A method for automatic arrangement of cardiomyocyte fusion based on micro-channel

Abstract

The application provides a kind of myocardial cell fusion automatic arrangement method based on micro channel, steps are as follows: initialization operation field, micro channel, myocardial cell and micro tube nozzle are placed in same field of view;Based on current field of view image, myocardial cell detection and micro channel, micro tube nozzle identification are carried out to field of view image photographed by camera using image processing technology, and myocardial cell is numbered;Myocardial cell, micro tube nozzle initial position is clarified, and micro tube nozzle target position when myocardial cell is moved and discharged to micro channel, and micro tube moving path is planned;In order to control myocardial cell position in liquid accompanying suction or discharge micro tube, sliding surface is constructed according to myocardial cell movement model, and sliding mode active disturbance rejection controller is designed to realize the control of stepping motor in the process of micro tube transferring myocardial cell;According to the number of myocardial cells, myocardial cells in field of view are sequentially moved and discharged into micro channel according to planned path through micro tube, and automatic arrangement is completed.

Description

Technical Field

[0001] This invention relates to the field of cell-level micromanipulation, and in particular to an automated method for fusion and arrangement of cardiomyocytes based on microchannels. Background Technology

[0002] The specific arrangement and fusion of cardiomyocytes within microchannels are crucial for studying intercellular communication and constructing myocardial tissue engineering. This process requires precise bulk cell capture without compromising cell viability and the ability to manipulate cell arrangement within microchannels. Traditional cell transport operations rely heavily on manual manipulation by professionals, requiring high technical skills, fine manipulation abilities, and extensive experience. Therefore, to improve this process, scientists are constantly exploring new technologies and methods. For example, Z. Zou et al. [Zou Z, Lee S, Ahn C HA polymer microfluidic chip with interdigitated electrode arrays for simultaneous dielectrophoretic manipulation and impedimetric detection of microparticles[J].IEEE Sensors Journal,2008,8(5):527-535.] fabricated a high-throughput polymer microfluidic biochip and used dielectrophoresis to separate and manipulate microparticles within the chip. However, achieving precise cell arrangement within microchannels using this method presents significant challenges in microchannel network design. G. Gibson et al. [Gibson G, Carberry DM, Whyte G, et al. Holographic assembly workstation for optical manipulation[J]. Journal of Optics A: Pure and Applied Optics, 2008, 10(4): 044009-044009.] used optical tweezers to fix and arrange cells; however, the intensity of the optical tweezers needs to be precisely controlled to avoid potential damage to the cells due to excessive intensity. Z. Lu et al. [Lu Z, Moraes C, Ye G, et al. Single cell deposition and patterning with a robotic system[J]. PloS One, 2010, 5(10): e13542] developed a visual feedback-based cell sorting system that uses micropipettes smaller than the cell diameter for cell manipulation. However, because this operation does not completely draw the entire cell into the micropipette, cell deformation may occur during cell holding, or the cell may become stuck in the micropipette pores during release, causing release failure. Therefore, for the manipulation of cardiomyocytes in the microchannels, it is necessary to develop a technology that can achieve precise cell capture and arrangement without damaging cell viability. Summary of the Invention

[0003] This invention addresses the shortcomings of existing technologies by providing an automated arrangement method for cardiomyocyte fusion based on microchannels. This method enables precise arrangement and fusion of cardiomyocytes within microchannels and is applicable to myocardial tissue engineering and intercellular communication research.

[0004] This invention achieves this objective through the following technical solution:

[0005] An automated method for fusion and arrangement of cardiomyocytes based on microchannels, comprising the following steps:

[0006] a. Place the culture dish containing the microchannel substrate and the cardiomyocyte suspension on the motorized stage of the micromanipulation system. The motorized micromanipulation arm controls the position of the microtube. The microtube is held by the needle holder. The inner diameter of the microtube is not less than the diameter of the cardiomyocyte. Initialize the field of view and place the microchannel, cardiomyocyte and microtube opening in the same field of view.

[0007] b. Based on the current field of view image, use image processing technology to detect myocardial cells and identify microchannels and microtubule openings in the field of view image captured by the high-speed industrial camera, and number the myocardial cells;

[0008] c. Determine the initial position of the cardiomyocytes and the microtube opening, as well as the target position of the microtube opening when the cardiomyocytes move through the microtubes and are discharged into the microchannels. Plan the path from the initial position of the microtube opening to the initial position of the cardiomyocytes and the path from the microtube opening after the cardiomyocytes are aspirated to the target position of the microtube opening when the cardiomyocytes are discharged into the microchannels. Based on the planned path, set the control law to control the movement of the electric stage or the electric micromanipulation arm that holds the microtubes.

[0009] d. The microtube is connected to the pneumatic injector via a catheter. The stepper motor connected to the lead screw of the pneumatic injector is controlled to control the position of cardiomyocytes in the liquid that is inhaled or exhaled from the microtube. A sliding surface is constructed based on the cardiomyocyte movement model, and a sliding mode active disturbance rejection controller is designed to control the stepper motor during the transfer of cardiomyocytes via the microtube.

