CN117421691B - Cell culture monitoring method and system based on artificial intelligence - Google Patents
Cell culture monitoring method and system based on artificial intelligence Download PDFInfo
- Publication number
- CN117421691B CN117421691B CN202311740189.3A CN202311740189A CN117421691B CN 117421691 B CN117421691 B CN 117421691B CN 202311740189 A CN202311740189 A CN 202311740189A CN 117421691 B CN117421691 B CN 117421691B
- Authority
- CN
- China
- Prior art keywords
- data
- impedance
- cell culture
- parameter
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000004113 cell culture Methods 0.000 title claims abstract description 56
- 238000012544 monitoring process Methods 0.000 title claims abstract description 45
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 20
- 238000007781 pre-processing Methods 0.000 claims abstract description 16
- 230000006870 function Effects 0.000 claims description 28
- 230000000694 effects Effects 0.000 claims description 25
- 230000008569 process Effects 0.000 claims description 14
- 239000000463 material Substances 0.000 claims description 12
- 239000003792 electrolyte Substances 0.000 claims description 9
- 239000000725 suspension Substances 0.000 claims description 9
- 230000006399 behavior Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 6
- 239000006285 cell suspension Substances 0.000 claims description 6
- 230000004044 response Effects 0.000 claims description 6
- 238000001228 spectrum Methods 0.000 claims description 6
- 238000012935 Averaging Methods 0.000 claims description 3
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 239000000084 colloidal system Substances 0.000 claims description 3
- 238000013461 design Methods 0.000 claims description 3
- 230000005684 electric field Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 239000011244 liquid electrolyte Substances 0.000 claims description 3
- 230000007704 transition Effects 0.000 claims description 3
- 230000009897 systematic effect Effects 0.000 abstract description 3
- 238000012545 processing Methods 0.000 abstract description 2
- 230000009471 action Effects 0.000 description 3
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000004660 morphological change Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/32—Circuit design at the digital level
- G06F30/33—Design verification, e.g. functional simulation or model checking
- G06F30/3323—Design verification, e.g. functional simulation or model checking using formal methods, e.g. equivalence checking or property checking
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
- G16B5/00—ICT specially adapted for modelling or simulations in systems biology, e.g. gene-regulatory networks, protein interaction networks or metabolic networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Health & Medical Sciences (AREA)
- Geometry (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Physiology (AREA)
- Biotechnology (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Investigating Or Analyzing Materials By The Use Of Electric Means (AREA)
Abstract
The invention discloses a cell culture monitoring method and system based on artificial intelligence. The invention belongs to the technical field of data processing, in particular to a cell culture monitoring method and system based on artificial intelligence, which are used for preprocessing original data, minimizing noise and systematic errors, extracting key parameters from impedance signals by utilizing a mixed equivalent circuit model, and then determining cell volume concentration; and adjusting inertial weight parameters based on the current optimal and worst fitness values, balancing the relation between global search and local search of parameter positions, updating the parameter positions based on individual experience optimal and global optimal, and judging search results based on a fitness threshold value and the maximum iteration times.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a cell culture monitoring method and system based on artificial intelligence.
Background
The cell culture monitoring method utilizes artificial intelligence technologies such as computer vision, machine learning and the like to realize functions of monitoring a cell culture process in real time, analyzing cell characteristics, predicting cell morphological changes and the like; however, in the general cell culture monitoring process, the problem of poor monitoring effect caused by noise interference and inaccurate impedance prediction exists; the general searching method has the problems that the global searching capability is poor, the searching is difficult to converge, and the local searching and the global searching cannot be balanced.
Disclosure of Invention
Aiming at the situation, in order to overcome the defects of the prior art, the invention provides a cell culture monitoring method and a system based on artificial intelligence, and aims at the problem of poor monitoring effect caused by noise interference and inaccurate impedance prediction in the cell culture monitoring process; aiming at the problems that the general searching method has poor global searching capability, the searching is difficult to converge and balance between local searching and global searching cannot be achieved, the scheme adjusts inertia weight parameters based on the current optimal and worst fitness values, balances the relationship between the global searching and the local searching of parameter positions, updates the parameter positions based on individual experience optimal and global optimal, and judges the searching result based on the fitness threshold and the maximum iteration times.
The technical scheme adopted by the invention is as follows: the invention provides an artificial intelligence-based cell culture monitoring method, which comprises the following steps:
step S1: collecting data;
step S2: preprocessing data;
step S3: predicting impedance;
step S4: calculating the cell volume concentration;
step S5: searching initial parameters;
step S6: cell culture monitoring.
