Microfluidic sorting system and method

Abstract

The application provides a micro-fluidic sorting system and method, and relates to the technical field of particle sorting.The micro-fluidic sorting system comprises a micro-fluidic sorting chip, a liquid driving device, an image acquisition device and a controller.The application provides a micro-fluidic sorting system and method, which can realize sorting of target particles from particles to be sorted based on the image acquisition device and deep learning technology without the need of fluorescent staining of the particles to be sorted, can reduce the influence and damage of fluorescent staining on the particles to be sorted, can improve the accuracy and reliability of subsequent experiments, can significantly reduce the cost required for cell sorting, can obtain multi-dimensional information such as cell size, roundness, edge features and morphological features, can better meet the multi-dimensional cell sorting requirement, can help more accurately and efficiently sort the target particles, can better meet the multi-dimensional cell sorting requirement, and can realize flexible sorting and non-destructive sorting of cells.

Description

Technical Field

[0001] This invention relates to the field of particle sorting technology, and in particular to a microfluidic sorting system and method. Background Technology

[0002] Microfluidic sorting systems are cell or particle sorting systems based on microfluidic technology. They offer advantages such as high throughput, automation, integration, and low cost, and are widely used in various biomedical research and applications.

[0003] In related technologies, traditional microfluidic sorting systems can sort target particles from microparticles to be sorted based on fluorescence-activated cell sorting analysis. However, when using traditional microfluidic sorting systems to sort target particles from microparticles to be sorted, the microparticles to be sorted must first be fluorescently stained before optical effects can be used to manipulate and separate the fluorescently stained microparticles to obtain the desired target particles.

[0004] However, fluorescent staining of cells can alter their behavior or activity. Therefore, target particles sorted using traditional microfluidic sorting systems may exhibit feature loss or modification, affecting the reliability of subsequent experiments. Thus, how to achieve cell sorting without fluorescent staining is a pressing technical problem in this field. Summary of the Invention

[0005] This invention provides a microfluidic sorting system and method to address the shortcomings of existing technologies that require fluorescent staining of cells during cell sorting. However, fluorescent staining alters the behavior or activity of the stained cells, leading to the loss or modification of characteristic features in the sorted cells, which in turn affects the reliability of subsequent experiments. This invention enables cell sorting without fluorescent staining of cells.

[0006] This invention provides a microfluidic sorting system, comprising: a microfluidic sorting chip, a liquid driving device, an image acquisition device, and a controller. The liquid driving device and the image acquisition device are electrically connected. The liquid driving device, in response to the controller's control, drives a sample solution containing particles to be sorted into the microfluidic sorting chip from a second inlet, and drives a sheath fluid into the microfluidic sorting chip from a first inlet, thereby mixing the sample solution and the sheath fluid within the microfluidic sorting chip. The liquid driving device is also, in response to the controller's control, drives a sorting liquid into the microfluidic sorting chip from a third inlet, thereby causing the mixture of the sample solution and the sheath fluid to be mixed within the microfluidic sorting chip under the impact of the sorting liquid. A sample flows into a collection bottle from an outlet, and the particles to be sorted include target particles. The image acquisition device is used to acquire images of the mixture within the microfluidic sorting chip at a preset frequency and send the acquired images of the mixture to a controller. The controller is used to control the liquid driving device to drive the sheath fluid and the sample fluid into the microfluidic sorting chip from the first and second inlets, respectively, when a cell sorting task is triggered. The controller is also used to perform image recognition on the mixed liquid image sent by the image acquisition device, and then, based on the recognition result of the mixed liquid image, determine whether to control the liquid driving device to drive the sorting liquid into the microfluidic sorting chip from the third inlet.

[0007] According to a microfluidic sorting system provided by the present invention, the controller is specifically configured to, upon receiving an image of the mixture sent by the image acquisition device, input the image of the mixture into a cell recognition model, obtain the cell recognition result output by the cell recognition model, and then, based on the cell recognition result, determine that a predefined sorting condition is met, control the liquid driving device to drive the sorting liquid to flow from the third inlet into the microfluidic sorting chip; wherein, the cell recognition model is obtained after training based on a sample image and the cell recognition result of the sample image; the cell recognition result of the sample image is the sample image labeled with the target microparticle; the sorting condition includes that the image of the mixture contains the target microparticle.

[0008] According to a microfluidic sorting system provided by the present invention, the microfluidic sorting chip includes: a first flow channel and an annular flow channel; the annular flow channel is provided with two flow holes, namely a first flow hole and a second flow hole; the first flow hole communicates with a first inlet disposed outside the annular flow channel, and the second flow hole communicates with a second inlet disposed inside the annular flow channel; a first end of the first flow channel communicates with the second flow hole, such that the sheath fluid flowing in from the first inlet is coated at the second flow hole. The sample liquid flowing into the sample inlet forms a mixture that flows into the first flow channel. The second end of the first flow channel is connected to the third inlet, the first outlet, and the second outlet of the microfluidic sorting chip, respectively. This allows the mixture to flow into the waste bottle through the second outlet without being impacted by the sorting liquid flowing in from the third inlet at the second end of the first flow channel. Conversely, the mixture to flow into the collection bottle through the first outlet under the impact of the sorting liquid flowing in from the third inlet at the second end of the first flow channel.

[0009] According to a microfluidic sorting system provided by the present invention, the liquid driving device includes: a pressurizing device; the pressurizing device is used to pressurize a first sealed container containing the sheath fluid and a second sealed container containing the sample fluid in response to the control of the controller, so that the sheath fluid and the sample fluid flow into the first inlet and the second inlet respectively through a first pipe and a second pipe under pressure; the pressurizing device is also used to pressurize a third sealed container containing the sorting liquid in response to the control of the controller, so that the sorting liquid flows into the third inlet through a third pipe; the first pipe is used to connect the first sealed container and the first inlet; the second pipe is used to connect the second sealed container and the second inlet; the third pipe is used to connect the third sealed container and the third inlet.

[0010] According to a microfluidic sorting system provided by the present invention, it further includes: a second flow channel, a third flow channel, a fourth flow channel, a fifth flow channel, and a sixth flow channel; the first flow channel and the third flow channel are coaxially arranged; the second end of the first flow channel is connected to the third sample inlet through the second flow channel, the second end of the first flow channel is connected to the second sample outlet through the third flow channel, and the second end of the first flow channel is connected to the first sample outlet through the fourth flow channel; the first sample inlet is connected to the first flow passage through the fifth flow channel, and the second sample inlet is connected to the second flow passage through the sixth flow channel; the first end of the fourth flow channel is connected to the second end of the first flow channel, and the second end of the fourth flow channel is connected to the first sample outlet, the diameter of the first end of the fourth flow channel is larger than the diameter of the second end of the fourth flow channel; the diameter of the third flow channel is larger than the diameter of the first end of the fourth flow channel.

[0011] According to a microfluidic sorting system provided by the present invention, the controller is further configured to control a first pressure value applied by the pressurizing device to the first sealed container based on the radius and specific resistance of the fifth channel and the annular channel and the density of the sheath fluid; the controller is further configured to control a second pressure value applied by the pressurizing device to the second sealed container based on the radius and specific resistance of the sixth channel and the density of the sample fluid; and the controller is further configured to control a third pressure value applied by the pressurizing device to the third sealed container based on the radius and specific resistance of the second channel and the density of the sorting fluid.

[0012] According to a microfluidic sorting system provided by the present invention, the diameter of the flow channel in the microfluidic sorting chip is determined based on the diameter of the particles to be sorted.

[0013] According to a microfluidic sorting system provided by the present invention, after the controller controls the liquid driving device to drive the sorting liquid into the microfluidic sorting chip from the third inlet, it is further configured to, upon receiving an image of the mixture sent by the image acquisition device, input the image of the mixture into a cell recognition model, obtain the cell recognition result output by the cell recognition model, and then, if it is determined based on the cell recognition result that the sorting conditions are not met, control the liquid driving device to stop driving the sorting liquid into the microfluidic sorting chip from the third inlet.

[0014] According to a microfluidic sorting system provided by the present invention, when the particles to be sorted are cells, the cells to be sorted live in a cell matrix.

[0015] This invention also provides a microfluidic sorting method based on any of the microfluidic sorting systems described above, comprising: upon determining that a cell sorting task is triggered, driving a sample solution and a sheath fluid containing particles to be sorted to flow into the microfluidic sorting chip from a first inlet and a second inlet, respectively, so that the sample solution and the sheath fluid are mixed within the microfluidic sorting chip; acquiring an image of the mixture of the sample solution and the sheath fluid within the microfluidic sorting chip; performing image recognition on the image of the mixture, and then, based on the mixture... The image recognition result of the mixture determines whether to drive the sorting liquid to flow into the microfluidic sorting chip from the third inlet. If the sorting liquid is not driven to flow into the microfluidic sorting chip from the third inlet, the mixture flows to the waste bottle through the second outlet of the microfluidic sorting chip. If the sorting liquid is driven to flow into the microfluidic sorting chip from the third inlet, the mixture flows to the collection bottle through the first outlet of the microfluidic sorting chip.

