Detection device, detection method, computer program product including detection program, and detection system

Abstract

The invention provides a detection apparatus, a detection method, a computer program product including a detection program, and a detection system. The detection device includes: an acquisition unit that acquires a plurality of pieces of detection data corresponding to each of a plurality of samples; and a calculation unit that processes the plurality of detection data. The calculation unit acquires an identification result indicating that peak information of the signal strength extracted from each of the plurality of pieces of detection data is identified as peak information of the signal strength corresponding to the target component, and when there is outlier peak information among the plurality of pieces of peak information identified in each of the plurality of pieces of detection data, the identification result indicates that the peak information of the signal strength extracted from each of the plurality of pieces of detection data is identified as the peak information of the signal strength corresponding to the target component. It is detected that the identification result corresponding to the outlier peak information is abnormal.

Description

Technical Field

[0001] The present disclosure relates to a detection device, a detection method, a computer program product including a detection program, and a detection system. Background Art

[0002] As disclosed in Japanese Patent Laid-Open No. 2023-012485, there is known a technique of chromatography that uses a chromatograph to separate components contained in a sample to be analyzed. Chromatography is applied to analysis in the field of metabolomics that comprehensively analyzes specific molecules produced by cell activities, or pesticide residue analysis, etc. The chromatograph of Japanese Patent Laid-Open No. 2023-012485 automatically identifies the target component by detecting the peak of the signal intensity of the chromatogram using a peak detection algorithm. Summary of the invention

[0003] The sample to be analyzed by the chromatograph may be mixed with an impurity component, or even if the sample is obtained under the same conditions, the type and concentration of the components contained in the sample may deviate. In the chromatograph described in Japanese Patent Publication No. 2023-012485, since the influence of the mixing of impurities or the deviation of components is not particularly considered, the target component may not be properly identified. For example, there is a concern that an erroneous peak may be identified as a peak of the target component, or a part of the peak shape may be identified as an erroneous shape.

[0004] As a method for judging whether a desired peak is properly identified as a peak of a target component by the action of a peak detection algorithm, it is necessary to confirm it through visual recognition by a user. In the analysis in the field of metabolomics and pesticide residue analysis, the number of samples and the number of target components sometimes reach hundreds. It is complicated to confirm whether each of the hundreds of target components contained in hundreds of samples is properly identified through visual recognition by a user.

[0005] The present disclosure is made to solve the above-mentioned problems, and an object of the present disclosure is to appropriately identify a peak in a chromatogram as a peak of a target component.

[0006] The detection device according to one aspect of the present disclosure includes: an acquisition unit that acquires a plurality of detection data corresponding to each of a plurality of samples detected by a chromatograph; and a calculation unit that processes the plurality of detection data acquired by the acquisition unit. The calculation unit acquires an identification result, wherein the identification result indicates that the peak information of the signal intensity extracted from each of the plurality of detection data is identified as the peak information of the signal intensity corresponding to the target component, and when there is outlier peak information among the plurality of peak information identified in each of the plurality of detection data, the identification result corresponding to the outlier peak information is detected to be abnormal.

[0007] According to another aspect of the present disclosure, a detection method is a method for detecting abnormalities in identification results of target components detected by a computer, which includes: a step of acquiring a plurality of detection data corresponding to each of a plurality of samples detected by a chromatograph; and a step of processing the plurality of detection data acquired by the acquisition step. The processing step includes: a step of acquiring an identification result, wherein the identification result indicates that the peak information of the signal intensity extracted from each of the plurality of detection data is identified as the peak information of the signal intensity corresponding to the target component; and a step of detecting that the identification result corresponding to the peak information of the outlier is abnormal when there is peak information of an outlier among the plurality of peak information identified in each of the plurality of detection data.

[0008] A computer program product including a detection program according to another aspect of the present disclosure is a computer program product including a detection program for detecting abnormalities in identification results of target components, which causes a computer to execute: a step of acquiring a plurality of detection data corresponding to each of a plurality of samples detected by a chromatograph; and a step of processing the plurality of detection data acquired by the acquisition step. The processing step includes: a step of acquiring an identification result, wherein the identification result indicates that peak information of signal intensity extracted from each of the plurality of detection data is identified as peak information of signal intensity corresponding to the target component; and a step of detecting that an identification result corresponding to the peak information of the outlier is abnormal when there is peak information of an outlier among the plurality of peak information identified in each of the plurality of detection data.

[0009] According to another aspect of the present disclosure, a detection system includes: a chromatograph; and a detection device for detecting abnormalities in identification results of target components. The detection device includes: an acquisition unit for acquiring multiple detection data corresponding to each of multiple samples detected by the chromatograph; and a calculation unit for processing the multiple detection data acquired by the acquisition unit. The calculation unit acquires an identification result, wherein the identification result indicates that the peak information of the signal intensity extracted from each of the multiple detection data is identified as the peak information of the signal intensity corresponding to the target component, and when there is outlier peak information in the multiple peak information identified in each of the multiple detection data, the abnormality of the identification result corresponding to the outlier peak information is detected.

