Cell cell real-time feedback analysis system based on multi-parameter fusion weighting

Abstract

The invention relates to the field of cell analysis, in particular to a cell chamber real-time feedback analysis system based on multi-parameter fusion weighting. Comprising a multi-parameter acquisition module for acquiring cell growth related parameter information in a cell chamber; an artificial intelligence algorithm model is built in the weight calculation module, and the weights of the multiple pieces of cell growth related parameter information are dynamically adjusted through machine learning; the parameter fusion module is used for fusing the multiple pieces of cell growth related parameter information according to the weight calculated by the weight calculation module; the real-time feedback module is used for generating feedback information in real time according to the fusion information; the visual display module is used for carrying out visual display on the feedback information; and the adaptive learning module is used for collecting experimental result information in the cell culture process, so that the weight calculation module optimizes the artificial intelligence algorithm model. According to the method, related parameter information of growth of each cell is dynamically adjusted according to the actual condition of cell culture, and fusion is carried out according to the weight, so that the requirements of modern cell research and application are met.

Description

Technical Field

[0001] The present invention relates to the field of cell analysis, and particularly to an analysis system for real-time feedback of cell chambers based on multi-parameter fusion weighting. Background Art

[0002] In the field of cell culture and analysis, with the in-depth research of life sciences and the rapid development of industries such as biopharmaceuticals, it has become crucial to accurately monitor and analyze the growth state of cells. During the cell culture process, the growth, metabolism, and differentiation of cells are comprehensively affected by various parameters such as temperature, pH value, oxygen concentration, carbon dioxide concentration, nutrient content, and the density of the cells themselves.

[0003] Currently, most cell analysis systems adopt a single-parameter monitoring method, which can only show the changes of a certain parameter in isolation and cannot reflect the combined effect of various parameters on cell growth. Even if some systems can collect multiple parameters, they often use simple data listing or fixed-weight fusion methods in data processing. This processing method has obvious defects. On the one hand, the fixed weight cannot adapt to the characteristics of the dynamic changes in the influence degree of each parameter at different cell culture stages. On the other hand, there is a lack of real-time analysis and effective feedback mechanism for parameter changes, making it difficult to give early warnings in time when the cell culture environment is abnormal, resulting in researchers being unable to quickly adjust the culture conditions, affecting the success rate of cell culture experiments and the accuracy of experimental results.

[0004] Therefore, there is an urgent need for a system that can effectively fuse and analyze multiple parameters in a cell chamber, flexibly adjust the parameter weights according to the actual situation of cell culture, and achieve real-time, accurate visualization analysis and feedback to meet the needs of modern cell research and applications. Summary of the Invention

[0005] In order to overcome the technical defect in the above-mentioned prior art that multiple parameters in a cell chamber cannot be effectively fused and analyzed, the purpose of the present invention is to provide an analysis system for real-time feedback of cell chambers based on multi-parameter fusion weighting, which dynamically adjusts the weights of each cell growth-related parameter information according to the actual situation of cell culture and performs fusion according to the weights to meet the needs of modern cell research and applications.

[0006] The present invention discloses an analysis system for real-time feedback of cell chambers based on multi-parameter fusion weighting, including: A multi-parameter acquisition module: used to collect multiple cell growth-related parameter information in the cell chamber based on a sensor system; A weight calculation module: connected to the multi-parameter acquisition module and built-in with an artificial intelligence algorithm model to dynamically adjust the weights of multiple cell growth-related parameter information through machine learning according to different stages of cell culture, preset rules, and cell growth-related parameter information; Parameter fusion module: Connected to the weight calculation module, it is used to fuse multiple cell growth-related parameter information according to the weights calculated by the weight calculation module to obtain fusion information; Real-time feedback module: Connected to the parameter fusion module, it is used to generate feedback information in real time according to the fusion information; Visualization display module: Connected to the real-time feedback module, it is used to visually display the feedback information; Adaptive learning module: Connected to the weight calculation module, it is used to collect experimental result information during the cell culture process and feedback the experimental result information to the weight calculation module in real time, so that the weight calculation module optimizes the artificial intelligence algorithm model according to the experimental result information.

