A multi-height position correction method and system based on cell electrical impedance measurement

Abstract

The application discloses a multi-height position correction method and system based on cell electric impedance measurement, and the correction method comprises the following steps: 1) acquiring an electric impedance signal by adopting a double-difference seven-electrode structure, and performing Gaussian filtering processing on the real part of the electric impedance signal; 2) obtaining characteristic points generated under different heights according to a function of finding extreme values, wherein the characteristic points are extreme points; 3) judging whether the characteristic points are'saddle' type waves according to the number of the characteristic points, and calculating a correction factor; and 4) after judging whether the characteristic points are'saddle' type waves, obtaining the correction factor of the waveform under different heights, and performing scaling on the waveform data under different heights according to a standard factor and the correction factor corresponding to the height, so as to obtain a corrected full-wave image. On the basis of the seven-electrode double-difference, the full-wave correction is performed, so that the seven-electrode double-difference is not only limited to low flow channels, but also can be applied to high flow channels, and the flux is improved.

Description

TECHNICAL FIELD

[0001] The application relates to the field of cell electrical impedance measurement, and particularly relates to a multi-height position correction method and system based on cell electrical impedance measurement. BACKGROUND

[0002] The size distribution of various biological samples in sub-micron to micron scale plays a very important role in biomedical and clinical applications, for example, the accurate size and count of bacteria are crucial in the sensitivity test of antibacterial drugs and the drug resistance test of antibacterial drugs. The size distribution of most common bacteria is in the range of sub-micron to micron scale, which makes it a key link to measure the size and quantity of cells with high precision. In addition, the apoptotic bodies in the apoptosis of cells are broken (0.1 μm-5 μm) and have been found to be related to immune regulation and tumorigenesis, and accurate measurement and sub-micron accuracy are required for the above biological samples.

[0003] The common methods for cell particle size and count are optical microscopy and image-based combination and Brownian motion detection using light scattering, the latter has been applied to commercial nanoparticle tracking analyzers (NTAs) to achieve high sensitivity and resolution at the nanoscale. Such methods can effectively observe the number and size of cells, play a very important role in biomedical and clinical applications, and are constantly evolving. However, the use of optical microscopy and image-based combination in rapid detection requires a large amount of time and labor, and is a labor-intensive method. In addition, the commercial nanoparticle tracking analyzer is difficult to detect living biological samples with self-flowing due to its method being very sensitive to the movement of objects. In recent years, impedance-based microfluidic flow cytometry has become an alternative method. It has the characteristics of label-free and high throughput, and has become an effective method for cell size measurement, counting and studying the dielectric properties of cells.

[0004] As the embodiment of cells on the electrical impedance signal, different reflections of the electrical signal reflect the cell number detection and size measurement. The basic principle of cell electrical impedance measurement is that when the current flows through the cell, it is resisted by the cell membrane, which affects the amplitude and size of the current flowing through the cell, so the voltage of the current flowing through the cell can be calculated to calculate the electrical impedance of the cell. The electrical impedance signal is associated with the size, shape, membrane capacitance, cytoplasm conductivity and other biophysical properties of the cell, and the analysis of the electrical impedance signal plays a crucial role. The detection of cell events usually uses a coplanar differential electrode as a sensitive unit. However, due to the exponential decay of the electric field strength in the normal direction of the coplanar electrode with distance, the spatial distribution of the current density and its non-uniformity. The consequences may be: the same cell flowing through the channel at different heights will produce different current amplitudes, resulting in misjudgment as different cells; different cells flowing through the channel at different heights will produce the same signal, resulting in misjudgment as the same cell. This signal interference caused by the height of the cell in the flow channel is called position dependence. Jianwei Zhong et al. proposed a seven-electrode double-differential method, which improved the non-uniformity of the electric field to some extent, and the double-differential signal has obvious signal characteristics of "saddle" type wave (passive differential signal component), bipolar Gaussian wave peak type (active differential signal component). The ratio of the difference between the peak and the valley of the "saddle surface" to the peak value of the cell event signal is used as the relative prominence, and the size correction is carried out by using it, which can reduce the influence of the signal difference of the cell at different heights at low flux. However, the limitation of this method is that when the height of the flow channel is large, the "saddle surface" of the double-differential signal waveform disappears, resulting in the failure to calculate the correction factor, so the algorithm is invalid. That is: the existing method is only applicable to shallow flow channels with "saddle surface" characteristic waveforms (flow channel height within 8μm). In actual application, it is necessary to increase the height of the flow channel to improve the detection flux, which will inevitably lead to the disappearance of the "saddle surface" signal characteristics, and the existing algorithm is invalid. Therefore, the present application proposes a full-wave correction algorithm, which is expected to effectively make up for the shortcomings of the existing correction method, and is applicable to conventional waveforms with "saddle surface" waveform characteristics and high-flux detection signal waveforms with "saddle surface" disappearing. SUMMARY

