A microfluidic biological detection method and system

Abstract

The present invention discloses a microfluidic biological detection method and system, which specifically relates to the field of biological detection technology. Analyzing droplet images based on a deep learning model is beneficial to ensuring the quality and consistency of droplet images, and is conducive to directly understanding whether droplet generation is consistent; through comprehensive analysis and evaluation of potential risk information of droplet generation, the droplet generation stability evaluation coefficient is calculated by comprehensively considering factors such as fluid flow rate abnormality ratio, droplet generation variation index and microvalve response action deviation index, which can further reveal the potential adverse effects on the consistency and stability of future droplet generation when generating normal droplet difference signals; through analysis of medium-risk droplet generation signals within the real-time risk monitoring interval, the potential instability risk of droplet generation is evaluated in a more detailed and real-time manner, which helps to promptly respond to the potential risks of droplet generation instability and ensure the accuracy and reliability of the experiment.

Description

Technical Field

[0001] The present invention relates to the field of biological detection technology, and more specifically, to a microfluidic biological detection method and system. Background Art

[0002] In the field of biological detection, researchers seek to gain a deep understanding of the gene expression patterns of each single cell in a tissue or organ in order to reveal cellular heterogeneity, changes in cell state during development, and the mechanisms of disease development. Single-cell separation, as a common operation in single-cell analysis, is crucial for studying the mechanisms of disease development, especially single-cell analysis plays an important role in in-depth studies of cellular heterogeneity, mutation status, and the development of drug resistance within tumors. Through this in-depth study, researchers can more accurately identify potential therapeutic targets and provide solid scientific support for personalized medicine and precision drug design.

[0003] Microfluidics is a field of science and technology that studies and controls the flow of liquids at the micrometer scale. Microfluidics involves the processing of tiny volumes of liquids, usually in the micrometer to millimeter scale. Microfluidics includes droplet microfluidics technology, which is a technology that uses droplet microfluidics chips to quickly generate immiscible droplets from two immiscible liquids for single cell separation. The use of droplet microfluidics for single cell separation has the advantages of being fast, simple, easy to operate and small in size.

[0004] In the operation of single-cell separation using droplet microfluidics, it is necessary to sort the droplets formed by droplet microfluidics to distinguish the droplets with different needs in biological detection, including using fluid focusing control structures and microvalve structures to achieve droplet generation, and controlling the size and distribution of droplets by adjusting the fluid flow rate and relative position as well as the state of the microvalve; finally, the formed droplets can be automatically sorted, but before droplet sorting, if the droplet generation is unstable or abnormal, the accuracy of droplet sorting will be reduced, making the subsequent sorting process difficult to accurately execute, resulting in a possible reduction in the credibility of the experimental results of biological detection, especially for single-cell analysis that requires high precision.

[0005] In order to solve the above problems, a technical solution is now provided. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide a microfluidic biological detection method and system to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A microfluidic biological detection method comprises the following steps:

[0009] Step S1: Analyze the droplet image based on the deep learning model to determine whether the size of the droplets generated by the microvalve is consistent;

[0010] Step S2: evaluating the stability of the fluid flow rate by analyzing the fluid flow rate in the droplet microfluidics process; evaluating the stability of droplet generation by analyzing the droplet generation situation; evaluating the stability of the response ability of the microvalve by analyzing the response of the microvalve;

[0011] Step S3: when the sizes of the droplets generated by the microvalve are consistent, a comprehensive analysis is performed on the stability of the flow rate of the fluid, the stability of the droplet generation, and the stability of the response capability of the microvalve to determine the degree of adverse effects on the droplet generation process;

[0012] Step S4: analyzing the proportion of the adverse effects in the droplet generation process, and evaluating the risk of instability in the droplet generation; when the risk of instability in the droplet generation is small, performing automated droplet sorting.

[0013] In a preferred embodiment, step S1 comprises the following steps:

[0014] Step S101: photographing a liquid droplet image generated by the current microvalve and preprocessing the liquid droplet image;

[0015] Step S102: analyzing the droplet image based on the deep learning model;

[0016] Step S103: Based on the analysis result of the droplet image by the deep learning model, determine whether the size of the droplets generated by the microvalve is consistent.

