Data processing device

Abstract

To reduce the workload of an operator involved in creating templates such as analysis reports or system check reports.SOLUTION: A data processing device comprises: a report acquisition unit 31 that acquires an existing report including a plurality of report items related to a predetermined analytical instrument 10; a layout analysis unit 32 that executes, using a learned model constructed by machine learning, processing to generate layout data indicating the layout of the plurality of report items in the existing report from the existing report; and a template generation unit 33 that generates a template used for creating a new report on the basis of the layout data.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a data processing device. [Background technology]

[0002] In recent years, computers with dedicated software installed are often used to process and manage various data obtained from analytical instruments such as gas chromatographs, liquid chromatographs, and mass spectrometers. Such computer-based analytical instrument management systems often include a function for automatically generating analytical reports containing graphs such as chromatograms or mass spectra, or numerical data collected during analysis. An analytical report includes various report items, such as sample information, analytical conditions, and graphs or tables showing analytical results. Operators vary in the layout of report items within the analytical report. Therefore, in analytical instrument management systems with such a report creation function, an operator creates a template that defines the layout of various report items within the analytical report and stores it in the computer. The operator then specifies the template to be used for creating the analytical report and a data file containing the information to be reported, and issues an instruction to create the analytical report. The analytical report is then automatically created with the specified report items inserted in the specified positions within the template (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6645513 Summary of the Invention [Problem to be solved by the invention]

[0004] In an analytical instrument management system equipped with the above-described report creation function, when an analytical instrument to be managed is replaced with a new instrument or the dedicated software is changed to new software, it is necessary to create a new template so that the newly introduced analytical instrument or software can generate the same analysis report as before, which places a burden on the management and operation of the system. Also, when a new analytical instrument management system such as the above is introduced or when a new operator uses an existing system, there is a problem in that the work of creating a new template takes time and effort.

[0005] These problems occur not only in analysis reports that show the results of sample analysis by an analytical instrument, but also in system check reports that show the results of operational checks of analytical instruments.

[0006] The present invention has been made in view of the above points, and an object of the present invention is to reduce the workload on an operator involved in creating templates for the above-mentioned analysis reports, system check reports, and the like. [Means for solving the problem]

[0007] In order to solve the above problems, the data processing device according to the present invention comprises: a report acquisition unit that acquires an existing report including a plurality of report items related to a predetermined analytical instrument; a layout analysis unit that executes a process of generating layout data indicating a layout of the plurality of report items in the existing report from the existing report using a trained model constructed by machine learning; a template generation unit that generates a template to be used for creating a new report based on the layout data; It has the following characteristics. [Effects of the Invention]

[0008] According to the data processing device of the present invention, a template for creating a new report with the same layout as the existing report can be easily created without the operator having to perform complicated operations, thereby reducing the workload of the operator involved in creating the template. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a schematic configuration diagram of an analysis system including a data processing device according to an embodiment of the present invention. [Figure 2] FIG. 10 is a diagram showing an example of image data of an existing report in the embodiment. [Figure 3] FIG. 4 is a schematic diagram showing layout data of the existing report. [Figure 4] 10 is a flowchart showing a procedure for generating a report template in the embodiment. [Figure 5] FIG. 10 is a diagram showing an example of image data of a reference report in the embodiment. [Figure 6] FIG. 6 is a schematic diagram showing layout data generated based on the reference report of FIG. 5; [Figure 7] FIG. 7 is a schematic diagram showing a report template generated based on the layout data of FIG. 6. [Figure 8] FIG. 10 is a diagram showing another example of image data of a reference report in the embodiment. [Figure 9] FIG. 9 is a schematic diagram showing layout data generated based on the reference report of FIG. 8; [Figure 10] FIG. 10 is a schematic diagram showing a report template generated based on the layout data of FIG. 9. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, an embodiment of a data processing device according to the present invention will be described with reference to the accompanying drawings.