[0010] The above and other objects, features, aspects and advantages of the present invention will become more apparent from the following detailed description of the present invention when taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 This is a diagram showing the configuration of a detection system and a detection device according to the first embodiment.

[0012] Figure 2 : is a flowchart showing the process of obtaining the identification result.

[0013] Figure 3 This is a diagram showing a display example of a chromatogram generated by the detection device in the first embodiment.

[0014] Figure 4 This is a diagram showing an example of a table of peak information in the first embodiment.

[0015] Figure 5 is a flowchart showing the abnormality detection process of the identification result.

[0016] Figure 6 It means executing Figure 5 A diagram showing an example of a chromatogram displayed after the flowchart.

[0017] Figure 7 It means executing Figure 5 A diagram showing an example of a table of peak information displayed after the flowchart.

[0018] Figure 8 This is a diagram showing a display example of a chromatogram generated by the detection device in the second embodiment.

[0019] Fig. 9 This is a diagram showing an example of a table of peak information in the second embodiment.

[0020] Fig.10 It means that in implementation mode 2, when executing Figure 5 A diagram showing an example of a chromatogram displayed after the flowchart.

[0021] Fig.11 It means that in implementation mode 2, Figure 5 A diagram showing an example of a table of peak information displayed after the flowchart. DETAILED DESCRIPTION

[0022] [Implementation Method 1]

[0023] The present embodiment will be described in detail with reference to the accompanying drawings. In addition, the same reference numerals are attached to the same or corresponding parts in the drawings, and their description will not be repeated in principle.

[0024] <Overall structure of the detection system and detection device>

[0025] Reference Figure 1 The configurations of the detection system 1 and the detection device 100 according to the first embodiment will be described. Figure 1 1 is a diagram showing the structure of the detection system 1 and the detection device 100 according to the first embodiment. Figure 1 As shown, the detection system 1 includes a chromatograph 10 and a detection device 100 .

[0026] The chromatograph 10 includes a container 11, a liquid delivery pump 12, a syringe 13, a column 14, and a detector 15. The container 11 contains a mobile phase. The liquid delivery pump 12 draws the mobile phase from the container 11 and feeds it at a constant flow rate. The syringe 13 injects a sample to be analyzed into the mobile phase fed by the liquid delivery pump 12.

[0027] The column 14 contains a stationary phase and separates various components contained in the sample injected by the syringe 13. The detector 15 detects the components eluted from the column 14. As the detector 15, for example, an absorbance detector (photodiode array (PDA) detector), a fluorescence detector, a differential refractive index detector, a conductivity detector, or a mass spectrometer is used. Detection data indicating the signal intensity corresponding to the components in the sample detected by the detector 15 is output to the detection device 100.

[0028] Furthermore, the chromatograph 10 of the first embodiment is a liquid chromatograph (LC: Liquid Chromatograph) using a liquid as a mobile phase, but the chromatograph 10 may be another chromatograph such as a gas chromatograph using a gas as a mobile phase.

[0029] In the chromatograph 10, a sample to be analyzed is injected into the mobile phase by the syringe 13. The injected sample reaches the column 14 by the flow of the mobile phase fed by the liquid delivery pump 12, and passes through the column 14. The time for various components contained in the sample to pass through the column 14 is different depending on the affinity between the stationary phase and the mobile phase. For example, the time (also called "retention time") for the components contained in the sample that are easily adsorbed to the stationary phase to pass through the column 14 is longer than that for the components that are not easily adsorbed to the stationary phase. As a result, the various components contained in the sample are separated in the time direction by the column 14.

[0030] The eluate containing the components separated in the column 14 is introduced from the column 14 to the detector 15. The detector 15 outputs detection data indicating the signal intensity corresponding to the concentration (amount) of the components introduced from the column 14. A chromatogram is generated by processing the detection data by the detection device 100. In addition, the solution that has passed through the detector 15 is discharged as waste liquid.

[0031] The detection device 100 may be a general-purpose computer or a computer dedicated to the detection system 1 for processing detection data from the chromatograph 10. The detection device 100 includes a computing device 101, a memory 102, a storage device 103, an interface 104, a display device 110, and an input device 120.

[0032] The computing device 101 is an example of a "computing unit". The computing device 101 is a computing subject (computer) that performs various processing by executing various programs. The computing device 101 includes, for example, a processor such as a central processing unit (CPU) or a microprocessor (MPU). In addition, the processor as an example of the computing device 101 has the function of performing various processing by executing programs, and a dedicated hardware circuit such as an application-specific integrated circuit (ASIC) or a field-programmable gate array (FPGA) can also be used to implement part or all of these functions.

[0033] The term "processor" is not limited to a processor in a narrow sense, which executes processing in a memory program manner like a CPU or an MPU, and may include a hard-wired circuit such as an ASIC or an FPGA. Therefore, the "processor" as an example of the computing device 101 may also be replaced by a processing circuit (processing circuitry) in which processing is predefined by a computer-readable code and / or a hard-wired circuit.

