JP2023121141A - Learning support method and learning support program - Google Patents

Abstract

To promote a user to input high-quality teacher data for updating an estimation model.SOLUTION: A learning support method includes: acquiring a first chromatogram (step S2); displaying a first chromatogram image (step S6); acquiring, from a chromatogram DB, a second chromatogram that is the same as or similar to the first chromatogram, and second peak information for identifying a peak of the second chromatogram (step S4); displaying a second chromatogram image and a second peak information image (step S8); receiving a user's input of first peak information for identifying a peak of the first chromatogram (step S10); and learning an estimation model on the basis of the first chromatogram and the first peak information (step S14).SELECTED DRAWING: Figure 9

Description

The present disclosure relates to learning support methods and learning support programs.

WO2017/040487 (Patent Document 1) discloses a chromatographic system. This chromatographic system detects peaks by AI (Artificial Intelligence) using an estimated model, and performs qualitative or quantitative analysis of a sample based on the peaks.

Patent Literature 1 discloses a technique that allows a user to input training data for learning an estimation model. Specifically, a technique is disclosed in which a user views an unseparated chromatogram in which peaks are not separated and inputs information for designating peaks into a chromatographic system as teacher data.

WO2017/040487

Creating an estimation model may require annotation of a large amount of training data, and it is desired to suppress variations in annotations and improve the quality of the estimation model.

The present disclosure has been made to solve such problems, and the purpose thereof is to suppress variation in annotations and improve the quality of estimation models.

The learning support method of the present disclosure is a method of causing a computer to execute processing for supporting learning work of an estimation model used for detecting peaks in signal waveforms acquired by an analysis device. The learning aid method comprises obtaining a first signal waveform produced by an analyzer. The learning aid method comprises displaying the first signal waveform on a display device. In the learning support method, a second signal waveform having a high degree of similarity to the first signal waveform and second peak information for specifying a peak of the second signal waveform are extracted from a storage device storing a plurality of annotated signals. Prepare to get. The learning support method comprises displaying a second signal waveform and a second peak information image showing second peak information on a display device. The learning aid method comprises accepting user input of first peak information for identifying peaks of a first signal waveform. The learning assistance method comprises learning an estimation model based on the first signal waveform and the first peak information.

The learning support program of the present disclosure is a program that causes a computer to execute processing that supports learning work of an estimation model used to detect peaks in signal waveforms acquired by an analysis device. The learning support program causes the computer to acquire the first signal waveform created by the analyzer. The learning support program causes the computer to display the first signal waveform on the display device. The learning support program provides a computer with a second signal waveform having a high degree of similarity to the first signal waveform and a second peak for identifying a peak of the second signal waveform from a storage device that stores a plurality of annotated signals. Get information and get it done. The learning support program causes the computer to display the second signal waveform and the second peak information image representing the second peak information on the display device. The learning support program causes the computer to accept input by the user of first peak information for identifying peaks of the first signal waveform. The learning support program causes the computer to learn the estimation model based on the first signal waveform and the first peak information.

Advantageous Effects of Invention According to the present disclosure, it is possible to suppress variations in annotations and improve the quality of an estimated model.

It is a figure which shows the structural example of an analysis system. FIG. 4 is a diagram showing an example in which different teacher data are given to the same chromatogram; 1 is a block diagram showing the hardware configuration of a learning support device according to this embodiment; FIG. It is a functional block diagram of a learning support device. It is an example of a chromatogram DB. It is an example of a screen displayed on a display device. It is an example of a screen displayed on a display device. It is an example of a screen displayed on a display device. It is a flow chart which shows processing of a learning support device. 9 is a flowchart of processing of the learning support device of the second embodiment; 11 is a flowchart of processing of the learning support device of the third embodiment; It is an example of a screen displayed on a display device.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. The same or corresponding parts in the drawings are denoted by the same reference numerals, and the description thereof will not be repeated.

<First embodiment>
[Analysis system]
TECHNICAL FIELD The present disclosure relates to a technique for assisting learning of an estimation model used for detecting peaks in signal waveforms created by an analysis device. Analytical devices include, for example, a gas chromatograph (GC) device, a liquid chromatograph (LC) device, a mass spectrometer, a spectrophotometer, and an X-ray spectrometer.

For example, the signal waveform may be a chromatographic waveform or a mass spectral waveform. Moreover, when the analyzer is a spectrophotometer, the signal waveform is an absorption spectrum waveform. If the analyzer is an X-ray analyzer, the signal waveform will be an X-ray spectral waveform.

Also, the learning (learning process) of the estimation model (estimation model 121 described later) includes a process of newly generating (building) an estimation model that has not been constructed and a process of updating an already constructed estimation model. . "Updating the estimation model" includes processing for updating the parameters of the estimation model. An estimation model updated (optimized) by learning processing is also referred to as a “learned model”. Also, pre-learning estimation models and trained estimation models are collectively referred to as "estimation models".

In this embodiment, an analysis apparatus employing a liquid chromatograph will be described. FIG. 1 is a diagram showing a configuration example of an analysis system 100. As shown in FIG. It includes an analysis device 10 , a data analysis device 25 , an input device 61 , a display device 62 and a learning support device 30 . Data analysis device 25 and learning support device 30 are configured by, for example, an information processing device (for example, a PC (Personal Computer)). Although the data analysis device 25 and the learning support device 30 are shown separately in the example of FIG. 1, they may be integrated.

The input device 61 is, for example, a pointing device such as a keyboard or a mouse, and receives commands from the user. The display device 62 is composed of, for example, a liquid crystal (LCD) panel. The display device 62 displays various images. When a touch panel is used as the user interface, the input device 61 and the display device 62 are integrally formed. The input device 61 is connected to the data analysis device 25 and the learning support device 30 . The display device 62 is also connected to the data analysis device 25 and the learning support device 30 .

The data analysis device 25 has a control section 20 . The control unit 20 controls the analysis device 10 . Analyzer 10 includes mobile phase container 11 , pump 12 , injector 13 , column 14 and detector 15 . The mobile phase container 11 contains a mobile phase. The pump 12 sucks the mobile phase contained in the mobile phase container 11 and supplies it to the column 14 at a substantially constant flow rate (or flow rate).

The injector 13 injects a predetermined amount of sample liquid prepared in advance into the mobile phase at a predetermined timing according to an instruction from the control unit 20 . The injected sample liquid is introduced into the column 14 along with the flow of the mobile phase. Various components contained in the sample liquid are separated in the time direction and eluted while passing through the column 14 . In other words, the column 14 separates the components contained in the sample liquid according to their retention times.

Detector 15 detects components in the eluate eluted from column 14 . The detector 15 outputs a detection signal having an intensity corresponding to the amount of the component to the data analysis device 25 . The detector 15 is, for example, an optical detector such as a photodiode array (PDA) detector.

The data analysis device 25 has a data collection section 110 , a peak detection processing section 111 , and an analysis section 117 in addition to the control section 20 described above.