[0016] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the microfluidic sorting method as described above.

[0017] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the microfluidic sorting method as described above.

[0018] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the microfluidic sorting method as described above.

[0019] The microfluidic sorting system and method provided by this invention include a microfluidic sorting chip, a liquid driving device, an image acquisition device, and a controller. Based on the image acquisition device and deep learning technology, it can sort the desired target particles from the microparticles to be sorted without fluorescent staining. This reduces the impact and damage of fluorescent staining on the microparticles to be sorted, improves the accuracy and reliability of subsequent experiments, and significantly reduces the cost of cell sorting. Furthermore, compared to traditional microfluidic sorting systems that rely on fluorescent staining, the microfluidic sorting system provided by this invention, based on the image acquisition device and deep learning technology, can acquire multi-dimensional information such as cell size, roundness, edge features, and morphological features. This better meets the multi-dimensional cell sorting needs, facilitating more accurate and efficient sorting of target particles. It enables flexible and non-destructive cell sorting, and is particularly suitable for cell recovery and culture for further analysis. The cell sorting operation is simple and has broad application prospects. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the microfluidic sorting system provided by the present invention.

[0022] Figure 2 This is a schematic diagram of the microfluidic sorting system provided by the present invention.

[0023] Figure 3 This is a schematic diagram of the fluid channel structure in the microfluidic sorting chip of the microfluidic sorting system provided by the present invention.

[0024] Figure 4 This is one of the images of the mixture in the microfluidic sorting system provided by the present invention.

[0025] Figure 5 This is the second image of the mixture in the microfluidic sorting system provided by the present invention.

[0026] Figure 6 This is a schematic diagram of the working principle of the liquid driving device in the microfluidic sorting system provided by the present invention.

[0027] Figure 7 This is a summary diagram of the control parameters in the microfluidic sorting system provided by the present invention.

[0028] Figure 8 This is one of the simulation results of the microfluidic sorting chip provided by the present invention.

[0029] Figure 9 This is the second schematic diagram of the simulation results of the microfluidic sorting chip provided by the present invention.

[0030] Figure 10 This is a schematic flowchart of the microfluidic sorting method provided by the present invention.

[0031] Figure 11 This is a schematic diagram of the structure of the electronic device provided by the present invention.

[0032] Figure label:

[0033] 101: Microfluidic sorting system; 102: Microfluidic sorting chip; 103: Liquid driving device; 104: Image acquisition device; 105: Controller; 301: First flow channel; 302: Annular flow channel; 303: First flow passage; 304: Second flow passage; 305: Second sample inlet; 306: First sample inlet; 307: Third sample inlet; 308: Second sample outlet; 309: First sample outlet; 310: Second flow channel; 311: Third flow channel; 312: Fourth flow channel; 313: Fifth flow channel; 314: Sixth flow channel. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0035] In the description of the invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0036] In the description of this application, the terms "first," "second," etc., are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, without limiting the number of objects; for example, a first object can be one or more. Furthermore, in the description of this application, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects have an "or" relationship.

[0037] It should be noted that traditional microfluidic sorting systems in related technologies can utilize optical effects to manipulate and separate fluorescently stained microparticles, thereby separating the desired target microparticles from the microparticles to be sorted.

[0038] In traditional microfluidic sorting systems, when manipulating and separating fluorescently stained microparticles, droplets encapsulate different cells or viruses, exhibiting varying fluorescence intensities through specific reactions. The fluorescence signal is acquired by a PMT (Polymer Transducer), converting the optical signal into an electrical signal, which is then acquired by an FPGA (Field-Programmable Gate Array) and displayed in software. Droplets matching the sorting criteria are identified, and the FPGA outputs a TTL-level signal. This electrical signal is amplified by a high-voltage amplifier and output to the microfluidic chip. A non-uniform electric field is formed on both sides of the chip channel. Under this electric field, the target droplet becomes polarized, causing it to move and thus achieving sorting.

[0039] However, the chemical reagents used in fluorescent staining of cells (such as fixatives and permeabilizers) may be toxic to cells, affecting their normal physiological functions. These chemicals may damage cell membranes, interfere with intracellular metabolic processes, or affect cell signaling. Furthermore, the chemical reagents and fixatives used in fluorescent staining may alter cell morphology and size, resulting in differences between observed cell morphology and actual physiological cell morphology. This morphological alteration may affect the accurate understanding of cell structure and function. During fluorescent staining, cells typically undergo fixation and permeabilization, steps that may damage cell viability and cause cells to lose their original functions. This is particularly true for cells that need to maintain viability (such as cells in live-cell imaging experiments), where fluorescent staining may cause irreversible damage.

[0040] Therefore, fluorescent staining of cells alters the behavior or activity of the stained cells, and target particles sorted using traditional microfluidic sorting systems may exhibit feature loss or modification, thus affecting the reliability of subsequent experiments. How to achieve cell sorting without fluorescent staining is a pressing technical problem to be solved in this field.

[0041] Furthermore, traditional microfluidic sorting systems rely entirely on fluorescent biomarkers and / or light scattering intensity when sorting cells using fluorescence-activated cell sorting analysis, without utilizing high-information-content image information. This results in a lack of analysis of the localization of subcellular organelles and multimorphic features, such as cell geometry, nucleus, and cytoskeleton shape. Such information is crucial for improving cell classification and identification accuracy, especially for rare cells and cells without known or unique biomarkers. The amount of cell morphology and information that fluorescence-activated cell sorting analysis can detect is extremely limited.

[0042] Furthermore, traditional microfluidic sorting systems are typically suited for oil-phase droplets, where cells or viruses are encapsulated within droplets, using the oil phase as the mobile phase. However, most cells thrive in a cell matrix, which is more compatible with the human environment. Some specialized cells (such as organoids) can only proliferate, differentiate, and grow three-dimensionally in a cell matrix (e.g., matrigel and collagen). However, these specialized cells cannot be sorted in an oil phase.

[0043] Meanwhile, for large particles such as organoid cells, the diameter of these large particles is approximately 100-1000 mm. Left and right. However, traditional microfluidic sorting systems are difficult to sort large particles.

[0044] The following is combined Figures 1-6 This invention describes the microfluidic sorting system provided by the present invention.

[0045] Figure 1 This is a schematic diagram of the microfluidic sorting system provided by the present invention. The following is in conjunction with... Figure 1 The microfluidic sorting system provided by this invention will be described. For example... Figure 1 As shown, the microfluidic sorting system 101 includes: a microfluidic sorting chip 102, a liquid driving device 103, an image acquisition device 104, and a controller 105. The liquid driving device 103 and the image acquisition device 104 are electrically connected.

[0046] The liquid driving device 103 is used in response to the control of the controller 105 to drive the sample liquid containing the microfluidic particles to be sorted into the microfluidic sorting chip 102 from the second inlet and drive the sheath liquid into the microfluidic sorting chip 102 from the first inlet, so that the sample liquid and the sheath liquid are mixed in the microfluidic sorting chip 102. The liquid driving device 103 is also used in response to the control of the controller 105 to drive the sorting liquid into the microfluidic sorting chip 102 from the third inlet, so that the mixture of sample liquid and sheath liquid flows into the collection bottle through the first outlet 309 of the microfluidic sorting chip 102 under the impact of the sorting liquid. The microparticles to be sorted include the target microparticles.

[0047] The image acquisition device 104 is used to acquire images of the mixture in the microfluidic sorting chip 102 at a preset frequency, and send the acquired images of the mixture to the controller 105;

[0048] The controller 105 is used to control the liquid driving device 103 to drive the sheath fluid and sample fluid to flow into the microfluidic sorting chip 102 from the first inlet and the second inlet when the cell sorting task is determined to be triggered. The controller 105 is also used to perform image recognition on the image of the mixture when the image acquisition device 104 sends the image of the mixture, and then determine whether to control the liquid driving device 103 to drive the sorting fluid to flow into the microfluidic sorting chip 102 from the third inlet based on the image recognition result of the mixture.

[0049] It should be noted that the microparticles to be sorted are the sorting objects of the microfluidic sorting system 101 provided by the present invention. The microparticles to be sorted include the target particles and non-target particles. The microfluidic sorting system 101 provided by the present invention can sort the sample liquid containing the microparticles to be sorted into a target liquid including the target particles and a waste liquid not including the target particles, thereby realizing the sorting of the microparticles to be sorted.

