CN116187294A

CN116187294A - Method and system for rapidly generating electronic file of informationized detection laboratory

Info

Publication number: CN116187294A
Application number: CN202310446592.9A
Authority: CN
Inventors: 李刚; 张宾武
Original assignee: Kaiyuan Huachuang Technology Group Co ltd
Current assignee: Kaiyuan Huachuang Technology Group Co ltd
Priority date: 2023-04-24
Filing date: 2023-04-24
Publication date: 2023-05-30
Anticipated expiration: 2043-04-24
Also published as: CN116187294B

Abstract

The application relates to the technical field of electronic file generation, and particularly discloses a method and a system for quickly generating an electronic file of an informationized detection laboratory, which are used for receiving experimental data transmitted by laboratory equipment, automatically braiding and formatting the experimental data based on a preset template and a specification to generate the laboratory electronic file, and quickly generating the laboratory electronic file in such a way.

Description

Method and system for rapidly generating electronic file of informationized detection laboratory

Technical Field

The application relates to the technical field of electronic file generation, and in particular relates to a method and a system for quickly generating an electronic file in an informationized detection laboratory.

Background

Laboratory electronic documents are an important component of laboratory work and record the purpose, course, results and conclusions of the experiment, as well as the instruments, materials and methods used. The rapid generation of laboratory electronic files can improve efficiency and quality, ensure traceability and credibility, and promote communication and collaboration.

In the prior art, a plurality of methods for generating an electronic file are disclosed, for example, in chinese patent with application No. 2015123056. X, the method for generating an electronic file includes the following steps: receiving an electronic file generation instruction and identity information; sending an information acquisition request corresponding to the electronic file generation instruction and the identity information to a server corresponding to the electronic file generation instruction; and when receiving the data information fed back by the server based on the electronic file generation instruction and the identity information, generating the electronic file according to the data information. The system of the to-be-handled mechanism receives the electronic file generation instruction and the identity information, and sends an information acquisition request to the server corresponding to other mechanisms related to the to-be-handled mechanism, so that the to-be-handled mechanism can directly acquire corresponding data information of other mechanism systems, and further, the to-be-handled mechanism can directly open electronic files corresponding to other mechanism systems related to the to-be-handled mechanism.

For another example, chinese patent application No. 201310041567.9 discloses a method and apparatus for generating an electronic file, where the method includes creating a hypertext markup language file, writing header information, name information and body information of the hypertext markup language file, writing a comment tag in the body information, where the body data of the comment tag is program data generated by an application program, writing summary information in the body information, and the summary information is information represented by the program data. The invention can generate the hypertext markup language file by the application program, and is convenient for people receiving the electronic file to read the main information in the electronic file.

Similar to the electronic file generation technology, a large amount of generation instructions and data information need to be received, quality detection on the data information is lacking, deviation is easy to occur, a preset template and a preset standard are lacking, the speed of generating the electronic file is low, and the user needs are difficult to meet.

Therefore, a rapid generation scheme for informationized detection laboratory electronic files is desired.

Disclosure of Invention

The present application has been made in order to solve the above technical problems. The embodiment of the application provides a method and a system for quickly generating an informationized detection laboratory electronic file, which are used for receiving experimental data transmitted by laboratory equipment, automatically braiding and formatting the experimental data based on a preset template and a specification to generate the laboratory electronic file, and quickly generating the laboratory electronic file in such a way.

According to one aspect of the present application, there is provided an informationized detection laboratory electronic document rapid generation system comprising:

the experimental data receiving module is used for receiving experimental data transmitted by laboratory equipment; and

and the file generation module is used for automatically braiding and formatting the experimental data based on a preset template and a specification to generate a laboratory electronic file.

In the escalator voice prompt system, the image feature extraction module is configured to: each layer of the convolutional neural network model using the feature extractor performs, in forward transfer of the layer, input data: carrying out convolution processing on input data to obtain a convolution characteristic diagram; pooling the convolution feature images based on the local feature matrix to obtain pooled feature images; performing nonlinear activation on the pooled feature map to obtain an activated feature map; the output of the last layer of the convolutional neural network serving as the feature extractor is the shallow feature map and the deep feature map, and the input of the first layer of the convolutional neural network serving as the feature extractor is the staircase monitoring image.

In the above system for rapidly generating electronic files in an informationized detection laboratory, the experimental data receiving module includes: the data acquisition unit is used for acquiring experimental data of a plurality of preset time points in a preset time period acquired by laboratory equipment, wherein the experimental data are image data; a quality evaluation unit for passing the image data of each predetermined point in time through an image quality evaluator based on an encoder-decoder structure to obtain an image quality timing input vector composed of a plurality of image quality score values; a time sequence difference unit, which is used for calculating the difference between the image quality score values of every two adjacent positions in the image quality time sequence input vector to obtain an image quality time sequence difference input vector; the image quality time sequence information integration unit is used for cascading the image quality time sequence input vector and the image quality time sequence difference input vector to obtain an image quality input vector; an image quality time sequence change feature extraction unit, configured to pass the image quality input vector through an image quality time sequence feature extractor including a first convolution layer and a second convolution layer to obtain an image quality time sequence feature vector; and the device performance detection unit is used for passing the image quality time sequence feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the performance of the laboratory device for providing the plurality of experimental data is normal or not.

In the above-mentioned information-based detection laboratory electronic file rapid generation system, the quality evaluation unit includes: an encoding subunit for extracting image features from the image data at the respective predetermined points in time using an encoder of the image quality evaluator to obtain a plurality of image features, wherein the encoder of the image quality evaluator comprises a plurality of convolutional layers; and a decoding subunit for decoding the plurality of image features using a decoder of the image quality evaluator to obtain the image quality timing input vector comprised of a plurality of image quality decoded values, wherein the decoder of the image quality evaluator comprises a plurality of deconvolution layers.

In the above system for rapidly generating an electronic file in an informationized detection laboratory, the image quality time sequence variation feature extraction unit includes: a first image quality timing feature extraction subunit configured to input the image quality input vector into a first convolution layer of the image quality timing feature extractor to obtain a first image quality timing feature vector, where the first convolution layer has a first one-dimensional convolution kernel of a first length; a second image quality temporal feature extraction subunit configured to input the image quality input vector into a second convolution layer of the image quality temporal feature extractor to obtain a second image quality temporal feature vector, where the second convolution layer has a second one-dimensional convolution kernel of a second length, the first length being different from the second length; and a cascading subunit, configured to concatenate the first image quality timing feature vector and the second image quality timing feature vector to obtain the image quality timing feature vector.

