Nucleic acid detection sample defect analysis method and system applied to nucleic acid detector

Abstract

The embodiment of the present application provides a nucleic acid detection sample defect analysis method and system applied to a nucleic acid detector, which can utilize a nucleic acid detection sample defect prediction model to determine the nucleic acid detection sample defect information of a target nucleic acid detection sample image based on the image abnormal point features and image area features of the target nucleic acid detection sample image, thereby improving the correlation between different image areas during the nucleic acid detection sample defect analysis process, and thus more accurately understanding the image features of the target nucleic acid detection sample image, thereby improving the reliability of the nucleic acid detection sample defect analysis. In addition, by performing model input differentiation of nucleic acid detection labels, the reliability of the nucleic acid detection sample defect analysis can be further improved.

Description

Technical Field

[0001] The present application relates to the technical field of nucleic acid detection, and more specifically, to a method and system for analyzing nucleic acid detection sample defects applied to a nucleic acid detector. Background Art

[0002] The nucleic acid detector searches for the presence of nucleic acid of foreign invading viruses in human respiratory specimens, blood or feces to determine whether the patient is infected with the new coronavirus. Therefore, once the nucleic acid test is positive, it can be proved that the virus exists in the patient's body. Before using the nucleic acid detector for nucleic acid testing, for some samples with high importance and long testing cycles, in order to ensure the reliability of subsequent nucleic acid test results, it is necessary to conduct a nucleic acid test sample defect analysis to re-collect nucleic acid test samples with serious defects. However, in the related art, the reliability of improving the defect analysis of nucleic acid test samples is poor. Summary of the invention

[0003] In view of this, the purpose of the present application is to provide a nucleic acid detection sample defect analysis method and system applied to a nucleic acid detector.

[0004] According to the first aspect of the present application, a method for analyzing nucleic acid detection sample defects applied to a nucleic acid detector is provided, which is applied to a cloud server platform, and the method comprises:

[0005] Determine image feature data corresponding to the target nucleic acid detection sample image, wherein the image feature data includes one or more nucleic acid detection labels and image features corresponding to each of the nucleic acid detection labels;

[0006] Loading the image feature data into a nucleic acid detection sample defect prediction model that meets the model convergence requirements, calling the nucleic acid detection sample defect prediction model to predict the image feature data to generate multiple nucleic acid detection sample defect information and confidence information corresponding to each nucleic acid detection sample defect information;

[0007] Based on the confidence information corresponding to each nucleic acid detection sample defect information, determine the target nucleic acid detection sample defect information whose corresponding confidence information meets the target output requirement from all the nucleic acid detection sample defect information, and determine it as the nucleic acid detection sample defect information corresponding to the target nucleic acid detection sample image;

[0008] Among them, the nucleic acid detection sample defect prediction model predicts all the nucleic acid detection sample defect information based on the image abnormal point features corresponding to each nucleic acid detection label and the image area features corresponding to each nucleic acid detection label.

[0009] In a possible implementation of the first aspect, the step of calling the nucleic acid detection sample defect prediction model to predict the image feature data to generate multiple nucleic acid detection sample defect information and confidence information corresponding to each nucleic acid detection sample defect information includes:

[0010] Based on the convolutional neural network corresponding to the nucleic acid detection sample defect prediction model, the image feature data is convoluted to obtain the convolution processing features corresponding to each nucleic acid detection label; the convolution processing features corresponding to each nucleic acid detection label include the image abnormal point features corresponding to each nucleic acid detection label and the image area features corresponding to each nucleic acid detection label;

[0011] Based on the feature processing network of the nucleic acid detection sample defect prediction model, feature mapping is performed on the convolution processing features corresponding to each of the nucleic acid detection labels to obtain a target image feature array corresponding to each of the nucleic acid detection labels;

[0012] Based on the feature dimensionality reduction network of the nucleic acid detection sample defect prediction model, feature dimensionality reduction processing is performed on the target image feature array corresponding to the nucleic acid detection label to obtain a reduced dimensionality image feature array corresponding to each nucleic acid detection label;

[0013] Based on the feature fusion network of the nucleic acid detection sample defect prediction model, the reduced-dimensional image feature arrays corresponding to all the nucleic acid detection labels are fused to obtain a fused feature distribution;

[0014] The prediction network based on the nucleic acid detection sample defect prediction model predicts the fusion feature distribution to generate multiple nucleic acid detection sample defect information and confidence information corresponding to each nucleic acid detection sample defect information.

[0015] In a possible implementation of the first aspect, the feature processing network of the nucleic acid detection sample defect prediction model performs feature mapping on the convolution processing features corresponding to each of the nucleic acid detection labels to obtain a target image feature array corresponding to each of the nucleic acid detection labels, including:

[0016] When the nucleic acid detection sample defect prediction model only includes one feature processing network, feature mapping is performed on the convolution processing features corresponding to each nucleic acid detection label based on the feature processing network to obtain a target image feature array corresponding to each nucleic acid detection label;

[0017] When the nucleic acid detection sample defect prediction model includes multiple image processing networks, for each nucleic acid detection label, the feature processing network associated with the nucleic acid detection label performs feature mapping on the convolution processing features corresponding to the nucleic acid detection label to obtain the target image feature array corresponding to the nucleic acid detection label.

