culture medium perfusion system. After receiving the signal, the control unit locks the current temperature, humidity, and carbon dioxide concentration settings of the incubator, and simultaneously locks the speed of the culture medium perfusion pump. The temperature sensor, humidity sensor, and carbon dioxide sensor monitor the parameters in the incubator in real time, and the flow sensor monitors the culture medium perfusion rate. If fluctuations occur, the control unit automatically fine-tunes to the current set value to keep all culture conditions unchanged.

[0119] In this embodiment, when it is determined that the frequency of the vesicle release event exceeds the upper limit of its normal fluctuation threshold range, and at the same time it is determined that the vesicle maturity index also exceeds the upper limit of its normal fluctuation threshold range, reading the culture environment parameters monitored by the biosensor and generating a normal functional enhancement state determination signal can be achieved by the following steps:

[0120] When it is determined that the frequency of the vesicle release event exceeds the upper limit of its normal fluctuation threshold range, and at the same time it is determined that the vesicle maturity index also exceeds the upper limit of its normal fluctuation threshold range, the glucose concentration and lactate concentration in the culture environment parameters monitored by the biosensor are read.

[0121] The Raman scattering spectra of extracellular vesicles were obtained, and the regularity of the characteristic peaks of the Raman scattering spectra of extracellular vesicles was extracted from the Raman scattering spectra.

[0122] The rate of change of glucose concentration and the rate of change of lactate concentration are detected, and the characteristics of cellular physiological metabolic activity are determined based on the rate of change of glucose concentration and the rate of change of lactate concentration.

[0123] A normal function enhancement state determination signal is generated based on the characteristic peak regularity and the cell physiological metabolic activity characteristics.

[0124] It should be noted that the characteristic peak regularity mentioned in this application represents a parameter reflecting the degree of regularity of the extracellular vesicle molecular structure; the cell physiological metabolic activity characteristics represent characteristics reflecting the cell's metabolic capacity; and the normal function enhancement state determination signal is a high-level valid signal used to trigger subsequent instructions to maintain the current culture conditions.

[0125] In practice, firstly, after determining that both the vesicle release event frequency and vesicle maturity index exceed their respective normal fluctuation thresholds, the detection probe is focused on the region where the extracellular vesicles are located, and Raman scattering spectra of the extracellular vesicles are collected. Background noise in the spectrum is filtered out to obtain clear Raman scattering spectra. Secondly, the obtained Raman scattering spectra are analyzed using a standardized Raman scattering spectral processing procedure. Fluorescence background interference is eliminated through an adaptive iterative baseline correction algorithm. Subsequently, a local maximum peak-finding algorithm is used to locate the coordinates of the characteristic peaks corresponding to the extracellular vesicles. The left and right half-peak width symmetry coefficients and the peak intensity variation coefficients of each characteristic peak are calculated. According to the preset feature weight allocation rules, the left and right half-peak width symmetry coefficients and peak intensity variation coefficients of each characteristic peak are scored independently. Finally, the single-valued peaks of all characteristic peaks are calculated. The scores were weighted and averaged, and the parameters obtained from the weighted average calculation were used as the characteristic peak regularity of the Raman scattering spectrum of extracellular vesicles. Next, glucose concentration and lactate concentration data at three consecutive time points were selected, and the rate of decrease of glucose concentration and the rate of increase of lactate concentration were calculated respectively. The two rate values ​​were directly combined as the characteristic of cell physiological metabolic activity. Finally, the obtained characteristic peak regularity of Raman scattering spectrum and the characteristic of cell physiological metabolic activity were input into a pre-trained support vector machine (SVM) binary classification model. The SVM binary classification model completed the automatic classification and identification of cell functional state. When the SVM binary classification model determined that the current cell state belonged to the normal functional enhancement category, a stable high-level enable signal was generated, and this high-level enable signal was used as the normal functional enhancement state determination signal.

