Automatic sound tweezer control and detection integrated system and method based on image recognition

Abstract

The invention discloses an automatic sound tweezer control and detection integrated system and method based on image recognition, and the system comprises a control system and a liquid drop generation device, and the control system is provided with an image recognition program; a flow channel is formed in the middle of the substrate, and the flow channel is connected with the liquid drop generation device; the substrate comprises a piezoelectric substrate, a plurality of interdigital electrodes are plated on the piezoelectric substrate and on the two sides of the flow channel and inside the flow channel, and the interdigital electrodes are connected with a control system. Initial liquid drops containing fluorescent particles are generated by the liquid drop generation device, the initial liquid drops are split into a plurality of small liquid drops through the flow channel, and high-flux generation of the small liquid drops is achieved; the interdigital electrode is controlled by the control system to generate surface acoustic waves to drive the small liquid drops to array and / or move in the flow channel, so that the operation of the small liquid drops is realized; meanwhile, the substrate is placed under a microscope, image recognition of the fluorescent particles is conducted through a control system so as to control opening or closing of the interdigital electrode of the flow channel where the small liquid drops are located, and detection and sorting of the small liquid drops are achieved.

Description

Technical Field

[0001] The invention relates to the field of image recognition and surface acoustic wave technology, and in particular to an automatic acoustic tweezers manipulation and detection integrated system based on image recognition. Background Art

[0002] Microfluidics is a technology that precisely controls and manipulates microscale fluids, especially submicron structures. Microfluidics is a new technology for automatically controlling fluids (liquids and gases), which covers fluid flow control at the micron scale. It applies this technology to fluid control tasks in medical instruments, analytical instruments, biosensors, micro-nano technologies, chemical reactors, and medical information systems, which can effectively improve performance, reduce costs, improve reliability, and increase flexibility and operability.

[0003] Acoustofluidics refers to the fusion of acoustics and microfluidics. Acoustofluidics can manipulate fluids in space and time by exploiting the interaction of sound waves with solids, liquids and / or gases in confined spaces for active and intensive mixing. Among the various examples of the convergence of dynamics and microfluidics, acoustically integrated microfluidic (i.e., acoustofluidic) mixers have attracted considerable attention due to their ability to manipulate fluids with high precision, adjustable speed, and cost-effectiveness, as well as contactless and non-invasively.

[0004] In recent years, cell separation and enrichment technologies have received extensive attention. Studies have confirmed that cancer cell separation and enrichment technologies can provide cell sources for the development of targeted drugs. In-depth biomolecular analysis of these cells can provide insights into tumor development, expected drug treatment responses, diagnosis of tumor metastasis, and prognostic evaluation and identification of diseased cells, paving the way for more personalized treatment. Therefore, cell sorting and purification have become key issues. Microfluidics is the most commonly used and effective technology for cell sorting and enrichment. It can achieve high-speed, high-recovery, and high-purity recovery of cancer cells. It has the advantages of high throughput, low cost, fast speed, and simple operation. Surface acoustic standing wave technology can quickly and easily achieve non-contact cell separation, with high throughput, and the acoustic energy used is harmless to cells. The chip production is relatively simple and can be reused.

[0005] In recent years, with the development of artificial intelligence, the method of automatically classifying microdroplets and cells using image recognition with fluorescence detection has become increasingly popular. Deep learning methods can enable models to automatically extract knowledge hidden in large-scale data and perform high-speed classification or prediction based on high-quality features.

[0006] Therefore, how to combine acoustofluidics technology with image recognition processing of fluorescence detection to achieve automated selective separation of microdroplets and further realize the detection and separation of biological cells has become a technical problem that the applicant needs to solve urgently. Summary of the invention

[0007] The purpose of the present invention is to provide an integrated system for automated acoustic tweezer manipulation and detection based on image recognition. The present invention can achieve high-throughput generation of droplets, can not only manipulate droplets, but also detect and sort droplets, and has the advantages of small size, light weight, and automation.

[0008] The technical solution of the present invention is: an integrated system for automated acoustic tweezer manipulation and detection based on image recognition, comprising a control system and a droplet generation device, wherein the control system has an image recognition program; and further comprising a substrate, wherein a flow channel is provided in the middle of the substrate, and the flow channel is connected to the droplet generation device;

[0009] The substrate comprises a piezoelectric substrate, on which a plurality of interdigital electrodes are plated at positions on both sides of the flow channel and at positions inside the flow channel, and the interdigital electrodes are connected to a control system.

