A functional detection method and system for fluorescent immunoassay analyzer

Abstract

The present invention discloses a function detection method and system for a fluorescent immunoassay analyzer, and relates to the field of function detection. The method comprises the following steps: using the interaction logic between the functional modules in the fluorescent immunoassay analyzer, constructing a detection task topology diagram showing the relationship between the detection tasks; obtaining key detection points from the detection task topology diagram based on a graph theory algorithm, and establishing a list of key detection points; using a priority search algorithm, generating several detection paths from the starting point to the end point, and each detection path covers different key detection points; obtaining information of each key detection point extracted from the historical detection record, obtaining a historical data set, and using a machine learning algorithm to train a detection path efficiency prediction model; inputting all detection paths into the detection path efficiency prediction model, comparing and selecting the detection path with the highest comprehensive efficiency as the recommended detection path. The present invention improves the overall speed and efficiency of detection.

Description

Technical Field

[0001] The present invention relates to the field of functional detection, and in particular to a functional detection method and system for a fluorescent immunoassay analyzer. Background Art

[0002] Fluorescence immunoassay analyzer is a precision medical device that uses fluorescence labeling technology combined with immunological principles to quantitatively detect specific antigens, antibodies or other biological molecules in biological samples such as blood, urine, and tissue fluid. This instrument is widely used in clinical diagnosis, disease monitoring, drug concentration monitoring and other fields. The basic principle of fluorescence immunoassay is to combine fluorescent dyes with antibodies to form fluorescent labeled antibodies. When this labeled antibody binds to the target antigen in the sample to be tested, it will emit a fluorescent signal of a specific wavelength. By detecting the intensity of the signal, the concentration of the antigen can be inferred.

[0003] Functional testing of fluorescent immunoassay analyzers refers to regular or on-demand inspections of the instrument's various functional modules to ensure that their performance meets established standards, thereby guaranteeing the accuracy, reliability, and repeatability of test results. Existing methods for functional testing of fluorescent immunoassay analyzers only perform quality control tests on equipment or reagent performance, and only test individual components or specific functions, ignoring the interactions and dependencies between the various functional modules, and failing to consider the efficiency of the entire testing process, affecting overall testing speed and resource utilization.

[0004] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0005] In view of the problems in the related art, the present invention proposes a function detection method and system for a fluorescent immunoassay analyzer to overcome the above technical problems existing in the existing related art.

[0006] To this end, the specific technical solution adopted by the present invention is as follows:

[0007] According to one aspect of the present invention, a function detection method for a fluorescent immunoassay analyzer is provided, and the function detection method for a fluorescent immunoassay analyzer comprises the following steps:

[0008] S1. Using the interaction logic between the functional modules in the fluorescent immunoassay analyzer, a detection task topology diagram is constructed to show the relationship between the detection tasks.

[0009] S2. Obtain key detection points from the detection task topology map based on graph theory algorithms and establish a list of key detection points.

[0010] S3. Using the priority search algorithm, several detection paths are generated from the starting point to the end point, and each detection path covers different key detection points.

[0011] S4. Obtain information of each key inspection point extracted from historical inspection records, obtain a historical data set, and use a machine learning algorithm to train an inspection path efficiency prediction model.

[0012] S5. Input all detection paths into the detection path efficiency prediction model, compare and select the detection path with the highest comprehensive efficiency as the recommended detection path.

[0013] The method of obtaining key detection points from the detection task topology map based on the graph theory algorithm and establishing a list of key detection points includes the following steps:

[0014] S21, using a single-source shortest path algorithm to calculate the shortest paths of all node pairs in the detection task topology graph, and obtain a weight matrix, where the weight matrix includes the shortest path length from the source node to each node;

[0015] S22, identifying key detection points according to the weight matrix;

[0016] S23, selecting a starting node from the detection task topology graph, greedily selecting key nodes according to the importance of the nodes, and setting a termination condition for the key node search;

[0017] S24, analyzing the neighborhood of each key node and updating the key detection point list;

[0018] S25. After multiple iterations, the final list of key inspection points is determined.

[0019] Furthermore, using the interaction logic between the functional modules in the fluorescent immunoassay analyzer, constructing a detection task topology diagram showing the relationship between the detection tasks includes the following steps:

[0020] S11, determining all functional modules in the fluorescence immunoassay analyzer;

[0021] S12. Define each detection task according to the interaction logic of each functional module, and determine the dependency relationship between each detection task;

[0022] S13. Draw a detection task topology graph by taking each detection task as a node and the dependency relationship as an edge.

[0023] Furthermore, the single-source shortest path algorithm is used to calculate the shortest paths of all node pairs in the detection task topology graph, and the weight matrix is ​​obtained, which includes the following steps:

[0024] S211, set the source node, and set the shortest path estimation values ​​of all nodes to infinity;

[0025] S212, creating a determined set and an undetermined set, the determined set is used to store nodes for which the shortest paths have been found, and the undetermined set is used to store nodes for which the shortest paths have not been found;

[0026] S213, selecting a node from the undetermined set, moving it to the determined set, and updating the shortest path estimation values ​​of all nodes in the undetermined set that are directly connected to the current node;

[0027] S214. For each neighbor of the current node, if the path through the current node to the neighbor is shorter than the known path, update the shortest path estimate of the neighbor. The update formula is:

[0028] ;

[0029] In the formula, new _ distance represents the new attempted distance from the source node to the neighbor;

[0030] current _ distance Represents the shortest path length from the source node to the current node;

[0031] weight _ of _ edge Represents the weight of the edge from the current node to its neighbors;

[0032] S215. Repeat S213-S214 until all nodes are moved to the determined set, and the shortest path to any other node is constructed starting from the source node by tracking the predecessor node of each node.

[0033] Furthermore, selecting a starting node from the detection task topology graph and greedily selecting key nodes according to the importance of the nodes includes the following steps:

[0034] S231, select any node as the starting node, and calculate the importance index for each node. The importance index calculation formula is:

[0035] ;

[0036] In the formula, Representation Node v Betweenness centrality of ;

[0037] Represents a slave node s To Node t The number of shortest paths;

[0038] represents the number of shortest paths passing through node v;

[0039] S232, creating an empty key node set for storing the selected key nodes;

[0040] S233, from the nodes that have not been selected, according to the importance index of the nodes, select the most critical nodes and add them to the critical node set;

[0041] S234: Repeat S233 until a specific stop condition is met.

[0042] Furthermore, the neighborhood of each key node is analyzed and the list of key detection points is updated, including the following steps:

[0043] S241, dividing the neighborhood of the key node into several subsets, and using a local search algorithm to find a key detection point in each subset;

[0044] S242. Find the local optimal set of key nodes in each subset;

[0045] S243. Compare the local optimal key node set with the current key node set, select a better set, and if the newly found key node set is better, update the key detection point list.

[0046] Furthermore, using a priority search algorithm, several detection paths are generated from the starting point to the end point, and each detection path covers different key detection points, including the following steps:

[0047] S31, selecting a starting point node in the detection task topology map as the starting point of the search, and initializing a stack structure for storing nodes accessed during the traversal process;

[0048] S32, push the starting point node into the stack and mark it as visited, take out the current node from the top of the stack, and traverse all unvisited adjacent nodes;

[0049] S33. For each adjacent node, if it is a key detection point and has not been visited, it is pushed into the stack and marked as visited, otherwise it is not pushed into the stack;

[0050] S34, backtracking the current node and recording the node combinations that have been visited;

[0051] S35, repeat S32-S34 until all key detection points have been visited, and then end the search;

[0052] S36. Output all generated detection paths as search results.

[0053] Furthermore, backtracking the current node and recording the visited node combinations includes the following steps:

[0054] S341. If all adjacent nodes of the current node have been visited, the current node is popped from the stack and backtracked. Each time the backtracking is performed, the path from the starting point to the current node is recorded.

[0055] S342. During the search process, a collection is used to record the node combinations that have been visited.

[0056] Further, obtaining information of each key detection point extracted from the historical detection records, obtaining a historical data set, and using a machine learning algorithm to train a detection path efficiency prediction model includes the following steps:

[0057] S41, extracting data from the historical test records of the fluorescent immunoassay analyzer, cleaning the data and extracting features to obtain a historical data set;

[0058] S42, normalizing the features in the historical data set, and dividing the historical data set into a training set and a test set;

[0059] S43. Select a machine learning algorithm, and use historical data sets to perform model training and verification on the selected machine learning algorithm to obtain a detection path efficiency prediction model.

[0060] Further, selecting a machine learning algorithm and using the historical data set to perform model training and verification on the selected machine learning algorithm includes the following steps:

[0061] S431, selecting a decision tree algorithm, and using a training set to train the decision tree algorithm to obtain a detection path efficiency prediction model;

[0062] S432. Use the test set to verify the detection path efficiency prediction model and evaluate the prediction accuracy and effect of the detection path efficiency prediction model.

[0063] According to another aspect of the present invention, a functional detection system for a fluorescent immunoassay analyzer is provided, which includes a topology map construction module, a key detection point acquisition module, a detection path generation module, a model training module and a recommended detection path module; the topology map construction module, the key detection point acquisition module, the detection path generation module, the model training module and the recommended detection path module are connected in sequence.

[0064] The topology map construction module is used to utilize the interaction logic between the functional modules in the fluorescent immunoassay analyzer to construct a detection task topology map that shows the relationship between the detection tasks.

[0065] The key detection point acquisition module is used to obtain key detection points from the detection task topology map based on graph theory algorithms and establish a list of key detection points.

[0066] The detection path generation module is used to generate several detection paths from the starting point to the end point using a priority search algorithm, and each detection path covers different key detection points.

[0067] The model training module is used to obtain information of key inspection points extracted from historical inspection records, obtain historical data sets, and use machine learning algorithms to train the inspection path efficiency prediction model.

[0068] The recommended detection path module is used to input all detection paths into the detection path efficiency prediction model, compare and select the detection path with the highest comprehensive efficiency as the recommended detection path.

[0069] The beneficial effects of the present invention are:

[0070] (1) The present invention provides a functional detection method and system for a fluorescent immunoassay analyzer, which optimizes the detection path, reduces unnecessary duplication and redundancy, and improves the overall speed and efficiency of the detection. It identifies key detection points and paths, reasonably allocates detection resources, and reduces resource waste. It improves the fault detection rate during the detection process through optimization based on historical data, ensures the reliability of the detection results, and uses machine learning algorithms to predict and select the best path to ensure the real-time and dynamic adaptability of the detection process.

[0071] (2) The present invention uses graph theory algorithms to identify the most critical detection points in the detection process of the fluorescent immunoassay analyzer. By identifying the key detection points, resource allocation can be optimized to ensure the operating efficiency and reliability of the key detection points. This is of great significance for the optimization of the functional detection of the fluorescent immunoassay analyzer. The discovery and priority treatment of key detection points can improve the efficiency and accuracy of detection, increase the fault detection rate, and save precious detection resources.

[0072] (3) The present invention can generate multiple detection paths through a priority search algorithm and ensure that each path covers different key detection points, thereby providing diversity for the subsequent selection of the optimal detection path.

[0073] (4) The present invention compares the comprehensive efficiency of different paths and selects the path with the highest efficiency to ensure the optimization of the detection process. The recommended path is based on the prediction of historical data and machine learning models, ensuring the reliability and real-time performance of the detection path. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0075] Figure 1 is a flow chart of a function detection method for a fluorescent immunoassay analyzer according to an embodiment of the present invention;

[0076] Figure 2 The present invention is a block diagram of a functional detection system for a fluorescent immunoassay analyzer according to an embodiment of the present invention.

[0077] In the figure:

[0078] 1. Topology map construction module; 2. Key detection point acquisition module; 3. Detection path generation module; 4. Model training module; 5. Recommended detection path module. DETAILED DESCRIPTION

[0079] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0080] According to an embodiment of the present invention, a function detection method and system for a fluorescent immunoassay analyzer are provided.

[0081] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to one embodiment of the present invention, a function detection method for a fluorescent immunoassay analyzer is provided, and the function detection method for a fluorescent immunoassay analyzer comprises the following steps:

[0082] S1. Using the interaction logic between the functional modules in the fluorescent immunoassay analyzer, a detection task topology diagram is constructed to show the relationship between the detection tasks.

[0083] In one embodiment, using the interaction logic between the functional modules in the fluorescent immunoassay analyzer, constructing a detection task topology diagram showing the relationship between the detection tasks includes the following steps:

[0084] S11. Determine all functional modules in the fluorescent immunoassay analyzer, such as a sample processing module, a reaction module, a detection module, a data processing module, etc., and obtain a functional module list.

[0085] S12, according to the interaction logic of each functional module, including data transmission, function call, dependency, etc., for example, the sample output by the sample processing module is transmitted to the reaction module for reaction;

[0086] Define each detection task, such as sample preparation, fluorescence reaction, signal detection, data analysis, etc., and determine the dependencies between the detection tasks. For example, the fluorescence reaction task depends on the completion of the sample preparation task.

[0087] S13. Draw a topological map of the detection tasks by taking each detection task as a node and the dependency relationship as an edge. At the same time, the accuracy of the topological map can be verified through actual operation and expert review, and possible errors or omissions can be corrected.

[0088] S2. Obtain key detection points from the detection task topology map based on graph theory algorithms and establish a list of key detection points.

[0089] In one embodiment, the method of obtaining key detection points from the detection task topology map based on a graph theory algorithm and establishing a list of key detection points includes the following steps:

[0090] S21. Use the single-source shortest path algorithm (Dijkstra algorithm) to calculate the shortest paths of all node pairs in the detection task topology graph to obtain a weight matrix, which contains the shortest path length from the source node to each node.

[0091] S22. According to the weight matrix, key detection points are identified, where the key detection points have the characteristics of high connectivity or critical path length.

[0092] S23, select the starting node from the detection task topology graph, greedily select the key nodes according to the importance of the nodes (such as degree centrality, betweenness, etc.), and set the termination condition of the key node search.

[0093] S24. Analyze the neighborhood of each key node and update the list of key detection points.

[0094] S25. After multiple iterations, determine the final list of key detection points. Specifically, determine whether the termination condition is met: if the termination condition is met (such as covering all key detection tasks or reaching a predetermined number of nodes), end the search. If the termination condition is not met, use a new set of key nodes as the starting point.

[0095] In one embodiment, using a single-source shortest path algorithm to calculate the shortest paths of all node pairs in the detection task topology graph to obtain a weight matrix includes the following steps:

[0096] S211. Set the source node and set the shortest path estimation values ​​of all nodes to infinity, so that the true shortest path can be found in subsequent calculations.

[0097] S212, creating a determined set and an undetermined set, the determined set is used to store nodes for which the shortest path has been found, and the undetermined set is used to store nodes for which the shortest path has not yet been found; by creating the determined set and the undetermined set, the node status is conveniently managed.

[0098] S213. Select a node from the undetermined set (usually the node with the smallest estimated value, because the node with the smallest estimated value is most likely to be the next node to find the shortest path), move it to the determined set, and update the shortest path estimates of all nodes in the undetermined set that are directly connected to the current node.

[0099] S214. For each neighbor of the current node, if the path through the current node to the neighbor is shorter than the known path, update the shortest path estimate of the neighbor. The update formula is:

[0100] ;

[0101] In the formula, new _ distance represents the new attempted distance from the source node to the neighbor; current _ distance Represents the shortest path length from the source node to the current node; weight _ of _ edge Represents the weight of the edge from the current node to its neighbors; that is, by comparing the new trial distance with the known shortest path length, determine whether to update the shortest path estimate.

[0102] S215, repeat S213-S214 until all nodes are moved to the determined set or the termination condition of the algorithm is reached, and the shortest path to any other node is constructed from the source node by tracking the predecessor node of each node. In each iteration, the shortest path is continuously approached by selecting the node with the minimum estimated value and updating the shortest path estimate of its neighbors. Finally, the shortest path to any other node is constructed from the source node by tracking the predecessor node of each node.

[0103] In one embodiment, selecting a starting node from a detection task topology graph and greedily selecting key nodes according to the importance of the nodes comprises the following steps:

[0104] S231. Select any node as the starting node, and calculate the importance index for each node. The importance index calculation formula is:

[0105] ;

[0106] In the formula, Representation Node vBetweenness centrality of ; Represents a slave node s To Node t The number of shortest paths; represents the number of shortest paths passing through node v;

[0107] Betweenness centrality is a measure of a node’s ability to control the flow of information in a graph. It is based on the v If many shortest paths pass through the node v ,So v The betweenness centrality of is very high, which means that it plays the role of a "bridge" in the graph and has greater control over the flow of information or resources.

[0108] S232: Create an empty key node set for storing selected key nodes.

[0109] S233. From the nodes that have not been selected, according to the importance index of the nodes, the most critical nodes are selected and added to the critical node set, and the states of these nodes are marked or updated in the weight matrix.

[0110] S234, repeat S233 until a specific stop condition is met, such as reaching a predetermined number of key nodes, or covering all key paths in the graph.

[0111] Among them, when selecting key nodes, the dependencies between nodes are considered to ensure that the selected nodes can represent the integrity of the detection task, and to evaluate whether the currently selected set of key nodes covers the main part of the detection task to ensure the comprehensiveness of key detection points.

[0112] In one embodiment, analyzing the neighborhood of each key node and updating the key detection point list includes the following steps:

[0113] S241, dividing the neighborhood of the key node into several subsets, and using a local search algorithm to find a key detection point in each subset;

[0114] It should be noted that the neighborhood of the key node is decomposed into smaller subsets for more efficient search. The local search algorithm is a method to find the optimal solution in a local range, which can find possible key detection points in each subset.

[0115] S242. Find the local optimal key node set in each subset. These sets may contain key detection points. Local optimality refers to the optimal solution found in the current subset. Use the local search algorithm to find the local optimal key node set in each subset. These sets may contain key detection points, but are not necessarily globally optimal.

[0116] S243, comparing the local optimal key node set with the current key node set, selecting a better set, and if the newly found key node set is better, updating the key detection point list;

[0117] It should be noted that a better set is selected from the local optimal key node set, and the key detection point list is updated. If the newly found key node set is better than the current key node set, the key detection point list is updated. The better set can be selected by comparing their performance indicators, such as detection accuracy, detection sensitivity and other indicators.

[0118] For example, the fluorescence immunoassay analyzer includes the following functional modules and detection tasks:

[0119] Functional modules: sample processing module, reagent processing module, reaction module, detection module, data processing module.

[0120] Detection tasks: T1, T2, T3, T4, T5, T6.

[0121] T1: Sample processing → reaction.

[0122] T2: Reagent treatment → reaction.

[0123] T3: Reaction → Detection.

[0124] T4: Reaction → data processing.

[0125] T5: Detection → data processing.

[0126] T6: Data processing → sample processing.

[0127] Select node T1 as the source node, the determined set: {T1}. The undetermined set: {T2, T3, T4, T5, T6}.

[0128] Select node T3 connected to T1 and update the shortest path estimate of T3:

[0129] T3:1 (T1→T3).

[0130] Determined set: {T1, T3}.

[0131] Undetermined set: {T2, T4, T5, T6}.

[0132] Select node T5 connected to T3 and update the shortest path estimate of T5:

[0133] T5:2 (T1→T3→T5).

[0134] The set has been determined: {T1, T3, T5}.

[0135] Undetermined set: {T2, T4, T6}.

[0136] Select node T6 connected to T5 and update the shortest path estimate of T6:

[0137] T6:3 (T1→T3→T5→T6).

[0138] The set has been determined: {T1, T3, T5, T6}.

[0139] Undetermined set: {T2, T4}.

[0140] Select the node connected to T6:

[0141] At this time, T6 is not further connected to any node in the undetermined set, so the shortest path estimation values ​​of other nodes are not updated.

[0142] Final shortest path estimates: T1: 0, T2: ∞, T3: 1, T4: ∞, T5: 2, T6: 3. Key checkpoints: T1, T3, T5, T6. Select key nodes based on betweenness centrality, where the initial set of key nodes is empty. Select T3 because it has the highest betweenness centrality (2). Update the neighborhood and select T5 (because of the next highest betweenness centrality of 1). After the neighborhood is updated, select T6 (no further selection). Final list of key checkpoints: T1, T3, T5, T6.

[0143] S3. Using the priority search algorithm, several detection paths are generated from the starting point to the end point, and each detection path covers different key detection points.

[0144] In one embodiment, a priority search algorithm is used to generate several detection paths from the starting point to the end point, and each detection path covers different key detection points, including the following steps:

[0145] S31, selecting a starting point node in the detection task topology map as the starting point of the search, and initializing a stack structure for storing nodes accessed during the traversal process;

[0146] The starting point is the starting node of the detection path search, which usually represents the beginning or entry point of the detection task. The starting point is generally fixed, such as the first step of the detection process; at the same time, there are multiple choices for the starting point, such as choosing a starting point from multiple initial detection tasks. The end point is the end node of the detection path search, representing the completion or exit point of the detection task. The end point is not necessarily a single node, but can also be a group of nodes. As long as any of these nodes is reached, the detection path can be considered complete.

[0147] S32: Push the starting point node into the stack and mark it as visited. At the same time, take out the current node from the top of the stack and traverse all unvisited adjacent nodes.

[0148] S33. For each adjacent node, if it is a key detection point and has not been visited, it is pushed into the stack and marked as visited, otherwise it is not pushed into the stack to avoid repeated searches.

[0149] S34, backtracking the current node and recording the node combinations that have been visited;

[0150] S35. Repeat S32-S34 until all key detection points have been visited, and then end the search.

[0151] S36. Output all generated detection paths as search results.

[0152] The priority search algorithm starts from the starting node and gradually expands the search scope to find multiple detection paths covering different key detection points.

[0153] In one embodiment, backtracking the current node and recording the visited node combinations includes the following steps:

[0154] S341, if all adjacent nodes of the current node have been visited, then pop the current node from the stack and backtrack, and record the path from the starting point to the current node each time when backtracking. When all adjacent nodes of the current node have been visited, it means that the exploration of the node has been completed.

[0155] S342. During the search process, a set is used to record the node combinations that have been visited to avoid generating duplicate paths.

[0156] In S2, we get a list of key checkpoints: T1, T3, T5, and T6. Starting from T1, mark T1 as visited, and check its adjacent node T3. Stack structure: [T1, T3]. Visited mark: {T1, T3}. T3 is a key node, so push it into the stack and continue. Next, visit T3's adjacent node T5. Stack structure: [T1, T3, T5]. Visited mark: {T1, T3, T5}. T5 is also a key node, so push it into the stack and continue. Finally, reach T6 from T5. Stack structure: [T1, T3, T5, T6]. Visited mark: {T1, T3, T5, T6}. T6 is also a key node as the end point, so push it into the stack. At this point, a complete path is found: T1→T3→T5→T6. Record this path and backtrack to search for other possible paths.

[0157] S4. Obtain the average time consumption, failure rate and other information of each key inspection point extracted from the historical inspection records to obtain the historical data set, and use the machine learning algorithm to train the inspection path efficiency prediction model, which can predict the comprehensive efficiency of different inspection paths (such as time consumption, resource consumption, fault detection rate, etc.).

[0158] Among them, comprehensive efficiency is a comprehensive indicator, which is calculated based on information such as time consumption, resource consumption, and fault detection rate. For example, by combining the three aspects of average time consumption, resource consumption, and fault detection rate with corresponding weights, a higher comprehensive efficiency means that a better balance has been achieved between these factors, achieving a higher efficiency level.

[0159] Duration: The total time required to execute the path.

[0160] Resource consumption: The total amount of resources consumed by each operation in the detection path.

[0161] Fault detection rate: The probability of detecting a fault on the path.

[0162] In one embodiment, obtaining information of each key detection point extracted from historical detection records to obtain a historical data set and using a machine learning algorithm to train a detection path efficiency prediction model includes the following steps:

[0163] S41. Extract data from the historical detection records of the fluorescent immunoassay analyzer, clean the data and extract features to obtain a historical data set, i.e., a detection point feature data set; the data includes but is not limited to the start and end time of each detection task, the key detection points involved, the amount of resources used, the type and frequency of faults that occur, etc.

[0164] S42. Normalize the features in the historical data set and divide the historical data set into a training set and a test set.

[0165] S43. Select a machine learning algorithm, and use historical data sets to perform model training and verification on the selected machine learning algorithm to obtain a detection path efficiency prediction model.

[0166] In one embodiment, selecting a machine learning algorithm and using a historical data set to perform model training and verification on the selected machine learning algorithm includes the following steps:

[0167] S431. Select a decision tree algorithm, and use a training set to train the decision tree algorithm to obtain a detection path efficiency prediction model.

[0168] S432. Use the test set to verify the detection path efficiency prediction model and evaluate the prediction accuracy and effect of the detection path efficiency prediction model.

[0169] S5. Input all detection paths into the detection path efficiency prediction model, compare and select one or more detection paths with the highest comprehensive efficiency as recommended detection paths.

[0170] Specifically, the sorted detection path data is input into the trained detection path efficiency prediction model, and a comprehensive efficiency prediction is performed on each detection path to generate an efficiency score for each path. All detection paths are sorted according to the comprehensive score, from high to low.

[0171] like Figure 2 As shown, according to another embodiment of the present invention, a functional detection system for a fluorescent immunoassay analyzer is also provided, the functional detection system for a fluorescent immunoassay analyzer comprises a topology map construction module 1, a key detection point acquisition module 2, a detection path generation module 3, a model training module 4 and a detection path recommendation module 5;

[0172] The topology map construction module 1, the key detection point acquisition module 2, the detection path generation module 3, the model training module 4 and the detection path recommendation module 5 are connected in sequence;

[0173] A topology map construction module 1 is used to construct a detection task topology map showing the relationship between various detection tasks by using the interaction logic between various functional modules in the fluorescent immunoassay analyzer;

[0174] The key detection point acquisition module 2 is used to obtain key detection points from the detection task topology map based on the graph theory algorithm and establish a key detection point list;

[0175] The detection path generation module 3 is used to generate several detection paths from the starting point to the end point by using a priority search algorithm, and each detection path covers different key detection points;

[0176] Model training module 4, used to obtain information of each key detection point extracted from the historical detection records, obtain the historical data set, and use the machine learning algorithm to train the detection path efficiency prediction model;

[0177] The detection path recommendation module 5 is used to input all detection paths into the detection path efficiency prediction model, compare and select the detection path with the highest comprehensive efficiency as the recommended detection path.

[0178] In summary, the present invention provides a method and system for functional detection of fluorescent immunoassay analyzer, which optimizes the detection path, reduces unnecessary repetition and redundancy, improves the overall speed and efficiency of detection, and reasonably allocates detection resources by identifying key detection points and paths, reduces resource waste, and improves the detection rate of faults in the detection process through optimization based on historical data, ensures the reliability of the detection results, and uses machine learning algorithms to predict and select the best path to ensure the real-time and dynamic adaptability of the detection process. The present invention uses a graph theory algorithm to identify the most critical detection points in the detection process of the fluorescent immunoassay analyzer. By identifying the key detection points, resource allocation can be optimized to ensure the operating efficiency and reliability of the key detection points. It is of great significance to the optimization of the functional detection of the fluorescent immunoassay analyzer. The discovery and priority treatment of key detection points can improve the efficiency and accuracy of detection, improve the detection rate of faults, and save precious detection resources. The present invention can generate multiple detection paths through a priority search algorithm, and ensure that each path covers different key detection points, providing diversity for the subsequent selection of the optimal detection path. The present invention compares the comprehensive efficiency of different paths, selects the most efficient path, and ensures the optimization of the detection process. The recommended path is based on historical data and predictions from machine learning models, ensuring the reliability and real-time nature of the detection path.

[0179] 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 principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A functional detection method for a fluorescent immunoassay analyzer, characterized in that: The functional detection method for the fluorescent immunoassay analyzer comprises the following steps: S1. Using the interaction logic between the functional modules in the fluorescent immunoassay analyzer, a detection task topology diagram is constructed to show the relationship between the detection tasks. S2, based on graph theory algorithm, obtain key detection points from the detection task topology map and establish a list of key detection points; S3, using a priority search algorithm, generate several detection paths from the starting point to the end point, and each detection path covers different key detection points; S4, obtaining information of each key inspection point extracted from the historical inspection records, obtaining a historical data set, and using a machine learning algorithm to train an inspection path efficiency prediction model; S5. Input all detection paths into the detection path efficiency prediction model, compare and select the detection path with the highest comprehensive efficiency as the recommended detection path; The method of obtaining key detection points from the detection task topology map based on the graph theory algorithm and establishing a list of key detection points includes the following steps: S21, using a single-source shortest path algorithm to calculate the shortest paths of all node pairs in the detection task topology graph, and obtain a weight matrix, where the weight matrix includes the shortest path length from the source node to each node; S22, identifying key detection points according to the weight matrix; S23, selecting a starting node from the detection task topology graph, greedily selecting key nodes according to the importance of the nodes, and setting a termination condition for the key node search; S24, analyzing the neighborhood of each key node and updating the key detection point list; S25. After multiple iterations, the final list of key inspection points is determined.

2. A functional detection method for a fluorescent immunoassay analyzer according to claim 1, characterized in that: The method of using the interaction logic between the functional modules in the fluorescent immunoassay analyzer to construct a detection task topology diagram showing the relationship between the detection tasks includes the following steps: S11, determining all functional modules in the fluorescence immunoassay analyzer; S12. Define each detection task according to the interaction logic of each functional module, and determine the dependency relationship between each detection task; S13. Draw a detection task topology graph by taking each detection task as a node and the dependency relationship as an edge.

3. A functional detection method for a fluorescent immunoassay analyzer according to claim 2, characterized in that: The method of using a single-source shortest path algorithm to calculate the shortest paths of all node pairs in the detection task topology graph to obtain a weight matrix includes the following steps: S211, set the source node, and set the shortest path estimation values ​​of all nodes to infinity; S212, creating a determined set and an undetermined set, the determined set is used to store nodes for which the shortest paths have been found, and the undetermined set is used to store nodes for which the shortest paths have not been found; S213, selecting a node from the undetermined set, moving it to the determined set, and updating the shortest path estimation values ​​of all nodes in the undetermined set that are directly connected to the current node; S214. For each neighbor of the current node, if the path through the current node to the neighbor is shorter than the known path, update the shortest path estimate of the neighbor. The update formula is: ； In the formula, new _ distance represents the new attempted distance from the source node to the neighbor; current _ distance Represents the shortest path length from the source node to the current node; weight _ of _ edge Represents the weight of the edge from the current node to its neighbors; S215. Repeat S213-S214 until all nodes are moved to the determined set, and the shortest path to any other node is constructed starting from the source node by tracking the predecessor node of each node.

4. A functional detection method for a fluorescent immunoassay analyzer according to claim 3, characterized in that: The step of selecting a starting node from the detection task topology graph and greedily selecting key nodes according to the importance of the nodes comprises the following steps: S231, select any node as the starting node, and calculate the importance index for each node. The importance index calculation formula is: ； In the formula, Representation Node v Betweenness centrality of ; Represents a slave node s To Node t The number of shortest paths; represents the number of shortest paths passing through node v; S232, creating an empty key node set for storing the selected key nodes; S233, from the nodes that have not been selected, according to the importance index of the nodes, select the most critical nodes and add them to the critical node set; S234: Repeat S233 until a specific stop condition is met.

5. A functional detection method for a fluorescent immunoassay analyzer according to claim 4, characterized in that: The analyzing the neighborhood of each key node and updating the key detection point list comprises the following steps: S241, dividing the neighborhood of the key node into several subsets, and using a local search algorithm to find a key detection point in each subset; S242. Find the local optimal set of key nodes in each subset; S243. Compare the local optimal key node set with the current key node set, select a better set, and if the newly found key node set is better, update the key detection point list.

6. A functional detection method for a fluorescent immunoassay analyzer according to claim 5, characterized in that: The method of using a priority search algorithm to generate several detection paths from the starting point to the end point, wherein each detection path covers different key detection points, includes the following steps: S31, selecting a starting point node in the detection task topology map as the starting point of the search, and initializing a stack structure for storing nodes accessed during the traversal process; S32, push the starting point node into the stack and mark it as visited, take out the current node from the top of the stack, and traverse all unvisited adjacent nodes; S33. For each adjacent node, if it is a key detection point and has not been visited, it is pushed into the stack and marked as visited, otherwise it is not pushed into the stack; S34, backtracking the current node and recording the node combinations that have been visited; S35, repeat S32-S34 until all key detection points have been visited, and then end the search; S36. Output all generated detection paths as search results.

7. A functional detection method for a fluorescent immunoassay analyzer according to claim 6, characterized in that: The step of backtracking the current node and recording the visited node combinations includes the following steps: S341. If all adjacent nodes of the current node have been visited, the current node is popped from the stack and backtracked. Each time the backtracking is performed, the path from the starting point to the current node is recorded. S342. During the search process, a collection is used to record the node combinations that have been visited.

8. A functional detection method for a fluorescent immunoassay analyzer according to claim 7, characterized in that: The method of obtaining information of each key detection point extracted from the historical detection records, obtaining a historical data set, and using a machine learning algorithm to train a detection path efficiency prediction model comprises the following steps: S41, extracting data from the historical test records of the fluorescent immunoassay analyzer, cleaning the data and extracting features to obtain a historical data set; S42, normalizing the features in the historical data set, and dividing the historical data set into a training set and a test set; S43. Select a machine learning algorithm, and use historical data sets to perform model training and verification on the selected machine learning algorithm to obtain a detection path efficiency prediction model.

9. A functional detection method for a fluorescent immunoassay analyzer according to claim 8, characterized in that: The method of selecting a machine learning algorithm and using a historical data set to perform model training and verification on the selected machine learning algorithm comprises the following steps: S431, selecting a decision tree algorithm, and using a training set to train the decision tree algorithm to obtain a detection path efficiency prediction model; S432. Use the test set to verify the detection path efficiency prediction model and evaluate the prediction accuracy and effect of the detection path efficiency prediction model.

10. A function detection system for a fluorescent immunoassay analyzer, used to implement the function detection method for a fluorescent immunoassay analyzer according to any one of claims 1 to 9, characterized in that: The functional detection system for the fluorescent immunoassay analyzer includes a topological map construction module, a key detection point acquisition module, a detection path generation module, a model training module and a detection path recommendation module; The topology map construction module, the key detection point acquisition module, the detection path generation module, the model training module and the recommended detection path module are connected in sequence; The topology map construction module is used to construct a detection task topology map showing the relationship between various detection tasks by using the interaction logic between various functional modules in the fluorescent immunoassay analyzer; The key detection point acquisition module is used to obtain key detection points from the detection task topology map based on a graph theory algorithm and establish a key detection point list; The detection path generation module is used to generate several detection paths from the starting point to the end point using a priority search algorithm, and each detection path covers different key detection points; The model training module is used to obtain information of each key detection point extracted from the historical detection records, obtain a historical data set, and use a machine learning algorithm to train a detection path efficiency prediction model; The recommended detection path module is used to input all detection paths into the detection path efficiency prediction model, compare and select the detection path with the highest comprehensive efficiency as the recommended detection path.

Images