[0017] In a preferred embodiment, in step S2, the fluid flow rate in the droplet microfluidics process is analyzed and reflected by the fluid flow rate stability information, and the fluid flow rate stability information includes the fluid flow rate abnormality ratio. The acquisition logic of the fluid flow rate abnormality ratio is:

[0018] Setting a fluid monitoring interval; obtaining a preset range interval of a flow rate of a fluid in a fluid focusing control structure;

[0019] Monitoring the flow rate value of the fluid in the fluid focusing control structure within the fluid monitoring interval;

[0020] Obtain the length of time during which the flow velocity value of the fluid in the fluid focusing control structure in the fluid monitoring interval is not within the preset range of the flow velocity, and mark the ratio of the length of time during which the flow velocity value of the fluid in the fluid focusing control structure in the fluid monitoring interval is not within the preset range of the flow velocity to the length of time corresponding to the fluid monitoring interval as the fluid flow velocity abnormality ratio.

[0021] In a preferred embodiment, the droplet generation situation in the droplet microfluidics process is analyzed by droplet generation stability information, and the droplet generation stability information includes a droplet generation variation index. The acquisition logic of the droplet generation variation index is:

[0022] Obtain the generation time point of m times of droplets closest to the real time, obtain the total generation time of m times of droplets, and mark the ratio of m to the total generation time of m times of droplets as the droplet generation frequency; obtain the preset droplet generation frequency, and mark the ratio of the deviation value of the droplet generation frequency and the preset droplet generation frequency to the preset droplet generation frequency as the droplet generation frequency deviation ratio; m is a positive integer;

[0023] According to the generation time points of m droplets, the time interval between the generation time points of every two adjacent droplets is calculated;

[0024] Obtaining a preset time interval; setting a stable evaluation time interval range based on the preset time interval;

[0025] A total of m-1 time intervals between generation time points of two adjacent droplets are obtained, the number of time intervals between generation time points of two adjacent droplets that are not within the stable evaluation time interval range is obtained, and the ratio of the number of time intervals between generation time points of two adjacent droplets that are not within the stable evaluation time interval range to m-1 is marked as the droplet generation interval offset ratio;

[0026] The droplet generation frequency deviation ratio and the droplet generation interval offset ratio are de-unitized, and the de-unitized droplet generation frequency deviation ratio and the droplet generation interval offset ratio are weighted summed to calculate the droplet generation variation index.

[0027] In a preferred embodiment, the response of the microvalve in the droplet microfluidics process is analyzed and reflected by the microvalve response information. The microvalve response information includes a response action deviation index, and its specific acquisition logic is:

[0028] A: Set the door response monitoring interval and obtain the response action of the microvalve within the door response monitoring interval;

[0029] B: obtaining the response time corresponding to the response action of the microvalve, and obtaining the preset response time corresponding to the response action of the microvalve;

[0030] C: Calculate the deviation between the response time corresponding to the response action of the microvalve within the gate response monitoring interval and the preset response time, and calculate the response action deviation index.

[0031] In a preferred embodiment, in step S3, when the size of the droplets generated by the microvalve is consistent, the fluid flow rate anomaly ratio, the droplet generation variation index, and the response action deviation index are normalized, and the droplet generation stability evaluation coefficient is calculated by the normalized fluid flow rate anomaly ratio, the droplet generation variation index, and the response action deviation index;

[0032] A first droplet generation assessment threshold and a second droplet generation assessment threshold are set; the first droplet generation assessment threshold is less than the second droplet generation assessment threshold; a droplet generation stability assessment coefficient is compared with the first droplet generation assessment threshold and the second droplet generation assessment threshold to generate a risk level signal, wherein the risk level signal includes a high risk droplet generation signal, a medium risk droplet generation signal, and a low risk droplet generation signal;

[0033] When the droplet generation stability assessment coefficient is greater than the droplet generation assessment second threshold, a high-risk droplet generation signal is generated;

[0034] When the droplet generation stability assessment coefficient is greater than or equal to the droplet generation assessment first threshold, and the droplet generation stability assessment coefficient is less than or equal to the droplet generation assessment second threshold, a medium-risk droplet generation signal is generated;

[0035] When the droplet generation stability assessment coefficient is less than the droplet generation assessment first threshold, a low risk droplet generation signal is generated.

[0036] In a preferred embodiment, in step S4, a real-time risk monitoring interval is set, and the presence of medium-risk droplet generation signals within the real-time risk monitoring interval is analyzed to assess the risk level of instability in droplet generation:

[0037] Obtaining the time length of generating a medium-risk droplet generation signal within the real-time risk monitoring interval, and marking the ratio of the time length of generating the medium-risk droplet generation signal within the real-time risk monitoring interval to the time length corresponding to the real-time risk monitoring interval as a droplet generation risk ratio;

[0038] When the droplet generation risk ratio is greater than the droplet generation risk ratio threshold, a high droplet generation risk signal is generated; when the droplet generation risk ratio is less than or equal to the droplet generation risk ratio threshold, a low droplet generation risk signal is generated.