[0005] The purpose of the present application is to provide a multi-height position correction method and system based on cell electrical impedance measurement. The correction method can be applied to the optimization of electrical impedance signals at different heights.

[0006] To achieve the above-mentioned purpose, the technical scheme of the present application is:

[0007] In a first aspect, the present application provides a multi-height position correction method based on cell electrical impedance measurement. The correction method is applied to the optimization of electrical impedance signals at different heights, and the correction method comprises the following steps:

[0008] 1) Obtain the electrical impedance signal by using a double-differential seven-electrode structure, and obtain the real part of the electrical impedance signal and perform Gaussian filtering processing;

[0009] 2) Obtain the characteristic points generated at different heights according to the function of finding extreme values based on the data after Gaussian filtering processing, and the characteristic points are extreme points;

[0010] 3) Determine whether it is a "saddle" type wave according to the number of characteristic points: determine the waveform shape according to the number of characteristic points, if the number of characteristic points is 8, there is a "saddle" type wave, find the maximum extreme point MAX and the minimum extreme point MIN of the "saddle" type wave, the wave valley B1 of the first saddle wave and the wave peak B2 of the second saddle wave, at this time, the correction factor is directly calculated according to formula (1);

[0011] if the number of characteristic points is 4, it is a "saddle" type wave disappearance type wave, find the maximum extreme point MAX and the minimum extreme point MIN of the "saddle" type wave disappearance type wave, there are four extreme points, the first extreme point is recorded as B1, and the fourth extreme point is recorded as B2, at this time, the correction factor is calculated according to formula (1);

[0012]

[0013] wherein β is the correction factor, B1, B2, MIN, MAX are the extreme point ordinates of the waveform;

[0014] 4) After determining whether it is a "saddle" type wave, the correction factor of the waveform at different heights is obtained, the waveform data at different heights is scaled according to the standard factor and the correction factor corresponding to the height, and the corrected full wave image is obtained;

[0015] The calculation process of the standard factor β0 is: selecting the waveforms with "saddle" type waves in different heights, taking the middle value of the height of the cells flowing through the flow channel of all the waveforms with "saddle" type waves as the standard height, and taking the correction factor corresponding to the standard height as the standard factor β0.

[0016] Further, the number of cells is determined according to the occurrence frequency of the maximum peak in the corrected full wave image.

[0017] Further, the calculation process of the standard factor is: selecting the middle value h0 of the flow channel height as the standard height according to formula (2), and taking the correction factor corresponding to the standard height as the standard factor,

[0018] (h max +h min ) / 2=h0 (2)

[0019] wherein h max , h minrespectively the maximum coordinate value and the minimum coordinate value of the corresponding height of the flow channel, h max The value range of h is 20-60 μm.

[0020] Further, the data of each point in the whole waveform of the cell flowing from the inlet to the outlet is normalized by dividing by the correction factor at the corresponding height, and then divided by the standard factor, so that the waveforms at all heights are all normalized to the data of the flow-in at the standard height.

[0021] In a second aspect, the application provides a multi-height position correction system based on cell electrical impedance measurement, comprising:

[0022] COMSOL software is used to simulate electrical signals at different flow-in heights, and simulated signals are obtained by cells flowing through flow channels of different heights;

[0023] COMSOL with MATLAB is used for real-time data processing, impedance signal data extraction of simulated signals, and real part extraction of impedance signal data;

[0024] MATLAB software is used for digital filtering, feature point acquisition, event judgment, correction factor calculation, and full-wave correction to eliminate height dependence of the data transmitted by COMSOL with MATLAB;

[0025] The formula for calculating the correction factor is:

[0026]

[0027] Where β is the correction factor, MIN and MAX are the ordinates of the maximum and minimum extreme points of the waveform; when the waveform is a "saddle" type wave, B1 and B2 are the ordinates of the first saddle wave trough and the second saddle wave peak, respectively; when the waveform is a "saddle" type wave disappearing wave, B1 and B2 are the ordinates of the first extreme point and the fourth extreme point in the waveform, respectively.