Further, in step S1, the data acquisition is to acquire impedance data, dielectric constant data, conductivity data, voltage data and time data during cell culture.
Further, in step S2, the data preprocessing specifically includes the following steps:
step S21: eliminating errors, namely, short-circuiting the two electrodes of all the sensors by clamping the two electrodes together to obtain sensor impedance, and subtracting the sensor impedance from a measured impedance result to obtain actual impedance; to solve the impedance problem introduced by the multi-channel system;
step S22: deleting error data, measuring 25 impedance data points at each frequency, averaging the 25 data points to obtain an impedance value at one frequency, calculating the standard deviation of each group of original impedance data, wherein the standard deviation represents the discrete degree of the data, comparing the standard deviation of each group of original impedance data with the average value, if the difference between the standard deviation of the frequency data and the average value is higher than 0.02, the data at the frequency has larger noise or abnormal value, and deleting 15 frequency data after the frequency.
Further, in step S3, the predicted impedance specifically includes the steps of:
step S31: a debye relaxation model is built, which shows maxwell-wagner dielectric relaxation when a colloid with cells suspended in a liquid electrolyte is subjected to an alternating electric field, and is described as:
;
wherein epsilon (·) is a complex dielectric constant spectrum, representing the response of the material to alternating current signals of different frequencies;epsilon is the relaxation intensity and represents the variation amplitude of the dielectric constant of the material in the relaxation process; j is the imaginary unit; omega is the radial frequency of the input ac signal; τ is the relaxation time constant, representing the time required for the relaxation process in the material; epsilon h Is the dielectric constant at the high frequency limit, representing the dielectric properties of the material at high frequencies;
step S32: establishing a conductivity debye relaxation model, taking into account the lossy nature of cell culture in a culture solution, and adopting the conductivity debye relaxation model to represent the following steps:
;
wherein σ (·) is the complex conductivity spectrum; sigma (sigma) 0 Is the conductivity at the low frequency limit;σ is the conductivity variation amplitude;
step S33: establishing an equivalent circuit, re-modeling the electrolyte suspension components into an equivalent circuit model, describing the characteristics and response of the electrolyte suspension components more accurately, and further combining the electrolyte suspension components with the voltage enhancement effect components, so that the accuracy and interpretation capability of a feature extraction algorithm are improved, wherein the equivalent circuit model is expressed as:
;
wherein Z is cl The impedance of the equivalent circuit, C is the capacitance of the equivalent circuit; c (C) 0 Is the capacitance value at the high frequency limit; at C 0 In the case of =1, C is equivalent to Δε in the debye relaxation model;
step S34: the conductivity debye relaxation model is re-expressed by means of dielectric constants using the following formula:
;
step S35: modeling is based on the EP effect, i.e. the voltage boosting effect, using the following formula:
;
wherein Z is EP Is a complex impedance of the EP effect; q (·) is an impedance function; n is the phase index of the CPE component describing the phase characteristics of the impedance function in the complex frequency domain; CPE component refers to a constant phase element component; when n=1, the CPE component describes pure capacitive behavior, and when n=0.5, the CPE component describes pure diffuse behavior;
step S36: the predicted impedance is calculated, since the electroosmotic effect occurs near the electrodes with most of the cell suspension in between, and thus the electroosmotic effect component and the cell suspension component are connected in series, expressed as:
;
in the method, in the process of the invention,is the predicted impedance.
Further, in step S4, the calculating the cell volume concentration specifically includes the steps of:
step S41: defining a distance function, balancing the impedance and the dielectric constant by defining the distance function, and taking the logarithm of the dielectric constant and the impedance to obtain the distance is kept unchanged in the transition, expressed as follows:
;
wherein ε 1 And epsilon 2 Is the dielectric constant of different media, Z 1 And Z 2 Is the impedance of different media;
step S42: the loss function L is defined using the formula:
;
in the method, in the process of the invention,is the true impedance, K is the total sensor data, K is the data index;
step S43: calculating the cell volume concentration, obtaining the determined values of relaxation intensity and relaxation time constant of the cell volume by minimizing a loss function, and further obtaining the cell volume concentration, wherein the formula is as follows:
logCc=logC1+C2logε+C3logτ;
where C1, C2 and C3 are calibration coefficients and Cc is the cell volume concentration.