In the above system for rapidly generating an electronic file in an informationized detection laboratory, the device performance detection unit is configured to: processing the image quality temporal feature vector using the classifier to obtain a classification result with the formula:

, wherein />

Is a weight matrix>

To->

For the bias vector +.>

Is an image quality timing feature vector.

The system for quickly generating the electronic file in the informationized detection laboratory further comprises a training module for training the image quality time sequence feature extractor comprising the first convolution layer and the second convolution layer and the classifier.

In the above system for rapidly generating electronic files in an informationized detection laboratory, the training module includes: the training data acquisition unit is used for acquiring training data, wherein the training data comprise training experiment data of a plurality of preset time points in a preset time period and a true value of whether the performance of the laboratory equipment is normal or not; a training quality evaluation unit for passing the training experiment data of each predetermined time point through the encoder-decoder structure-based image quality evaluator to obtain a training image quality time sequence input vector composed of a plurality of training image quality score values; the training time sequence difference unit is used for calculating the difference between training image quality score values of every two adjacent positions in the training image quality time sequence input vector to obtain a training image quality time sequence difference input vector; the training image quality time sequence information integration unit is used for cascading the training image quality time sequence input vector and the training image quality time sequence difference input vector to obtain a training image quality input vector; a training image quality time sequence change feature extraction unit, configured to pass the training image quality input vector through the image quality time sequence feature extractor including the first convolution layer and the second convolution layer to obtain a training image quality time sequence feature vector; the classification loss unit is used for passing the training image quality time sequence feature vector through the classifier to obtain a classification loss function value; and a model training unit, configured to train the image quality time sequence feature extractor including the first convolution layer and the second convolution layer and the classifier based on the classification loss function value and through propagation in a direction of gradient descent, where in each iteration of the training, a feature affinity space affine learning iteration is performed on a weight matrix of the classifier.

In the above system for rapidly generating electronic files in an informationized detection laboratory, the classification loss unit is configured to: and calculating a cross entropy loss function value between the training classification result and a true value of whether the performance of the laboratory equipment is normal or not as the classification loss function value.

In the rapid generation system of the informationized detection laboratory electronic file, in each round of iteration of the training, carrying out characteristic affinity space affine learning iteration on the weight matrix of the classifier according to the following optimization formula; wherein, the optimization formula is:

wherein ,

a weight matrix representing said classifier, +.>

Two norms of a weight matrix representing the classifier +.>

A kernel norm of a weight matrix representing the classifier, and +.>

Is the scale of the weight matrix of the classifier,

represents a logarithmic function value based on 2, < +.>

An exponential operation representing a matrix representing the calculation of a natural exponential function value raised to a power by a characteristic value at each position in the matrix,/v>

Representing multiplication by location +. >

And representing the weight matrix of the classifier after iteration.

According to another aspect of the present application, there is provided a method for rapidly generating an electronic file in an informationized detection laboratory, including:

receiving experimental data transmitted by laboratory equipment; and

the experimental data is automatically woven and formatted to generate a laboratory electronic file based on a preset template and specification.

According to still another aspect of the present application, there is provided an electronic apparatus including: a processor; and a memory having stored therein computer program instructions that, when executed by the processor, cause the processor to perform the informationized detection laboratory electronic file rapid generation method as described above.

According to yet another aspect of the present application, there is provided a computer readable medium having stored thereon computer program instructions which, when executed by a processor, cause the processor to perform the informative detection laboratory electronic file rapid generation method as described above.

Compared with the prior art, the method and the system for quickly generating the electronic file of the informationized detection laboratory, which are provided by the application, receive experimental data transmitted by laboratory equipment, automatically weave and format the experimental data based on a preset template and a specification to generate the electronic file of the laboratory, and quickly generate the electronic file of the laboratory in such a way.

Drawings

The foregoing and other objects, features and advantages of the present application will become more apparent from the following more particular description of embodiments of the present application, as illustrated in the accompanying drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.

FIG. 1 is a block diagram of an informationized detection laboratory electronic file rapid generation system according to an embodiment of the present application;

FIG. 2 is a block diagram of an informationized detection laboratory electronic file rapid generation system according to an embodiment of the present application;

FIG. 3 is a system architecture diagram of an inference module in an informationized detection laboratory electronic file rapid generation system according to an embodiment of the present application;

FIG. 4 is a system architecture diagram of a training module in an informationized detection laboratory electronic file rapid generation system according to an embodiment of the present application;

FIG. 5 is a block diagram of an experimental data receiving module in an informationized detection laboratory electronic file rapid generation system according to an embodiment of the present application;

FIG. 6 is a block diagram of a quality assessment unit in an informationized detection laboratory electronic file rapid generation system according to an embodiment of the present application;

FIG. 7 is a block diagram of an image quality time sequence variation feature extraction unit in an informationized detection laboratory electronic file rapid generation system according to an embodiment of the present application;

FIG. 8 is a flow chart of a method for rapid generation of an informative detection laboratory electronic file according to an embodiment of the present application;

fig. 9 is a block diagram of an electronic device according to an embodiment of the present application.

Description of the embodiments

Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application and not all of the embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.

Summary of the application

As mentioned above, laboratory electronic files are an important component of laboratory work, recording the purpose, process, results and conclusions of the experiment, as well as the instruments, materials and methods used. The rapid generation of laboratory electronic files can improve efficiency and quality, ensure traceability and credibility, and promote communication and collaboration. Therefore, a rapid generation scheme for informationized detection laboratory electronic files is desired.

Specifically, in the technical solution of the present application, a method for quickly generating laboratory electronic files is proposed, which uses a professional laboratory electronic file management system (LIMS). LIMS can help laboratory manage and control various links of experiments including sample management, instrument management, data acquisition, data analysis, data reporting, and the like. LIMS can quickly generate laboratory electronic files by: LIMS can be connected with laboratory instruments and equipment to automatically collect and transmit data, thereby reducing human error and delay. The LIMS can automatically compile and format a data report according to a preset template and a preset standard, so that the consistency and standardization of data are improved. LIMS can support multiple data formats and languages, and adapt to different experimental requirements and user preferences. LIMS can provide a variety of data presentation and visualization means, such as charts, images, video, etc., to enhance the expressive and legibility of data. LIMS can protect the security and integrity of data by means of password, encryption, backup and the like, and prevent data leakage or damage.