[0010] In the above-mentioned integrated system of automated acoustic tweezer manipulation and detection based on image recognition, the interdigital electrodes include array interdigital electrodes with multiple fingers arranged on both sides of the flow channel and sorting interdigital electrodes with multiple fingers arranged inside the flow channel.

[0011] In the aforementioned integrated system of automated acoustic tweezer manipulation and detection based on image recognition, the array interdigitated electrodes are electrodes of equal finger width, and the two electrodes are arranged symmetrically; the sorting interdigitated electrodes include focusing interdigitated electrodes, adsorption interdigitated electrodes and pushing interdigitated electrodes; the sorting interdigitated electrodes are arranged equidistantly in the transverse direction.

[0012] The aforementioned integrated system of automated acoustic tweezers manipulation and detection based on image recognition, wherein the flow channel includes an oil phase inlet, a water phase inlet, a droplet generation point, a droplet splitting point, an array electrode area, a sorting electrode area, and a plurality of supporting cylinders for preventing the flow channel from collapsing; the oil phase inlet and the water phase inlet are respectively connected to the droplet generation device; the droplet generation point is arranged at the connection point between the oil phase inlet and the water phase inlet; the droplet splitting point is connected to the downstream direction of the droplet generation point; the array electrode area is used to set the array finger electrodes; the sorting electrode area is used to set the sorting finger electrodes.

[0013] In the aforementioned integrated system of automated acoustic tweezer manipulation and detection based on image recognition, a detachable film is provided on the piezoelectric substrate and is located directly below the flow channel; the detachable film is completely consistent with the flow channel; a flow channel positioning lead is provided on the piezoelectric substrate; the detachable film is bonded to the piezoelectric substrate so that the edge of the flow channel is aligned with the flow channel positioning lead.

[0014] The aforementioned integrated system for automated acoustic tweezer manipulation and detection based on image recognition, the control system includes a first signal generator, a second signal generator, a first power amplifier, a second power amplifier, a microscope, a computer, a single-chip microcomputer and a relay; the computer is respectively connected to the microscope, the single-chip microcomputer and the first signal generator; the first signal generator is connected to the first power amplifier; the relay is respectively connected to the first power amplifier and the single-chip microcomputer; the second power amplifier is connected to the second power amplifier.

[0015] The aforementioned integrated system of automated acoustic tweezers manipulation and detection based on image recognition also includes a control substrate, the substrate is arranged on the control substrate, and a plurality of interfaces are arranged on the control substrate; the array finger electrodes are connected to the second power amplifier via the corresponding interfaces; the sorting finger electrodes are connected to the relay via the corresponding interfaces; the substrate on the control substrate is arranged at the observation position of the microscope.

[0016] In the aforementioned integrated system of automated acoustic tweezer manipulation and detection based on image recognition, the droplet generation device comprises an oil pump and a water pump; the oil pump is connected to the oil phase inlet of the flow channel; and the water pump is connected to the water phase inlet.

[0017] In the aforementioned integrated system of automated acoustic tweezers manipulation and detection based on image recognition, the piezoelectric substrate is made of Y-cut 128° lithium niobate; the material of the flow channel and the detachable film is polydimethylsiloxane; the thickness of the interdigitated electrodes is 200±20nm; the material of the interdigitated electrodes is a combination of one or more materials selected from aluminum, gold and silver.

[0018] The aforementioned method of the integrated system for automated acoustic tweezer manipulation and detection based on image recognition comprises the following steps: a droplet generating device generates an initial droplet containing fluorescent particles, and the initial droplet is split in equal proportion through a flow channel to form a plurality of small droplets, thereby realizing high-throughput generation of small droplets; a control system is used to control the finger electrodes to generate surface acoustic waves to drive the small droplets to form an array and / or move in the flow channel, thereby realizing manipulation of the small droplets; at the same time, the substrate is placed under a microscope, and the control system is used to perform image recognition of the fluorescent particles, so as to control the opening or closing of the finger electrodes in the flow channel where the small droplets are located, thereby realizing detection and sorting of the small droplets.