[0034] In addition, the computing device 101 may be composed of one chip or multiple chips. Furthermore, the processor and the associated processing circuits may include multiple computers connected to each other in a wired or wireless manner via a local area network or a wireless network. The processor and the associated processing circuits may also include a cloud computer that performs calculations remotely based on input data and outputs the calculation results to another device located in a remote location.

[0035] The memory 102 includes a volatile storage area (e.g., a work area) that temporarily stores program codes or work memory when the computing device 101 executes various programs. Examples of the memory 102 include volatile memories such as dynamic random access memory (DRAM) and static random access memory (SRAM), or nonvolatile memories such as read-only memory (ROM) and flash memory.

[0036] The storage device 103 stores various programs or various data executed by the computing device 101. The storage device 103 may be one or more non-transitory computer readable mediums, or one or more computer readable storage mediums. Examples of the storage device 103 include a hard disk drive (HDD) and a solid state drive (SSD).

[0037] The storage device 103 of Embodiment 1 stores a detection program 130 for executing a process for detecting abnormalities in the detection data acquired by the operation device 101 from the chromatograph 10 , and an acquisition program 135 for acquiring identification results of target components described later.

[0038] The interface 104 transmits and receives data with an external device or an external machine via wired communication or wireless communication. For example, the interface 104 acquires detection data output from the chromatograph 10 by communicating with the chromatograph 10. In addition, the interface 104 may be a communication device that communicates with a cloud server (not shown) to send the detection data acquired from the chromatograph 10 to the cloud server, or sends the execution result of the detection processing of the abnormal data of the computing device 101 to the cloud server.

[0039] Furthermore, the interface 104 can transmit and receive data with the display device 110 or the input device 120 as a user interface via wired communication or wireless communication. The detection device 100 is not limited to one interface 104, and may include a plurality of interfaces 104 according to the number of communication objects.

[0040] The display device 110 is, for example, a display including a liquid crystal panel, etc., and displays the execution result of the detection process based on the abnormal data of the detection device 100. The input device 120 is, for example, a pointing device such as a keyboard or a mouse, and receives instructions from the user. In the case of using a touch panel as a user interface, the display device 110 and the input device 120 may be formed integrally. In addition, the display device 110 and the input device 120 may be provided separately from the detection device 100.

[0041] <Acquisition of identification results>

[0042] In the detection system 1 configured as described above, the detection device 100 acquires detection data indicating the signal intensity corresponding to the components in the sample detected by the chromatograph 10 in a time series via the interface 104. The detection device 100 generates a chromatogram indicating the time variation of the signal intensity based on the acquired detection data. The chromatogram includes a plurality of peak-shaped waveforms (also referred to as "peaks" for short). The plurality of peaks contained in the chromatogram correspond to the plurality of components in the sample, respectively.

[0043] The detection device 100 in Embodiment 1 uses a peak detection algorithm to automatically determine which peak among the multiple peaks included in the chromatogram corresponds to a target component. The target component is a component to be analyzed among the multiple components included in the sample and is pre-set by the user of the detection system 1.

[0044] As described above, the components in the sample are separated by the column 14. Peaks in the chromatogram are formed at a retention time specific to each component. The theoretical retention time of the target component (also referred to as "theoretical retention time") is input in advance by the user and stored in the storage device 103.

[0045] The detection device 100 automatically determines a peak formed within a retention time close to the theoretical retention time of the target component among the multiple peaks in the chromatogram as a peak corresponding to the target component. In this way, the act of determining one of the multiple peaks contained in the chromatogram as the peak of the target component is called "identification".

[0046] The detection device 100 obtains information related to the peak of the peak identified as the target component (also referred to as "peak information") as the identification result. The peak information includes, for example, the maximum value of the signal intensity, the holding time when the signal intensity is the maximum value (also referred to as the "peak holding time"), the holding time when the signal intensity starts to rise (referred to as the "peak start holding time"), the holding time when the signal intensity ends to fall (referred to as the "peak end holding time"), and the area of ​​the peak shape (referred to as the "peak area"), the position of the baseline described later, and the intensity obtained by subtracting the intensity of the baseline from the maximum value of the signal intensity (also referred to as the "peak height") and other information.

[0047] The detection device 100 uses the peak height, peak area, and a pre-prepared calibration curve to calculate the concentration of the target component contained in the sample. As described above, the detection device 100 of embodiment 1 obtains an identification result indicating that the peak information of the signal intensity extracted from the detection data is identified as the peak information of the signal intensity corresponding to the target component.

[0048] Figure 2 : is a flowchart showing the process of obtaining the identification result. Figure 2The processing steps shown in FIG. 1 are implemented by the computing device 101 executing the identification result acquisition program 135 .

[0049] like Figure 2 As shown, the computing device 101 acquires detection data indicating the signal intensity corresponding to the component in the sample detected by the chromatograph 10 (step S110). The computing device 101 generates a chromatogram based on the detection data (step S120).

[0050] The computing device 101 performs a correction process for the retention time (step S130). Even when the same component passes through the column 14, the retention time of the chromatogram may vary due to degradation of the column 14, etc. The process of step S130 is a process for correcting the variation of the retention time caused by degradation of the column 14, etc.