Further, in step S5, the initial parameter search specifically includes the following steps:
step S51: initializing, namely initializing a parameter position based on an initial parameter search space, taking the cell volume concentration correct rate obtained based on the parameter position as an adaptability value of the parameter position, wherein a formula for initializing the parameter position is as follows:
;
wherein x is I,J Is the parameter position of individual I in the J dimension, L J Is the lower limit of the J dimension of the search space, U J Is the upper bound of the search space J dimension, rand is a random number from 0 to 1;
step S52: the self-adaptive inertia weight coefficient W is designed, and is a core parameter of the balance parameter global searching capability and the local searching capability, and the formula is as follows:
;
wherein f (t) is the fitness value of the parameter position, t is the current iteration number, f w Is the worst fitness value in the t-th iteration, f b Is the optimal fitness value in the t-th iteration, W max Is the maximum weight, W min Is the minimum weight;
step S53: the design location is updated using the following formula:
;
;
in the middle ofV is the speed, r 1 、r 2 And r 3 Is a random number of 0 to 1 independent of each other, x is a parameter position, g is a global optimum position, p is an individual experience optimum position, k1 is an individual variation of the parameter, c 1 Is a local learning factor, c 2 Is a global learning factor, c 3 Is a vitality factor;
step S54: designing iterative search, presetting an adaptability threshold, and outputting a parameter position when the adaptability value of the parameter position is higher than the adaptability threshold; if the maximum iteration times are reached, reinitializing the parameter positions for searching; otherwise, continuing the iterative search.
Further, in step S6, the cell culture monitoring is to obtain an optimal parameter based on the parameter position output in step S5, so as to monitor the cell volume concentration in the cell culture process in real time.
The invention provides an artificial intelligence-based cell culture monitoring system which comprises a data acquisition module, a data preprocessing module, an impedance prediction module, a cell volume concentration calculation module, an initial parameter search module and a cell culture monitoring module, wherein the data acquisition module is used for acquiring a cell volume concentration of a cell;
the data acquisition module acquires impedance data, dielectric constant data, conductivity data, voltage data and time data in the cell culture process and sends the data to the data preprocessing module;
the data preprocessing module is used for eliminating errors and deleting error data of the acquired data and sending the data to the impedance prediction module;
the impedance prediction module establishes an equivalent circuit based on the debye relaxation model and the conductivity debye relaxation model, re-represents the conductivity debye relaxation model in a dielectric constant mode, models based on an EP effect, calculates predicted impedance, and sends data to the cell volume concentration calculation module;
the cell volume concentration calculating module obtains the determined values of the relaxation intensity and the relaxation time constant of the cell volume by minimizing the loss function based on the defined distance function and the loss function, further obtains the cell volume concentration, and sends the data to the initial parameter searching module;
the initial parameter searching module updates the position based on the designed self-adaptive inertia weight coefficient, searches and judges based on the maximum iteration number and the adaptability threshold value, and sends data to the cell culture monitoring module;
the cell culture monitoring module obtains optimal parameters based on the parameter positions output by the initial parameter searching module, so that the cell volume concentration in the cell culture process is monitored in real time.
By adopting the scheme, the beneficial effects obtained by the invention are as follows:
(1) Aiming at the problem of poor monitoring effect caused by noise interference and inaccurate impedance prediction in the cell culture monitoring process, the scheme preprocesses the original data, minimizes noise and systematic errors, extracts key parameters from impedance signals by using a mixed equivalent circuit model, and then determines the cell volume concentration.
(2) Aiming at the problems that the general searching method has poor global searching capability, the searching is difficult to converge and balance between local searching and global searching cannot be achieved, the scheme adjusts inertia weight parameters based on the current optimal and worst fitness values, balances the relationship between the global searching and the local searching of parameter positions, updates the parameter positions based on individual experience optimal and global optimal, and judges the searching result based on the fitness threshold and the maximum iteration times.
Drawings
FIG. 1 is a schematic flow chart of an artificial intelligence based cell culture monitoring method provided by the invention;
FIG. 2 is a schematic diagram of an artificial intelligence based cell culture monitoring system provided by the present invention;
FIG. 3 is a flow chart of step S3;
fig. 4 is a flow chart of step S5.
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention; all other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the description of the present invention, it should be understood that the terms "upper," "lower," "front," "rear," "left," "right," "top," "bottom," "inner," "outer," and the like indicate orientation or positional relationships based on those shown in the drawings, merely to facilitate description of the invention and simplify the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus should not be construed as limiting the invention.