[0007] Preferably, the sensor system includes at least two of a temperature sensor, a pH sensor, an oxygen concentration sensor, a carbon dioxide concentration sensor, and a cell density sensor.

[0008] Preferably, the temperature sensor is a thermocouple temperature sensor, and the pH sensor is a glass electrode pH sensor.

[0009] Preferably, the weight calculation module calculates the weights of multiple cell growth-related parameter information based on the following calculation formula: Where, represents the weight of the i-th parameter, represents the influence factor of the i-th parameter in the current cell culture stage, and n represents the total number of cell growth-related parameter information collected, represents the influence factor of the j-th parameter in the current cell culture stage.

[0010] Preferably, the influence factor is dynamically set according to the historical data of cell culture and the cell growth-related parameter information.

[0011] Preferably, the parameter fusion module fuses multiple cell growth-related parameter information based on weighted summation, and the calculation formula of the fusion information P is: Where, represents the weight of the i-th parameter, represents the collected value of the i-th parameter, and n represents the total number of the collected cell growth-related parameter information.

[0012] Preferably, when the parameter fusion module calculates the fusion information P, a time decay factor is introduced, and the calculation formula of the fusion information P after introducing the decay factor is: Among them, , t represents the duration value of the current cell culture, represents a preset key time point, and e represents the natural constant.

[0013] Preferably, before performing weighted summation, the parameter fusion module preprocesses multiple cell growth-related parameter information.

[0014] Preferably, the real-time feedback module compares the value of the fusion information with a preset interval range threshold to generate real-time feedback information; The interval range threshold is set according to the cell type and culture stage.

[0015] Preferably, when generating real-time feedback information, the real-time feedback module performs quantitative evaluation based on the risk assessment index value R, and the calculation formula is: Among them, is the value of the k-th parameter after fusion, represents the middle value of the preset interval range threshold of the k-th parameter, m represents the number of parameters participating in the evaluation, represents the risk weight coefficient of the k-th parameter, and ; Compare the risk assessment index value R with a preset risk threshold; to generate feedback information according to the comparison result.

[0016] After adopting the above technical solutions, compared with the prior art, the present invention dynamically adjusts the weight of each cell growth-related parameter information according to the actual situation of cell culture, and performs fusion according to the weight to meet the needs of modern cell research and application; collects multiple cell growth-related parameter information through a multi-parameter acquisition module, combines a weight calculation module to dynamically adjust the weight, so that the parameter fusion module scientifically fuses cell growth-related parameter information. Compared with the traditional fixed weight method, it realizes the intelligent and effective fusion between multiple cell growth-related parameter information, and more accurately reflects the generation state of cells; an artificial intelligence algorithm model is built in, and according to the cell culture stage, preset rules and cell growth-related parameter information, the weight is dynamically adjusted through machine learning, and the artificial intelligence algorithm model is optimized through the experimental result information collected by the adaptive learning module, realizing weight adaptive learning, and solving the problem that the traditional fixed weight cannot adapt to the dynamic changes of cell culture; the real-time feedback module generates feedback information in real time based on the fusion information, and combines the preset interval range threshold to realize the real-time monitoring of the cell culture state; the visualization display module is used to display the feedback information, which is convenient for researchers to quickly obtain key information. Description of the Drawings

[0017] Figure 1This is a schematic flowchart of the analysis system for real-time feedback of cell chambers based on multi-parameter fusion and weighting in the present invention. Specific embodiments

[0018] The advantages of the present invention will be further elaborated below in conjunction with the accompanying drawings and specific embodiments.

[0019] Exemplary embodiments will be described in detail herein, and examples thereof are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0020] The terms used in the present disclosure are for the purpose of describing specific embodiments only and are not intended to limit the present disclosure. The singular forms "a", "the", and "said" used in the present disclosure and the appended claims are also intended to include the plural forms unless the context clearly dictates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0021] It should be understood that although the terms first, second, third, etc. may be used in the present disclosure to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present disclosure, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "when" or "while" or "in response to determining".

[0022] In the description of the present invention, it should be understood that the orientation or positional relationships indicated by the terms "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. are based on the orientation or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation of the present invention.