Accordingly, it is considered that the LIMS may automatically collect data of laboratory equipment through connection with laboratory instruments and equipment, but if the laboratory equipment fails or the laboratory equipment performance deviates, the source data deviates, which in turn causes useless idle running of the whole laboratory electronic file rapid generation system, and affects the effectiveness of the whole laboratory electronic file rapid generation system.

Based on this, in the technical solution of the present application, it is desirable to evaluate whether the performance of the laboratory device is normal by analyzing the experimental data collected by the laboratory device. In particular, considering that data collected by laboratory equipment in a time dimension are mostly presented in the form of images, in the technical scheme of the application, experimental data are taken as an example of image data, image quality time sequence change implicit characteristic information in the image data is extracted, and detection of the experimental data is performed, so that performance of the laboratory equipment is evaluated. However, since the image quality is difficult to evaluate and the time-series variation characteristic of the image quality is fine implicit variation characteristic information of a small scale, the variation of the image quality characteristic in the time dimension is difficult to capture. Therefore, in this process, it is difficult to mine the time-series related feature distribution information about the image quality features in the image data, so as to accurately perform quality detection of the image data, that is, detect whether deviation occurs in the experimental source data, so as to accurately evaluate whether the performance of the laboratory equipment is normal.

In recent years, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, text signal processing, and the like. The development of deep learning and neural networks provides new solutions and solutions for mining time-series-associated feature distribution information about image quality features in the image data.

Specifically, in the technical solution of the present application, first, experimental data at a plurality of predetermined time points within a predetermined period of time are collected by laboratory equipment. It will be appreciated that if the laboratory device malfunctions or the source data deviates due to laboratory device performance deviation, this will be apparent at the laboratory data end, and that the laboratory device performance detection based on the feature analysis of the image data will be based on taking the laboratory data as an example of the image data, considering that the laboratory device data collected in the time dimension are mostly presented in the form of images.

Next, it is considered that there is performance characteristic information about the laboratory equipment in the image data due to the respective predetermined time points, and if there is no deviation in the performance of the laboratory equipment, there is no deviation in the source data reflected in the laboratory equipment. Thus, the performance characteristics of the laboratory device are represented in the image quality characteristics of the image data at the respective predetermined time points. Based on this, in the technical solution of the present application, the image data at each predetermined time point is subjected to quality evaluation of the experimental data at each predetermined time point by an image quality evaluator based on an encoder-decoder structure to determine whether there is a deviation, thereby obtaining an image quality time sequence input vector composed of a plurality of image quality score values.

Further, in consideration of the fact that the image quality score value of the experimental data of the laboratory equipment has a dynamic change rule in the time dimension, in order to accurately evaluate the image quality, dynamic change characteristics of the image quality of the experimental data in the time dimension are required to be extracted, in consideration of the fact that the change condition of the image quality score value of the experimental data in time sequence is weak, the weak change characteristics are small-scale change characteristic information relative to the image quality score value, if the time sequence dynamic change characteristic extraction of the image quality is performed by absolute change information, not only the calculated amount is large, but also the small-scale weak change characteristics of the image quality score value in the time dimension are difficult to perceive, and the accuracy of subsequent classification is further affected.

Based on the above, in the technical solution of the present application, the dynamic change feature extraction of the image quality is performed comprehensively by using the time sequence relative change feature and the absolute change feature of the image quality score. Specifically, first, the difference between the image quality score values of every adjacent two positions in the image quality time series input vector is calculated to obtain an image quality time series differential input vector. Next, it is considered that there is an association relationship with time-series dynamic change with respect to the image quality in the experimental data between time-series relative change characteristics and time-series absolute change characteristics due to the image quality score value. Therefore, in order to fully explore the dynamic change rule of the image quality in the time dimension so as to accurately detect the quality of experimental data, so that the laboratory equipment is subjected to performance accurate evaluation, in the technical scheme of the application, the image quality time sequence input vector and the image quality time sequence differential input vector are cascaded to obtain the image quality input vector.

Then, since the dynamic change law of the image quality score value in the time dimension has volatility and uncertainty, it presents different pattern change characteristic information at different time period spans within the predetermined time period. Therefore, in order to accurately detect the image quality in the experimental data and accurately evaluate the performance of the laboratory equipment, in the technical scheme of the application, the image quality input vector is further passed through an image quality time sequence feature extractor comprising a first convolution layer and a second convolution layer to obtain an image quality time sequence feature vector. In particular, here, the first convolution layer and the second convolution layer have different feature receptive fields to perform feature mining on the image quality input vector using different one-dimensional convolution kernels, so as to extract dynamic multi-scale associated feature information of the image quality score value under different time spans, that is, time-series multi-scale dynamic change feature information of the image quality.

And then, the image quality time sequence feature vector is further used as a classification feature vector to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the performance of laboratory equipment for providing the plurality of experimental data is normal or not. That is, in the technical solution of the present application, the tag of the classifier includes a normal performance (first tag) of the laboratory device that provides the plurality of experimental data, and an abnormal performance (second tag) of the laboratory device that provides the plurality of experimental data, wherein the classifier determines to which classification tag the classification feature vector belongs through a soft maximum function. It is noted that the first tag p1 and the second tag p2 do not include the concept of human settings, and in fact, during the training process, the computer model does not have the concept of "whether the performance of the laboratory equipment providing the plurality of experimental data is normal", which is simply that there are two kinds of classification tags and the probability that the output feature is under the two kinds of classification tags, i.e., the sum of p1 and p2 is one. Thus, the classification result of whether the performance of the laboratory device providing the plurality of experimental data is normal is actually converted into a classified probability distribution conforming to the natural law by classifying the tags, and the physical meaning of the natural probability distribution of the tags is essentially used instead of the linguistic text meaning of "whether the performance of the laboratory device providing the plurality of experimental data is normal". It should be understood that, in the technical solution of the present application, the classification label of the classifier is a control policy label for determining whether the performance of the laboratory device providing the plurality of experimental data is normal, so after the classification result is obtained, the quality detection of the image data can be accurately performed based on the classification result, that is, whether the deviation of the experimental source data occurs is detected, so as to accurately evaluate whether the performance of the laboratory device is normal.

