Waveform analysis method, waveform analysis device, and analyzer

Abstract

To provide more accurate peak information when performing peak detection automatically.SOLUTION: One aspect of a waveform analysis method pertaining to the present invention is a waveform analysis method for analyzing a signal waveform which is a chromatogram or a spectrum. The method includes: a model creation step for creating a trained model that identifies a peak section included in an inputted waveform, by machine learning using a plurality of reference waveforms in which the positions of peak sections are known; a region estimation step (S11-S14) for dividing the waveform to be analyzed into a plurality of partial waveforms, determining whether or not each of a plurality of partial waveforms of the waveform to be analyzed is a peak section, by using the trained model, and estimating a plurality of different types of regions in the waveform to be analyzed, including a peak region and a non-peak region, on the basis of the determination result; and a detection step (S15) for determining whether or not a row of each estimated region along the horizontal axis of the waveform to be analyzed conforms to a prescribed rule, and detecting a section which is inappropriate as a region estimation result.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to a method and device for analysing a signal waveform acquired by an analytical device, and to an analytical device equipped with such a waveform analyser. [Background technology]

[0002] In a chromatogram obtained by an analytical device such as a gas chromatograph (GC) or a liquid chromatograph (LC), peaks originating from components contained in a sample appear. In a data device provided in such an analytical device, peaks are generally detected by performing waveform processing on the chromatogram obtained by analysis, and peaks of a target compound are identified by performing identification processing on the detected multiple peaks. In addition, the concentration or content of the compound corresponding to the identified peak is calculated from the area or height of the identified peak.

[0003] To date, various methods have been put into practical use as methods for detecting peaks. In recent years, a new peak detection method that utilizes machine learning has been proposed and put into practical use (see Patent Document 1 and Non-Patent Document 1).

[0004] In the waveform analysis method described in Patent Document 1, a set including a plurality of partial waveforms created by finely dividing a reference waveform in which the position of the peak portion is known in the time axis direction is prepared for each reference waveform, and a trained model is created by machine learning using the partial waveforms for each of the many sets of reference waveforms to identify partial waveforms corresponding to peak portions included in an input waveform. The analysis target waveform is also divided into a plurality of partial waveforms in the same manner as the reference waveform, and a trained model is used to determine whether each of the plurality of partial waveforms is a peak portion. Then, based on the determination result, the peak region and other regions of the analysis target waveform are determined. By performing machine learning using partial waveforms corresponding to the peak start point and peak end point in addition to the peak portion, partial waveforms corresponding to the peak start point and peak end point can also be found among the plurality of partial waveforms in the analysis target waveform. Furthermore, by further dividing the peak portion into a single peak portion and an unseparated peak portion in which a plurality of peaks are superimposed, it is also possible to determine the single peak region and the unseparated peak region in the analysis target waveform. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] International Publication No. 2021 / 064924 [Non-patent literature]

[0006] [Non-Patent Document 1] "Peakintelligence for LCMS LabSolutions LCMS, optional waveform processing software for LabSolutions Insight," [online], [searched March 16, 2023], Shimadzu Corporation, Internet <URL:https: / / www.an.shimadzu.co.jp / products / liquid-chromatograph-mass-spectrometry / lc-ms-software / peakintelligence / index.html> [Non-Patent Document 2] Olaf Ronneberger and 2 others, "U-Net: Convolutional Networks for Biomedical Image Segmentation", [online], [Submitted on 18 May 2015], arXiv.org, Internet.<URL: https: / / arxiv.org / pdf / 1505.04597.pdf> Summary of the Invention [Problem to be solved by the invention]

[0007] Peak detection using machine learning is a very useful method in that it eliminates or reduces the need for users to set tedious parameters. In addition, in many cases, it is possible to detect peaks quite accurately for various chromatogram waveforms. Nevertheless, depending on the shape of the waveform to be analyzed, peak regions, etc. may not be determined accurately, and inappropriate peak detection results that lack rationality and go against common technical common sense may be output.

[0008] An object of the present invention is to provide a waveform analysis method and a waveform analysis device that can automatically recognize inappropriate peak portions by waveform processing performed after peak detection processing using machine learning, even if peaks are not properly detected by the peak detection processing, and can inform the user of this or correct it without human intervention. [Means for solving the problem]

[0009] One aspect of the waveform analysis method according to the present invention is a waveform analysis method for analyzing a signal waveform that is a chromatogram or a spectrum, comprising: A model creation step of creating a trained model for identifying peak portions contained in an input waveform by machine learning using a plurality of reference waveforms whose peak portion positions are known; a region estimation step of dividing the waveform to be analyzed into a plurality of partial waveforms, determining whether each of the plurality of partial waveforms is a peak portion using the trained model, and estimating a plurality of different types of regions, including peak regions and non-peak regions, in the waveform to be analyzed based on the results of the determination; a detection step for detecting an inappropriate region estimation result by determining whether or not the arrangement of each region estimated in the region estimation step along the horizontal axis of the analysis target waveform complies with a predetermined rule; has.

[0010] One aspect of the waveform analysis device according to the present invention is a waveform analysis device that analyzes a signal waveform that is a chromatogram or a spectrum, comprising: a region estimation unit that divides a waveform to be analyzed into a plurality of partial waveforms, and determines whether each of the plurality of partial waveforms of the waveform to be analyzed is a peak portion or not using a trained model created by machine learning using a plurality of reference waveforms whose peak portion positions are known, and estimates a plurality of different types of regions, including peak regions and non-peak regions, in the waveform to be analyzed based on the results of the determination; a detection unit that detects inappropriate region estimation results by determining whether or not the arrangement of each region estimated by the region estimation unit along the horizontal axis of the waveform to be analyzed complies with a predetermined rule; Equipped with.