[0051] In Embodiment 1, a standard substance (e.g., alkane) is used to correct the variation of the retention time. In Embodiment 1, the standard substance is mixed in a sample and a chromatogram is generated. The retention time of the standard substance in a state where the column 14 is not degraded is input to the detection device 100 in advance. The detection device 100 uses the retention time of the standard substance in the generated chromatogram and the retention time of the standard substance in a state where the column 14 is not degraded to determine the degree of degradation of the column 14, so as to correct the retention time of the entire chromatogram.

[0052] In one aspect, the retention time can be corrected by using the retention time of an arbitrary component when the column 14 is not degraded and the retention time when the arbitrary component is actually detected without using a standard substance. On the other hand, the retention time can be corrected by correcting the flow rate of the mobile phase. In this way, the detection device 100 of Embodiment 1 takes into account the change in the retention time caused by external factors such as the degradation of the column 14, and performs a correction process on the retention time of the chromatogram in advance.

[0053] return Figure 2 The computing device 101 uses a peak detection algorithm to obtain the identification result of the target component from the multiple peaks included in the chromatogram whose retention time is corrected in step S130 (step S140), and ends the processing. Figure 2 The processing of steps S110 to S140 shown in FIG. 1 is a processing for obtaining an identification result of one target component for one sample.

[0054] <Identification results for multiple samples>

[0055] exist Figure 2The process of obtaining the identification result of one target component in one sample is described in the embodiment 1. In the embodiment 1, the identification result of the target component is obtained from a plurality of samples. That is, the detection device 100 of the embodiment 1 executes the detection according to the number of the plurality of samples prepared in advance. Figure 2 As described above, in the analysis in the field of metabolomics and pesticide residue analysis, the number of samples reaches hundreds.

[0056] More specifically, in a pesticide residue analysis for analyzing whether the shipped agricultural products contain pesticides, a portion of the shipped products is randomly selected as samples. The following describes an example of selecting 100 shipped products as samples. In this case, the detection device 100 performs a detection on each of the 100 shipped products. Figure 2 Flow chart of . The 100 shipments are agricultural products of the same type. More specifically, when the shipment is bananas, for example, the 100 shipments do not include agricultural products other than bananas, and the 100 shipments are all bananas. Therefore, the 100 samples have substantially the same components.

[0057] In the first embodiment, the case where the target component A, which is a pesticide, is set as the target component is described. The theoretical retention time of the target component A is 7.14 minutes. In the first embodiment, the control is performed for each of the 100 shipped products. Figure 2 The detection device 100 performs the following identification processing on each of the 100 chromatograms: the peak formed within the retention time close to 7.14 minutes is automatically determined as the peak corresponding to the target component A.

[0058] Figure 3 FIG. 1 is a diagram showing a display example of a chromatogram generated by the detection device 100 in the first embodiment. Figure 3 Graphs C1, C2, and C3 corresponding to three samples S1, S2, and S3 among the 100 samples are shown in FIG. In each of the chromatograms C1, C2, and C3, the horizontal axis represents the retention time (minutes) and the vertical axis represents the signal intensity (absorbance). Figure 3 Chromatograms C1 to C3 shown in FIG. 1 are displayed to the user by the display device 110 as image data.

[0059] As described above, the 100 samples are agricultural products of the same type. Therefore, most of the components contained in the 100 samples have common components. In the example of Embodiment 1, an example is described in which most of the 100 samples contain target component A. However, Figure 3 As shown, the peak shapes included in the chromatograms C1 to C3 may differ due to mixing of foreign components into the sample or deviation of components.

[0060] exist Figure 3 In the example, in chromatograms C1 and C2, a peak with a signal intensity exceeding "5000" is formed at the holding time "7.14 minutes". On the other hand, in chromatogram C3, a peak with a signal intensity less than "5000" and about "2000" is formed at the holding time "7.14 minutes". In addition, in chromatogram C3, a peak with a signal intensity of about "3000" is formed at the holding time "7.30 minutes". On the other hand, in chromatograms C1 and C2, no peak is formed at the holding time "7.30 minutes".

[0061] Thus, even if the same type of samples S1, S2, and S3 are collected under the same conditions, the peak shapes are different due to the mixing of foreign components or the deviation of components. As described above, the detection device 100 of the first embodiment uses the peak detection algorithm to identify the target component A in each chromatogram C1, C2, and C3. Figure 3 In the figure, white triangular images P1, P2, and P3 are displayed in the peaks of the chromatograms C1 to C3, respectively. The images P1, P2, and P3 indicate that the peaks shown by the images P1, P2, and P3 are identified as the target component A.

[0062] That is, image P1 indicates that the peak at the retention time "7.14 minutes" of chromatogram C1 is the peak of target component A. Image P2 indicates that the peak at the retention time "7.14 minutes" of chromatogram C2 is the peak of target component A. On the other hand, image P3 indicates that the peak at the retention time "7.30 minutes" of chromatogram C3 is the peak of target component A.

[0063] like Figure 3 As shown, in chromatogram C3, an erroneous peak (a peak with a signal intensity of about "3000") is identified as target component A. This is because the concentration of target component A contained in sample S3 is lower than that in other samples, and due to the deviation of the components, an impurity component with a retention time of "7.30 minutes" is mixed into sample S3.