[0023] In the description of the present invention, unless otherwise specified and defined, it should be noted that the terms "mounted", "connected", and "connected" should be understood in a broad sense. For example, it may be a mechanical connection or an electrical connection, or it may be the communication inside two elements. It may be directly connected, or indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms may be understood according to specific circumstances.

[0024] In the following description, suffixes such as "module", "component", or "unit" used to denote elements are only for facilitating the description of the present invention and have no specific meaning in themselves. Therefore, "module" and "component" can be used interchangeably.

[0025] This embodiment provides an analysis system for real-time feedback of cell chambers based on multi-parameter fusion weighting, including: a multi-parameter acquisition module: used to collect multiple cell growth-related parameter information in the cell chamber based on a sensor system; a weight calculation module: connected to the multi-parameter acquisition module and built with an artificial intelligence algorithm model to dynamically adjust the weights of multiple cell growth-related parameter information through machine learning according to different stages of cell culture, preset rules, and cell growth-related parameter information; a parameter fusion module: connected to the weight calculation module, used to fuse multiple cell growth-related parameter information according to the weights calculated by the weight calculation module to obtain fusion information; a real-time feedback module: connected to the parameter fusion module, used to generate feedback information in real time according to the fusion information; a visualization display module: connected to the real-time feedback module, used to visually display the feedback information; an adaptive learning module: connected to the weight calculation module, used to collect experimental result information during the cell culture process and feedback the experimental result information to the weight calculation module in real time, so that the weight calculation module optimizes the artificial intelligence algorithm model according to the experimental result information.

[0026] Refer to Figure 1 As shown, the analysis system will be described in detail in this embodiment as follows: The multi-parameter acquisition module collects multiple cell growth-related parameter information in the cell chamber through the sensor system. The sensor system includes at least two of, but is not limited to, a temperature sensor, a pH sensor, an oxygen concentration sensor, a carbon dioxide concentration sensor, and a cell density sensor. Each sensor in the sensor system works in cooperation to comprehensively collect various cell growth-related parameter information in the cell chamber. For example, in a basic cell culture monitoring scenario, a temperature sensor, a pH sensor, and a cell density sensor can be selected; while in a cell culture experiment sensitive to the gas environment, an oxygen concentration sensor and a carbon dioxide concentration sensor can be added to meet diverse experimental needs. In this embodiment, the working principles of the temperature sensor, the pH sensor, and the cell density sensor among them will be explained in detail. The temperature sensor uses a thermocouple temperature sensor, which, based on the Seebeck effect, converts temperature changes into thermoelectric potential signals. The pH sensor selects a glass electrode pH sensor, which uses the principle of ion-selective electrodes to measure the pH value of a solution by generating a potential difference due to the hydrogen ion concentration difference on both sides of the glass membrane. The cell density sensor can be classified into sensors based on the impedance principle or the optical principle according to its working principle. The impedance-based cell density sensor reflects changes in cell quantity and size by detecting changes in the impedance of the cell suspension; the optical cell density sensor uses the absorption, scattering, or fluorescence characteristics of light to accurately measure cell concentration and morphological changes, providing quantitative data for evaluating the cell growth state.

[0027] The sensor system continuously performs real-time detection on the inside of the cell chamber, collecting cell growth-related parameter information at a high frequency (such as 1 - 10 times per second) to ensure that subtle changes in the parameters are captured. The analog signals (such as voltage and current signals) collected by the sensor system are converted into digital signals by a high-precision A / D converter (resolution ≥ 16 bits) for easy processing and transmission. At the same time, digital filtering algorithms (such as Kalman filtering and median filtering) are used to process the converted digital signals to remove abnormal data generated by factors such as environmental interference and sensor noise, thereby improving the accuracy and reliability of the cell growth-related parameter information. In this multi-parameter acquisition module, when transmitting the cell growth-related parameters to the weight calculation module, they are transmitted through a standardized data transmission protocol (such as SPI, I2C, RS-485, or wireless transmission protocols Wi-Fi and Bluetooth). It should be noted that this standardized transmission method is also applicable to the transmission between other modules in this embodiment.