In particular, in the technical solution of the present application, here, for the image quality input vector, since it is obtained by directly concatenating the image quality timing input vector and the image quality timing difference input vector, which are timing-series arrangement of an absolute value of image quality and a variation value of image quality, respectively, which are not perfectly aligned in a timing sequence dimension, there is a significant discontinuity of the sequence distribution of the image quality timing feature vector as a timing-related feature distribution when feature extraction is performed by an image quality timing feature extractor including a first convolution layer and a second convolution layer. In this way, when the image quality time sequence feature vector is classified by the classifier, the relevance between the local weight value distributions of the weight matrix of the classifier is also insufficient, and the training speed of the classifier is affected.

Based on the above, in the technical solution of the present application, each time the weight matrix iterates, the weight matrix is mapped

Feature affinity spatial affine learning is performed, expressed as:

wherein ,

representing the two norms of the weight matrix, i.e +.>

Maximum eigenvalue of>

Represents the kernel norm of the weight matrix, i.e. the sum of the eigenvalues of the weight matrix, and +.>

Is the scale of the weight matrix, i.e. width times height.

Here, the feature affinity spatial affine learning performs affine migration based on spatial transformation with relatively low-resolution information characterization by performing detailed structured information expression in a low-dimensional eigensubspace on high-resolution information characterization in a weight value distribution space of the weight matrix, thereby implementing super-resolution (e.g., weight-by-weight) activation of weight distribution local to each weight based on affinity (affinity) dense simulation between weight value characterization to enhance training speed of the classifier by enhancing correlation between each local weight distribution of the weight matrix. Therefore, the quality detection of the experimental data of the informationized detection laboratory can be accurately carried out so as to detect whether the experimental source data deviate or not, and the performance of the informationized detection laboratory equipment is evaluated, so that the normal operation of a laboratory electronic file rapid generation system is ensured.

Based on this, the application proposes an informationized detection laboratory electronic file rapid generation system, which comprises: the experimental data receiving module is used for receiving experimental data transmitted by laboratory equipment; and the file generation module is used for automatically braiding and formatting the experimental data based on a preset template and a specification to generate a laboratory electronic file.

Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.

Exemplary System

FIG. 1 is a block diagram of an informationized detection laboratory electronic file rapid generation system according to an embodiment of the present application. As shown in fig. 1, an informationized detection laboratory electronic document rapid generation system 300 according to an embodiment of the present application includes: an experimental data receiving module 310 for receiving experimental data transmitted by laboratory equipment; and a file generation module 320, configured to automatically weave and format the experimental data based on a preset template and specification to generate a laboratory electronic file.

Wherein the experimental data receiving module 310 is configured to receive experimental data transmitted by laboratory equipment; and a file generation module 320, configured to automatically weave and format the experimental data based on a preset template and specification to generate a laboratory electronic file.

Specifically, during operation of the rapid information detection laboratory electronic file generation system 300, the experimental data receiving module 310 is configured to receive experimental data transmitted by laboratory equipment. In the technical scheme of the application, whether the performance of the laboratory equipment is normal is evaluated by collecting experimental data for the laboratory equipment and analyzing.

Fig. 5 is a block diagram of an experimental data receiving module in an informationized detection laboratory electronic file rapid generation system according to an embodiment of the present application. As shown in fig. 5, the experimental data receiving module 310 includes: a data acquisition unit 3110 for acquiring experimental data at a plurality of predetermined time points within a predetermined period of time acquired by a laboratory apparatus, wherein the experimental data is image data; a quality evaluation unit 3120 for passing the image data of the respective predetermined time points through an image quality evaluator based on an encoder-decoder structure to obtain an image quality timing input vector composed of a plurality of image quality score values; a time sequence difference unit 3130 for calculating a difference between the image quality score values of each adjacent two positions in the image quality time sequence input vector to obtain an image quality time sequence difference input vector; an image quality timing information integration unit 3140, configured to concatenate the image quality timing input vector and the image quality timing difference input vector to obtain an image quality input vector; an image quality timing variation feature extraction unit 3150 for passing the image quality input vector through an image quality timing feature extractor comprising a first convolution layer and a second convolution layer to obtain an image quality timing feature vector; the device performance detecting unit 3160 is configured to pass the image quality timing feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether the performance of the laboratory device that provides the plurality of experimental data is normal.

Fig. 3 is a system architecture diagram of an inference module in an informationized detection laboratory electronic document rapid generation system according to an embodiment of the present application. As shown in fig. 3, in the network architecture, first, experimental data of a plurality of predetermined time points within a predetermined period of time acquired by a laboratory device are acquired by the data acquisition unit 3110, wherein the experimental data are image data; next, the quality evaluation unit 3120 passes the image data of each predetermined time point acquired by the data acquisition unit 3110 through an image quality evaluator based on an encoder-decoder structure to obtain an image quality timing input vector composed of a plurality of image quality score values; the timing difference unit 3130 calculates a difference between the image quality score values of each adjacent two positions in the image quality timing input vector obtained by the quality evaluation unit 3120 to obtain an image quality timing difference input vector; the image quality timing information integration unit 3140 concatenates the image quality timing input vector obtained by the quality evaluation unit 3120 and the image quality timing difference input vector calculated by the timing difference unit 3130 to obtain an image quality input vector; then, the image quality timing variation feature extraction unit 3150 passes the image quality input vector obtained by the image quality timing information integration unit 3140 through an image quality timing feature extractor including a first convolution layer and a second convolution layer to obtain an image quality timing feature vector; further, the device performance detecting unit 3160 passes the image quality time series feature vector obtained by the image quality time series change feature extracting unit 3150 through a classifier to obtain a classification result indicating whether the performance of the laboratory device that provided the plurality of experimental data is normal.

More specifically, during operation of the rapid information detection laboratory electronic file generation system 300, the data acquisition unit 3110 and the quality evaluation unit 3120 are configured to acquire experimental data at a plurality of predetermined time points within a predetermined period of time acquired by laboratory equipment, wherein the experimental data is image data. It will be appreciated that if the laboratory device malfunctions or the source data deviates due to laboratory device performance deviation, this will be apparent at the laboratory data end, and that the laboratory device performance detection based on the feature analysis of the image data will be based on taking the laboratory data as an example of the image data, considering that the laboratory device data collected in the time dimension are mostly presented in the form of images.