[0028] The full-wave correction specifically includes: after obtaining the waveform data corresponding to different heights, the normalized data is obtained by multiplying the correction factor, and then the normalized data at different heights is normalized to the height data corresponding to the standard factor, so that the standard normalization operation is realized, thereby eliminating the height dependence.

[0029] The process of full-wave correction is: selecting the waveforms with "saddle" type waves at different heights, taking the middle value of the flow channel height of all cells with "saddle" type waves as the standard height, and taking the correction factor corresponding to the standard height as the standard factor β0,

[0030] Divide the data of each point in the whole waveform by the correction factor under the corresponding height to obtain normalized data, and then divide by the standard factor, standardize all the waveforms of all heights to the inflow data under the standard height, and eliminate the height dependence.

[0031] The COMSOL software adopts a double-difference seven-electrode design for modeling the microfluidic chip structure, adopts a high-frequency sinusoidal alternating input voltage, and utilizes double-difference signals for noise weakening; when cells flow through the flow channel, the current is hindered by the cells to generate deformation, and the measured impedance amplitude will be in a linear relationship with the size of the bioimpedance of the cells; the number of maximum peaks of the signal is proportional to the number of cells flowing through, and the number of cells is calculated; the signal of the cells entering the detection channel is used for analyzing the position characteristics and electrical characteristics of the cells;

[0032] The double-difference seven-electrode structure is adopted, the current difference when the cells flow through the left and right electrodes is used to obtain the impedance signal, and the impedance characteristics and height signal information of the cells are obtained.

[0033] The double-difference seven-electrode structure adopts a structure with an electrode width of 10-200 mu m and a spacing of 10-50 mu m, and the flow channel size is 20-2000 mu m in height and 100-2000 mu m in length.

[0034] The double-difference seven-electrode wiring selection is composed of a negative excitation voltage, a floating electrode, a ground electrode, a positive excitation voltage, a ground electrode, a floating electrode and a negative excitation voltage, and the specific signal is obtained by the ground electrode difference, and the excitation voltage adopts a sinusoidal alternating current input with a frequency selection of 100K-40MHz and an initial phase of 0.

[0035] Compared with the prior art, the beneficial effects of the present application are:

[0036] On the basis of the data processing of the electrical impedance flow cytometry analysis, the present application proposes a full-wave correction for the case where the multi-height algorithm fails, a correction factor is obtained by collecting the characteristic points of the data through waveform event judgment, and the full wave (full wave refers to all waveform characteristics using passive difference and active difference, which is applicable to both saddle waves and non-saddle waves) of multiple heights is optimized according to the standard factor and the correction factor, the full wave correction is realized, and the existing half wave correction (only using the "saddle surface" signal waveform of the passive difference signal) method is obviously beneficial.

[0037] The present application performs full-wave correction on the basis of the seven-electrode double-difference, so that the seven-electrode double-difference is not only limited to low flow channels, but also can be applied to high flow channels, and the flux is improved. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1Flowchart of the multi-height position correction system based on cell impedance measurement of the application.

[0039] Figure 2 Modeling flowchart for COMSOL software.

[0040] Figure 3 Modeling flowchart for COMSOL software.

[0041] Figure 4 Structural diagram of the microfluidic chip structure.

[0042] Figure 5 Multi-position measurement comparison diagram for cell impedance measurement.

[0043] Figure 6 Diagram for explaining the significance of feature points.

[0044] Figure 7 Waveform comparison diagram for the existence and disappearance of "saddle" type waves.

[0045] Figure 8 Real-time processing algorithm diagram.