[0064] Therefore, the detection device 100 mistakenly identifies the peak corresponding to the inclusion component formed at the holding time "7.30 minutes" as the target component A. Figure 3 In the example shown, since the target component A is automatically identified using the peak detection algorithm, an erroneous peak in some samples is identified as the target component A. Therefore, in the first embodiment, the erroneous identification is detected as an abnormality by the method shown below.

[0065] <Abnormality detection and processing of identification results>

[0066] The detection device 100 acquires the peak information of the target component A identified in each of the chromatograms C1 to C3 . Figure 4 This is a diagram showing an example of a table of peak information in the first embodiment. Figure 4 The table shown is similar to Figure 3 Similarly, the image data is displayed to the user by the display device 110. In the first embodiment, the peak information is the peak holding time. The detection device 100 obtains the peak holding time corresponding to the identified peak and generates Figure 4 The table shown. Figure 4 The table shown is stored by the storage device 103 .

[0067] like Figure 4 As shown, the peak retention time of the target component A in the chromatogram C1 is the retention time "7.14 minutes". The peak retention time of the target component A in the chromatogram C2 is the retention time "7.14 minutes". The peak retention time of the target component A in the chromatogram C3 is the retention time "7.30 minutes". Although the illustration is omitted, Figure 4 The table shown includes the peak holding times for 100 samples.

[0068] Then, the detection device 100 executes the detection program 130 for performing abnormality detection processing. Figure 5 is a flowchart showing the abnormality detection process of the identification result. Figure 5 The processing steps shown are implemented by the computing device 101 executing the detection program 130 .

[0069] The computing device 101 Figure 4 In the first embodiment, the computing device 101 uses Hoteling's T 2 More specifically, the computing device 101 calculates the peak holding time x of each sample of the total number of samples N (100 in the first embodiment). i In addition, the computing device 101 calculates the unbiased variance σ of the peak holding time of all 100 samples. 2 The unbiased variance σ 2 It is expressed by the following formula 1.

[0070] [Formula 1]

[0071]

[0072] The computing device 101 uses the mean μ and the unbiased variance σ 2 To calculate the peak holding time of each wave, χi The abnormality a is expressed by the following formula 2.

[0073] [Formula 2]

[0074]

[0075] When the abnormality a of the peak holding time exceeds a predetermined threshold, the computing device 101 determines that the peak holding time is an outlier. When the abnormality a of the peak holding time is below the threshold, the computing device 101 determines that the peak holding time is not an outlier. The threshold is 10, for example.

[0076] return Figure 5 , the computing device 101 ends the processing when there is no outlier (NO in step S220). When there is an outlier (YES in step S220), the computing device 101 detects the identification result corresponding to the outlier as abnormal data (step S230). The computing device 101 emphasizes and displays the abnormal data (step S240). More specifically, the computing device 101 displays the peak holding time of the outlier on the display device 110 in a more emphasized manner than other peak holding times different from the peak holding time of the outlier.

[0077] The computing device 101 obtains a representative value based on a peak holding time that is different from the peak holding time of the outlier (step S250). The so-called representative value is a value that represents a normal value that is not an outlier, and is, for example, an average value of a peak holding time that is different from the peak holding time of the outlier. Specifically, in the example of implementation mode 1, when only the peak holding time of sample S3 is detected as an outlier, the average value of the peak holding time in 99 samples other than sample S3 is calculated as the representative value. In the example of implementation mode 1, the holding time "7.14 minutes" that is the same as the theoretical holding time is calculated as the representative value. Thereafter, the computing device 101 replaces the peak holding time of the outlier with the representative value (step S260).

[0078] Figure 6 It means executing Figure 5 The following figure shows an example of a chromatogram displayed after the flowchart. Figure 6 As shown, the computing device 101 displays an image P3C instead of the triangular image P3 indicating an erroneous peak. Unlike the image P3, the image P3C is added with color information. For example, the image P3C is displayed in red. As described above, since the images P1 and P2 are white, the red image P3C is highlighted. Thus, in the detection device 100, the user can easily recognize that the peak shown in the image P3 is erroneous.

[0079] Furthermore, if Figure 6 As shown, the computing device 101 uses the representative value replaced in step S260 to re-identify the peak existing near the representative value as the peak of the target component A. Figure 6 , an image P3A representing a peak identified based on a representative value is illustrated. Image P3A is displayed larger than images P1 and P2. Thus, in the detection device 100, a user can easily grasp the position of a peak after an outlier is replaced by a representative value. Furthermore, in the example of embodiment 1, an arrow-shaped image Ar1 is displayed, which represents the position of a peak before and after an outlier is replaced by a representative value. Thus, in embodiment 1, by displaying the arrow-shaped image Ar1, a user can easily grasp the position of a peak before and after an outlier is replaced by a representative value.

[0080] Figure 7 It means executing Figure 5 An example of a table of peak information displayed after the flowchart. Figure 7 As shown in FIG. 1 , color information is added to the background of the cell representing the peak retention time of the chromatogram C3 in which the erroneous identification was performed. Specifically, Figure 7 In the table, the background of the cell indicating the peak holding time "7.30 minutes" is red. Thus, in the detection device 100, the user can easily recognize that the peak holding time "7.30 minutes" is an erroneous peak holding time.

[0081] In this way, the detection device 100 of embodiment 1 can detect the identification result that is erroneously identified using an outlier as abnormal data. Thus, in embodiment 1, by displaying the abnormal data to the user and prompting the user to correct the abnormal data, the peak of the chromatogram can be appropriately identified as the peak of the target component. In addition, as described above, the detection device 100 automatically replaces the outlier with a representative value, and the peak of the chromatogram can be automatically and appropriately identified as the peak of the target component without requiring the user to perform a correction operation. Furthermore, in the example of embodiment 1, since the abnormality is judged by whether it is an outlier, it is possible to suppress the situation where the judgment of whether the identification result is abnormal is different due to different users, compared with the situation where the presence or absence of an abnormality is judged by the user's visual recognition.

[0082] [Implementation Method 2]

[0083] In the first embodiment, an example is described in which the peak holding time is used to determine whether there is an outlier and only one target component A in a sample is used as the analysis object. In the second embodiment, an example is described in which information different from the peak holding time is used to determine whether there is an outlier and multiple target components are used as the analysis objects. In addition, in the second embodiment, the structure that overlaps with the structure of the first embodiment is not repeatedly described.

[0084] In Embodiment 2, the peak information used for outlier detection is the peak start holding time and the peak end holding time. As described above, the so-called peak start holding time is the holding time when the signal intensity starts to rise, and the so-called peak end holding time is the holding time when the signal intensity ends to decrease. The peak start holding time and the peak end holding time are automatically detected by the peak detection algorithm that reads the chromatogram. More specifically, the detection device 100 detects the holding time when the signal intensity increases from a state where the signal intensity is lower than a prescribed threshold and reaches a signal intensity above the prescribed threshold as the "peak start holding time". In addition, the detection device 100 detects the holding time when the signal intensity decreases from a state where the signal intensity is higher than a prescribed threshold and reaches a signal intensity below the prescribed threshold as the "peak end holding time".

[0085] In Embodiment 2, target component B is analyzed in addition to target component A. The theoretical retention time of target component B is "8.05 minutes". In addition, the theoretical peak start retention time of target component B is "8.01 minutes", and the theoretical peak end retention time is "8.09 minutes".

[0086] Figure 8 FIG. 2 is a diagram showing a display example of a chromatogram generated by the detection device 100 in Embodiment 2. Figure 8 Zhongyu Figure 3 Similarly, chromatograms C1, C2, and C3 corresponding to three samples S1, S2, and S3 among the 100 samples are shown. Figure 8 In FIG. 1 , only the peak identified as the target component A is illustrated, and the peak identified as the target component B is omitted.

[0087] like Figure 8 As shown, in the chromatogram C1, in the peak identified as the target component A, the signal intensity starts to rise from the retention time "6.10 minutes" time point, and the decrease in signal intensity ends at the retention time "6.21 minutes". Image PS1 shows that the peak of chromatogram C1 starts at the retention time "6.10 minutes". Image PE1 shows that the peak of chromatogram C1 ends at the retention time "6.21 minutes". Figure 8 As shown, the detection device 100 draws a baseline Bs1 between the retention time when the signal intensity starts to rise and the retention time when the signal intensity ends to fall in the chromatogram C1. The baseline Bs1 is a line used when calculating the peak area and peak height, and is drawn by the detection device 100.

[0088] In addition, in the chromatogram C2, in the peak identified as the target component A, the signal intensity starts to rise from the retention time "6.11 minutes" time point, and the decrease in signal intensity ends at the retention time "6.23 minutes". Image PS2 shows that the peak of chromatogram C2 starts at the retention time "6.11 minutes". Image PE2 shows that the peak of chromatogram C2 ends at the retention time "6.23 minutes". Figure 8 As shown, in the chromatogram C2, a baseline Bs2 is drawn between the retention time at which the signal intensity starts to increase and the retention time at which the signal intensity ends to decrease.

[0089] Furthermore, in chromatogram C3, in the peak identified as target component A, the signal intensity starts to rise from the retention time "6.11 minutes" and the decrease in signal intensity ends at the retention time "6.34 minutes". Image PS3 shows that the peak of chromatogram C3 starts at the retention time "6.11 minutes". Image PE3 shows that the peak of chromatogram C3 ends at the retention time "6.34 minutes". Figure 8 As shown, in the chromatogram C3, a baseline Bs3 is drawn between the retention time at which the signal intensity starts to increase and the retention time at which the signal intensity ends to decrease.

[0090] In Embodiment 2, an impurity component with a peak between the retention time "6.25 minutes" and the retention time "6.35 minutes" is mixed into the sample S3. Therefore, the detection device 100 detects the peak end retention time of the peak of the target component A in the chromatogram C3 as "6.34 minutes" which is relatively longer than that of the chromatograms C1 and C2. Therefore, the detection device 100 incorrectly calculates the peak area of ​​the target component A in the chromatogram C3. That is, the detection device 100 obtains an incorrect identification result in the chromatogram C3.