[0010] e. Based on the cardiomyocyte number, the cardiomyocytes in the field of view are sequentially arranged into the microchannels by moving the microtubes to the target location according to the planned path, thus completing the automated arrangement.

[0011] Furthermore, in step a, the specific operation is as follows: in the initial state, adjust the position of the manually controlled robotic arm and the stage so that the microchannel is vertical and located on one side of the field of view, and the microtube is horizontal and located on the other side of the field of view with the microtube opening facing the direction of the microchannel, and adjust the focal plane to the cell plane.

[0012] Further, in step b, a target neighborhood point boundary tracking method based on binary images is used to detect cell contours and extract cell center points. The specific steps are as follows: binarize the grayscale image of the current field of view; traverse the image pixels and calculate the standard deviation of the pixels in the 3×3 neighborhood of each pixel. When the standard deviation is greater than 0.3, it is considered a cell region, and the pixel value of the corresponding pixel is set to 255, while the values ​​of the remaining pixels are all set to 0; perform morphological operations of closing and opening operations on the image; detect contours and count them; extract the contours in the image and calculate the contour area. When the contour area is between 10-30μm, the corresponding contour is considered a cell contour. The center point of the least squares fitted ellipse is taken as the cell center point, and half the width of the outer rectangle of the cell contour is taken as the cell radius; sort the cells detected in the field of view; and record the information of each cell after sorting.

[0013] Furthermore, the cardiomyocytes detected in the field of view are sorted vertically from bottom to top in the microchannels. If two adjacent cardiomyocytes are less than the width of the microchannel in the vertical direction, the cardiomyocytes closer to the microchannel are sorted first, and the cardiomyocytes farther away from the microchannel are sorted later. The information of each cell after sorting is recorded.

[0014] Furthermore, in step b, a line detection algorithm based on Hough transform is used to identify the microchannel location. The specific steps are as follows: edge detection is performed on the grayscale image of the current field of view to obtain the edge image; line detection is performed using probabilistic Hough transform, and the slope of the line is calculated; the detected lines are filtered, and the average slope of the two contour lines of the detected microchannel is calculated as the slope of the microchannel in the field of view; the target position of all myocardial cells to be arranged is calculated based on the starting point of the target position of myocardial cell arrangement and the slope of the microchannel.

[0015] Furthermore, the target position planning of the microtube opening in step c is as follows: taking the bottom endpoint of the microchannel contour line identified near the center of the field of view as the reference, the numbered cardiomyocytes are arranged sequentially from bottom to top, so that the center point of the cardiomyocyte is located on the microchannel contour line. The coordinates of the center point of each cardiomyocyte on the contour line are used as the target position of the center position of the outermost diameter contour line of the microtube opening after the cardiomyocyte is inhaled.

[0016] Furthermore, the automatic control method described in step d for cell localization within the microtubule specifically employs sliding mode active disturbance rejection control based on an expansion state observer to maintain cardiomyocytes at the microtubule opening. Considering the interactions between the motor, air pressure, fluid, and cardiomyocytes, and taking the cardiomyocytes as the control target, a dynamic model of the cardiomyocytes is established as follows:

[0017]

[0018] Where x1 is the cell position, x2 is the cell movement velocity, u is the motor speed (i.e., the system control input), A1 is the syringe cross-sectional area, A2 is the microtube cross-sectional area, P0 is the initial chamber pressure of the system's pneumatic syringe, V0 is the initial gas volume when the syringe piston is stationary, m is the cell mass, r is the cell radius, and μ is the dynamic viscosity coefficient. The instantaneous rate of change of pressure in the closed air chamber is represented by the disturbance term f(·), which is a lumped uncertainty, including the uncertainty of model parameters, unmodeled dynamics, the hysteresis effect of the closed air pressure, and other disturbances.

[0019] Let θ1 = 6πμr / m, θ2 = V0 / P0A2, b = A1 / A2, the second formula above can be written as:

[0020]

[0021] Let b0 = bθ1, The above formula can be written as:

[0022]

[0023] The system control model can now be written in the following form, where F represents the total disturbance of the system, and h represents the derivative of the total disturbance:

[0024]

[0025] in It represents the differential of the total system disturbance F, that is, the rate of change of the disturbance with time;

[0026] A linear extended state observer based on a second-order system is used to estimate the state variables, where It is x = [x1, x2, x3] T The estimated value:

[0027]

[0028] Where β1, β2, and β3 are the observer gain parameters;

[0029] The characteristic polynomial of the observer is:

[0030] λ(s)=s 3 +β1s 2 +β2s+β3

[0031] Define β1 = 3w0, β2 = 3w0 2 β3=w0 3 Then the above characteristic polynomial is equal to (s+w0). 3 The pole is configured at -w0, where w0 is the bandwidth parameter of the extended state observer;

[0032] In this control system, x1 represents the actual position of the cell, x d The desired trajectory of the cell is represented by the positional error defined as:

[0033] e = x1 - x d

[0034] Define the state estimation error as:

[0035]

[0036] The sliding surface is defined as follows:

[0037]

[0038] For the aforementioned sliding surface, a sliding mode active disturbance rejection controller can be designed to make the cell position tracking error converge to zero:

[0039]

[0040] Where k p η is the control gain, β1 and β2 are the observer gains, sat(σ) is the saturation function, β represents the boundary layer thickness, and β > 0.