[0046] Figure 9 Parameter point optimization diagram; (a) optimized point diagram; (b) original point diagram

[0047] Figure 10 Comparison diagram of original data and full-wave correction to eliminate height dependence; (a) original data diagram; (b) optimized correction diagram; (c) "saddle" type wave recovery diagram after full-wave correction. DETAILED DESCRIPTION

[0048] The application will be further explained in conjunction with the embodiments and the accompanying drawings, but it is not intended to limit the scope of protection of the present application. As shown in Figure 1 Flowchart of the multi-height position correction system based on cell impedance measurement of the application, comprising:

[0049] COMSOL software, used for electrical signal simulation of different inflow heights, and simulated signals are obtained by cells flowing through different height flow channels; COMSOL software is a multi-physical field simulation software based on finite element method, and modeling, parameter selection, physical field construction, etc. are performed by COMSOL software, and simulated signals are obtained by cells flowing through different height flow channels.

[0050] COMSOL with MATLAB, used for data extraction, completing real-time data processing, and COMSOL with MATLAB is used for impedance signal data extraction of simulated signals, and real part acquisition is performed for collection;

[0051] MATLAB software is used for digital filtering, feature point acquisition, event judgment, correction factor calculation and full-wave correction to eliminate height dependence of COMSOL with MATLAB transmitted data.

[0052] COMSOL software is used for COMSOL simulation, and the COMSOL software in this embodiment is COMSOL 6.0 for modeling and parameter selection.

[0053] As shown in Figure 2 The modeling process of COMSOL software is shown in the schematic diagram, which is: starting, creating a mathematical model, setting boundary conditions, setting local parameters, dividing the grid, then judging whether the model passes the test, if not, returning to create a mathematical model; if it passes, performing result analysis, and judging whether it reaches the expectation, if it reaches the expectation, exporting data, if it does not reach the expectation, returning to create a mathematical model. In the software modeling, two-dimensional modeling method is first used to improve the calculation speed; the electrode boundary is set as conductive, and the rest of the boundary is set as insulating; the local parameters are set as: the width of the electrode is 20 μm, the electrode interval is 20 μm, and the software modeling model is shown in Figure 3 The wiring selection of the double-difference seven-electrode is -1V excitation voltage, floating electrode, ground electrode, 1V excitation voltage, ground electrode, floating electrode, -1V excitation voltage, and the specific signal is the ground electrode difference value, the excitation voltage uses a sine alternating current input with a frequency of 500 KHz and an initial phase of 0, the cell parameter selection is a radius of 2.5 μm, and the dielectric constant and conductivity are 9.8131-38485i and 1.07 S / m respectively. The flow channel height is selected as 50 μm, and the cell inflow height is selected as 10 μm, 15 μm, 20 μm, 25 μm, 30 μm, and 35 μm. The grid division uses custom grid to obtain more accurate results, the maximum unit size is selected as 1.25 μm, the minimum unit size is selected as 0.026 μm, the maximum unit growth rate is selected as 1.1, the curvature factor is selected as 0.2, and the number of units for the detection area, i.e. the electrode geometry, is selected as 20 to obtain more accurate results for the detection area. The corresponding parameter selection is that the dielectric constant and conductivity of the solution in the flow channel are 80.2 and 1.6 S / m respectively; after the related condition setting is completed, the model test is judged, the model is observed whether the geometric parameter and other conditions are correctly set, when the test is passed, the result analysis is performed, whether the electric field reaches the expected effect is observed, if the electric field does not reach the expected effect, the model is further optimized, otherwise the data is extracted, and the simulation software process is completed.

[0054] As shown in Figure 4A schematic diagram of the microfluidic chip structure, which is composed of a PDMS layer and a glass layer bonded together, where the electrode pieces above the glass layer are responsible for implementing the electric field and detecting changes in the electrical signal. When the cells flow into the microfluidic chip, the electrical resistance of the cells in the electric field is measured: the electrical resistance of the cells in the electric field depends on the size, shape, structure and membrane conductive properties of the cells. According to Ohm's law, the resistance can be expressed as the ratio of voltage to current:

[0055] R = V / I (1)

[0056] where R represents the electrical resistance of the cells in the electric field, V represents the voltage, and I represents the current.