[0091] Fig. 9 1 is a diagram showing an example of a table of peak information in Embodiment 2. The detection device 100 in Embodiment 2 obtains the peak start retention time and the peak end retention time of the peak of the target component A identified in each of the chromatograms C1 to C3 as peak information. The detection device 100 uses a peak detection algorithm to obtain the peak start retention time and the peak end retention time of the identified peak, and generates Fig. 9 The table shown. Fig. 9 The table shown is similar to Figure 4 The same is stored in the storage device 103 .

[0092] like Fig. 9As shown, the peak start retention time of the target component A in the chromatogram C1 is the retention time "6.10 minutes". In addition, the peak end retention time of the target component A in the chromatogram C1 is the retention time "6.21 minutes". The peak start retention time of the target component A in the chromatogram C2 is the retention time "6.11 minutes". In addition, the peak end retention time of the target component A in the chromatogram C2 is the retention time "6.23 minutes". The peak start retention time of the target component A in the chromatogram C3 is the retention time "6.11 minutes". In addition, the peak end retention time of the target component A in the chromatogram C3 is the retention time "6.34 minutes".

[0093] As described above, in the detection device 100 of the second embodiment, in addition to the target component A, the target component B is also taken as the analysis object. Similar to the target component A, the detection device 100 of the second embodiment uses the peak detection algorithm to automatically identify the peak corresponding to the target component B, and generates a peak detection algorithm based on the identification result. Fig. 9 The table shown.

[0094] like Fig. 9 As shown, the peak start holding time of the target component B in the chromatogram C1 is the holding time "8.01 minutes". In addition, the peak end holding time of the target component B in the chromatogram C1 is the holding time "8.09 minutes". The peak start holding time of the target component B in the chromatogram C2 is the holding time "8.00 minutes". In addition, the peak end holding time of the target component B in the chromatogram C2 is the holding time "8.08 minutes". The peak start holding time of the target component B in the chromatogram C3 is the holding time "8.03 minutes". In addition, the peak end holding time of the target component B in the chromatogram C3 is the holding time "8.08 minutes". Fig. 9 In the example of chromatograms C1 to C3, no abnormality in the identification result of the target component B occurs.

[0095] In the second embodiment, the detection device 100 also performs Figure 5 In the second embodiment, the peak information for outlier detection is not the peak holding time, but the peak start holding time and the peak end holding time. In addition, in the second embodiment, since the two components, target component A and target component B, are used as target components, two steps corresponding to target component A and target component B are performed for one sample. Figure 5 Flowchart of the process.

[0096] Fig.10 It means that in implementation mode 2, when executing Figure 5 The following figure shows an example of a chromatogram displayed after the flowchart. Fig.10As shown, the computing device 101 displays image PE3C instead of the triangular image PE3 which indicates the wrong peak as the peak end holding time of the target component A. Image PE3C is highlighted in red. Thus, in the second embodiment, in the detection device 100, the user can also recognize that the peak end holding time shown in image PE3 is wrong.

[0097] Furthermore, if Fig.10 As shown, the computing device 101 uses the representative value replaced in step S260 to re-identify the peak end retention time existing near the representative value as the peak end retention time of the target component A. Fig.10 , an image PE3A representing a peak identified based on a representative value is shown. The image PE3A is displayed with greater emphasis than other triangular images. Thus, in the detection device 100, the user can easily grasp the peak end holding time after replacement with the representative value. Fig.10 As shown, the detection device 100 draws the baseline Bs3A based on the change of the peak end holding time based on the representative value. Furthermore, in the second embodiment, by displaying the arrow-shaped image Ar2, the user can easily understand the position of the peak before and after the outlier is replaced by the representative value.

[0098] Fig.11 It means that in implementation mode 2, Figure 5 An example of a table of peak information displayed after the flowchart. Fig.11 As shown in FIG. 1 , color information is added to the background of the peak end retention time of the chromatogram C3 that was incorrectly identified. Specifically, Fig.11 In the table, the background of the cell with the peak end holding time of "6.34 minutes" is red and highlighted.

[0099] In addition, in Embodiment 2, the T based on the Hotelling's 2 The abnormality degree is calculated by the method. Fig.11 As shown, the abnormality of the peak start holding time of the target component A in the chromatogram C1 is the holding time "0.01", and the abnormality of the peak end holding time of the target component A in the chromatogram C1 is the holding time "0.04". In addition, the abnormality of the peak start holding time of the target component A in the chromatogram C2 is the holding time "0.03", and the abnormality of the peak end holding time of the target component A in the chromatogram C2 is the holding time "0.05".

[0100] Furthermore, the abnormality of the peak start holding time of the target component A in the chromatogram C3 is the holding time "0.03", and the abnormality of the peak end holding time of the target component A in the chromatogram C3 is the holding time "10.01". The detection device 100 determines that the peak end holding time of the target component A in the chromatogram C3 whose abnormality exceeds the threshold value "10" is abnormal.