[0028] The weight calculation module is connected to the multi-parameter acquisition module to receive multiple cell growth-related parameter information collected by the multi-parameter acquisition module, and an artificial intelligence algorithm model is built into the weight calculation module to dynamically adjust the weights of the multiple cell growth-related parameter information through machine learning according to different stages of the cell culture phase, preset rules, and the cell growth-related parameter information.

[0029] For different cell culture stages, such as the initial stage of culture, logarithmic growth phase, stationary phase, etc., the weights of multiple cell growth-related parameter information are automatically adjusted. Therefore, in this embodiment, the weights of multiple cell growth-related parameter information can be adjusted according to the actual culture stage of cell culture. Therefore, when culturing a certain cell, the weights of multiple cell growth-related parameter information of the cell in different culture stages are different. For example, in the initial stage of culture, the weights of temperature and pH value are relatively high (affecting cell adhesion and survival), and in the logarithmic growth phase, the weights of nutrient concentration and oxygen are increased (supporting rapid proliferation). For this preset rule, based on the cell type (such as stem cells, tumor cells) and experimental purpose, a basic weight strategy is set, and the basic weight strategy refers to the basic weights of multiple cell growth-related parameter information of a certain cell in different culture stages. Therefore, in this embodiment, the correlation between the fluctuations of cell growth-related parameter information and the cell state is analyzed by machine learning to optimize the weights in real time so as to adjust according to the actual situation.

[0030] In addition, it should be noted that in this weight calculation module, the experimental result information (such as cell viability, metabolite concentration) fed back by the following adaptive learning module will also be combined to continuously optimize the artificial intelligence algorithm model for weight calculation. For example: if it is found that the change of a certain parameter (such as glucose concentration) is highly correlated with the cell apoptosis rate, the system will automatically increase the weight of this parameter. Another example is to form an artificial intelligence algorithm model for personalized weight calculation for a specific cell type through historical data accumulation.

[0031] A parameter fusion module, which is connected to the weight calculation module, is used to fuse multiple cell growth-related parameter information according to the weights calculated by the weight calculation module to obtain fusion information. Specifically, the parameter fusion module deeply processes the cell growth-related parameter information collected by the multi-parameter acquisition module based on the weights output by the weight calculation module, providing accurate data support for the subsequent real-time feedback module and visualization display module. The following will describe it in detail from aspects such as its functions and principles.

[0032] The function of this parameter fusion module is to fuse multiple cell growth-related parameter information collected by the multi-parameter acquisition module according to the weights of each cell growth-related parameter information determined by the weight calculation module, integrate the originally scattered and independent parameters into an index that can comprehensively reflect the cell growth state, eliminate the limitations of single-parameter analysis, and lay a foundation for accurately judging the cell culture state subsequently. This parameter fusion module fuses multiple cell growth-related parameter information based on the method of weighted summation, and the calculation of its fusion information P is: 。

[0033] Among them, denoted as the weight of the \(i\)-th parameter, which is calculated by the weight calculation module according to the cell culture stage, historical data, preset rules, and cell growth-related parameter information, etc. denoted as the acquisition value of the \(i\)-th parameter, which is obtained by real-time acquisition of the multi-parameter acquisition module. \(n\) represents the total number of cell growth-related parameter information collected. This calculation method makes the fused information \(P\) more conform to the actual cell growth state by assigning weights that match the importance of different cell growth-related parameter information. For example, in a specific stage of cell culture, if temperature has a greater impact on cell growth, the weight calculation module assigns a higher weight to the temperature parameter. After weighted summation calculation, the temperature parameter occupies a larger proportion in the fused parameter \(P\) after fusion, thus highlighting its impact on the cell growth state.