[0019] Compared with the prior art, the present invention uses a flow channel to make the droplets split into four (or eight) small droplets after being generated by two (or three) equal-proportion splits, thereby entering the subsequent flow channel, thereby achieving high-throughput generation of micro-droplets or small droplets, and then can detect and manipulate droplets as much as possible at one time, and the initial droplets contain fluorescent particles. After splitting multiple times, some micro-droplets no longer have fluorescent particles, and the other part still contains fluorescent particles, which is ready for later sorting. At the same time, the present invention is connected to the interdigital electrode electrode through a control substrate, which is used to control the interdigital electrode to generate acoustic surface waves to drive the droplets to array and / or move in the flow channel, thereby realizing the manipulation of small droplets; further, the substrate of the present invention is placed under a microscope, the microscope is connected to a computer, the computer has image recognition software and programming control of the output signal (frequency, amplitude, phase) of the first signal generator, and the sorting interdigital electrode is connected to a relay, the relay is connected to a single-chip microcomputer, and the single-chip microcomputer is connected to the computer, so that image recognition of fluorescence detection can be performed in real time, and the sorting interdigital electrode of the flow channel where it is located is controlled to be turned on or off, thereby realizing automatic detection and sorting. Therefore, the present invention can not only solve the problems of traditional droplet sorting, and has the advantages of small size and light weight, but also realize high-throughput, selective and automated detection and sorting of microdroplets and cells under an image recognition control system, and has a very broad application prospect. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a general system connection diagram in a specific embodiment of the present invention;

[0021] Figure 2 It is a schematic diagram of the front structure of the control substrate and the base in a specific embodiment of the present invention;

[0022] Figure 3 It is a schematic diagram of the front structure of a substrate (including a piezoelectric substrate and the first two interdigital electrodes) in a specific embodiment of the present invention;

[0023] Figure 4 It is a schematic diagram of the front structure of a substrate (including a piezoelectric substrate and a third type of interdigital electrode) in a specific embodiment of the present invention;

[0024] Figure 5 It is a schematic diagram of the front structure of a flow channel in a specific embodiment of the present invention;

[0025] Figure 6 It is a schematic diagram of the front structure of two other flow channels in a specific embodiment of the present invention;

[0026] Figure 7 It is a schematic diagram of the structure decomposition of a detachable film of a device in a specific embodiment of the present invention;

[0027] Figure 8 A diagram of the droplet generation and splitting process in a specific embodiment of the present invention;

[0028] Fig. 9 This is a flowchart of controlling the array interdigital electrodes in a specific embodiment of the present invention;

[0029] Fig.10 This is a working principle diagram of an array interdigitated electrode in a specific embodiment of the present invention;

[0030] Fig.11 It is a working flow chart of controlling the sorting interdigital electrodes 202 in the control system;

[0031] Fig.12 It is a working principle diagram of an imaging system in a specific embodiment of the present invention;

[0032] Fig.13 This is a flow chart of the image recognition principle in a specific embodiment of the present invention;

[0033] Fig.14 The working principle of the focusing sorting electrode;

[0034] Fig.15 It is the working principle of adsorption type sorting electrode;

[0035] Fig.16 It is the working principle of push-type sorting electrode;

[0036] Fig.17 FIG. 2 is another imaging schematic diagram in a specific embodiment of the present invention.

[0037] Reference numerals

[0038] 1. Lithium niobate piezoelectric substrate; 2. Electrode; 201. Array electrode; 202. Sorting electrode; 203. Array interdigital electrode flow channel positioning lead; 3. Flow channel; 301. Oil phase inlet; 302. Water phase inlet; 303. Droplet generation; 304. Droplet splitting; 305. Array electrode area; 306. Sorting electrode area; 307. Support cylinder; 4. Removable PDMS film; 5. Control substrate; 501. Interface; Control system devices: 601. First signal generator; 602. First power amplifier; 603. Microscope; 604. Computer; 605. Single chip microcomputer; 606. Relay; 607. Second signal generator; 608. Second power amplifier; 609. COMS CCD camera; 7. Wire; 8. Oil pump: 9. Water pump; 1001. Rayleigh wave; 1002. Acoustic valve; 1003. Acoustic trap. DETAILED DESCRIPTION

[0039] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not limited to the specific implementation methods.