[0057] Cell electrical impedance: Cell electrical impedance refers to the impedance exhibited by a cell in an alternating electric field. Cell electrical impedance can be represented as a complex number of the phase difference and amplitude ratio of voltage to current:

[0058] Z = R + jX (2)

[0059] where R is the electrical resistance of the cells in the electric field, X is the inductance or capacitance of the cells, and j is the imaginary unit. Since the electrical resistance and capacitance of the cells in an alternating electric field at different frequencies are different, by measuring the changes in cell electrical impedance at different frequencies, the size, shape and structure of the cells can be evaluated.

[0060] From the above formula, the corresponding values of cell electrical impedance can be obtained from the measured differential current data (differential current is the current image analyzed, as double differential detection is used, it is called differential current).

[0061] Figure 5 A schematic diagram of the cells flowing through different heights inside the chip is shown, which is used for comparison of multi-position cell impedance measurement. COMSOL with MATLAB is used for real-time processing, MATLAB's simulink interface is used to change the model parameters of COMSOL model, and the model results are extracted and collected. The change of model parameters uses:

[0062] model.param.set('hch','50[um]');(3)

[0063] where hch is the height of the flow channel in the model, and this function is used to change the height of the flow channel in the COMSOL model.

[0064] model.param.set('y0','25[um]');(4)

[0065] Where y0 represents the height of the cell flow through the channel. Comparative data were obtained by changing the height of the cell flow through the channel when the channel height was 50 μm. The height of the cell flow through the channel was set to 10 μm, 15 μm, 20 μm, 25 μm, 30 μm, and 35 μm, respectively.

[0066] like Figure 6 The diagram illustrates the process of full-wave optimization correction. Specifically, after extracting data from the calculation results of each set of parameters, the real part of the complex impedance signal is separated. To avoid sharp waveform fluctuations causing errors in event judgment, Gaussian filtering is performed to obtain the corresponding dataset. After obtaining the dataset, a peak-finding function is used to obtain local extrema, thereby acquiring feature point data (including the number and coordinates of feature points). The waveform shape is then determined based on the number of feature points. Figure 7 The waveform comparison diagram showing the presence and disappearance of the "saddle" shaped wave indicates that a "saddle" shaped wave exists when the number of extreme value judgment points is 8. Figure 7 In the case where the cell flows through a channel with a height of 15 μm and a channel height of 50 μm, the maximum and minimum extreme points of the "saddle" shaped wave are found, as well as the trough of the first saddle wave (B1) and the peak of the second saddle wave (B2). At this time, the correction factor is directly calculated according to formula (1), and then full-wave correction optimization is performed. If the number of extreme points is 4, it is a "saddle" shaped wave disappearing type wave, such as Figure 7 In the case where the cell flows through a channel with a height of 30 μm and the channel height is 50 μm, the three points of the saddle wave with the "saddle" shape are combined into one point. The first extreme point and the fourth extreme point are combined into one point, so that the extreme points on both sides of the "saddle" shape are the same as the extreme point in the middle, that is, A1 and C1 are the same as B1, and A2 and C2 are the same as B2. That is, the "saddle" shape disappearance wave is associated with the "saddle" shape wave. The correction factor is calculated using formula (1). The correction factor corresponding to the height of the inflow channel is compared with the standard factor to obtain the scaling factor. The data of each point in the entire waveform from the inlet to the outlet of the cell is processed according to the scaling factor to obtain the image after full wave correction. The image is compared with the original data image to determine the feasibility and optimality of the full wave correction optimization method.

[0067] like Figure 8 The figure shows the parameter significance of each extreme point as a characteristic point in a "saddle" shaped wave. There are 8 extreme points in the figure, which are named A1, B1, C1, MIN, MAX, C2, B2, and A2 in order of distance. Among them, MIN is the minimum extreme point and MAX is the maximum extreme point. The factor is calculated according to formula (1):

[0068]

[0069] Wherein β is the correction factor, B1, B2, MIN, MAX are the extreme points of the waveform, corresponding to the image as shown in Figure 8 .

[0070] The standard factor is calculated as follows: in the flow channel height value range is (20-60 μm), the correction factor of the middle value of the flow channel height is selected, and the corresponding height is the standard height h0:

[0071] (h max +h min ) / 2=h0 (2)

[0072] Wherein h max , h min are the maximum coordinate value and the minimum coordinate value of the corresponding height of the flow channel; in this embodiment, the standard height is selected as 25 μm.