[0011] FIG. 1 is a schematic diagram of an analytical system including a data processing device 20 according to this embodiment. The analytical system according to this embodiment has a configuration in which an analytical device 10 that performs physical or chemical analysis on a sample and collects data, and a data processing device 20 that processes data output from the analytical device 10 are connected to each other via a dedicated cable or a network such as a LAN. For convenience of explanation, the analytical device 10 is assumed to be a liquid chromatograph, but the type of analytical device 10 in this embodiment is not particularly limited and may be, for example, a gas chromatograph, a mass spectrometer, a spectrophotometer, or the like. Furthermore, the number of analytical devices 10 connected to the data processing device 20 is not limited to one, and multiple analytical devices 10 may be connected.

[0012] The data processing device 20 comprises, as functional blocks, an analysis control unit 21 that controls the operation of the analytical device 10, an analysis data acquisition unit 22 (corresponding to the data acquisition unit in the present invention) that acquires data output from the analytical device 10, an analysis data processing unit 23 that performs predetermined analytical processing based on the data, a report creation unit 24 that creates an analysis report that describes the results of the analysis performed by the analytical device 10, and an automatic template generation unit 30 that generates a template (report template) used to create the analysis report. The automatic template generation unit 30 further comprises, as subordinate functional blocks, a reference report acquisition unit 31 (corresponding to the report acquisition unit in the present invention), a layout analysis unit 32, and a template generation unit 33.

[0013] The data processing device 20 further includes a storage unit 40. The storage unit 40 includes an analysis data storage unit 41 in which an analysis data file (described later) is stored, a report storage unit 42 in which an analysis report generated by the report creation unit 24 is stored, an analysis model storage unit 43 in which an analysis model used by the layout analysis unit 32 is stored, and a template storage unit 44 in which a report template generated by the automatic template generation unit 30 is stored. The analysis data file stored in the analysis data storage unit 41 is a file in which various information related to the analysis of a sample by the analytical instrument 10 is recorded. Here, "various information related to the analysis of a sample" includes, for example, information input by an operator when performing the analysis (e.g., information about the sample, analysis conditions in the analytical instrument 10, or data analysis conditions in the analytical data processing unit 23), data output from the analytical instrument 10 in connection with the analysis and acquired by the analytical data acquisition unit 22, various calculated values ​​obtained by processing the data in the analytical data processing unit 23, or the processing results thereof.

[0014] The data processing device 20 is actually a computer such as a personal computer equipped with a CPU, memory, and a large-capacity storage device such as a hard disk drive (HDD) or a solid-state drive (SSD). A predetermined program is pre-installed in the computer, and the above-mentioned functional blocks are realized in software form when the CPU executes this program. The functions of the storage unit 40 are also realized by the large-capacity storage device. The computer is connected to a display unit 51 for displaying various information, an operation unit 52 for a user to perform operations such as inputting various instructions, and a printing unit 53. The display unit 51 is, for example, a liquid crystal display, and the operation unit 52 is, for example, a keyboard, a pointing device such as a mouse, or a touch panel attached to the display unit 51.

[0015] The term "computer" as used herein refers not only to a standalone computer but also to a computer system consisting of multiple computers interconnected via a network. Therefore, for example, the functions of analysis control unit 21, analysis data acquisition unit 22, analysis data processing unit 23, report creation unit 24, analysis data storage unit 41, and report storage unit 42 may be assigned to a first computer, and the functions of automatic template generation unit 30 and analysis model storage unit 43 may be shared by a second computer (e.g., a cloud server) connected to the first computer directly or via a network such as a LAN or the Internet. In this case, the function of template storage unit 44 may be assigned to either the first computer or the second computer.

[0016] [Creating an analytical model] As described above, the analytical model storage unit 43 of the data processing device 20 according to this embodiment stores an analytical model (corresponding to a trained model in the present invention) used in the layout analysis unit 32. Note that the analytical model in this embodiment is a mathematical model that receives an image of an analytical report as input and outputs layout data indicating the structure of the analytical report, and is generated by machine learning. A method for generating this analytical model will be described below.