[0011] Furthermore, one aspect of the analysis device according to the present invention is any one of a chromatography device, a mass spectrometer, and an optical measurement device, which includes any aspect of the waveform analysis device according to the present invention as a data analysis section. Effect of the Invention

[0012] According to the above aspects of the waveform analysis method and waveform analysis device of the present invention, even if peaks are not properly detected by the peak detection process using machine learning, the waveform processing performed after the peak detection process can automatically recognize inappropriate peak parts and notify the user of the inappropriate peak parts or correct them without human intervention. As a result, while automating peak detection using machine learning, it is possible to eliminate or reduce the output of inappropriate peak detection results that lack rationality and would not occur if peak detection was based on human judgment. [Brief description of the drawings]

[0013] [Figure 1] 1 is a schematic diagram showing the configuration of an embodiment of an analysis device for carrying out a waveform analysis method according to the present invention. [Diagram 2] 11 is a flowchart showing the steps of a trained model creation process in the analysis device of this embodiment. [Diagram 3] 4 is a flowchart showing the procedure of a peak detection process on a chromatogram waveform to be analyzed in the analyzer of the present embodiment. [Figure 4] FIG. 4 is a diagram showing an example of a peak waveform on a chromatogram. [Diagram 5] FIG. 5 shows the results of determining the likelihood of a region using a trained model for the peak waveform shown in FIG. 4. [Figure 6] 6 is a diagram showing a region estimation result based on the determination result shown in FIG. 5 (when estimated appropriately). [Figure 7] FIG. 13 shows another example of peak waveforms on a chromatogram (when tailing is present). [Figure 8] A figure showing the results of judging the region likelihood using a trained model for the peak waveform shown in Figure 7 (when the region is not estimated appropriately). DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] In the waveform analysis method and the waveform analysis device according to the above-mentioned aspects of the present invention, the chromatogram includes a chromatogram obtained by a GC device including a gas chromatograph mass spectrometer (GC-MS), a chromatogram obtained by an LC device including a liquid chromatograph mass spectrometer (LC-MS), an electropherogram obtained by an electrophoresis device, etc. In a GC device or an LC device in which the detector is a mass spectrometer, the chromatogram includes a total ion (total ion current) chromatogram and an extracted ion chromatogram. In addition, the spectrum includes a mass spectrum obtained by a mass spectrometer (a profile spectrum not subjected to centroid processing), a time-of-flight spectrum obtained by a time-of-flight mass spectrometer before being converted into a mass spectrum, a light intensity spectrum obtained by an optical measurement device such as a spectrometer or a fluorescence measurement device, an X-ray intensity spectrum obtained by an X-ray analysis device, etc.

[0015] Hereinafter, an LC system, which is one embodiment of an analytical device including a waveform analyzer for carrying out a waveform analysis method according to the present invention, will be described with reference to the accompanying drawings. FIG. 1 is a schematic configuration diagram of an LC system A of this embodiment and a system B that creates a trained model used in the LC system A.

[0016] The LC system A in this embodiment includes an LC measurement unit 1, a data analysis unit 2, an input unit 24, and a display unit 25. Although not shown, the LC measurement unit 1 includes a liquid delivery pump, an injector, a column, a column oven, a detector, etc., and performs LC analysis on a given sample to obtain chromatogram data indicating the temporal change in signal intensity obtained by the detector. There is no particular limit to the type or type of detector, and for example, a mass spectrometer, a photodiode array (PDA) detector, etc. can be used.

[0017] The data analysis unit 2 includes functional blocks such as a data collection unit 20, a peak detection processing unit 21, a qualitative / quantitative analysis unit 22, and a display processing unit 23. The peak detection processing unit 21 includes functional blocks such as a waveform preprocessing unit 210, a judgment unit 211, a trained model storage unit 212, a region tentative determination unit 213, a non-prescribed region detection unit 214, and a region correction unit 215.

[0018] In the data analysis section 2, the data collection section 20 collects and stores the chromatogram data obtained in the LC measurement section 1. The peak detection processing section 21 automatically detects peaks in a chromatogram waveform formed by the collected chromatogram data in response to an instruction received from a user at the input section 24, and outputs peak information including the start and end positions (retention time) of the detected peaks and the range of the peak region. The qualitative and quantitative analysis section 22 identifies components (compounds) corresponding to each peak based on the peak information provided by the peak detection processing section 21, calculates peak height values ​​and peak area values, and calculates quantitative values, which are the concentrations or content amounts of each component, from the values. The display processing section 23 displays the peak detection results and calculated values ​​such as quantitative values ​​calculated based on them on the display section 25 in a predetermined format.

[0019] In general, the entity of the data analysis unit 2 is a personal computer or a more powerful workstation in which a specific software (computer program) is installed, or a computer system including a high-performance computer connected to such a computer via a communication line. That is, the functions of each block included in the data analysis unit 2 are realized by executing software installed on a single computer or a computer system including multiple computers on the computer. Of course, some of these functions may be executed using hardware circuits specialized for specific calculations, such as a digital signal processor.

[0020] The computer program can be provided to the user in a form stored in a computer-readable non-transitory storage medium, such as a CD-ROM, a DVD-ROM, a memory card, a USB memory (dongle), etc. The program can also be provided to the user in the form of data transfer via a communication line such as the Internet. Furthermore, the program can be pre-installed in a computer that is part of the system (strictly speaking, a storage device that is part of the computer) at the time the user purchases the system.

[0021] 1, system B provided separately from LC system A includes a model creation unit 3, and the model creation unit 3 includes, as functional blocks, a learning data storage unit 30, a learning execution unit 31, and a model storage unit 32. The trained model created in this model creation unit 3 is stored in a trained model storage unit 212 in the data analysis unit 2 of LC system A.

[0022] Although the details will be described later, in general, the task of creating a trained model requires a huge amount of calculation. Therefore, the entity of system B is a high-performance computer, and the functions of each block are realized by executing software installed on the computer. Of course, system B may be integrated with LC system A.

[0023] Next, the peak detection process executed mainly by the peak detection processing unit 21 of the data analysis unit 2 will be described. In very general terms, the peak detection processing unit 21 visualizes a chromatogram waveform composed of chromatogram data, and then detects the positions or ranges of peak regions, peak start regions, peak end regions, non-peak regions, etc. by using a semantic segmentation technique based on deep learning, which is a machine learning technique that detects the category and position of objects present in the image.

[0024] [Creating a trained model] As is well known, in machine learning methods, it is necessary to construct a trained model in advance using a large amount of learning data (training data and validation data). As described above, the task of constructing this trained model is not performed in the data analysis unit 2 that is part of the LC system A, but is performed in the model creation unit 3 that is configured by a separate computer system, and the result is stored in the trained model storage unit 212. FIG. 2 is a flowchart showing an outline of the procedure of the trained model creation process performed in the model creation unit 3.

[0025] When creating a trained model, first, training data based on a reference waveform is prepared (step S1). Here, a large number of diverse chromatogram waveforms are used as reference waveforms. The diverse chromatogram waveforms referred to here are preferably chromatogram waveforms including elements such as various noise contamination, baseline fluctuation (drift), overlapping of multiple peaks, or deformation of peak shape, which may appear in a chromatogram waveform when actually performing peak detection. However, this chromatogram waveform data may not be data collected by actual LC analysis, but may be data created by simulation.