The data collector 110 samples the detection signal output from the detector 15 at predetermined time intervals and converts it into digital data. Data collection unit 110 stores the digital data in a predetermined storage area (not shown). This digital data is data indicating a chromatogram waveform (hereinafter also referred to as "chromatogram data").

The peak detection processing unit 111 uses AI (Artificial Intelligence) to estimate (derive) chromatogram peaks based on the chromatogram data collected by the data collection unit 110 .

In the present embodiment, peak detection processing section 111 has model storage section 114 and peak determination section 116 . The model storage unit 114 also stores an estimation model 121 (neural network) generated by machine learning, for example. This estimation model 121 is expressed by, for example, a predetermined function. The predetermined function is, for example, an EMG (Exponentially Modified Gaussian) function.

Also, the peak determination unit 116 inputs the chromatogram based on the chromatogram data collected by the data collection unit 110 to the estimation model 121 . The estimation model 121 outputs chromatogram peaks. In this way, the peak detection processing unit 111 estimates peaks of the chromatogram based on the chromatogram data collected by the data collection unit 110 and outputs the peaks to the analysis unit 117 .

The time at which a peak is observed (retention time) corresponds to the type of component. The chromatogram is sent to a data analysis device. The data analysis device identifies components from the retention times of peaks included in the chromatogram. This identification is also called "qualitative analysis".

Also, the peak height and peak area of the chromatogram correspond to the concentration or content of the component of the sample. The data analysis device specifies the concentration and content of the component of the sample from the peak height or area value included in the chromatogram. This determination is also called "quantitative analysis".

The analysis unit 117 obtains the position (time) of the peak top of the peak output from the peak determination unit 116 and the area value (or height) of the peak. The analysis unit 117 identifies the component from information on the position of each peak on the chromatogram. Also, the analysis unit 117 uses a calibration curve prepared in advance to calculate the content of each component from the peak area value (or height value). In this way, analysis unit 117 performs qualitative analysis and quantitative analysis of each component contained in the sample. The analysis unit 117 displays the qualitative analysis results and the quantitative analysis results on the display device 62 .

[Learning support device]
Next, the learning support device 30 will be described. As described above, the learning support device 30 optimizes the estimation model 121 in order to improve the accuracy of peak detection by the peak detection processing unit 111 . Moreover, in the present embodiment, the manufacturer may optimize the estimation model 121 in the manufacturing stage of the analysis system 100 . Alternatively, the analysis system 100 may be shipped to a user for the user to optimize the estimation model 121 . In this case, the user prepares learning data for optimizing the estimation model 121 and executes the annotation process himself/herself. Therefore, the user can generate the estimation model 121 desired by the user.

In general, the performance of the machine-learned estimation model 121 is not perfect, and is operated on the assumption that some error occurs in peak detection. As described above, in the present embodiment, the user can learn the estimation model 121 by himself/herself, so that the user's convenience can be improved.

Generally, the performance of the estimation model 121 greatly depends on the quality of learning data. In particular, it is preferable that various chromatograms can be covered and accurate teacher labels be assigned to the chromatograms.

In this embodiment, there are two methods for the user to optimize the estimation model 121 . A first method is a method in which the user corrects the peaks detected by the peak detection processing unit 111 . Specifically, the analysis system 100 displays the chromatogram image of the chromatogram and the peak information image of the peak information assigned to the chromatogram on the display device 62 . This peak information image corresponds to a "detected peak image" described later. A chromatogram image is an image showing a chromatogram. A peak information image is an image showing peak information. The peak information is information for specifying a chromatogram peak. Analysis system 100 accepts user modification of the displayed peak information.

A second method is a method in which the user designates a peak for a chromatogram newly collected by the data collection unit 110 (a chromatogram in which no peak is detected). Specifically, the analysis system 100 displays the chromatogram image on the display device 62 and does not display the peak information image. Analysis system 100 then receives input of peak information for the displayed chromatogram image. Also, the input peak information becomes a teacher label or teacher data for optimizing the estimation model 121 .

Thus, in both the first method and the second method, analysis system 100 accepts input of peak information from the user as teacher data. As a result of the user inputting peak information, the parameters of the estimation model 121 are updated using the input peaks and chromatograms as new learning data. The peak detection processing unit 111 can use the updated estimation model 121 . However, if the user makes annotations that are inconsistent (varied) with past annotations, the accuracy of the estimation model 121 may decrease.

Also, creating the estimation model 121 may require annotation of a large amount of teacher data. FIG. 2 is a diagram showing an example in which different teacher data are assigned to the same chromatogram (inconsistent annotation is performed). FIG. 2A is a diagram showing that the peaks of the chromatogram are widely specified by the user. FIG. 2B is a diagram showing that the peak of the chromatogram is narrowly specified by the user. In FIG. 2 and the figures showing chromatograms described later, the horizontal axis indicates time, and the vertical axis indicates signal intensity. Also, in the present embodiment, the peak information (teaching data) input by the user is the peak information 92A and the peak information 92B.

For example, the user inputs peak information 92A (see FIG. 2A) for the first chromatogram, and the user inputs peak information 92B (see FIG. 2B) for the second chromatogram. ) is entered. In this case, different peak information (teaching data) is input for the same chromatogram, which means that annotation variations occur. The quality of the estimation model 121 may be degraded when the annotations vary in this way.

Therefore, the learning support device 30 of the present embodiment encourages the user to input peak information so as to suppress variations in annotations (provide consistency with past peak information). Thereby, the learning support device 30 can support learning of the estimation model 121 by the user.

[Hardware configuration of learning support device]
FIG. 3 is a block diagram showing the hardware configuration of the learning support device 30 according to this embodiment. As shown in FIG. 3, the learning support device 30 includes a control device 51, a storage device 52, a media reading device 17, a display interface 18, and an input interface 26 as main hardware elements.

The control device 51 executes learning of the estimation model 121 as described later. The control device 51 includes, for example, a CPU (Central Processing Unit), an FPGA (Field Programmable Gate Array), a GPU (Graphics Processing Unit), and the like. Note that the control device 51 may be composed of at least one of a CPU, an FPGA, and a GPU, or may be composed of all of a CPU and an FPGA, an FPGA and a GPU, a CPU and a GPU, or a CPU, an FPGA, and a GPU. good too. Controller 51 may also be comprised of processing circuitry.

The storage device 52 includes a volatile storage area (for example, a working area) that temporarily stores program codes, work memory, etc. when the control device 51 executes an arbitrary program. For example, the storage device 52 is configured with a volatile memory device such as a DRAM (Dynamic Random Access Memory) or SRAM (Static Random Access Memory). In addition, storage device 52 includes non-volatile storage areas. For example, the storage device 52 is composed of a non-volatile memory device such as a hard disk or SSD (Solid State Drive).

In this embodiment, an example in which the volatile storage area and the nonvolatile storage area are included in the same storage device 52 is shown. It may be contained in different storage devices. For example, the control device 51 may contain a volatile memory area and the memory device 52 may contain a non-volatile memory area. The learning support device 30 may have a microcomputer including a control device 51 and a storage device 52 .