In an embodiment, referring to fig. 1, the method for monitoring cell culture based on artificial intelligence provided by the invention comprises the following steps:
step S1: collecting data;
step S2: preprocessing data;
step S3: predicting impedance;
step S4: calculating the cell volume concentration;
step S5: searching initial parameters;
step S6: cell culture monitoring.
In step S1, the data acquisition is to acquire impedance data, dielectric constant data, conductivity data, voltage data and time data during cell culture, as described above with reference to fig. 1.
In the third embodiment, referring to fig. 1, the data preprocessing specifically includes the following steps in step S2, where the steps are as follows:
step S21: eliminating errors, namely, short-circuiting the two electrodes of all the sensors by clamping the two electrodes together to obtain sensor impedance, and subtracting the sensor impedance from a measured impedance result to obtain actual impedance; to solve the impedance problem introduced by the multi-channel system;
step S22: deleting error data, measuring 25 impedance data points at each frequency, averaging 25 data points to obtain an impedance value at one frequency, calculating the standard deviation of each group of original impedance data, wherein the standard deviation represents the discrete degree of the data, namely the discrete degree of the data points relative to the mean value, comparing the standard deviation of each group of original impedance data with the ratio of the mean value, if the difference between the standard deviation of the frequency data and the mean value is higher than 0.02, the data at the frequency has larger noise or abnormal value, and deleting 15 frequency data after the frequency.
In the fourth embodiment, referring to fig. 1 and 3, the predicted impedance in step S3 specifically includes the following steps:
step S31: a debye relaxation model is built, which shows maxwell-wagner dielectric relaxation when a colloid with cells suspended in a liquid electrolyte is subjected to an alternating electric field, and is described as:
;
wherein epsilon (·) is a complex dielectric constant spectrum, representing the response of the material to alternating current signals of different frequencies;epsilon is the relaxation intensity and represents the variation amplitude of the dielectric constant of the material in the relaxation process; j is the imaginary unit; omega is the radial frequency of the input ac signal; τ is the relaxation time constant, representing the time required for the relaxation process in the material; epsilon h Is the dielectric constant at the high frequency limit, representing the dielectric properties of the material at high frequencies;
step S32: establishing a conductivity debye relaxation model, taking into account the lossy nature of cell culture in a culture solution, and adopting the conductivity debye relaxation model to represent the following steps:
;
wherein σ (·) is the complex conductivity spectrum; sigma (sigma) 0 Is the conductivity at the low frequency limit;σ is the conductivity variation amplitude;
step S33: establishing an equivalent circuit, re-modeling the electrolyte suspension components into an equivalent circuit model, describing the characteristics and response of the electrolyte suspension components more accurately, and further combining the electrolyte suspension components with the voltage enhancement effect components, so that the accuracy and interpretation capability of a feature extraction algorithm are improved, wherein the equivalent circuit model is expressed as:
;
wherein Z is cl The impedance of the equivalent circuit, C is the capacitance of the equivalent circuit; c (C) 0 Is the capacitance value at the high frequency limit; at C 0 In the case of =1, C is equivalent to Δε in the debye relaxation model;
step S34: the conductivity debye relaxation model is re-expressed by means of dielectric constants using the following formula:
;
step S35: modeling is based on the EP effect, i.e. the voltage boosting effect, using the following formula:
;
wherein Z is EP Is a complex impedance of the EP effect; q (·) is an impedance function; n is the phase index of the CPE component describing the phase characteristics of the impedance function in the complex frequency domain; CPE component refers to a constant phase element component; when n=1, the CPE component describes pure capacitive behavior, and when n=0.5, the CPE component describes pure diffuse behavior;
step S36: the predicted impedance is calculated, since the electroosmotic effect occurs near the electrodes with most of the cell suspension in between, and thus the electroosmotic effect component and the cell suspension component are connected in series, expressed as:
;
in the method, in the process of the invention,is the predicted impedance.
Embodiment five, referring to fig. 1, based on the above embodiment, in step S4, calculating the cell volume concentration specifically includes the following steps:
step S41: defining a distance function, balancing the impedance and the dielectric constant by defining the distance function, and taking the logarithm of the dielectric constant and the impedance to obtain the distance is kept unchanged in the transition, expressed as follows:
;
wherein ε 1 And epsilon 2 Is the dielectric constant of different media, Z 1 And Z 2 Is the impedance of different media;
step S42: the loss function L is defined using the formula:
;
in the method, in the process of the invention,is the true impedance, K is the total sensor data, K is the data index;
step S43: calculating the cell volume concentration, obtaining the determined values of relaxation intensity and relaxation time constant of the cell volume by minimizing a loss function, and further obtaining the cell volume concentration, wherein the formula is as follows:
logCc=logC1+C2logε+C3logτ;
where C1, C2 and C3 are calibration coefficients and Cc is the cell volume concentration.
