Bubble correction method and device, blood cell analyzer and storage medium

Abstract

The application provides a bubble correction method and device, a blood cell analyzer and a storage medium, and relates to the technical field of blood detection. The liquid volume calibration method comprises the following steps: acquiring a target image; pre-processing the target image to obtain a pre-processed image; detecting whether bubbles exist in the pre-processed image; if yes, compensating for the total volume V1 of the reagent to obtain the actual volume V0 of the reagent; and when bubbles exist in the reagent, identifying the bubbles in the reagent in the TIP head in the target image and compensating for the volume of the reagent to obtain the actual volume of the reagent. That is, the actual volume of the reagent is corrected by using the method of machine vision, without the need to specially set up a hardware structure to remove bubbles, so that the use cost is low and the control logic is simple. Moreover, the accurate detection result can be directly obtained by compensation, without the need for the user to remove the bubbles and then perform detection again, so that the operation is simple.

Description

Technical Field

[0001] This invention relates to the field of blood testing technology, and in particular to a bubble correction method, device, blood cell analyzer, and storage medium. Background Technology

[0002] In routine blood tests, blood samples need to be diluted and lysed sequentially to prepare the test sample. During preparation, diluent and lysing agent are quantitatively transferred using a pipette pump. In this process, the pipette pump is prone to generating air bubbles. These air bubbles reduce the volume of the test sample, resulting in a smaller actual sample volume and affecting the accuracy of the blood cell analyzer. Therefore, it is necessary to eliminate the interference caused by air bubbles to obtain accurate test results.

[0003] Referring to CN110095624B, this bubble detection and removal device uses a first sensor to acquire information about the sample inside the cylinder head. When the first sensor detects an air bubble at the cylinder head, the pump first pushes the bubble back into the cylinder, and then the pump restarts to draw the sample into the cavity. It also includes a second sensor connected to the control components and used to detect the liquid level in the cylinder. Both the first and second sensors are through-beam photoelectric sensors. This bubble detection and removal device requires a pump to expel the air bubbles before calculating the volume. The pump, as hardware, increases the operating cost, and the control logic is complex.

[0004] See CN205539016U. This multi-purpose medical testing instrument uses a pressure sensor to detect changes in air pressure inside the syringe to determine the presence of air bubbles or clots in the sample. If a problem is detected, the test will terminate and an alarm will sound. This multi-purpose medical testing instrument only alerts the user to the presence of air bubbles; the user must then manually remove the bubbles or directly replace the blood sample, which is cumbersome.

[0005] In view of this, a new technical solution is needed to solve the above-mentioned technical problems. Summary of the Invention

[0006] To eliminate the influence of air bubbles on the preparation of blood samples, the present invention aims to provide an air bubble correction method, device, blood cell analyzer, and storage medium, which has low cost, simple control logic, and simple operation.

[0007] To achieve the above objectives, the present invention employs the following technical means:

[0008] The first aspect of this invention discloses a bubble correction method for calculating the actual volume of reagent inside a tip tip. The bubble correction method includes the following steps:

[0009] Acquire the target image;

[0010] The target image is preprocessed to obtain a preprocessed image;

[0011] Detect whether there are air bubbles in the preprocessed image;

[0012] If so, calculate the volume V2 corresponding to the bubble, and then compensate the total volume V1 of the reagent to obtain the actual volume V0 = V1 - V2 of the reagent.

[0013] Optionally, the preprocessing of the target image includes the following steps: denoising, sharpening, and adaptive binarization of the target image.

[0014] Optionally, detecting whether bubbles exist in the preprocessed image includes the following steps:

[0015] Edge extraction is performed on the preprocessed image to obtain the image edges;

[0016] Determine whether the image edges include bubble edges;

[0017] If so, then it is determined that there are bubbles in the preprocessed image.

[0018] Optionally, determining whether the image edge includes bubble edges includes the following steps:

[0019] The outermost edge of the image is selected as the reagent edge;

[0020] If other image edges fall within the area defined by the reagent edge, then the image edge is determined to be a bubble edge.

[0021] Optionally, after detecting whether bubbles exist in the preprocessed image, the method further includes the following steps:

[0022] Determine the shape of the bubble;

[0023] Calculate the bubble volume V2 based on the bubble shape, and compensate for the total reagent volume V1 to obtain the actual reagent volume V0 = V1 - V2.