The storage device 52 stores an estimation model 121 , a control program 122 and a chromatogram DB (Data Base) 123 . The estimation model 121 includes a neural network and parameters used in processing in the neural network. The estimation model 121 is composed of a convolution neural network (CNN) or the like. The control program 122 is a program executed by the control device 51 .

The estimation model 121 includes at least a program capable of machine learning, and parameters are optimized (adjusted) by performing machine learning based on learning data (teacher data). The learning support device 30 transmits the optimized estimation model 121 to the data analysis device 25 . The data analysis device 25 updates the estimated model 121 stored in the model storage unit 114 to the transmitted estimated model 121 (optimized estimated model). Thus, the peak detection processing unit 111 can improve the estimation accuracy by estimating the peak using the updated estimation model 121 .

The media reader 17 receives a recording medium 130 such as a removable disk and acquires data stored in the recording medium 130 . The data is, for example, a control program. Also, the control program 122 may be stored in a recording medium 130 (eg, removable disk) and distributed as a program product. Alternatively, the control program 122 may be provided by an information provider as a downloadable program product via the so-called Internet. The control device 51 reads a program provided by the recording medium 130 or the Internet. The control device 51 stores the read program in a predetermined storage area (storage area of the storage device 52). The control device 51 executes a learning support process, which will be described later, by executing the stored control program 122 .

The recording medium 130 is not limited to a DVD-ROM (Digital Versatile Disk Read Only Memory), CD-ROM (compact disc read-only memory), FD (Flexible Disk), hard disk, magnetic tape, cassette tape, optical disk (MO ( Semiconductors such as Magnetic Optical Disc) / MD (Mini Disc) / DVD (Digital Versatile Disc), optical cards, mask ROM, EPROM (Electronically Programmable Read-Only Memory), EEPROM (Electronically Erasable Programmable Read-Only Memory), flash ROM, etc. It may be a medium that holds the program fixedly, such as a memory, etc. The recording medium 130 is a computer-readable non-temporary medium for the control program 122 and the like.

The display interface 18 is an interface for connecting the display device 62 and realizes input/output of data between the learning support device 30 and the display device 62 . The input interface 26 is an interface for connecting the input device 61 and realizes input/output of data between the learning support device 30 and the input device 61 .

[Functional Blocks of Learning Support Device]
FIG. 4 is a functional block diagram of the learning support device 30. As shown in FIG. As described above, the learning support device 30 has the control device 51 and the storage device 52 . The control device 51 further has an acquisition unit 32 , a processing unit 34 and an output unit 36 .

The acquisition unit 32 acquires information input by the user through the input device 61 . The information is, for example, the peak information described above. In addition, the data collection unit 110 transmits the chromatogram data to the learning support device 30 every time it collects a chromatogram. Acquisition unit 32 acquires chromatogram data from data collection unit 110 . The information (peak information and chromatogram data) acquired by the acquisition unit 32 is output to the processing unit 34 .

The processing unit 34 executes processing according to the type of information output from the acquisition unit 32 . When peak information is input by the acquisition unit 32, the parameters of the estimation model 121 are updated based on the peak information. Also, when chromatogram data is input by the acquisition unit 32, the chromatogram DB 123 is updated. The processing unit 34 also executes various other processes.

The output unit 36 outputs various signals or information. For example, the output unit 36 transmits image data of an image to be displayed on the display device 62 to the display device 62 . A display device 62 displays an image based on the image data. Moreover, the output unit 36 outputs the updated estimation model to the model storage unit 114 each time the estimation model 121 is updated. The model storage unit 114 stores the updated estimation model.

[Chromatogram DB]
FIG. 5 is an example of the chromatogram DB 123. FIG. The chromatogram DB 123 is stored in the storage device 52 as described above. The chromatogram DB 123 associates chromatogram IDs (identifications) with chromatogram data, stored peak information, chromatogram feature amounts indicated by the chromatogram data, and analysis results. Chromatogram data and stored peak information are also referred to as "annotated signals".

The chromatogram DB 123 also stores S (S is an integer equal to or greater than 1) chromatogram IDs. The chromatogram DB 123 stores chromatograms created in the past by the analyzer 10 or a device equivalent to the analyzer 10, feature amounts of the chromatograms, analysis results derived from the chromatograms, and the like. . In other words, the chromatogram DB 123 stores one or more (or multiple) annotated signals.

A chromatogram ID is information for identifying a chromatogram. The chromatogram data is data representing a chromatogram, and is digital data collected by the data collection unit 110 . The chromatogram data corresponds to the "stored signal waveform" of this disclosure.

The stored peak information is data for specifying peaks included in the chromatogram. The peak information is indicated by, for example, a second peak information image 253 shown in FIG. 6, which will be described later. The stored peak information is information indicating peaks detected by the peak detection processing unit 111 or peak information input by the user.

A chromatogram feature amount is a feature amount of a chromatogram indicated by chromatogram data. In the example of FIG. 5, the chromatogram features include the number of chromatogram peaks, the point-to-point slope, and the area value. The slope between two points refers to the slope of a line segment connecting the start point of a peak and the end point of the peak. The area value is the area value of the area surrounded by the line segment forming the peak of the chromatogram and the line segment due to the gradient between the two points. Also, the peak indicated by the chromatogram feature amount is the peak indicated by the stored peak information.

In addition, the two-point gradient and the area value are also peak feature amounts (hereinafter also referred to as "peak feature amounts"). There are as many peak feature amounts as there are peaks. For example, the chromatogram whose chromatogram ID is C1 has three peaks. Also, E11, E12, and E13 are shown as point-to-point slopes of the three peaks. Also, M11, M12, and M13 are shown as the area values of the three peaks.

The analysis result is a result derived based on peaks detected by the peak detection processing unit 111 using the current estimation model 121 as chromatogram peaks of chromatogram data corresponding to the analysis result. The analysis results include at least one of qualitative analysis results and quantitative analysis results. A qualitative analysis result is a result showing a component identified by a chromatogram. A quantitative analysis result is a result showing the amount of the component. As a modification, the analysis results may include the qualitative analysis results but not the quantitative analysis results.

In the example of FIG. 5, chromatogram data D1, number of peaks G1, gradient E1 between two points, area value M1, qualitative analysis result P1, and quantitative analysis result Q1 correspond to chromatogram ID C1. is associated with the stored peak information R1.

Thus, the chromatogram DB 123 stores at least one stored signal waveform created by the analyzer and at least one stored peak information for identifying each peak of the at least one stored signal waveform. .

[Learning support]
Next, a method of learning support for the user by the learning support device 30 of the present embodiment will be described. The learning support device 30 performs learning support for the user by displaying various images on the display area 62A of the display device 62 . 6 to 8 are diagrams showing examples of the various images of this embodiment. In this embodiment, the image in FIG. 6 is displayed, then the image in FIG. 7 is displayed, and then the image in FIG. 8 is displayed.

When the user performs a predetermined operation on the input device 61, the mode of the analysis system 100 is shifted to the learning mode. In the learning mode, the estimation model 121 is updated by the user inputting first peak information, which will be described later.