More specifically, during operation of the informative inspection laboratory electronic document rapid generation system 300, the quality assessment unit 3120 is configured to pass the image data of each predetermined point in time through an image quality assessor based on an encoder-decoder structure to obtain an image quality temporal input vector composed of a plurality of image quality score values. It is considered that the performance characteristic information about the laboratory device exists in the image data due to the respective predetermined time points, and if there is no deviation in the performance of the laboratory device, there is no deviation in the source data reflected in the laboratory device. Thus, the performance characteristics of the laboratory device are represented in the image quality characteristics of the image data at the respective predetermined time points. Based on this, in the technical solution of the present application, the image data at each predetermined time point is subjected to quality evaluation of the experimental data at each predetermined time point by an image quality evaluator based on an encoder-decoder structure to determine whether there is a deviation, thereby obtaining an image quality time sequence input vector composed of a plurality of image quality score values.

Fig. 6 is a block diagram of a quality assessment unit in an informationized detection laboratory electronic file rapid generation system according to an embodiment of the present application. As shown in fig. 6, the quality evaluation unit 3120 includes: an encoding subunit 3121 for extracting image features from the image data at the respective predetermined points in time using an encoder of the image quality evaluator, wherein the encoder of the image quality evaluator comprises a plurality of convolutional layers, to obtain a plurality of image features; and a decoding subunit 3122 for decoding the plurality of image features using a decoder of the image quality evaluator to obtain the image quality timing input vector comprised of a plurality of image quality decoded values, wherein the decoder of the image quality evaluator comprises a plurality of deconvolution layers.

More specifically, during operation of the rapid information detection laboratory electronic file generation system 300, the time series difference unit 3130 is configured to calculate a difference between the image quality score values of each two adjacent positions in the image quality time series input vector to obtain an image quality time series difference input vector. In consideration of the fact that the image quality score value of the experimental data of the laboratory equipment has a dynamic change rule in the time dimension, in order to accurately evaluate the image quality, dynamic change characteristics of the image quality of the experimental data in the time dimension are required to be extracted, in consideration of the fact that the change condition of the image quality score value of the experimental data in time sequence is weak, the weak change characteristics are small-scale change characteristic information relative to the image quality score value, if the time sequence dynamic change characteristic extraction of the image quality is carried out by absolute change information, not only the calculated amount is large, but also the small-scale weak change characteristics of the image quality score value in the time dimension are difficult to perceive, and the accuracy of subsequent classification is further affected. Therefore, in the technical scheme of the application, the dynamic change feature extraction of the image quality is comprehensively performed by adopting the time sequence relative change feature and the absolute change feature of the image quality score value. Specifically, first, the difference between the image quality score values of every adjacent two positions in the image quality time series input vector is calculated to obtain an image quality time series differential input vector.

More specifically, during operation of the rapid information detection laboratory electronic file generation system 300, the image quality timing information integration unit 3140 is configured to concatenate the image quality timing input vector and the image quality timing difference input vector to obtain an image quality input vector. Consider that there is an association between a time-series relative change characteristic and a time-series absolute change characteristic due to the image quality score value with respect to time-series dynamic change of the image quality in the experimental data. Therefore, in order to fully explore the dynamic change rule of the image quality in the time dimension so as to accurately detect the quality of experimental data, so that the laboratory equipment is subjected to performance accurate evaluation, in the technical scheme of the application, the image quality time sequence input vector and the image quality time sequence differential input vector are cascaded to obtain the image quality input vector.

More specifically, during operation of the rapid information detection laboratory electronic document generation system 300, the image quality temporal variation feature extraction unit 3150 is configured to pass the image quality input vector through an image quality temporal feature extractor comprising a first convolution layer and a second convolution layer to obtain an image quality temporal feature vector. In the technical scheme of the application, in order to obtain different mode change characteristic information presented by the image quality score under different time period spans, the image quality input vector is further processed through an image quality time sequence characteristic extractor comprising a first convolution layer and a second convolution layer to obtain the image quality time sequence characteristic vector. In particular, here, the first convolution layer and the second convolution layer have different feature receptive fields to perform feature mining on the image quality input vector using different one-dimensional convolution kernels, so as to extract dynamic multi-scale associated feature information of the image quality score value under different time spans, that is, time-series multi-scale dynamic change feature information of the image quality.

Fig. 7 is a block diagram of an image quality time-series variation feature extraction unit in an informationized detection laboratory electronic file rapid generation system according to an embodiment of the present application. As shown in fig. 7, the image quality timing variation feature extraction unit 3150 includes: a first image quality timing feature extraction subunit 3151 configured to input the image quality input vector into a first convolution layer of the image quality timing feature extractor to obtain a first image quality timing feature vector, where the first convolution layer has a first one-dimensional convolution kernel of a first length; a second image quality timing feature extraction subunit 3152 for inputting the image quality input vector into a second convolution layer of the image quality timing feature extractor to obtain a second image quality timing feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel of a second length, the first length being different from the second length; and a concatenation subunit 3153, configured to concatenate the first image quality timing feature vector and the second image quality timing feature vector to obtain the image quality timing feature vector.

More specifically, in the informatization detection laboratory electronic file rapid generation During operation of the system 300, the device performance detection unit 3160 is configured to pass the image quality timing feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether the performance of the laboratory device that provides the plurality of experimental data is normal. That is, after the image quality time series feature vector is obtained, it is further passed through a classifier as a classification feature vector to obtain a classification result, specifically, the image quality time series feature vector is processed using the classifier in the following formula:

wherein M is a weight matrix, +.>

To->

For the bias vector +.>

Is an image quality timing feature vector. Specifically, the classifier includes a plurality of fully connected layers and a Softmax layer cascaded with a last fully connected layer of the plurality of fully connected layers. In the classification processing of the classifier, multiple full-connection encoding is carried out on the classification feature vectors by using multiple full-connection layers of the classifier to obtain encoded classification feature vectors; further, the image quality temporal feature vector is input to a Softmax layer of the classifier, i.e. the encoded classification feature vector is classified using the Softmax classification function to obtain a classification label. In the technical solution of the present application, the label of the classifier includes a normal performance (first label) of the laboratory device providing the plurality of experimental data, and an abnormal performance (second label) of the laboratory device providing the plurality of experimental data, wherein the classifier determines to which classification label the classification feature vector belongs through a soft maximum function. It is noted that the first tag p1 and the second tag p2 do not include any artificial means The concept of calibration, which is not actually a concept of "whether the performance of the laboratory equipment providing the plurality of experimental data is normal" in the training process, is simply two kinds of classification labels and the probability that the output feature is under the two classification labels, i.e., the sum of p1 and p2 is one. Thus, the classification result of whether the performance of the laboratory device providing the plurality of experimental data is normal is actually converted into a classified probability distribution conforming to the natural law by classifying the tags, and the physical meaning of the natural probability distribution of the tags is essentially used instead of the linguistic text meaning of "whether the performance of the laboratory device providing the plurality of experimental data is normal". It should be understood that, in the technical solution of the present application, the classification label of the classifier is a control policy label for determining whether the performance of the laboratory device providing the plurality of experimental data is normal, so after the classification result is obtained, the quality detection of the image data can be accurately performed based on the classification result, that is, whether the deviation of the experimental source data occurs is detected, so as to accurately evaluate whether the performance of the laboratory device is normal.

Specifically, during the operation of the rapid information detection laboratory electronic file generating system 300, the file generating module 320 is configured to automatically weave and format the experimental data based on a preset template and a specification to generate a laboratory electronic file. It should be appreciated that laboratory electronic files are an important component of laboratory work, recording the purpose, process, results and conclusions of the experiment, as well as the instruments, materials and methods used. The rapid generation of laboratory electronic files can improve efficiency and quality. That is, in the technical solution of the present application, after the experimental data is obtained, the experimental data is further automatically woven and formatted to generate a laboratory electronic file to help laboratory manage and control each link of the experiment, including sample management, instrument management, data acquisition, data analysis, data reporting, and the like.

It will be appreciated that the image quality temporal feature extractor comprising the first and second convolution layers and the classifier need to be trained prior to the inference using the neural network model described above. That is, in the informationized detection laboratory electronic file rapid generation system of the present application, the system further comprises a training module for training the image quality time sequence feature extractor comprising the first convolution layer and the second convolution layer and the classifier. The training of deep neural networks mostly adopts a back propagation algorithm, and the back propagation algorithm updates the parameters of the current layer through errors transmitted by the later layer by using a chained method, which can suffer from the problem of gradient disappearance or more broadly, the problem of unstable gradient when the network is deep.

Fig. 2 is a block diagram of an informationized detection laboratory electronic file rapid generation system according to an embodiment of the present application. As shown in fig. 2, the informationized detection laboratory electronic file rapid generation system 300 according to an embodiment of the present application further includes a training module 400, which includes: a training data acquisition unit 410; training a quality assessment unit 420; training the timing difference unit 430; a training image quality timing information integration unit 440; a training image quality time-series variation feature extraction unit 450; a classification loss unit 460; model training unit 470.

The training data acquisition unit 410 is configured to acquire training data, where the training data includes training experiment data of a plurality of predetermined time points within a predetermined period of time, and a true value of whether the performance of the laboratory device is normal; the training quality evaluation unit 420 is configured to pass the training experimental data at each predetermined time point through the image quality evaluator based on the encoder-decoder structure to obtain a training image quality time sequence input vector composed of a plurality of training image quality score values; the training time sequence difference unit 430 is configured to calculate a difference between training image quality score values of every two adjacent positions in the training image quality time sequence input vector to obtain a training image quality time sequence difference input vector; the training image quality timing information integrating unit 440 is configured to concatenate the training image quality timing input vector and the training image quality timing difference input vector to obtain a training image quality input vector; the training image quality time sequence variation feature extraction unit 450 is configured to pass the training image quality input vector through the image quality time sequence feature extractor including the first convolution layer and the second convolution layer to obtain a training image quality time sequence feature vector; the classification loss unit 460 is configured to pass the training image quality timing feature vector through the classifier to obtain a classification loss function value; the model training unit 470 is configured to train the image quality timing feature extractor including the first convolution layer and the second convolution layer and the classifier based on the classification loss function value and propagating in a direction of gradient descent, where in each iteration of the training, a feature affinity space affine learning iteration is performed on a weight matrix of the classifier.

Fig. 4 is a system architecture diagram of a training module in an informationized detection laboratory electronic file rapid generation system according to an embodiment of the present application. As shown in fig. 4, in the system architecture of the rapid information detection laboratory electronic file generating system 300, in a training module 400, training data is first acquired through the training data acquisition unit 410, where the training data includes training experiment data at a plurality of predetermined time points within a predetermined period of time, and a true value of whether the performance of the laboratory device is normal; next, the training quality evaluation unit 420 passes the training experimental data of each predetermined time point acquired by the training data acquisition unit 410 through the encoder-decoder structure-based image quality evaluator to obtain a training image quality time sequence input vector composed of a plurality of training image quality score values; the training time sequence difference unit 430 calculates a difference between training image quality score values of every two adjacent positions in the training image quality time sequence input vector obtained by the training quality evaluation unit 420 to obtain a training image quality time sequence difference input vector; the training image quality timing information integrating unit 440 concatenates the training image quality timing input vector obtained by the training quality evaluating unit 420 and the training image quality timing difference input vector calculated by the training timing difference unit 430 to obtain a training image quality input vector; then, the training image quality time sequence variation feature extraction unit 450 passes the training image quality input vector obtained by the training image quality time sequence information integration unit 440 through the image quality time sequence feature extractor including the first convolution layer and the second convolution layer to obtain a training image quality time sequence feature vector; the classification loss unit 460 passes the training image quality time sequence feature vector obtained by the training image quality time sequence change feature extraction unit 450 through the classifier to obtain a classification loss function value; further, the model training unit 470 trains the image quality timing feature extractor including the first convolution layer and the second convolution layer and the classifier based on the classification loss function value and traveling in the direction of gradient descent, wherein in each round of the training, a feature affinity space affine learning iteration is performed on a weight matrix of the classifier.