[0024] Optionally, calculating the volume V2 of the bubble based on its shape includes the following steps:

[0025] Based on the shape of the bubble, determine whether the bubble is frustum-shaped or spherical;

[0026] If the bubble is frustum-shaped, the volume V2' of the frustum-shaped bubble is calculated using the formula V2'=1 / 3n1πh(r2+πR22+rR2); where n1 is the number of frustum-shaped bubbles, r is the radius of the upper surface circle of the frustum, R2 is the radius of the lower surface circle of the frustum, and h is the height of the frustum.

[0027] If the bubble is spherical, the volume V2” of the spherical bubble is calculated using the formula V2”=n2(4 / 3)πR13; where R1 is the radius of the bubble and n2 is the number of spherical bubbles;

[0028] The bubble volume V2 = V2' + V2 is obtained.

[0029] A second aspect of the present invention discloses a bubble correction device for implementing the bubble correction method described in any of the preceding claims, comprising:

[0030] The light source is positioned towards the TIP head and is used to illuminate the reagent inside the TIP head.

[0031] Telecentric cameras are used to acquire images of targets;

[0032] The calculation module, electrically connected to the telecentric camera, is used to preprocess the target image, detect whether there are bubbles in the preprocessed image, and calculate the volume V2 corresponding to the bubbles, thereby compensating for the total volume V1 of the reagent to obtain the actual volume V0 of the reagent = V1 - V2.

[0033] Optionally, there are multiple telecentric cameras arranged around the TIP head on the same horizontal plane to identify whether the bubble is a spherical bubble or a frustum-shaped bubble.

[0034] A third aspect of the present invention discloses a blood cell analyzer, including any of the above-described bubble correction devices, which are used to implement any of the above-described bubble correction methods.

[0035] A fourth aspect of the present invention discloses a storage medium including a processor and a memory, wherein the memory stores a computer program, and the processor is configured to execute the computer program to implement the steps of any of the methods described above.

[0036] Compared with existing technologies, the present invention brings the following technical effects:

[0037] The bubble correction method of this invention identifies bubbles in the reagent within the TIP head of the target image and compensates for the reagent's volume to obtain its actual volume. This method uses machine vision to correct the actual volume of the reagent, eliminating the need for dedicated hardware to remove bubbles, resulting in low operating costs and simple control logic. Furthermore, accurate test results can be obtained directly through compensation, without requiring the user to remove bubbles and retest, simplifying operation for blood cell analyzers. Attached Figure Description

[0038] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 A schematic diagram of the internal structure of a blood cell analyzer according to an embodiment is shown;

[0040] Figure 2 It shows Figure 1 A magnified view of a portion of point A in the middle;

[0041] Figure 3 A schematic diagram of a bubble calibration apparatus according to an embodiment is shown;

[0042] Figure 4 A flowchart of a bubble calibration method according to one embodiment is shown;

[0043] Figure 5 It shows Figure 4 Flowchart of step S30;

[0044] Figure 6 A schematic diagram of a bubble calibration apparatus according to another embodiment is shown;

[0045] Figure 7 A flowchart of a bubble calibration method according to one embodiment is shown;

[0046] Figure 8 A flowchart of a bubble calibration method according to one embodiment is shown.

[0047] Explanation of key component symbols:

[0048] 10-Bottom surface; 20-Adjustment mechanism; 30-Tray; 40-Frame; 50-Pipette mechanism; 51-Drive component; 511-Guide shaft; 52-Connection structure; 53-Pipette pump; 531-Pipette tip; 60-Calculation module; 100-Consumables holder; 110-TIP tip; 1-Reagent; 2-Bubble; 3-Frustum-shaped bubble; 4-Spherical bubble; 120-Sample tube; 130-Sample cell; 140-Counting plate; 200-Light source; 300-Telecentric camera; 310-Telecentric lens. Detailed Implementation

[0049] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0050] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other. Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout.

[0051] Example 1

[0052] Please see Figures 1-2 This embodiment discloses a blood cell analyzer, which includes: an adjustment mechanism 20, a tray 30, a frame 40, a pipette pump 53, and a bubble 2 correction device.

[0053] The blood cell analyzer has a bottom surface 10, which serves as the mounting carrier for the adjustment mechanism 20 and the frame 40. The adjustment mechanism 20 is connected to the tray 30 to drive the tray 30 to move along the x-axis and y-axis directions, thereby controlling the movement of the tray 30 on the working surface parallel to the bottom surface 10.

[0054] Specifically, the adjustment mechanism 20 includes two drive motors, a transmission component, and a tray connection structure 52. One drive motor (not shown) is connected to the transmission component, and the transmission component is fixedly connected to the tray 30 through the tray connection structure 52 to drive the tray 30 to move along the x-axis. The other drive motor is directly connected to the tray 30 to drive the tray 30 to move along the y-axis.