[0040] Embodiment: An integrated system for automated acoustic tweezer manipulation and detection based on image recognition, such as Figure 1 As shown, it includes a control system and a droplet generating device, wherein the control system has an image recognition program, which is conventional image recognition software or algorithm and can be obtained commercially; it also includes a substrate, wherein a flow channel 3 is provided in the middle of the substrate, and the flow channel 3 is connected to the droplet generating device;

[0041] like Figure 2 As shown, the substrate includes a piezoelectric substrate 1, which is made of Y-cut 128° lithium niobate and presents a lithium niobate piezoelectric wafer. A plurality of interdigital electrodes 2 are plated on the piezoelectric substrate 1 and on both sides of the flow channel 3 and inside the flow channel 3; Figure 3 and Figure 4 As shown, the interdigitated electrode 2 includes an array interdigitated electrode 201 with multiple fingers at both sides of the detachable flow channel and four (or eight) sorting interdigitated electrodes 202 with multiple fingers at the internal position of the detachable flow channel, wherein the array interdigitated electrode 201 is an electrode of equal finger width, and the two electrodes are arranged symmetrically; the sorting interdigitated electrode 202 is a focusing interdigitated electrode (in other embodiments, the sorting interdigitated electrode 202 can be an adsorption interdigitated electrode or a push interdigitated electrode), and these four (or eight) electrodes are arranged equidistantly in the horizontal direction. The thickness of the interdigitated electrode 2 is 200±20nm. The material of the interdigitated electrode 2 is silver, and in other embodiments it can be a combination of one or more materials selected from aluminum, gold and silver.

[0042] like Figure 5 As shown, the piezoelectric substrate 1 is provided with a detachable film 4 located just below the flow channel 3, and the detachable film 4 is completely consistent with the flow channel 3, so as to realize the detachable replacement of the flow channel 3; Figure 5 A flow channel structure is given. Figure 6 Two other flow channel structures are given, and the three flow channel structures correspond to three types of sorting interdigital electrodes 202. In this embodiment, the flow channel 3 and the removable film 4 are made of polydimethylsiloxane (PDMS), and the material thickness of the flow channel 3 is 3000um-5000um. The flow channel 3 includes an oil phase inlet 301, a water phase inlet 302, a droplet generation point 303, a droplet splitting point 304, an array electrode area 305, a sorting electrode area 306 and a number of supporting cylinders 307 for preventing the flow channel from collapsing; the oil phase inlet 301 and the water phase inlet 302 are respectively connected to the droplet generation device; the droplet generation point 303 is set at the connection point between the oil phase inlet 301 and the water phase inlet 302; the droplet splitting point 304 is connected to the downstream direction of the droplet generation point 303, and is used to split the droplets; the boundary of the removable film 4 should be completely consistent with the flow channel 3 when it is cut; as shown in FIG. Figure 7As shown, the detachable film 4 is detachable and bonded to the piezoelectric substrate 1; Figure 4 As shown, the piezoelectric substrate 1 is provided with a flow channel positioning lead 203, and the flow channel 3 is bonded to the piezoelectric substrate 1 with the edge of the flow channel 3 aligned with the positioning lead 203 of the array finger electrode 201 as a standard, so that the array finger electrode 201 and the sorting finger electrode 202 are located in their respective areas.

[0043] like Figure 1 Shown and Figure 5 As shown, the droplet generation device includes an oil pump 8 and a water pump 9; the oil pump 8 is connected to the oil phase inlet 301 of the flow channel 3, and the water pump 9 is connected to the water phase inlet 302 of the flow channel 3; the oil pump 8 and the water pump 9 are used to pump oil and water into the flow channel 3, so as to generate oil-in-water droplets at the droplet generation location 303, and the oil pump 8 and the water pump 9 are both constant flow pumps, which can adjust the flow rate of oil and water according to specific needs. Different flow rates and oil and water flow rate ratios can generate droplets of different spacings and sizes. The droplet generation process is shown in FIG. Figure 8 As shown, the two oil phases squeeze the water phase to generate droplets at the droplet generation point 303, and then flow to the droplet splitting point 304 to split into two, thereby generating the final required micro droplets (small droplets).