[0039] In a preferred embodiment, a microfluidic biological detection system includes an image acquisition module, a size determination module, a potential risk assessment module, a droplet generation assessment module, and a risk level determination module;

[0040] The image acquisition module obtains the droplet image generated by the current microvalve and pre-processes the droplet image;

[0041] The size judgment module analyzes the droplet image based on the deep learning model to determine whether the size of the droplets generated by the microvalve is consistent;

[0042] The potential risk assessment module evaluates the stability of the fluid flow rate by analyzing the fluid flow rate in the droplet microfluidics process; evaluates the stability of droplet generation by analyzing the droplet generation situation; and evaluates the stability of the response ability of the microvalve by analyzing the response of the microvalve;

[0043] When the size of the droplets generated by the microvalve is consistent, the droplet generation evaluation module comprehensively analyzes the stability of the flow rate of the fluid, the stability of the droplet generation, and the stability of the response ability of the microvalve to determine the degree of adverse effects on the droplet generation process;

[0044] The risk level judgment module evaluates the risk level of instability in droplet generation by analyzing the proportion of adverse effects on the droplet generation process over a period of time; when the risk level of instability in droplet generation is small, automatic droplet sorting is performed.

[0045] The technical effects and advantages of a microfluidic biological detection method and system of the present invention are as follows:

[0046] 1. Analyzing droplet images based on deep learning models is beneficial to ensuring the quality and consistency of droplet images. Based on deep learning models, through structures such as convolutional neural networks, droplet images are analyzed to further determine the consistency of the size of droplets generated by microvalves, which is beneficial to directly understand whether the droplet generation is consistent, thereby determining whether it will have an adverse effect on droplet sorting and single cell separation.

[0047] 2. Through comprehensive analysis and evaluation of potential risk information of droplet generation, factors such as fluid velocity abnormality ratio, droplet generation variation index, and microvalve response action deviation index are comprehensively considered. By normalizing these factors and calculating the droplet generation stability evaluation coefficient, the stability of droplet generation can be evaluated more carefully when the droplet difference normal signal is generated, and the potential adverse effects on the consistency and stability of droplet generation in the future can be further revealed. By setting the first threshold of droplet generation evaluation and the second threshold of droplet generation evaluation, the evaluation results are further divided into high-risk, medium-risk, and low-risk droplet generation signals, providing operators with more specific risk level information, which helps to take appropriate measures in time to ensure the consistency and stability of droplet generation.

[0048] 3. By analyzing the medium-risk droplet generation signals within the real-time risk monitoring interval, a more detailed and real-time assessment of the potential instability risk of droplet generation is carried out. By calculating the droplet generation risk ratio, the risk degree of droplet generation instability can be quantified more objectively. The droplet generation risk ratio threshold is set, and the droplet generation risk ratio is compared with the threshold to generate a signal of high or low droplet generation risk. This helps to take timely measures in the case of unstable droplet generation, helps to respond to the potential risks of droplet generation instability in a timely manner, and ensures the accuracy and reliability of the experiment. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic diagram of a microfluidic biological detection method of the present invention;

[0050] Figure 2 The schematic diagram is a structural diagram of a microfluidic biological detection system of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] Example 1

[0053] Figure 1 The present invention provides a microfluidic biological detection method, which comprises the following steps:

[0054] Step S1: Analyze the droplet image based on the deep learning model to determine whether the size of the droplets generated by the microvalve is consistent.

[0055] Step S2: Evaluate the stability of the fluid flow rate by analyzing the fluid flow rate in the droplet microfluidics process; evaluate the stability of droplet generation by analyzing the droplet generation situation; evaluate the stability of the response ability of the microvalve by analyzing the response of the microvalve.

[0056] Step S3: When the size of the droplets generated by the microvalve is consistent, a comprehensive analysis is performed on the stability of the flow rate of the fluid, the stability of the droplet generation, and the stability of the response capability of the microvalve to determine the degree of adverse effects on the droplet generation process.

[0057] Step S4: analyzing the proportion of the adverse effects in the droplet generation process, and evaluating the risk of instability in the droplet generation; when the risk of instability in the droplet generation is small, performing automated droplet sorting.

[0058] Wherein, step S1 comprises the following steps:

[0059] Step S101: photograph the droplet image generated by the current microvalve and pre-process the droplet image:

[0060] Use the camera to capture the image of the droplets generated by the current microvalve. Ensure that the droplet image contains enough droplets. Preprocess the captured image, including but not limited to:

[0061] Image denoising to reduce the impact of environmental noise.

[0062] Adjust image brightness and contrast to ensure consistent image quality.

[0063] Images are cropped or scaled to ensure consistent image size for input into the model.