[0026] Peak detection is performed in advance on the chromatogram waveform, which is the reference waveform, and the start and end points of one or more peaks on the waveform are accurately determined. This chromatogram waveform is visualized with the signal intensity, i.e., the vertical axis of the graph, normalized, and further divided into a predetermined number of partial original waveforms in the horizontal axis, i.e., the time axis direction. This division number is determined so that the width (length in the time axis direction) of each partial waveform is at least smaller than the peak width. Therefore, the division number can be appropriately set according to the minimum value of the expected peak width. Note that the signal waveform may be visualized and treated as one-dimensional array data rather than two-dimensional array data.

[0027] A chromatogram waveform is composed of many partial waveforms. Data constituting each partial waveform is associated with characteristic information indicating which of a plurality of types of regions the partial waveform belongs to. For example, characteristic information indicating a peak region is associated with a partial waveform corresponding to a peak portion on a chromatogram waveform, characteristic information indicating a peak start region is associated with a partial waveform including a peak start point, and characteristic information indicating a non-peak region (baseline region) is associated with a partial waveform not including any peak portion. There are at least four types of regions: a peak region, a peak start region, a peak end region, and a non-peak region. However, the peak region may be divided into a single peak region where only one peak exists (no overlapping with other peaks) and an unseparated peak region where multiple peaks overlap, depending on the state of the peak in the peak portion. The unseparated peak region may also be divided into more finely as described later. The learning data storage unit 30 stores partial waveform data constituting each of a plurality of chromatogram waveforms together with the characteristic information associated therewith. The learning data may be divided into training data and verification data in advance, or may be used as either training data or verification data as appropriate when learning is performed without such a division.

[0028] When the start of creating a learning model is instructed, the learning execution unit 31 prepares an unlearned learning model (step S2). Various learning models capable of executing semantic segmentation can be used for this learning model. Semantic segmentation is generally used to analyze an image composed of pixel data distributed two-dimensionally, but here, it is applied to the analysis of a chromatogram waveform composed of data arranged one-dimensionally along the time axis. Here, U-Net (see Non-Patent Document 2) is used as a learning model capable of executing semantic segmentation, but other learning models such as SeGNet and PSPNet may also be used.

[0029] Next, the learning execution unit 31 reads the learning data (partial waveform data and characteristic information) from the learning data storage unit 30 (step S3). The learning execution unit 31 performs machine learning using the read learning data, and constructs a learning model for estimating to which region a given partial waveform corresponds (step S4). The learning procedure will not be described in detail here, but a trained model can be constructed according to the procedure described in Patent Document 1, for example.

[0030] The model storage unit 32 stores the trained model created by machine learning using a large amount of training data (step S5). The trained model stored in the model storage unit 322 is transmitted to the trained model storage unit 212 in the LC system A via, for example, a communication line and stored therein.

[0031] It should be noted that the trained model and the method for creating the trained model used in this embodiment are merely examples, and the present invention is not limited to this description. That is, the present invention can be widely applied to trained models in general that potentially have the problems described in this specification. A specific example of another trained model is a trained model using a general regression model. A general regression model is a technology for predicting a numerical value for input data, and is a model that performs more limited processing than the trained model that estimates an area such as a peak area described above.

[0032] [Peak detection processing for the waveform to be analyzed] Next, a process for detecting peaks on a chromatogram waveform obtained for a target sample, which is executed in the data analysis unit 2 in the LC system A, will be described. Fig. 3 is a flow chart showing an outline of the flow of the peak detection process performed in the peak detection processing unit 21.

[0033] First, the waveform preprocessing unit 210 reads the chromatogram waveform data to be analyzed from the data collecting unit 20 (step S11). The waveform preprocessing unit 210 normalizes the signal intensity of the read data, visualizes it (or one-dimensionally arranges it), and further divides the visualized chromatogram waveform into a predetermined number of partial waveforms in the horizontal axis (time axis) direction (step S12). This number of divisions may be the same as the number of divisions in the learning data, but may be a different number as long as the width of the partial waveform is smaller than the peak width.

[0034] Next, the determination unit 211 reads out the trained model from the trained model storage unit 212, and sequentially inputs the partial waveforms to the trained model. The trained model determines whether the input partial waveform corresponds to each region, such as a peak region, a peak start region, a peak end region, a non-peak region, etc. Specifically, the determination unit 211 uses the trained model to calculate probability information indicating the possibility that each partial waveform corresponds to each region as a numerical value (step S13). The higher the numerical value of this probability, the more likely the partial waveform is to be in that region. In this way, the determination unit 211 outputs all partial waveforms constituting the input chromatogram waveform with the probability information for each region, such as a peak region, a peak start region, etc., added thereto.

[0035] The area tentative determination unit 213 receives the output from the determination unit 211, determines the area showing the highest probability for each partial waveform as the area corresponding to that partial waveform, and determines the type of area corresponding to each partial waveform (step S14). As a result, all partial waveforms constituting the entire chromatogram waveform are classified into one of the areas.

[0036] A specific example will now be given. The waveform to be analyzed is assumed to be the chromatogram waveform shown in Fig. 4. As can be seen from Fig. 4, this chromatogram has two peaks that appear to overlap each other. Fig. 5 shows the accuracy of each region obtained by determination unit 211 for this chromatogram waveform. In Fig. 5, the vertical axis indicates normalized accuracy, and the higher this value (the closer to 1), the more likely it is that region. In this example, the peak regions are divided into two, single peak regions and unseparated peak regions, and the number of regions is five.

[0037] FIG. 6 is a diagram showing the regions determined by the region tentative determination unit 213 superimposed on FIG. 5. In this case, when the region determination result in FIG. 6 is viewed in the time course direction (from left to right), it can be seen that the order is non-peak region → peak start region → unseparated peak region → peak end region → peak start region → unseparated peak region → peak end region → non-peak region. That is, although the bottoms of the two peaks overlap as shown in FIG. 4, in the result of FIG. 6, the two peaks are detected as unseparated peak regions, and the peak start region exists at the left end of each unseparated peak region, and the peak end region exists at the right end. In other words, each unseparated peak region is sandwiched between the peak start region and the peak end region. Therefore, at least from the viewpoint of the arrangement of the regions, it can be said that the peak detection result is rational and valid.