[0126] It should be noted that the support vector machine binary classification model for cell state recognition is trained in the following way: sample data of cells in normal and enhanced functional states are collected; the regularity of the characteristic peaks of Raman scattering spectra and the characteristics of cell physiological and metabolic activity are constructed into two-dimensional input feature vectors; the feature vectors are standardized and preprocessed; corresponding classification labels are labeled; and radial basis kernel functions are selected to complete model training and parameter optimization.

[0127] Therefore, this application demonstrates that the cell culture status can be determined and automatically controlled based on the vesicle maturity index, the vesicle release event frequency, and the culture environment parameters monitored by the biosensor. Firstly, a biosensor is installed inside the culture container. A laser beam is highly focused onto a preset capture point within the culture container using an objective lens to form a three-dimensional gradient force trap. The field of view is monitored in real time, and vesicles are captured and stably bound to the focal point as they diffuse. This allows for in-situ, non-contact dynamic capture of extracellular vesicles without disrupting the culture environment or interfering with normal cell culture, overcoming the limitations of offline detection methods that require intervention in the culture process. Secondly, excitation light is focused onto the area where the captured vesicles are located, and Raman scattering spectra are collected. After baseline correction and noise reduction, the lipid phase transition temperature and nucleic acid characteristic peak intensity are analyzed. This enables real-time, non-destructive, and rapid analysis of the core physicochemical characteristics of vesicles, effectively addressing the technical pain points of existing detection methods, which are complex, time-consuming, and difficult to rapidly capture dynamic changes in vesicle characteristics. Furthermore, the total number of vesicles within a preset time window is counted to obtain the vesicle release event frequency, which is then combined with the lipid phase transition temperature deviation and nucleic acid... By employing fuzzy processing and fuzzy inference to quantify peak intensity deviation, the dynamic release behavior and intrinsic maturity characteristics of vesicles are transformed into quantifiable vesicle maturity indicators. This establishes a precise quantitative correlation system between vesicle characteristics and cell physiological state, overcoming the bottleneck of existing technologies that cannot accurately bind vesicle characteristics to actual cell state. Furthermore, when the frequency of vesicle release events increases but the maturity indicator decreases, a stress-induced hypersecretion state judgment signal is generated through biosensor-monitored culture environment parameters and multi-dimensional feature verification, and antioxidants are precisely injected. When both indicators increase simultaneously, a normal function enhancement state judgment signal is generated and culture conditions are maintained, achieving precise cell state judgment and targeted automatic regulation based on in-situ dynamic markers of vesicles. Finally, through the entire process of in-situ capture, real-time analysis, quantitative characterization, and closed-loop regulation, a complete technical process for in-situ dynamic tracking of extracellular vesicles and precise regulation of cell state is constructed. This makes extracellular vesicles truly precise markers reflecting cell physiological state and secretory function, providing real-time and reliable state feedback for the precise control of automated cell culture, and significantly improving the scientific rigor and targeting of cell culture regulation.

[0128] In summary, the technical solution adopted in this application can realize in-situ dynamic tracking of extracellular vesicles, thereby providing precise markers for cell state regulation.

[0129] Example 2: This application provides an automatic control system for cell culture processes, referring to... Figure 4 As shown in the figure, this is a modular structure diagram of an automatic control system for cell culture process according to this embodiment of the present application. The automatic control system includes:

[0130] The extracellular vesicle capture module 100 is used to form a non-contact optical trap inside the culture container by focusing a laser beam, and to capture extracellular vesicles suspended in the culture medium using the optical trap.

[0131] Extracellular vesicle feature extraction module 200 is used to apply excitation light to the captured extracellular vesicles and collect their Raman scattering spectra, and to analyze the lipid phase transition temperature and nucleic acid characteristic peak intensity of the extracellular vesicles from the characteristic peaks of the Raman scattering spectra.

[0132] The vesicle maturity analysis module 300 is used to count the arrival frequency of captured extracellular vesicles within a preset time window as the vesicle release event frequency, and to determine the vesicle maturity index based on the lipid phase transition temperature and nucleic acid characteristic peak intensity of extracellular vesicles within the preset time window.