[0034] It should be noted that to avoid interference of different cell growth-related parameter information on the fused information \(P\) due to dimensional differences (such as the temperature unit is \(^{\circ}C\), the pH value is dimensionless, and the nutrient concentration unit is diverse), before weighted fusion, the parameter fusion module will perform normalization processing on each cell growth-related parameter information, mapping the values of each cell growth-related parameter information to a unified interval (such as the interval \([0, 1]\)), so that different cell growth-related parameter information is on the same measurement scale, ensuring the scientificity and accuracy of the fusion calculation. Taking the temperature parameter (range assumed to be \(30^{\circ}C - 40^{\circ}C\)), pH value parameter (range \(0 - 14\)), and nutrient concentration parameter (assumed range is \(0mg / L - 100mg / L\)) as examples, after normalization processing, the two have the same mathematical basis in the fusion calculation, avoiding a certain parameter having a different reasonable dominant effect on the fused information \(P\) due to different dimensions. Specifically, for example, the collected temperature parameter is \(37.2^{\circ}C\), the pH value parameter is \(7.3\), and the nutrient concentration parameter is \(80mg / L\) for the three parameters. Then the normalized temperature value is equal to , the normalized pH value is equal to , and the normalized nutrient concentration value is equal to .

[0035] For example, in a cell culture experiment, assume that the multi-parameter acquisition module collects the temperature parameter is \(37.2^{\circ}C\), the pH value parameter is \(7.3\), and the nutrient concentration parameter is \(80mg / L\) for the three parameters; the weight calculation module calculates the temperature weight \( = 0.4\), the pH value weight \( = 0.3\), and the nutrient concentration weight = 0.3. The parameter fusion module first normalizes these three parameters, maps the temperature, pH value, and nutrient concentration to the interval [0, 1] respectively to obtain the normalized values, and then calculates the fusion information P = 0.4×(0.72)+0.3×(0.52)+0.3×(0.8) according to the weighted summation formula, so as to obtain the value of the fusion information P as 0.684. After this calculation, it reflects the influence of temperature, pH value, and nutrient concentration on the cell growth state in the current cell culture environment, and can be used to judge whether the cell environment is suitable and whether the culture conditions need to be adjusted in the follow-up.

[0036] The real-time feedback module, connected to the parameter fusion module, is used to receive the fusion information and generate real-time feedback information according to the fusion information. The real-time feedback module generates real-time feedback information by comparing the value of the fusion information P with the preset interval range threshold. For example, the interval range threshold of the temperature parameter is set to 36.5°C - 37.5°C. When the value of the temperature parameter in the fusion information P exceeds the maximum value of this interval range threshold, the real-time feedback module generates a feedback message of "The current temperature is abnormal and needs to be adjusted in time".

[0037] The visualization display module, connected to the real-time feedback module, is used to receive and visually display the feedback information. For example, when the real-time feedback module generates a feedback message of "The current temperature is abnormal and needs to be adjusted in time", the visualization display module displays this feedback message of "The current temperature is abnormal and needs to be adjusted in time" so that the experimental personnel can make corresponding adjustments quickly.

[0038] The adaptive learning module is connected to the weight calculation module, and is used to collect experimental result information during the cell culture process, and feed back the experimental result information to the weight calculation module in real time, so that the weight calculation module optimizes the artificial intelligence algorithm model according to the experimental result information. When the adaptive learning module collects experimental result information during the cell culture process, it includes but is not limited to obtaining experimental result information through sensor data, test detection data, and image analysis data. Sensor data is real-time monitoring data directly obtained from cell density sensors, metabolite sensors, and metabolite sensors (such as lactic acid and amino acid concentration sensors). Experimental detection data is non-real-time data such as manually entered cell activity detection results (such as MTT experimental data) and sequencing reports. Image analysis data is to analyze the morphological images of cells taken by a microscope and extract characteristic parameters such as cell size, morphology, and degree of differentiation. Statistics are used in the adaptive module to eliminate abnormal data (such as outliers caused by sensor failures), and normalize data of different dimensions to the [0,1] interval to ensure data quality. Machine learning algorithms (such as correlation analysis and regression analysis) are used to explore the potential relationship between experimental results and the weights of each parameter. For example, through analysis, it was found that when the glucose concentration increased by 20, the cell proliferation rate increased significantly. For another example, by setting an optimization threshold (such as a cell survival rate fluctuation exceeding 10%), when an abnormality in a key indicator is detected, an optimization instruction is automatically generated.