[0041] Furthermore, the specific control methods and steps for the localization control of cardiomyocytes within microtubules are as follows:

[0042] Step S1: Microtube detection: Extract the microtube opening template, magnify its area as the region of interest for tracking, and set it as the background image B1(x,y);

[0043] Step S2: Obtain the current frame's field of view image F(x,y), and set the target image X. N For pixels (x, y) where all are 0, the current frame image is subtracted within the ROI using the background image. The result is then thresholded and assigned to the target image to identify the position of the moving object. Before the next frame arrives, the background is updated to improve robustness. The background update method is as follows:

[0044] B N+1 (x,y)=αF(x,y)+(1-α)B N (x,y),0≤α≤1

[0045] Where α is the background update weight;

[0046] The subtraction operation method is as follows:

[0047]

[0048] Where T is the threshold for the subtraction operation;

[0049] Step S3: Calculate the current cell position and the desired position P. c If the error is less than a certain threshold, control is terminated; otherwise, the microtube positioning controller output is calculated, and the output signal is transmitted to the motor to control the target to move.

[0050] Step S4: Update the error information, proceed to the next frame, and return to step S2.

[0051] Compared with the prior art, the beneficial effects of this invention are as follows:

[0052] This invention realizes an automated method for the fusion and arrangement of cardiomyocytes based on microchannels. First, image processing is used to solve the key localization technologies for cell detection and microchannel identification. Then, based on automatic control technology, the key control technologies for stage control and cell positioning control within microtubes are solved. Finally, an automated arrangement process for cardiomyocytes based on microchannels is designed to achieve automated arrangement and fusion growth of cardiomyocytes within microchannels. This invention uses micropipettes larger than the cell diameter to manipulate cells, avoiding damage caused by cell deformation during cell capture. It presents key technologies based on image processing and automatic control, including cell detection and microchannel identification, as well as stage control and cell positioning control within microtubes. This achieves an automated operation process for the directional arrangement and fusion of cardiomyocytes within microchannels, enabling precise cell capture while avoiding damage caused by cell deformation during capture, and precisely controlling cell arrangement within microchannels to achieve directional arrangement and fusion of cardiomyocytes within microchannels. This automated method can be used in the directional arrangement and fusion experiments of other somatic cells, and can also be used for the automated and precise construction of myocardial tissue engineering and research on intercellular communication. Attached Figure Description

[0053] Figure 1 This is a flowchart of the method of the present invention;

[0054] Figure 2 It is a microoperating system used in experiments on the automated arrangement and fusion of cardiomyocytes based on microchannels;

[0055] Figure 3 This is a diagram illustrating the cell detection process of the present invention;

[0056] Figure 4 This is a diagram illustrating the microchannel identification process of the present invention;

[0057] Figure 5 This is a diagram illustrating the motion control results of the stage;

[0058] Figure 6 This is a diagram illustrating the results of cell localization control within microtubules;

[0059] Figure 7 It is a flowchart of cardiomyocyte arrangement based on microchannels;

[0060] Figure 8 This is a diagram illustrating the arrangement and fusion of cardiomyocytes. Detailed Implementation

[0061] Exemplary embodiments of the present invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present invention and to fully convey the scope of the invention to those skilled in the art. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0062] See Figures 1 to 8 This invention discloses an embodiment of an automated arrangement method for cardiomyocyte fusion based on microchannels. This embodiment uses the method to evaluate the arrangement and fusion of cardiomyocytes in microchannels.

[0063] The flowchart of the automated arrangement method for cardiomyocyte fusion based on microchannels of this invention is as follows: Figure 1 As shown.

[0064] The cells used in this example were primary cardiomyocytes extracted from SD suckling mice aged 1-3 days after birth. The method for obtaining the cardiomyocytes is as follows:

[0065] SD suckling mice were ordered from an animal testing center. The mice were anesthetized by immersion in 75% alcohol. Using large forceps in the left hand, the skin on the back of the mouse's neck was firmly pinched to fully expose the chest. Using a pair of ophthalmic curved scissors in the right hand, the skin was cut open, and the ribs were cut upwards along the lower left edge of the sternal manubrium. Then, a transverse cut was made in the middle of the incision along the sternum. Next, using ophthalmic straight forceps in the left hand and ophthalmic straight scissors in the right, the chest cavity was gently pushed open until the heart beat. The ophthalmic straight forceps were used to grasp the heart, and the ventricular portion was cut directly from the middle of the heart using the ophthalmic straight scissors and placed in an ice bath of D-Hank's solution. This process was repeated until all samples were collected. Using ophthalmic straight forceps in both hands, blood clots and fibrous tissue around the heart in the culture dish were removed, and the heart was placed in a fresh ice bath of D-Hank's solution for rinsing. After rinsing, the D-Hank's solution was removed, and the heart tissue was minced to approximately 1 mm using ophthalmic straight scissors. 3For the heart sludge, add 4 ml of trypsin (0.25%) to small pieces of sludge and transfer them together with the sludge to a 15 ml centrifuge tube. Add trypsin to bring the total volume to 5 ml. Place the centrifuge tube in a 37°C water bath, shake for 5 minutes, and then let it stand for 5 minutes. After standing, carefully aspirate the supernatant from the tube using a pipette and transfer it to a new centrifuge tube. Add an equal volume of complete culture medium to the new centrifuge tube to stop digestion. Add the aspirated trypsin to the centrifuge tube containing the heart sludge to bring the total volume to 5 ml. Repeat the shaking, standing, and supernatant aspiration steps 7-8 times until the heart sludge is observed to be completely digested. After digestion, discard the supernatant from the first digestion and collect all the supernatant in a 50 ml centrifuge tube. Centrifuge at 1000 rpm for 6 minutes. Discard the supernatant and resuspend the cells in 10 ml of complete culture medium containing 20% ​​serum. Filter the cell suspension through a 20 μm cell filter to obtain a cell suspension containing cardiomyocytes, fibroblasts, and endothelial cells. Seed this suspension into 75 cm⁻¹ cells. 2 The cells were placed in a plastic culture flask and incubated at 37°C and 5% CO2 concentration for 2 hours. After that, the cells were removed, the adherent cells were discarded, and the non-adherent cell suspension was taken out. The cells were repeatedly aspirated with a pipette to mix them evenly in the culture medium, thus obtaining a cardiomyocyte suspension for cell arrangement.

[0066] The microscopic operating system used in the implementation adopts Figure 2 The NK-MR901 micro-operating system shown is based on a standard inverted microscope (Eclipse TI-E, Nikon, Japan) and is equipped with a high-speed industrial camera (ACA640-120GM / GC, Basler, Germany) to acquire microscopic grayscale images with a resolution of 640×480 pixels and a frame rate of 50 frames / s. In addition, the system includes a motorized stage (ProScan III, Prior, UK) for placing culture dishes; two three-DOF motorized micromanipulation arms (MP285, Sutter, USA) for controlling the microtube position; a needle holder (HI-7, Narishige, Japan) for holding the microtube; a pneumatic syringe (Narishige, IM-11B, Japan) connected to the microtube via a catheter; a stepper motor (Sanyo, 103H546-0410, Japan) connected to the pneumatic syringe lead screw via a coupling; a syringe motor controller (Vince, VSMD101_025T, China) for driving the stepper motor; and an industrial computer connected to the computer and multiple devices on the worktable for motion control and processing of microscopic images.

[0067] In the experiment, the microtubes used had an inner diameter of 20 μm, and the microchannels were grooves 20 μm wide and 20 μm deep on polydimethylsiloxane plates. The cells used were suspended cardiomyocytes, approximately 10-20 μm in diameter, freshly digested and extracted from the hearts of suckling rats. The experiment used the right manipulator to move the microtubes, and the movement of the cells and microchannels was controlled by the movement of the stage. Before the experiment, image detection was performed to detect the microtubes, cells, and microchannels separately to obtain positional information. Then, the stage was used to control the transport of cells and microchannels, and cell picking and placement were controlled by cell positioning within the microtubes. The cells were cyclically manipulated until a predetermined number of cells were arranged.

[0068] The present invention will be further described below with reference to the accompanying drawings. The specific implementation of the method provided by the present invention is as follows:

[0069] Step 1: Initialize the operating field of view. In the initial state, manually control the robotic arm and stage to move them to the appropriate positions so that the microtubes are located in the lower right part of the field of view, roughly horizontal, with multiple cells present, and the microchannels are located in the left half of the field of view, roughly vertical. Adjust the focal plane to the cell plane.

[0070] Second, based on the current field of view image, image processing technology is used to detect the location of cardiomyocytes and microchannels.

[0071] Image processing techniques are used to detect cells and microchannels. Specifically, for the detection and sorting of cardiomyocytes, a target neighborhood point boundary tracking method based on binary images is considered to detect cell boundaries and extract cell contours. The implementation process is as follows:

[0072] Step S1: Binarize the grayscale image of the current field of view. Traverse the image pixels and calculate the standard deviation of the pixels in the 3×3 neighborhood of each pixel. When the calculated standard deviation is greater than 0.3, it is considered a cell region, and the pixel value of the corresponding pixel is set to 255. The values ​​of the remaining pixels are all set to 0.

[0073] Step S2: Perform morphological operations such as closing and opening operations on the image.