[0064] Step S102: Analyze the droplet image based on a deep learning model, such as a convolutional neural network (CNN), specifically:

[0065] Model input: x represents the input image, which is a tensor containing the pixel values ​​of the image; assuming the image size is M*N, then x can be expressed as x∈R M*N*C , where C is the number of channels of the image, R represents the real number domain, M is the number of rows (height) of the image, and N is the number of columns of the image.

[0066] Model output: y represents the output of the model, that is, the prediction of droplet information. Specifically, if the droplet information includes multiple attributes such as position, size, shape, etc., it can be expressed as y = [y1, y2, ..., y k ], where k is the dimension of the droplet information.

[0067] Model mapping: y = f(x; θ), where y is the output of the model, describing the model's mapping from the input image to the droplet information. This mapping is achieved through a deep learning network, which contains multiple layers (convolutional layers, pooling layers, fully connected layers, etc.); θ is the model's parameters, including the weights and biases of each layer in the network.

[0068] Step S103: Based on the analysis result of the droplet image by the deep learning model, determine whether the size of the droplets generated by the micro valve is consistent:

[0069] The output y of the model is compared with the actual label using the mean squared error as the loss function. The mean squared error measures the squared difference between the predicted value and the actual value. A smaller mean squared error indicates that the model fits the actual droplet size better, that is, better consistency. Where n is the number of samples, y iis the model's predicted value, label i is the actual label; the actual label refers to the true output or target value corresponding to each sample in the training data. i is the number of the predicted value of the model, i = 1, 2, 3, 4, ..., n, i and n are both positive integers. MSE (y, label) is the mean square error loss value. The larger the mean square error loss value, the worse the consistency of the droplet size and shape in the droplet image, the greater the subsequent impact on single cell separation, and the normal progress of biological detection.

[0070] The mean square error loss value is compared with the loss threshold: when the mean square error loss value is greater than the loss threshold, a large droplet difference signal is generated; this indicates that in the droplet microfluidics process, the inconsistency of the droplet shape will have a greater impact on the accuracy of single-cell separation; based on the generated large droplet difference signal, an alarm is issued, the operation of the droplet microfluidics is immediately stopped, and professional technicians are arranged to carry out maintenance, including whether the equipment is abnormal and whether the experimental environment meets the standards.

[0071] When the mean square error loss value is less than or equal to the loss threshold, a normal droplet difference signal is generated; at this time, in direct observation, the shape consistency of the droplet generated by the microvalve is normal; but the potential risk factors of droplet generation are not considered, and it may not necessarily fully reveal the potential problems in the droplet generation process.

[0072] The loss threshold is set by professionals in this field based on the size of the mean square error loss value, as well as other actual conditions such as the requirements for the consistency of droplet size and shape in practice, and will not be elaborated here.

[0073] In step S2, potential risk information of the droplet microfluidics process is analyzed, and the potential risk information includes fluid flow rate stability information, droplet generation stability information, and microvalve response information.

[0074] The analysis of the fluid flow rate in the droplet microfluidics process is reflected by the fluid flow rate stability information. The fluid flow rate stability information includes the fluid flow rate abnormality ratio. The acquisition logic of the fluid flow rate abnormality ratio is:

[0075] A fluid monitoring interval is set, and the stability of the flow rate of the fluid in the fluid focusing control structure within the fluid monitoring interval is analyzed to evaluate the potential factors of instability in droplet generation.

[0076] The time length corresponding to the fluid monitoring interval is a fixed value. The fluid monitoring interval is a real-time monitoring interval, that is, the end point of the fluid monitoring interval is always a real-time time point, that is, the monitoring range corresponding to the fluid monitoring interval changes with time.

[0077] The preset range interval of the flow rate of the fluid in the fluid focus control structure is obtained. The preset range interval of the flow rate of the fluid in the fluid focus control structure refers to the range of the flow rate value of the fluid in the fluid focus control structure allowed by the system in a stable state. When the flow rate of the fluid in the fluid focus control structure exceeds the range of the flow rate value of the fluid in the fluid focus control structure allowed by the system, it may cause instability in droplet generation.

[0078] Monitor the flow rate value of the fluid in the fluid focusing control structure within the fluid monitoring interval.

[0079] Obtain the length of time during which the flow velocity value of the fluid in the fluid focusing control structure in the fluid monitoring interval is not within the preset range of the flow velocity, and mark the ratio of the length of time during which the flow velocity value of the fluid in the fluid focusing control structure in the fluid monitoring interval is not within the preset range of the flow velocity to the length of time corresponding to the fluid monitoring interval as the fluid flow velocity abnormality ratio.

[0080] The larger the fluid flow rate anomaly ratio is, the greater the adverse effect on droplet generation, which may be caused by inaccurate or unstable flow rate regulation within the fluid focusing control structure.

[0081] The analysis of droplet generation in the droplet microfluidics process is reflected by droplet generation stability information. The droplet generation stability information includes the droplet generation variation index. The acquisition logic of the droplet generation variation index is:

[0082] The generation time point of m droplets closest to the real time is obtained, the total generation time of these m droplets is obtained, and the ratio of m to the total generation time of these m droplets is marked as the droplet generation frequency; the preset droplet generation frequency is obtained, and the ratio of the deviation value of the droplet generation frequency to the preset droplet generation frequency to the preset droplet generation frequency is marked as the droplet generation frequency deviation ratio. m is a positive integer.

[0083] According to the generation time points of m droplets, the time interval between the generation time points of every two adjacent droplets is calculated.

[0084] A preset time interval is obtained, where the preset time interval is the reciprocal of a preset droplet generation frequency.

[0085] The stable evaluation time interval range is set based on a preset time interval. For example, if the preset time interval is 3 seconds, the stable evaluation time interval range can be set to 2.9 seconds to 3.1 seconds.

[0086] When the time interval between the generation time points of every two adjacent droplets is not within the stable evaluation time interval, it indicates that the generation of the droplets is unstable.

[0087] A total of m-1 time intervals between the generation time points of two adjacent droplets are obtained, the number of time intervals between the generation time points of two adjacent droplets that are not within the stable evaluation time interval range is obtained, and the ratio of the number of time intervals between the generation time points of two adjacent droplets that are not within the stable evaluation time interval range to m-1 is marked as the droplet generation interval offset ratio.

[0088] The droplet generation frequency deviation ratio and the droplet generation interval offset ratio are de-unitized, and the weighted sum of the de-unitized droplet generation frequency deviation ratio and the droplet generation interval offset ratio is performed to calculate the droplet generation variation index, which is expressed as follows: sad = a*dax+b*fsa; wherein sad, dax, and fsa are the droplet generation variation index, the droplet generation frequency deviation ratio, and the droplet generation interval offset ratio, respectively; a and b are the weight coefficients of the droplet generation frequency deviation ratio and the droplet generation interval offset ratio, respectively, and a and b are both greater than 0.

[0089] The larger the droplet generation variation index is, the worse the stability of droplet generation is, and the greater the adverse effect on droplet consistency is.

[0090] The preset droplet generation frequency is obtained according to the droplet generation frequency set by the droplet microfluidic system.

[0091] m is a positive integer and is set according to actual conditions. For example, m can be set to 20 according to the monitoring requirements for droplets.

[0092] It is worth noting that the droplets are generated by microvalves.

[0093] The analysis of the response of the microvalve in the droplet microfluidics process is reflected by the microvalve response information. The microvalve response information includes the response action deviation index, and its specific acquisition logic is:

[0094] A: Set the door response monitoring interval and obtain the response action of the microvalve within the door response monitoring interval.

[0095] B: Obtain the response time corresponding to the response action of the micro valve, and obtain the preset response time corresponding to the response action of the micro valve.

[0096] C: Calculate the deviation between the response time corresponding to the response action of the microvalve within the gate response monitoring interval and the preset response time, calculate the response action deviation index, and evaluate whether the response capability of the microvalve is stable.

[0097] Among them, the expression of the response action deviation index is specifically: Where xpz is the response action deviation index, W and q are the number of response actions of the microvalve in the door response monitoring interval and the number of response actions of the microvalve in the door response monitoring interval, respectively, W and q are positive integers, and q = 1, 2, 3, 4, ..., W, xst q is the response time corresponding to the response action of the qth microvalve in the gate response monitoring interval; q It is the preset response time corresponding to the response action of the qth microvalve within the door response monitoring interval.

[0098] The larger the response action deviation index is, the more unstable the response ability of the microvalve is.

[0099] Door response monitoring interval: The time length corresponding to the door response monitoring interval is a fixed value. The door response monitoring interval is a real-time monitoring interval, that is, the end point of the door response monitoring interval is always the real-time time point, that is, the monitoring range corresponding to the door response monitoring interval changes with time.

[0100] The response action of a microvalve refers to the process by which the microvalve reacts to a control signal, thereby achieving control over the flow of a tiny liquid or gas. The response action of a microvalve includes opening, closing, and regulating the microvalve.

[0101] The preset response time is the standard value of the completion time corresponding to the response action of the microvalve.

[0102] If the microvalve responds too quickly, it may cause oscillation or shock in the system, affecting the stability of the system. Rapid closing of the microvalve may cause fluid shock, resulting in pressure waves and fluid instability, affecting the stability of the microfluidic system. Too slow a response speed may reduce the accuracy of the system for real-time regulation, especially in applications that require high-frequency switching. Slow response speed may cause a delay effect in the system, resulting in a significant time difference between the control signal and the actual execution. This will lead to unstable droplet generation.

[0103] In step S3, when a normal droplet difference signal is generated, although the size consistency of the droplets generated by the microvalve is normal under direct observation at this time; however, the potential risk information of the droplet microfluidics process will have an adverse effect on the consistency and stability of the future droplet generation.

[0104] Therefore, when a normal droplet difference signal is generated, the potential risk information including the fluid flow rate stability information, droplet generation stability information and microvalve response information is comprehensively analyzed to evaluate the potential adverse impact on droplet generation stability, specifically:

[0105] The fluid flow rate anomaly ratio, droplet generation variation index and response action deviation index are normalized, and the droplet generation stability assessment coefficient is calculated based on the normalized fluid flow rate anomaly ratio, droplet generation variation index and response action deviation index. The larger the droplet generation stability assessment coefficient, the greater the potential adverse effect on the stability of droplet generation.

[0106] For example, the present invention can use the following formula to calculate the droplet generation stability evaluation coefficient, which is expressed as: Yswx=ln(α1*lyb+α2*sad+α3*xpz+1); wherein, Yswx and lyb are the droplet generation stability evaluation coefficient and the fluid flow rate anomaly ratio, respectively; α1, α2, and α3 are the preset proportional coefficients of the fluid flow rate anomaly ratio, the droplet generation variation index, and the response action deviation index, respectively, and α1, α2, and α3 are all greater than 0.

[0107] A first droplet generation assessment threshold and a second droplet generation assessment threshold are set; the first droplet generation assessment threshold is less than the second droplet generation assessment threshold; the droplet generation stability assessment coefficient is compared with the first droplet generation assessment threshold and the second droplet generation assessment threshold to generate a risk level signal, the risk level signal includes a high risk droplet generation signal, a medium risk droplet generation signal and a low risk droplet generation signal.

[0108] When the droplet generation stability assessment coefficient is greater than the droplet generation assessment second threshold, a high-risk droplet generation signal is generated.

[0109] When the droplet generation stability assessment coefficient is greater than or equal to the first droplet generation assessment threshold, and the droplet generation stability assessment coefficient is less than or equal to the second droplet generation assessment threshold, a medium-risk droplet generation signal is generated.

[0110] When the droplet generation stability assessment coefficient is less than the droplet generation assessment first threshold, a low risk droplet generation signal is generated.

[0111] The first threshold value for droplet generation evaluation and the second threshold value for droplet generation evaluation are set by professional and technical personnel in this field according to the size of the droplet generation stability evaluation coefficient and other actual conditions such as the actual requirements for droplet generation, and will not be repeated here.

[0112] When a high-risk droplet generation signal is generated, it indicates that the potential adverse impact on the stability of droplet generation is relatively large. At this time, the generation of droplets is relatively unstable. According to the generated high-risk droplet generation signal, it is necessary to immediately stop the generation of droplets by microfluidics and arrange professional technicians to carry out maintenance.

[0113] When a medium-risk droplet generation signal is generated, it indicates that there is a potential adverse effect on the droplet generation stability at this time, but the degree of the effect is small.

[0114] When a low risk droplet generation signal is generated, it indicates that there is no potential adverse effect on droplet generation stability at this time.

[0115] In step S4, the proportion of medium-risk droplet generation signals within a period of time is analyzed to determine the risk level of potential adverse effects on droplet generation stability:

[0116] A real-time risk monitoring interval is set, and the risk level of unstable droplet generation is evaluated by analyzing the presence of medium-risk droplet generation signals within the real-time risk monitoring interval.

[0117] The time length corresponding to the real-time risk monitoring interval is a fixed value. The real-time risk monitoring interval is a real-time monitoring interval, that is, the end point of the real-time risk monitoring interval is always the real-time time point, that is, the monitoring range corresponding to the real-time risk monitoring interval changes with time.

[0118] The time length of generating a medium-risk droplet generation signal within the real-time risk monitoring interval is obtained, and the ratio of the time length of generating a medium-risk droplet generation signal within the real-time risk monitoring interval to the time length corresponding to the real-time risk monitoring interval is marked as the droplet generation risk ratio. The larger the droplet generation risk ratio, the greater the risk of unstable and inconsistent droplet generation.