By executing the operation, the method and the device for monitoring the cell culture have the advantages that the problem that the monitoring effect is poor due to noise interference and inaccurate impedance prediction in the cell culture monitoring process is solved, the original data are preprocessed, noise and systematic errors are minimized, key parameters are extracted from impedance signals by using a mixed equivalent circuit model, and then the cell volume concentration is determined.
Embodiment six, referring to fig. 1 and 4, based on the above embodiment, in step S5, the initial parameter search specifically includes the following steps:
step S51: initializing, namely initializing a parameter position based on an initial parameter search space, taking the cell volume concentration correct rate obtained based on the parameter position as an adaptability value of the parameter position, wherein a formula for initializing the parameter position is as follows:
;
wherein x is I,J Is the parameter position of individual I in the J dimension, L J Is the lower limit of the J dimension of the search space, U J Is the upper bound of the search space J dimension, rand is a random number from 0 to 1;
step S52: the self-adaptive inertia weight coefficient W is designed, and is a core parameter of the balance parameter global searching capability and the local searching capability, and the formula is as follows:
;
wherein f (t) is the fitness value of the parameter position, t is the current iteration number, f w Is the worst fitness value in the t-th iteration, f b Is the optimal fitness value in the t-th iteration, W max Is the maximum weight, W min Is the minimum weight;
step S53: the design location is updated using the following formula:
;
;
wherein v is the velocity, r 1 、r 2 And r 3 Is a random number of 0 to 1 independent of each other, x is a parameter position, g is a global optimum position, p is an individual experience optimum position, k1 is an individual variation of the parameter, c 1 Is a local learning factor, c 2 Is a global learning factor, c 3 Is a vitality factor;
step S54: designing iterative search, presetting an adaptability threshold, and outputting a parameter position when the adaptability value of the parameter position is higher than the adaptability threshold; if the maximum iteration times are reached, reinitializing the parameter positions for searching; otherwise, continuing the iterative search.
By executing the above operation, aiming at the problems that the general searching method has poor global searching capability, the searching is difficult to converge and can not reach balance between the local searching and the global searching, the scheme adjusts the inertia weight parameter based on the current optimal and worst fitness value, balances the relation between the global searching and the local searching of the parameter position, updates the parameter position based on the individual experience optimal and the global optimal, and judges the searching result based on the fitness threshold and the maximum iteration times.
Embodiment seven, referring to fig. 1, the embodiment is based on the above embodiment, and in step S6, the cell culture monitoring obtains the optimal parameters based on the parameter positions output in step S5, so as to monitor the cell volume concentration in the cell culture process in real time.
An eighth embodiment, referring to fig. 2, is based on the above embodiment, and the artificial intelligence-based cell culture monitoring system provided by the present invention includes a data acquisition module, a data preprocessing module, an impedance prediction module, a cell volume concentration calculation module, an initial parameter search module, and a cell culture monitoring module;
the data acquisition module acquires impedance data, dielectric constant data, conductivity data, voltage data and time data in the cell culture process and sends the data to the data preprocessing module;
the data preprocessing module is used for eliminating errors and deleting error data of the acquired data and sending the data to the impedance prediction module;
the impedance prediction module establishes an equivalent circuit based on the debye relaxation model and the conductivity debye relaxation model, re-represents the conductivity debye relaxation model in a dielectric constant mode, models based on an EP effect, calculates predicted impedance, and sends data to the cell volume concentration calculation module;
the cell volume concentration calculating module obtains the determined values of the relaxation intensity and the relaxation time constant of the cell volume by minimizing the loss function based on the defined distance function and the loss function, further obtains the cell volume concentration, and sends the data to the initial parameter searching module;
the initial parameter searching module updates the position based on the designed self-adaptive inertia weight coefficient, searches and judges based on the maximum iteration number and the adaptability threshold value, and sends data to the cell culture monitoring module;
the cell culture monitoring module obtains optimal parameters based on the parameter positions output by the initial parameter searching module, so that the cell volume concentration in the cell culture process is monitored in real time.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
The invention and its embodiments have been described above with no limitation, and the actual construction is not limited to the embodiments of the invention as shown in the drawings. In summary, if one of ordinary skill in the art is informed by this disclosure, a structural manner and an embodiment similar to the technical solution should not be creatively devised without departing from the gist of the present invention.