[0055] It should be noted that the above-described driving device and transmission structure are merely exemplary and do not represent a limitation on the present invention. In other embodiments, other equivalent structures may also be used.

[0056] The tray 30 is provided with a consumables holder 100, which has two placement slots (not shown). One placement slot is used to place the sample tube 120, and the other placement slot is used to place the TIP head 110.

[0057] The sample tube 120 is a reagent tube with a cap and serves as a carrier for the blood sample. The TIP head 110 is a cylindrical plastic tube with openings at both ends. The TIP head 110 prevents the blood sample from directly contacting the pipette pump 53, avoiding cross-contamination between different blood samples and improving the reliability of the blood cell analyzer's test results.

[0058] The consumable holder 100 also has nine sample pools 130 arranged in a square. The sample pools 130 are used to hold diluent and hemolysing agent. Alternatively, some sample pools 130 may be left empty and used to mix blood samples with diluent. The diluent is physiological saline.

[0059] The pipetting mechanism 50 is mounted on the frame 40 and is capable of lifting and lowering to transfer diluents and hemolysing agents to prepare test samples.

[0060] Specifically, the pipetting mechanism 50 includes a drive element 51, two guide shafts 511, a connecting structure 52, and a pipetting pump 53. Both guide shafts are located on the lower surface of the drive element 51 and extend vertically. The connecting structure 52 is fitted onto the two guide shafts 511 and can move along the direction of the guide shafts 511. The pipetting pump 53 is connected to the connecting structure 52 and is mounted on the side wall of the connecting structure 52, thus maintaining a certain distance between the pipetting pump 53 and the guide shafts 511. This distance allows the pipetting pump 53 to move vertically without interference.

[0061] It should be noted that the above-described pipetting mechanism 50 is merely exemplary and does not represent a limitation of the present invention. In other embodiments, it may be configured with other equivalent structures.

[0062] The lower end of the pipette pump 53 is provided with a pipette tip 531. The pipette tip 531 is cylindrical in shape and its radius is slightly smaller than the radius of the upper opening of the TIP tip 110, so that the TIP tip 110 can be fixed securely and reliably.

[0063] The bubble correction device (not shown) is electrically connected to the adjustment mechanism 20 and the pipetting mechanism 50, respectively. The calculation module 60 controls the movement of the adjustment mechanism 20 on the working surface parallel to the bottom surface 10 and controls the lifting and lowering movement of the pipetting mechanism 50, including the degree of lifting and lowering, thus completing the preparation of the sample to be tested.

[0064] The preparation process of the test samples includes preparing red blood cell test samples, white blood cell test samples, and platelet test samples. The following steps take the preparation of red blood cell test samples as an example:

[0065] Control the pipette pump 53 to descend until the pipette tip 531 of the pipette pump 53 is precisely engaged with the upper opening of the TIP head 110, so as to install the TIP head 110.

[0066] Aspirate 40 μL of blood sample from sample tube 120 and place it into one of the empty sample chambers 130 and mix well to form a diluted blood sample.

[0067] Take 25 μL of diluted blood sample and transfer it to another sample chamber 130 containing physiological saline. Mix well to form a mixture. Then take another 25 μL of the mixture and add it to the red blood cell chamber of the counting chamber 140.

[0068] Air bubbles 2 may be generated during the process of pipetting pump 53 drawing blood samples, physiological saline, and diluting blood samples. By setting up an air bubble 2 correction device (not shown), the air bubble correction device will compensate for the volume of reagent 1 in TIP head 110, so that the actual volume of reagent 1 is used as the calculated data, thereby improving the detection accuracy of the blood cell analyzer.

[0069] Example 2

[0070] Please see Figure 3 Air bubbles 2 may form in the reagent 1 inside the TIP head 110. Therefore, this embodiment discloses an air bubble correction device for the above-mentioned blood cell analyzer to correct the volume of reagent 1 inside the TIP head 110, thereby improving the accuracy of blood testing.

[0071] The bubble correction device includes:

[0072] The light source 200 is positioned towards the TIP head 110 and is used to illuminate the reagent 1 inside the TIP head 110.

[0073] Telecentric camera 300, used to acquire target images;

[0074] The calculation module 60 is electrically connected to the telecentric camera 300. It is used to preprocess the target image, detect whether there is a bubble 2 in the preprocessed image, and calculate the volume V2 corresponding to the bubble. Then, it performs volume compensation on the total volume V1 of the reagent to obtain the actual volume V0 = V1 - V2 of the reagent.