[0044] like Figure 1As shown, the substrate is arranged on the control substrate 5, and a plurality of interfaces 501 are arranged around the control substrate 5; the control system includes a first signal generator 601, a second signal generator 607, a first power amplifier 602, a second power amplifier 608, a microscope 603, a computer 604, a single-chip microcomputer 605 and a relay 606; the computer 604 is provided with image recognition software or program; the computer 604 is respectively connected to the microscope 603, the single-chip microcomputer 605 and the first signal generator 601; the first signal generator 601 is connected to the first power amplifier 602; the relay 606 is respectively connected to the first power amplifier 602 and the single-chip microcomputer 605; the control The substrate on the control substrate 5 is set at the observation position of the microscope 603; wherein the sorting interdigital electrode 202 is connected to the relay 606 via the corresponding interface 501; the first power amplifier 602 and the second power amplifier 608 are respectively connected to a power supply, and the power supply is used to provide power for the power amplifier; the first signal generator 601 and the second signal generator 607 are respectively used to generate signals, the first power amplifier 602 and the second power amplifier 608 are used to amplify signals, and the output end of the first power amplifier 602 and the single-chip microcomputer 605 are connected to the input end of the relay 606; the output end of the relay 606 is connected to the interface 501 corresponding to the sorting interdigital electrode 202 in the control substrate, and the relay 606 acts as a switch. The second power amplifier 608 is connected to the second power amplifier 608. The array interdigital electrode 201 is connected to the second power amplifier 608 via the corresponding interface 501; all components are connected through wires 7.

[0045] The working process of the control system of this embodiment includes the working process of controlling the array interdigital electrodes and the working process of controlling the sorting interdigital electrodes; Fig. 9 The flowchart is a work flow chart for controlling the array of interdigital electrodes 201 in the control system; the second power amplifier 608 is connected to the power supply, the second signal generator 607 and the array of interdigital electrodes 201; the power supply is used to supply power, the second signal generator 607 is used to generate the signal required by the array of interdigital electrodes 201, and the signal is amplified by the second power amplifier 608, transmitted to the array of interdigital electrodes 201, and generates an inverse piezoelectric effect on the piezoelectric substrate 1, and the piezoelectric film generates a certain deformation and vibration to generate an acoustic wave, which is a Rayleigh wave 1001, which is a kind of standing wave. When the array of interdigital electrodes 201 on both sides of the flow channel 3 generate Rayleigh waves at the same time, the two waves will intersect, and the intersection is an acoustic pressure node, causing the droplets to be arranged at the acoustic pressure node, such as Fig.10 shown.

[0046] Fig.11The flowchart is a control system for controlling the sorting interdigital electrodes 202; similarly, the first power amplifier 602 is respectively connected to the power supply, the first signal generator 601 and the relay 606; the relay 606 is connected to the single-chip microcomputer 605 and the control substrate interface 501 of the sorting interdigital electrodes 202; the computer 604 is connected to the microscope 603, the single-chip microcomputer 605 and the signal generator 601; the image captured by the microscope 603 is transmitted to the computer 604, and the fluorescence recognition is performed by the image recognition software, so that the single-chip microcomputer 605 controls the relay 606 to be turned on or off, so that the signal amplified by the first power amplifier 602 is transmitted to the sorting interdigital electrodes 202, and an inverse piezoelectric effect is generated on the piezoelectric substrate 1, and the piezoelectric substrate 1 produces a certain deformation and vibration to generate sound waves, thereby driving the droplets to move; at the same time, the computer 604 can also program and control the output (frequency, phase, amplitude) of the first signal generator 601 through another software according to specific needs.

[0047] Fig.12 The working principle diagram of the imaging system is shown in Figure 6. The digital signal from the microscope 603 is transmitted to the microprocessor chip through the digital video port (DVP), and then the parallel data is converted into USB data packets and sent to the computer (i.e., the data processing PC end in the figure); then, the imaging data is processed by the computer image recognition software based on a self-designed program for fluorescence detection. In addition, there is a voltage regulator in the system to adjust the feedback power of the system.

[0048] Image recognition program (software) workflow Fig.13 As shown, at the beginning, relay 606 is in a completely closed state, and the microscope transmits the acquired real-time image to the computer. Before the droplets reach the sorting area, the computer performs fluorescence recognition through image recognition software. If a droplet with fluorescent particles is recognized, the output is "1", and the single-chip microcomputer 605 controls relay 606 to open, and the sorting fork electrode 202 generates sound waves to move the droplet to another outlet; if the droplet does not carry fluorescent particles, the output is "0", the relay 606 is closed, and the system continues to acquire images, thus ensuring real-time sorting of droplets; and the aforementioned droplet generation device can be used to control the droplet generation rate by controlling the flow rate of the two phases to ensure that the droplet spacing satisfies the previous droplet recognition completion and the next droplet enters the image acquisition area.