Claims (6)
1. The cell culture monitoring method based on artificial intelligence is characterized by comprising the following steps of: the method comprises the following steps:
step S1: collecting data;
step S2: preprocessing data;
step S3: predicting impedance;
step S4: calculating the cell volume concentration;
step S5: searching initial parameters;
step S6: cell culture monitoring;
step S31: a debye relaxation model is built, which shows maxwell-wagner dielectric relaxation when a colloid with cells suspended in a liquid electrolyte is subjected to an alternating electric field, and is described as:
;
wherein epsilon (·) is a complex dielectric constant spectrum, representing the response of the material to alternating current signals of different frequencies;is the relaxation intensity, which represents the variation amplitude of dielectric constant in the relaxation process of the material; j is the imaginary unit; omega is the radial frequency of the input ac signal; τ is the relaxation time constant, representing the time required for the relaxation process in the material; epsilon h Is the dielectric constant at the high frequency limit, representing the dielectric properties of the material at high frequencies;
step S32: establishing a conductivity debye relaxation model, taking into account the lossy nature of cell culture in a culture solution, and adopting the conductivity debye relaxation model to represent the following steps:
;
wherein σ (·) is the complex conductivity spectrum; sigma (sigma) 0 Is the conductivity at the low frequency limit;σ is the conductivity variation amplitude;
step S33: establishing an equivalent circuit, re-modeling the electrolyte suspension components into an equivalent circuit model, describing the characteristics and response of the electrolyte suspension components more accurately, and further combining the electrolyte suspension components with the voltage enhancement effect components, so that the accuracy and interpretation capability of a feature extraction algorithm are improved, wherein the equivalent circuit model is expressed as:
;
wherein Z is cl The impedance of the equivalent circuit, C is the capacitance of the equivalent circuit; c (C) 0 Is the capacitance value at the high frequency limit; at C 0 In the case of =1, C is equivalent to Δε in the debye relaxation model;
step S34: the conductivity debye relaxation model is re-expressed by means of dielectric constants using the following formula:
;
step S35: modeling is based on the EP effect, i.e. the voltage boosting effect, using the following formula:
;
wherein Z is EP Is a complex impedance of the EP effect; q (·) is an impedance function; n is the phase index of the CPE component describing the phase characteristics of the impedance function in the complex frequency domain; CPE component refers to a constant phase element component; when n=1, the CPE component describes pure capacitive behavior, and when n=0.5, the CPE component describes pure diffuse behavior;
step S36: the predicted impedance is calculated, since the electroosmotic effect occurs near the electrodes with most of the cell suspension in between, and thus the electroosmotic effect component and the cell suspension component are connected in series, expressed as:
;
in the method, in the process of the invention,is the predicted impedance;
step S51: initializing, namely initializing a parameter position based on an initial parameter search space, taking the cell volume concentration correct rate obtained based on the parameter position as an adaptability value of the parameter position, wherein a formula for initializing the parameter position is as follows:
;
wherein x is I,J Is the parameter position of individual I in the J dimension, L J Is the lower limit of the J dimension of the search space, U J Is the upper bound of the search space J dimension, rand is a random number from 0 to 1;
step S52: the self-adaptive inertia weight coefficient W is designed, and is a core parameter of the balance parameter global searching capability and the local searching capability, and the formula is as follows:
;
wherein f (t) is the fitness value of the parameter position, t is the current iteration number, f w Is the worst fitness value in the t-th iteration, f b Is the optimal fitness value in the t-th iteration, W max Is the maximum weight, W min Is the minimum weight;
step S53: the design location is updated using the following formula:
;
;
wherein v is the velocity, r 1 、r 2 And r 3 Is a random number of 0 to 1 independent of each other, x is a parameter position, g is a global optimum position, p is an individual experience optimum position, k1 is an individual variation of the parameter, c 1 Is a local learning factor, c 2 Is a global learning factor, c 3 Is a vitality factor;
step S54: designing iterative search, presetting an adaptability threshold, and outputting a parameter position when the adaptability value of the parameter position is higher than the adaptability threshold; if the maximum iteration times are reached, reinitializing the parameter positions for searching; otherwise, continuing to iterate the search;
step S41: defining a distance function, balancing the impedance and the dielectric constant by defining the distance function, and taking the logarithm of the dielectric constant and the impedance to obtain the distance is kept unchanged in the transition, expressed as follows:
;
wherein ε 1 And epsilon 2 Is the dielectric constant of different media, Z 1 And Z 2 Is the impedance of different media;
step S42: the loss function L is defined using the formula:
;
in the method, in the process of the invention,is the true impedance, K is the total sensor data, K is the data index;
step S43: calculating the cell volume concentration, obtaining the determined values of relaxation intensity and relaxation time constant of the cell volume by minimizing a loss function, and further obtaining the cell volume concentration, wherein the formula is as follows:
logCc=logC1+C2logε+C3logτ;
where C1, C2 and C3 are calibration coefficients and Cc is the cell volume concentration.