[0050] It should be noted that the microparticles in the embodiments of the present invention can be cells, viruses, enzymes, proteins and other types of microparticles. Cells can include single cells and organoid cells, etc.

[0051] It is understood that the particles to be sorted and the target particles in the embodiments of the present invention can be determined based on actual needs.

[0052] For example, in the embodiments of the present invention, the target particle is an organoid cell of a specific morphology, and the particles to be sorted include organoid cells of various morphologies. The microfluidic sorting system 101 provided by the present invention can sort the sample liquid containing organoid cells of various morphologies into a target liquid including the organoid cells of the above-mentioned specific morphology and a waste liquid not including the organoid cells of the above-mentioned specific morphology.

[0053] For example, in this embodiment of the invention, the target particle is a specific type of organoid cell, and the particles to be sorted include multiple types of organoid cells. The microfluidic sorting system 101 provided by this invention can sort a sample solution containing multiple types of organoid cells into a target solution including the aforementioned specific type of organoid cells and a waste solution not including the aforementioned specific type of organoid cells. In this embodiment of the invention, the particles to be sorted and the target particles are not specifically limited.

[0054] The microfluidic sorting chip 102 integrates basic operational units involved in fields such as chemistry and biology, including sample preparation, reaction, separation, detection, cell culture, sorting, and lysis, onto a single micrometer-scale chip. It utilizes the special properties of fluids within microchannels, such as laminar flow and diffusion effects, to manipulate and sort cells or particles.

[0055] The microfluidic sorting chip 102 is mainly composed of a chip substrate, fluid channels, microstructures, electrode layers, dielectric layers, and hydrophobic layers.

[0056] The chip substrate is the main structure of the microfluidic sorting chip 102, and is typically made of silicon, glass, or polymer materials (such as PMMA, PDMS, etc.). These materials have good processing performance and chemical stability, and can withstand the pressure, temperature, and chemical environment in microfluidic operation. Tiny fluid channels and reaction chambers are fabricated on the substrate to control and guide the flow of fluids and to sort cells or particles.

[0057] Fluid channels are the core component of the microfluidic sorting chip 102, used for transporting and mixing fluid samples. These channels are typically micrometer-sized, enabling precise control of the fluid flow direction and velocity. The channel design can be customized to specific sorting needs, such as incorporating reservoirs, main channels, focusing channels, detection channels, and collection channels, to achieve precise cell sorting and collection.

[0058] The microstructures include miniature components such as reaction chambers, valves, pumps, and sensors. In the microfluidic sorting chip 102, these microstructures are used to control the flow and mixing of fluids, as well as to monitor physical and chemical changes during the reaction process in real time. For example, valves and pumps can be used to control the flow rate and direction of the fluid to ensure uniform distribution and precise sorting of cells in the channels. Sensors are used to monitor parameters such as cell concentration and flow rate in the fluid so that sorting conditions can be adjusted in a timely manner.

[0059] Optionally, in this embodiment of the invention, the chip substrate of the microfluidic sorting chip 102 is polydimethylsiloxane (PDMS).

[0060] The microfluidic sorting chip 102 in this embodiment of the invention is configured with three inlet ports and two outlet ports, namely a first inlet port, a second inlet port and a third inlet port, as well as a first outlet port 309 and a second outlet port.

[0061] The first inlet of the microfluidic sorting chip 102 is used to inject sheath fluid into the microfluidic sorting chip 102. The second inlet of the microfluidic sorting chip 102 is used to inject sample solution containing particles to be sorted into the microfluidic sorting chip 102. The third inlet of the microfluidic sorting chip 102 is used to inject sorting solution into the microfluidic sorting chip 102.

[0062] The first outlet 309 of the microfluidic sorting chip 102 is used for the outflow of target liquid containing target particles. The second outlet of the microfluidic sorting chip 102 is used for the outflow of waste liquid that does not contain target particles.

[0063] In this embodiment of the invention, the controller 105 refers to a device capable of controlling and regulating a circuit or system according to a predetermined program or instruction. The controller 105 typically consists of a program counter, an instruction register, an instruction decoder, a timing generator, and an operation controller 105, etc. These parts work together to enable the controller 105 to issue commands and coordinate and direct the operation of the entire system.

[0064] In this embodiment of the invention, the controller 105 determines to trigger a cell sorting task upon receiving a control command instructing the triggering of the cell sorting task. The control command instructing the triggering of the cell sorting task may be input by a user.

[0065] Figure 2 This is a schematic diagram of the microfluidic sorting system provided by the present invention. Figure 2 As shown, when the controller 105 determines that the cell sorting task is triggered, it can control the liquid driving device 103 to drive the sample liquid containing the microparticles to be sorted to flow into the microfluidic sorting chip 102 from the second inlet of the microfluidic sorting chip 102, and control the liquid driving device 103 to drive the sheath fluid to flow into the microfluidic sorting chip 102 from the first inlet of the microfluidic sorting chip 102.

[0066] The sheath fluid and sample fluid flow into the microfluidic sorting chip 102 through the first and second inlets and mix within the microfluidic sorting chip 102 to form a mixture of sample fluid and sheath fluid.

[0067] The image acquisition device 104 can acquire images of the mixture in the microfluidic sorting chip 102 at a preset frequency, and can send the acquired images of the mixture to the controller 105.

[0068] After receiving the image of the mixture sent by the image acquisition device 104, the controller 105 can perform image recognition on the image of the mixture based on deep learning technology. If the target particles are identified in the image of the mixture, it determines that the mixture corresponding to the image of the mixture contains the target particles. Then, it can send a sorting signal to the liquid driving device 103. Upon receiving the sorting signal, the liquid driving device 103 drives the sorting liquid to flow into the microfluidic sorting chip 102 from the third inlet. Under the impact of the sorting liquid, the mixture flows through the first outlet 309 of the microfluidic sorting chip 102 into the collection bottle, thereby obtaining the target liquid containing the target particles in the collection bottle.

[0069] It is understandable that when the controller 105 does not control the liquid driving device 103 to drive the sorting liquid to flow into the microfluidic sorting chip 102 from the third inlet, the sample liquid is not impacted by the sorting liquid and will flow directly into the waste bottle through the second outlet of the microfluidic sorting chip 102.

[0070] It should be noted that the liquid driving device 103 in the embodiments of the present invention can drive the sample liquid, sheath liquid and sorting liquid into the microfluidic sorting chip 102 in a variety of ways. For example, the liquid driving device 103 can drive the sample liquid, sheath liquid and sorting liquid into the microfluidic sorting chip 102 by means of air pressure driving, gravity driving and mechanical driving.

[0071] It should be noted that the microfluidic sorting chip 102 in this embodiment of the invention can achieve mixing of sample liquid and sheath liquid through the fluid channel, and the mixture of sample liquid and sheath liquid flows out through the first sample outlet 309 of the microfluidic sorting chip 102 under the impact of the sorting liquid. The specific structure of the fluid channel in the microfluidic sorting chip 102 is not limited in this embodiment of the invention.

[0072] Optionally, the image acquisition device 104 in this embodiment of the invention may include a fluorescence microscope and a high-resolution high-speed camera (Fast camera, CMOS). After the light enters the detection optical path through the objective lens of the fluorescence microscope, it is split into two by a beam splitter on the detection optical path. After stray light is filtered out by a filter, a portion of the light is detected by the high-speed camera and converted into an image signal input to the controller 105.

[0073] It should be noted that the resolution of the high-speed camera and the preset frequency at which the high-speed camera acquires images of the mixture within the microfluidic sorting chip 102 in this embodiment of the invention can be determined based on prior knowledge and / or actual conditions. This embodiment of the invention does not impose specific limitations on the aforementioned preset frequency and the resolution of the high-speed camera.

[0074] Understandably, using a high-resolution, high-speed camera to capture images of the mixture within the microfluidic sorting chip 102 ensures that the captured images of the mixture are clear and distortion-free.

[0075] Optionally, in this embodiment of the invention, the preset frequency can be in the range of 400fps to 600fps, for example, the preset frequency can be 400fps, 500fps or 600fps.

[0076] The microfluidic sorting system in this invention includes a microfluidic sorting chip, a liquid driving device, an image acquisition device, and a controller. Based on the image acquisition device and deep learning technology, it can sort the desired target particles from the microparticles to be sorted without requiring fluorescent staining. This reduces the impact and damage caused by fluorescent staining on the microparticles, improves the accuracy and reliability of subsequent experiments, and significantly reduces the cost of cell sorting. Furthermore, compared to traditional microfluidic sorting systems that rely on fluorescent staining, the microfluidic sorting system provided by this invention, based on the image acquisition device and deep learning technology, can acquire multi-dimensional information such as cell size, roundness, edge features, and morphological features. This better meets the multi-dimensional cell sorting needs, facilitating more accurate and efficient sorting of target particles. It enables flexible and non-destructive cell sorting, making it particularly suitable for cell recovery and culture for further analysis. The cell sorting operation is simple and has broad application prospects.