[0133] The culture status analysis and automatic control module 400 determines the cell culture status and performs automatic control based on the vesicle maturity index, the frequency of vesicle release events, and the culture environment parameters monitored by the biosensor.

[0134] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0135] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compactdisc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0136] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

Claims

1. An automated control method for cell culture process, characterized in that, The automatic control method includes the following steps: Step S1: Set up a biosensor inside the culture container and form a non-contact optical trap inside the culture container by focusing a laser beam to capture extracellular vesicles suspended in the culture medium. Step S2: Apply excitation light to the captured extracellular vesicles and collect their Raman scattering spectra. Analyze the lipid phase transition temperature and nucleic acid characteristic peak intensity of the extracellular vesicles from the characteristic peaks of the Raman scattering spectra. Step S3: Statistically determine the arrival frequency of captured extracellular vesicles within the preset time window as the vesicle release event frequency, and determine the vesicle maturity index based on the lipid phase transition temperature and nucleic acid characteristic peak intensity of extracellular vesicles within the preset time window. Step S4: Determine the cell culture status and automatically adjust it based on the vesicle maturity index, the frequency of vesicle release events, and the culture environment parameters monitored by the biosensor.

2. The automatic control method for cell culture process as described in claim 1, characterized in that, Step S1 specifically includes: The laser beam is highly focused onto a preset capture point inside the culture container using an objective lens to form a three-dimensional gradient force-optical trap. The field of view near the preset capture point is monitored in real time. When extracellular vesicles are detected to diffuse past the preset capture point, the three-dimensional gradient force light trap captures and stably binds them to the center of the focal point, thus obtaining the captured extracellular vesicles.

3. The automatic control method for cell culture process as described in claim 1, characterized in that, Step S2 specifically includes: Excitation light is focused onto the region where the captured extracellular vesicles are located, and Raman scattering spectra generated after vesicle molecules are scattered are collected. The Raman scattering spectrum was baseline corrected and noise-reduced, and the peak position shift of lipid-related characteristic peaks was extracted and analyzed. The lipid phase transition temperature of extracellular vesicles was then determined from the peak position shift. The peak areas of nucleic acid-related characteristic peaks were extracted and analyzed, and their values ​​were used as the intensity of nucleic acid characteristic peaks of extracellular vesicles.

4. The automatic control method for cell culture process as described in claim 1, characterized in that, Step S3 specifically includes: Record the total number of vesicles successfully captured and spectral measurements completed within a preset time window, and determine the vesicle release event frequency based on the total number of vesicles and the duration of the preset time window; The deviation of the lipid phase transition temperature of extracellular vesicles is determined based on the lipid phase transition temperature of extracellular vesicles within a preset time window. The deviation of nucleic acid peak intensity of extracellular vesicles is determined based on the intensity of the characteristic nucleic acid peaks of extracellular vesicles within a preset time window; The lipid phase transition temperature deviation and the nucleic acid peak intensity deviation are fuzzified to obtain the lipid phase transition temperature membership value and the nucleic acid peak intensity membership value. Based on a preset fuzzy reasoning rule base, fuzzy reasoning is performed on the membership values ​​of the lipid phase transition temperature and the nucleic acid peak intensity to obtain the vesicle maturity index.

5. The automatic control method for cell culture process as described in claim 1, characterized in that, Step S4 specifically includes: Operation A: When the frequency of vesicle release events increases but the vesicle maturity index decreases, the cell is determined to be in a state of high stress secretion based on the culture environment parameters monitored by the biosensor and the antioxidant injection procedure is initiated. Operation B: When the frequency of vesicle release events increases synchronously with the vesicle maturity index, the cells are determined to be in a state of normal functional enhancement based on the culture environment parameters monitored by the biosensor, and the current culture conditions are maintained.