[0075] Among them, light source 200 is a surface light source, that is, multiple LED arrays are arranged in a high density to form a light-emitting surface. The surface light source emits mutually parallel diffused light (such as... Figure 3 (As indicated by the middle arrow). The surface light source emits light uniformly, providing the telecentric camera 300 with a uniform and bright imaging environment.

[0076] Furthermore, the telecentric camera 300, the TIP head 110, and the light source 200 are arranged in a straight line. The diffused light emitted by the light source 200 passes through reagent 1 and then directly enters the telecentric camera 300. With this configuration, the light emitted by the light source 200 can enter the telecentric camera 300 to the maximum extent, providing the telecentric camera 300 with a good imaging environment.

[0077] In this scenario, the observed object is bubble 2, which appears at a random location on reagent sample 1 and has a three-dimensional shape. This means that bubble 2 is at varying distances from the telecentric camera 300. Using a traditional camera would result in bubbles appearing larger when closer and smaller when farther away, making it impossible to obtain the true volume of bubble 2 and affecting the accuracy of bubble 2 correction.

[0078] The telecentric camera 300 ensures that the magnification of the image remains constant within a certain object distance range. The image formed by the telecentric camera 300 maintains the same size and shape as the bubble 2 to guarantee image quality.

[0079] Specifically, the telecentric camera 300 includes a telecentric lens 310 and a CCD (not shown). The telecentric lens 310 is equivalent to setting an aperture in the optical path, which is used to limit the width of the imaging beam and the size of the imaging range; the CCD is used to receive the light signal received by the telecentric lens 310 and to convert the light signal into a digital signal. In other embodiments, a CMOS sensor can be used instead of a CCD.

[0080] Please see again Figure 2 The calculation module 60 is located on the bottom surface 10 and is electrically connected to the adjustment mechanism 20 and the pipetting mechanism 50, respectively. The calculation module 60 is used to control the movement of the adjustment mechanism 20 on the working surface parallel to the bottom surface 10 and to control the lifting and lowering movement of the pipetting mechanism 50, thus completing the preparation of the sample to be tested. The specific control logic of the calculation module 60 is described in Example 1 and will not be repeated here.

[0081] In addition, the calculation module 60 is also used to preprocess the target image, detect whether there are air bubbles 2 in the preprocessed image, and perform volume compensation on the total reagent volume V1 to obtain the actual volume V0 of reagent 1. The preprocessing will be described in detail in Example 3.

[0082] Example 3

[0083] Please see Figure 4 , Figure 5 and Figure 6 This embodiment discloses a bubble correction method, which is implemented using the bubble correction device of Embodiment 2. See also... Figure 4 The bubble correction method includes the following steps:

[0084] Step S10: Acquire the target image;

[0085] Step S20: Preprocess the target image to obtain a preprocessed image;

[0086] Step S30: Detect whether there are air bubbles in the preprocessed image;

[0087] Step S40: If yes, perform volume compensation on the total reagent volume V1 to obtain the actual reagent volume V0.

[0088] The bubble correction method of this invention identifies bubbles in the reagent within the TIP head of the target image and compensates for the reagent's volume to obtain its actual volume. This method uses machine vision to correct the actual volume of the reagent, eliminating the need for dedicated hardware to remove bubbles, resulting in low operating costs and simple control logic. Furthermore, accurate test results can be obtained directly through compensation, without requiring the user to remove bubbles and retest, simplifying operation for blood cell analyzers.

[0089] The target image is an image with a TIP header.

[0090] In practical applications, the acquired target images may be unclear. Unclear images prevent the calculation module from identifying bubbles based on features. Therefore, it is necessary to preprocess the target images to obtain preprocessed images that meet the usage requirements.

[0091] In one specific implementation, step S20 includes the following steps:

[0092] Step S21: Denoise, sharpen, and adaptively binarize the target image.

[0093] Denoising, sharpening, and adaptive binarization can all improve the clarity of the target image.

[0094] Specifically, denoising is suitable for situations where the target image has multiple noise points, such as Gaussian noise points or salt-and-pepper noise points. By filtering the target image, noise points are eliminated, making the target image clearer.

[0095] Sharpening is suitable for situations where the edges and contours of the target image are unclear. By filtering the target image, it compensates for the image's contours, enhances the edges and areas with abrupt grayscale changes, making the target image clearer.