[0074] Step S3: Detect and count contours. Extract contours from the image and calculate the contour area. When the contour area is between 10-30 μm, the corresponding contour is considered a cell contour. The center point of the least squares fitted ellipse is taken as the cell center point, and half the width of the outer rectangle of the cell contour is taken as the cell radius.

[0075] Step S4: Sort the cells detected in the field of view. To avoid collisions between microtubules and unsorted cells during stage movement, sort the cells in order from bottom to top along the y-axis. If two adjacent cells are less than the width of the microtubule in the y-axis direction, select the cardiomyocyte closer to the microchannel and place it first. Record the information of each cell after sorting.

[0076] 2.2 The process of identifying microchannels in the current field of view and calculating the target position of cell arrangement is as follows:

[0077] Step S1: Perform edge detection on the grayscale image of the current field of view to obtain the edge image.

[0078] Step S2: Use the probabilistic Hough transform to detect the line and calculate the slope of the line.

[0079] Step S3: Filter the detected lines. To simplify subsequent operations, set the slope of vertical lines to 0 and remove lines with an absolute slope value less than 1.

[0080] Step S4: Calculate the average slope of the remaining straight lines, and use this as the slope of the microchannel in the field of view. If there is a straight line with a slope of 0, then the slope of the microchannel is also set to 0. Select the lower endpoint of the rightmost straight line among the remaining lines as the lower right corner of the microchannel, which is the starting point of the target cell arrangement position.

[0081] Step S5: Based on the target starting point and the slope of the microchannels, calculate the target positions for all cells to be arranged, i.e., the positions where the cells are transported from the microtube openings, in order from bottom to top along the y-axis. Where p i Let l be the i-th target point, and p be the i-th target point. i The distance between the lower endpoint of the groove on the right side of the microchannel in the field of view and the straight line, where k is the slope of the microchannel.

[0082]

[0083] p ix =p 0x p iy =p 0y -il, k=0

[0084] p ix and p iy For p i The x and y coordinates of the point, p 0x and p 0y The coordinates of the lower endpoint of the groove on the right side of the microchannel along the outline line are used as a reference point for the target position of cell arrangement.

[0085] Based on the current field of view image, the location of cardiomyocytes in the field of view is detected and obtained by using a target neighborhood point tracking boundary tracing method based on binary images. Figure 3 The cell detection results were presented, among which... Figure 3 (a) Original image of the cell, Figure 3 (b) is the original image after binarization and morphological operations. Figure 3 (c) shows the cell contour extraction results. Figure 3 (d) shows the cell sorting results. Based on the current field-of-view image, the position of the microchannels in the field of view is detected and obtained using a line detection algorithm based on Hough transform. Figure 4 The microchannel detection results were presented, among which... Figure 4 (a) is the original image of the microchannel. Figure 4 (b) shows the edge detection results. Figure 4 (c) shows the results of the line detection. Figure 4 (d) represents the target sorting result.

[0086] Third, based on the position information of cells and microchannels, a proportional-integral-derivative (PID) controller and a sliding mode active disturbance rejection control based on an extended state observer are used for the motion of the stage and the positioning of cells in the microtubes, respectively. Figure 5 and Figure 6 The results of stage motion control and microtubule intracellular cell localization control are presented, among which... Figure 5 (a) is a comparison diagram of the cell movement trajectory along the x-axis and the expected trajectory. Figure 5 (b) is a comparison diagram of the cell movement trajectory along the y-axis and the expected trajectory.

[0087] Based on the positional information of cells and microchannels, an automatic control system is designed to control the stage and the positioning of cells within the microtubes.

[0088] Based on the obtained cell and microchannel position information data, an automatic control system is designed to control the movement of the stage to transport the cells or microchannels to the microtube opening, and to control the positioning of the cells inside the microtube at the microtube opening. The microtube opening coordinates are the coordinates of the center point of the leftmost opening diameter contour line identified after microtube detection.

[0089] For the motion control of the stage, a proportional-integral-derivative (PID) controller is used. It calculates the deviation e(t) based on the given value and the actual output value, and then converts the proportional (P), integral (I), and derivative (D) of the deviation into three control parameters K. p K i and K d After linear combination, a control variable u(t) is formed to control the controlled object. Its control law is as follows:

[0090]

[0091] The process of motion control for the stage or micromanipulation arm is as follows:

[0092] Step S1: The coordinates of a myocardial cell in the pixel coordinate system of the visual field image are P. c In the pixel coordinate system, the coordinates of the microtube opening are P. p The objective lens calibration factor is C, used for the conversion between the pixel coordinate system and the world coordinate system. Since microchannel recognition only needs to be performed once, while cell recognition may be performed in multiple fields of view, when switching fields of view, as the stage moves, the pixel coordinate system of the new field of view image will correspond to the new world coordinate system coordinates. The cardiomyocytes to be arranged are positioned at the microtube opening by horizontally moving the stage or micromanipulation arm.