As shown in FIGS. 6 to 8, the display area 62A includes an edit area 131A and a past result area 132A adjacent to the edit area 131A. In other words, the editing area 131A and the past result area 132A are set on the same screen. In the editing area 131A, an image (hereinafter also referred to as a "first chromatogram image 211") of a chromatogram derived by the analyzer 10 (hereinafter also referred to as a "first chromatogram") is displayed. be. The first chromatogram corresponds to the "first signal waveform" of the present disclosure. The first chromatogram image 211 corresponds to the "first waveform image" of the present disclosure. Upon obtaining the chromatogram data collected by the data collection unit 110 , the learning support device 30 displays the chromatogram image of the chromatogram data as the first chromatogram image 211 .

In addition, when acquiring the first chromatogram, the learning support device 30 identifies a second chromatogram that is identical or similar to the first chromatogram from the chromatogram DB (FIG. 5), and the identified second chromatogram to get A specific example technique for a second chromatogram that is the same or similar to the first chromatogram will now be described.

The learning support device 30 calculates a first similarity between each of the S stored chromatogram data stored in the chromatogram DB and the first chromatogram. That is, the learning support device 30 calculates S first similarities. The first similarity indicates the degree of similarity between one piece of stored chromatogram data and the first chromatogram. The greater the similarity between one piece of stored chromatogram data and the first chromatogram, the greater the value of the first similarity. In this embodiment, the first similarity is expressed in %. Also, the first similarity of the second chromatogram that is the same as the first chromatogram is 100%.

The first similarity is calculated from the following two viewpoints. As a first aspect, the learning support device 30 combines the feature amount of the first chromatogram (hereinafter also referred to as "first feature amount") and the feature amount (first feature amount) of the stored chromatogram. A first similarity is calculated based on the feature amount of the same type (hereinafter also referred to as "second feature amount").

In the present embodiment, the first feature amount is a feature amount derived based on peaks detected by the peak detection processing unit 111 using the current estimation model 121 as the peaks of the first chromatogram. In the present embodiment, it is the feature quantity of the same type as the second feature quantity of the storage chromatogram of FIG. Specifically, the first feature amount and the second feature amount are the number of peaks, the gradient between two points, and the area value. Note that the first feature amount and the second feature amount may be of the same type, and may be of another type. Other types may be at least one of peak width, peak separation, peak leading, peak tailing, and the like.

As a second aspect, the learning support device 30 calculates the first similarity based on the analysis result (derived) indicated by the first chromatogram and the analysis result indicated by the storage chromatogram. Here, the analysis results indicated by the storage chromatogram include qualitative analysis results and quantitative analysis results corresponding to the storage chromatogram, as shown in FIG. Also, the analysis results indicated by the first chromatogram are the qualitative analysis results and quantitative analysis results indicated by the first chromatogram (that is, analysis results of the same type as the analysis results indicated by the storage chromatogram).

For example, the learning support device 30 compares the qualitative analysis result indicated by the first chromatogram with the qualitative analysis result corresponding to the stored chromatogram, and if both qualitative analysis results are different, the first similarity is set to " 0”. On the other hand, if both the qualitative analysis results are the same, the learning support device 30 calculates the degree of similarity between the two quantitative analysis results (hereinafter also referred to as "first provisional similarity"). Further, the learning support device 30 calculates a second provisional similarity based on the above-described first feature amount and the above-described second feature amount. The second provisional similarity is calculated using, for example, a correlation coefficient. Also, the second provisional similarity may be calculated as a similarity between the shape of the first chromatogram and the shape of the stored chromatogram.

Then, the learning support device 30 calculates the first similarity based on the first temporary similarity and the second temporary similarity. In this way, the learning support device 30 integrates the similarity calculated from the first viewpoint (first similarity) and the similarity calculated from the second viewpoint (first similarity), S (the number of chromatogram IDs stored in the chromatogram DB) first similarities are calculated.

As a modification, the learning support device 30 may calculate the first similarity from either the first viewpoint or the second viewpoint.

The learning support device 30 identifies a first similarity that is equal to or greater than the first threshold from the S calculated first similarities. The learning support device 30 determines the stored chromatogram that is the specified first degree of similarity as the second chromatogram. Here, the first threshold is a predetermined threshold, for example, 70%. In other words, the learning support device 30 can determine a chromatogram (second chromatogram) having a degree of similarity of 70% or more with the first chromatogram in terms of the first feature amount and the analysis result. Furthermore, the learning support device 30 acquires peak information (second peak information) corresponding to the chromatogram ID of the second chromatogram by referring to the chromatogram DB 123 .

When learning support device 30 acquires the first chromatogram, the second chromatogram, and the second peak information, as shown in FIG. A second chromatogram image 212 of two chromatograms and a second peak information image 253 of the second peak information are displayed. In the example of FIG. 6, the second chromatogram images 212A, 212B, 212C, 212D of each of the four second chromatograms and the second peak information image 253 of each of the four second chromatogram images 212 are displayed. It is In the following, the four second chromatogram images 212A, 212B, 212C, 212D are also referred to as "second chromatogram images 212".

As described above, the second chromatogram image 212 is the image of the second chromatogram having a high degree of similarity (greater than or equal to the first threshold) with the first chromatogram. Therefore, the user can recognize the peaks of the second chromatogram by viewing the second peak information image 253 of the chromatogram that is highly similar to the first chromatogram.

As shown in FIG. 6, in a state where the first chromatogram image 211 is displayed, the learning support device 30 displays peak information for specifying peaks of the first chromatogram indicated by the first chromatogram image 211. Accepts input from the user. The peak information corresponds to "first peak information" of the present disclosure.

FIG. 7 shows an image when the user inputs the first peak information. In the example of FIG. 7, an image representing the input first peak information is displayed as the first peak information image 220. In the example of FIG. The learning support device 30 updates the parameters of the estimation model 121 so that the peak detection processing unit 111 detects the peak specified by the input first peak information as the peak of the first chromatogram.

Thus, in this embodiment, the user can input the first peak information consistent with the past second peak information (the second peak information displayed in the past result area 132A). That is, for example, input of different peak information (see FIG. 2) despite the same chromatogram is suppressed. The learning support device 30 can then update the estimation model 121 based on the first peak information. Therefore, from the viewpoint of improving the accuracy of specifying the peak of the first signal waveform, the accuracy of updating the estimation model 121 can be improved.

In addition, the learning support apparatus 30 calculates the first similarity between the first chromatogram and each of the S stored chromatograms, and stores the stored chromatograms having the first similarity equal to or higher than the first threshold as the first Acquire as 2 chromatograms. Then, as shown in FIG. 6, the learning support device 30 displays the first chromatogram image 211, the acquired second chromatogram image 212 of the second chromatogram, and the second peak information image 253. FIG.

After that, as shown in FIG. 7, the learning support device 30 receives the input of the first peak information. Therefore, while visually recognizing the second chromatogram that is the same as or similar to the first chromatogram and was created in the past, and the second peak information image 253 that indicates the peak of the second chromatogram, the first You can enter the first peak information of the chromatogram. Therefore, the learning support device 30 can prompt the user to input teacher data (first peak information) in which variations in annotations are suppressed. As a result, the learning support device 30 can improve the quality of the estimation model by suppressing variations in annotations.

