Rapid detection method and system for pathogens

Abstract

According to the rapid pathogen detection method and system provided by the invention, when the to-be-analyzed pathogen sample data is processed, in order to obtain a better pathogen sample data processing effect, selection of multi-state pathogen attribute knowledge fields is performed on the to-be-analyzed pathogen sample data; according to the pathogen attribute knowledge field of each state, evaluating the credibility coefficient, namely the influence of the information of each state on the pathogen sample data processing instruction, and according to the pathogen attribute knowledge field of each state and the corresponding credibility coefficient, evaluating the credibility of the pathogen sample data processing instruction; the target pathogen knowledge fields are finally used for determining all states of pathogen detection results of the pathogen sample data. According to the method provided by the embodiment of the invention, by estimating the credible coefficient of the information of each state, pathogens can be rapidly detected, the types of the pathogens can be accurately obtained, obligational workers can treat patients in time according to the types of the pathogens, and the life safety of the patients can be guaranteed.

Description

Technical Field

[0001] The present application relates to the technical field of rapid detection of pathogens, and more specifically, to a method and system for rapid detection of pathogens. Background Art

[0002] Diseases caused by emerging pathogens are emerging at an unprecedented rate (one new disease per year on average) and spreading across geographical boundaries around the world. Emerging infectious diseases are prone to cause major social impacts due to the lack of existing analytical detection technology and diagnostic and treatment methods, as well as the general lack of immunity among the population. Therefore, a rapid detection method for pathogens is urgently needed to solve the above technical problems. Summary of the invention

[0003] In order to improve the technical problems existing in the related technologies, the present application provides a method and system for rapid detection of pathogens.

[0004] In a first aspect, a method for rapid detection of pathogens is provided, comprising: Obtaining pathogen sample data to be analyzed; In combination with the pathogen sample data to be analyzed, selecting pathogen attribute knowledge fields of at least two states corresponding to the pathogen sample data to be analyzed; For each state of the pathogen attribute knowledge field, the influence of the pathogen attribute knowledge field of the state is determined according to the pathogen attribute knowledge field of the state, and the influence of the pathogen attribute knowledge field of each state represents the credibility coefficient of the pathogen attribute knowledge field of the state for the pathogen sample data processing instruction; Determine the target pathogen knowledge field for each state based on the pathogen attribute knowledge field of each state and the corresponding influence; Based on the target pathogen knowledge field of each state, the pathogen detection result of the pathogen sample data to be analyzed corresponding to the pathogen sample data processing instruction is determined.

[0005] In an independently implemented embodiment, determining the target pathogen knowledge field of each state according to the pathogen attribute knowledge field of each state and the corresponding influence includes: For each state of the pathogen attribute knowledge field, determining a matching result between the pathogen attribute knowledge field of the state and each of the pathogen attribute knowledge fields of the at least two states; For each state of the pathogen attribute knowledge field, based on the matching result between the pathogen attribute knowledge field of the state and the pathogen attribute knowledge field of each state in the pathogen attribute knowledge fields of at least two states, weighted concatenation is performed on the pathogen attribute knowledge fields of each state in the pathogen attribute knowledge fields of at least two states to obtain the concatenated knowledge field of the state; The influence corresponding to each state is used to weight the concatenated knowledge fields of the corresponding states to obtain the target pathogen knowledge field of each state.

[0006] In an independently implemented embodiment, the selecting of pathogen attribute knowledge fields of at least two states corresponding to the pathogen sample data to be analyzed in combination with the pathogen sample data to be analyzed includes: A knowledge field is selected for at least one of the following information corresponding to the pathogen sample data to be analyzed to obtain an information knowledge field for each information, and the information knowledge field for each information is determined as a pathogen attribute knowledge field of a state: The pathogen sample data to be analyzed; The pathogen event corresponding to the pathogen sample data to be analyzed; The continuous recorded data corresponding to each pathogen sample data, the continuous recorded data corresponding to each pathogen sample data is the pathogen sample data range where a target item in the pathogen sample data to be analyzed is located.

[0007] In an independently implemented embodiment, the selecting of pathogen attribute knowledge fields of at least two states corresponding to the pathogen sample data to be analyzed in combination with the pathogen sample data to be analyzed includes: Selecting knowledge fields for the pathogen sample data to be analyzed to obtain all pathogen sample pathogen attribute knowledge fields for the pathogen sample data to be analyzed; In combination with the pathogen attribute knowledge fields of all pathogen samples, a target pathogen sample data range associated with the pathogen sample data processing instruction in the pathogen sample data to be analyzed is estimated; The target pathogen sample data range is subjected to knowledge field selection to obtain a key pathogen sample pathogen attribute knowledge field, and the key pathogen sample pathogen attribute knowledge field is determined as a pathogen attribute knowledge field in a state.

[0008] In an independently implemented embodiment, the pathogen attribute knowledge fields of all pathogen samples include a plurality of knowledge field relationship networks, and the step of estimating the target pathogen sample data range associated with the pathogen sample data to be analyzed and the pathogen sample data processing instruction in combination with the pathogen attribute knowledge fields of all pathogen samples includes: The knowledge field values ​​of the knowledge field points at the same position in the plurality of knowledge field relationship networks are spliced ​​to obtain a spliced ​​knowledge field relationship network; Determine the influence of each knowledge field point in the spliced ​​knowledge field relationship network in combination with the spliced ​​knowledge field relationship network, wherein the influence of each knowledge field point represents the degree of association between the pathogen sample data range corresponding to the knowledge field point and the pathogen sample data processing instruction; According to the influence of each knowledge field point in the spliced ​​knowledge field relationship network, the target pathogen sample data range in the pathogen sample data to be analyzed is determined.

[0009] In an independently implemented embodiment, the at least one type of information includes the pathogen event, and the knowledge field selection is performed on various types of information in the at least one type of information to obtain the information knowledge field of each type of information, including: Selecting knowledge fields for each pathogen event unit in the pathogen event to obtain partial knowledge fields of each pathogen event unit; Select all knowledge fields for the pathogen event to obtain all knowledge fields of the pathogen event; All knowledge fields of the pathogen event and partial knowledge fields of each pathogen event unit are spliced ​​to obtain the pathogen event knowledge field of the pathogen event, and the pathogen event knowledge field is determined as a pathogen attribute knowledge field of a state.

[0010] In an independently implemented embodiment, the pathogen event knowledge field of the pathogen event is obtained by concatenating all knowledge fields of the pathogen event and partial knowledge fields of each pathogen event unit, including: splicing all knowledge fields of the pathogen event and part of the knowledge fields of each pathogen event unit to obtain a spliced ​​knowledge field; By performing knowledge field selection again on the spliced ​​knowledge field, the spliced ​​knowledge field is switched to the knowledge field coordinate system corresponding to the pathogen sample data processing instruction, and the re-selected knowledge field is determined as the pathogen event knowledge field of the pathogen event.

[0011] In an independently implemented embodiment, for each state of the pathogen attribute knowledge field, determining the influence of the pathogen attribute knowledge field of the state according to the pathogen attribute knowledge field of the state includes: If the pathogen attribute knowledge field of the state is the pathogen attribute knowledge field of the pathogen sample, based on the various knowledge field relationship networks in the pathogen attribute knowledge field of the pathogen sample, determine the influence of each knowledge field relationship network; using the influence of each knowledge field relationship network in the pathogen attribute knowledge field of the pathogen sample, weight the corresponding knowledge field relationship network, and determine the influence of the pathogen attribute knowledge field of the pathogen sample based on the weighted knowledge field relationship networks; If the pathogen attribute knowledge field of the state is a pathogen event knowledge field, the influence of the pathogen event knowledge field is determined according to the knowledge field value of each knowledge field point in the pathogen event knowledge field.

[0012] In an independently implemented embodiment, the decision network includes a first artificial intelligence network and a second artificial intelligence network, wherein the first artificial intelligence network is pre-configured; for each of the pathogen sample data examples, the selecting of knowledge field examples of at least two states corresponding to the pathogen sample data example includes: Selecting, by the first artificial intelligence network, a first pathogen event knowledge field, a sample pathogen attribute knowledge field, and a portion of pathogen sample pathogen attribute knowledge fields corresponding to the pathogen sample data example, which are within the pathogen sample data range where each target item in at least one target item in the pathogen sample data example is located; Performing knowledge field switching on the first pathogen event knowledge field through the second artificial intelligence network to obtain a second pathogen event knowledge field, and identifying an example target range associated with a pathogen sample data processing instruction in the pathogen sample data example through the second artificial intelligence network, and selecting an important knowledge field example of the example target range; The second pathogen event knowledge field, the sample pathogen attribute knowledge field, the pathogen attribute knowledge field of some pathogen samples in the pathogen sample data range of each target item, and the important knowledge field examples are respectively determined as knowledge field examples of a state; wherein the influence of the knowledge field example of each state and the first pathogen detection result of the pathogen sample data example are determined by a second artificial intelligence network; The debugging of the network coefficients of the decision network includes: debugging the network coefficients of the second artificial intelligence network.

[0013] In an independently implemented embodiment, for each of the pathogen sample data examples, selecting knowledge field examples of at least two states corresponding to the pathogen sample data example includes: Select all pathogen attribute knowledge fields of the pathogen sample data example; According to all pathogen sample pathogen attribute knowledge fields of the pathogen sample data example, the example target range in the pathogen sample data example is determined, and important knowledge field examples of the example target range are selected, and the important knowledge field examples are determined as knowledge field examples of a state.

[0014] In an independently implemented embodiment, determining the influence of each knowledge field relationship network based on each knowledge field relationship network in the pathogen attribute knowledge field of the pathogen sample includes: For each knowledge field relationship network, the average value of the knowledge field values ​​of each knowledge field point in the knowledge field relationship network is determined as the influence of the knowledge field relationship network; The determining of the influence of the pathogen attribute knowledge field of the pathogen sample based on the weighted knowledge field relationship network includes: determining the average of the knowledge field values ​​of all knowledge field points in the weighted knowledge field relationship network as the influence of the pathogen attribute knowledge field of the pathogen sample.

[0015] In an independently implemented embodiment, the method is implemented by calling a configured pathogen sample data processing network, wherein the pathogen sample data processing network is configured in the following manner: Obtaining a set of configured pathogen examples, the set of configured pathogen examples comprising a plurality of pathogen sample data examples with a pathogen subject, the pathogen subject of each of the pathogen sample data examples representing a real-time pathogen detection result of the pathogen sample data example corresponding to the pathogen sample data instruction; Repeat the configuration operation on the decision network in combination with the configured pathogen example set until the configuration termination requirement is met, and determine the decision network that meets the configuration termination requirement as the configured pathogen sample data processing network, wherein the configuration operation includes the following steps: For each of the pathogen sample data examples, select knowledge field examples of at least two states corresponding to the pathogen sample data example; For each of the pathogen sample data examples, determining the influence of the knowledge field example of each state of the pathogen sample data example according to the knowledge field example of the state of the pathogen sample data example; For each of the pathogen sample data examples, based on the knowledge field example of each state of the pathogen sample data example and the corresponding influence, determine a target knowledge field example of each state of the pathogen sample data example; For each of the pathogen sample data examples, based on the target knowledge field examples of each state of the pathogen sample data example, estimating a first pathogen detection result of the pathogen sample data example corresponding to the pathogen sample data processing instruction; Determine a first configuration quality indicator based on the pathogen theme of each of the pathogen sample data examples and the influence corresponding to each state, and determine a second configuration quality indicator based on the pathogen theme of each of the pathogen sample data examples and the first pathogen detection result; For each of the pathogen sample data examples, based on all pathogen sample pathogen attribute knowledge fields corresponding to the important knowledge field examples of the pathogen sample data example, estimate the second pathogen detection result corresponding to the pathogen sample data example; Determining a third configuration quality indicator according to the pathogen subject corresponding to each of the pathogen sample data examples and the second pathogen detection result; Obtaining a configuration total quality indicator according to the first configuration quality indicator, the second configuration quality indicator and the third configuration quality indicator; If the configuration termination requirement is not met, the network coefficients of the decision network are debugged in combination with the configuration total quality index.

[0016] In a second aspect, a rapid pathogen detection system is provided, comprising a processor and a memory communicating with each other, wherein the processor is used to read a computer program from the memory and execute the computer program to implement the above method.

[0017] The rapid detection method and system of pathogens provided in the embodiment of the present application, when processing the pathogen sample data to be analyzed, in order to obtain a better pathogen sample data processing effect, will select the pathogen attribute knowledge fields of the pathogen sample data to be analyzed in multiple states, and on this basis, will further evaluate the credibility coefficient of each state of information on the pathogen sample data processing instruction, that is, the influence, according to the pathogen attribute knowledge fields of each state, and obtain the target pathogen knowledge fields of each state that are finally used to determine the pathogen detection results of the pathogen sample data according to the pathogen attribute knowledge fields of each state and the corresponding credibility coefficients. The method provided in the embodiment of the present application, by estimating the credibility coefficient of the information of each state, can quickly detect pathogens, can accurately obtain the type of pathogens, medical workers can provide timely treatment to patients according to the type of pathogens, and can ensure the life safety of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A flowchart of a method for rapid detection of pathogens provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to better understand the above technical scheme, the technical scheme of the present application is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical scheme of the present application, rather than limitations on the technical scheme of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.

[0021] See also Figure 1 , shows a method for rapid detection of pathogens, which may include the technical solutions described in the following steps S110-S150.

[0022] Step S110: Obtain pathogen sample data to be analyzed.

[0023] Exemplarily, the pathogen sample data to be analyzed is obtained by examining patients to obtain information such as pathogen properties.

[0024] The specific source of the pathogen sample data to be analyzed is not limited in the present embodiment of the application. The pathogen sample data to be analyzed can be an independent piece of pathogen sample data, or can be pathogen sample data in a video.

[0025] Step S120: Based on the pathogen sample data to be analyzed, select pathogen attribute knowledge fields of at least two states corresponding to the pathogen sample data to be analyzed.

[0026] Among them, knowledge fields can be understood as features.

[0027] In an embodiment of the present application, for the pathogen sample data to be analyzed, the pathogen attribute knowledge fields of at least two states corresponding to the pathogen sample data can be obtained by selecting information from different sources corresponding to the pathogen sample data to be analyzed. Optionally, the information from different sources may include but is not limited to the pathogen sample data itself to be analyzed, the pathogen events corresponding to the pathogen sample data to be analyzed, and the partial range of pathogen sample data in the pathogen sample data to be analyzed. Then, all the pathogen sample pathogen attribute knowledge fields of the pathogen sample data can be obtained by selecting all the knowledge fields of the pathogen sample data to be analyzed, and the pathogen event knowledge fields of the pathogen sample data can be obtained by selecting the knowledge fields through the pathogen events, and the pathogen event knowledge fields of the pathogen sample data can be obtained by performing the knowledge field selection on the partial range of pathogen sample data. The pathogen attribute knowledge fields of the pathogen sample data corresponding to the pathogen sample data can be obtained by performing the knowledge field selection on the pathogen event. The pathogen attribute knowledge fields of the pathogen sample data corresponding to the pathogen sample data can be obtained. The above-mentioned pathogen attribute knowledge fields of all pathogen samples, pathogen event knowledge fields, and pathogen attribute knowledge fields of the pathogen sample data can be respectively determined as pathogen attribute knowledge fields of one state.

[0028] The pathogen event corresponding to the pathogen sample data to be analyzed may include the pathogen event information contained in the pathogen sample data to be analyzed or the pathogen event description of the pathogen sample data generated based on the pathogen sample data to be analyzed. Optionally, when the pathogen sample data to be analyzed includes the pathogen event, the pathogen event corresponding to the pathogen sample data can be obtained by performing pathogen event identification on the pathogen sample data to be analyzed. If the pathogen sample data to be analyzed does not include the pathogen event, the pathogen event description corresponding to the pathogen sample data can be generated through a configured decision network, and the pathogen event description is determined as the pathogen event corresponding to the pathogen sample data.

[0029] In an optional embodiment of the present application, the above-mentioned selection of pathogen attribute knowledge fields of at least two states corresponding to the pathogen sample data to be analyzed based on the pathogen sample data to be analyzed may include: selecting knowledge fields for at least one of the following information corresponding to the pathogen sample data to be analyzed, obtaining the information knowledge field of each information, and determining the information knowledge field of each information as a pathogen attribute knowledge field of one state: pathogen sample data to be analyzed; pathogen events corresponding to the pathogen sample data to be analyzed; continuous record data corresponding to each pathogen sample data, and the continuous record data corresponding to each pathogen sample data is the pathogen sample data range where a target item in the pathogen sample data to be analyzed is located.

[0030] Among them, the pathogen attribute knowledge field of each state can be implemented by a corresponding decision network. The continuous record data corresponding to the above pathogen sample data is the pathogen sample data in the range where the target item in the pathogen sample data to be analyzed is located, that is, the sub-pathogen sample data. The target item specifically refers to which or which types of items, which can be configured in advance according to actual application requirements. Optionally, the target item can include but is not limited to at least one of the specified items or specified parts of the items in the pathogen sample data to be analyzed. For example, the specified item can be a person or a body part of a person. If there are several target items in the pathogen sample data to be analyzed, the pathogen sample data range where each target item is located in the pathogen sample data to be analyzed can be respectively determined as a continuous record data corresponding to a pathogen sample data, and the knowledge field can be selected for the continuous record data corresponding to each pathogen sample data to obtain the pathogen sample pathogen attribute knowledge field corresponding to the continuous record data corresponding to each pathogen sample data. The pathogen sample pathogen attribute knowledge field corresponding to the continuous record data corresponding to each pathogen sample data can be determined as a pathogen attribute knowledge field of a state.

[0031] Relative to the knowledge field of the pathogen sample data to be analyzed (i.e., all pathogen sample pathogen attribute knowledge fields of the pathogen sample data to be analyzed, referred to as the first pathogen sample pathogen attribute knowledge field here), the pathogen sample pathogen attribute knowledge field corresponding to the continuous record data corresponding to the pathogen sample data can be understood as the partial pathogen sample pathogen attribute knowledge field of the pathogen sample data to be analyzed. Optionally, before obtaining the partial pathogen sample pathogen attribute knowledge field, the pathogen sample data to be analyzed can be detected first to detect the range where each target item in the pathogen sample data is located, and then the knowledge field selection network is used to select the knowledge field of the range to obtain the partial pathogen sample pathogen attribute knowledge field of the pathogen sample data range where each target item is located.

[0032] For pathogen events corresponding to the pathogen sample data to be analyzed, the pathogen event identification can be performed on the pathogen sample data to be analyzed through an artificial intelligence recognition network or other methods to obtain the pathogen event information of the pathogen sample data to be analyzed. Thereafter, the corresponding pathogen event knowledge field can be obtained by selecting the knowledge field of the pathogen event information.

[0033] Optionally, at least one of the above-mentioned information includes a pathogen event corresponding to the pathogen sample data to be analyzed. The information knowledge field of this information, namely the pathogen event knowledge field, can be obtained in the following manner: selecting the knowledge field of each pathogen event unit in the pathogen event to obtain a partial knowledge field of each pathogen event unit; selecting all the knowledge fields of the pathogen event to obtain the entire knowledge field of the pathogen event; splicing all the knowledge fields of the pathogen event and the partial knowledge fields of each pathogen event unit to obtain the pathogen event knowledge field of the pathogen event, and determining the pathogen event knowledge field as a pathogen attribute knowledge field of a state.

[0034] Optionally, the above-mentioned splicing of all the knowledge fields of the pathogen event and part of the knowledge fields of each pathogen event unit to obtain the pathogen event knowledge field of the pathogen event may include: splicing all the knowledge fields of the pathogen event and part of the knowledge fields of each pathogen event unit to obtain the spliced ​​knowledge field; performing knowledge field selection on the spliced ​​knowledge field again, switching the spliced ​​knowledge field to the knowledge field coordinate system corresponding to the pathogen sample data processing instruction, and determining the re-selected knowledge field as the pathogen event knowledge field of the pathogen event.

[0035] Optionally, the concatenated knowledge field can be mapped to a specific knowledge field coordinate system through one or more cascaded fully connected layers, and the mapped knowledge field can be determined as the pathogen event knowledge field of the pathogen sample data to be analyzed.

[0036] In the pathogen sample data processing scenario, the selection of knowledge fields is generally considered to be an upstream knowledge field selection instruction, and obtaining the pathogen detection results of the pathogen sample data based on the selected knowledge fields is a downstream instruction. The processing method provided by the present application can be implemented by a configured pathogen sample data processing network, wherein the pathogen sample data processing network may include a pathogen event knowledge field selection module for selecting the above-mentioned pathogen event knowledge field, a knowledge field selection module for selecting knowledge fields of other states, and a neural network structure for implementing subsequent steps. When configuring the pathogen sample data processing network, in order to improve the configuration efficiency and effect of the network, the pathogen event knowledge field selection module can be configured in advance, that is, when configuring the entire pathogen sample data processing network, the network coefficient of the pathogen event knowledge field selection module can be fixed, but this method may cause the pathogen event knowledge field selected by the pathogen event knowledge field selection module to be a general knowledge field, which may not be well applicable to the current pathogen sample data processing instruction. In order to solve this problem, the solution provided by the present application connects one or more knowledge field mapping modules (such as one or more fully connected layers) after the pathogen event knowledge field selection module for general knowledge field selection. Through configuration, the knowledge field mapping module can switch the general knowledge field selected by the knowledge field selection module to a pathogen event knowledge field that is more suitable for the current pathogen sample data processing instruction, so as to further improve the pathogen sample data processing effect.

[0037] Optionally, the pathogen attribute knowledge fields of at least two states include pathogen sample pathogen attribute knowledge fields (key pathogen sample pathogen attribute knowledge fields) of a target pathogen sample data range (key pathogen sample data range) in the pathogen sample data to be analyzed, wherein the key pathogen sample pathogen attribute knowledge fields of the key pathogen sample data range can be obtained in the following manner: performing knowledge field selection on the pathogen sample data to be analyzed to obtain all pathogen sample pathogen attribute knowledge fields of the pathogen sample data to be analyzed (here referred to as the second pathogen sample pathogen attribute knowledge fields); estimating the target pathogen sample data range associated with the pathogen sample data processing instructions in the pathogen sample data to be analyzed based on all pathogen sample pathogen attribute knowledge fields; performing knowledge field selection on the target pathogen sample data range to obtain the key pathogen sample pathogen attribute knowledge fields, and determining the key pathogen sample pathogen attribute knowledge fields as a pathogen attribute knowledge field in one state.

[0038] Among them, the second pathogen sample pathogen attribute knowledge field and the first pathogen sample pathogen attribute knowledge field are all pathogen sample pathogen attribute knowledge fields of the pathogen sample data to be analyzed, and the second pathogen sample pathogen attribute knowledge field and the first pathogen sample pathogen attribute knowledge field can be the same knowledge field or different. In other words, the pathogen sample data to be analyzed can be input into the configured pathogen sample data selection network to obtain all pathogen sample pathogen attribute knowledge fields of the pathogen sample data. This knowledge field can be determined as the first pathogen sample pathogen attribute knowledge field, and can also be determined as the second pathogen sample pathogen attribute knowledge field. It is also possible to select all knowledge fields of the pathogen sample data to be analyzed through two different pathogen sample pathogen attribute knowledge field selection networks, and obtain the first pathogen sample pathogen attribute knowledge field and the second pathogen sample pathogen attribute knowledge field, respectively.

[0039] In the embodiment of the present application, the target pathogen sample data range refers to the range in the pathogen sample data to be analyzed that is highly correlated with the pathogen sample data to be analyzed instruction, that is, the pathogen sample data range that plays an important role in determining the pathogen detection result of the pathogen sample data to be analyzed. Among them, the detection of the target pathogen sample data range and the selection of knowledge fields can be implemented using a configured decision network. Optionally, during configuration, the pathogen sample data example can be annotated with the target pathogen sample data range in the pathogen sample data example (that is, the key pathogen sample data range of the pathogen sample data example is known), and the decision network can be configured based on a large number of pathogen sample data examples, so that the decision network can adaptively learn the ability to detect the key pathogen sample data range in the pathogen sample data based on the target pathogen sample data range annotated in the pathogen sample data example. After obtaining the configured network, the target pathogen sample data range in the pathogen sample data to be analyzed can be detected by the network, so that the knowledge field can be selected for the range to obtain the key pathogen sample pathogen attribute knowledge field.

[0040] Optionally, when detecting the target pathogen sample data range in the pathogen sample data to be analyzed, the above-mentioned all pathogen sample pathogen attribute knowledge fields include several knowledge field relationship networks, and the above-mentioned combination of the above-mentioned all pathogen sample pathogen attribute knowledge fields to estimate the target pathogen sample data range associated with the pathogen sample data processing instruction to be analyzed can include: splicing the knowledge field values ​​of the knowledge field points with the same positioning in several knowledge field relationship networks to obtain a spliced ​​knowledge field relationship network; based on the spliced ​​knowledge field relationship network, determining the influence of each knowledge field point in the spliced ​​knowledge field relationship network, the influence of each knowledge field point represents the degree of association between the pathogen sample data range corresponding to the knowledge field point and the pathogen sample data processing instruction; according to the influence of each knowledge field point in the spliced ​​knowledge field relationship network, determining the target pathogen sample data range in the pathogen sample data to be analyzed.

[0041] Each knowledge field point in the knowledge field relationship network corresponds to a range in the pathogen sample data to be analyzed (the specific size of the range is determined by the receptive field of the knowledge field relationship network). Correspondingly, the knowledge field value of each knowledge field point reflects the semantic information of the pathogen sample data range corresponding to the knowledge field point. Therefore, after obtaining a spliced ​​knowledge field relationship network that splices knowledge field relationship networks of several dimensions, the importance of each knowledge field point in the knowledge field relationship network to the pathogen sample data processing instructions, that is, the above-mentioned influence, can be estimated based on the spliced ​​knowledge field relationship network. That is, the importance of the pathogen sample data range corresponding to each knowledge field point in the current pathogen sample data processing requirements, and then, according to the influence corresponding to each knowledge field point, the key range in the spliced ​​knowledge field relationship network can be determined, and the pathogen sample data range corresponding to the key range in the pathogen sample data to be analyzed is the key pathogen sample data range in the pathogen sample data to be analyzed.

[0042] Since the pathogen attribute knowledge field of key pathogen samples is information on the pathogen sample data range that is highly correlated with the pathogen sample data processing instructions, determining the pathogen attribute knowledge field of key pathogen samples as a pathogen attribute knowledge field in a state and integrating it into the processing of the pathogen sample data to be analyzed can further effectively improve the pathogen sample data processing effect.

[0043] In actual implementation, optionally, the above-mentioned pathogen attribute knowledge fields of at least two states include the first pathogen attribute knowledge field of the pathogen sample data to be analyzed, the pathogen event knowledge field, at least one partial pathogen sample pathogen attribute knowledge field and the key pathogen sample pathogen attribute knowledge field.

[0044] Step S130: For each state of the pathogen attribute knowledge field, determine the influence of the pathogen attribute knowledge field of the state based on the pathogen attribute knowledge field of the state. The influence of the pathogen attribute knowledge field of each state represents the credibility coefficient of the pathogen attribute knowledge field of the state for the pathogen sample data processing instructions.

[0045] Among them, the pathogen sample data processing instruction refers to the purpose of processing the pathogen sample data, that is, what kind of processing is to be performed on the pathogen sample data. The specific instructions of the pathogen sample data processing instruction are determined by the actual application requirements and are configured according to the requirements. The embodiment of the present application does not limit it. In theory, the pathogen sample data processing instruction in the embodiment of the present application can be any processing requirement based on the pathogen sample data. Optionally, the pathogen sample data processing instruction may include but is not limited to pathogen sample data classification, pathogen sample data segmentation, pathogen sample data detection or pathogen sample data generation (for example, generating pathogen sample data of other styles based on the pathogen sample data to be analyzed, such as cartoon pathogen sample data) and other instructions. Among them, when the knowledge field of the pathogen sample data to be analyzed is selected, which of the above-mentioned at least two states specifically includes which states can be configured according to the specific pathogen sample data processing instruction.

[0046] In practical applications, for a pathogen sample data to be analyzed, the information from different sources of the pathogen sample data usually plays different roles in the pathogen sample data processing process. For example, the pathogen sample data processing instruction is to classify the pathogen sample data. To determine whether there are specific factors in the pathogen sample data, for the identification result of the pathogen sample data, the continuous record data corresponding to the pathogen sample data where the matters related to the factor in the pathogen sample data to be analyzed are located or the pathogen event information display related to the factor is more important. The pathogen attribute knowledge field of the state related to the factor should have a relatively greater influence. In order to improve the accuracy of pathogen sample data processing, after obtaining the pathogen attribute knowledge fields of various states, the importance, that is, the influence, of the pathogen attribute knowledge fields of each state can be determined. When determining the pathogen detection results of the pathogen sample data to be analyzed corresponding to the pathogen sample data processing instructions based on the pathogen attribute knowledge fields of each state, the pathogen attribute knowledge fields of different states can be controlled to play corresponding roles according to the influence of the pathogen attribute knowledge fields of each state, such as weakening the role of the pathogen attribute knowledge fields of the states that are weakly associated with the pathogen sample data processing instructions, and enhancing the role of the pathogen attribute knowledge fields of the states that are highly associated with the pathogen sample data processing instructions.

[0047] Among them, the specific determination method of the influence of the pathogen attribute knowledge field of each state is not uniquely limited in the embodiment of the present application. For example, the influence corresponding to each state can be estimated through a configured decision network. Optionally, for the pathogen attribute knowledge field of each state, the influence of the pathogen attribute knowledge field of the state can be determined in the following way: if the pathogen attribute knowledge field of the state is the pathogen attribute knowledge field of the pathogen sample, based on the various knowledge field relationship networks in the pathogen attribute knowledge field of the pathogen sample, the influence of each knowledge field relationship network is determined, and based on each knowledge field relationship network and the influence of each knowledge field relationship network, the influence of the pathogen attribute knowledge field of the pathogen sample is determined; if the pathogen attribute knowledge field of the state is the pathogen event knowledge field, the influence of the pathogen event knowledge field is determined according to the knowledge field value of each knowledge field point in the pathogen event knowledge field.

[0048] Among them, for the pathogen sample pathogen attribute knowledge field in the pathogen sample data state (that is, the information source of the knowledge field is the pathogen sample data), the pathogen sample pathogen attribute knowledge field in each state includes multiple knowledge field relationship networks, that is, several knowledge field relationship networks of identification trajectories, that is, several knowledge fields of different dimensions. Since the functions of knowledge field relationship networks of different identification trajectories are usually different, when determining the influence of the pathogen attribute knowledge field in the pathogen sample data state, the importance, that is, the influence, of each knowledge field relationship network in the state can be determined first, and the influence of each knowledge field relationship network can be used to weight the knowledge field relationship network. After that, the importance of the pathogen sample pathogen attribute knowledge field in the state for the pathogen sample data processing instructions can be determined based on the weighted knowledge field relationship networks.

[0049] Optionally, for the pathogen attribute knowledge field of the pathogen sample data state, the influence of each knowledge field relationship network in the pathogen attribute knowledge field can be determined in the following way: for each knowledge field relationship network in the pathogen attribute knowledge field of the pathogen sample, determine the mean of the knowledge field values ​​of each knowledge field point in the knowledge field relationship network, and determine the mean as the influence of the knowledge field relationship network.

[0050] For the pathogen sample pathogen attribute knowledge field in the pathogen sample data state, after obtaining the weighted knowledge field relationship network corresponding to the pathogen sample pathogen attribute knowledge field, the mean of the knowledge field values ​​of all knowledge field points in the weighted knowledge field relationship network can be calculated, and the mean can be determined as the influence of the pathogen sample pathogen attribute knowledge field.

[0051] Optionally, for the pathogen attribute knowledge field of the pathogen event status, since the pathogen event information is different from the pathogen sample data and does not have a spatial knowledge field, the average of the knowledge field values ​​of each knowledge field point in the pathogen event knowledge field (the knowledge field representation of the pathogen event) can be determined as the influence of the pathogen event knowledge field.

[0052] Step S140: Determine the target pathogen knowledge field for each state according to the pathogen attribute knowledge field of each state and the corresponding influence.

[0053] Step S150: Based on the target pathogen knowledge field of each state, determine the pathogen detection result of the pathogen sample data to be analyzed corresponding to the pathogen sample data processing instruction.

[0054] For the pathogen attribute knowledge field of each state, after determining the influence corresponding to each state, the pathogen attribute knowledge field of the state can be updated according to the influence corresponding to the state to obtain its target pathogen knowledge field, so that the pathogen attribute knowledge field of each state can produce a corresponding effect according to its importance to the pathogen sample data processing instruction. Determined as an optional solution, for each state, the pathogen attribute knowledge field of the state can be weighted by the influence of the state to obtain the target pathogen knowledge field of the state.

[0055] As another optional scheme, the target pathogen knowledge field of each state can be determined in the following manner: for the pathogen attribute knowledge field of each state, determine the matching result between the pathogen attribute knowledge field of the state and the pathogen attribute knowledge field of each state in the pathogen attribute knowledge fields of at least two states; for the pathogen attribute knowledge field of each state, based on the matching result between the pathogen attribute knowledge field of the state and the pathogen attribute knowledge field of each state in the pathogen attribute knowledge fields of at least two states, perform weighted splicing of the pathogen attribute knowledge fields of each state in the pathogen attribute knowledge fields of at least two states to obtain the spliced ​​knowledge field of the state; use the influence corresponding to each state to weight the spliced ​​knowledge field of the corresponding state to obtain the target pathogen knowledge field of each state.

[0056] In practical applications, since the pathogen attribute knowledge fields of various states of the pathogen sample data to be analyzed are obtained based on the information of different dimensions of the pathogen sample data to be analyzed, there is a correlation between the information of each dimension. For example, the pathogen event corresponding to the pathogen sample data to be analyzed is usually a text describing the pathogen sample data information of the pathogen sample data to be analyzed. There is a connection between the pathogen event and the pathogen sample data. Similarly, the pathogen attribute knowledge fields of all pathogen samples of the pathogen sample data to be analyzed and the pathogen attribute knowledge fields of some pathogen samples in the scope of the target items in the pathogen sample data are also related. Therefore, in order to improve the expression ability of the pathogen attribute knowledge fields of each state and improve the processing effect of the pathogen sample data, after obtaining the pathogen attribute knowledge fields of each state of the pathogen sample data to be analyzed, the knowledge fields can be spliced ​​according to the matching results between the pathogen attribute knowledge fields of each state to obtain the spliced ​​knowledge fields corresponding to each state, that is, the updated knowledge fields mentioned above. The updated knowledge fields corresponding to each state not only contain the information of the state, but also incorporate the relevant information of other states.

[0057] Optionally, when calculating the matching results between the pathogen attribute knowledge fields of two states, the correlation between the pathogen attribute knowledge fields of the two states can be directly calculated, or the pathogen attribute knowledge fields of each state can be re-selected to obtain the re-selected knowledge fields corresponding to each state, and the matching results between the two states can be calculated based on the re-selected knowledge fields corresponding to the two states.

[0058] For the pathogen attribute knowledge field of each state, when calculating the matching result between the pathogen attribute knowledge field of the state and the pathogen attribute knowledge field of each state, the matching result between the state itself may be included or not. For example, the matching result between the pathogen attribute knowledge field of each state and the pathogen attribute knowledge field of the state may be set to a set value, and only the matching result between the pathogen attribute knowledge field of the state and the pathogen attribute knowledge fields of other states except the state is calculated. Alternatively, the matching result between the pathogen attribute knowledge field of each state and the pathogen attribute knowledge field of the state may be calculated. For example, for each state, different knowledge field selection modules may be used to further select knowledge fields for the pathogen attribute knowledge field of the state to obtain two different re-selected knowledge fields corresponding to the pathogen attribute knowledge field of the state. By calculating the matching result between the two re-selected knowledge fields, the matching result between the pathogen attribute knowledge fields of the state itself is obtained.

[0059] After obtaining the updated knowledge fields corresponding to each state, the influence corresponding to each state can be used to weight the updated knowledge fields of the corresponding state to obtain the target pathogen knowledge field of each state. The target pathogen knowledge field of each state not only considers the information of the state, but also considers the semantic association between the information of the state and the information of other states, and also considers the credibility coefficient of the information of the state for the pathogen sample data processing instructions. Therefore, based on the target pathogen knowledge field of each state, more accurate pathogen detection results of pathogen sample data can be obtained.

[0060] In order to better improve the performance of the network, the embodiment of the present application also provides a method for configuring a pathogen sample data processing network, which may include: Step S210: obtaining a set of configured pathogen examples, the set of configured pathogen examples including a plurality of pathogen sample data examples with a pathogen theme, the pathogen theme of each pathogen sample data example representing a real-time pathogen detection result of the pathogen sample data example corresponding to the pathogen sample data instruction; Step S220: Repeat the configuration operation of step S221 to step S225 on the decision network based on the configured pathogen example set until the configuration termination requirement is met, and determine the decision network that meets the configuration termination requirement as the pathogen sample data processing network, wherein the configuration operation includes the following steps: Step S221: for each pathogen sample data example, select knowledge field examples of at least two states corresponding to the pathogen sample data example, and determine the influence of the knowledge field example of each state according to the knowledge field example of the state; Step S222: for each pathogen sample data example, based on the knowledge field example of each state of the pathogen sample data example and the corresponding influence, determine the target knowledge field example of each state of the pathogen sample data example; Step S223: for each pathogen sample data example, based on the target knowledge field examples of each state of the pathogen sample data example, estimating the first pathogen detection result of the pathogen sample data example corresponding to the pathogen sample data processing instruction; Step S224: determining a first configuration quality index based on the pathogen theme of each pathogen sample data example and the influence corresponding to each state, determining a second configuration quality index based on the pathogen theme of each pathogen sample data example and the first pathogen detection result, and obtaining a configuration total quality index according to the first configuration quality index and the second configuration quality index; Step S225: If the configuration termination requirement is not met, the network coefficients of the decision network are debugged based on the configuration total quality index.

[0061] Among them, the above-mentioned configuration termination requirements can be configured according to actual application needs, and may include but are not limited to the configuration times reaching a set number or the quality indicator function corresponding to the network converging, such as the total configuration quality indicator of the network is less than the set value or the total configuration quality indicators of consecutive times are all less than the set value, etc.

[0062] The configuration method provided in the embodiment of the present application, the total configuration quality index of the network includes two parts, one part is the difference between the pathogen detection results of each pathogen sample data example obtained through the network (i.e., the first processing) and the real-time pathogen detection results of the pathogen sample data example (i.e., the pathogen topic) (the second configuration quality index), and the other part is the difference between the influence of the pathogen attribute knowledge field of each state of each pathogen sample data example estimated by the network and the pathogen topic of the pathogen sample data example (the first configuration quality index). Among them, for each pathogen sample data example, the influence pathogen topic of the pathogen attribute knowledge field of each state of the pathogen sample data example can be determined according to the pathogen topic of the pathogen sample data example, that is, the real-time influence, and the second configuration quality index can be calculated according to the difference between the real-time influence of each state and the influence estimated by the network. If the total configuration quality index of the network does not meet the preset requirements or the number of configurations does not reach the set number, the network coefficient can be debugged, and the debugged network can be repeatedly configured.

[0063] In the configuration scheme provided by the present application, the configuration of the network is constrained by the above two parts of quality indicators, so that the configured network can objectively estimate the influence corresponding to the pathogen attribute knowledge field of each state of the pathogen sample data to be analyzed, and can correct the knowledge field of each state according to the influence, so as to more accurately estimate the pathogen detection results of the pathogen sample data to be analyzed.

[0064] Optionally, the decision network includes a first artificial intelligence network and a second artificial intelligence network, wherein the first artificial intelligence network is pre-configured; for each pathogen sample data example, selecting knowledge field examples of at least two states corresponding to the pathogen sample data example, including: selecting, through the first artificial intelligence network, a first pathogen event knowledge field, a sample pathogen attribute knowledge field, and a portion of pathogen sample pathogen attribute knowledge fields within the pathogen sample data range where each target item in at least one target item in the pathogen sample data example is located; performing knowledge field switching (knowledge field mapping) on ​​the first pathogen event knowledge field through the second artificial intelligence network to obtain a second pathogen event knowledge field, and selecting, through the second artificial intelligence network, a first pathogen event knowledge field, a sample pathogen attribute knowledge field, and a partial pathogen sample pathogen attribute knowledge field within the pathogen sample data range where each target item in the pathogen sample data example is located. The intelligent network identifies the example target range associated with the pathogen sample data processing instruction in the pathogen sample data example (that is, the key pathogen sample data range in the pathogen sample data example), and selects important knowledge field examples of the example target range; the second pathogen event knowledge field, the sample pathogen attribute knowledge field, the partial pathogen sample pathogen attribute knowledge fields and the important knowledge field examples in the pathogen sample data range where each target item is located are respectively determined as knowledge field examples of a state; wherein the influence of the knowledge field example of each state and the first pathogen detection result of the pathogen sample data example are also determined by the second artificial intelligence network; the above-mentioned debugging of the network coefficient of the decision network includes: debugging the network coefficient of the second artificial intelligence network.

[0065] That is to say, it is determined as an optional solution in actual implementation. The first artificial intelligence network in the decision network can be a pre-configured general network. In the above configuration operation, only the network coefficient of the second artificial intelligence network can be debugged. The adoption of this solution can speed up the configuration effect of the network. In addition, during the configuration process, for the pathogen event knowledge field of the pathogen sample data example, the first pathogen event knowledge field of the pathogen event can be first selected through the pre-configured first artificial intelligence network, and then the first pathogen event knowledge field can be selected through the second artificial intelligence network to be configured to obtain the second pathogen event knowledge field. Since the network coefficient of the second artificial intelligence network is continuously optimized and debugged during the configuration process, the first pathogen event knowledge field is switched to the second pathogen event knowledge field through the second artificial intelligence network, so that the configured decision network can obtain the pathogen event knowledge field that is more consistent with the pathogen sample data instruction processing.

[0066] In the implementation, the knowledge fields of the pathogen sample data examples or the multiple states of the pathogen sample data to be analyzed may include at least two of the knowledge fields of the various states described above. For example, as a choice, the knowledge field examples of the pathogen sample data examples may include the second pathogen event knowledge field, the sample pathogen attribute knowledge field, the pathogen sample pathogen attribute knowledge field of the pathogen sample data range where each target item is located, and at least two of the important knowledge field examples. Correspondingly, the structure of the decision network may also be debugged accordingly. For example, when the knowledge field examples do not include the pathogen attribute knowledge field of some pathogen samples, the decision network may not include the network structure for item identification and selection of the pathogen attribute knowledge field of some pathogen samples in the partial range where the item is located. Optionally, the first artificial intelligence network may include the pathogen event knowledge field selection module, the pathogen sample pathogen attribute knowledge field selection module, and the item knowledge field selection module. Optionally, the second artificial intelligence network may include the pathogen event knowledge field mapping module, the key information detection module, the state discrimination module, the multi-state splicing module, and the result estimation module.

[0067] Determined as an optional implementation, the second pathogen event knowledge field of each pathogen sample data example can be obtained in the following manner: the first pathogen event knowledge field corresponding to the pathogen sample data example is selected by a pathogen event knowledge field selection module, and the first pathogen event knowledge field is selected by a pathogen event knowledge field mapping module to obtain the second pathogen event knowledge field.

[0068] Among them, the pathogen event knowledge field selection module may include a pathogen event recognition network and a pathogen event knowledge field selection network. The pathogen event knowledge field selection network may include one or several fully connected layers, or other network structures with knowledge field mapping functions.

[0069] The sample pathogen attribute knowledge fields of the pathogen sample data example may be obtained by selecting all knowledge fields of the pathogen sample data example through the above-mentioned pathogen sample pathogen attribute knowledge field selection module.

[0070] For some graphic knowledge fields of the pathogen sample data example, the item knowledge field selection module can be used to identify and select knowledge fields for the pathogen sample data range where at least one target item in the pathogen sample data example is located, and obtain some pathogen sample pathogen attribute knowledge fields for the pathogen sample data range where each target item is located. Optionally, the item knowledge field selection module can include an item recognition network and a pathogen sample pathogen attribute knowledge field selection network. The item recognition network can detect the location of the target item in the pathogen sample data example, and the pathogen sample pathogen attribute knowledge field selection network can perform knowledge field selection on some pathogen sample data ranges with corresponding locations, and obtain some pathogen sample pathogen attribute knowledge fields.

[0071] Among them, for each pathogen sample data example, after obtaining the knowledge field examples of each state corresponding to the pathogen sample data example, the state discrimination module can be used to determine the influence of the knowledge field examples of each state of the pathogen sample data example. The multi-state splicing module can determine the target knowledge field examples of each state of the pathogen sample data example based on the knowledge field examples of each state of the pathogen sample data example and the corresponding influence. The result estimation module is used to estimate the first pathogen detection result of the pathogen sample data example based on the target knowledge field examples of each state of the pathogen sample data example.

[0072] Correspondingly, the above-mentioned debugging of the network coefficients of the decision network based on the configuration total quality index refers to: debugging the network coefficients of the pathogen event knowledge field mapping module, the key information detection module, the state discrimination module, the multi-state splicing module and the result estimation module based on the configuration total quality index.

[0073] The above configuration method provided in the embodiment of the present application can speed up the configuration speed of the network, and the pathogen event knowledge field mapping module can switch the same first pathogen event knowledge field selected by the pre-configured pathogen event knowledge field selection module to a second pathogen event knowledge field that is more suitable for specific pathogen sample data processing instructions through configuration learning. Optionally, when the pathogen event knowledge field selection module performs knowledge field selection on the pathogen sample data example, it can first perform pathogen event identification on the pathogen sample data example, perform all knowledge fields on the identified pathogen event, and perform partial knowledge field selection on each pathogen event unit in the pathogen event, and splice the selected partial knowledge fields and all knowledge fields to obtain the first pathogen event knowledge field. After that, the first pathogen event knowledge field can be switched to the second pathogen event knowledge field through the pathogen event knowledge field mapping module (such as one or several cascaded fully connected layers).

[0074] Optionally, the knowledge field examples of at least two states of the pathogen sample data example include important knowledge field examples. In order to further improve the accuracy of the decision network in detecting the range of key pathogen sample data in the pathogen sample data (i.e., the performance of the key information detection module obtained by configuration), in an optional embodiment of the present application, the above-mentioned configuration operation may also include: for each pathogen sample data example, based on all pathogen sample pathogen attribute knowledge fields corresponding to the important knowledge field examples of the pathogen sample data example (i.e., all pathogen sample pathogen attribute knowledge fields selected by the decision network to determine the example target range), estimating the second pathogen detection result corresponding to the pathogen sample data example; determining the third configuration quality indicator based on the pathogen theme and the second pathogen detection result corresponding to each pathogen sample data example; accordingly, obtaining the configuration total quality indicator based on the first configuration quality indicator and the second configuration quality indicator includes: obtaining the configuration total quality indicator based on the first configuration quality indicator, the second configuration quality indicator and the third configuration quality indicator.

[0075] On the basis of the above, a rapid detection device for pathogens is provided, the device comprising: A data acquisition module, used to obtain pathogen sample data to be analyzed; An attribute selection module, used to select pathogen attribute knowledge fields of at least two states corresponding to the pathogen sample data to be analyzed in combination with the pathogen sample data to be analyzed; A coefficient determination module is used to determine the influence of the pathogen attribute knowledge field in each state according to the pathogen attribute knowledge field in the state, and the influence of the pathogen attribute knowledge field in each state represents the credibility coefficient of the pathogen attribute knowledge field in the state for the pathogen sample data processing instruction; A knowledge field determination module is used to determine the target pathogen knowledge field of each state according to the pathogen attribute knowledge field of each state and the corresponding influence; The result detection module is used to determine the pathogen detection result of the pathogen sample data to be analyzed corresponding to the pathogen sample data processing instruction based on the target pathogen knowledge field of each state.

[0076] Based on the above, a rapid detection system for pathogens is shown, including a processor and a memory that communicate with each other, and the processor is used to read a computer program from the memory and execute it to implement the above method.

[0077] Based on the above, a computer-readable storage medium is also provided, on which a computer program stored implements the above method when running.

[0078] In summary, based on the above scheme, when processing the pathogen sample data to be analyzed, in order to obtain a better pathogen sample data processing effect, the pathogen attribute knowledge fields of the pathogen sample data to be analyzed will be selected in multiple states. On this basis, the information of each state will be further evaluated according to the pathogen attribute knowledge fields of each state. The credibility coefficient, that is, the influence of the pathogen sample data processing instructions, and the target pathogen knowledge fields of each state that are finally used to determine the pathogen detection results of the pathogen sample data are obtained according to the pathogen attribute knowledge fields of each state and the corresponding credibility coefficients. The method provided in the embodiment of the present application estimates the credibility coefficient of the information of each state, so that the pathogen can be quickly detected, the type of pathogen can be accurately obtained, and medical workers can provide timely treatment to patients according to the type of pathogen, which can ensure the life safety of patients.

[0079] It should be understood that the system and its modules shown above can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or a dedicated design hardware. Those skilled in the art will understand that the above methods and systems can be implemented using computer executable instructions and / or included in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of the present application can not only be implemented by hardware circuits such as ultra-large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but can also be implemented by software such as executed by various types of processors, and can also be implemented by a combination of the above hardware circuits and software (e.g., firmware).

[0080] It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects that may be produced may be any one or a combination of the above, or any other beneficial effects that may be obtained.

Claims

1. A method for rapid detection of pathogens, characterized in that: The method comprises: Obtaining pathogen sample data to be analyzed; In combination with the pathogen sample data to be analyzed, selecting pathogen attribute knowledge fields of at least two states corresponding to the pathogen sample data to be analyzed; For each state of the pathogen attribute knowledge field, the influence of the pathogen attribute knowledge field of the state is determined according to the pathogen attribute knowledge field of the state, and the influence of the pathogen attribute knowledge field of each state represents the credibility coefficient of the pathogen attribute knowledge field of the state for the pathogen sample data processing instruction; Determine the target pathogen knowledge field for each state based on the pathogen attribute knowledge field of each state and the corresponding influence; Based on the target pathogen knowledge field of each state, the pathogen detection result of the pathogen sample data to be analyzed corresponding to the pathogen sample data processing instruction is determined.

2. The method according to claim 1, characterized in that The target pathogen knowledge field of each state is determined according to the pathogen attribute knowledge field of each state and the corresponding influence, including: For each state of the pathogen attribute knowledge field, determining a matching result between the pathogen attribute knowledge field of the state and each of the pathogen attribute knowledge fields of the at least two states; For each state of the pathogen attribute knowledge field, based on the matching result between the pathogen attribute knowledge field of the state and the pathogen attribute knowledge field of each state in the pathogen attribute knowledge fields of at least two states, weighted concatenation is performed on the pathogen attribute knowledge fields of each state in the pathogen attribute knowledge fields of at least two states to obtain the concatenated knowledge field of the state; The influence corresponding to each state is used to weight the concatenated knowledge fields of the corresponding states to obtain the target pathogen knowledge field of each state.

3. The method according to claim 1, characterized in that The selecting of pathogen attribute knowledge fields of at least two states corresponding to the pathogen sample data to be analyzed in combination with the pathogen sample data to be analyzed includes: A knowledge field is selected for at least one of the following information corresponding to the pathogen sample data to be analyzed to obtain an information knowledge field for each information, and the information knowledge field for each information is determined as a pathogen attribute knowledge field of a state: The pathogen sample data to be analyzed; The pathogen event corresponding to the pathogen sample data to be analyzed; The continuous recorded data corresponding to each pathogen sample data, the continuous recorded data corresponding to each pathogen sample data is the pathogen sample data range where a target item in the pathogen sample data to be analyzed is located.

4. The method according to claim 3, characterized in that The selecting of pathogen attribute knowledge fields of at least two states corresponding to the pathogen sample data to be analyzed in combination with the pathogen sample data to be analyzed includes: Selecting knowledge fields for the pathogen sample data to be analyzed to obtain all pathogen sample pathogen attribute knowledge fields for the pathogen sample data to be analyzed; In combination with the pathogen attribute knowledge fields of all pathogen samples, a target pathogen sample data range associated with the pathogen sample data processing instruction in the pathogen sample data to be analyzed is estimated; Performing knowledge field selection on the target pathogen sample data range to obtain a key pathogen sample pathogen attribute knowledge field, and determining the key pathogen sample pathogen attribute knowledge field as a pathogen attribute knowledge field in a state; The pathogen attribute knowledge fields of all pathogen samples include a plurality of knowledge field relationship networks, and the estimation of the target pathogen sample data range associated with the pathogen sample data to be analyzed and the pathogen sample data processing instruction in combination with the pathogen attribute knowledge fields of all pathogen samples includes: The knowledge field values ​​of the knowledge field points at the same position in the plurality of knowledge field relationship networks are spliced ​​to obtain a spliced ​​knowledge field relationship network; Determine the influence of each knowledge field point in the spliced ​​knowledge field relationship network in combination with the spliced ​​knowledge field relationship network, wherein the influence of each knowledge field point represents the degree of association between the pathogen sample data range corresponding to the knowledge field point and the pathogen sample data processing instruction; According to the influence of each knowledge field point in the spliced ​​knowledge field relationship network, the target pathogen sample data range in the pathogen sample data to be analyzed is determined.

5. The method according to claim 3, characterized in that: The at least one type of information includes the pathogen event, and knowledge fields are selected for various types of information in the at least one type of information to obtain information knowledge fields for each type of information, including: Selecting knowledge fields for each pathogen event unit in the pathogen event to obtain partial knowledge fields of each pathogen event unit; Select all knowledge fields for the pathogen event to obtain all knowledge fields of the pathogen event; splicing all knowledge fields of the pathogen event and part of the knowledge fields of each pathogen event unit to obtain the pathogen event knowledge field of the pathogen event, and determining the pathogen event knowledge field as a pathogen attribute knowledge field of a state; The pathogen event knowledge field of the pathogen event is obtained by splicing all the knowledge fields of the pathogen event and part of the knowledge fields of each pathogen event unit, including: splicing all knowledge fields of the pathogen event and part of the knowledge fields of each pathogen event unit to obtain a spliced ​​knowledge field; By performing knowledge field selection again on the spliced ​​knowledge field, the spliced ​​knowledge field is switched to the knowledge field coordinate system corresponding to the pathogen sample data processing instruction, and the re-selected knowledge field is determined as the pathogen event knowledge field of the pathogen event.

6. The method according to claim 1, characterized in that For each state of the pathogen attribute knowledge field, determining the influence of the pathogen attribute knowledge field in the state according to the pathogen attribute knowledge field in the state includes: If the pathogen attribute knowledge field of the state is the pathogen attribute knowledge field of the pathogen sample, based on the various knowledge field relationship networks in the pathogen attribute knowledge field of the pathogen sample, determine the influence of each knowledge field relationship network; using the influence of each knowledge field relationship network in the pathogen attribute knowledge field of the pathogen sample, weight the corresponding knowledge field relationship network, and determine the influence of the pathogen attribute knowledge field of the pathogen sample based on the weighted knowledge field relationship networks; If the pathogen attribute knowledge field of the state is a pathogen event knowledge field, the influence of the pathogen event knowledge field is determined according to the knowledge field value of each knowledge field point in the pathogen event knowledge field; The determining of the influence of each knowledge field relationship network based on each knowledge field relationship network in the pathogen attribute knowledge field of the pathogen sample includes: For each knowledge field relationship network, the average value of the knowledge field values ​​of each knowledge field point in the knowledge field relationship network is determined as the influence of the knowledge field relationship network; The determining of the influence of the pathogen attribute knowledge field of the pathogen sample based on the weighted knowledge field relationship network includes: determining the average of the knowledge field values ​​of all knowledge field points in the weighted knowledge field relationship network as the influence of the pathogen attribute knowledge field of the pathogen sample.

7. The method according to claim 1, characterized in that The decision network includes a first artificial intelligence network and a second artificial intelligence network, wherein the first artificial intelligence network is pre-configured; for each pathogen sample data example, selecting knowledge field examples of at least two states corresponding to the pathogen sample data example, including: The first artificial intelligence network selects the first pathogen event knowledge field, the sample pathogen attribute knowledge field, and the pathogen attribute knowledge field of some pathogen samples corresponding to the pathogen sample data example, which are within the pathogen sample data range where each target item in at least one target item is located, wherein the first pathogen event knowledge field represents the descriptive features of all pathogens in the first pathogen event knowledge segment, and can represent the corresponding pathogen; Performing knowledge field switching on the first pathogen event knowledge field through the second artificial intelligence network to obtain a second pathogen event knowledge field, and identifying an example target range associated with a pathogen sample data processing instruction in the pathogen sample data example through the second artificial intelligence network, and selecting an important knowledge field example of the example target range; The second pathogen event knowledge field, the sample pathogen attribute knowledge field, the pathogen attribute knowledge field of some pathogen samples in the pathogen sample data range of each target item, and the important knowledge field examples are respectively determined as knowledge field examples of a state; wherein the influence of the knowledge field example of each state and the first pathogen detection result of the pathogen sample data example are determined by a second artificial intelligence network; Debugging the network coefficients of the decision network includes: debugging the network coefficients of the second artificial intelligence network.

8. The method according to claim 1, characterized in that For each of the pathogen sample data examples, knowledge field examples of at least two states corresponding to the pathogen sample data example are selected, including: Select all pathogen attribute knowledge fields of the pathogen sample data example; According to all pathogen sample pathogen attribute knowledge fields of the pathogen sample data example, the example target range in the pathogen sample data example is determined, and important knowledge field examples of the example target range are selected, and the important knowledge field examples are determined as knowledge field examples of a state.

9. The method according to claim 1, characterized in that: The method is implemented by calling a configured pathogen sample data processing network, and the pathogen sample data processing network is configured in the following manner: Obtaining a set of configured pathogen examples, the set of configured pathogen examples comprising a plurality of pathogen sample data examples with a pathogen subject, the pathogen subject of each of the pathogen sample data examples representing a real-time pathogen detection result of the pathogen sample data example corresponding to the pathogen sample data instruction; Repeat the configuration operation on the decision network in combination with the configured pathogen example set until the configuration termination requirement is met, and determine the decision network that meets the configuration termination requirement as the configured pathogen sample data processing network, wherein the configuration operation includes the following steps: For each of the pathogen sample data examples, select knowledge field examples of at least two states corresponding to the pathogen sample data example; For each of the pathogen sample data examples, determining the influence of the knowledge field example of each state of the pathogen sample data example according to the knowledge field example of the state of the pathogen sample data example; For each of the pathogen sample data examples, based on the knowledge field example of each state of the pathogen sample data example and the corresponding influence, determine a target knowledge field example of each state of the pathogen sample data example; For each of the pathogen sample data examples, based on the target knowledge field examples of each state of the pathogen sample data example, estimating a first pathogen detection result of the pathogen sample data example corresponding to the pathogen sample data processing instruction; Determine a first configuration quality indicator based on the pathogen theme of each of the pathogen sample data examples and the influence corresponding to each state, and determine a second configuration quality indicator based on the pathogen theme of each of the pathogen sample data examples and the first pathogen detection result; For each of the pathogen sample data examples, based on all pathogen sample pathogen attribute knowledge fields corresponding to the important knowledge field examples of the pathogen sample data example, estimate the second pathogen detection result corresponding to the pathogen sample data example; Determining a third configuration quality indicator according to the pathogen subject corresponding to each of the pathogen sample data examples and the second pathogen detection result; Obtaining a configuration total quality indicator according to the first configuration quality indicator, the second configuration quality indicator and the third configuration quality indicator; If the configuration termination requirement is not met, the network coefficients of the decision network are debugged in combination with the configuration total quality index.

10. A rapid detection system for pathogens, characterized in that: The invention comprises a processor and a memory communicating with each other, wherein the processor is used to read a computer program from the memory and execute the computer program to implement the method according to any one of claims 1 to 9.

Images