[0096] Adaptive binarization is suitable for situations where different parts of a target image have different brightness levels. Adaptive binarization involves defining multiple small regions on the image and assigning a different threshold to each region based on its brightness, thereby making the target image clearer even with varying brightness levels.

[0097] It should be noted that, depending on the specific characteristics of the target image, denoising, sharpening, and adaptive binarization can be used individually or in combination.

[0098] See Figure 5 In one specific implementation, step S30 includes the following steps:

[0099] Step S31: Extract edges from the preprocessed image to obtain the image edges;

[0100] Step S32: Determine whether the image edges include bubble edges;

[0101] See Figure 6 Specifically, the implementation process of step S32 includes the following steps:

[0102] Step S321: Select the outermost image edge as the reagent edge, wherein the reagent edge is frustum-shaped;

[0103] Step S322: When an image edge falls within the frustum-shaped area defined by the reagent edge, the edge is determined to be a bubble edge;

[0104] Step S323: If an image edge falls outside the frustum-shaped region of the reagent edge, then that edge is considered noise.

[0105] Determining whether an edge is a bubble edge by its relative position to the reagent edge is highly accurate and practical.

[0106] Step S32 effectively selects bubble edges from the image edges, thereby improving the success rate of bubble edge recognition and thus improving the accuracy of bubble volume calculation.

[0107] Edge relationships and quantities are relatively easy parameters to obtain in machine vision. Identifying bubble edges and reagent edges based on edge relationships and quantities is a simple and reliable operation.

[0108] Step S33: If yes, then determine that there are bubbles in the preprocessed image.

[0109] Please refer to it again. Figure 3 There is a clear edge between bubble 2 and reagent 1. Step S31 extracts the edge as a feature, which can quickly and effectively identify bubble 2 and reagent 1, and also provides a basis for calculating the volume of bubble 2 and reagent 1 later.

[0110] by Figure 3 Only one frustum-shaped bubble is used as an example.

[0111] In step S40, the volume V2 of the bubble is calculated. The specific implementation is as follows: The radius r of the upper surface circle of the frustum, the radius R2 of the lower surface circle of the frustum, and the height h of the frustum are obtained from the bubble's edge. Then, the volume of the bubble is calculated using the formula for a frustum: V2' = 1 / 3πh(r² + πR²² + rR²).

[0112] In step 40, before performing volume compensation on the total reagent volume V1, it is necessary to calculate the total reagent volume V1, which is achieved through the following steps: obtaining the height H of the reagent through the edge of the reagent, the radius R0 of the upper surface of the reagent circle, and V2 = 1 / 3πR0. 2H. Among them, the TIP head 110 is a cone, and the reagent is distributed in a cone shape inside the TIP head 110.

[0113] In other embodiments, the total reagent volume V1 is set by a computer program and does not need to be calculated separately.

[0114] Example 4

[0115] In practical applications, there are scenarios where multiple bubbles 2 exist, with some bubbles spherical and others frustum-shaped. Please refer to [link / reference]. Figure 7 For example, reagent 1 contains one frustum-shaped bubble 3 and three spherical bubbles 4.

[0116] Please see Figure 7 In order to accurately calculate the volume of bubble 2, this embodiment discloses a bubble correction device for the above-mentioned blood cell analyzer, so as to correct the volume of reagent 1 in TIP head 110, thereby improving the accuracy of blood detection.

[0117] Unlike Embodiment 2, the bubble correction device in this embodiment includes two telecentric cameras 300, which are arranged around the TIP head 110 on the same horizontal plane to identify whether the bubble 2 is a spherical bubble 4 or a frustum-shaped bubble 3.

[0118] In the diagram, the arrows indicate the direction of the light path. The light path extends from the light source 200 to reagent 1 inside the TIP head 110, and then into the telecentric camera 300. There is a shortest straight line between each telecentric camera 300 and the TIP head 110, and the angle α between two adjacent shortest straight lines is 90°.

[0119] By setting the included angle α to 90°, images from different angles can be obtained through two telecentric cameras 300, and the algorithm can be used to distinguish between spherical bubble 4 and frustum-shaped bubble 3.

[0120] In one specific implementation, the method for determining the bubble shape includes the following steps:

[0121] The first image is acquired by a telecentric camera;

[0122] Another telecentric camera acquires a second image;

[0123] Edge extraction is performed on the first image to obtain the first edge;

[0124] Perform edge extraction on the second selected edge to obtain the second edge;

[0125] Determine the shapes of the first and second edges;

[0126] If both the first and second edges are circular, then the bubble is determined to be spherical.