[0038] Next, consider the case where the chromatogram waveform of the analysis target waveform is as shown in FIG. 7. This chromatogram waveform has a long tailing and noise (or other very small peaks) on the tailing. FIG. 8 shows the accuracy in each region obtained by the determination unit 211 for this chromatogram waveform. In this case, a peak start region exists at the left end of the unseparated peak region, but a peak end region does not exist at the right end of the unseparated peak region, and the unseparated peak region and the non-peak region are adjacent to each other. Since a peak should always have a start point and an end point as technical common sense, the above peak detection result is irrational and inappropriate in this respect. One reason why the above-mentioned determination result is obtained as a result of region determination using a trained model is that the trained model is not trained on the rule that a peak start region and a peak end region always exist at the ends of a peak region when the trained model is created.

[0039] In this embodiment, as a post-processing of the above-mentioned judgment using machine learning, a process of correcting an irrational judgment result is performed. Specifically, one or more rules related to the arrangement of regions are registered in advance in the non-standard area detection unit 214. When the non-standard area detection unit 214 receives the judgment result of the region based on the partial waveform from the region tentative determination unit 213, it recognizes the arrangement of the regions in the time axis direction, that is, from left to right of the chromatogram waveform, and judges whether the arrangement matches the registered rule. Then, if there is a part that does not match the rule, it identifies the position (step S15).

[0040] In machine learning, statistical processing is performed from given learning data without explicitly stating the relationship between input and output or the rules. Therefore, although a trained model created by machine learning can handle phenomena that do not appear in the rules and irregular inputs, there is a slight possibility that the model may output a result that does not follow the rules. In response to this, by providing the processing of step S15, after the trained model outputs an estimated result, the consistency of the result with the rules is judged and the result is adapted to the rules, thereby improving the output accuracy of the present system.

[0041] As the rules for arranging the above-mentioned regions, various conditions that are technically reasonable can be registered. For example, the following rules can be considered: <Rule 1> A peak start region exists at the left end of a peak region, and a peak end region exists at the right end of the peak region. <Rule 2> When viewed from the left, only one type of peak region exists between the start region of one peak and the end region of the next peak (for example, a single peak region and an unresolved peak region do not coexist). <Rule 3> A peak region and a non-peak region are not adjacent to each other. Of course, the above rules are just examples, and any other reasonable rules can be added.

[0042] For example, in the example shown in FIG. 8, when the non-standard area detection unit 214 checks the arrangement of the areas, it recognizes that there is no peak end area at the right end of the unseparated peak area, so rule 1 is not satisfied. In addition, it recognizes that the unseparated peak area and the non-peak area are adjacent to each other, so rule 3 is also not satisfied. Therefore, the non-standard area detection unit 214 detects the right end of the unseparated peak area, that is, the boundary between the unseparated peak area and the non-peak area adjacent to its right, as a part that does not match the rules. On the other hand, in the example shown in FIG. 6, the non-standard area detection unit 214 determines that the arrangement of the areas at least satisfies the above rules 1 to 3, so it outputs a result that there is no part that does not match the rules.

[0043] When the area correction unit 215 receives a result that the non-standard area detection unit 214 has detected a part that does not match the rules, the area correction unit 215 corrects the area determined in step S14 above or a part of it in accordance with a predetermined rule corresponding to each rule (step S16).

[0044] For example, when it is determined that either one or both of the peak start region and the peak end region does not exist and rule 1 is not met, the region correction unit 215 changes the region corresponding to the partial waveform located at the left end and / or the right end of the peak region to the peak start region and / or the peak end region. When a peak region of a different type exists between one peak start region and the next peak end region and rule 2 is not met, the region correction unit 215 determines whether or not one of the different types of peak regions is wider (longer in the time direction), and replaces the narrower peak region with the wider peak region. As a result, the region between the pair of peak start region and peak end region is unified into one type of peak region. When a peak region and a non-peak region are adjacent to each other and rule 3 is not met, the region correction unit 215 changes the region corresponding to the partial waveform located at the boundary between the peak region and the non-peak region to the peak start region or the peak end region.

[0045] After the region is corrected as necessary, the display processing unit 23 displays the peak detection result by the peak detection processing unit 21 on the screen of the display unit 25 (step S17). If the setting is such that the qualitative analysis is automatically performed based on the peak detection result, the qualitative and quantitative analysis unit 22 obtains, for example, the retention time of the peak top of each detected peak, and identifies the component corresponding to the peak based on the retention time. If the setting is such that the quantitative analysis is automatically performed based on the peak detection result, the qualitative and quantitative analysis unit 22 obtains the peak area value or height value for each detected peak, and calculates the concentration (content) of the component corresponding to the peak by referring to the area value or height value against a calibration curve created in advance. The display processing unit 23 displays the results of such qualitative or quantitative analysis on the screen of the display unit 25 together with the peak detection result.

[0046] As described above, in the LC system of this embodiment, when peak information of peaks automatically detected using machine learning lacks rationality in light of common technical common sense, the system can detect inappropriate peak information, appropriately correct it, and provide it to the user.

[0047] In the above explanation, the portion detected by the non-standard area detection unit 214 as not conforming to the rules is automatically corrected, but there may be cases where it is better to discard the peak detection result itself rather than correcting it. Therefore, instead of correcting it automatically, when the non-standard area detection unit 214 determines that it does not conform to the rules, the detected portion may first be displayed on the display unit 25 to notify the user. Then, in response to an instruction from the user who has confirmed the result, some of the area may be corrected, or the peak detection result itself may be discarded.

[0048] Next, some preferred additional configurations will be described. The configurations described below can be appropriately combined with those described in the above embodiment. In addition, multiple additional configurations can be combined as long as they do not perform contradictory processes.

[0049] In the above embodiment, the overlapped peak, in which multiple peaks overlap as shown in FIG. 4, is regarded as the unseparated peak region, but when the concentration or content of a compound is obtained from the area or height of the peak, a process of dividing the overlapped peak into multiple peaks is required. Several methods for the division are known, and the appropriate division method differs depending on the way the peaks overlap, etc. Specifically, the division of the overlapped peak includes, for example, a tailing process in which the start point to the end point of the target peak is divided into two peaks by dividing the peak into one peak and superimposing another peak on the peak, a complete separation in which the start point, the minimum point, and the end point of the target peak are connected in order to separate the peak into two peaks, and a vertical division in which the two peaks are separated by a perpendicular line passing through the minimum point of the target peak.