[0018] In a possible implementation of the first aspect, determining image feature data corresponding to the target nucleic acid detection sample image includes:

[0019] Based on one or more preset nucleic acid detection labels, extract the image encoding vector corresponding to each nucleic acid detection label in the target nucleic acid detection sample image, and determine it as the image feature corresponding to the nucleic acid detection label;

[0020] Furthermore, before the image feature data is loaded into a nucleic acid detection sample defect prediction model that meets the model convergence requirements, and the nucleic acid detection sample defect prediction model is called to predict the image feature data to generate a plurality of nucleic acid detection sample defect information and confidence information corresponding to each nucleic acid detection sample defect information, the method further includes:

[0021] Pre-processing the image feature data;

[0022] The pre-processing of the image feature data includes:

[0023] For each image feature corresponding to the nucleic acid detection label, analyze whether there is a redundant image coding vector that meets the noise feature in the image feature. When there is a redundant image coding vector that meets the noise feature, eliminate the redundant image coding vector in the image feature.

[0024] In a possible implementation of the first aspect, analyzing, for each image feature corresponding to the nucleic acid detection label, whether there is a redundant image coding vector that meets the noise feature in the image feature, includes:

[0025] For each image feature corresponding to the nucleic acid detection label, analyzing whether there is a first image coding vector associated with a pixel point category and a preset first pixel point category in the image feature; if the first image coding vector exists in the image feature, determining whether there is a redundant image coding vector that meets the noise feature in the image feature, and the redundant image coding vector includes the first image coding vector; and / or

[0026] For each image feature corresponding to the nucleic acid detection label, analyzing whether there is a second image coding vector in the image feature that is associated and has the same pixel category and the pixel category is associated with a preset second pixel category; if the second image coding vector exists in the image feature, determining whether there is a redundant image coding vector in the image feature that meets the noise feature, and the redundant image coding vector includes part of the second image coding vector;

[0027] Before removing the redundant image coding vector in the image feature, the method further includes:

[0028] Analyze whether the suppression parameter value of the defect analysis of the redundant image coding vector in the image feature is greater than the target parameter value; when the suppression parameter value of the defect analysis of the image feature is not greater than the target parameter value, implement the operation of eliminating the redundant image coding vector in the image feature.

[0029] In a possible implementation of the first aspect, before determining the image feature data corresponding to the target nucleic acid detection sample image, the method further includes:

[0030] When the initial nucleic acid detection sample image to be subjected to defect analysis includes two or more detection sample image blocks, the nucleic acid detection element information of each of the detection sample image blocks is determined, wherein the nucleic acid detection element information of each of the detection sample image blocks includes the nucleic acid detection essential element corresponding to the detection sample image block and / or the nucleic acid detection auxiliary element corresponding to the detection sample image block;

[0031] Based on the nucleic acid detection element information of each of the detection sample image blocks, determining one or more target nucleic acid detection sample images from the initial nucleic acid detection sample image;

[0032] Wherein, determining one or more target nucleic acid detection sample images from the initial nucleic acid detection sample image based on the nucleic acid detection element information of each detection sample image block includes:

[0033] When the nucleic acid detection element information of all detection sample image blocks matches, determining the initial nucleic acid detection sample image as the target nucleic acid detection sample image;

[0034] When the nucleic acid detection element information of all detection sample image blocks is not evenly matched, all the detection sample image blocks are classified based on the nucleic acid detection element information of each of the detection sample image blocks to obtain two or more target nucleic acid detection sample images, wherein if a certain target nucleic acid detection sample image includes two or more detection sample image blocks, the nucleic acid detection element information of all the detection sample image blocks in the target nucleic acid detection sample image is matched.

[0035] In a possible implementation of the first aspect, determining the nucleic acid detection element information of each of the detection sample image blocks includes:

[0036] Determine one or more image elements corresponding to each of the detection sample image blocks, and determine a key image feature area that meets target requirements in each of the detection sample image blocks;

[0037] For each of the detection sample image blocks, based on the block mapping information of each of the image elements in the detection sample image block and the block mapping information of the key image feature area in the detection sample image block, determine the image element label of each of the image elements in the detection sample image block, and determine it as the nucleic acid detection element information of the detection sample image block;

[0038] The determining of the key image feature area that meets the target requirements in each of the detection sample image blocks includes:

[0039] For each of the detection sample image blocks, analyze whether there is a first image feature area for introducing the nucleic acid detection target in the detection sample image block. When the first image feature area for introducing the nucleic acid detection target exists in the detection sample image block, determine the first image feature area as a key image feature area that meets the target requirements in the detection sample image block. When the first image feature area for introducing the nucleic acid detection target does not exist in the detection sample image block, determine the second image feature area in the detection sample image block for expressing the nucleic acid detection feature and determine it as the key image feature area that meets the target requirements in the detection sample image block.

[0040] According to the second aspect of the present application, a cloud server platform is provided, wherein the nucleic acid detector comprises a machine-readable storage medium and a processor, wherein the machine-readable storage medium stores machine-executable instructions, and when the processor executes the machine-executable instructions, the cloud server platform implements the aforementioned nucleic acid detection sample defect analysis method applied to the nucleic acid detector.

[0041] According to the third aspect of the present application, a computer-readable storage medium is provided, in which computer-executable instructions are stored. When the computer-executable instructions are executed, the aforementioned nucleic acid detection sample defect analysis method applied to the nucleic acid detector is implemented.

[0042] Based on any of the above aspects, in the present application, the nucleic acid detection sample defect prediction model can be used to determine the nucleic acid detection sample defect information of the target nucleic acid detection sample image based on the image outlier features and image area features of the target nucleic acid detection sample image, thereby improving the correlation between different image areas during the nucleic acid detection sample defect analysis process, and thus more accurately understanding the image features of the target nucleic acid detection sample image, thereby improving the reliability of the nucleic acid detection sample defect analysis. In addition, by distinguishing the model input of the nucleic acid detection label, the reliability of the nucleic acid detection sample defect analysis can be further improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] 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 creative work.

[0044] Figure 1 A schematic diagram of a flow chart of a method for analyzing nucleic acid detection sample defects applied to a nucleic acid detector provided in an embodiment of the present application;

[0045] Figure 2 A schematic diagram of the component structure of a cloud server platform provided in an embodiment of the present application for implementing the above-mentioned nucleic acid detection sample defect analysis method applied to a nucleic acid detector is shown. Implementation

[0046] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below based on the drawings in the embodiments of the present application. It should be understood that the drawings in the present application are only for the purpose of illustration and description, and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowcharts used in this application illustrate the operations implemented in some embodiments of the embodiments of the present application. It should be understood that the operations of the flowchart may not be implemented in sequence, and steps without logical contextual relationships may be reversed in order or implemented simultaneously. In addition, under the guidance of the VR running behavior of the present application, those skilled in the art may add one or more other operations to the flowchart, or may destroy one or more operations from the flowchart.

[0047] Figure 1 The flowchart of the method for analyzing the defects of nucleic acid detection samples applied to nucleic acid detectors provided in the embodiment of the present application is shown. It should be understood that in other embodiments, the order of some steps in the method for analyzing the defects of nucleic acid detection samples applied to nucleic acid detectors in this embodiment can be shared with each other according to actual needs, or some steps can be omitted or maintained. The details of the method for analyzing the defects of nucleic acid detection samples applied to nucleic acid detectors include:

[0048] Step S101, determining image feature data corresponding to the target nucleic acid detection sample image.

[0049] The image feature data includes one or more nucleic acid detection tags and the image features corresponding to each nucleic acid detection tag. The nucleic acid detection tags may include various types of nucleic acid detection categories.

[0050] Determining the image feature data corresponding to the target nucleic acid detection sample image may include:

[0051] Based on the preset one or more nucleic acid detection labels, the image coding vector corresponding to each nucleic acid detection label in the target nucleic acid detection sample image is extracted to determine the image feature corresponding to the nucleic acid detection label.

[0052] Therefore, the image coding vector in the target nucleic acid detection sample image can be extracted based on multiple nucleic acid detection labels, and the image features in the target nucleic acid detection sample image can be accurately understood.

[0053] Step S102: load the image feature data into a nucleic acid detection sample defect prediction model that meets the model convergence requirements, call the nucleic acid detection sample defect prediction model to predict the image feature data to generate multiple nucleic acid detection sample defect information and confidence information corresponding to each nucleic acid detection sample defect information.

[0054] The nucleic acid detection sample defect prediction model predicts all nucleic acid detection sample defect information based on the image abnormal point features corresponding to each nucleic acid detection label and the image region features corresponding to each nucleic acid detection label, wherein the image abnormal point features corresponding to each nucleic acid detection label represent the image abnormal information corresponding to each image coding vector in the image feature corresponding to the nucleic acid detection label, and the image region features corresponding to each nucleic acid detection label represent the block mapping information corresponding to each image coding vector in the image feature corresponding to the nucleic acid detection label. In this way, by determining the nucleic acid detection sample defect information of the target nucleic acid detection sample image based on the image abnormal point features and the image region features, the correlation between different image regions in the nucleic acid detection sample defect analysis process is improved, and the image features of the target nucleic acid detection sample image can be understood more accurately.

[0055] Among them, the step of calling the nucleic acid detection sample defect prediction model to predict the image feature data to generate multiple nucleic acid detection sample defect information and confidence information corresponding to each nucleic acid detection sample defect information may include:

[0056] The convolution neural network corresponding to the nucleic acid detection sample defect prediction model performs convolution processing on the image feature data to obtain the convolution processing features corresponding to each nucleic acid detection label; the convolution processing features corresponding to each nucleic acid detection label include the image abnormal point features corresponding to each nucleic acid detection label and the image region features corresponding to each nucleic acid detection label;

[0057] The feature processing network of the nucleic acid detection sample defect prediction model performs feature mapping on the convolution processing features corresponding to each nucleic acid detection label to obtain a target image feature array corresponding to each nucleic acid detection label;

[0058] The feature dimension reduction network of the nucleic acid detection sample defect prediction model performs feature dimension reduction processing on the target image feature array corresponding to the nucleic acid detection label to obtain a reduced dimension image feature array corresponding to each nucleic acid detection label;

[0059] The feature fusion network of the nucleic acid detection sample defect prediction model performs fusion on the reduced-dimensional image feature arrays corresponding to all nucleic acid detection labels to obtain a fusion feature distribution;

[0060] The step of using the prediction network of the nucleic acid detection sample defect prediction model to predict the fusion feature distribution to generate multiple nucleic acid detection sample defect information and confidence information corresponding to each nucleic acid detection sample defect information.

[0061] In some exemplary design ideas, the feature processing network of the nucleic acid detection sample defect prediction model performs feature mapping on the convolution processing features corresponding to each nucleic acid detection label to obtain the target image feature array corresponding to each nucleic acid detection label, which may include:

[0062] When the nucleic acid detection sample defect prediction model only includes one feature processing network, the feature processing network performs feature mapping on the convolution processing features corresponding to each nucleic acid detection label to obtain a target image feature array corresponding to each nucleic acid detection label;

[0063] When the nucleic acid detection sample defect prediction model includes multiple image processing networks, for each nucleic acid detection label, the feature processing network associated with the nucleic acid detection label performs feature mapping on the convolution processing features corresponding to the nucleic acid detection label to obtain the target image feature array corresponding to the nucleic acid detection label.

[0064] In some exemplary design ideas, different nucleic acid detection tags correspond to different feature processing networks.

[0065] Step S103: Based on the confidence information corresponding to each nucleic acid detection sample defect information, determine the target nucleic acid detection sample defect information whose corresponding confidence information meets the target output requirements from all nucleic acid detection sample defect information, and determine it as the nucleic acid detection sample defect information corresponding to the target nucleic acid detection sample image.

[0066] In some exemplary design ideas, based on the confidence information corresponding to each nucleic acid detection sample defect information, determining the target nucleic acid detection sample defect information whose corresponding confidence information meets the target output requirements from all nucleic acid detection sample defect information, and determining the nucleic acid detection sample defect information corresponding to the target nucleic acid detection sample image, may include:

[0067] Based on the confidence information corresponding to each of the nucleic acid detection sample defect information, the nucleic acid detection sample defect information with the maximum confidence represented by the corresponding confidence information among all the nucleic acid detection sample defect information is determined as the target nucleic acid detection sample defect information that meets the target output requirements, and is determined as the nucleic acid detection sample defect information corresponding to the target nucleic acid detection sample image.

[0068] In this way, the nucleic acid detection sample defect information with the highest confidence can be used as the nucleic acid detection sample defect information of the target nucleic acid detection sample image, thereby improving the accuracy of the nucleic acid detection sample defect analysis.

[0069] In some exemplary design ideas, before determining the image feature data corresponding to the target nucleic acid detection sample image, the method may also include:

[0070] When the initial nucleic acid detection sample image to be subjected to defect analysis includes two or more detection sample image blocks, the nucleic acid detection element information of each detection sample image block is determined, wherein the nucleic acid detection element information of each detection sample image block includes the nucleic acid detection essential element corresponding to the detection sample image block and / or the nucleic acid detection auxiliary element corresponding to the detection sample image block;

[0071] Based on the nucleic acid detection element information of each detection sample image block, one or more target nucleic acid detection sample images are determined from the initial nucleic acid detection sample image.

[0072] In some exemplary design ideas, based on the nucleic acid detection element information of each detection sample image block, determining one or more target nucleic acid detection sample images from the initial nucleic acid detection sample image may include:

[0073] When the nucleic acid detection element information of all detection sample image blocks matches, the initial nucleic acid detection sample image is determined as the target nucleic acid detection sample image;

[0074] When the nucleic acid detection element information of all detection sample image blocks is not evenly matched, all detection sample image blocks are classified based on the nucleic acid detection element information of each detection sample image block to obtain two or more target nucleic acid detection sample images, wherein if a target nucleic acid detection sample image includes two or more detection sample image blocks, the nucleic acid detection element information of all detection sample image blocks in the target nucleic acid detection sample image is matched.

[0075] In some exemplary design ideas, when the nucleic acid detection element information of all detection sample image blocks is not evenly matched, the method may further include:

[0076] After predicting and obtaining the nucleic acid detection sample defect information corresponding to each of the target nucleic acid detection sample images, the nucleic acid detection sample defect information corresponding to the initial nucleic acid detection sample image is determined based on the nucleic acid detection sample defect information corresponding to each of the target nucleic acid detection sample images.

[0077] In some exemplary design ideas, determining the nucleic acid detection element information of each detection sample image block may include:

[0078] Determine one or more image elements corresponding to each detection sample image block, and determine a key image feature area that meets the target requirements in each detection sample image block;

[0079] For each detection sample image block, based on the block mapping information of each image element in the detection sample image block and the block mapping information of the key image feature area in the detection sample image block, the image element label of each image element in the detection sample image block is determined, and the nucleic acid detection element information of the detection sample image block is determined.

[0080] Therefore, the image element label of the image element can be determined based on the block mapping information of the image element and the key image feature area in the detection sample image block, and the nucleic acid detection element information of the detection sample image block can be obtained, thereby improving the accuracy of determining the nucleic acid detection element information of the detection sample image block.

[0081] In some exemplary design ideas, determining the key image feature area that meets the target requirements in each detection sample image block may include:

[0082] For each of the detection sample image blocks, analyze whether there is a first image feature area for introducing the nucleic acid detection target in the detection sample image block. When the first image feature area for introducing the nucleic acid detection target exists in the detection sample image block, determine the first image feature area as a key image feature area that meets the target requirements in the detection sample image block. When the first image feature area for introducing the nucleic acid detection target does not exist in the detection sample image block, determine the second image feature area in the detection sample image block for expressing the nucleic acid detection feature and determine it as the key image feature area that meets the target requirements in the detection sample image block.

[0083] Based on the above steps, the nucleic acid detection sample defect prediction model can be used to determine the nucleic acid detection sample defect information of the target nucleic acid detection sample image based on the image outlier features and image area features of the target nucleic acid detection sample image, thereby improving the correlation between different image areas during the nucleic acid detection sample defect analysis process, and thus more accurately understanding the image features of the target nucleic acid detection sample image, thereby improving the reliability of the nucleic acid detection sample defect analysis. In addition, by distinguishing the model input of the nucleic acid detection label, the reliability of the nucleic acid detection sample defect analysis can be further improved.

[0084] In another embodiment, the nucleic acid detection sample defect analysis method applied to the nucleic acid detector may include the following operations:

[0085] Step S201, determine the image feature data corresponding to the target nucleic acid detection sample image, where the image feature data includes one or more nucleic acid detection labels and the image features corresponding to each nucleic acid detection label.

[0086] Step S202: pre-process the image feature data.

[0087] As an optional implementation, pre-processing the image feature data may include:

[0088] For the image feature corresponding to each nucleic acid detection label, analyze whether there is a redundant image coding vector that meets the noise feature in the image feature. When there is a redundant image coding vector that meets the noise feature, eliminate the redundant image coding vector in the image feature.

[0089] In some exemplary design ideas, for each image feature corresponding to a nucleic acid detection label, analyzing whether there is a redundant image coding vector that meets the noise feature in the image feature may include:

[0090] For each image feature corresponding to the nucleic acid detection label, analyzing whether there is a first image coding vector associated with a pixel point category and a preset first pixel point category in the image feature, if the first image coding vector exists in the image feature, determining whether there is a redundant image coding vector that meets the noise feature in the image feature, wherein the redundant image coding vector includes the first image coding vector; and / or,

[0091] For the image feature corresponding to each nucleic acid detection label, analyze whether there is a second image coding vector in the image feature that is associated and has the same pixel category and a pixel category associated with a preset second pixel category. If there is a second image coding vector in the image feature, determine whether there is a redundant image coding vector in the image feature that meets the noise feature, wherein the redundant image coding vector includes part of the second image coding vector.

[0092] Before removing the redundant image coding vectors in the image feature, the method may further include:

[0093] Analyze whether the suppression parameter value of the redundant image coding vector in the image feature for the defect analysis of the image feature is greater than the target parameter value; when the suppression parameter value of the redundant image coding vector in the image feature for the defect analysis of the image feature is not greater than the target parameter value, trigger the above-mentioned operation of removing the redundant image coding vector in the image feature.

[0094] Step S203: load the image feature data into a nucleic acid detection sample defect prediction model that meets the model convergence requirements, call the nucleic acid detection sample defect prediction model to predict the image feature data to generate multiple nucleic acid detection sample defect information and confidence information corresponding to each nucleic acid detection sample defect information.

[0095] Step S204: Based on the confidence information corresponding to each nucleic acid detection sample defect information, determine the target nucleic acid detection sample defect information whose corresponding confidence information meets the target output requirements from all nucleic acid detection sample defect information, and determine it as the nucleic acid detection sample defect information corresponding to the target nucleic acid detection sample image.

[0096] Figure 2 The diagram schematically shows a cloud server platform 100 that can be used to implement the various embodiments described in the present application.

[0097] For one embodiment, Figure 2 A cloud server platform 100 is shown having one or more processors 102, a control module (chip set) 104 coupled to one or more of the (one or more) processors 102, a memory 106 coupled to the control module 104, a non-volatile memory (NVM) / storage device 108 coupled to the control module 104, one or more input / output devices 110 coupled to the control module 104, and a network interface 112 coupled to the control module 106.

[0098] The processor 102 may include one or more single-core or multi-core processors, and the processor 102 may include any combination of general-purpose processors or special-purpose processors (such as graphics processors, application processors, baseband processors, etc.). In some exemplary design ideas, the cloud server platform 100 can be used as a cloud server platform device such as a gateway described in the embodiments of the present application.

[0099] In some exemplary design concepts, the cloud server platform 100 may include one or more computer-readable media (e.g., memory 106 or NVM / storage device 108) having instructions 114 and one or more processors 102 integrated with the one or more computer-readable media and configured to execute instructions 114 to implement a module to perform the actions described in the present disclosure.

[0100] For one embodiment, the control module 104 may include any suitable interface controller to provide any suitable interface to one or more of the processor(s) 102 and / or any suitable device or component in communication with the control module 104 .

[0101] The control module 104 may include a memory controller module to provide an interface to the memory 106. The memory controller module may be a hardware module, a software module, and / or a firmware module.

[0102] The memory 106 may be used, for example, to load and store data and / or instructions 114 for the cloud server platform 100. For one embodiment, the memory 106 may include any suitable volatile memory, such as a suitable DRAM.

[0103] For one embodiment, control module 104 may include one or more input / output controllers to provide interfaces to NVM / storage device 108 and input / output device(s) 110 .

[0104] For example, NVM / storage 108 may be used to store data and / or instructions 114. NVM / storage 108 may include any suitable non-volatile memory (e.g., flash memory) and / or may include any suitable non-volatile storage device(s) (e.g., one or more hard disk drives (HDDs), one or more compact disk (CD) drives, and / or one or more digital versatile disk (DVD) drives).

[0105] NVM / storage device 108 may include storage resources that are physically part of the device on which cloud server platform 100 is installed, or it may be accessible to the device without being part of the device. For example, NVM / storage device 108 may be accessed via (one or more) input / output devices 110 over a network.

[0106] (One or more) input / output devices 110 may provide an interface for the cloud server platform 100 to communicate with any other appropriate device, and the input / output device 110 may include a communication component, a phonetic component, a sensor component, etc. The network interface 112 may provide an interface for the cloud server platform 100 to communicate based on one or more networks, and the cloud server platform 100 may wirelessly communicate with one or more components of a wireless network based on any of one or more wireless network standards and / or protocols, such as accessing a wireless network based on a communication standard.

[0107] For one embodiment, one or more of the processor(s) 102 may be loaded together with the logic of one or more controllers (e.g., a memory controller module) of the control module 104. For one embodiment, one or more of the processor(s) 102 may be loaded together with the logic of one or more controllers of the control module 104 to form a system level load. For one embodiment, one or more of the processor(s) 102 may be integrated on the same die with the logic of one or more controllers of the control module 104. For one embodiment, one or more of the processor(s) 102 may be integrated on the same die with the logic of one or more controllers of the control module 104 to form a system on chip (SoC).

[0108] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for analyzing nucleic acid detection sample defects applied to a nucleic acid detector, characterized in that: Applied to a cloud server platform, the method includes: Determine image feature data corresponding to the target nucleic acid detection sample image, wherein the image feature data includes one or more nucleic acid detection labels and image features corresponding to each of the nucleic acid detection labels; Loading the image feature data into a nucleic acid detection sample defect prediction model that meets the model convergence requirements, calling the nucleic acid detection sample defect prediction model to predict the image feature data to generate multiple nucleic acid detection sample defect information and confidence information corresponding to each nucleic acid detection sample defect information; Based on the confidence information corresponding to each nucleic acid detection sample defect information, determine the target nucleic acid detection sample defect information whose corresponding confidence information meets the target output requirement from all the nucleic acid detection sample defect information, and determine it as the nucleic acid detection sample defect information corresponding to the target nucleic acid detection sample image; The nucleic acid detection sample defect prediction model predicts all the nucleic acid detection sample defect information based on the image abnormal point features corresponding to each nucleic acid detection label and the image region features corresponding to each nucleic acid detection label; Before determining the image feature data corresponding to the target nucleic acid detection sample image, the method further includes: When the initial nucleic acid detection sample image to be subjected to defect analysis includes two or more detection sample image blocks, the nucleic acid detection element information of each of the detection sample image blocks is determined, wherein the nucleic acid detection element information of each of the detection sample image blocks includes the nucleic acid detection essential element corresponding to the detection sample image block and / or the nucleic acid detection auxiliary element corresponding to the detection sample image block; Based on the nucleic acid detection element information of each of the detection sample image blocks, determining one or more target nucleic acid detection sample images from the initial nucleic acid detection sample image; Wherein, determining one or more target nucleic acid detection sample images from the initial nucleic acid detection sample image based on the nucleic acid detection element information of each detection sample image block includes: When the nucleic acid detection element information of all detection sample image blocks matches, determining the initial nucleic acid detection sample image as the target nucleic acid detection sample image; When the nucleic acid detection element information of all detection sample image blocks is not evenly matched, all the detection sample image blocks are classified based on the nucleic acid detection element information of each of the detection sample image blocks to obtain two or more target nucleic acid detection sample images, wherein if a certain target nucleic acid detection sample image includes two or more detection sample image blocks, the nucleic acid detection element information of all the detection sample image blocks in the target nucleic acid detection sample image is matched.

2. The method for analyzing nucleic acid detection sample defects applied to a nucleic acid detector according to claim 1, characterized in that: The step of calling the nucleic acid detection sample defect prediction model to predict the image feature data to generate multiple nucleic acid detection sample defect information and confidence information corresponding to each nucleic acid detection sample defect information includes: Based on the convolutional neural network corresponding to the nucleic acid detection sample defect prediction model, the image feature data is convoluted to obtain the convolution processing features corresponding to each nucleic acid detection label; the convolution processing features corresponding to each nucleic acid detection label include the image abnormal point features corresponding to each nucleic acid detection label and the image area features corresponding to each nucleic acid detection label; Based on the feature processing network of the nucleic acid detection sample defect prediction model, feature mapping is performed on the convolution processing features corresponding to each of the nucleic acid detection labels to obtain a target image feature array corresponding to each of the nucleic acid detection labels; Based on the feature dimensionality reduction network of the nucleic acid detection sample defect prediction model, feature dimensionality reduction processing is performed on the target image feature array corresponding to the nucleic acid detection label to obtain a reduced dimensionality image feature array corresponding to each nucleic acid detection label; Based on the feature fusion network of the nucleic acid detection sample defect prediction model, the reduced-dimensional image feature arrays corresponding to all the nucleic acid detection labels are fused to obtain a fused feature distribution; The prediction network based on the nucleic acid detection sample defect prediction model predicts the fusion feature distribution to generate multiple nucleic acid detection sample defect information and confidence information corresponding to each nucleic acid detection sample defect information.

3. The method for analyzing nucleic acid detection sample defects applied to a nucleic acid detector according to claim 2, characterized in that: The feature processing network of the nucleic acid detection sample defect prediction model performs feature mapping on the convolution processing features corresponding to each nucleic acid detection label to obtain a target image feature array corresponding to each nucleic acid detection label, including: When the nucleic acid detection sample defect prediction model only includes one feature processing network, feature mapping is performed on the convolution processing features corresponding to each nucleic acid detection label based on the feature processing network to obtain a target image feature array corresponding to each nucleic acid detection label; When the nucleic acid detection sample defect prediction model includes multiple image processing networks, for each nucleic acid detection label, the feature processing network associated with the nucleic acid detection label performs feature mapping on the convolution processing features corresponding to the nucleic acid detection label to obtain the target image feature array corresponding to the nucleic acid detection label.

4. The method for analyzing nucleic acid detection sample defects applied to a nucleic acid detector according to any one of claims 1 to 3, characterized in that: The step of determining image feature data corresponding to the target nucleic acid detection sample image includes: Based on one or more preset nucleic acid detection labels, extract the image encoding vector corresponding to each nucleic acid detection label in the target nucleic acid detection sample image, and determine it as the image feature corresponding to the nucleic acid detection label; Furthermore, before the image feature data is loaded into a nucleic acid detection sample defect prediction model that meets the model convergence requirements, and the nucleic acid detection sample defect prediction model is called to predict the image feature data to generate a plurality of nucleic acid detection sample defect information and confidence information corresponding to each nucleic acid detection sample defect information, the method further includes: Pre-processing the image feature data; The pre-processing of the image feature data includes: For each image feature corresponding to the nucleic acid detection label, analyze whether there is a redundant image coding vector that meets the noise feature in the image feature. When there is a redundant image coding vector that meets the noise feature, eliminate the redundant image coding vector in the image feature.

5. The method for analyzing nucleic acid detection sample defects applied to a nucleic acid detector according to claim 4, characterized in that: The step of analyzing, for each image feature corresponding to the nucleic acid detection label, whether there is a redundant image coding vector that meets the noise feature in the image feature includes: For each image feature corresponding to the nucleic acid detection label, analyzing whether there is a first image coding vector associated with a pixel point category and a preset first pixel point category in the image feature; if the first image coding vector exists in the image feature, determining whether there is a redundant image coding vector that meets the noise feature in the image feature, and the redundant image coding vector includes the first image coding vector; and / or For each image feature corresponding to the nucleic acid detection label, analyzing whether there is a second image coding vector in the image feature that is associated and has the same pixel category and the pixel category is associated with a preset second pixel category; if the second image coding vector exists in the image feature, determining whether there is a redundant image coding vector in the image feature that meets the noise feature, and the redundant image coding vector includes part of the second image coding vector; Before removing the redundant image coding vector in the image feature, the method further includes: Analyze whether the suppression parameter value of the defect analysis of the redundant image coding vector in the image feature is greater than the target parameter value; when the suppression parameter value of the defect analysis of the image feature is not greater than the target parameter value, implement the operation of eliminating the redundant image coding vector in the image feature.

6. The method for analyzing nucleic acid detection sample defects applied to a nucleic acid detector according to claim 5, characterized in that: The determining of nucleic acid detection element information of each detection sample image block includes: Determine one or more image elements corresponding to each of the detection sample image blocks, and determine a key image feature area that meets target requirements in each of the detection sample image blocks; For each of the detection sample image blocks, based on the block mapping information of each of the image elements in the detection sample image block and the block mapping information of the key image feature area in the detection sample image block, determine the image element label of each of the image elements in the detection sample image block, and determine it as the nucleic acid detection element information of the detection sample image block; The determining of the key image feature area that meets the target requirements in each of the detection sample image blocks includes: For each of the detection sample image blocks, analyze whether there is a first image feature area for introducing the nucleic acid detection target in the detection sample image block. When the first image feature area for introducing the nucleic acid detection target exists in the detection sample image block, determine the first image feature area as a key image feature area that meets the target requirements in the detection sample image block. When the first image feature area for introducing the nucleic acid detection target does not exist in the detection sample image block, determine the second image feature area in the detection sample image block for expressing the nucleic acid detection feature and determine it as the key image feature area that meets the target requirements in the detection sample image block.

7. A nucleic acid detection sample defect analysis system applied to a nucleic acid detector, characterized in that: The nucleic acid detection sample defect analysis system applied to the nucleic acid detector includes a cloud server platform and a nucleic acid detector connected to the cloud server platform for communication. The cloud server platform is specifically used for: Determine image feature data corresponding to the target nucleic acid detection sample image, wherein the image feature data includes one or more nucleic acid detection labels and image features corresponding to each of the nucleic acid detection labels; Loading the image feature data into a nucleic acid detection sample defect prediction model that meets the model convergence requirements, calling the nucleic acid detection sample defect prediction model to predict the image feature data to generate multiple nucleic acid detection sample defect information and confidence information corresponding to each nucleic acid detection sample defect information; Based on the confidence information corresponding to each nucleic acid detection sample defect information, determine the target nucleic acid detection sample defect information whose corresponding confidence information meets the target output requirement from all the nucleic acid detection sample defect information, and determine it as the nucleic acid detection sample defect information corresponding to the target nucleic acid detection sample image; The nucleic acid detection sample defect prediction model predicts all the nucleic acid detection sample defect information based on the image abnormal point features corresponding to each nucleic acid detection label and the image region features corresponding to each nucleic acid detection label; Before determining the image feature data corresponding to the target nucleic acid detection sample image, the cloud server platform is further used to: When the initial nucleic acid detection sample image to be subjected to defect analysis includes two or more detection sample image blocks, the nucleic acid detection element information of each of the detection sample image blocks is determined, wherein the nucleic acid detection element information of each of the detection sample image blocks includes the nucleic acid detection essential element corresponding to the detection sample image block and / or the nucleic acid detection auxiliary element corresponding to the detection sample image block; Based on the nucleic acid detection element information of each of the detection sample image blocks, determining one or more target nucleic acid detection sample images from the initial nucleic acid detection sample image; Wherein, determining one or more target nucleic acid detection sample images from the initial nucleic acid detection sample image based on the nucleic acid detection element information of each detection sample image block includes: When the nucleic acid detection element information of all detection sample image blocks matches, determining the initial nucleic acid detection sample image as the target nucleic acid detection sample image; When the nucleic acid detection element information of all detection sample image blocks is not evenly matched, all the detection sample image blocks are classified based on the nucleic acid detection element information of each of the detection sample image blocks to obtain two or more target nucleic acid detection sample images, wherein if a certain target nucleic acid detection sample image includes two or more detection sample image blocks, the nucleic acid detection element information of all the detection sample image blocks in the target nucleic acid detection sample image is matched.

8. A cloud server platform, characterized in that: The cloud server platform includes a processor and a computer-readable storage medium, wherein the computer-readable storage medium stores machine-executable instructions, and when the machine-executable instructions are executed by a computer, the nucleic acid detection sample defect analysis method applied to a nucleic acid detector as described in any one of claims 1-6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores machine executable instructions, and when the machine executable instructions are executed by a computer, the nucleic acid detection sample defect analysis method applied to a nucleic acid detector as described in any one of claims 1-6 is implemented.

Images