[0017] When creating an analysis model, the operator first prepares image data of various existing analysis reports (hereinafter referred to as existing reports) generated by the analytical instrument 10 for which the template is to be created, and layout data indicating the structure of each existing report. Here, layout data refers to data indicating what is written where in the existing report, i.e., data indicating the types of various report items included in the existing report and their layout. Figure 2 shows an example of image data of the existing report, and Figure 3 shows an example of layout data for the same existing report. As shown in Fig. 2, this existing report includes, as report items, the creation date and page number 61 of the report, a report title 62, sample information 63 which is information about the sample and analysis conditions, a chromatogram 64, and a peak table 65 which describes information about peaks included in the chromatogram 64. The corresponding layout data, as shown in Fig. 3, is data representing the layout of each report item in the existing report, i.e., the vertical and horizontal sizes of areas (writing areas) 71 to 75 in which each of the report items is described, the arrangement of the writing areas 71 to 75, and the types of report items described in the writing areas 71 to 75. A large number of sets of image data and layout data of such existing reports are prepared, and these are used as teacher data (also referred to as learning data or training data) to create an analysis model using a machine learning algorithm. Various methods are known as such supervised machine learning algorithms, such as support vector machines, random forests, GANs (generative adversarial networks), Adaboost, naive Bayes, k-nearest neighbors, or deep learning including neural networks, and any of these methods may be used. For example, in deep learning using a convolutional neural network (CNN), the filter coefficients of each of many convolutional layers, weighting coefficients in nodes and fully connected layers, and the like are learned as CNN parameters.

[0018] [Create a report template] Next, the creation of a report template, which is a characteristic operation of the analysis system according to this embodiment, will be described with reference to the flowchart in Figure 4. In the data processing device 20 according to this embodiment, the automatic template generation unit 30 generates a report template based on an image of an analysis report (hereinafter referred to as a reference report) prepared by the operator and used as a reference for creating the template. Note that the reference report does not necessarily have to be an analysis report created in connection with an actual analysis by the analytical device 10, but may also be a virtual analysis report created by the operator using, for example, word processing software, spreadsheet software, image editing software, or the like.

[0019] When creating a report template, first, the reference report acquisition unit 31 acquires image data of the reference report (step 1). At this time, the reference report acquisition unit 31 may acquire, for example, image data of an analysis report designated by an operator from image data of analysis reports stored in advance in the storage unit 40 as the image data of the reference report, or may acquire image data obtained by reading a predetermined analysis report with a scanner (not shown) connected to the computer as the image data of the reference report.

[0020] The image data of the reference report acquired by the reference report acquisition unit 31 is input to the layout analysis unit 32. The layout analysis unit 32 inputs the image data of the reference report to the above-mentioned analysis model, and acquires layout data indicating the structure of the reference report as an output from the analysis model (step 2).

[0021] Next, the template generation unit 33 generates a report template based on the layout data (step 3). Here, the report template is a model used to create a new analysis report (new report), and includes one or more information insertion fields into which information is inserted, and specifies the layout of each information insertion field, the vertical and horizontal sizes of each information insertion field, the types of report items to be inserted into each information insertion field, etc.