2. The artificial intelligence based cell culture monitoring method of claim 1, wherein: in step S2, the data preprocessing specifically includes the following steps:
step S21: eliminating errors, namely, short-circuiting the two electrodes of all the sensors by clamping the two electrodes together to obtain sensor impedance, and subtracting the sensor impedance from a measured impedance result to obtain actual impedance; to solve the impedance problem introduced by the multi-channel system;
step S22: deleting error data, measuring 25 impedance data points at each frequency, averaging 25 data points to obtain an impedance value at one frequency, calculating the standard deviation of each group of original impedance data, wherein the standard deviation represents the discrete degree of the data, namely the discrete degree of the data points relative to the mean value, comparing the standard deviation of each group of original impedance data with the ratio of the mean value, if the difference between the standard deviation of the frequency data and the mean value is higher than 0.02, the data at the frequency has larger noise or abnormal value, and deleting 15 frequency data after the frequency.
3. The artificial intelligence based cell culture monitoring method of claim 1, wherein: in step S1, the data acquisition is to acquire impedance data, dielectric constant data, conductivity data, voltage data and time data in the cell culture process.
4. The artificial intelligence based cell culture monitoring method of claim 1, wherein: in step S6, the cell culture monitoring is to obtain an optimal parameter based on the parameter position output in step S5, so as to monitor the cell volume concentration in the cell culture process in real time.
5. An artificial intelligence based cell culture monitoring system for implementing an artificial intelligence based cell culture monitoring method according to any one of claims 1-4, characterized in that: the system comprises a data acquisition module, a data preprocessing module, an impedance prediction module, a cell volume concentration calculation module, an initial parameter searching module and a cell culture monitoring module.
6. The artificial intelligence based cell culture monitoring system of claim 5, wherein:
the data acquisition module acquires impedance data, dielectric constant data, conductivity data, voltage data and time data in the cell culture process and sends the data to the data preprocessing module;
the data preprocessing module is used for eliminating errors and deleting error data of the acquired data and sending the data to the impedance prediction module;
the impedance prediction module establishes an equivalent circuit based on the debye relaxation model and the conductivity debye relaxation model, re-represents the conductivity debye relaxation model in a dielectric constant mode, models based on an EP effect, calculates predicted impedance, and sends data to the cell volume concentration calculation module;
the cell volume concentration calculating module obtains the determined values of the relaxation intensity and the relaxation time constant of the cell volume by minimizing the loss function based on the defined distance function and the loss function, further obtains the cell volume concentration, and sends the data to the initial parameter searching module;
the initial parameter searching module updates the position based on the designed self-adaptive inertia weight coefficient, searches and judges based on the maximum iteration number and the adaptability threshold value, and sends data to the cell culture monitoring module;
the cell culture monitoring module obtains optimal parameters based on the parameter positions output by the initial parameter searching module, so that the cell volume concentration in the cell culture process is monitored in real time.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311740189.3A CN117421691B (en) | 2023-12-18 | 2023-12-18 | Cell culture monitoring method and system based on artificial intelligence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311740189.3A CN117421691B (en) | 2023-12-18 | 2023-12-18 | Cell culture monitoring method and system based on artificial intelligence |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117421691A CN117421691A (en) | 2024-01-19 |
CN117421691B true CN117421691B (en) | 2024-03-15 |
Family
ID=89528784
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311740189.3A Active CN117421691B (en) | 2023-12-18 | 2023-12-18 | Cell culture monitoring method and system based on artificial intelligence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117421691B (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115222007A (en) * | 2022-05-31 | 2022-10-21 | 复旦大学 | Improved particle swarm parameter optimization method for glioma multitask integrated network |