[0073] As shown in Figure 9 , it is a parameter point optimization schematic diagram; wherein (a) is an optimized point diagram; (b) is an original point diagram.

[0074] As shown in Figure 9 , it is a parameter point optimization schematic diagram; in order to verify the universality of the full-wave correction, the number of cells flowing through the channel is increased, the flow channel height is changed to 100 μm, and the relationship between the point number and the cell flow channel height is studied, from (a) figure, it can be obtained that there is a more obvious piecewise function between the optimized point number and the flow channel height, that is, when the distance between the cell flow channel position and the electrode is less than the center distance and is located above the center distance, it presents obvious linear relationship respectively, which presents more obvious contrast with the original data point diagram (b). When the electrical parameters of the cells flowing through the flow channel are obtained, the cell flow channel height can be calculated according to the full-wave correction optimization method in the application, so as to obtain the impedance signal diagram after standard normalization (standard normalization refers to the operation of recovering to the saddle wave characteristic), and the full-wave optimization obtained according to the height can make the "saddle" wave diagram disappear due to the height dependence, so as to carry out the next step of electrical property calculation.

[0075] As shown in Figure 10The figure shows the comparison between the original data and the full-wave optimization after eliminating the height dependence; figure (a) is the original data figure; (b) is the full-wave correction optimization figure; (c) is the "saddle" type wave recovery figure. Among them, from figure (a), when the cells flow through the position with relatively polarized channel height, the signal error generated by the cells is larger than the signal error near the center position of the channel, so that the signal amplitude data generates larger error when used as analysis data, and the "saddle" type wave pattern disappears at the position far from the electrode (such as the images at 25 μm, 30 μm, 35 μm). Figure (b) is the correction factor optimization figure. After the correction factor optimization, the optimized image generated by the cells flowing through different channel heights has the characteristics of normalization (that is, the difference of the peak value of the electrical signal is not obvious). The data of each point in the whole waveform from the inlet to the outlet of the cells is divided by the correction factor at the corresponding height to perform the normalization operation. Further, the standard image (referring to the waveform at the standard height / the standard factor at the standard height) is used to recover the feature points, and the electrical signal figure after processing obtained from figure (c) recovers the "saddle type" wave pattern, that is, the standard normalization operation (observe Figure 10 the images at 25 μm, 30 μm, 35 μm in (b) and (c), increase the image feature points thereof), so as to obtain the parameters of the "saddle" type wave segment (the standard height of the standard factor is the position capable of presenting the saddle wave form), realize the full-wave correction, eliminate the influence of the height dependence, and then be used for the electrical property analysis of the cells in the later stage.

[0076] The present application is based on the COMSOL theory simulation data and MATLAB algorithm, and utilizes COMSOL with MATLAB for real-time series combination, so as to realize the full-wave optimization under different heights. The characteristics of the present application are as follows:

[0077] 1) The present application performs the standard normalization processing on the waveforms under different channel heights, so that the algorithm application scene has universality;

[0078] 2) The present application utilizes COMSOL with MATLAB for real-time series combination, can realize the integrated processing of the data acquisition and processing optimization, is not limited by the labor and time, and can efficiently realize the elimination of the height dependence;

[0079] 3) The present application can realize the full-wave optimization, judges the size after the full-wave correction, greatly increases the accuracy of the cell size judgment, and at the same time, retains the original waveform, so as to realize the accurate realization of the cell size detection and counting.

[0080] The COMSOL software microfluidic chip structure modeling adopts a double-difference seven-electrode design, adopts a high-frequency sinusoidal alternating input voltage, and utilizes double-difference signals to weaken noise; when cells flow through the flow channel, the current is hindered by the cells to produce deformation, and the measured impedance amplitude will be in a linear relationship with the size of the biological impedance of the cells; the number of signal peaks is proportional to the number of cells flowing through, and the number of cells is calculated; the cells flowing through the height different from the electric signal present obvious correlation; therefore, the signal of the cells entering the detection channel can be used for analyzing the position characteristics and electrical characteristics of the cells.