[0022] The above-described report template is created, for example, by converting the layout data generated in step 2 into a file format usable for creating a new report in the report creation unit 24. However, if the layout data generated in step 2 includes repetitions of the same report items, the report template reflects the repetition pattern. Specifically, for example, if the reference report in step 1 includes two or more pairs of one chromatogram 64 and one peak table 65 arranged vertically as shown in FIG. 5, the layout data generated in step 2 includes the same number of pairs of chromatogram entry areas 74 and peak table entry areas 75 as the number of chromatograms 64 and peak tables 65 in the reference report, as shown in FIG. 6. The report template generated in step 3 specifies that pairs of information insertion fields 84 into which chromatograms are inserted and information insertion fields 85 into which peak tables are inserted are arranged vertically, the number of which is the same as the number of chromatograms and peak tables to be reported, as shown in FIG. 7. Also, for example, if the reference report in step 1 is one in which a plurality of peak tables 65 are arranged below a plurality of (two in this figure) chromatograms 64 as shown in Fig. 8, the layout data generated in step 2 will have the same number of chromatogram entry areas 74 as the chromatograms 64 in the reference report, and below that, the same number of peak table entry areas 75 as the peak tables 65 in the reference report, as shown in Fig. 9. The report template generated in step 3 will specify that information insertion fields 84 into which chromatograms are inserted are arranged vertically in the number of chromatograms to be reported, and below that, information insertion fields 85 into which peak tables are inserted are arranged vertically in the number of peak tables to be reported, as shown in Fig. 10.Here, "the number of chromatograms and peak tables to be reported" and "the number of chromatograms to be reported" and "the number of peak tables to be reported" refer to the number of chromatograms and the number of peak tables included in the analysis data file designated as the subject of reporting by the new report when creating the new report described below.

[0023] With the above steps, the report template to be used for creating a new analysis report is completed and stored in the template storage unit 44 (step 4).

[0024] [Create analysis report] When creating a new analysis report using the report template created above, the operator performs a predetermined operation on the operation unit 52 to specify the report template to be used for report creation and the analysis data file containing the information to be included in the report, and then issues a command to create the analysis report. The report creation unit 24 then reads the specified report template from the template storage unit 44 and the specified analysis data file from the analysis data storage unit 41. The analysis report is then created by extracting information corresponding to the types of report items specified to be inserted into the information insertion fields 84 and 85 on the report template from the information contained in the analysis data file (i.e., the above-mentioned "various information related to sample analysis") and inserting the information into the information insertion fields 84 and 85. This allows the creation of a new analysis report containing the same report items as those described in the reference report, in the same layout as the reference report. The analysis report created as described above is stored in the report storage unit 42. The analysis report stored in the report storage unit 42 can be displayed on the screen of the display unit 51 or printed by the printing unit 53 in response to an instruction from an operator via the operation unit 52.

[0025] Here, an analysis report is created for a past analysis performed by the analytical device 10, but this is not limited to this. Prior to performing an analysis using the analytical device 10, a report template to be used for creating the analysis report and an analysis data file in which data for the analysis to be performed are stored can be specified, and an analysis report can be automatically created upon completion of the analysis.

[0026] While specific examples of embodiments of the present invention have been described above, the present invention is not limited to these embodiments and may be modified as appropriate within the spirit and scope of the present invention. For example, in the above embodiment, the present invention is applied to creating a template for an analysis report showing the results of a sample analysis performed by the analytical instrument 10 and creating a new analysis report using the template. However, the present invention is not limited to this, and can also be applied to creating a template for a report (i.e., a system check report) showing information related to an operation check performed to confirm whether the analytical instrument 10 is in a state where it can operate normally and creating a new system check report using the template. Here, "information related to the operation check" includes, for example, information input by an operator when performing the operation check (e.g., the operating conditions of the analytical instrument 10), data output from the analytical instrument 10 in connection with the operation check and acquired by the analytical data acquisition unit 22, and various calculated values ​​and processing results obtained by processing the data in the analytical data processing unit 23. When the present invention is applied to creating such a system check report template, an analysis model is generated by machine learning using image data and layout data of an existing system check report as training data and stored in the analysis model storage unit 43. As the reference report when creating a report template, an existing system check report that serves as a reference for creating the template is used.

[0027] In the above, the analysis model is configured to output layout data of a reference report in response to input of image data of the reference report, but this is not limiting, and the layout data of the reference report may be generated based on a reference report in a predetermined file format (e.g., PDF format) other than an image file. When generating such an analysis model, a set of data and layout data of an existing report in the predetermined file format is used as training data for machine learning, instead of the set of image data and layout data of the existing report as described above.

[0028] In addition, in the above embodiment, a program for realizing the functions of the data processing device 20 is pre-installed on the computer, but the program can also be provided by being stored on a non-transitory computer-readable recording medium.