CN115343345A (en) * | 2022-09-15 | 2022-11-15 | 南京工业大学 | High-precision detection method of glucolase biosensor based on adaptive Kalman filtering algorithm |
WO2023288314A1 (en) * | 2021-07-15 | 2023-01-19 | Georgia Tech Research Corporation | Methods of determining viable cell count and impedance-based biosensors for the same |
CN117035001A (en) * | 2023-08-07 | 2023-11-10 | 大连理工大学 | Conversion force sensor temperature compensation method based on Kriging interpolation |
CN117114144A (en) * | 2023-10-24 | 2023-11-24 | 青岛农业大学 | Rice salt and alkali resistance prediction method and system based on artificial intelligence |
-
2023
- 2023-12-18 CN CN202311740189.3A patent/CN117421691B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2023288314A1 (en) * | 2021-07-15 | 2023-01-19 | Georgia Tech Research Corporation | Methods of determining viable cell count and impedance-based biosensors for the same |
CN115222007A (en) * | 2022-05-31 | 2022-10-21 | 复旦大学 | Improved particle swarm parameter optimization method for glioma multitask integrated network |
CN115343345A (en) * | 2022-09-15 | 2022-11-15 | 南京工业大学 | High-precision detection method of glucolase biosensor based on adaptive Kalman filtering algorithm |
CN117035001A (en) * | 2023-08-07 | 2023-11-10 | 大连理工大学 | Conversion force sensor temperature compensation method based on Kriging interpolation |
CN117114144A (en) * | 2023-10-24 | 2023-11-24 | 青岛农业大学 | Rice salt and alkali resistance prediction method and system based on artificial intelligence |
Non-Patent Citations (5)
Title |
---|
Electrical impedance spectroscopy for detection of bacterial cells in suspensions using interdigitated microelectrodes;Yang, LJ;TALANTA;20080215;全文 * |
基于生物阻抗谱的生物细胞活性免标记检测方法;徐恺杰等;生物化学与生物物理进展;20230314;第50卷(第11期);全文 * |
基于电化学阻抗谱层析成像的微流道内细胞分布检测研究;刘夏移;中国博士硕士学位论文集信息科技辑;20210115;全文 * |
微流控电阻抗检测***构建及其应用;张思祥;竭霞;胡雪迎;李默;张英杰;周围;;分析科学学报;20200420(02);全文 * |
面向血液葡萄糖测量的生物组织阻抗模型;邵平;张晓东;邹雪城;;激光生物学报;20160615(03);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN117421691A (en) | 2024-01-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116150676B (en) | Equipment fault diagnosis and identification method and device based on artificial intelligence | |
EP3358508A1 (en) | Abnormality detection apparatus, abnormality detection method, and program | |
JP2020501238A (en) | Face detection training method, apparatus and electronic equipment | |
CN110850164B (en) | Multi-harmonic responsibility division method considering background harmonic voltage fluctuation and impedance change | |
EP3769047B1 (en) | Sensor calibration | |
CN111046327A (en) | Prony analysis method suitable for low-frequency oscillation and subsynchronous oscillation identification | |
CN109815988B (en) | Model generation method, classification method, device and computer-readable storage medium | |
CN110766058A (en) | Battlefield target detection method based on optimized RPN (resilient packet network) | |
CN109142865B (en) | Frequency domain spectrum identification method considering polarization equivalent circuit parameters of oiled paper insulation interface | |
CN109657147B (en) | Microblog abnormal user detection method based on firefly and weighted extreme learning machine | |
CN112990330A (en) | User energy abnormal data detection method and device | |
CN108829878A (en) | A kind of industry experiment data abnormal point detecting method and device | |
CN110673208B (en) | First arrival picking method and system for high-dimensional feature constraint under machine learning framework | |
CN107505358A (en) | A kind of detection method of the fermentation of black tea index of quality based on electrical characteristic parameter | |
CN115166839A (en) | Magnetotelluric sounding method power spectrum intelligent selection method based on deep learning classification | |
CN104376078A (en) | Abnormal data detection method based on knowledge entropy | |
CN117421691B (en) | Cell culture monitoring method and system based on artificial intelligence | |
CN104516991A (en) | Gamma sensor full-temperature range compensation method | |
CN107846402A (en) | A kind of BGP stability method for detecting abnormality, device and electronic equipment | |
CN109408498A (en) | The identification of time series feature and decomposition method based on eigenmatrix decision tree | |
CN110956112B (en) | Novel high-reliability slewing bearing service life assessment method | |
CN116702005A (en) | Neural network-based data anomaly diagnosis method and electronic equipment | |
CN110969238A (en) | Method and device for calibrating electricity consumption data | |
JP2018055402A (en) | Equivalent circuit estimation method | |
CN116484184A (en) | Method and device for enhancing partial discharge defect sample of power equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |