A path-driven cleaning optimization method for microfluidic lab-on-a-chip systems

Abstract

The application discloses a path-driven cleaning optimization method for a micro-fluidic lab-on-chip system, takes a sequence diagram of biochemical assay experiments, a chip architecture and an experiment scheduling strategy as input, calculates an execution program with a cleaning operation as output, minimizes the number of cleaning operations, the total length of a cleaning path and the total execution time of biochemical assay experiments, reduces the influence of the insertion of the cleaning operation on the biochemical assay experiments, and ensures the accuracy of experimental results.

Description

Technical Field

[0001] This invention belongs to the field of computer-aided design technology, specifically relating to a path-driven cleaning optimization method for microfluidic on-chip laboratory systems. Background Technology

[0002] The emergence of microfluidic laboratory-on-a-chip (Lab-on-a-Chip) systems has revolutionized traditional biological and biochemical experiments. By precisely controlling nanoliter volumes of liquid samples / reagents, complex experiments can be integrated onto a coin-sized chip and executed automatically according to the experimental protocol. Furthermore, Lab-on-a-Chip systems offer advantages such as high efficiency, high reliability, and low cost, and have been widely applied in experiments such as protein crystallization and nucleic acid isolation.

[0003] Figure 1 (a) illustrates the internal structure of the microfluidic chip, comprising a flow layer and a control layer, each with its own channel network. In the flow layer, liquid samples / reagents are injected into the chip via flow ports and transported to target components through flow channels to perform various biochemical operations such as mixing and heating. In the control layer, an elastic membrane (also known as a microvalve) is deployed in the overlapping area with the flow layer. The control ports guide the airflow; when the airflow passes through the elastic membrane, as... Figure 1 As shown in (b), the elastic membrane squeezes the flow channel downwards, the micro-valve closes, and the transport of fluid in the flow channel is blocked; when the airflow is released, the elastic membrane returns to its original state, the micro-valve opens, and the transport of fluid resumes.

[0004] Despite the advantages of lab-on-a-chip (Lab-on-Chip) experiments, cross-contamination between different fluids can occur anywhere on the chip, potentially leading to erroneous results if not handled properly. For example, in in vitro diagnostic experiments, chemiluminescence immunoassay is a commonly used method for identifying various medical indicators, such as tumor markers. When two fluids carrying different luminescent agents pass through a channel sequentially, the second fluid may be contaminated by residues from the first, causing abnormal luminescence during the reaction and resulting in incorrect detection results. Therefore, cleaning procedures are necessary in Lab-on-Chip experiments to prevent cross-contamination between fluids.

[0005] Performing a cleaning operation requires injecting cleaning agent into the chip through the flow port and rinsing the contaminated points along the cleaning path. A waste liquid port needs to be installed at the end of the cleaning path to release waste fluid and air from the channel. Therefore, a complete cleaning path should include the flow port, the contaminated point, the waste liquid port, and the channel connecting them: flow port → contaminated point → waste liquid port. Consequently, when developing a cleaning strategy, the resource usage on the chip by biochemical operations, liquid transport operations, and excess liquid removal operations must be considered simultaneously, making the design of the cleaning strategy extremely complex. An inappropriate cleaning strategy will severely impact the total execution time of biochemical assays and may even lead to erroneous experimental results.

[0006] Several methods for cleaning optimization already exist. In "K. Hu, T.-Y. Ho, and K. Chakrabarty, “Wash optimization and analysis for cross-contamination removal under physical constraints in flow-based microfluidic biochips,” IEEE Trans.Comput.-Aided Design Integr. Circuits Syst., vol. 35, no. 4, pp. 559–572, 2015,” a graph-based method was proposed, transforming the cleaning optimization problem into a collision set problem. However, this method requires a given set of contamination points, making it impossible to integrate the generation of cleaning strategies into the experimental process in real time. In “WH Minhass, J. McDaniel, M. Raagaard, P. Brisk, P. Pop, and J. Madsen, “Scheduling and fluid routing for flow-based microfluidicclaboratories-on-a-chip,” IEEE Trans. Comput.-Aided Design Integr. CircuitsSyst., vol. 37, no. 3, pp. 615–628, 2018,” a demand-driven heuristic algorithm was proposed to implement this during mapping. However, because this method postpones the cleaning operation as much as possible until the contaminated resource needs to be reused, conflicts between the cleaning operation and fluid transport frequently occur, severely delaying the completion time of biochemical experiments.Recently, a delay-aware cleaning optimization algorithm was proposed in "WH Minhass, J. McDaniel, M. Raagaard, P. Brisk, P. Pop, and J. Madsen, “Scheduling and fluid routing for flow-based microfluidic laboratories-on-a-chip,” IEEE Trans. Comput.-Aided Design Integr. Circuits Syst., vol. 37, no.3, pp. 615–628, 2018.”, achieving efficient cleaning in a distributed channel storage architecture. However, this method does not consider resource sharing when generating cleaning paths, resulting in higher chip costs and longer overall experimental execution time. With the increasing integration of on-chip laboratory experiments, the experiments conducted on the chip are becoming increasingly complex, and existing cleaning optimization algorithms are gradually failing to meet the experimental requirements. Summary of the Invention

[0007] To overcome the shortcomings of existing technologies, this invention provides a path-driven cleaning optimization method for microfluidic laboratory-on-a-chip systems. Taking the sequence diagram of biochemical assay experiments, chip architecture, and experimental scheduling strategy as input, an execution program with cleaning operations is calculated as output, which minimizes the number of cleaning operations, the total length of the cleaning path, and the total execution time of the biochemical assay experiments. This reduces the impact of inserted cleaning operations on the biochemical assay experiments and ensures the accuracy of experimental results.

[0008] The technical solution adopted by this invention to solve its technical problem includes the following steps: Step 1: Transform the cleaning optimization problem into an integer linear programming problem, establish the corresponding ILP model, and solve it using a solver. The model establishment process is as follows: Step 1-1: Based on the chip architecture used for biochemical assays, use a chip of size [size missing]. virtual grid This indicates the chip layout, where components and channels are placed. On the cell; because the original schedule of the experiment will change after the cleaning operation is introduced, the start time of each task in the biochemical assay experiment will be recalculated. Steps 1-2: For each operation To ensure the accuracy of the results, sufficient execution time should be allowed, which is constrained to be:

[0009] in, and They represent The start and end times, It is a sequence diagram The provisions of China Execution time; Steps 1-3: For sequence diagrams Each edge on ,operate Need to operate Execute after completion, and be constrained as follows:

[0010] in It is a 0-1 variable, representing the sequence diagram. middle, Is it? child nodes; It is a constant used to... The two indicated cases are converted into linear constraints; Steps 1-4: Use and They respectively represent being bound to Components and The output fluid, when two operations and When bound to the same component, they cannot execute simultaneously; this is a constraint:

[0011] in It is a 0-1 variable that represents the execution order of two operations; Steps 1-5: For sequence maps Each edge on Output liquid From components To Component The transportation task was undertaken by It means that it is necessary to Completed before the start time, and is constrained as follows:

[0012] in and These are transportation tasks The start and end times.

[0013] Steps 1-6: After completing the transportation task Next, it is necessary to remove the component. Excess liquid, use This indicates the removal of excess liquid, and this also needs to be done during the operation. Completed before starting, and constrained as follows:

[0014] in, and These are the tasks for removing excess liquid. The start and end times; Steps 1-7: To ensure the task and Execute correctly. and The duration should be longer than the duration specified in the given measurement strategy. and This indicates that since some excess liquid removal tasks can be integrated into the cleaning operation, 0-1 variables are used. To indicate whether or not to It is integrated into the cleaning operation, and the value of this variable can be determined after the cleaning path is established, subject to the following constraints:

[0015] Steps 1-8: To avoid conflicts, transport / removal tasks using the same cell cannot be executed simultaneously, subject to the following constraints:

[0016] in It is a 0-1 variable used to represent the execution order of two transport / removal tasks. It is the task specified in the given scheduling policy. The flow path; Steps 1-9: Use and Represent the collection of contaminated cells and the cells respectively. The time point of contamination; For contaminated cells Use 0-1 variables This indicates whether the cell meets the first type of cleaning avoidance condition, that is, if the cell... It will not be used for any transportation tasks in subsequent steps. This can avoid cleaning cells. Therefore, the following constraints apply:

[0017] Similarly, using 0-1 variables and To represent the contaminated cells respectively Does it meet the second and third cleaning avoidance conditions, i.e., does it flow through the cell again? The fluid will not be contaminated by residual liquid on the cell, or the cell will only be used for waste disposal in subsequent steps, subject to the following constraints:

[0018] in It is a 0-1 variable used to indicate a transportation task. Is the transported fluid the same type of fluid as the pollutant? It means Whether it is used for waste liquid removal is a 0-1 variable; Steps 1-10: Using 0-1 variables Indicates whether the contaminated cell needs to be cleaned. It is constrained as follows:

[0019] Steps 1-11: After completing the analysis of the necessity of cleaning for the transportation / removal task, determine the set of cleaning targets. A collection of cleaning operations Next, an effective cleaning path is established on the chip. For each cleaning operation Its cleaning path It should begin at a flow port and end at a waste port, and is constrained as follows:

[0020] in, and These are a set of stream ports and a set of waste ports on a given chip architecture. It is a 0-1 variable used to indicate whether to open the stream port. Assigned to cleaning path ; Steps 1-12: At the flow port and port Assigned to Afterwards, in the adjacent cells of the two ports, only one cell was cleaned along the path. Therefore, the following constraints apply to its use:

[0021] in and These are virtual meshes superior and The set of adjacent cells, It is a 0-1 variable, representing the cell. Is the path being cleaned? use; Steps 1-13: For the cleaning path Except for the cells adjacent to the flow port / waste port, the remaining cells... There should be exactly two adjacent cells that are also cleaned. Usage is subject to the following constraints:

[0022] in It is a virtual grid upper cell The set of adjacent cells; Steps 1-14: Assumptions It is a set of cleaning operations. The cells to be cleaned should be those corresponding to the cleaning path. Coverage is constrained as follows:

[0023] Steps 1-15: Establish the cleaning path Next, calculations are performed to execute the cleaning operation. Time window; use to indicate The point in time when all cells in the table were contaminated. It is proposed to clean The earliest time a cell becomes contaminated is subject to the following constraints:

[0024] in and These are the cleaning operations. The start and end times; Steps 1-16: Calculate cleaning operations based on the rinsing time of the cleaning agent and the dissolution time of the contaminants. The duration is calculated as follows:

[0025] in , and These are the cleaning paths. The length, fluid flow rate on the chip, and dissolution time of contaminants; Steps 1-17: To ensure the cleaning operation is fully performed, The execution should last for a sufficiently long time; therefore, the following constraints apply:

[0026] Steps 1-18: Cleaning Operation The occupied cell cannot be used by other transport / remove tasks or another cleaning operation at the same time, subject to the following constraints:

[0027] in and It is a 0-1 variable used to represent the execution order of two operations; Step 2: For the cleaning operation and excess liquid removal task Assuming a liquid transportation task A path is required. The next task is to determine the variables in equation (7). Formulated as:

[0028] Step 3: In After all operations in the process are completed, by The completion time of the experiment is determined by the following constraints:

[0029] Step 4: Use The number of washing operations represented is calculated as follows:

[0030] in It is a 0-1 variable used to represent the task. Does the path need to be cleaned, and can it be formalized as:

[0031] Step 5: From The total length of the washing path is calculated as follows:

[0032] Step 6: Determine a cleaning optimization strategy that minimizes the experiment's completion time, the total length of the washing path, and the number of cleaning operations by solving the following optimization problem:

[0033] The beneficial effects of this invention are as follows: 1. This invention comprehensively considers biochemical operations, liquid transportation tasks, and cleaning requirements, and analyzes the necessity of cleaning operations from a global perspective, reducing the number of cleaning operations and additional resource consumption.

[0034] 2. This invention integrates the cleaning operation with the removal of excess liquid, thus effectively reducing the channel occupancy caused by the cleaning operation and improving the efficiency of biochemical assays.

[0035] 3. This invention calculates the cleaning path and insertion time window of the cleaning operation from the perspective of path, so as to minimize the total execution time of the biochemical assay experiment. Attached Figure Description

[0036] Figure 1 The diagram shows the structure of a microfluidic biochip: (a) a schematic diagram of the flow layer and control layer, and (b) a schematic diagram of the microvalve in the closed state.

[0037] Figure 2 This is a sequence diagram of the biochemical assay experiment in an embodiment of the present invention.

[0038] Figure 3 Chip layout for biochemical assay experiments in this embodiment of the invention.

[0039] Figure 4 The scheduling strategy for biochemical assay experiments in this invention embodiment.

[0040] Figure 5 The present invention provides a highly efficient cleaning optimization scheme. Detailed Implementation

[0041] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0042] The purpose of this invention is to develop an efficient cleaning strategy for biochemical assays on microfluidic on-chip laboratory systems, design a path-driven cleaning optimization method, and transform the cleaning optimization problem into an integer linear programming (ILP) problem for solution.

[0043] This invention requires a sequence diagram of biochemical assay experiments. Taking chip architecture and experimental scheduling strategies as input, this invention calculates an execution program with cleaning operations as output, minimizing the number of cleaning operations, the total length of the cleaning path, and the total execution time of the biochemical assay. This invention transforms the cleaning optimization problem into an integer linear programming problem, establishes a corresponding ILP model, and uses a solver to solve it. The model establishment process is described below.

[0044] Based on the chip architecture used in biochemical assays, this invention uses a chip of size [size missing]. Virtual grid The chip layout is represented by cells R, with components and channels placed on them. Because the original experimental schedule changes after the introduction of the cleaning operation, the start times of each task in the biochemical assay need to be recalculated to ensure correct execution of the experiment.

[0045] For each operation To ensure the accuracy of the results, sufficient execution time should be allowed, which can be constrained as follows:

[0046] in, and They represent The start and end times, It is a sequence diagram The provisions of China Execution time.

[0047] For sequence diagrams Each edge on ,operate Need to operate Execute after completion, which can be constrained as follows:

[0048] in It is a 0-1 variable, representing the sequence diagram. middle, Is it? The child nodes. It is a very large constant used to... The two indicated cases are converted into linear constraints.

[0049] For ease of expression, use and To represent being bound to Components and The output fluid. When two operations and When bound to the same component, they cannot execute simultaneously; this can be constrained as follows:

[0050] in It is a 0-1 variable that represents the execution order of two operations.

[0051] For sequence diagrams Each edge on Output liquid From components To Component The transportation task (by) (This indicates that) it is necessary to It must be completed before the start time, which can be constrained as follows:

[0052] in and These are transportation tasks The start and end times.

[0053] After completing the transportation task Next, it is necessary to remove the component. Excess liquid, use This indicates the removal of excess liquid, and this also needs to be done during the operation. Complete before starting, which can be constrained as follows:

[0054] in, and These are the tasks for removing excess liquid. The start and end times.

[0055] To ensure the mission and For them to be executed correctly, they should be longer than the duration specified in the given measurement strategy (by...). and (This is indicated by the text). Furthermore, since some excess liquid removal tasks can be integrated into the cleaning operation, 0-1 variables are used. To indicate whether or not to It is integrated into the cleaning operation, and the value of this variable can be determined after the cleaning path is established, subject to the following constraints:

[0056] Furthermore, to avoid conflicts, transport / removal tasks using the same cell cannot be executed simultaneously, subject to the following constraints:

[0057] in It is a 0-1 variable used to represent the execution order of two transport / removal tasks. It is the task specified in the given scheduling policy. The flow path.

[0058] After completing the liquid transport / removal task, the contaminated cell can be identified. Using... and Represent the collection of contaminated cells and the cells respectively. The time point when it was contaminated.

[0059] For contaminated cells Use 0-1 variables This indicates whether a cell meets the first type of cleaning avoidance condition, i.e., if the cell... It will not be used for any transportation tasks in subsequent steps. This can avoid cleaning the cells. Therefore, the following constraints apply:

[0060] Similarly, using 0-1 variables and To represent the contaminated cells respectively Does it meet the second and third cleaning avoidance conditions, i.e., does it flow through the cell again? The fluid will not be contaminated by residual liquid on the cell, or the cell will only be used for waste disposal in subsequent steps. Therefore, the following constraints apply:

[0061] in It is a 0-1 variable used to indicate a transportation task. Is the transported fluid the same type of fluid as the pollutant? It means Whether it is used for waste liquid removal is a 0-1 variable.

[0062] Then, use 0-1 variables. To indicate whether the contaminated cells need to be cleaned. This can be constrained as follows:

[0063] After completing the analysis of the necessity of cleaning for the transportation / removal task, the set of cleaning targets can be determined. A collection of cleaning operations Next, an effective cleaning path needs to be established on the chip. For each cleaning operation Its cleaning path It should begin at a flow port and end at a waste port, which can be constrained as follows:

[0064] in, and These are a set of stream ports and a set of waste ports on a given chip architecture. It is a 0-1 variable used to indicate whether to open the stream port. Assigned to cleaning path .

[0065] At the stream port and port Assigned to After that, in the adjacent cells of the two ports, only one cell can be cleaned along the path. Use. Therefore, the following constraints apply:

[0066] in and These are virtual meshes superior and The set of adjacent cells, It is a 0-1 variable used to represent a cell. Is the path being cleaned? use.

[0067] Next, regarding the cleaning path Except for the cells adjacent to the flow port / waste port, the remaining cells... There should be exactly two adjacent cells that are also cleaned. Use. Subject to the following constraints:

[0068] in It is a virtual grid upper cell The set of adjacent cells.

[0069] Assumption It is a set of cleaning operations. The cells to be cleaned should be those corresponding to the cleaning path. Coverage, which can be constrained as follows:

[0070] Establishing a cleaning path Next, it is necessary to calculate the parameters used to perform the cleaning operation. The time window. Use to indicate The point in time when all cells in the table were contaminated. It is proposed to clean The earliest time a cell becomes contaminated is subject to the following constraints:

[0071] in and These are the cleaning operations. The start and end times.

[0072] Next, based on the rinsing time of the cleaning agent and the dissolution time of the contaminants... [8] To calculate cleaning operations The duration, which can be calculated as:

[0073] in , and These are the cleaning paths. The length, fluid flow rate on the chip, and dissolution time of contaminants.

[0074] To ensure that the cleaning operation is fully performed, The execution should last for a sufficiently long time. Therefore, the following constraints apply:

[0075] To avoid resource conflicts, the cleaning operation An occupied cell cannot be occupied by another transport / removal task or another cleaning operation simultaneously. Therefore, the following constraints apply:

[0076] in and It is a 0-1 variable used to represent the execution order of two operations.

[0077] Using the above results, a cleaning path can be generated. Combining the cleaning operation with the excess liquid removal operation, using the cleaning operation to simultaneously treat contaminants and remove excess liquid, can further optimize the existing scheduling strategy. For the cleaning operation... and excess liquid removal task Assuming a liquid transportation task A path is required. The next task, then the variables in (7) It can be formalized as:

[0078] exist After all operations in the process are completed, by The indicated completion time of the experiment can be determined by the following constraints:

[0079] Next, using The number of washing operations can be represented as:

[0080] in It is a 0-1 variable used to represent the task. Does the path need to be cleaned, and can it be formulated as:

[0081] In addition, by The total length of the washing path can be calculated as follows:

[0082] Finally, a cleaning optimization strategy that minimizes the experiment's completion time, the total length of the washing path, and the number of cleaning operations can be determined by solving the following optimization problem:

[0083] This invention uses eight benchmark datasets for testing to validate the proposed algorithm. Table 1 shows the basic information of the benchmark datasets, where the parameters |O|, |E|, and |device| are the sequence diagrams, respectively. The number of biological operations and edges in the middle, and the number of corresponding components in the component library.

[0084] Table 1. Details of the benchmarks used

[0085] To verify the superiority of this invention in the cleaning optimization problem, it is compared with a recently proposed Delay-Aware Cleaning Optimization (DAWO) algorithm. This method uses a list scheduling algorithm to introduce cleaning operations, calculates cleaning paths using a depth-first search algorithm, and finally allocates time windows for cleaning operations using a scanline method, achieving efficient cleaning. To ensure a fair comparison, this invention and the DAWO algorithm were implemented in Python and run on the same computer using the aforementioned benchmark dataset to verify the advancement of the path-driven approach used in this invention.

[0086] Table 2 shows the comparison results between the present invention and the DAWO algorithm in terms of the number of cleaning operations and the total length of the cleaning path. Compared with the DAWO algorithm, the present invention reduces the number of cleaning operations by an average of 17.73%. In addition, the present invention achieves an optimization of the total length of the cleaning path by 13.33%-30.43%, with an improvement rate of 24.56%. The reasons for this result are as follows: 1) Since the present invention analyzes the necessity of cleaning, unnecessary cleaning operations are completely avoided; and 2) Since the present invention formulates the cleaning optimization problem as an ILP problem, it establishes a more efficient cleaning path in a global manner compared with the heuristic algorithm used in DAWO.

[0087] Table 2 Comparison of the number of cleaning cycles and total cleaning path length between the present invention and DAWO

[0088] As shown in Table 3, compared with the DAWO algorithm, the present invention improves the cleaning operation execution time by 22.33% and the average waiting time by 17.07%. As shown in Table 4, the experimental delay time and total experimental execution time are reduced by an average of 33.10% and 9.28%, respectively, thus proving the effectiveness of the proposed method.

[0089] Table 3 Comparison of cleaning operation execution time and average operation waiting time between the present invention and DAWO

[0090] Table 4 Comparison of experimental delay time and total experimental execution time between the present invention and DAWO

[0091] Example: Biochemical assay protocols typically use sequence diagrams. It means that, among them It is a set of biochemical operations with a specific execution time. It is used to represent the dependencies between these operations. Figure 2 This diagram illustrates the sequence of a biochemical assay that used two input reagents, r1 and r2, and involved seven biochemical operations. These are processed to obtain the corresponding measurement results. To enable automated execution of this biochemical assay in a microfluidic chip laboratory, Figure 3 and Figure 4 The chip layout generated using the DAWO algorithm and the scheduling strategy for biochemical assays are shown respectively. All flow paths are listed in Table 5. It can be seen that all biochemical operations are completed within 30 seconds.

[0092] Table 5 Path Table

[0093] A. Analyze the necessity of the cleaning operation use Figure 3 and Figure 4 The chip layout and scheduling scheme shown can effectively execute... Figure 2 The biochemical assay shown does not account for cross-contamination between fluids, which may lead to incorrect results. However, if all contaminated resources are cleaned immediately after contamination, a large number of channels will be used for cleaning operations for an extended period, significantly delaying the completion time of the experiment. In fact, some channels / components do not require cleaning even if contaminated; these are analyzed in three categories below.

[0094] Type 1: Contaminated components / channels that will no longer be used in subsequent experimental steps. For example, in Figure 2 Operations in After completion, the component Contaminated. Then, in the... Use of flow path in seconds The result The output liquid from Transport to This further contaminated the flow path. However, due to components It will not be used for any biochemical operations, and the contaminated path will not be used by any subsequent transport missions, thus avoiding the need for corresponding cleaning operations.

[0095] Type 2: Contaminated channels / components used for transporting / operating the same type of liquid. For example, in operations... After completion, as follows Figure 4 As shown, using flow paths The resulting liquid from Transport to In the middle, the flow path was contaminated. The path will then be in the second... seconds are used to The output liquid from Transport to In this process, the liquids that pass through the flow path are the same type of liquid, thus avoiding the need for corresponding cleaning operations. In particular, if the residual liquid remaining in the component is the same type of liquid as the subsequently input liquid, the component cleaning operation can also be avoided.

[0096] Type 3: Contaminated channels used for transporting waste liquids. For example, after completing the operation... Afterwards, as Figure 4 As shown, using flow paths The resulting liquid from Transport to In the middle, the flow path was contaminated. However, since this flow path is only used to discharge the wastewater... The waste liquid generated can avoid the need for corresponding cleaning operations.

[0097] B. Integrate cleaning operations with excess liquid removal operations A complete cleaning path should include the flow port, contamination point, waste liquid port, and the connecting channels. Introducing cleaning operations may result in additional resource consumption, and if not handled properly, conflicts between cleaning operations and liquid transport tasks may frequently occur, significantly delaying the completion time of bioassay experiments. On the other hand, when transporting liquid to the target component, excess liquid will remain at both ends of the component due to the need to purge air from within the component. This excess liquid needs to be removed using a separate flow path. However, combining cleaning operations with excess liquid removal operations, using cleaning operations to simultaneously handle contaminants and excess liquid, can optimize resource consumption and improve execution efficiency.

[0098] exist Figure 3 and Figure 4 In the chip layout and scheduling strategy, reagents via flow path Injected Excess liquid remains in the channel section. In, it is necessary to use The second path listed is removed separately. Afterwards, the path... In the Seconds are used for transportation The mixture requires further rinsing after removing excess liquid, resulting in two separate liquid manipulation tasks. However, combining the rinsing operation with the excess liquid removal operation introduces a flow path... The cleaning process can remove excess liquid and contaminants simultaneously.

[0099] C. Optimize the allocation of cleaning paths and cleaning time windows. To minimize experimental delays caused by the introduction of cleaning operations, it is necessary to calculate optimized cleaning paths and time windows to execute the cleaning operations efficiently without introducing additional conflicts. We will still use... Figure 3 and Figure 4 The chip layout and scheduling strategies described in [the text] are used as examples. In operation... After completion, as follows Figure 4 As shown, using flow paths The resulting liquid from Transport to This led to contamination of the pathway. In the future The output liquid moves to This path needs to be cleaned beforehand. Therefore, the optimal starting point for cleaning the path is... And there are three candidates that can serve as its endpoint, including , and If you choose As an endpoint, although it can produce a shorter cleaning path, it will lead to changes in the cleaning operation and the use of the flow path. and The conflict with the transportation mission significantly delayed the operation. The execution time. Additionally, if you choose... As the endpoint, this will create a section that needs to be traversed. The cleaning path is very long, and the cleaning operation will be used in conjunction with the flow path. The transportation tasks conflict. Conversely, if the choice is... As the endpoint, it can not only generate a shorter washing path, that is... Furthermore, the cleaning operation can be combined with the use of flow paths. The transportation tasks are carried out simultaneously.

[0100] Based on the above analysis, Figure 5 This invention demonstrates a scheduling strategy with cleaning operations, which introduces only three cleaning operations that are executed concurrently with several other liquid manipulation tasks. Furthermore, it utilizes a path... , and The task of removing excess liquid was integrated with the cleaning operation, resulting in a delay of only one second in the completion time of the experiment.

Claims

1. A path-driven cleaning optimization method for microfluidic on-chip laboratory systems, characterized in that, Includes the following steps: Step 1: Transform the cleaning optimization problem into an integer linear programming problem, establish the corresponding ILP model, and solve it using a solver. The model establishment process is as follows: Step 1-1: Based on the chip architecture used for biochemical assays, use a chip of size [size missing]. Virtual grid This indicates the chip layout, where components and channels are placed. On the cell; because the original schedule of the experiment will change after the cleaning operation is introduced, the start time of each task in the biochemical assay experiment will be recalculated. Steps 1-2: For each operation It is constrained as follows: in, and They represent The start and end times, It is a sequence diagram The provisions of China Execution time; Steps 1-3: For sequence diagrams Each edge on ,operate Need to operate Execute after completion, and be constrained as follows: in It is a 0-1 variable, representing the sequence diagram. middle, Is it? child nodes; It is a constant used to... The two indicated cases are converted into linear constraints; Steps 1-4: Use and They respectively represent being bound to Components and The output fluid, when two operations and When bound to the same component, they cannot execute simultaneously; this is a constraint: in It is a 0-1 variable that represents the execution order of two operations; Steps 1-5: For sequence maps Each edge on Output liquid From components To Component The transportation task was undertaken by It means that it is necessary to Completed before the start time, and is constrained as follows: in and These are transportation tasks The start and end times; Steps 1-6: After completing the transportation task Next, it is necessary to remove the component. Excess liquid, use This indicates the operation of removing excess liquid. Completed before starting, and constrained as follows: in, and These are the tasks for removing excess liquid. The start and end times; Steps 1-7: To ensure the task and Execute correctly. and The duration must be greater than the duration specified in the given measurement strategy. and Use 0-1 variables To indicate whether or not to Integrated into the cleaning operation, the value of this variable is determined after the cleaning path is established, subject to the following constraints: Steps 1-8: To avoid conflicts, transport or removal tasks using the same cell cannot be executed simultaneously, subject to the following constraints: in It is a 0-1 variable used to represent the execution order of two transport or removal tasks. It is the task specified in the given scheduling policy. The flow path; It is the task specified in the given scheduling policy. The flow path; Steps 1-9: Use and Represent the collection of contaminated cells and the cells respectively. The time point of contamination; For contaminated cells Use 0-1 variables This indicates whether the cell meets the first type of cleaning avoidance condition, that is, if the cell... It will not be used for any transportation tasks in subsequent steps. This can avoid cleaning cells. Therefore, the following constraints apply: Use 0-1 variables and To represent the contaminated cells respectively Does it meet the second and third cleaning avoidance conditions, i.e., does it flow through the cell again? The fluid will not be contaminated by residual liquid on the cell, or the cell will only be used for waste disposal in subsequent steps, subject to the following constraints: in It is a 0-1 variable used to indicate a transportation task. Is the transported fluid the same type of fluid as the pollutant? It means Whether it is used for waste liquid removal is a 0-1 variable; Steps 1-10: Using 0-1 variables Indicates whether the contaminated cell needs to be cleaned. It is constrained as follows: Steps 1-11: Determine the set of cleaning targets A collection of cleaning operations Next, a set of effective cleaning paths L is established on the chip, for each cleaning operation. Its cleaning path It starts at a flow port and ends at a waste port, and is constrained as follows: in, and These are a set of stream ports and a set of waste ports on a given chip architecture. It is a 0-1 variable used to indicate whether to open the stream port. Assigned to cleaning path ; It is a 0-1 variable used to indicate whether the waste liquid port is open. Assigned to cleaning path Steps 1-12: At the flow port and waste liquid port Assigned to Afterwards, in the adjacent cells of the two ports, only one cell was cleaned along the path. Therefore, the following constraints apply to its use: in and These are virtual meshes superior and The set of adjacent cells, It is a 0-1 variable, representing the cell. Is the path being cleaned? use; Steps 1-13: For the cleaning path Except for the cells adjacent to the flow port or waste liquid port, the remaining cells... Only two adjacent cells are also cleaned in the path. Usage is subject to the following constraints: in It is a virtual grid upper cell The set of adjacent cells; Steps 1-14: Assumptions It is a set of cleaning operations. The cells to be cleaned are those corresponding to the cleaning path. Coverage is constrained as follows: Steps 1-15: Establish the cleaning path Next, calculations are performed to execute the cleaning operation. Time window; use to indicate The point in time when all cells in the table were contaminated. It is proposed to clean The earliest time a cell becomes contaminated is subject to the following constraints: in and These are the cleaning operations. The start and end times; Steps 1-16: Calculate cleaning operations based on the rinsing time of the cleaning agent and the dissolution time of the contaminants. The duration is calculated as follows: in , and These are the cleaning paths. The length, fluid flow rate on the chip, and dissolution time of contaminants; Steps 1-17: To ensure the cleaning operation is fully performed, the following constraints apply: Steps 1-18: Cleaning Operation The occupied cell cannot be occupied by other transport or remove tasks or another cleanup operation at the same time, subject to the following constraints: in and It is a 0-1 variable used to represent the execution order of two operations; Step 2: For the cleaning operation and excess liquid removal task Assuming a liquid transportation task A path is required. The next task is to determine the variables in equation (7). Formulated as: Step 3: In After all operations in the process are completed, by The completion time of the experiment is determined by the following constraints: Step 4: Use The number of washing operations represented is calculated as follows: in It is a 0-1 variable used to represent the task. Does the path need to be cleaned, and can it be formalized as: Step 5: From The total length of the washing path is calculated as follows: Step 6: Determine a cleaning optimization strategy that minimizes the experiment's completion time, the total length of the washing path, and the number of cleaning operations by solving the following optimization problem:

Images