[0050] In the tailing process, the first peak start point and the second peak start point are present in order from the side with the shortest retention time, followed by the first and second peak end points. On the other hand, in complete separation and vertical division, the first peak start point and end point are present from the side with the shortest retention time, followed by the second peak start point and end point. Therefore, the unseparated peak region can be set to three regions: a vertically divided peak region, a completely separated peak region, and a tailing process peak region. In this case, a trained model can be created by machine learning using a plurality of sets of partial waveforms created by dividing a reference waveform including overlapping peaks that have been peak-separated by each of the tailing process, complete separation, and vertical division along the time axis.

[0051] In addition, in the above embodiment, the arrangement of the area determined based on the judgment result by the judgment unit 211 was used to judge the irrationality and inappropriateness of the peak detection result, but various other factors can be added to the judgment.

[0052] For example, it is possible to change part of the rules for determining the sequence of a region or change the rules for modifying a part of a region by using numerical information reflecting the shape and characteristics of the peak in the range corresponding to the peak region. Examples of numerical information reflecting the shape and characteristics of the peak that can be used here include the S / N ratio, the degree of peak separation, the symmetry coefficient, the percentage of missing peaks, the peak height, and the peak width.

[0053] For example, even if a region is determined to be a peak region, if the S / N ratio calculated from the signal intensity in the peak region and non-peak region is equal to or less than a predetermined threshold (e.g., 10), there is a high possibility that noise has been recognized as a peak, and the reliability of the peak detection result itself is low. Therefore, if the S / N ratio of a portion determined to be a peak region is equal to or less than a predetermined threshold, the peak region and the peak start region or peak end region at either end of the peak region may all be replaced with non-peak regions.

[0054] For example, the symmetry coefficient is an index showing the symmetry of a peak, but if the symmetry of the peak is poor, there is a possibility that a small peak is located at the base of a large peak. Therefore, if the symmetry coefficient is greater than a predetermined threshold, even if the determination result of the region corresponding to the peak portion is a single peak region, or even if the ratio of the unseparated peak region is small, the entire peak portion may be replaced with an unseparated peak region. Also, the same processing can be performed when the peak width is too wide.

[0055] Regarding the degree of resolution, the Japanese Pharmacopoeia states that a value of 1.5 or more indicates complete separation. Based on this degree of resolution, it is possible to correct a completely separated peak region to an unresolved peak region, or vice versa. Also, if a peak exists at the end of a chromatogram and part of the peak is missing, there is no problem with quantification if the missing proportion is small, but if the missing proportion is large, it is not suitable for quantification. Therefore, if the missing proportion is large, it can be modified so that it is not detected as a peak.

[0056] It is also possible to calculate the slope of the baseline based on the partial waveform determined to be a non-peak region, and use the calculated value to change part of the rules for determining the arrangement of regions, or to change the rules for modifying part of a region. Specifically, if the slope of the baseline changes from a positive value from large to small over time, it is assumed that the baseline is upwardly convex. In this case, even if the overlapping peaks present on such a baseline are vertically separated peak regions, it can be assumed that the bottom of the valley of the overlapping peak is in contact with the baseline, and therefore it can be changed to a completely separated peak region.

[0057] Furthermore, when making a judgment by comparing a numerical value reflecting the shape or characteristics of a peak or a numerical value reflecting the baseline shape with a threshold, it is preferable to set the threshold from the input unit 24. This allows the user to appropriately change the region change criteria depending on the purpose of analysis, the type of sample, etc.

[0058] Mass spectrometers can usually observe multiple types of ions (called quantitative ions and confirmatory ions) with different m / z values ​​generated from one component. Therefore, in GC-MS and LC-MS, it is possible to create an extracted ion chromatogram of the quantitative ions and one or more confirmed ions for the same component. Since these are chromatograms derived from the same component, the waveforms should be similar. Therefore, it is possible to compare the area estimation results in the quantitative ion chromatograms and the confirmed ion chromatograms for the same component, and change the rules for determining the area or the rules for correcting the area based on the comparison results.

[0059] For example, if the arrangement of regions determined in peak detection for a chromatogram of a quantitative ion does not conform to the rules, it is possible to refer to the peak detection results for a chromatogram of a confirmation ion of the same component, and if the results conform to the rules, then that result can be adopted.

[0060] It is also possible to compare the region estimation results in signal waveforms such as chromatograms and spectra obtained for an unknown sample containing the same target component and a standard sample containing the target component at a known concentration, or to additionally use the region estimation results in signal waveforms such as chromatograms and spectra obtained for a blank sample not containing the target component.

[0061] In addition, compound information on the target compound in the sample that can be known in advance can be additionally used to determine the validity of the region. The compound information here includes, for example, the concentration of the compound, as well as structural information such as whether or not the compound has isomers, and information such as whether or not derivatives generated during pretreatment or the like are present.

[0062] For example, when the presence of isomers in a target compound is known as compound information, even if the isomers have the same molecular weight, the peaks on a chromatogram often shift in time due to differences in structure, etc. Therefore, for a peak estimated to correspond to the target compound, even if the peak portion is determined to be a single peak region, it is possible to modify the region by replacing it with an unresolved peak region, for example.

[0063] In this way, by using additional information when assessing the validity of the region arrangement determined using the trained model or when correcting the regions, the accuracy of peak detection can be further improved.

[0064] The above-mentioned embodiment and modified examples are merely examples and can be modified as appropriate in accordance with the spirit of the present invention. In the above-mentioned embodiment, the waveform analyzer for detecting peaks is incorporated into the same measurement system A as the measurement unit, but it can be configured as a waveform analyzer independent of the LC system A. In that case, it is sufficient to read the chromatogram data previously acquired by the LC measurement unit 1 and perform the analysis.

[0065] Furthermore, while the above embodiment has been described with reference to an example in which a chromatogram waveform is processed, it is clear that the present invention can be applied to the waveform analysis of signal waveforms whose signal intensity can change in response to changes in the value of a specified parameter, which are obtained by various analytical instruments, such as electropherograms obtained by an electrophoresis instrument, mass spectra (profile spectra) obtained by a mass spectrometer, spectroscopic spectra obtained by a spectrometer, fluorescence spectra obtained by a fluorescence measuring instrument, and X-ray intensity spectra obtained by an X-ray analyzer.

[0066] [Various aspects] It will be appreciated by those skilled in the art that the exemplary embodiments described above are examples of the following aspects.