In particular, in the technical solution of the present application, here, for the image quality input vector, since it is obtained by directly concatenating the image quality timing input vector and the image quality timing difference input vector, which are timing-series arrangement of an absolute value of image quality and a variation value of image quality, respectively, which are not perfectly aligned in a timing sequence dimension, there is a significant discontinuity of the sequence distribution of the image quality timing feature vector as a timing-related feature distribution when feature extraction is performed by an image quality timing feature extractor including a first convolution layer and a second convolution layer. In this way, when the image quality time sequence feature vector is classified by the classifier, the relevance between the local weight value distributions of the weight matrix of the classifier is also insufficient, and the training speed of the classifier is affected. Based on the above, in the technical solution of the present application, each time the weight matrix iterates, the weight matrix is mapped

Feature affinity spatial affine learning is performed, expressed as:

wherein ,

a weight matrix representing said classifier, +.>

Two norms of a weight matrix representing the classifier +.>

A kernel norm of a weight matrix representing the classifier, and +.>

Is the scale of the weight matrix of the classifier,

represents a logarithmic function value based on 2, < +.>

Representing multiplication by location +.>

And representing the weight matrix of the classifier after iteration. Here, the feature affinity spatial affine learning performs affine migration based on spatial transformations with relatively low-resolution information characterizations by detailed structured information expression in low-dimensional eigensubspaces for high-resolution information characterizations in weight-value distribution space of the weight matrix, thereby enabling super-resolution (e.g., weight-by-weight) activation of weight distribution local to individual weights based on affinity (affinity) dense simulations between weight-value characterizations, by increasing the weightThe relevance between the local weight distributions of the weight matrix improves the training speed of the classifier. Therefore, the quality detection of the experimental data of the informationized detection laboratory can be accurately carried out so as to detect whether the experimental source data deviate or not, and the performance of the informationized detection laboratory equipment is evaluated, so that the normal operation of a laboratory electronic file rapid generation system is ensured.

In summary, an informationized detection laboratory electronic document rapid generation system 300 is illustrated that receives experimental data transmitted by laboratory equipment and automatically weaves and formats the experimental data based on preset templates and specifications to generate laboratory electronic documents in such a way that laboratory electronic document rapid generation is performed.

As described above, the informationized detection laboratory electronic file rapid generation system according to the embodiments of the present application can be implemented in various terminal devices. In one example, the informative detection laboratory electronic file rapid generation system 300 according to embodiments of the present application may be integrated into a terminal device as a software module and/or hardware module. For example, the informative detection laboratory electronic file rapid generation system 300 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the informative laboratory electronic file rapid generation system 300 may also be one of a number of hardware modules of the terminal device.

Alternatively, in another example, the informative detection laboratory electronic document rapid generation system 300 and the terminal device may be separate devices, and the informative detection laboratory electronic document rapid generation system 300 may be connected to the terminal device through a wired and/or wireless network and transmit interactive information in a agreed data format.

Exemplary method

Fig. 8 is a flowchart of a method for quickly generating an electronic file in an informationized detection laboratory according to an embodiment of the present application. As shown in fig. 8, the method for quickly generating the electronic file of the informationized detection laboratory according to the embodiment of the application comprises the following steps: s110, receiving experimental data transmitted by laboratory equipment; and S120, automatically braiding and formatting the experimental data based on a preset template and a specification to generate a laboratory electronic file.

In one example, in the above method for quickly generating an electronic file in an informationized detection laboratory, the step S110 includes: acquiring experimental data of a plurality of preset time points in a preset time period acquired by laboratory equipment, wherein the experimental data are image data; passing the image data of each predetermined point in time through an image quality evaluator based on an encoder-decoder structure to obtain an image quality temporal input vector consisting of a plurality of image quality score values; calculating the difference between the image quality score values of every two adjacent positions in the image quality time sequence input vector to obtain an image quality time sequence differential input vector; cascading the image quality time sequence input vector and the image quality time sequence difference input vector to obtain an image quality input vector; passing the image quality input vector through an image quality temporal feature extractor comprising a first convolution layer and a second convolution layer to obtain an image quality temporal feature vector; and passing the image quality time sequence feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the performance of laboratory equipment for providing the plurality of experimental data is normal or not. Wherein passing the image data at each predetermined point in time through an image quality evaluator based on an encoder-decoder structure to obtain an image quality timing input vector comprised of a plurality of image quality score values, comprises: extracting image features from the image data at the respective predetermined points in time using an encoder of the image quality evaluator to obtain a plurality of image features, wherein the encoder of the image quality evaluator comprises a plurality of convolutional layers; and decoding the plurality of image features using a decoder of the image quality evaluator to obtain the image quality temporal input vector comprised of a plurality of image quality decoded values, wherein the decoder of the image quality evaluator comprises a plurality of deconvolution layers; and passing the image quality input vector through an image quality temporal feature extractor comprising a first convolution layer and a second convolution layer to obtain an image quality temporal feature vector, comprising: inputting the image quality input vector into a first convolution layer of the image quality timing feature extractor to obtain a first image quality timing feature vector, wherein the first convolution layer has a first one-dimensional convolution kernel of a first length; inputting the image quality input vector into a second convolution layer of the image quality timing feature extractor to obtain a second image quality timing feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel of a second length, the first length being different from the second length; cascading the first image quality time sequence feature vector and the second image quality time sequence feature vector to obtain the image quality time sequence feature vector; more specifically, passing the image quality timing feature vector through a classifier to obtain a classification result indicating whether or not the performance of the laboratory apparatus providing the plurality of experimental data is normal, comprising: processing the image quality temporal feature vector using the classifier to obtain a classification result with the formula:

Wherein M is a weight matrix, +.>

To->

For the bias vector +.>

Is an image quality timing feature vector.

In summary, an informationized detection laboratory electronic file rapid generation method according to embodiments of the present application is illustrated that receives experimental data transmitted by laboratory equipment and automatically weaves and formats the experimental data based on preset templates and specifications to generate laboratory electronic files, in such a way that laboratory electronic file rapid generation is performed.

Exemplary electronic device

Next, an electronic device according to an embodiment of the present application is described with reference to fig. 9.

Fig. 9 illustrates a block diagram of an electronic device according to an embodiment of the present application.

As shown in fig. 9, the electronic device 10 includes one or more processors 11 and a memory 12.

The processor 11 may be a Central Processing Unit (CPU) or other form of processing unit having data processing and/or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.

Memory 12 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, random Access Memory (RAM) and/or cache memory (cache), and the like. The non-volatile memory may include, for example, read Only Memory (ROM), hard disk, flash memory, and the like. On which one or more computer program instructions may be stored that may be executed by the processor 11 to implement the functions in the informative detection laboratory electronic file rapid generation system and/or other desired functions of the various embodiments of the present application described above. Various contents such as image quality time series differential input vectors may also be stored in the computer readable storage medium.

In one example, the electronic device 10 may further include: an input device 13 and an output device 14, which are interconnected by a bus system and/or other forms of connection mechanisms (not shown).

The input means 13 may comprise, for example, a keyboard, a mouse, etc.

The output device 14 may output various information including the classification result and the like to the outside. The output means 14 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, etc.

Of course, only some of the components of the electronic device 10 relevant to the present application are shown in fig. 9 for simplicity, components such as buses, input/output interfaces, etc. being omitted. In addition, the electronic device 10 may include any other suitable components depending on the particular application.