[0119] When the droplet generation risk ratio is greater than the droplet generation risk ratio threshold, a droplet generation risk high signal is generated. At this time, according to the generated droplet generation risk high signal, it is necessary to stop the microfluidics from generating droplets and arrange professional technicians for maintenance.

[0120] When the droplet generation risk ratio is less than or equal to the droplet generation risk ratio threshold, a low droplet generation risk signal is generated. At this time, the risk of instability in droplet generation is small and no measures need to be taken. At this time, the next step of automated droplet sorting can be carried out.

[0121] The droplet generation risk ratio threshold is set by professionals in the field according to the actual conditions such as the size of the droplet generation risk ratio and the required standards for droplet generation.

[0122] Example 2

[0123] The difference between Example 2 of the present invention and Example 1 is that this example introduces a microfluidic biological detection system.

[0124] Figure 2 A structural schematic diagram of a microfluidic biological detection system of the present invention is given, which includes an image acquisition module, a size judgment module, a potential risk assessment module, a droplet generation assessment module and a risk degree judgment module.

[0125] The image acquisition module obtains the droplet image generated by the current microvalve and pre-processes the droplet image.

[0126] The size judgment module analyzes the droplet image based on the deep learning model to determine whether the size of the droplets generated by the microvalve is consistent.

[0127] The potential risk assessment module evaluates the stability of the fluid flow rate by analyzing the fluid flow rate in the droplet microfluidics process; evaluates the stability of droplet generation by analyzing the droplet generation situation; and evaluates the stability of the response ability of the microvalve by analyzing the response of the microvalve.

[0128] When the size of droplets generated by the microvalve is consistent, the droplet generation evaluation module conducts a comprehensive analysis of the stability of the fluid flow rate, the stability of droplet generation, and the stability of the microvalve's response capability to determine the degree of adverse effects on the droplet generation process.

[0129] The risk level judgment module evaluates the risk level of instability in droplet generation by analyzing the proportion of adverse effects in the droplet generation process over a period of time. When the risk level of instability in droplet generation is low, automatic droplet sorting is performed.