[0093] Step S2: Obtain the current position information of the stage and calculate the error between the target position and the current position of the stage. If the error is less than a certain set value, exit the control thread; otherwise, proceed to the next step.

[0094] Step S3: Input the error information into the proportional-integral controller, calculate the control output, i.e. the movement speed of the stage in the x and y directions, and transmit it to the stage so that it moves according to the speed.

[0095] Step S4: Update the error information, including the integral of the error, return to step S2, and obtain the current stage position information.

[0096] For the localization control of cells within microtubules, a sliding mode active disturbance rejection control based on an expansion state observer is considered. A pneumatic injector system is used to control cell localization. In this system, a stepper motor is rigidly connected to the injector piston via a lead screw and guide rail. The forward and reverse rotation of the motor drives the piston to move back and forth, causing changes in the pressure within the injector. The gas-liquid interface within the microtubule also changes with the pressure, thus causing the cells to move with the fluid. In this system, the cell position obtained through visual feedback is used as the system output, and the stepper motor speed is used as the system input. Considering the interactions between the motor, air pressure, fluid, and cells, the cell is taken as the control target, and a cell dynamic model is established as follows:

[0097]

[0098] Where x1 is the cell position, x2 is the cell movement velocity, u is the motor speed (i.e., system control input), A1 is the cross-sectional area of ​​the syringe, A2 is the cross-sectional area of ​​the microtube, P0 is the initial cavity pressure of the system's pneumatic syringe, V0 is the initial gas volume when the syringe piston is stationary, m is the cell mass, r is the cell radius, and μ is the dynamic viscosity coefficient. The instantaneous rate of change of pressure in the closed air chamber is represented by the disturbance term f(·), which is a lumped uncertainty, including the uncertainty of model parameters, unmodeled dynamics, the hysteresis effect of the closed air pressure, and other disturbances.

[0099] Let θ1 = 6πμr / m, θ2 = V0 / P0A2, b = A1 / A2, the second formula above can be written as:

[0100]

[0101] Let b0 = bθ1, The above formula can be written as:

[0102]

[0103] The system control model can now be written in the following form, where F represents the total disturbance of the system, and h represents the derivative of the total disturbance:

[0104]

[0105] in It represents the differential of the total system disturbance F, that is, the rate of change of the disturbance over time.

[0106] Extended state observers are a crucial component of active disturbance rejection control (ADRC). They estimate disturbances caused by model inaccuracies, external interference, and internal coupling, enabling real-time compensation. This paper considers using a linear extended state observer based on a second-order system to estimate the state variables, where... It is x = [x1, x2, x3] T The estimated values ​​are: x1 represents the actual position of the cell, x2 represents the cell velocity, and x3 represents the integral state of the total system disturbance F.

[0107]

[0108] Where β1, β2, and β3 are observer gain parameters used to adjust the convergence speed of the observer.

[0109] The characteristic polynomial of the observer is:

[0110] λ(s)=s 3 +β1s 2 +β2s+β3

[0111] Define β1 = 3w0, β2 = 3w0 2 β3=w0 3 Then the above characteristic polynomial is equal to (s+w0). 3The pole is configured at -w0. w0 is the bandwidth parameter of the extended state observer; w0>0, this parameter does not affect the structure of the sliding surface. Adjusting w0 enables the extended state observer to estimate quickly and accurately, and then adjusting the sliding mode parameter optimizes the tracking performance.

[0112] In this control system, w1 represents the actual position of the cell, x d The desired trajectory of the cell is represented by the positional error defined as:

[0113] e = x1 - x d

[0114] Define the state estimation error as:

[0115]

[0116] The sliding surface is defined as follows:

[0117]

[0118] For the aforementioned sliding surface, a sliding mode active disturbance rejection controller can be designed to make the cell position tracking error converge to zero:

[0119]

[0120] Where k p To control the gain, η is the switching gain, β1 and β2 are the observer gains, sat(σ) is the saturation function, β represents the boundary layer thickness, and β > 0. The optimal value of β is determined through experimental debugging.

[0121] The above describes the specific control method for the localization control of cells within microtubules. The following are the implementation steps:

[0122] Step S1: Perform microtube detection. Extract the microtube opening template, magnify its area as the region of interest (ROI) for tracking, and set it as the background image B1(x,y).

[0123] Step S2: Obtain the current frame's field of view image F(x,y), and set the target image X. N The (x, y) pixels are all 0. The current frame image is subtracted within the ROI using the background image, and the result, after thresholding, is assigned to the target image to identify the position of the moving object. Before the next frame arrives, the background is updated to improve robustness. The background update method is as follows:

[0124] B N+1 (x,y)=αF(x,y)+(1-α)B N (x,y),0≤α≤1

[0125] Where α is the background update weight. The subtraction operation method is as follows, where T is the subtraction operation threshold.

[0126]

[0127] Step S3: Calculate the error between the current position and the desired position of the cell. If it is less than a certain threshold, exit control. Otherwise, calculate the output of the microtube positioning controller and transmit the output signal to the motor to control the target to move.