Exemplary computer program product and computer readable storage Medium

In addition to the methods and apparatus described above, embodiments of the present application may also be a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform the steps in the functions in the rapid generation of informative detection laboratory electronic files according to the various embodiments of the present application described in the "exemplary systems" section of this specification.

The computer program product may write program code for performing the operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server.

Furthermore, embodiments of the present application may also be a computer-readable storage medium, having stored thereon computer program instructions that, when executed by a processor, cause the processor to perform steps in the functions of the informative detection laboratory electronic file rapid generation method according to various embodiments of the present application described in the "exemplary systems" section of this specification.

The computer readable storage medium may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not limiting, and these advantages, benefits, effects, etc. are not to be considered as necessarily possessed by the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not intended to be limited to the details disclosed herein as such.

The block diagrams of the devices, apparatuses, devices, systems referred to in this application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.

It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent to the present application.

The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.

Claims

1. An informationized detection laboratory electronic file rapid generation system, comprising:

2. The rapid generation system of informationized inspection laboratory electronic files of claim 1, wherein the experimental data receiving module comprises:

the data acquisition unit is used for acquiring experimental data of a plurality of preset time points in a preset time period acquired by laboratory equipment, wherein the experimental data are image data;

a quality evaluation unit for passing the image data of each predetermined point in time through an image quality evaluator based on an encoder-decoder structure to obtain an image quality timing input vector composed of a plurality of image quality score values;

a time sequence difference unit, which is used for calculating the difference between the image quality score values of every two adjacent positions in the image quality time sequence input vector to obtain an image quality time sequence difference input vector;

the image quality time sequence information integration unit is used for cascading the image quality time sequence input vector and the image quality time sequence difference input vector to obtain an image quality input vector;

An image quality time sequence change feature extraction unit, configured to pass the image quality input vector through an image quality time sequence feature extractor including a first convolution layer and a second convolution layer to obtain an image quality time sequence feature vector;

and the device performance detection unit is used for passing the image quality time sequence feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the performance of the laboratory device for providing the plurality of experimental data is normal or not.

3. The rapid generation system of informationized inspection laboratory electronic files according to claim 2, characterized in that said quality assessment unit comprises:

an encoding subunit for extracting image features from the image data at the respective predetermined points in time using an encoder of the image quality evaluator to obtain a plurality of image features, wherein the encoder of the image quality evaluator comprises a plurality of convolutional layers; and

a decoding subunit for decoding the plurality of image features using a decoder of the image quality evaluator to obtain the image quality temporal input vector consisting of a plurality of image quality decoded values, wherein the decoder of the image quality evaluator comprises a plurality of deconvolution layers.

4. The rapid generation system of an informationized inspection laboratory electronic file according to claim 3, wherein said image quality time series variation feature extraction unit comprises:

a first image quality timing feature extraction subunit configured to input the image quality input vector into a first convolution layer of the image quality timing feature extractor to obtain a first image quality timing feature vector, where the first convolution layer has a first one-dimensional convolution kernel of a first length;

a second image quality temporal feature extraction subunit configured to input the image quality input vector into a second convolution layer of the image quality temporal feature extractor to obtain a second image quality temporal feature vector, where the second convolution layer has a second one-dimensional convolution kernel of a second length, the first length being different from the second length; and

and the cascading subunit is used for cascading the first image quality time sequence feature vector and the second image quality time sequence feature vector to obtain the image quality time sequence feature vector.

5. The rapid generation system of informationized inspection laboratory electronic files of claim 4, wherein the device performance inspection unit is configured to: processing the image quality temporal feature vector using the classifier to obtain a classification result with the formula:

, wherein ,/>

Is a weight matrix>

To->

For the bias vector +.>

Is an image quality timing feature vector.

6. The rapid information detection laboratory electronic file generation system of claim 5, further comprising a training module for training said image quality timing feature extractor comprising a first convolution layer and a second convolution layer and said classifier.

7. The rapid generation system of informationized detection laboratory electronic files of claim 6, wherein the training module comprises:

the training data acquisition unit is used for acquiring training data, wherein the training data comprise training experiment data of a plurality of preset time points in a preset time period and a true value of whether the performance of the laboratory equipment is normal or not;

a training quality evaluation unit for passing the training experiment data of each predetermined time point through the encoder-decoder structure-based image quality evaluator to obtain a training image quality time sequence input vector composed of a plurality of training image quality score values;

the training time sequence difference unit is used for calculating the difference between training image quality score values of every two adjacent positions in the training image quality time sequence input vector to obtain a training image quality time sequence difference input vector;

The training image quality time sequence information integration unit is used for cascading the training image quality time sequence input vector and the training image quality time sequence difference input vector to obtain a training image quality input vector;

a training image quality time sequence change feature extraction unit, configured to pass the training image quality input vector through the image quality time sequence feature extractor including the first convolution layer and the second convolution layer to obtain a training image quality time sequence feature vector;

the classification loss unit is used for passing the training image quality time sequence feature vector through the classifier to obtain a classification loss function value;

and a model training unit, configured to train the image quality time sequence feature extractor including the first convolution layer and the second convolution layer and the classifier based on the classification loss function value and through propagation in a direction of gradient descent, where in each iteration of the training, a feature affinity space affine learning iteration is performed on a weight matrix of the classifier.

8. The rapid generation system of informationized detection laboratory electronic files of claim 7, wherein said classification loss unit is configured to:

Processing the training image quality time sequence feature vector by using the classifier to obtain training classification results, and

and calculating a cross entropy loss function value between the training classification result and a true value of whether the performance of the laboratory equipment is normal or not as the classification loss function value.

9. The rapid generation system of informationized inspection laboratory electronic files of claim 8 wherein, in each iteration of the training, a feature affinity spatial affine learning iteration is performed on the weight matrix of the classifier with the following optimization formula;

wherein, the optimization formula is:

；

wherein ,

a weight matrix representing said classifier, +.>

Two norms of a weight matrix representing the classifier +.>

A kernel norm of a weight matrix representing the classifier, and +.>

Is the scale of the weight matrix of the classifier, < >>

Represents a logarithmic function value based on 2, < +.>

Representing multiplication by location +.>

And representing the weight matrix of the classifier after iteration.

10. The method for rapidly generating the electronic file of the informationized detection laboratory is characterized by comprising the following steps of:

Receiving experimental data transmitted by laboratory equipment; and