[0067] (Item 1) One aspect of the waveform analysis method according to the present invention is a waveform analysis method for analyzing a signal waveform that is a chromatogram or a spectrum, comprising: A model creation step of creating a trained model for identifying peak portions contained in an input waveform by machine learning using a plurality of reference waveforms whose peak portion positions are known; a region estimation step of dividing the waveform to be analyzed into a plurality of partial waveforms, determining whether each of the plurality of partial waveforms is a peak portion using the trained model, and estimating a plurality of different types of regions, including peak regions and non-peak regions, in the waveform to be analyzed based on the results of the determination; a detection step for detecting an inappropriate region estimation result by determining whether or not the arrangement of each region estimated in the region estimation step along the horizontal axis of the analysis target waveform complies with a predetermined rule; has.

[0068] (2) In the waveform analysis method described in (1), the model creation step can create a trained model that identifies peak portions contained in an input waveform by machine learning using multiple sets of partial waveforms created by dividing a reference waveform in which the positions of the peak portions are known.

[0069] (Item 15) One aspect of the waveform analysis device according to the present invention is a waveform analysis device that analyzes a signal waveform that is a chromatogram or a spectrum, comprising: a region estimation unit that divides a waveform to be analyzed into a plurality of partial waveforms, and determines whether each of the plurality of partial waveforms of the waveform to be analyzed is a peak portion or not using a trained model created by machine learning using a plurality of reference waveforms whose peak portion positions are known, and estimates a plurality of different types of regions, including peak regions and non-peak regions, in the waveform to be analyzed based on the results of the determination; a detection unit that detects inappropriate region estimation results by determining whether or not the arrangement of each region estimated by the region estimation unit along the horizontal axis of the waveform to be analyzed complies with a predetermined rule; Equipped with.

[0070] (Item 16) In the waveform analysis device described in Item 15, the machine learning model may be created by machine learning using multiple sets of partial waveforms created by dividing a reference waveform whose peak positions are known.

[0071] According to the waveform analysis methods described in paragraphs 1 and 2 and the waveform analysis devices described in paragraphs 15 and 16, even if peaks are not properly detected by the peak detection process using machine learning, the waveform processing performed after the peak detection process can automatically recognize inappropriate peak parts and notify the user of the inappropriate peak parts or correct them without human intervention. As a result, while automating peak detection by using machine learning, it is possible to eliminate or reduce the output of inappropriate peak detection results that lack rationality and would not occur if peak detection was based on human judgment.

[0072] (Item 3) The waveform analysis method according to item 1 may further include a correction step of correcting the estimation result of the region in the invalid portion detected in the detection step in accordance with a predetermined rule.

[0073] (Item 17) The waveform analysis device according to item 15 may further include a correction unit that corrects the estimation result of the area in the invalid portion detected by the detection unit in accordance with a predetermined rule.

[0074] In the waveform analysis method described in paragraph 3 and the waveform analysis device described in paragraph 17, peak information of a portion of a signal waveform where peaks are not properly detected by the peak detection process using machine learning can be corrected to proper information without relying on manual labor. This prevents erroneous peak information from being provided to the user, and improves qualitative accuracy by reducing undetected peaks.

[0075] (4) The waveform analysis method described in paragraph 1 may further include a notification step of notifying a user of the estimated area of ​​the invalid portion detected in the detection step in a format that is visible to the user.

[0076] (Item 18) The waveform analysis device described in item 15 may further include a notification unit that notifies a user of the estimation result of the area in the invalid part detected by the detection unit in a format that is visible to the user.

[0077] In the waveform analysis method described in paragraph 4 and the waveform analysis device described in paragraph 18, even if peaks are not properly detected by the peak detection process using machine learning, the user can be sure of this fact. This allows the user to appropriately correct erroneous peak information or delete unnecessary information based on their own judgment, experience, etc., and obtain more accurate peak information.

[0078] (5) In the waveform analysis method according to the first aspect, the trained model is created by machine learning using a plurality of reference waveforms whose peak start and end positions are known, The plurality of different types of regions include a peak region, a non-peak region, a peak start region, and a peak end region, The predetermined rule may be that a peak start region exists at one end of a peak region and a peak end region exists at the other end of a peak region.

[0079] (Item 19) In the waveform analysis device according to item 15, the trained model is created by machine learning using a plurality of reference waveforms whose peak start and end positions are known, The plurality of different types of regions include a peak region, a non-peak region, a peak start region, and a peak end region, The predetermined rule may be that a peak start region exists at one end of a peak region and a peak end region exists at the other end of a peak region.

[0080] According to the waveform analysis method described in paragraph 5 and the waveform analysis device described in paragraph 19, it is possible to accurately find a peak whose starting point and / or ending point has not been detected.

[0081] (Item 6) In the waveform analysis method described in Item 5, the detection step may search for pairs of peak start and end regions along the horizontal axis of the waveform to be analyzed in ascending order of parameter values ​​on the horizontal axis, and if a peak region is sandwiched between the pairs of peak start and end regions found, the region estimation result may be determined to be valid.

[0082] (Clause 20) In the waveform analysis device described in Clause 19, the detection unit may search for pairs of peak start regions and peak end regions along the horizontal axis of the waveform to be analyzed in ascending order of parameter values ​​on the horizontal axis, and when a peak region is sandwiched between the pairs of peak start regions and peak end regions found, determine that the region estimation result is valid.

[0083] In the waveform analysis method described in paragraph 6 and the waveform analyzer described in paragraph 20, the "parameter on the horizontal axis" is time in a chromatogram or a time-of-flight spectrum, the m / z value in a mass spectrum, the wavelength or wavenumber in a light intensity spectrum such as a spectroscopic spectrum, and the wavelength or energy in an X-ray intensity spectrum. According to the waveform analysis method described in paragraph 6 and the waveform analyzer described in paragraph 20, it is possible to accurately find, for example, a peak whose start point and end point have both not been detected, a peak whose waveform shape is so unique that it is difficult to determine whether it is a single peak region or an unresolved peak region, and the like.

[0084] (Item 7) The waveform analysis method described in item 5 may further include a correction step of designating the ends of a peak region as the peak start region and / or the peak end region when it is detected in the detection step that a peak start region or a peak end region does not exist at either end of the peak region.

[0085] (Clause 21) In the waveform analysis device described in clause 19, the detection unit may further include a correction unit which, when it is detected that a peak start region or a peak end region does not exist at either end of a peak region, sets the ends of the peak region as the peak start region and / or the peak end region.