[0039] For ease of understanding, we will take the collected temperature parameters of 37.2°C, pH value of 7.3, and oxygen concentration of 5% as an example. At this time, the manually entered cell differentiation rate is 35% (the test result on the seventh day). Through the adaptive learning module, it was found that when the oxygen concentration weight increased from 0.2 to 0.3, the cell differentiation rate increased significantly. The automatic generation of instructions adjusted the oxygen concentration influence factor from 0.3 to 0.4, and the temperature influence factor from 0.4 to 0.35. After the weight calculation module was updated, the cell differentiation rate in the next batch of experiments increased to 42%, thus verifying the effect of this optimization.

[0040] Furthermore, the weight calculation module calculates the weights of multiple cell growth related parameter information based on the following calculation formula: .in, is represented as the weight of the i-th parameter, It represents the influencing factor of the i-th parameter in the current cell culture stage, and n represents the total number of cell growth-related parameter information collected. It is expressed as the influencing factor of the jth parameter in the current cell culture stage.

[0041] In this embodiment, the weight calculation module will be described in detail. When the weight calculation module calculates the weight of the cell growth related parameter information, the calculation formula Calculate. Denoted as the weight of the \(i\)-th parameter, it is a calculated value with a range from 0 to 1, reflecting the relative importance of the \(i\)-th parameter among all the cell growth-related parameter information collected. As mentioned above, for the three parameters of temperature, pH value, and nutrient concentration, represents the weight of the temperature parameter, denotes the weight of the pH value parameter, represents the weight of the nutrient concentration parameter, and + + = 1. Denoted as the influence factor of the \(i\)-th parameter in the current cell culture stage, this factor is a quantitative indicator measuring the importance of the parameter, and its setting is dynamically determined based on the historical data of cell culture and the cell growth-related parameter information. For example, in the initial stage of cell culture, according to the historical data and the cell growth-related parameter information, temperature has a greater impact on cell survival, and the influence factor of the temperature parameter can be set to 0.4. \(n\) represents the total number of cell growth-related parameter information collected. Denoted as the influence factor of the \(j\)-th parameter in the current cell culture stage, it reflects the biological relevance of the parameter in a specific culture stage. For example, in the initial stage of cell culture, temperature may be crucial for cell adhesion and survival, so a relatively high influence factor (such as 0.4) is given; in the logarithmic growth phase, the nutrient concentration parameter may become a key limiting factor, and the influence factor may be adjusted to 0.5. It should be noted that it is not fixed but is dynamically adjusted according to the historical data and the real-time cell growth-related parameter information.

[0042] The calculation logic in this embodiment is as follows: When calculating the weight of a certain parameter, first determine the influence factor of this parameter, then add up the influence factors of all cell growth-related parameters to obtain , and finally divide by to obtain the weight of the \(i\)-th parameter. For example, when collecting three parameters of temperature, pH value, and nutrient concentration, and their influence factors are respectively = 0.4, = 0.3, = 0.3, first calculate = 0.4 + 0.3 + 0.3 = 1, and then calculate the weights respectively, then = 0.4, = 0.3, = 0.3.

[0043] Further, the real-time feedback module compares the value of the fusion information with a preset range threshold to generate real-time feedback information; the range threshold is set according to the cell type and the culture stage.

[0044] In this embodiment, the real-time feedback module will be described in detail. The real-time feedback module compares the value of the fusion information with a preset range threshold to generate feedback information. Refer to the above embodiment for description.

[0045] The range threshold is set according to the cell type and the culture stage. Different types of cells have different physiological characteristics and requirements for the culture environment, and the suitable ranges of parameters such as growth temperature, pH value, and oxygen concentration are different. For example, the suitable growth temperature of mammalian cells is usually between 36.5°C and 37.5°C, and the pH value is between 7.2 and 7.4; while the suitable temperature and pH value ranges of plant cells are different. At the same time, the requirements of the same cell for environmental parameters also change at different culture stages. In the initial stage of cell culture, during the cell adaptation stage, higher stability requirements are placed on temperature and pH value parameters; after the logarithmic growth phase, the cells are metabolically active and are more sensitive to parameters such as nutrient concentration and oxygen supply. Therefore, the preset range threshold needs to be set according to the cell type and the culture stage to ensure the accuracy of the system in judging the cell culture state.