[0049] In this embodiment, the working principles of the three types of separation interdigital electrodes 202 are as follows: Figure 14-16 As shown, corresponding to Figure 5 as well as Figure 6 Three electrode structures shown in (a) and (b). Fig.14This is the working principle of the aforementioned focusing sorting electrode. When the relay 606 is closed, the electrode does not receive any signal, no sound waves are generated, and the droplets flow into the right outlet in the original direction. After receiving the signal, the focusing sorting electrode will generate sound waves, which are emitted toward the center of the circle and focused at the center of the circle, causing the liquid in the flow channel to vibrate and form a sound wave barrier - the sound valve 1002, thereby blocking the original travel path of the droplets and forcing the droplets to flow into the left outlet, thereby achieving sorting. Fig.15 This is the working principle of the aforementioned adsorption-type sorting electrode. When the relay 606 is closed, the electrode does not receive any signal, no sound waves are generated, and the droplets flow into the right outlet in the original direction; after receiving the signal, the adsorption-type sorting electrode generates sound waves in the two longitudinal directions. The sound waves push the liquid to flow out in the two longitudinal directions and flow in in the two lateral directions, forming a "butterfly"-shaped fluid, and two "acoustic traps" 1003 are formed on the left and right sides of the electrode where the liquid flows in and out. When the liquid flows in on both sides in the lateral direction, the droplets are adsorbed and finally adsorbed to the "acoustic trap", so that they finally flow out from the left outlet to complete the sorting. Fig.16 This is the working principle of the aforementioned push-type sorting electrode. When the relay 606 is closed, the electrode does not receive any signal, no sound waves are generated, and the droplets flow into the right outlet in the original direction. After receiving the signal, the push-type sorting electrode emits sound waves in four directions, two longitudinal and two lateral directions. The sound waves push the liquid to flow out in the four directions and flow into the four diagonal directions of the electrode. When the droplets travel to the left side of the electrode, they are pushed out by the sound waves, causing them to move to the left outlet, thereby completing the sorting.

[0050] Furthermore, the microscope in this embodiment can be replaced by a COMS CCD camera 609, such as Fig.12 As shown, it is installed directly below the chip to obtain images.

[0051] It can be seen that this embodiment provides an integrated system and method for automated acoustic tweezer manipulation and detection based on image recognition, which is used for the generation, arraying and sorting of droplets. When images of different droplets (containing fluorescent particles and not containing fluorescent particles) are captured, image recognition can be judged according to its algorithm, and the judgment result is transmitted to the single-chip microcomputer to control the opening or closing of the electrodes in each channel, which can not only realize automation, but also achieve real-time selective sorting.

[0052] In this embodiment, the design of the flow channel can make the droplets eventually be evenly divided into 4 or 8 micro-droplets, thereby achieving high-throughput droplet generation, and the size and spacing between the droplets can be controlled.

[0053] In this embodiment, an algorithm is used to perform fluorescence recognition on droplets. When the droplets are recognized, the computer outputs a corresponding signal, which is transmitted to the single-chip microcomputer to control the relay switch, and the algorithm is trained so that it can meet the needs of subsequent biological reagent sorting experiments.

[0054] In this embodiment, in addition to performing image recognition on the droplets, the computer can also perform programmable control on the output signal (frequency, amplitude, phase) of the signal amplifier, thereby adjusting different output signals as needed.

[0055] In summary, the present invention can not only solve the problems of traditional droplet sorting, and has the advantages of small size and light weight, but also can realize high-throughput, real-time, selective and automated detection and sorting of microdroplets and cells under an image recognition control system, and has a very broad application prospect.

Claims

1. An integrated system for automated acoustic tweezer manipulation and detection based on image recognition, comprising a control system and a droplet generation device, wherein the control system has an image recognition program; characterized in that: It also includes a substrate, a flow channel (3) is provided in the middle of the substrate, and the flow channel (3) is connected to the droplet generating device; The substrate comprises a piezoelectric substrate (1), on which a plurality of interdigital electrodes (2) are plated at positions on both sides of a flow channel (3) and at positions inside the flow channel (3), and the interdigital electrodes (2) are connected to a control system.

2. The automatic acoustic tweezers manipulation and detection integrated system based on image recognition according to claim 1 is characterized in that: The interdigitated electrode (2) comprises an array interdigitated electrode (201) having multiple fingers arranged at both sides of the flow channel (3) and a sorting interdigitated electrode (202) having multiple fingers arranged inside the flow channel (3).