[0127] If both the first and second edges are two parallel straight lines, the bubble shape is determined to be frustum-shaped. The relationship and number of edges are relatively easy parameters to obtain in machine vision. Determining the bubble edge and reagent edge by analyzing the relationship and number of edges is a simple and reliable operation.

[0128] By using two telecentric cameras to observe the bubble from two different angles, the accuracy of bubble shape judgment and volume compensation can be further improved.

[0129] Example 5

[0130] Please see Figure 8 This embodiment discloses a bubble correction method, which is implemented by the bubble correction device of embodiment 2 or embodiment 4.

[0131] The bubble correction method includes the following steps:

[0132] Step S100: Acquire the target image;

[0133] Step S200: Preprocess the target image to obtain a preprocessed image;

[0134] Step S300: Detect whether there are air bubbles in the preprocessed image;

[0135] Step S400: If yes, determine the bubble shape;

[0136] Step S500: Calculate the bubble volume V2 according to the bubble shape and compensate for the total reagent volume V1 to obtain the actual reagent volume V0 = V1 - V2.

[0137] The design concept of this embodiment is to use machine vision to identify frustum-shaped bubbles in a TIP head, a specific scenario.

[0138] The inside of the tip is a confined space where the reagent is stationary. In addition to gravity and buoyancy, the bubble is also subjected to hydrostatic pressure exerted by the sidewalls of the tip. This hydrostatic pressure creates an external force field, balancing the pressure inside the gas and the internal pressure of the reagent. Therefore, besides spherical bubbles, the bubble inside the tip may also maintain a frustum shape, following the shape of the tip and positioned between the two reagent segments (see...). Figure 7 In short, the static pressure exerted on the bubble by the sidewall of the tip, the surface tension of the bubble itself, and the pressure exerted on the bubble by the reagent can be balanced, allowing the bubble to exist in a frustum shape.

[0139] The specific process of frustum-shaped air bubble formation is as follows: During pipetting, the reagent level continuously drops until it falls below the lower opening of the tip. If the pipetting pump has not received a stop signal from the computing module at this point, it will draw in a large amount of air. This air will form an air column inside the tip. Because the tip is frustum-shaped, this air column follows the shape of the tip and is also frustum-shaped.

[0140] Generally, when unaffected by an external force field, bubbles are spherical. This is because, under the influence of surface tension, the surface area of ​​a bubble tends to be minimized, causing the surface to bend. Since the curvature of a sphere is uniform at all points, the forces acting on it are balanced, and the sphere has the smallest surface area, resulting in the lowest and most stable total surface energy of the bubble.

[0141] For example, see CN113628213A. According to paragraphs 23 and 29 of its specification, the bubble is located in the liquid path of the sample analyzer. The liquid path does not apply an external force field to the bubble; therefore, the bubble in the liquid path is spherical.

[0142] For example, see CN109073517A, paragraph 39 of which states that the bubble is located in the space between the substrate and the cover. The space between the substrate and the cover does not provide an external force field to the bubble; therefore, the bubble in the space between the substrate and the cover is spherical.

[0143] In summary, frustum-shaped bubbles will form within the TIP head. It is necessary for the applicant to determine the shape of these bubbles and specifically configure a calculation method for frustum-shaped bubbles based on the cell shape to improve the accuracy of bubble volume calculation. Conversely, in the space between the interfaces of the liquid path, substrate, and cap, bubbles of shapes other than spheres will not exist, and there is no incentive to determine the shape of the bubbles or change the bubble volume calculation method based on their shape. In machine vision technology, the difficulty of completing the task must be judged from the machine's perspective; otherwise, incorrect estimates and judgments of the task's difficulty will be made.

[0144] Specifically, machine vision technology simulates human visual function through computers. However, unlike human visual function, machine vision also possesses some of the functions of the human brain, namely, extracting information from images of objective things, processing and understanding it, and ultimately using it for practical detection, measurement, and control.

[0145] In some scenarios, the human visual system is more effective than machine vision.

[0146] For example, the human visual system can quickly and accurately determine the shape of a bubble. However, machines need to use algorithms to measure, verify, and calculate features in order to determine the shape of a bubble.

[0147] In some scenarios, machine vision can be more effective than the human visual system.

[0148] For example, the human visual system struggles to perform quantitative calculations, such as determining the volume of a bubble; however, machine vision, through algorithms that measure, verify, and calculate features, can obtain the volume of a bubble, especially in determining whether it is a frustum or a sphere.

[0149] It should be noted that the algorithm needs to be specially designed according to the specific scenario, and cannot be simply obtained.