[0086] According to the waveform analysis method described in paragraph 7 and the waveform analysis device described in paragraph 21, even if there is a peak whose start point and / or end point has not been detected by peak detection using machine learning, the start point and / or end point of the peak can be identified and accurate peak information can be provided.

[0087] (Item 8) In the waveform analysis method according to item 5, the peak region includes a single peak region having one peak and an unseparated peak region having multiple overlapping peaks, The detection step may determine that the region estimation result is invalid when the two or more different types of peak regions are adjacent to each other along the horizontal axis of the waveform to be analyzed.

[0088] (Item 22) In the waveform analyzer according to item 19, the peak region includes a single peak region having one peak and an unseparated peak region having multiple overlapping peaks, The detection section may be configured to determine that the region estimation result is invalid when the two or more different types of peak regions are adjacent to each other along the horizontal axis of the waveform to be analyzed.

[0089] According to the waveform analysis method described in paragraph 8 and the waveform analysis device described in paragraph 22, it is possible to accurately detect, for example, peaks whose start and end points have not been detected, and peaks whose waveform shapes are so unique that it is difficult to determine whether they are a single peak region or an unseparated peak region.

[0090] (Item 9) The waveform analysis method described in item 5 may further include a correction step of unifying the two or more different types of peak regions into a peak region of the type having the longest region among them when it is determined in the detection step that the region estimation result is invalid because the two or more different types of peak regions are adjacent along the horizontal axis of the waveform to be analyzed.

[0091] (Clause 23) In the waveform analysis device described in clause 19, the detection unit may further include a correction unit that, when it is determined that the region estimation result is invalid because the two or more different types of peak regions are adjacent to each other along the horizontal axis of the waveform to be analyzed, unifies the two or more different types of peak regions into a peak region of the type having the longest region among them.

[0092] According to the waveform analysis method described in paragraph 9 and the waveform analysis device described in paragraph 23, even in the case of, for example, a peak whose start point and end point have not been detected, or a peak whose waveform shape is so special that it is difficult to determine whether it is a single peak region or an unseparated peak region, it is possible to identify the start point and / or end point of the peak, or to determine whether it is a single peak region or an unseparated peak region, and provide accurate peak information.

[0093] (Item 10) In the waveform analysis method described in item 1, the detection step may calculate one or more of the signal-to-noise ratio, resolution, symmetry coefficient, percentage of missing peaks, peak height, or peak width in the peak region, or the baseline slope in the non-peak region, and the calculation results may be used together with the determination of the validity of the region.

[0094] According to the waveform analysis method described in paragraph 10, in addition to the information on the sequence of the region obtained based on a judgment using machine learning, it is possible to more accurately detect peaks and obtain peak information by referring to numerical information obtained from the peak shape and its characteristics, or the baseline shape, etc.

[0095] (Item 11) In the waveform analysis method described in item 10, a threshold value that is a criterion for using the calculation result for judgment can be set by a user.

[0096] According to the waveform analysis method described in paragraph 11, more appropriate peak detection is possible depending on the type of sample, the purpose of the analysis, the status of the device, the analysis conditions, etc., and highly accurate peak information can be provided.

[0097] (Item 12) In the waveform analysis method described in item 1, the signal waveform is a chromatogram obtained by chromatography mass spectrometry, and in the detection step, a region estimation result in a chromatogram of a quantitative ion for the same component is compared with a region estimation result in a chromatogram of a confirmation ion, and the comparison result can be used together with the region validity judgment.

[0098] The quantification ion and confirmation ion derived from the same component should show peaks with roughly similar shapes in the extracted ion chromatograms. Therefore, even if a peak is not properly detected in one chromatogram for some reason, it is possible to obtain accurate peak information by using the peak detection result in the other chromatogram.

[0099] (Item 13) In the waveform analysis method described in item 1, the detection step can include comparing area estimation results in signal waveforms obtained from at least two of an unknown sample containing the target component, a standard sample containing the target component at a known concentration, and a blank sample not containing the target component, and using the comparison results together when determining the validity of the area.

[0100] A peak corresponding to the target component should appear in both chromatograms obtained for an unknown sample and a standard sample containing the same target component. If overlapping peaks and the effects of noise are ignored, the shapes of the peaks should be roughly similar. Therefore, even if a peak is not properly detected in one chromatogram due to some factor, it is possible to obtain highly accurate peak information by using the peak detection results in the other chromatogram.

[0101] On the other hand, the chromatogram obtained for the blank sample should have a baseline that is approximately the same as that of the chromatograms obtained for the unknown and standard samples, respectively. Therefore, even if the baseline is significantly inclined, the peak detection results in the chromatogram obtained for the blank sample can be used to judge the peak detection results based on the chromatograms obtained for the unknown or standard samples, and highly accurate peak information can be obtained.

[0102] (Item 14) In the waveform analysis method according to item 1, in the detection step, compound information on the target compound may also be used when determining the validity of the region.

[0103] The "compound information" referred to here can include not only information about the compound itself contained in the sample, such as concentration values, but also information about isomers of the compound, and information about derivatives generated by pretreatment, etc.

[0104] According to the waveform analysis method described in item 14, more appropriate peak detection is possible by utilizing compound information, and highly accurate peak information can be provided.

[0105] (Item 24) Furthermore, one aspect of the analytical device of the present invention may be any one of a chromatograph, a mass spectrometer, and a spectrometer, which includes a waveform analysis device described in any one of items 15 to 23 as a data analysis unit.

[0106] According to the analytical device described in paragraph 24, high quality qualitative and quantitative performance can be achieved by utilizing highly accurate peak information. [Explanation of symbols]

[0107] A…Analysis system 1...LC measurement section 2. Data analysis section 20…Data collection section 21...Peak detection processing unit 210...Waveform preprocessing unit 211…Judgment section 212…Model memory section 213...Tentative area determination unit 214...non-standard area detection unit 215…Area correction section 22...Qualitative / Quantitative Analysis Department 23...Display processing unit 24...Input section 25…Display section 3. Model Creation Section 30…Learning data storage unit 31…Learning Execution Department 32…Model memory section

Claims

1. A waveform analysis method for analyzing a signal waveform that is a chromatogram or a spectrum, comprising: A model creation step of creating a trained model for identifying peak portions contained in an input waveform by machine learning using a plurality of reference waveforms whose peak portion positions are known; a region estimation step of dividing the waveform to be analyzed into a plurality of partial waveforms, determining whether each of the plurality of partial waveforms is a peak portion using the trained model, and estimating a plurality of different types of regions, including peak regions and non-peak regions, in the waveform to be analyzed based on the results of the determination; a detection step for detecting an inappropriate region estimation result by determining whether or not the arrangement of each region estimated in the region estimation step along the horizontal axis of the analysis target waveform complies with a predetermined rule; The waveform analysis method includes the steps of:

2. 2. The waveform analysis method according to claim 1, wherein the model creation step creates a trained model that identifies peak portions contained in an input waveform by machine learning using multiple sets of partial waveforms created by dividing a reference waveform whose peak positions are known.