[0081] The double-difference seven-electrode structure is adopted, the current difference when the cells flow through the left and right electrodes is used for acquisition, the impedance characteristics of the cells can be effectively characterized, the impedance characteristics and height signal information of the cells are obtained, and the optimization result is obtained through data processing.

[0082] The double-difference seven-electrode structure adopts an electrode width of 10-200 mu m, a spacing of 10-50 mu m, a flow channel size design of 20-2000 mu m in height and 100-2000 mu m in length to make the cells flow through.

[0083] The double-difference seven-electrode wiring selection is composed of a negative excitation voltage, a floating electrode, a ground electrode, a positive excitation voltage, a ground electrode, a floating electrode and a negative excitation voltage, and the specific signal is obtained by the ground electrode difference, the excitation voltage adopts a frequency selection of 100K-40MHz, and the initial phase is 0.

[0084] The COMSOL software and MATLAB software have a serial interface, so that the signal can be acquired in real time, and the COMSOL software can be used to calculate and obtain multi-height data for real-time and rapid acquisition and processing.

[0085] Multi-height position correction is a high-throughput real-time full-wave correction method combining microfluidic technology and impedance detection technology; the characteristic points in the original data are used for elimination of height dependence through data processing and feature point extraction and correction factor calculation, so as to obtain normalized signal data, and then the morphology of the cells can be accurately analyzed, and the throughput and accuracy of the cells flowing through the channel are effectively improved.

[0086] The full-wave correction specifically comprises multiplying and scaling the data after processing corresponding to different heights, that is, normalization operation, standard normalization of signal data of different heights to height data corresponding to a standard factor (one height corresponds to one correction factor, the waveform / correction factor of the whole section from the outlet to the outlet of the cells is normalized, and the normalized data is divided by the standard factor, and all the waveforms of all heights are normalized to the data flowing into the standard height), so that the height dependence is eliminated.

[0087] In the high-throughput (high flow channel height) measurement scene, only the maximum peak of the signal can be used as a reference to determine the number of cells, and the size of the cells cannot be determined, so the measurement result has obvious height dependence, and the algorithm failure caused by height dependence cannot be avoided. In order to eliminate the influence of height dependence under high throughput, the present application uses the generated signal image to propose an algorithm based on the feature points of the electrical impedance signal, different processing methods are proposed for different waveforms to calculate the correction factor, and then full-wave correction is performed, the feature points possessed by all waveforms are used for correction, the full-wave correction can be performed for the high-throughput large-height flow channel, the influence of height dependence is eliminated, and high-throughput real-time detection of cell counting and size detection is realized. The high-throughput range of the present application can be effective in the range of 20-2000 μm, preferably 40-500 μm, and the present application is also applicable to low-throughput conditions, providing a new solution for cell number and size research in low-throughput conditions.

[0088] The unmentioned part of the present application is applicable to the prior art.

Claims

1. A multi-height position correction method based on cell electrical impedance measurement, characterized in that, The correction method is applied to the optimization of electrical impedance signals at different heights, and the correction method comprises the following steps: 1) Obtain the electrical impedance signal by using a double-difference seven-electrode structure, and obtain the real part of the electrical impedance signal and perform Gaussian filtering processing; 2) Obtain the characteristic points generated at different heights according to the function of finding extreme values after Gaussian filtering processing of the data, and the characteristic points are extreme points; 3) Determine whether it is a "saddle" type wave according to the number of characteristic points: determine the waveform shape according to the number of characteristic points, if the number of characteristic points is 8, there is a "saddle" type wave, find the maximum extreme point MAX and the minimum extreme point MIN of the "saddle" type wave, the wave valley B1 of the first saddle wave and the wave peak B2 of the second saddle wave, and directly calculate the correction factor according to formula (1) at this time; If the number of characteristic points is 4, it is a "saddle" type wave disappearance type wave, find the maximum extreme point MAX and the minimum extreme point MIN of the "saddle" type wave disappearance type wave, and there are four extreme points, the first extreme point is recorded as B1 and the fourth extreme point is recorded as B2, and the correction factor is calculated according to formula (1) at this time; Wherein β is the correction factor, B1, B2, MIN and MAX are the extreme point coordinates of the waveform; 4) After determining whether it is a "saddle" type wave, the correction factor of the waveform at different heights is obtained, and the waveform data at different heights is scaled according to the standard factor and the correction factor corresponding to the height, to obtain the corrected full wave image; The calculation process of the standard factor β0 is: selecting the waveforms with "saddle" type waves in different heights, taking the middle value of the height of the cells flowing through the flow channel as the standard height, and taking the correction factor corresponding to the standard height as the standard factor β0; The data of each point in the whole waveform of the cells flowing from the inlet to the outlet is divided by the correction factor corresponding to the height, and then divided by the standard factor, so that all the waveforms at different heights are normalized to the data flowing into the standard height.