6. The automatic control method for cell culture process as described in claim 5, characterized in that, Operation A specifically includes: Operation A1: Obtain the frequency of vesicle release events and the vesicle maturity index for the current time window, and compare them with their respective normal fluctuation threshold ranges; Operation A2: When it is determined that the frequency of the vesicle release event exceeds the upper limit of its normal fluctuation threshold range, and at the same time it is determined that the vesicle maturity index is lower than the lower limit of its normal fluctuation threshold range, the culture environment parameters monitored by the biosensor are read, and a stress hypersecretion state determination signal is generated. Operation A3: Based on the stress hypersecretion state determination signal, turn on the culture medium micro-injection pump to extract and inject a preset volume of antioxidant from the antioxidant storage bottle into the culture container.

7. The automatic control method for cell culture process as described in claim 6, characterized in that, Operation A2 specifically includes: When it is determined that the frequency of the vesicle release event exceeds the upper limit of its normal fluctuation threshold range, and at the same time it is determined that the vesicle maturity index is lower than the lower limit of its normal fluctuation threshold range, the concentration of reactive oxygen free radicals in the culture environment parameters monitored by the biosensor is read. Contour extraction and surface texture analysis were performed on extracellular vesicle images to extract the in-situ morphological features of extracellular vesicles. By detecting transient changes in membrane potential at extracellular vesicle release sites, the potential coupling phase characteristics of the extracellular secretion process can be extracted. A stress-induced hypersecretion state determination signal is generated based on the concentration of reactive oxygen species, the in-situ morphological characteristics, and the potential coupling phase characteristics.

8. The automatic control method for cell culture process as described in claim 6, characterized in that, Operation A3 specifically includes: Based on the stress hypersecretion state determination signal, the total volume of the culture medium in the current culture container is obtained, and the amount of antioxidant to be injected is calculated based on the preset final concentration of antioxidant. A start command is sent to the culture medium micro-injection pump connected to the antioxidant storage bottle, causing the culture medium micro-injection pump to draw and deliver the antioxidant into the culture container at a constant flow rate; After delivery, turn off the culture medium micro-injection pump and record the injection time and volume of the antioxidant.

9. The automatic control method for cell culture process as described in claim 5, characterized in that, Operation B specifically includes: Obtain the frequency of vesicle release events and the vesicle maturity index for the current time window, and compare them with their respective normal fluctuation threshold ranges. When it is determined that the frequency of the vesicle release event exceeds the upper limit of its normal fluctuation threshold range, and at the same time it is determined that the vesicle maturity index also exceeds the upper limit of its normal fluctuation threshold range, the culture environment parameters monitored by the biosensor are read, and a normal functional enhancement state determination signal is generated. Based on the normal function enhancement state determination signal, the temperature, humidity, carbon dioxide concentration, and culture medium perfusion rate of the incubator are kept constant.

10. An automated control system for cell culture processes, used to execute an automated control method for cell culture processes as described in any one of claims 1 to 9, characterized in that, Automatic control systems include: An extracellular vesicle capture module is used to set up a biosensor inside a culture container. A non-contact optical trap is formed inside the culture container by a focused laser beam, and the optical trap is used to capture extracellular vesicles suspended in the culture medium. An extracellular vesicle feature extraction module is used to apply excitation light to the captured extracellular vesicles and collect their Raman scattering spectra, and to analyze the lipid phase transition temperature and nucleic acid characteristic peak intensity of the extracellular vesicles from the characteristic peaks of the Raman scattering spectra. The vesicle maturity analysis module is used to count the arrival frequency of captured extracellular vesicles within a preset time window as the vesicle release event frequency, and to determine the vesicle maturity index based on the lipid phase transition temperature and nucleic acid characteristic peak intensity of extracellular vesicles within the preset time window. The culture status analysis and automatic control module determines the cell culture status and performs automatic control based on the vesicle maturity index, the frequency of vesicle release events, and the culture environment parameters monitored by the biosensor.

Images