[0128] Step S4: Update the error information, proceed to the next frame, and return to step S2.

[0129] 4. The specific arrangement process is as follows: Figure 7 As shown, microtubule detection and microchannel identification are first performed to obtain the coordinates of the microtubule opening and the lower endpoint coordinates of the right side groove of the microchannel along the contour line, which serve as reference coordinates for the target position. The stage position information is then acquired, and the cardiomyocytes within the microscopic field of view are numbered and their target positions planned. The microtubules or stage are controlled to position the cardiomyocytes at the microtubule opening. After the microtubule draws in a cell, it moves the opening to the target position of that cardiomyocyte, aligning the microtubule opening coordinates with the target position coordinates of the cardiomyocyte. The microtubule then ejects the cell into the microchannel, moving the cells sequentially according to their numbering. After switching fields of view, the stage returns to its origin after each cell movement, becoming the stage origin for the new field of view.

[0130] 5. The results of cardiomyocyte arrangement and fusion are as follows: Figure 8 As shown, where Figure 8 (a) shows an image of cardiomyocytes arranged sequentially into microchannels according to their cardiomyocyte number. Figure 8 (b) Image of cells fused within the microchannels. Experimental results show that the method of the present invention can achieve precise arrangement and fusion of cardiomyocytes within the microchannels.

[0131] The present invention has been described in detail above through embodiments, but the content described is only an exemplary embodiment of the present invention and should not be considered as limiting the scope of the present invention. The scope of protection of the present invention is defined by the claims. Any technical solutions designed by those skilled in the art using the technical solutions described in the present invention, or similar technical solutions designed by those skilled in the art under the inspiration of the technical solutions of the present invention, within the substance and scope of protection of the present invention, to achieve the above-mentioned technical effects, or equivalent changes and improvements made to the scope of the application, should still fall within the patent protection scope of the present invention. It should be noted that, for clarity, descriptions of some components and processes that are not directly and obviously related to the scope of protection of the present invention but are known to those skilled in the art have been omitted in the description of the present invention.

Claims

1. An automated method for fusion and arrangement of cardiomyocytes based on microchannels, characterized in that, The method includes the following steps: a. Place the culture dish containing the microchannel substrate and the cardiomyocyte suspension on the motorized stage of the micromanipulation system. The motorized micromanipulation arm controls the position of the microtube. The microtube is held by the needle holder. The inner diameter of the microtube is not less than the diameter of the cardiomyocyte. Initialize the field of view and place the microchannel, cardiomyocyte and microtube opening in the same field of view. b. Based on the current field of view image, use image processing technology to detect myocardial cells and identify microchannels and microtubule openings in the field of view image captured by the high-speed industrial camera, and number the myocardial cells; c. Determine the initial position of the cardiomyocytes and the microtube opening, as well as the target position of the microtube opening when the cardiomyocytes move through the microtubes and are discharged into the microchannels. Plan the path from the initial position of the microtube opening to the initial position of the cardiomyocytes and the path from the microtube opening after the cardiomyocytes are aspirated to the target position of the microtube opening when the cardiomyocytes are discharged into the microchannels. Based on the planned path, set the control law to control the movement of the electric stage or the electric micromanipulation arm that holds the microtubes. d. The microtube is connected to the pneumatic injector via a catheter. The stepper motor connected to the lead screw of the pneumatic injector is controlled to control the position of cardiomyocytes in the liquid that is inhaled or exhaled from the microtube. A sliding surface is constructed based on the cardiomyocyte movement model, and a sliding mode active disturbance rejection controller is designed to control the stepper motor during the transfer of cardiomyocytes via the microtube. e. Based on the cardiomyocyte number, the cardiomyocytes in the field of view are sequentially arranged into the microchannels by moving the microtubes to the target location according to the planned path, thus completing the automated arrangement.

2. The method according to claim 1, characterized in that, The specific operation in step a is as follows: In the initial state, adjust the position of the manually controlled robotic arm and the stage so that the microchannel is vertical and located on one side of the field of view, and the microtube is horizontal and located on the other side of the field of view with the microtube opening facing the direction of the microchannel, and adjust the focal plane to the cell plane.

3. The method according to claim 1, characterized in that, Step b employs a target neighborhood point boundary tracking method based on binary images to detect cell contours and extract cell center points. The specific steps are as follows: Binarize the grayscale image of the current field of view; traverse the image pixels, calculate the standard deviation of pixels within a 3×3 neighborhood of each pixel, and consider any pixel with a standard deviation greater than 0.3 as a cell region, setting the corresponding pixel value to 255 and setting the values ​​of the remaining pixels to 0; perform morphological operations (closing and opening) on ​​the image; detect and count contours; extract contours from the image, calculate the contour area, and consider any contour area between 10-30 μm as a cell contour, using the center point of the least-squares fitted ellipse as the cell center point and half the width of the bounding rectangle of the cell contour as the cell radius; sort the detected cells in the field of view; and record the information of each cell after sorting.