3. The automatic acoustic tweezers manipulation and detection integrated system based on image recognition according to claim 2 is characterized in that: The array interdigital electrodes (201) are electrodes of equal finger width, and the two electrodes are arranged symmetrically; the sorting interdigital electrodes (202) include focusing interdigital electrodes, adsorption interdigital electrodes and push-type interdigital electrodes; the sorting interdigital electrodes (202) are arranged in a transversely equidistant manner.

4. The automatic acoustic tweezers manipulation and detection integrated system based on image recognition according to claim 2 is characterized in that: The flow channel (3) comprises an oil phase inlet (301), a water phase inlet (302), a droplet generation point (303), a droplet splitting point (304), an array electrode area (305), a sorting electrode area (306) and a plurality of supporting cylinders (307) for preventing the flow channel (3) from collapsing; the oil phase inlet (301) and the water phase inlet (302) are respectively connected to a droplet generation device; the droplet generation point (303) is arranged at the connection point between the oil phase inlet (301) and the water phase inlet (302); the droplet splitting point (304) is connected to the downstream direction of the droplet generation point (303); the array electrode area (305) is used to set the array interdigital electrode (201); and the sorting electrode area (306) is used to set the sorting interdigital electrode (202).

5. The automatic acoustic tweezers manipulation and detection integrated system based on image recognition according to claim 1 is characterized in that: The piezoelectric substrate (1) is provided with a detachable film (4) located directly below the flow channel (3); the detachable film (4) is completely consistent with the flow channel (3); the piezoelectric substrate (1) is provided with a flow channel positioning lead (203); the detachable film (4) is bonded to the piezoelectric substrate (1) based on the alignment of the edge of the flow channel (3) with the flow channel positioning lead (203).

6. The automatic acoustic tweezers manipulation and detection integrated system based on image recognition according to claim 2 is characterized in that: The control system comprises a first signal generator (601), a second signal generator (607), a first power amplifier (602), a second power amplifier (608), a microscope (603), a computer (604), a single-chip microcomputer (605) and a relay (606); the computer (604) is respectively connected to the microscope (603), the single-chip microcomputer (605) and the first signal generator (601); the first signal generator (601) is connected to the first power amplifier (602); the relay (606) is respectively connected to the first power amplifier (602) and the single-chip microcomputer (605); and the second power amplifier (608) is connected to the second power amplifier (608).

7. The automatic acoustic tweezers manipulation and detection integrated system based on image recognition according to claim 6 is characterized in that: It also comprises a control substrate (5), the substrate being arranged on the control substrate (5), and a plurality of interfaces (501) being arranged on the control substrate (5); the array interdigitated electrodes (201) being connected to a second power amplifier (608) via corresponding interfaces (501); the sorting interdigitated electrodes (202) being connected to a relay (606) via corresponding interfaces (501); and the substrate on the control substrate (5) being arranged at an observation position of a microscope (603).

8. The automatic acoustic tweezers manipulation and detection integrated system based on image recognition according to claim 4 is characterized in that: The droplet generating device comprises an oil pump (8) and a water pump (9); the oil pump (8) is connected to the oil phase inlet (301) of the flow channel (3); and the water pump (9) is connected to the water phase inlet (302).

9. The automatic acoustic tweezers manipulation and detection integrated system based on image recognition according to claim 5 is characterized in that: The piezoelectric substrate (1) is made of Y-cut 128° lithium niobate; the material of the flow channel (3) and the detachable film (4) is polydimethylsiloxane; the thickness of the interdigitated electrode (2) is 200±20nm; the material of the interdigitated electrode (2) is a combination of one or more materials selected from aluminum, gold and silver.

10. The method of the integrated system of automated acoustic tweezers manipulation and detection based on image recognition according to any one of claims 1 to 9, characterized in that: An initial droplet containing fluorescent particles is generated by a droplet generation device, and the initial droplet is split into multiple small droplets in equal proportion through a flow channel, thereby realizing high-throughput generation of small droplets; a control system is used to control the interdigitated electrodes to generate surface acoustic waves to drive the small droplets to form an array and / or move in the flow channel, thereby realizing manipulation of the small droplets; at the same time, the substrate is placed under a microscope, and the control system is used to perform image recognition of the fluorescent particles to control the opening or closing of the interdigitated electrodes in the flow channel where the small droplets are located, thereby realizing detection and sorting of the small droplets.

Images