Further, the learning support device 30 calculates the first similarity based on the first feature amount of the first chromatogram and the second feature amount of the same type as the first feature amount of the stored chromatogram. Therefore, the learning support device 30 can calculate the first similarity with a relatively simple calculation.

Further, the learning support device 30 calculates the first similarity based on the analysis result indicated by the first chromatogram and the analysis result indicated by the storage chromatogram. Therefore, the learning support device 30 can calculate the first similarity with a relatively simple calculation.

6 and 7, the learning support device 30 displays the first similarity of the second chromatogram in association with the second chromatogram image 212 showing the second chromatogram. In the examples of FIGS. 6 and 7, a similarity image 271 indicating the first similarity is displayed. In the examples of FIGS. 6 and 7, for example, an image of "98%" is displayed as the similarity image 271 in association with the second chromatogram image 212A. Therefore, the user can visually recognize the first similarity of the second chromatogram.

Furthermore, the learning support device 30 displays the second feature amount image 275 in association with the second chromatogram image 212 in the past result area 132A. Here, the second feature amount image 275 indicates the feature amount of the peak specified by the second peak information indicated by the second peak information image 253 . The second feature amount image 275 includes an image 272 representing the gradient between two points and an image 273 representing the area value. Also, the slope between two points and the area value are feature amounts of each of one or more peaks included in the second chromatogram. In the examples of FIGS. 6 and 7, the chromatogram of second chromatogram image 212A contains three peaks. Corresponding to the second chromatogram image 212A, the point-to-point gradients A1, A2, S3 of each of the three peaks are displayed, and the area values B1, B2, B3 of each of the three peaks are displayed. is displayed.

Further, in the present embodiment, the second chromatogram includes a high similarity chromatogram and a low similarity chromatogram having a lower first degree of similarity than the high similarity chromatogram. For example, in the example of FIG. 7, a second chromatogram image 212A of a chromatogram with a first similarity of 98% is displayed as an example of a highly similar chromatogram. Also, as an example of a low similarity chromatogram, a second chromatogram image 212D of a chromatogram with a first similarity of 80% is displayed. Furthermore, in the past result area 132A, the second chromatogram image 212A showing the high similarity chromatogram is displayed with priority over the chromatogram image 212B showing the low similarity chromatogram. In this way, the learning support device 30 displays the second chromatogram image 212 and the second peak information image 253 with priority according to the first similarity. In the examples of FIGS. 6 and 7, high similarity chromatograms are displayed above low similarity chromatograms.

Therefore, the user can preferentially view the second chromatogram, which has a high degree of similarity with the first chromatogram, over the second chromatogram, which has a low degree of similarity. Therefore, the learning support device 30 can easily visually recognize the second peak information of the second chromatogram that has a high degree of similarity with the first chromatogram.

In the example of FIG. 7, the second peak information image 253 is a linear image (hereinafter also referred to as a "line image"). Also, the area surrounded by the second chromatogram image 212 and the second peak information image 253 is the area showing the peaks of the second chromatogram.

In addition, the learning support device 30, in a state in which the first chromatogram image 211, the second chromatogram image 212, and the second peak information image 253 are displayed, Accepts input (designation). In this embodiment, a cursor 217 of the input device 61 (mouse) is displayed. The user can specify the first point 201 by moving the cursor 217 to a desired position and clicking the mouse. Also, the user can specify the second point 202 by moving the cursor 217 to another desired position and clicking the mouse. When the first point 201 and the second point 202 are specified, the learning support device 30 displays a line image connecting the first point 201 and the second point 202 as the first peak information image 220 . Then, the learning support device 30 selects the area surrounded by the line connecting the first point 201 and the second point 202 and the line included in the first chromatogram image 211 as the first peak information image 220 . It is recognized as a peak area (teaching data) specified by peak information. According to such a configuration, by designating the first point 201 and the second point 202 of the first chromatogram image 211, the first peak information can be input. Therefore, the user can intuitively input the first peak information, thereby improving convenience for the user. The line image connecting the first point 201 and the second point 202 in FIG. 7 is also referred to as the "peak baseline".

FIG. 8 is a diagram showing a screen after the user inputs the first peak information. As shown in FIG. 8, when the first peak information is input, the learning support device 30 calculates a first feature amount and displays a first feature amount image 231 of the first feature amount. Here, the first feature amount image 231 indicates the feature amount of the peak specified by the first peak information indicated by the first peak information image 220 . In the example of FIG. 8, the first feature amount image 231 is an image showing the two-point gradient of this peak and the area value of this peak. In the example of FIG. 8, "X1" is displayed as the value of the gradient between two points, and "Y1" is displayed as the area value.

As described above, as shown in FIGS. 6 to 8, the learning support device 30 displays the second feature amount image 275 in the past result area 132A. Thus, the learning support device 30 displays the first feature image 231 and the second feature image 275 . Therefore, after inputting the first peak information, the user can view the peak indicated by the first peak information from the viewpoint of the feature amount (the gradient between two points and the area value in the present embodiment) in the past. peak can be compared.

Also, the first feature image 231 is displayed for the number of peaks specified by the first peak information input by the user. For example, when the number of peaks specified by the first peak information input by the user is "3", the first feature image 231 of each of the three peaks is displayed.

[flowchart]
FIG. 9 is a flow chart showing the processing of the learning support device 30. As shown in FIG. In step S2, the learning support device 30 acquires the first chromatogram. Next, in step S4, the learning support device 30 acquires the second chromatogram and the second peak information. Step S4 includes step S42, step S44 executed after step S42, step S46 executed after step S44, step S48 executed after step S46, and step S48 executed after step S48. and step S50.

In step S42, the learning support device 30 extracts the first feature amount of the first chromatogram and the second feature amount of the stored chromatogram. As described above, the first feature quantity and the second feature quantity are the number of peaks, the point-to-point gradient, and the area value.

In step S44, the learning support device 30 extracts the first analysis result indicated by the first chromatogram and the second analysis result indicated by the storage chromatogram. The first analysis result and the second analysis result are the qualitative analysis result and the quantitative analysis result.

In step S46, the learning support device 30 calculates a first similarity between the first chromatogram and the stored chromatogram. The learning support device 30 calculates the first similarity based on the first feature amount, the second feature amount, the first analysis result, and the second analysis result.

In step S48, the learning support device 30 refers to the chromatogram DB 123 and acquires a stored chromatogram whose first similarity is greater than or equal to the first threshold as a second chromatogram.

In step S50, the learning support device 30 refers to the chromatogram DB 123 and acquires stored peak information corresponding to the second chromatogram as second peak information.

When the process of step S4 is finished, in step S6, the learning support device 30 displays the first chromatogram image 211 of the first chromatogram acquired in step S2.

Next, in step S<b>8 , the learning support device 30 displays the second chromatogram image 212 and the second peak information image 253 . Step S8 includes step S82 and step S84 executed after step S82.