[0077] Figure 3 This is a schematic diagram of the fluid channel structure in the microfluidic sorting chip of the microfluidic sorting system provided by the present invention. Figure 3 As shown, in an optional embodiment, the microfluidic sorting chip 102 includes: a first flow channel 301 and an annular flow channel 302;

[0078] The annular flow channel 302 is provided with two flow holes, namely the first flow hole 303 and the second flow hole 304.

[0079] The first flow passage 303 is connected to the first sample inlet 306 located outside the annular flow channel 302, and the second flow passage 304 is connected to the second sample inlet 305 located inside the annular flow channel 302.

[0080] The first end of the first flow channel 301 is connected to the second flow hole 304, so that the sheath liquid flowing in from the first inlet 306 covers the sample liquid flowing in from the second inlet 305 at the second flow hole 304, forming a mixed liquid that flows into the first flow channel 301.

[0081] The second end of the first flow channel 301 is connected to the third inlet 307, the first outlet 309, and the second outlet 308 of the microfluidic sorting chip 102, respectively. This allows the mixture to flow into the waste bottle through the second outlet 308 without being impacted by the sorting liquid flowing in from the third inlet 307 at the second end of the first flow channel 301. The mixture, under the impact of the sorting liquid flowing in from the third inlet 307 at the second end of the first flow channel 301, flows into the collection bottle through the first outlet 309.

[0082] Optionally, the annular flow channel 302 in the embodiments of the present invention can be a circular annular flow channel 302, an elliptical or rectangular annular flow channel 302, or a polygonal annular flow channel 302. Figure 3 The annular flow channel 302 shown is a polygonal annular flow channel 302. The specific shape of the annular flow channel 302 is not limited in this embodiment of the invention.

[0083] Preferably, the annular flow channel 302 in this embodiment of the invention is as follows: Figure 3 The polygon shown is composed of rectangles and triangles.

[0084] As an optional embodiment, the microfluidic sorting chip 102 further includes: a second channel 310, a third channel 311, a fourth channel 312, a fifth channel 313, and a sixth channel 314; the first channel 301 and the third channel 311 are coaxially arranged.

[0085] The second end of the first flow channel 301 is connected to the third inlet 307 through the second flow channel 310, the second end of the first flow channel 301 is connected to the second outlet 308 through the third flow channel 311, and the second end of the first flow channel 301 is connected to the first outlet 309 through the fourth flow channel 312.

[0086] The first injection port 306 is connected to the first flow passage 303 through the fifth flow channel 313, and the second injection port 305 is connected to the second flow passage 304 through the sixth flow channel 314.

[0087] The first end of the fourth flow channel 312 is connected to the second end of the first flow channel 301, and the second end of the fourth flow channel 312 is connected to the first sample outlet 309. The diameter of the first end of the fourth flow channel 312 is larger than the diameter of the second end of the fourth flow channel 312.

[0088] The diameter of the third flow channel 311 is larger than the diameter of the first end of the fourth flow channel 312.

[0089] Optionally, in this embodiment of the invention, the cross-sectional shape of the flow channel in the microfluidic sorting chip 102 can be circular, rectangular, trapezoidal, or semi-circular, etc. This embodiment of the invention does not limit the cross-sectional shape of the flow channel in the microfluidic sorting chip 102.

[0090] It is understood that the flow channels in the microfluidic sorting chip 102 in this embodiment of the invention include a first flow channel 301, a second flow channel 310, a third flow channel 311, a fourth flow channel 312, a fifth flow channel 313 and a sixth flow channel 314.

[0091] As an alternative embodiment, the diameter of the channels in the microfluidic sorting chip 102 is determined based on the diameter of the particles to be sorted.

[0092] Understandably, the diameter of the flow channel in the microfluidic sorting chip 102 is larger than the diameter of the particles to be sorted; for example, the diameter of the particles to be sorted is 180. In this case, the diameter of the flow channel in the microfluidic sorting chip 102 can be 300. .

[0093] It should be noted that the flow channels in the microfluidic sorting chip 102 in this embodiment of the invention can be designed differently to accommodate particles of different sizes (100-1000). The sorting of microfluidic sorting chips. For example, in the microfluidic sorting chip 102, the length and channel depth are both 30 mm. In this case, 30 can be achieved Single-cell sorting; the length and channel depth of the microfluidic sorting chip 102 are 300. In this case, 300 can be achieved Organoid cell sorting. By changing the size of the flow channels in the microfluidic sorting chip 102, the sorting of particles of different sizes can be achieved, making it applicable to various sizes, especially suitable for the sorting of large particles (such as organoid cells).

[0094] Optionally, the diameter of the first end of the fourth flow channel 312 can be determined based on the diameter of the second end of the fourth flow channel 312. For example, the diameter of the first end of the fourth flow channel 312 can be twice the diameter of the second end of the fourth flow channel 312, where the diameter of the second end of the fourth flow channel 312 is 300. In this case, the diameter of the first end of the fourth flow channel 312 can be 400. .

[0095] Optionally, such as Figure 3 As shown, the first end of the fourth flow channel 312 is funnel-shaped, and the diameter of the first end of the fourth flow channel 312 gradually decreases as it moves away from the second end of the first flow channel 301 until it becomes the same as the diameter of the second end of the fourth flow channel 312. The length of the first end of the fourth flow channel 312 can be determined based on prior knowledge and / or actual conditions. In this embodiment of the invention, the length of the first end of the fourth flow channel 312 is not limited.

[0096] The design of the fourth flow channel 312, with its first end diameter larger than the second end diameter, helps to gradually focus and accelerate the fluid during the sorting process, thereby improving the accuracy and efficiency of sorting. It also reduces fluid diffusion and loss within the flow channel, further improving reagent utilization.

[0097] Optionally, such as Figure 3 As shown, the first flow channel 301, the second flow channel 310, the third flow channel 311, the fifth flow channel 313 and the sixth flow channel 314 are all straight-through channels, while the fourth flow channel 312 is corner-shaped.

[0098] Optionally, such as Figure 3 As shown, the first flow channel 301 is located on the central axis of the annular flow channel 302, the fifth flow channel 313, the sixth flow channel 314, the first flow channel 301 and the third flow channel 311 are coaxially arranged, and the second flow channel 310 and the first end of the fourth flow channel 312 are coaxially arranged.

[0099] As an optional embodiment, the image acquisition device 104 is used to acquire images of the mixture in the first flow channel 301 at a preset frequency.

[0100] It should be noted that the image acquisition device 104 in this embodiment of the invention can be located near the first flow channel 301. The image acquisition device 104 can be used to acquire images of the mixture of sample liquid and sheath liquid in the first flow channel 301 at a preset frequency, and send the acquired images of the mixture to the controller 105.

[0101] The microfluidic sorting chip in this embodiment of the invention achieves the encapsulation of sample solution by sheath fluid through an annular flow channel. The two ends of the first flow channel are connected to the annular flow channel and the third sample inlet, the first sample outlet and the second sample outlet, respectively. This enables the separation of the mixture of sheath fluid and sample solution under the impact of the sorting liquid, thereby achieving the sorting of target particles. The microfluidic sorting chip has a simple structure and simple manufacturing process. It can sort target particles more accurately and efficiently, and can achieve flexible and non-destructive sorting of cells. It is particularly suitable for cell recovery and culture for further analysis and has great application prospects.

[0102] As an optional embodiment, the controller 105 is specifically used to input the image of the mixture into the cell recognition model when it receives the image of the mixture sent by the image acquisition device 104, obtain the cell recognition result output by the cell recognition model, and then control the liquid driving device 103 to drive the sorting liquid to flow from the third inlet 307 into the microfluidic sorting chip 102 when it is determined based on the cell recognition result that the predefined sorting conditions are met.

[0103] The cell recognition model is obtained by training based on sample images and cell recognition results of sample images; the cell recognition results of sample images are sample images labeled with target particles; the sorting conditions include that the image of the mixture contains target particles.

[0104] Specifically, when the controller 105 receives an image of the mixture sent by the image acquisition device 104, it can first perform image preprocessing on the image of the mixture to obtain a preprocessed image of the mixture. This image preprocessing may include image scaling and / or image cropping according to a preset image size, as well as image denoising and image filtering.