[0150] In one specific implementation, step S500, which calculates the bubble volume V2 based on the bubble shape, includes the following steps:

[0151] Step S510: Determine whether the bubble is frustum-shaped or spherical;

[0152] For example, a center point is selected within the edge of the bubble, and the distance between the center point and the edge is determined based on pixels. When the distances from the center point to the edge are all equal, the bubble is spherical. When the distances from the center point to the edge are unequal, the bubble is frustum-shaped.

[0153] Step S520: If the bubble is frustum-shaped, calculate the volume V2' of the frustum-shaped bubble using the formula V2' = 1 / 3n1πh(r2 + πR22 + rR2); where n1 is the number of frustum-shaped bubbles, r is the radius of the upper surface circle of the frustum, R2 is the radius of the lower surface circle of the frustum, and h is the height of the frustum.

[0154] A telecentric camera can only capture one cross-section of the bubble.

[0155] The cross-section of the frustum-shaped bubble is an isosceles trapezoid. The number of frustum-shaped bubbles, n1, is determined by obtaining the number of isosceles trapezoids extracted from the image. The length of the upper side of the isosceles trapezoid is obtained, and half of the upper side is used as the radius r of the upper circle of the frustum. The length of the lower side of the isosceles trapezoid is obtained, and half of the lower side is used as the radius R2 of the upper circle of the frustum. The height h of the frustum is determined by the length of the isosceles trapezoid.

[0156] Step S530: If the bubble is spherical, calculate the volume V2” of the spherical bubble using the formula V2”=n2(4 / 3)πR13; where R1 is the radius of the bubble and n2 is the number of spherical bubbles;

[0157] The cross-section of the spherical bubble is circular. The number of frustum-shaped bubbles, n2, is determined by obtaining the number of circles extracted from the image. The radius R1 of the circles is also obtained.

[0158] Step S540: Obtain the bubble volume V 2＝ V2'+V2".

[0159] By using steps S510 to S540, specifically by setting the algorithm described above, the volume of frustum-shaped and spherical bubbles can be accurately calculated, thus improving the accuracy of volume compensation.

[0160] In addition, the applicant conducted a comparative experiment to examine the accuracy of different calculation methods and results. The comparative experiment involved applying three different calculation methods to cases with spherical and cylindrical bubbles, all other things being equal.

[0161] The first to third groups contained both cylindrical and spherical air bubbles. The first group used a spherical bubble calculation method, the second group used a frustum-shaped bubble calculation method, and the third group used a combination of both. The accuracy of the test results was based on the number of red blood cells in the sample. Ten blood samples were collected from each of the first to third groups. The injection volume was 20 μL, and the reagent volume was the average of the reagent volumes from the ten blood samples.

[0162] It should be explained that: In the first group, the calculation method for spherical bubbles calculates the volume of both spherical and frustum-shaped bubbles. In the second group, the calculation method for frustum-shaped bubbles calculates the volume of both spherical and spherical bubbles. In the third group, the calculation method for spherical bubbles will be used to calculate the volume of spherical bubbles, and the calculation method for frustum-shaped bubbles will be used to calculate the volume of frustum-shaped bubbles.

[0163] In addition, a fourth group was set up as a control group. The fourth group calculated the actual volume of the reagent after directly removing air bubbles. This technique is relatively mature and accurate, making it suitable as a control group for comparison with the first three groups.

[0164] Similarly, in the fourth group, 10 blood samples were used, with an injection volume of 20 μL. The average reagent volume was the average of the volumes of the 10 blood samples.

[0165] The fourth group uses a method for calculating reagent volume to directly calculate the actual volume of the reagent. This method is: V3 = 1 / 3πR3 2 d, where R3 is the radius of the upper surface of the reagent after the air bubbles are removed, and d is the height of the reagent after the air bubbles are removed.

[0166] The calculation results are shown in the table below:

[0167]

[0168]

[0169] As shown in the table above, the difference between the average reagent volume of group four and group one is 0.4, the difference between the average reagent volume of group four and group two is 0.3, and the difference between the average reagent volume of group four and group three is 0.1. That is, among the three groups of data, the difference between the average reagent volume of group four and group three is the smallest. Therefore, we can conclude that in reagent scenarios involving both spherical and frustum-shaped bubbles, the calculation method combining spherical and frustum-shaped bubbles yields the most accurate reagent volume.