In step S82, the learning support device 30 displays the second chromatogram image 212 and the second peak information image 253 with priority according to the first similarity. In step S84, the similarity image 271 and the second feature amount image 275 are displayed.

As described above, the image shown in FIG. 6 is displayed by executing the processes of steps S2 to S8.

Next, in step S10, the learning support device 30 determines whether or not the user has input the first peak information. Step S10 includes the process of step S102. In step S102, learning support device 30 determines whether first point 201 and second point 202 have been input by the user. Learning support device 30 repeats the process of step S102 until first point 201 and second point 202 are input. If YES is determined in step S102, the process proceeds to step S12.

At step S12, the learning support device 30 displays the first peak information image of the first peak information determined to have been input at step S10. Step S12 includes step S122. In step S<b>122 , the learning support device 30 displays the first peak information image 220 and the first feature amount image 231 . Note that the image shown in FIG. 7 is displayed by the learning support device 30 displaying the first peak information image 220 in step S122. Further, in step S122, the learning support device 30 displays the first feature amount image 231, thereby displaying the image shown in FIG.

Next, in step S13, the learning support device 30 determines whether or not the user has performed an end operation. A termination operation is an operation performed by the user on the input device 61 . The end operation is, for example, the user's operation on the end button (not shown) displayed on the screens shown in FIGS.

For the first chromatogram image 211, the user can input one or more first peak information. When the user finishes inputting the first peak information, the user performs an end operation. If NO is determined in step S13, the process returns to step S102. On the other hand, if YES is determined in step S13, the process proceeds to step S14.

In step S14, the learning support device 30 learns the estimation model 121 based on the first chromatogram acquired in step S2 and the first peak information determined to be input in step S10. Also, the estimation model 121 may be learned with a plurality of sets of chromatograms and peak information.

As shown in FIG. 9, the learning support device 30 accepts input of the first peak information by the user after steps S6 and S8. As described above, step S6 is processing for displaying the first chromatogram image 211 on the display device 62 . Step S8 is a process of displaying the second chromatogram image 212 and the second peak information image 253 on the display device 62. FIG.

With the configuration as shown in FIG. 9, the user can view the second chromatogram similar to the first chromatogram and created in the past, and the second peak information indicating the peak of the second chromatogram. , can input the first peak information of the first chromatogram. Therefore, it is possible to improve the user's convenience in inputting the first peak information.

<Second embodiment>
FIG. 10 is a flow chart of processing of the learning support device 30 of the second embodiment. In FIG. 10, the learning support device 30 executes the process of step S6 after completing the process of step S2. Next, the learning support device 30 executes the process of step S10. Here, the user inputs the first peak information while the first chromatogram image 211 is displayed but the second chromatogram image 212 and the second peak information image 253 are not displayed. . When the process of step S10 is executed, the processes of steps S12 and S13 are executed. YES is determined in step S13. The process proceeds to step S4A. Step S4A differs from step S4 in that step S48 is replaced with step S52 and step S54.

In step S52, the learning support device 30 stores S pieces of each of the first peak information input in step S10 and the S pieces of stored peak information R (see FIG. 5) stored in the chromatogram DB 123. A degree of similarity (hereinafter also referred to as a “second degree of similarity”) is calculated. The second degree of similarity indicates the degree of similarity between one piece of stored peak information and the first peak information. The second degree of similarity takes a larger value as the piece of stored peak information and the first peak information are more similar. For example, the learning support device 30 calculates a parameter (for example, a correlation coefficient) regarding each of the S pieces of stored peak information and the first peak information as the second degree of similarity.

Next, in step S54, the learning support device 30 identifies a peak from the stored peak information in which the first similarity is equal to or greater than the first threshold and the second similarity is equal to or greater than the second threshold. Acquire the stored chromatogram as the second chromatogram. Here, the second threshold is a predetermined threshold. That is, in step S54, a second chromatogram having similar first peak information and chromatogram is acquired. Next, in step S50, second peak information corresponding to the second chromatogram obtained in step S54 is obtained.

In the example of FIG. 10, the learning support device 30 executes the process of step 4A after steps S6 and S10. Step S6 is processing for displaying the first chromatogram image 211 on the display device 62 . Step S10 is a process of accepting input by the user of the first peak information. Step 4A is a process of acquiring the second chromatogram and the second peak information from the chromatogram DB 123. FIG. Further, step S4A includes step S46, step S52 and step S54.

Step S52 is processing for calculating the first degree of similarity, and step S54 is processing for calculating the second degree of similarity. A step S56 converts the stored chromatogram whose peak is specified by the stored peak information whose first similarity is equal to or higher than the first threshold and whose second similarity is equal to or higher than the second threshold to the second chromatogram. It is a process to acquire as a gram.

According to the second embodiment, the learning support device 30 identifies peaks from memory peak information in which the first similarity is the first threshold or more and the second similarity is the second threshold or more. The stored chromatogram obtained can be displayed as the second chromatogram. Therefore, the user can confirm such a second chromatogram.

<Third Embodiment>
FIG. 11 is a flow chart of processing of the learning support device 30 of the third embodiment. In FIG. 11, the process of step S150 is executed after the process of step S6 of FIG. The process of step S150 is a process of displaying the detected peak image. The detected peak image is peak information (hereinafter also referred to as "detected peak information") for identifying the peaks of the first chromatogram (first chromatogram acquired in step S2) detected using the current estimation model 121. is called.). That is, the temporary peak information of the first chromatogram acquired in step S2 is displayed as the detected peak image. Thus, the first chromatogram image and the detected peak image (temporary peak image of the first chromatogram) are displayed by executing the processes of steps S6 and S150. After that, the processes of steps S8 and S10 are executed.

According to the third embodiment, the user can input the first peak information while referring to the detected peak information image. Furthermore, the user can input the first peak information while referring to the detected peak information image and the second peak information image. Therefore, user convenience can be improved.

Note that in step S102 (step S10), when the user determines that the first peak information image may remain the detected peak image, the user performs a predetermined operation (for example, an operation of pressing an OB button (not shown)). As a result, YES is determined in step S102, and the process proceeds to the next step S122. Further, in step S102 (step S10), if the user wishes to modify the detected peak image, he or she newly inputs the first peak information (specifies the first point 201 and the second point 202) to correct the detected peak image. YES is determined in step S102, and the process proceeds to the next step S122.

[Modification]
(1) In the process of FIG. 11, the learning support device 30 performs the process in the order of steps S2, S6, S150, S4A (see FIG. 10), S8, S10, and S12. good too. In step S52 of step S4A, S second similarities between the detected peak information and the S stored chromatograms are calculated. Therefore, in step S54, the learning support device 30 identifies a peak based on the stored peak information in which the first similarity is equal to or greater than the first threshold and the second similarity is equal to or greater than the second threshold. Acquire the stored chromatogram as the second chromatogram. That is, the first chromatogram to which the detected peak information is added and the second chromatogram image having the same or similar chromatogram and peak information are displayed. The learning support device 30 can improve convenience for the user by displaying such a second chromatogram image.