[0046] Further, when generating real-time feedback information, the real-time feedback module performs a quantitative evaluation based on the risk assessment index value R, and the calculation formula is: . Wherein, is the value of the k-th parameter after fusion, represents the middle value of the preset range threshold of the k-th parameter, m represents the number of parameters participating in the evaluation, represents the risk weight coefficient of the k-th parameter, and ; compare the risk assessment index value R with a preset risk threshold; and generate feedback information according to the comparison result.

[0047] In this embodiment, the real-time feedback module will be described in detail again. The real-time feedback module will perform a quantitative evaluation based on the risk assessment index value R, and the calculation formula of the risk assessment index value R is: Wherein, is the value of the k-th parameter after fusion, represents the middle value of the preset range threshold of the k-th parameter, m represents the number of parameters participating in the evaluation, represents the risk weight coefficient of the k-th parameter, and .

[0048] Compare the risk assessment index value R with a preset risk threshold. When the risk assessment index value R is greater than or equal to the preset risk threshold, the real-time feedback module generates a high-risk warning feedback message, indicating a relatively high risk. When the risk assessment index value R is less than the preset risk threshold, the real-time feedback module generates a risk-stable feedback message, indicating no risk.

[0049] Example of the calculation steps of the risk assessment index value R: Assume that the system monitors three parameters: temperature, pH value, and glucose concentration. = 37.8 °C, = 6.85, = 2.1 g / L. The central value of the preset threshold = 37 °C, = 7.2, = 3.75 g / L. Risk weight coefficient = 0.5, = 0.3, = 0.2. Calculate the single-risk contribution, temperature = 0.5 × , pH value = 0.3 × , glucose = 0.22 × , then R = + + = 0.134.

[0050] By setting the risk threshold and comparing it with the risk assessment index value R, different levels of warnings are triggered. For example, when the risk assessment index value R is less than or equal to the risk threshold, it is in a normal state and no intervention is required. When the risk assessment index value R is greater than the risk threshold and less than twice the risk threshold, it is in a warning state. However, when the risk assessment index value R is greater than or equal to twice the risk threshold, the culture conditions need to be adjusted immediately.

[0051] In the artificial intelligence algorithm model described in this embodiment, the weights of multiple cell growth-related parameter information are calculated dynamically based on the neural network algorithm. Specifically, the neural network algorithm is mainly used to process the non-linear relationship between multiple cell growth-related parameter information and the cell state to calculate the weights. More specifically, through the input layer of the neural network algorithm, parameters such as temperature and pH value are received. After feature transformation in the hidden layer, the weights of each parameter are generated through the output layer. It should be noted that the number of neurons in the input layer is equal to the number of cell growth-related parameter information collected. The hidden layer adopts a multi-layer perceptron architecture, usually including 1 - 2 hidden layers. When processing the non-linear relationship between multiple cell growth-related parameter information and the cell state, the ReLU or Sigmoid function is used to introduce non-linear transformation ability. For example: ReLU(x) = max(0, x). The input layer and the hidden layer are connected through a weight matrix Connection, the output of the hidden layer neurons is: Activation function ( ). Among them, represents the output of the j-th hidden layer neuron; j represents 1, 2,..., m, where m is the connection weight of the j-th neuron in the hidden layer, is the normalized parameter value of the i-th input layer (i = 1, 2,..., n, where n is the number of cell growth related parameter information), represents the bias term of the j-th neuron in the hidden layer. The hidden layer learns the correlation between each cell growth related parameter information (such as the association between temperature and cell metabolic rate).

[0052] Output layer output weight. The number of neurons in the output layer is equal to n. Each neuron corresponds to the weight of a cell growth related parameter information , and . The logic of its weight calculation is that the hidden layer and the output layer are connected through the weight matrix , and the calculation formula of the neurons in the output layer is , where, . Among them, represents the weight of the i-th cell growth related parameter information in the output layer (final result), represents the input value of the i-th neuron in the output layer, represents the connection weight from the k-th neuron in the hidden layer to the i-th neuron in the output layer, represents the output of the k-th neuron in the hidden layer. If the output value of the output layer neuron is [0.45, 0.3, 0.25], then the weights correspond to [0.4, 0.3, 0.3], which are the weights of temperature, pH value, and nutrient concentration respectively.