4. The method according to claim 1, characterized in that, The cardiomyocytes detected in the field of view are sorted vertically from bottom to top in the microchannels. If two adjacent cardiomyocytes are less than the width of the microchannel in the vertical direction, the cardiomyocytes closer to the microchannel are sorted first, and the cardiomyocytes farther away from the microchannel are sorted later. The information of each cell after sorting is recorded.

5. The method according to claim 1, characterized in that, In step b, a line detection algorithm based on Hough transform is used to identify the microchannel location. The specific steps are as follows: edge detection is performed on the grayscale image of the current field of view to obtain the edge image; line detection is performed using probabilistic Hough transform, and the slope of the line is calculated; the detected lines are filtered, and the average slope of the two contour lines of the detected microchannel is calculated as the slope of the microchannel in the field of view; the target position of all myocardial cells to be arranged is calculated based on the starting point of the target position of myocardial cell arrangement and the slope of the microchannel.

6. The method according to claim 1, characterized in that, In step c, the target position of the microtube opening is planned as follows: Taking the bottom endpoint of the microchannel contour line identified near the center of the field of view as the reference, the numbered cardiomyocytes are arranged sequentially from bottom to top, so that the center point of the cardiomyocyte is located on the microchannel contour line. The coordinates of the center point of each cardiomyocyte on the contour line are used as the target position of the center position of the outermost diameter contour line of the microtube opening after the cardiomyocyte is inhaled.

7. The method according to claim 1, characterized in that, In step d, the automatic control method for cell localization within the microtubule specifically employs sliding mode active disturbance rejection control based on an expansion state observer. This maintains the cardiomyocytes at the microtubule opening. Considering the interactions between the motor, air pressure, fluid, and cardiomyocytes, and taking the cardiomyocytes as the control target, a dynamic model of the cardiomyocytes is established as follows: Where x1 is the cell position, x2 is the cell movement velocity, u is the motor speed (i.e., the system control input), A1 is the syringe cross-sectional area, A2 is the microtube cross-sectional area, P0 is the initial chamber pressure of the system's pneumatic syringe, V0 is the initial gas volume when the syringe piston is stationary, m is the cell mass, r is the cell radius, and μ is the dynamic viscosity coefficient. The instantaneous rate of change of pressure in the closed air chamber is represented by the disturbance term f(·), which is a lumped uncertainty, including the uncertainty of model parameters, unmodeled dynamics, the hysteresis effect of the closed air pressure, and other disturbances. Let θ1 = 6πμr / m, θ2 = V0 / P0A2, b = A1 / A2, the second formula above can be written as: Let b0 = bθ1, The above formula can be written as: The system control model can now be written in the following form, where F represents the total disturbance of the system, and h represents the derivative of the total disturbance: in It represents the differential of the total system disturbance F, that is, the rate of change of the disturbance with time; A linear extended state observer based on a second-order system is used to estimate the state variables, where It is x = [x1, x2, x3] T The estimated value: Where β1, β2, and β3 are the observer gain parameters; The characteristic polynomial of the observer is: λ(s)=s 3 +β1s 2 +β2s+β3 Define β1 = 3w0, β2 = 3w0 2 β3=w0 3 Then the above characteristic polynomial is equal to (s+w0). 3 The pole is configured at -w0, where w0 is the bandwidth parameter of the extended state observer; In this control system, x1 represents the actual position of the cell, x d The desired trajectory of the cell is represented by the positional error defined as: and=x1-x d Define the state estimation error as: The sliding surface is defined as follows: For the aforementioned sliding surface, a sliding mode active disturbance rejection controller can be designed to make the cell position tracking error converge to zero: Where k p η is the control gain, β1 and β2 are the observer gains, sat(σ) is the saturation function, β represents the boundary layer thickness, and β > 0.

8. The method according to claim 7, characterized in that, The specific control method and steps for the localization control of cardiomyocytes within microtubules are as follows: Step S1: Microtubule detection: Extract the microtubule opening template, magnify its area as the region of interest for tracking, and set it as the background image B1(x,y); Step S2: Obtain the current frame's field of view image F(x,y), and set the target image X. N For pixels (x, y) where all are 0, the current frame image is subtracted within the ROI using the background image. The result is then thresholded and assigned to the target image to identify the position of the moving object. Before the next frame arrives, the background is updated to improve robustness. The background update method is as follows: B N+1 (x,y)=αF(x,y)+(1-α)B N (x,y),0≤α≤1 Where α is the background update weight; The subtraction operation method is as follows: Where T is the threshold for the subtraction operation; Step S3: Calculate the current cell position and the desired position P. c If the error is less than a certain threshold, control is terminated; otherwise, the microtube positioning controller output is calculated, and the output signal is transmitted to the motor to control the target to move. Step S4: Update the error information, proceed to the next frame, and return to step S2.

Images