[0105] Figure 4 This is one of the images of the mixture in the microfluidic sorting system provided by the present invention. Figure 5 This is the second image of the mixture in the microfluidic sorting system provided by the present invention. Figure 4 These are images of pancreatic ductal adenocarcinoma cells in different gels. Figure 5 These are images of different cells within collagen.

[0106] After obtaining the image of the preprocessed mixture, the image of the preprocessed mixture can be input into the cell recognition model. The cell recognition model can then identify the target particles in the image of the preprocessed mixture, thereby obtaining the cell recognition results output by the cell recognition model.

[0107] It should be noted that the cell recognition result output by the cell recognition model in this embodiment of the invention is an image of a mixture labeled with target microparticles.

[0108] It should be noted that the cell recognition model in this embodiment of the invention can be built based on the YOLO v4 (You Only Look Once, Version 4) network model.

[0109] The YOLO v4 network model is a deep learning model for object detection that significantly improves detection accuracy while maintaining fast detection speed and is able to identify more object categories. The YOLO v4 network model treats object detection as a single regression problem, achieving object detection by dividing the image into a grid and predicting bounding boxes and class probabilities on each grid.

[0110] Compared to two-stage deep learning object detectors, the YOLO v4 network model can identify objects in images much faster. The YOLO v4 network model uses anchor boxes to detect object categories in images. For each anchor box, the YOLO v4 network model predicts the following three attributes: Joint Intersection (IoU)—predicting the objectivity score for each anchor box; Anchor Box Offset—optimizing the anchor box position; and Class Probability—predicting the class label assigned to each anchor box. The YOLO v4 network model runs deep learning on the input image to produce network predictions.

[0111] Optionally, due to 32 The YOLO v4 network model with a convolution kernel of 32 has a value of 14. Therefore, when the controller 105 receives the image of the mixture sent by the image acquisition device 104, it can output 1280... The image of the mixture of 730 changed to 448. 448, to improve the computational speed of the cell recognition model.

[0112] In this embodiment of the invention, an image including the target particles can be acquired under focusing conditions, and then images with poor image quality, unclear target particles, or incorrectly identified target particles can be deleted, thereby determining the retained images as sample images.

[0113] By labeling target particles in sample images, the sample images labeled with target particles can be identified as cell recognition results of sample images, thus obtaining a sample set for model training.

[0114] The sample images in the sample set were randomly divided into a 70% training subset, a 30% validation subset, and a 10% test subset. The initial model based on the YOLO v4 network model was trained, tested, and validated to obtain a trained cell recognition model.

[0115] It should be noted that when the controller 105 receives an image of the mixture sent by the image acquisition device 104, it can store the image data and the image acquisition time in a stack, and can call the image data in the stack according to the image acquisition time sequence through the MATLAB image processing program.

[0116] It should be noted that the sorting conditions in the embodiments of the present invention may also include the number of target particles in the image of the mixture exceeding a preset value, etc.

[0117] The controller 105 can determine whether the predefined sorting conditions are met based on the cell recognition results output by the cell recognition model. If the controller 105 determines that the sorting conditions are met, it can send a sorting signal to the liquid drive device 103. It can also send a sorting signal to the liquid drive device 103 after a delayed sorting delay period.

[0118] It should be noted that since it takes a certain amount of time for the mixed liquid after image acquisition to flow to the second end of the first flow channel 301, and it also takes a certain amount of time for the controller 105 to determine whether to send the liquid to the liquid driving device 103 based on the image of the mixed liquid, if the two times are equal, the controller 105 determines the time when the sorting conditions are met based on the image of the mixed liquid. The mixed liquid corresponding to the image of the mixed liquid just flows to the second end of the first flow channel 301. The controller 105 can determine the time when the sorting conditions are met based on the image of the mixed liquid as the sorting time, and then send a sorting signal to the liquid driving device 103 at the sorting time, so that the mixed liquid corresponding to the image of the mixed liquid can flow into the first sample outlet 309 under the impact of the sorting liquid.

[0119] However, if the first flow channel 301 is long or the flow rate of the mixture within the first flow channel 301 is slow, when the controller 105 determines that the sorting conditions are met based on the image of the mixture, the mixture corresponding to the image has not yet flowed to the second end of the first flow channel 301. Therefore, a sorting delay is required before sending the sorting signal to the liquid driving device 103. In this embodiment of the invention, the sorting delay can be calculated based on the length and diameter of the first flow channel 301, the length and diameter of the second flow channel 310, and the flow rate of the mixture within the first flow channel 301.

[0120] It should be noted that, in this embodiment of the invention, different labels can be added to the target particles in the sample image. After obtaining the trained cell recognition model, control commands can be used to instruct the cell recognition model to recognize target particles with specified labels.

[0121] As an optional embodiment, after the controller 105 controls the liquid driving device 103 to drive the sorting liquid into the microfluidic sorting chip 102 from the third inlet 307, it is also used to input the image of the mixture into the cell recognition model when it receives the image of the mixture sent by the image acquisition device 104, obtain the cell recognition result output by the cell recognition model, and then control the liquid driving device 103 to stop driving the sorting liquid into the microfluidic sorting chip 102 from the third inlet 307 when it is determined based on the cell recognition result that the sorting conditions are not met.

[0122] The controller in this embodiment of the invention can input the image of the mixture into the cell recognition model when it receives the image of the mixture sent by the image acquisition device 104, obtain the cell recognition result output by the cell recognition model, and then control the liquid driving device 103 to drive the sorting liquid to flow from the third inlet into the microfluidic sorting chip when it is determined that the predefined sorting conditions are met based on the cell recognition result. It can more accurately and efficiently identify whether the mixture contains target particles based on deep learning technology, thereby more accurately and efficiently determining the sorting time for cell sorting, better meeting the multi-dimensional cell sorting needs, and further improving the accuracy and efficiency of cell sorting.

[0123] As an optional embodiment, the liquid drive device 103 includes: a pressurization device;

[0124] The pressurizing device is used to pressurize the first sealed container containing sheath fluid and the second sealed container containing sample fluid in response to the control of the controller 105, so that the sheath fluid and the sample fluid flow into the first inlet 306 and the second inlet 305 respectively through the first pipe and the second pipe under pressure. The pressurizing device is also used to pressurize the third sealed container containing sorting liquid in response to the control of the controller 105, so that the sorting liquid flows into the third inlet 307 through the third pipe.

[0125] The first conduit connects the first sealed container to the first inlet 306; the second conduit connects the second sealed container to the second inlet 305; and the third conduit connects the third sealed container to the third inlet 307.

[0126] Figure 6 This is a schematic diagram illustrating the working principle of the liquid driving device in the microfluidic sorting system provided by this invention. Figure 6 As shown, the working principle of the liquid driving device 103 in this embodiment of the invention includes: a certain amount of solution is contained in a sealed container, and a certain pressure of gas is introduced into the sealed container. The solution in the sealed container will enter the microfluidic sorting chip 102 along the liquid pipe under the action of pressure.

[0127] Optionally, the pressurizing device in this embodiment of the invention may include a gas source and a gas pump. The gas source is connected to the gas pump, and the gas pump increases the pressure in the sealed container by injecting gas from the gas source into the sealed container.

[0128] Optionally, the gas source can be nitrogen or an air compressor. The air pump can be a Fluigent air pump.

[0129] The air pump can pressurize the first sealed container to a first pressure value, the second sealed container to a second pressure value, and the third sealed container to a third pressure value by injecting different volumes of gas into the first sealed container, the second sealed container, and the third sealed container, respectively.

[0130] Optionally, precision pressure sensors are provided at the first inlet 306, the second inlet 305, and the third inlet 307 to monitor the pressure at the first inlet 306, the second inlet 305, and the third inlet 307, thereby adjusting the volume of gas injected into the first sealed container, the second sealed container, and the third sealed container by the gas pump.

[0131] Optionally, a solenoid valve can be installed between the air pump and the sealed container, which can control the volume of gas input into the sealed container by the air pump.

[0132] The liquid driving device in this embodiment of the invention, through the control of the controller, can precisely pressurize a sealed container containing sheath fluid, sample fluid, and sorting fluid, ensuring that the liquid flows stably into the corresponding inlet under pressure. This achieves high-precision control of the liquid flow rate within the microfluidic sorting chip. By precisely controlling the flow rates of the sheath fluid, sample fluid, and sorting fluid within the microfluidic sorting chip, the analytical throughput and efficiency of the microfluidic sorting chip can be improved. The use of a sealed container and piping system can effectively prevent contamination and evaporation of the sheath fluid, sample fluid, and sorting fluid during the flow process, ensuring the purity and stability of the sheath fluid, sample fluid, and sorting fluid. The liquid driving device has a relatively simple structure, is relatively easy to maintain and service, reduces the operating cost of the microfluidic sorting system, and improves the reliability and service life of the microfluidic sorting system.