[0130] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0131] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0132] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0133] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0134] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0135] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0136] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0137] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0138] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0139] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A microfluidic biological detection method, characterized in that: The steps include: Step S1: Analyze the droplet image based on the deep learning model to determine whether the size of the droplets generated by the microvalve is consistent; Step S2: evaluating the stability of the fluid flow rate by analyzing the fluid flow rate in the droplet microfluidics process; evaluating the stability of droplet generation by analyzing the droplet generation situation; evaluating the stability of the response ability of the microvalve by analyzing the response of the microvalve; Step S3: when the sizes of the droplets generated by the microvalve are consistent, a comprehensive analysis is performed on the stability of the flow rate of the fluid, the stability of the droplet generation, and the stability of the response capability of the microvalve to determine the degree of adverse effects on the droplet generation process; Step S4: analyzing the proportion of adverse effects in the droplet generation process and evaluating the risk of instability in the droplet generation; Automated droplet sorting when there is a small risk of unstable droplet generation; In step S2, the fluid flow rate in the droplet microfluidics process is analyzed and reflected by the fluid flow rate stability information. The fluid flow rate stability information includes the fluid flow rate abnormality ratio. The acquisition logic of the fluid flow rate abnormality ratio is: Setting a fluid monitoring interval; obtaining a preset range interval of a flow rate of a fluid in a fluid focusing control structure; Monitoring the flow rate value of the fluid in the fluid focusing control structure within the fluid monitoring interval; Obtaining the length of time during which the flow velocity value of the fluid in the fluid focusing control structure in the fluid monitoring interval is not within the preset range of the flow velocity, and marking the ratio of the length of time during which the flow velocity value of the fluid in the fluid focusing control structure in the fluid monitoring interval is not within the preset range of the flow velocity to the length of time corresponding to the fluid monitoring interval as the fluid flow velocity abnormality ratio; The droplet generation situation in the droplet microfluidics process is analyzed through the droplet generation stability information, which includes the droplet generation variation index. The acquisition logic of the droplet generation variation index is: Obtain the generation time point of m droplets closest to the real time, obtain the total generation time of the m droplets, and mark the ratio of m to the total generation time of the m droplets as the droplet generation frequency; Obtaining a preset droplet generation frequency, marking the ratio of the deviation value between the droplet generation frequency and the preset droplet generation frequency to the preset droplet generation frequency as a droplet generation frequency deviation ratio; m is a positive integer; According to the generation time points of m droplets, the time interval between the generation time points of every two adjacent droplets is calculated; Obtaining a preset time interval; setting a stable evaluation time interval range based on the preset time interval; A total of m-1 time intervals between generation time points of two adjacent droplets are obtained, the number of time intervals between generation time points of two adjacent droplets that are not within the stable evaluation time interval range is obtained, and the ratio of the number of time intervals between generation time points of two adjacent droplets that are not within the stable evaluation time interval range to m-1 is marked as the droplet generation interval offset ratio; The droplet generation frequency deviation ratio and the droplet generation interval offset ratio are subjected to unit removal processing, and the droplet generation frequency deviation ratio and the droplet generation interval offset ratio after the unit removal processing are weighted summed to calculate the droplet generation variation index; The response of the microvalve in the droplet microfluidics process is analyzed through the microvalve response information. The microvalve response information includes the response action deviation index, and its specific acquisition logic is as follows: A: Set the door response monitoring interval and obtain the response action of the microvalve within the door response monitoring interval; B: obtaining the response time corresponding to the response action of the microvalve, and obtaining the preset response time corresponding to the response action of the microvalve; C: Calculate the deviation between the response time corresponding to the response action of the microvalve within the gate response monitoring interval and the preset response time, and calculate the response action deviation index; In step S3, when the sizes of the droplets generated by the microvalve are consistent, the fluid flow rate anomaly ratio, the droplet generation variation index, and the response action deviation index are normalized, and the droplet generation stability evaluation coefficient is calculated by the normalized fluid flow rate anomaly ratio, the droplet generation variation index, and the response action deviation index; Setting a first droplet generation evaluation threshold and a second droplet generation evaluation threshold; the first droplet generation evaluation threshold is less than the second droplet generation evaluation threshold; Comparing the droplet generation stability assessment coefficient with the first droplet generation assessment threshold and the second droplet generation assessment threshold to generate a risk level signal, wherein the risk level signal includes a high risk droplet generation signal, a medium risk droplet generation signal, and a low risk droplet generation signal; When the droplet generation stability assessment coefficient is greater than the droplet generation assessment second threshold, a high-risk droplet generation signal is generated; When the droplet generation stability assessment coefficient is greater than or equal to the droplet generation assessment first threshold, and the droplet generation stability assessment coefficient is less than or equal to the droplet generation assessment second threshold, a medium-risk droplet generation signal is generated; When the droplet generation stability assessment coefficient is less than the droplet generation assessment first threshold, a low risk droplet generation signal is generated; In step S4, a real-time risk monitoring interval is set, and the presence of medium-risk droplet generation signals within the real-time risk monitoring interval is analyzed to assess the risk level of instability in droplet generation: Obtaining the time length of generating a medium-risk droplet generation signal within the real-time risk monitoring interval, and marking the ratio of the time length of generating the medium-risk droplet generation signal within the real-time risk monitoring interval to the time length corresponding to the real-time risk monitoring interval as a droplet generation risk ratio; When the droplet generation risk ratio is greater than the droplet generation risk ratio threshold, a droplet generation risk high degree signal is generated; When the droplet generation risk ratio is less than or equal to the droplet generation risk ratio threshold, a droplet generation risk low degree signal is generated.

2. A microfluidic biological detection method according to claim 1, characterized in that: Step S1 includes the following steps: Step S101: photographing a liquid droplet image generated by the current microvalve and preprocessing the liquid droplet image; Step S102: analyzing the droplet image based on the deep learning model; Step S103: Based on the analysis result of the droplet image by the deep learning model, determine whether the size of the droplets generated by the microvalve is consistent.

3. A microfluidic biological detection system, used to implement a microfluidic biological detection method according to any one of claims 1 to 2, characterized in that: It includes an image acquisition module, a size judgment module, a potential risk assessment module, a droplet generation assessment module and a risk level judgment module; The image acquisition module obtains the droplet image generated by the current microvalve and pre-processes the droplet image; The size judgment module analyzes the droplet image based on the deep learning model to determine whether the size of the droplets generated by the microvalve is consistent; The potential risk assessment module evaluates the stability of the fluid flow rate by analyzing the fluid flow rate in the droplet microfluidics process; and evaluates the stability of droplet generation by analyzing the droplet generation situation; The stability of the microvalve's response capability is evaluated by analyzing the microvalve's response; When the size of the droplets generated by the microvalve is consistent, the droplet generation evaluation module comprehensively analyzes the stability of the flow rate of the fluid, the stability of the droplet generation, and the stability of the response ability of the microvalve to determine the degree of adverse effects on the droplet generation process; The risk level judgment module evaluates the risk level of instability in droplet generation by analyzing the proportion of adverse effects on the droplet generation process over a period of time; When there is a small risk of unstable droplet generation, automated droplet sorting is performed.

Images