2. The multi-height location correction method based on cell electrical impedance measurement according to claim 1, wherein, The number of cells is determined according to the occurrence frequency of the maximum peak in the corrected full wave image.

3. The multi-height location correction method based on cell electrical impedance measurement of claim 1, wherein, The calculation process of the standard factor is: selecting the middle value h0 of the height of the flow channel as the standard height according to formula (2), and taking the correction factor corresponding to the standard height as the standard factor, (h max +h min ) / 2=h0 (2) wherein h max , h min are the maximum and minimum coordinate values of the corresponding height of the flow channel, respectively, and h max is in the range of 20-60 μm.

4. A multi-height position correction system based on cell electrical impedance measurement, characterized in that, It comprises: COMSOL software is used to simulate electrical signals at different flow heights, and the simulated signals are obtained by the cells flowing through the flow channels at different heights; COMSOLwith MATLAB is used for real-time data processing, impedance signal data extraction of the simulated signals, and extraction of the real part of the impedance signal data; MATLAB software is used for digital filtering, characteristic point acquisition, event determination, correction factor calculation and full wave correction to eliminate height dependence of the data transmitted by COMSOLwith MATLAB; The formula for calculating the correction factor is: Wherein β is the correction factor, MIN, MAX are the maximum and minimum extreme points of the waveform; when the waveform is a "saddle" type wave, B1, B2 are the minimum and maximum of the first and second saddle waves, respectively; when the waveform is a "saddle" type wave disappearing wave, B1, B2 are the first and fourth extreme points of the waveform, respectively; The full-wave correction is specifically: selecting waveforms with different heights and "saddle" type waves, taking the middle value of the height of the flow channel of the cells with all "saddle" type waves as the standard height, and the correction factor corresponding to the standard height as the standard factor β0; After obtaining the waveform data corresponding to different heights, the normalized data is obtained by multiplying the correction factor, and then the normalized data of different heights is normalized to the height data corresponding to the standard factor, to realize the standard normalization operation; Divide the data of each point in the whole waveform by the correction factor corresponding to the height to obtain the normalized data, and then divide by the standard factor, to standardize all height waveforms to the inflow data at the standard height, eliminating the height dependence.

5. The multi-height location correction system based on cell electrical impedance measurement of claim 4, wherein, The COMSOL software adopts a double-difference seven-electrode design for microfluidic chip structure modeling, uses a high-frequency sinusoidal alternating input voltage, and uses double-difference signals to weaken noise; when the cells flow through the flow channel, the current is hindered by the cells to produce deformation, and the measured impedance amplitude will be linearly related to the size of the biological impedance of the cells; the number of maximum peaks is proportional to the number of cells flowing through, and the number of cells is calculated; According to the signal of the cells entering the detection channel, the position characteristics and electrical characteristics of the cells are analyzed; A double-difference seven-electrode structure is used to obtain impedance signals by using the current difference when the cells flow through the left and right electrodes, and to obtain the impedance characteristics and height signal information of the cells.

6. The multi-height location correction system based on cell electrical impedance measurement of claim 5, wherein, The double-difference seven-electrode structure adopts an electrode width of 10-200 μm and a spacing of 10-50 μm, and the flow channel size is 20-2000 μm high and 100-2000 μm long; The double-difference seven-electrode wiring is selected to be composed of a negative excitation voltage, a floating electrode, a ground electrode, a positive excitation voltage, a ground electrode, a floating electrode, and a negative excitation voltage, and the specific signal is obtained by the ground electrode difference, and the excitation voltage uses a frequency of 100K-40MHz and an initial phase of 0 sinusoidal alternating current input.

Images