[0133] As an optional embodiment, the controller 105 is also configured to control a first pressure value applied by the pressurizing device to the first sealed container based on the radius and specific resistance of the fifth flow channel 313 and the annular flow channel 302 and the density of the sheath fluid; the controller 105 is also configured to control a second pressure value applied by the pressurizing device to the second sealed container based on the radius and specific resistance of the sixth flow channel 314 and the density of the sample fluid; and the controller 105 is also configured to control a third pressure value applied by the pressurizing device to the third sealed container based on the radius and specific resistance of the second flow channel 310 and the density of the sorting fluid.

[0134] It should be noted that the pressurizing device in this embodiment of the invention can generate different air pressures in response to the control of the controller 105.

[0135] Based on the radius and specific resistance of the fifth flow channel 313 and the annular flow channel 302, the density of the sheath fluid, and the expected flow velocity of the sheath fluid in the fifth flow channel 313 and the annular flow channel 302, the controller 105 can calculate the first pressure value to be pressurized into the first sealed container, and then control the pressurizing device to apply the first pressure value to the first sealed container, so that the sheath fluid flows from the first sealed container into the first inlet 306.

[0136] The first pressure value is calculated using the following formula:

[0137]

[0138] in, Indicates the first pressure value; This indicates the radius of the annular flow channel 302; This indicates the radius of the fifth flow channel 313; This indicates the specific drag of the annular flow channel 302; This indicates the specific drag of the fifth flow channel 313; Indicates the density of the sheath fluid; This indicates the expected flow rate of the sheath fluid in the fifth flow channel 313 and the annular flow channel 302.

[0139] Based on the radius and specific resistance of the sixth flow channel 314, the density of the sample liquid, and the expected flow rate of the sample liquid in the sixth flow channel 314, the controller 105 can calculate the second pressure value to be applied to the second sealed container, and then control the pressurizing device to apply the second pressure value to the second sealed container, so that the sample liquid flows from the second sealed container into the second inlet 305.

[0140] The second pressure value is calculated using the following formula:

[0141]

[0142] in, Indicates the second pressure value; This indicates the radius of the sixth flow channel 314; This indicates the specific drag of the sixth flow channel 314; Indicates the density of the sample solution; This indicates the expected flow rate of the sample solution in the sixth channel 314.

[0143] Based on the radius and specific resistance of the second flow channel 310, the density of the sorting liquid, and the expected flow rate of the sorting liquid in the second flow channel 310, the controller 105 can calculate the third pressure value to be applied to the third sealed container. Then, when it is determined that the sorting conditions are met, the controller 105 can control the pressurizing device to apply the third pressure value to the third sealed container, so that the sorting liquid flows from the third sealed container into the third inlet 307.

[0144] The third pressure value is calculated using the following formula:

[0145]

[0146] in, This indicates the third pressure value; This indicates the radius of the second flow channel 310; This indicates the specific drag of the second flow channel 310; Indicates the density of the separating solution; This indicates the expected flow rate of the sorting fluid in the second channel 310.

[0147] It should be noted that the expected flow rates of the sheath fluid in the fifth channel 313 and the annular channel 302, the expected flow rates of the sample fluid in the sixth channel 314, and the expected flow rates of the sorting fluid in the second channel 310 can be determined based on prior knowledge and / or actual conditions.

[0148] The microfluidic sorting chip in this embodiment of the invention highly integrates multiple inlets, outlets, channels, and flow orifices onto a single tiny chip, achieving a high degree of integration in sample processing. This design not only reduces the chip's size but also improves its portability and operability. Through different channels and connection structures, the microfluidic sorting chip modularizes the functions of sample liquid coating, mixing, sorting, and collection, allowing each functional unit to operate independently without interference. This design enhances the chip's flexibility and versatility, facilitating selection and adjustment by users according to experimental needs.

[0149] In this embodiment of the invention, the microfluidic sorting chip encapsulates the sample liquid flowing in from the second inlet at the second flow passage through a sheath fluid entering from the first inlet, forming a mixture. This helps reduce the diffusion and loss of the sample liquid in the flow channel, improving sorting accuracy. Simultaneously, the sheath fluid also protects and stabilizes the sample liquid, preventing damage during the sorting process. At the second end of the first flow channel, the mixture is impacted by the sorting liquid flowing in from the third inlet, thereby achieving the sorting of the particles. This impact method is highly efficient and rapid, significantly improving sorting efficiency. Furthermore, by adjusting the flow rate and impact force of the sorting liquid, precise sorting of particles of different sizes and properties can be achieved.

[0150] The flow channel design in the microfluidic sorting chip of this invention has been carefully optimized, including the size, curvature and connection structure of the flow channel, to ensure smooth flow and uniform distribution of fluid in the flow channel, which helps to reduce fluid resistance and loss in the flow channel and improve fluid utilization and sorting efficiency.

[0151] Due to the integrated and miniaturized design of the microfluidic sorting chip in this embodiment of the invention, the required amount of reagents is greatly reduced, which helps to lower experimental costs and reduce environmental pollution. Furthermore, by precisely controlling the amount and flow rate of reagents, the accuracy and reliability of the experiment can be further improved.

[0152] The microfluidic sorting chip in this invention is small in size, easy to carry and store, and can be used in different experimental environments, improving the flexibility and convenience of experiments. Due to the highly integrated and automated design of the microfluidic sorting chip, experimental operations become simple and easy to understand. Users only need to add samples and reagents according to the specified steps to automatically complete the sorting operation. This design reduces the requirements for operator expertise and experience, enabling more people to use this chip for experiments and research.

[0153] Figure 7 This is a summary diagram of the control parameters in the microfluidic sorting system provided by this invention. For example... Figure 7 As shown, the control parameters in the microfluidic sorting system 101 include sample loading parameters, image parameters, and sorting parameters. The sample loading parameters include a first pressure value applied by the pressurizing device to the first sealed container, a second pressure value applied by the pressurizing device to the second sealed container, and a third pressure value applied by the pressurizing device to the third sealed container. The image parameters may include a preset frequency, image size, and the label of the target particle. The sorting parameters may include a sorting delay time and a sorting time.

[0154] In the microfluidic sorting system 101 provided by this invention, a microfluidic sorting chip 102 is located below the bright field light source. The flow of sample liquid, sheath liquid, and sorting liquid in the microfluidic sorting chip 102 is controlled by a liquid driving device 103. Below the microfluidic sorting chip 102 is an optical module, which consists of an objective lens and a filter. The fluorescence field light source is located in the middle of the image acquisition device 104, and a controller 105 is located to the right. The controller 105 can be divided into a host computer and a slave computer. The host computer is mainly used to control the image acquisition device 104 to acquire, store, and process images of the mixed liquid. The slave computer is implemented by a printed circuit board (PCB) and mainly drives the liquid driving device 103 and the optical path switching device by receiving control commands from the host computer. The microfluidic sorting system 101 has the advantages of simple operation, fast detection, and high accuracy.

[0155] As an alternative embodiment, when the microparticles to be sorted are cells, the cells to be sorted live in a cell matrix.

[0156] In the embodiments of the present invention, the microparticles to be sorted live in the cell matrix, which enables aqueous phase sorting of the microparticles to be sorted. It is better suited for the application of the microparticles that cannot be sorted in the oil phase, and can be used for the sorting of various biological components (such as cells, viruses, enzymes, proteins, etc.) in the aqueous phase at the same time.

[0157] As an optional embodiment, the microfluidic sorting system 101 further includes a display device; the display device is used to display images of the mixture and other content in response to the control of the controller 105.

[0158] By performing physical simulation on the structure of the microfluidic sorting chip 102 in the microfluidic sorting system 101, it can be verified that the microfluidic sorting chip 102 has no structural defects or other potential design defects in terms of fluid mechanics. This allows for optimization of the design of the microfluidic sorting chip 102, ensuring the stability and reliability of the microfluidic sorting system 101 in practical applications. Figure 8 This is one of the simulation results of the microfluidic sorting chip provided by the present invention. Figure 9 This is the second schematic diagram illustrating the simulation results of the microfluidic sorting chip provided by this invention. The simulation results of the flow control sorting chip are as follows: Figure 8 and Figure 9 As shown.

[0159] Regarding the software design of the controller 105, a driver program was written to enable seamless integration between the controller 105 and the liquid driving device 103, ensuring accurate, rapid, and safe flow of sample liquid, sheath fluid, and sorting fluid into the microfluidic sorting chip 102. A user-friendly visual interface was developed, allowing operators to intuitively control and monitor the entire sorting process.