(2) In the above embodiment, the configuration in which the user inputs the first peak information by designating the first point 201 and the second point 202 has been described. However, the method of inputting the first peak information may be another method. For example, the user may input coordinates for specifying a desired peak in the displayed first chromatogram image 211 .

(3) In the example of FIG. 7, the configuration in which the user inputs the first point 201 and the second point 202 to input the "peak baseline (first peak information image 220)" has been described. However, in the baseline, the learning support device 30 may not be able to identify the peak.

FIG. 12 is a diagram showing an example of the first chromatogram image 211 when the learning support device 30 cannot identify peaks in the baseline. The first chromatogram image 211 in FIG. 12 shows peak Pa, peak Pb, and peak Pc. In the example of FIG. 12, the peak Pb and the peak Pc are connected, and although the starting point Pb1 of the peak Pb is displayed, the ending point of the peak Pb is not displayed (see point S in FIG. 12). Also, although not shown, there are cases where peaks are displayed for which neither the start point nor the end point are displayed. Peaks without such start and/or end points may be identified by vertical lines.

A vertical line is a line that is perpendicular or substantially perpendicular to the horizontal axis (time axis) of the first chromatogram image 211 . In the example of FIG. 12, a vertical line is displayed as the first peak information image 220 by specifying the first point 201 and the second point 202 by the user. Thus, the first peak information image 220 may include at least one of the baseline shown in FIG. 7 and the vertical lines shown in FIG.

In the example of FIG. 12, the details of the past result area 132A are not described, but for example, the similarity image 271, the second feature amount image 275, and the second peak information image 253 such as a vertical line are displayed. be.

[Aspect]
It will be appreciated by those skilled in the art that the multiple exemplary embodiments described above are specific examples of the following aspects.

(Section 1) A learning support method according to one aspect is a method of causing a computer to execute a process of supporting learning work of an estimation model used for detecting peaks in a signal waveform acquired by an analysis device. The learning aid method comprises obtaining a first signal waveform produced by an analyzer. The learning aid method comprises displaying the first signal waveform on a display device. In the learning support method, a second signal waveform having a high degree of similarity to the first signal waveform and second peak information for specifying a peak of the second signal waveform are extracted from a storage device storing a plurality of annotated signals. Prepare to get. The learning support method comprises displaying a second signal waveform and a second peak information image showing second peak information on a display device. The learning aid method comprises accepting user input of first peak information for identifying peaks of a first signal waveform. The learning assistance method comprises learning an estimation model based on the first signal waveform and the first peak information.

According to such a configuration, the user can input the first peak information consistent with the past second peak information, and update the estimation model based on the first peak information. Therefore, the object is to encourage the user to input training data in which variations in annotations are suppressed. As a result, it is possible to suppress variations in annotations and improve the quality of the estimated model.

(Section 2) In the learning support method according to Section 1, obtaining the second signal waveform and the second peak information from the storage device includes the first signal waveform and the plurality of annotated signals. and obtaining a stored signal waveform whose first similarity is greater than or equal to a first threshold as a second signal waveform. After displaying the first signal waveform on the display device and displaying the second signal waveform and the second peak information image on the display device, the input of the first peak information by the user is accepted.

According to such a configuration, the user can visually recognize the second signal waveform similar to the first signal waveform and created in the past, and the second peak information indicating the peak of the second signal waveform. You can enter the first peak information of one chromatogram. Therefore, it is possible to improve the user's convenience in inputting the first peak information.

(Section 3) In the learning support method according to Section 2, after displaying the first signal waveform on the display device and accepting user input of the first peak information, the storage device to acquire the second signal waveform and the second peak information from. Obtaining the second signal waveform and the second peak information from the storage device comprises first similarity between the first signal waveform and each of a plurality of stored signal waveforms included in the plurality of annotated signals. calculating a second degree of similarity between the first peak information and each of a plurality of pieces of stored peak information included in the plurality of annotated signals; Acquiring, as the second signal waveform, a signal waveform that is equal to or greater than a threshold and whose peak is specified by stored peak information in which the second similarity is equal to or greater than the second threshold.

According to such a configuration, the stored signal waveform whose peak is specified by the stored peak information having the first similarity greater than or equal to the first threshold and the second similarity greater than or equal to the second threshold is generated. It can be displayed as a second signal waveform. Therefore, the user can confirm such a second signal waveform.

(Item 4) In the learning support method according to item 2 or 3, the second signal waveform includes a high similarity signal waveform and a low similarity signal waveform having a first similarity lower than the high similarity signal waveform. including. Displaying the second signal waveform and the second peak information image on the display device includes displaying a waveform image showing a high similarity signal waveform with priority over a waveform image showing a low similarity signal waveform.

According to such a configuration, the user can preferentially view the second signal waveform having high similarity with the first signal waveform over the second signal waveform having low similarity.

(Section 5) In the learning support method according to any one of Sections 2 to 4, displaying the second signal waveform and the second peak information image on the display device includes: is displayed in association with a second signal waveform image showing the second signal waveform.

According to such a configuration, the user can visually recognize the first similarity of the second signal waveform.

(Section 6) In the learning support method according to any one of Sections 2 to 5, the calculation of the first similarity includes the first characteristic amount of the first signal waveform and the It includes calculating a first similarity based on the first feature amount and a second feature amount of the same type.

According to such a configuration, it is possible to calculate the first similarity with a relatively simple calculation.
(Section 7) In the learning support method according to any one of Sections 2 to 6, the calculation of the first similarity is based on the analysis result indicated by the first signal waveform and the stored signal waveform. and calculating a first degree of similarity based on the indicated analysis results.

According to such a configuration, it is possible to calculate the first similarity with a relatively simple calculation.
(Item 8) In the learning support method according to any one of items 1 to 7, the learning support method further comprises displaying a first peak information image showing the first peak information on the display device. Prepare.

According to such a configuration, the first peak information input by the user can be recognized on the display device.

(Section 9) In the learning support method according to Section 8, displaying the first peak information image on the display device includes: Displaying a first feature image indicative of the quantity. Further, displaying the second signal waveform and the second peak information image on the display device is a second feature amount image showing the feature amount of the peak specified by the second peak information indicated by the second peak information image. including displaying

According to such a configuration, after the user inputs the first peak information, the feature quantity of the peak specified by the peak of the first peak information and the feature quantity of the peak specified by the second peak information are determined. recognizable. Therefore, the user can confirm whether or not the input first peak information is appropriate.

(Item 10) In the learning support method according to any one of items 1 to 9, the learning support method includes detection for specifying a peak of the first signal waveform detected using the estimation model. It further comprises displaying a detected peak information image showing the peak information. After displaying the detected peak information image and the first signal waveform, the input of the first peak information by the user is accepted.

According to such a configuration, the user can input the first peak information while referring to the detected peak information.

(Item 11) In the learning support method according to any one of items 1 to 10, the learning support method includes: while the first signal waveform is displayed on the display device, The method further comprises accepting user input with two points. The second peak information image is a line image. The area surrounded by the second signal waveform and the line image is the area showing the peak of the second signal waveform. The peak area specified by the first peak information is an area surrounded by a line connecting the first point and the second point and a line included in the first signal waveform.