3. 2. The waveform analysis method according to claim 1, further comprising a correction step of correcting an estimated result of an area in the invalid portion detected in the detection step in accordance with a predetermined rule.

4. 2. The waveform analysis method according to claim 1, further comprising a notification step of notifying a user of an estimated region in the invalid portion detected in the detection step in a form that is visible to the user.

5. The trained model is created by machine learning using a plurality of reference waveforms whose peak start and end positions are known, The plurality of different types of regions include a peak region, a non-peak region, a peak start region, and a peak end region, 2. The waveform analysis method according to claim 1, wherein the predetermined rule is that a peak start region exists at one end of one peak region and a peak end region exists at the other end of one peak region.

6. 6. The waveform analysis method according to claim 5, wherein in the detection step, pairs of peak start regions and peak end regions are searched for along the horizontal axis of the waveform to be analyzed in ascending order of parameter values ​​on the horizontal axis, and if a peak region is sandwiched between the pairs of peak start regions and peak end regions found, the region estimation result is determined to be valid.

7. 6. The waveform analysis method according to claim 5, further comprising a correction step of designating the ends of a peak region as the peak start region and / or the peak end region when it is detected in the detection step that one or both of a peak start region and a peak end region do not exist at both ends of a peak region.

8. The peak region includes a single peak region having one peak and an unresolved peak region having multiple overlapping peaks, 6. The waveform analysis method according to claim 5, wherein in said detection step, when the two or more different types of peak regions are adjacent to each other along the horizontal axis of the waveform to be analyzed, it is determined that the region estimation result is invalid.

9. 6. The waveform analysis method according to claim 5, further comprising a correction step of unifying the two or more different types of peak regions into a peak region of a type having the longest region among them when it is determined in the detection step that the region estimation result is invalid because the two or more different types of peak regions are adjacent to each other along the horizontal axis of the waveform to be analyzed.

10. 2. The waveform analysis method according to claim 1, wherein in the detection step, one or more of the S / N ratio, resolution, symmetry coefficient, percentage of missing peaks, peak height, or peak width in the peak region, or baseline slope in the non-peak region are calculated, and the calculation results are used together when determining the validity of the region.

11. The waveform analysis method according to claim 10 , wherein a threshold value that is a criterion for using the calculation result for judgment can be set by a user.

12. 2. The waveform analysis method according to claim 1, wherein the signal waveform is a chromatogram obtained by chromatography mass spectrometry, and in the detection step, a region estimation result in a chromatogram of a quantitative ion for the same component is compared with a region estimation result in a chromatogram of a confirmation ion, and the comparison result is used together with the region estimation result when determining the validity of the region.

13. 2. The waveform analysis method according to claim 1, wherein in the detection step, region estimation results in signal waveforms obtained from at least two of an unknown sample containing the target component, a standard sample containing the target component at a known concentration, and a blank sample not containing the target component are compared, and the comparison results are used together when determining the validity of the region.

14. 2. The waveform analysis method according to claim 1, wherein in the detection step, compound information regarding the target compound is also used when determining the validity of the region.

15. A waveform analyzer for analyzing a signal waveform that is a chromatogram or a spectrum, comprising: a region estimation unit that divides a waveform to be analyzed into a plurality of partial waveforms, and determines whether each of the plurality of partial waveforms of the waveform to be analyzed is a peak portion or not using a trained model created by machine learning using a plurality of reference waveforms whose peak portion positions are known, and estimates a plurality of different types of regions, including peak regions and non-peak regions, in the waveform to be analyzed based on the results of the determination; a detection unit that detects inappropriate region estimation results by determining whether or not the arrangement of each region estimated by the region estimation unit along the horizontal axis of the waveform to be analyzed complies with a predetermined rule; A waveform analysis device comprising:

16. The waveform analysis device according to claim 15, wherein the machine learning model is created by machine learning using multiple sets of multiple partial waveforms created by dividing a reference waveform whose peak positions are known.

17. The waveform analysis device according to claim 15, further comprising a correction unit that corrects an estimation result of an area in the invalid portion detected by the detection unit in accordance with a predetermined rule.

18. 16. The waveform analysis device according to claim 15, further comprising a notification unit that notifies a user of an estimation result of the region in the invalid portion detected by the detection unit in a form that is visible to the user.

19. The trained model is created by machine learning using a plurality of reference waveforms whose peak start and end positions are known, The plurality of different types of regions include a peak region, a non-peak region, a peak start region, and a peak end region, 16. The waveform analyzer according to claim 15, wherein the predetermined rule is that a peak start region exists at one end of one peak region and a peak end region exists at the other end of one peak region.

20. 20. The waveform analysis device according to claim 19, wherein the detection unit searches for pairs of peak start regions and peak end regions along the horizontal axis of the waveform to be analyzed in ascending order of parameter values ​​on the horizontal axis, and when a peak region is sandwiched between the pairs of peak start regions and peak end regions found, the detection unit determines that the region estimation result is valid.

21. 20. The waveform analysis device according to claim 19, wherein the detection unit further comprises a correction unit that, when it is detected that one or both of a peak start region and a peak end region do not exist at both ends of a peak region, sets the ends of the peak region as the peak start region and / or the peak end region.

22. The peak region includes a single peak region having one peak and an unresolved peak region having multiple overlapping peaks, 20. The waveform analysis device according to claim 19, wherein the detection section determines that the region estimation result is invalid when the two or more different types of peak regions are adjacent to each other along the horizontal axis of the waveform to be analyzed.

23. 20. The waveform analysis device according to claim 19, wherein the detection unit further comprises a correction unit that, when it is determined that the region estimation result is invalid because the two or more different types of peak regions are adjacent to each other along the horizontal axis of the waveform to be analyzed, unifies the two or more different types of peak regions into a peak region of a type that has the longest region among them.

24. An analysis device, which is any one of a chromatography device, a mass spectrometer, and a spectrometer, comprising the waveform analysis device according to any one of claims 15 to 23 as a data analysis unit.

Images