[0160] The performance of the microfluidic sorting system 101 provided by this invention was verified through sorting experiments, including sorting efficiency, accuracy, and repeatability. Based on the experimental feedback, the design, hardware configuration, and software algorithm of the microfluidic sorting system 101 were continuously adjusted and optimized until the microfluidic sorting system 101 achieved the expected performance indicators.

[0161] The microfluidic sorting system 101 provided by the present invention can determine the diameter of the flow channel in the microfluidic sorting chip 102 based on the diameter of the microparticle to be sorted, and is suitable for sorting microparticles of different sizes, such as sorting single cells and organoid cells respectively.

[0162] The microfluidic sorting system 101 provided by the present invention uses an air pump pressurization method to control the flow of sheath fluid, sample fluid and sorting fluid in microfluidic sorting chip 102, which can realize aqueous phase sorting and is suitable for sorting various biological components (such as cells, viruses, enzymes, proteins, etc.) in the aqueous phase at the same time.

[0163] The microfluidic sorting system 101 provided by the present invention has the characteristics of multiple parameters, and can provide high-content images of cell morphology and intercellular interactions. Combined with image features, it can quickly analyze and sort cell populations.

[0164] The microfluidic sorting system 101 provided by this invention develops real-time imaging technology to achieve high-speed, real-time imaging and flexible sorting at the cell level.

[0165] The microfluidic sorting system 101 provided by this invention is based on deep continuation-assisted processing of high-throughput data, and is particularly suitable for various omics analyses and massive database research.

[0166] The microfluidic sorting system 101 provided by the present invention can achieve flexible and non-destructive sorting, and is particularly suitable for cell recovery and culture for further analysis.

[0167] The microfluidic sorting system provided by this invention integrates a microfluidic sorting chip and a liquid driving device to form a complete sorting system. This highly integrated design makes the system more compact, lightweight, and easy to carry and operate. The microfluidic sorting system achieves automatic inflow and mixing of sample solution, sheath fluid, and sorting solution through the pressurization action of the liquid driving device. Users only need to set the relevant parameters, and the microfluidic sorting system can automatically complete the sorting process, greatly reducing operational difficulty and labor intensity. The microfluidic sorting system can adjust parameters such as the flow rate and pressure of the sample solution, sheath fluid, and sorting solution according to different experimental needs to adapt to different sorting conditions and particle characteristics. The design of the microfluidic sorting chip is easy to expand and modify; more inlets, outlets, and channels can be added to meet more complex sorting requirements. Simultaneously, the liquid driving device can also be upgraded and expanded as needed to improve the system's performance and functionality. The microfluidic sorting system has a reasonable structural design, making it easy to disassemble and clean. This helps extend the system's service life and maintain its good working condition. The liquid-driven device employs mature technologies such as pressurization devices and air pumps, ensuring stable and reliable performance. The microfluidic sorting system can also be equipped with corresponding safety protection measures, such as pressure sensors and overcurrent protection, to ensure safe system operation.

[0168] Figure 10 This is a schematic flowchart of the microfluidic sorting method provided by the present invention. The microfluidic sorting method provided by the present invention is implemented based on the microfluidic sorting system 101 described above. Figure 10 As shown, the method includes the following steps: Step 1001: When it is determined that the cell sorting task is triggered, the sheath fluid and the sample solution containing the particles to be sorted are driven to flow into the microfluidic sorting chip from the first inlet and the second inlet, respectively, so that the sample solution and the sheath fluid are mixed in the microfluidic sorting chip.

[0169] Step 1002: Obtain an image of the mixture of sample solution and sheath fluid within the microfluidic sorting chip;

[0170] Step 1003: Perform image recognition on the image of the mixture, and then, based on the image recognition result of the mixture, determine whether to drive the sorting liquid to flow into the microfluidic sorting chip from the third inlet. When the sorting liquid is not driven to flow into the microfluidic sorting chip from the third inlet, the mixture flows to the waste bottle through the second outlet of the microfluidic sorting chip. When the sorting liquid is driven to flow into the microfluidic sorting chip from the third inlet, the mixture flows to the collection bottle through the first outlet of the microfluidic sorting chip.

[0171] It should be noted that the execution entity of the microfluidic sorting method in the embodiments of the present invention can be the controller 105 in the microfluidic sorting system 101. The microfluidic sorting method provided by the present invention is implemented based on the microfluidic sorting system 101 described above. The specific execution steps of the microfluidic sorting method can be found in the contents of the above embodiments, and will not be repeated in the embodiments of the present invention.

[0172] In this embodiment of the invention, when a cell sorting task is triggered, sheath fluid and sample solution containing particles to be sorted are driven into the microfluidic sorting chip through the first and second inlets, respectively. This allows the sample solution and sheath fluid to mix within the microfluidic sorting chip. An image of the mixture is acquired within the microfluidic sorting chip, and image recognition is performed on the image. Based on the recognition result, it is determined whether to drive the sorting fluid into the microfluidic sorting chip through the third inlet. When the sorting fluid is not driven into the microfluidic sorting chip through the third inlet, the mixture flows to a waste bottle through the second outlet. When the sorting fluid is driven into the microfluidic sorting chip through the third inlet, the mixture flows to a collection bottle through the first outlet. This process can be achieved through the image... Using image acquisition and deep learning technologies, this invention enables the separation of target microparticles from unsorted microparticles without the need for fluorescent staining. This reduces the impact and damage caused by fluorescent staining, improves the accuracy and reliability of subsequent experiments, and significantly reduces the cost of cell sorting. Furthermore, compared to traditional microfluidic sorting systems that rely on fluorescent staining, the microfluidic sorting system provided by this invention, based on image acquisition devices and deep learning technology, can acquire multi-dimensional information such as cell size, roundness, edge features, and morphological characteristics. This better meets the multi-dimensional cell sorting requirements, facilitating more accurate and efficient sorting of target microparticles. It enables flexible and non-destructive cell sorting, making it particularly suitable for cell recovery and culture for further analysis. The cell sorting operation is simple and has broad application prospects.

[0173] Figure 11 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 11As shown, the electronic device may include: a processor 1110, a communications interface 1120, a memory 1130, and a communications bus 1140, wherein the processor 1110, the communications interface 1120, and the memory 1130 communicate with each other through the communications bus 1140. The processor 1110 can call logic instructions in the memory 1130 to execute a microfluidic sorting method, which includes: when a cell sorting task is determined to be triggered, driving sheath fluid and sample fluid containing particles to be sorted to flow into the microfluidic sorting chip from the first and second inlets, respectively, so that the sample fluid and sheath fluid are mixed within the microfluidic sorting chip; acquiring an image of the mixture of sample fluid and sheath fluid within the microfluidic sorting chip; performing image recognition on the image of the mixture, and then, based on the recognition result of the image of the mixture, determining whether to drive the sorting fluid to flow into the microfluidic sorting chip from the third inlet, such that if the sorting fluid is not driven to flow into the microfluidic sorting chip from the third inlet, the mixture flows to a waste bottle through the second outlet of the microfluidic sorting chip; and if the sorting fluid is driven to flow into the microfluidic sorting chip from the third inlet, the mixture flows to a collection bottle through the first outlet of the microfluidic sorting chip.

[0174] Furthermore, the logical instructions in the aforementioned memory 1130 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0175] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the microfluidic sorting method provided by the above methods. The method includes: when a cell sorting task is determined to be triggered, driving sheath fluid and a sample solution containing particles to be sorted to flow into the microfluidic sorting chip from a first inlet and a second inlet, respectively, so that the sample solution and the sheath fluid are mixed within the microfluidic sorting chip; obtaining microfluidic sorted particles. An image of the mixture of sample solution and sheath fluid inside the chip; image recognition is performed on the image of the mixture, and based on the recognition result, it is determined whether to drive the sorting liquid to flow into the microfluidic sorting chip from the third inlet. When the sorting liquid is not driven to flow into the microfluidic sorting chip from the third inlet, the mixture flows to the waste bottle through the second outlet of the microfluidic sorting chip. When the sorting liquid is driven to flow into the microfluidic sorting chip from the third inlet, the mixture flows to the collection bottle through the first outlet of the microfluidic sorting chip.

[0176] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the microfluidic sorting method provided by the methods described above. The method includes: upon determining that a cell sorting task is triggered, driving sheath fluid and a sample solution containing particles to be sorted to flow into the microfluidic sorting chip from a first inlet and a second inlet, respectively, so that the sample solution and sheath fluid are mixed within the microfluidic sorting chip; and obtaining a diagram of the mixture of sample solution and sheath fluid within the microfluidic sorting chip. For example, image recognition is performed on the image of the mixture, and based on the recognition result, it is determined whether to drive the sorting liquid to flow into the microfluidic sorting chip from the third inlet. When the sorting liquid is not driven to flow into the microfluidic sorting chip from the third inlet, the mixture flows to the waste bottle through the second outlet of the microfluidic sorting chip. When the sorting liquid is driven to flow into the microfluidic sorting chip from the third inlet, the mixture flows to the collection bottle through the first outlet of the microfluidic sorting chip.