[0053] It should be noted that the embodiments of the present invention have better implementability and are not any form of limitation to the present invention. Any person skilled in the art may use the disclosed technical content to change or modify it into equivalent effective embodiments. However, as long as it does not depart from the technical solution of the present invention, any modification, equivalent change, or modification made to the above embodiments based on the technical essence of the present invention still falls within the scope of the technical solution of the present invention.

Claims

1. An analysis system for real-time feedback of cell chambers based on multi-parameter fusion weighting, characterized in that, Including: Multi-parameter acquisition module: used to acquire multiple pieces of cell growth-related parameter information in the cell chamber based on a sensor system; Weight calculation module: connected to the multi-parameter acquisition module and built with an artificial intelligence algorithm model to dynamically adjust the weights of multiple pieces of the cell growth-related parameter information through machine learning according to different stages of cell culture, preset rules, and the cell growth-related parameter information; Parameter fusion module: connected to the weight calculation module, used to fuse multiple pieces of the cell growth-related parameter information according to the weights calculated by the weight calculation module to obtain fusion information; Real-time feedback module: connected to the parameter fusion module, used to generate feedback information in real time according to the fusion information; Visualization display module: connected to the real-time feedback module, used to visually display the feedback information; Adaptive learning module: connected to the weight calculation module, used to collect experimental result information during the cell culture process and real-time feedback the experimental result information to the weight calculation module, so that the weight calculation module optimizes the artificial intelligence algorithm model according to the experimental result information.

2. The analysis system for real-time feedback of cell chambers based on multi-parameter fusion and weighting according to claim 1, wherein The sensor system includes at least two of a temperature sensor, a pH sensor, an oxygen concentration sensor, a carbon dioxide concentration sensor, and a cell density sensor.

3. The analysis system for real-time feedback of cell chambers based on multi-parameter fusion and weighting according to claim 2, characterized in that The temperature sensor is a thermocouple temperature sensor, and the pH sensor is a glass electrode pH sensor.

4. The analysis system for real-time feedback of cell chambers based on multi-parameter fusion and weighting according to claim 1, characterized in that, The weight calculation module calculates the weights of multiple pieces of the cell growth-related parameter information based on the following calculation formula: Among them, represents the weight of the i-th parameter, represents the influence factor of the i-th parameter in the current cell culture stage, and n represents the total number of the collected cell growth-related parameter information, represents the influence factor of the j-th parameter in the current cell culture stage.

5. The analysis system for real-time feedback of cell chambers based on multi-parameter fusion weighting according to claim 4, characterized in that, The influence factor It is dynamically set according to the historical data of cell culture and the information on the cell growth-related parameters.

6. The analysis system for real-time feedback of cell chambers based on multi-parameter fusion and weighting according to claim 1, wherein The parameter fusion module fuses multiple pieces of the cell growth-related parameter information based on weighted summation, and the calculation formula for the fusion information P is: Among them, represents the weight of the i-th parameter, represents the acquisition value of the i-th parameter, and n represents the total number of the cell growth-related parameter information acquired.

7. The analysis system for real-time feedback of cell chambers based on multi-parameter fusion weighting according to claim 6, wherein Before performing weighted summation, the parameter fusion module preprocesses multiple pieces of the cell growth-related parameter information.

8. The analysis system for real-time feedback of cell chambers based on multi-parameter fusion weighting according to claim 1, characterized in that The real-time feedback module compares the value of the fusion information with a preset interval range threshold to generate real-time feedback information; The interval range threshold is set according to the type and culture stage of the cells.

9. The analysis system for real-time feedback of cell chambers based on multi-parameter fusion weighting according to claim 8, wherein When generating the real-time feedback information, the real-time feedback module performs quantitative evaluation based on the risk assessment index value R, and the calculation formula is: Among them, is the value of the k-th parameter after fusion, represents the intermediate value of the preset interval range threshold of the k-th parameter, and m represents the number of parameters participating in the evaluation. represents the risk weight coefficient of the k-th parameter, and ; Compare the risk assessment index value R with a preset risk threshold to generate the feedback information according to the comparison result.

Images