[0170] Red blood cells have the same density in a blood sample, and the volume of the reagent is directly proportional to the number of red blood cells. Therefore, it can be deduced that using a combination of spherical and frustum-shaped air bubbles to calculate the reagent volume results in relatively accurate red blood cell counts, thus improving the precision of the blood cell analyzer.

[0171] Example 6

[0172] A storage medium includes a processor and a memory, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the bubble detection method in the above embodiments.

[0173] A processor may include one or more processing cores, such as a 4-core processor or a 6-core processor. The processor can be implemented using at least one of the following hardware forms: DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), or PLA (Programmable Logic Array).

[0174] The memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, memory device, or other volatile solid-state storage device.

[0175] The memory of this embodiment stores a computer program that can run on a processor. When the processor executes the computer program, it can implement all or part of the implementation steps of the bubble correction method of the present invention or the bubble correction device described above, and / or other content described in the text.

[0176] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom still fall within the scope of protection of this invention.

Claims

1. A bubble correction method for calculating the actual volume of reagent inside a tip, characterized in that, The bubble correction method includes the following steps: Acquire the target image; The target image is preprocessed to obtain a preprocessed image; Detect whether there are air bubbles in the preprocessed image; If so, calculate the volume V2 corresponding to the bubble, and then compensate the total reagent volume V1 to obtain the actual reagent volume V0 = V1 - V2; The preprocessing of the target image includes the following steps: The target image is denoised, sharpened, and adaptively binarized. The process of detecting whether air bubbles exist in the preprocessed image includes the following steps: Edge extraction is performed on the preprocessed image to obtain the image edges; Determine whether the image edges include bubble edges; If so, then determine that there are bubbles in the preprocessed image; The process of determining whether the image edge includes bubble edges includes the following steps: The outermost edge of the image is selected as the reagent edge; If other image edges fall within the area defined by the reagent edge, then the other image edges are determined to be bubble edges; If an image edge falls outside the reagent edge, that edge is considered noise. After detecting whether there are bubbles in the preprocessed image, the following steps are also included: Determine the shape of the bubble; The method for determining the shape of a bubble is as follows: The first image is acquired by a telecentric camera; Another telecentric camera acquires a second image; Edge extraction is performed on the first image to obtain the first edge; Perform edge extraction on the second selected edge to obtain the second edge; Determine the shapes of the first and second edges; If both the first and second edges are circular, then the bubble is determined to be spherical. If both the first and second edges are two parallel straight lines, then the bubble is determined to be truncated cone-shaped.

2. The bubble correction method according to claim 1, characterized in that, After detecting whether there are bubbles in the preprocessed image, the following steps are also included: Calculate the bubble volume V2 based on the bubble shape, and compensate for the total reagent volume V1 to obtain the actual reagent volume V0 = V1 - V2.

3. The bubble correction method according to claim 2, characterized in that, The calculation of the bubble volume V2 based on the bubble shape includes the following steps: If the bubbles are frustoconical, the frustoconical bubble volume V2' is calculated by the formula V2' = 1 / 3n1πh(r 2 + πR2 2 + rR2), where n1 is the number of frustoconical bubbles, r is the radius of the upper frustoconical surface circle, R2 is the radius of the lower frustoconical surface circle, and h is the height of the frustoconical bubble. If the bubble is spherical, then the formula V2”=n2(4 / 3)πR1 is used. 3 Calculate the volume V2” of the spherical bubble; where R1 is the radius of the bubble and n2 is the number of spherical bubbles; The bubble volume V2 = V2' + V2 is obtained.

4. A bubble correction device for implementing the bubble correction method according to any one of claims 1 to 3, characterized in that, The bubble correction device includes: The light source is positioned towards the TIP head and is used to illuminate the reagent inside the TIP head. A telecentric camera is used to acquire images of the target; two telecentric cameras are positioned on the same horizontal plane around the TIP head to identify whether the bubble is a spherical bubble or a frustum-shaped bubble. The calculation module, electrically connected to the telecentric camera, is used to preprocess the target image, detect whether there are bubbles in the preprocessed image, and calculate the volume V2 corresponding to the bubbles, thereby compensating for the total volume V1 of the reagent to obtain the actual volume V0 of the reagent = V1 - V2.

5. The bubble correction device according to claim 4, characterized in that, The telecentric cameras are multiple and arranged around the TIP head on the same horizontal plane to identify whether the bubble is spherical or frustum-shaped.

6. A blood cell analyzer, characterized in that, Includes the bubble correction device as described in claim 4 or 5.

7. A storage medium, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program, and the processor is used to execute the computer program to implement the steps of the method according to any one of claims 1 to 3.

Images