[0177] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0178] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0179] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A microfluidic sorting system, characterized in that, include: The microfluidic sorting chip, the liquid driving device, the image acquisition device, and the controller are electrically connected, wherein the liquid driving device and the image acquisition device are electrically connected. The liquid driving device is used in response to the control of the controller to drive the sample liquid containing the particles to be sorted into the microfluidic sorting chip from the second inlet of the microfluidic sorting chip, drive the sheath fluid into the microfluidic sorting chip from the first inlet of the microfluidic sorting chip, and drive the sorting liquid into the microfluidic sorting chip from the third inlet of the microfluidic sorting chip, so that the sample liquid and the sheath fluid are mixed in the microfluidic sorting chip. Under the impact of the sorting liquid, the mixture of the sample liquid and the sheath fluid flows into the collection bottle through the first outlet of the microfluidic sorting chip. The particles to be sorted include target particles, which include cells, viruses, enzymes and proteins. The cells include single cells and organoid cells. The image acquisition device is used to acquire images of the mixture in the microfluidic sorting chip at a preset frequency, and send the acquired images of the mixture to the controller. The image acquisition device includes a fluorescence microscope and a high-speed camera. The controller is used to identify the image of the mixture and determine whether to drive the sorting liquid to flow in from the third inlet; The microfluidic sorting chip includes: a first channel, an annular channel, a second channel, a third channel, and a fourth channel; The second end of the first flow channel is connected to the third sample inlet through the second flow channel, the second end of the first flow channel is connected to the second sample outlet through the third flow channel, and the second end of the first flow channel is connected to the first sample outlet through the fourth flow channel; The first end of the fourth flow channel is connected to the second end of the first flow channel, the second end of the fourth flow channel is connected to the first sample outlet, and the diameter of the first end of the fourth flow channel is larger than the diameter of the second end of the fourth flow channel. The diameter of the third flow channel is larger than the diameter of the first end of the fourth flow channel; The first end of the fourth flow channel is coaxially arranged with the second flow channel, the fourth flow channel is corner-shaped, and the first end of the fourth flow channel is funnel-shaped; The liquid-driven device includes: a pressurizing device; the pressurizing device includes an air source and an air pump; the pressurizing device is used to apply a third pressure value to a third sealed container containing the sorting liquid in response to the control of the controller, the third pressure value being calculated by the following formula: ； in, This indicates the third pressure value; Indicates the radius of the second flow channel; Indicates the specific drag of the second flow channel; Indicates the density of the separating solution; This indicates the expected flow rate of the sorting fluid in the second flow channel.

2. The microfluidic sorting system according to claim 1, characterized in that, The controller is specifically used to input the image of the mixture into the cell recognition model when it receives the image of the mixture sent by the image acquisition device, obtain the cell recognition result output by the cell recognition model, and then control the liquid driving device to drive the sorting liquid to flow from the third inlet into the microfluidic sorting chip when it is determined based on the cell recognition result that the predefined sorting conditions are met. The cell recognition model is obtained by training based on sample images and the cell recognition results of the sample images; the cell recognition results of the sample images are sample images labeled with the target particles; the sorting conditions include the inclusion of the target particles in the image of the mixture.

3. The microfluidic sorting system according to claim 1, characterized in that, The pressurizing device, in response to the control of the controller, pressurizes the first sealed container containing the sheath fluid and the second sealed container containing the sample solution, respectively, so that the sheath fluid and the sample solution flow under pressure into the first inlet and the second inlet, respectively, through the first pipe and the second pipe. The pressurizing device is also used to pressurize the third sealed container containing the sorting liquid in response to the control of the controller, so that the sorting liquid flows into the third inlet through the third pipe; The first conduit is used to connect the first sealed container and the first inlet; the second conduit is used to connect the second sealed container and the second inlet; and the third conduit is used to connect the third sealed container and the third inlet.

4. The microfluidic sorting system according to claim 1, characterized in that, The annular flow channel is provided with two flow holes, namely the first flow hole and the second flow hole; The first flow passage is connected to the first sample inlet located outside the annular flow channel, and the second flow passage is connected to the second sample inlet located inside the annular flow channel. The first end of the first flow channel is connected to the second flow hole, so that the sheath fluid flowing in from the first inlet covers the sample fluid flowing in from the second inlet at the second flow hole, forming the mixture flowing into the first flow channel; The second end of the first flow channel is connected to the third inlet, the first outlet, and the second outlet of the microfluidic sorting chip, respectively, so that the mixture flows into the waste bottle through the second outlet without being impacted by the sorting liquid flowing in from the third inlet at the second end of the first flow channel, and the mixture flows into the collection bottle through the first outlet under the impact of the sorting liquid flowing in from the third inlet at the second end of the first flow channel.

5. The microfluidic sorting system according to claim 3, characterized in that, Also includes: The fifth and sixth flow channels; the first flow channel and the third flow channel are coaxially arranged; The first injection port is connected to the first flow hole through the fifth flow channel, and the second injection port is connected to the second flow hole through the sixth flow channel.

6. The microfluidic sorting system according to claim 5, characterized in that, The controller is further configured to control a first pressure value applied by the pressurizing device to the first sealed container based on the radius and specific resistance of the fifth flow channel and the annular flow channel and the density of the sheath fluid; the controller is further configured to control a second pressure value applied by the pressurizing device to the second sealed container based on the radius and specific resistance of the sixth flow channel and the density of the sample fluid; and the controller is further configured to control a third pressure value applied by the pressurizing device to the third sealed container based on the radius and specific resistance of the second flow channel and the density of the sorting fluid.

7. The microfluidic sorting system according to claim 6, characterized in that, The diameter of the flow channel in the microfluidic sorting chip is determined based on the diameter of the particles to be sorted.

8. The microfluidic sorting system according to claim 2, characterized in that, After controlling the liquid driving device to drive the sorting liquid into the microfluidic sorting chip from the third inlet, the controller is also used to input the image of the mixture into the cell recognition model when it receives the image of the mixture sent by the image acquisition device, obtain the cell recognition result output by the cell recognition model, and then control the liquid driving device to stop driving the sorting liquid into the microfluidic sorting chip from the third inlet when it is determined based on the cell recognition result that the sorting conditions are not met.

9. The microfluidic sorting system according to any one of claims 1 to 8, characterized in that, In the case where the microparticles to be sorted are cells, the cells to be sorted live in a cell matrix.

10. A microfluidic sorting method based on the microfluidic sorting system as described in claim 3, characterized in that, include: When the cell sorting task is triggered, the sample solution containing the particles to be sorted is driven to flow into the microfluidic sorting chip from the second inlet of the microfluidic sorting chip, and the sheath fluid is driven to flow into the microfluidic sorting chip from the first inlet of the microfluidic sorting chip, so that the sample solution and the sheath fluid are mixed in the microfluidic sorting chip; Acquire an image of the mixture of the sample solution and the sheath fluid within the microfluidic sorting chip; Image recognition is performed on the image of the mixture, and based on the recognition result, it is determined whether to drive the sorting liquid to flow into the microfluidic sorting chip from the third inlet. When the sorting liquid is not driven to flow into the microfluidic sorting chip from the third inlet, the mixture flows to the waste bottle through the second outlet of the microfluidic sorting chip. When the sorting liquid is driven to flow into the microfluidic sorting chip from the third inlet, the mixture flows to the collection bottle through the first outlet of the microfluidic sorting chip. The pressurizing device is controlled to pressurize the first sealed container to a first pressure value, the second sealed container to a second pressure value, and the third sealed container to a third pressure value. The first pressure value is calculated using the following formula: ； in, Indicates the first pressure value; This indicates the radius of the annular flow channel 302; This indicates the radius of the fifth flow channel 313; This indicates the specific drag of the annular flow channel 302; This indicates the specific drag of the fifth flow channel 313; Indicates the density of the sheath fluid; This indicates the expected flow velocity of the sheath fluid in the fifth flow channel 313 and the annular flow channel 302; The second pressure value is calculated using the following formula: ； in, Indicates the second pressure value; This indicates the radius of the sixth flow channel 314; This indicates the specific drag of the sixth flow channel 314; Indicates the density of the sample solution; This indicates the expected flow rate of the sample solution in the sixth channel 314.

Images