According to such a configuration, the first peak information can be input by designating the first point and the second point of the first signal waveform. Therefore, since the user can input the first peak information relatively easily, user convenience can be improved.

(Section 12) A learning support program according to one aspect is a program that causes a computer to execute a process of supporting learning work of an estimation model used for detecting peaks in a signal waveform acquired by an analysis device. The learning support program causes the computer to acquire the first signal waveform created by the analyzer. The learning support program causes the computer to display the first signal waveform on the display device. The learning support program provides a computer with a second signal waveform having a high degree of similarity to the first signal waveform and a second peak for identifying a peak of the second signal waveform from a storage device that stores a plurality of annotated signals. Get information and get it done. The learning support program causes the computer to display the second signal waveform and the second peak information image representing the second peak information on the display device. The learning support program causes the computer to accept input by the user of first peak information for identifying peaks of the first signal waveform. The learning support program causes the computer to learn the estimation model based on the first signal waveform and the first peak information.

According to such a configuration, the user can input the first peak information consistent with the past second peak information, and update the estimation model based on the first peak information. Therefore, the object is to encourage the user to input training data in which variations in annotations are suppressed. As a result, it is possible to suppress variations in annotations and improve the quality of the estimated model.

It should be noted that, regarding the above-described embodiments and modifications, it is possible to appropriately combine the configurations described in the embodiments within a range that does not cause any inconvenience or contradiction, including combinations not mentioned in the specification at the time of filing. is scheduled from

It should be considered that the embodiments disclosed this time are illustrative in all respects and not restrictive. The scope of the present invention is indicated by the scope of the claims rather than the above description, and is intended to include all modifications within the scope and meaning of equivalents of the scope of the claims.

10 analyzer, 11 mobile phase container, 12 pump, 13 injector, 14 column, 15 detector, 18 display interface, 20 control unit, 25 data analysis device, 26 input interface, 30 learning support device, 32 acquisition unit, 34 processing Unit, 36 Output Unit, 51 Control Device, 52 Storage Device, 61 Input Device, 62 Display Device, 62A Display Area, 100 Analysis System, 110 Data Collection Unit, 111 Peak Detection Processing Unit, 114 Model Storage Unit, 116 Peak Determination Unit , 117 analysis unit, 121 estimation model, 122 control program, 130 recording medium, 131A editing area, 132A past result area, 201 first point, 202 second point, 211 first chromatogram image, 212 second chromatogram image, 217 cursor, 220 first peak information image, 231 first feature image, 253 second peak information image, 271 similarity image.

Claims (12)

A learning support method for causing a computer to execute a process for supporting learning work of an estimation model used for detecting peaks in a signal waveform acquired by an analysis device, comprising:
obtaining a first signal waveform produced by an analyzer;
displaying the first signal waveform on a display device;
Acquiring a second signal waveform highly similar to the first signal waveform and second peak information for specifying a peak of the second signal waveform from a storage device storing a plurality of annotated signals; ,
displaying the second signal waveform and a second peak information image showing the second peak information on the display device;
receiving user input of first peak information for identifying a peak of the first signal waveform;
learning the estimation model based on the first signal waveform and the first peak information. Acquiring the second signal waveform and the second peak information from the storage device includes:
calculating a first similarity between the first signal waveform and each of a plurality of stored signal waveforms included in the plurality of annotated signals;
acquiring as the second signal waveform a stored signal waveform whose first similarity is greater than or equal to a first threshold;
displaying the first signal waveform on the display device;
After displaying the second signal waveform and the second peak information image on the display device,
2. The learning support method according to claim 1, wherein input by a user of said first peak information is received. displaying the first signal waveform on the display device;
After receiving input by the user of the first peak information,
obtaining the second signal waveform and the second peak information from the storage device;
Acquiring the second signal waveform and the second peak information from the storage device includes:
calculating a first similarity between the first signal waveform and each of a plurality of stored signal waveforms included in the plurality of annotated signals;
calculating a second degree of similarity between the first peak information and each of a plurality of stored peak information included in the plurality of annotated signals;
A signal waveform having a first similarity greater than or equal to a first threshold and a stored signal waveform having a peak specified by stored peak information having a second similarity greater than or equal to a second threshold is defined as the second signal waveform. The learning support method according to claim 1, comprising obtaining as The second signal waveform includes a high-similarity signal waveform and a low-similarity signal waveform having a lower degree of similarity than the high-similarity signal waveform,
Displaying the second signal waveform and the second peak information image on the display device includes displaying a waveform image showing the high-similarity signal waveform with priority over a waveform image showing the low-similarity signal waveform. The learning support method according to claim 2 or 3, comprising: Displaying the second signal waveform and the second peak information image on the display device corresponds the first similarity of the second signal waveform to a second signal waveform image showing the second signal waveform. 5. The learning support method according to any one of claims 2 to 4, further comprising displaying with an attachment. Calculating the first similarity is based on a first feature amount of the first signal waveform and a second feature amount of the same type as the first feature amount of the stored signal waveform. The learning support method according to any one of claims 2 to 5, comprising calculating the degree. Calculating the first degree of similarity includes calculating the first degree of similarity based on an analysis result indicated by the first signal waveform and an analysis result indicated by the stored signal waveform. The learning support method according to any one of claims 2 to 6. The learning support method according to any one of claims 1 to 7, further comprising displaying a first peak information image showing the first peak information on the display device. Displaying the first peak information image on the display device includes displaying a first feature amount image indicating the feature amount of the peak specified by the first peak information indicated by the first peak information image. ,
Displaying the second signal waveform and the second peak information image on the display device is a second feature amount indicating the feature amount of the peak specified by the second peak information indicated by the second peak information image. 9. A learning support method according to claim 8, comprising displaying an image. The learning support method further comprises displaying a detected peak information image indicating detected peak information for identifying a peak of the first signal waveform detected using the estimated model,
The learning support method according to any one of claims 1 to 9, wherein after the detected peak information image and the first signal waveform are displayed, user input of the first peak information is accepted. The learning support method further comprises accepting user input of a first point and a second point while the first signal waveform is displayed on the display device,
The second peak information image is a line image,
an area surrounded by the second signal waveform and the line image is an area showing the peak of the second signal waveform;
2. The peak region specified by the first peak information is a region surrounded by a line connecting the first point and the second point and a line included in the first signal waveform. The learning support method according to any one of claims 1 to 10. A learning support program that causes a computer to execute processing that supports learning work of an estimation model used for detecting peaks in a signal waveform acquired by an analysis device,
to said computer;
obtaining a first signal waveform produced by an analyzer;
displaying the first signal waveform on a display device;
Acquiring a second signal waveform highly similar to the first signal waveform and second peak information for specifying a peak of the second signal waveform from a storage device storing a plurality of annotated signals; ,
displaying the second signal waveform and a second peak information image showing the second peak information on the display device;
receiving user input of first peak information for identifying a peak of the first signal waveform;
A learning support program for learning the estimation model based on the first signal waveform and the first peak information.

Images