[0029] [Aspect] It will be apparent to those skilled in the art that the above-described exemplary embodiments are examples of the following aspects.

[0030] (Item 1) A data processing device according to one aspect of the present invention comprises: a report acquisition unit that acquires an existing report including a plurality of report items related to a predetermined analytical instrument; a layout analysis unit that executes a process of generating layout data indicating a layout of the plurality of report items in the existing report from the existing report using a trained model constructed by machine learning; a template generation unit that generates a template to be used for creating a new report based on the layout data; It has the following characteristics.

[0031] According to the data processing device of paragraph 1, a template for creating a new report with the same layout as the existing report can be easily created without the operator having to perform complicated operations, thereby reducing the workload on the operator in creating templates for analysis reports, system check reports, etc.

[0032] (2) The data processing device according to paragraph 2 is the data processing device according to paragraph 1, further comprising: a data acquisition unit that acquires data output from the predetermined analytical device; a report creation unit that creates the new report by inserting the data acquired by the data acquisition unit or a processing result obtained by performing a predetermined process on the data into the template; and It has the following characteristics.

[0033] According to the data processing device of the second aspect, it is further possible to automatically create new reports using the templates, thereby further reducing the workload of the operator.

[0034] (Clause 3) The data processing device according to clause 3 is a data processing device according to clause 1 or 2, If the layout data indicates that a predetermined type of report item is described multiple times in the existing report, The template generation unit defines the template so that the number of report items of the predetermined type to be included in the new report is the same as the number of data that corresponds to the report items of the predetermined type and is specified as data to be included in the new report.

[0035] According to the data processing device of paragraph 3, when the predetermined type of report item is written multiple times in the existing report, it is possible to create a template that can create a new report with an appropriate layout regardless of the number of data corresponding to the report item to be reported.

[0036] (4) The program related to paragraph 4 is Computer, a report acquisition unit that acquires an existing report including a plurality of report items related to a predetermined analytical instrument; a layout analysis unit that executes a process of generating layout data indicating a layout of the plurality of report items in the existing report based on the existing report, using a trained model constructed by machine learning; a template generator that generates a template to be used for creating a new report based on the layout data; It functions as a [Explanation of symbols]

[0037] 10…Analytical equipment 20...Data processing device 21...Analysis control unit 22...Analysis data acquisition section 23...Analysis data processing section 24...Report Writing Department 30...Automatic template generation section 31...Reference Report Acquisition Department 32...Layout analysis section 33...Template generation unit 40...Storage section 41...Analysis data storage unit 42...Report memory section 43...Analysis model memory section 44...Template memory section 51...Display section 52...Operation unit 53…Printing Department 63...Sample information 64...chromatogram 65...Peak Table 71~75...Description area 84, 84...Information insertion field

Claims

1. a report acquisition unit that acquires an existing report including a plurality of report items related to a predetermined analytical instrument; a layout analysis unit that executes a process of generating layout data indicating a layout of the plurality of report items in the existing report from the existing report using a trained model constructed by machine learning; a template generation unit that generates a template to be used for creating a new report based on the layout data; A data processing device having:

2. Furthermore, a data acquisition unit that acquires data output from the predetermined analytical device; a report creation unit that creates the new report by inserting the data acquired by the data acquisition unit or a processing result obtained by performing a predetermined process on the data into the template; and 2. The data processing apparatus according to claim 1, comprising:

3. If the layout data indicates that a predetermined type of report item is described multiple times in the existing report, 3. The data processing device according to claim 1, wherein the template generation unit defines the template so that the number of report items of the predetermined type to be included in the new report is equal to the number of data items that correspond to the report items of the predetermined type and that are designated as data to be included in the new report.

4. Computer, a report acquisition unit that acquires an existing report including a plurality of report items related to a predetermined analytical instrument; a layout analysis unit that executes a process of generating layout data indicating a layout of the plurality of report items in the existing report based on the existing report, using a trained model constructed by machine learning; a template generator that generates a template to be used for creating a new report based on the layout data; A program that functions as a

Images