[0101] In Embodiment 2, in addition to the cell of the peak end holding time "6.34 minutes", the detection device 100 also highlights the background of the cell representing the sample "Sample S3 (Chromatogram C3)" and the cell of the abnormality "10.01" in red. Thus, in Embodiment 2, the user can easily understand which sample is abnormal and the degree of abnormality when compared with other samples. In addition, in the case where multiple samples are detected as outliers, the detection device 100 can sort the display order according to the degree of abnormality. Thus, in Embodiment 2, the samples can be displayed to the user in order from low to high abnormality.

[0102] Thus, in Embodiment 2, it is also easy to detect an identification result that is erroneously identified using an outlier as abnormal. Thus, in Embodiment 2, a peak of a chromatogram can also be appropriately identified as a peak of a target component.

[0103] <Modification>

[0104] Hereinafter, another embodiment in which a part of the above-described embodiment is modified will be described.

[0105] In the above example, the peak information includes the peak holding time, the peak start holding time, the peak end holding time, the peak area, etc. However, the information included in the peak information is not limited to these. In one aspect, the peak information includes: the period from the peak start holding time to the peak end holding time (also called "peak width"), the period from the peak start holding time to the peak holding time (also called "peak front half width"), the period from the peak holding time to the peak end holding time (also called "peak rear half width"), the magnitude of the signal intensity when the signal intensity starts to rise (also called "peak start height"), and the magnitude of the signal intensity when the signal intensity ends to fall (also called "peak end height"). Furthermore, the peak information may include: the peak width at a specified intensity (N%), the peak front half width, the peak rear half width, and the peak rear half width. In addition, it may also include: the separation degree indicating the degree of separation from another adjacent peak, or the separation coefficient, the holding coefficient, the theoretical plate number indicating the column efficiency, the symmetry coefficient indicating the degree of symmetry of the peak, etc.

[0106] In the example described, for T in Hotelling 2Although the example in which the threshold value is "10" is described, the threshold value may be a number other than "10". For example, the threshold value may be "5", "20", or "30", and may be determined based on experiments.

[0107] In the example of Embodiment 1, the example of using the peak holding time as the peak information and the detection target of the outlier is described. However, the peak information used as the detection target of the outlier may be other peak information described above, for example, the difference between the peak holding time and the theoretical holding time.

[0108] In the above example, an example using shipped agricultural products as samples is described. However, as samples, for example, standard samples prepared in advance may be used to generate a calibration curve for determining the target component from peak information. That is, part of the plurality of samples may be standard samples artificially mixed with the target component in order to obtain information on the target component, rather than actual samples such as agricultural products actually shipped to the market.

[0109] In the above example, Hotelling T is used as the method for detecting the presence or absence of outliers. 2 However, in one aspect, outliers can be detected by other methods, for example, simply using the variance or the distance from the mean, or using the k-nearest neighbor method.

[0110] In the above example, the method of highlighting the image is described as adding color information to the image P3C and displaying the image P3A in an enlarged manner. However, the method of highlighting the image may be other methods, for example, it may also include: flashing the image, changing the shape of the image from a triangle, etc. In addition, in the above example, Figure 7 as well as Fig.11 The display of the table only shows the outliers before being replaced by the representative values, but the outliers can also be displayed together with the representative values. Furthermore, in the example, an example of adding red color information when the abnormality exceeds the threshold value "10" is described. However, color information can be added in stages according to the abnormality. That is, when the abnormality exceeds the threshold value "5" and is below the threshold value "10", the detection device 100 can add yellow color information, and when the abnormality exceeds the threshold value "10", red color information can be added.

[0111] [Example]

[0112] Those skilled in the art should understand that the multiple exemplary embodiments are specific examples of the following embodiments.

[0113] (First Item)

[0114] A detection device according to one embodiment includes: an acquisition unit that acquires a plurality of detection data corresponding to each of a plurality of samples detected by a chromatograph; and

[0115] a computing unit, for processing the plurality of detection data acquired by the acquiring unit,

[0116] The calculation unit acquires an identification result indicating that the peak information of the signal intensity extracted from each of the plurality of detection data is identified as the peak information of the signal intensity corresponding to the target component,

[0117] Furthermore, when there is outlier peak information among the plurality of peak information identified in each of the plurality of detection data, it is detected that the identification result corresponding to the outlier peak information is abnormal.

[0118] By using the detection device described in the first item, the peak of the chromatogram can be appropriately identified as the peak of the target component.

[0119] (Second Item)

[0120] The detection device 100 according to the first item, wherein the calculation unit displays the peak information of the outlier on the display device in a manner more emphasized than other peak information different from the peak information of the outlier.

[0121] The detection device described in the second item can allow the user to recognize that the identified peak is wrong.

[0122] (Third Item)

[0123] The detection device 100 according to the first or second item, wherein the computing unit

[0124] acquiring a representative value based on other peak information different from the peak information of the outlier,

[0125] And the peak information of the outlier value is replaced by the representative value.

[0126] By using the detection device described in the third item, outliers can be replaced with representative values ​​that represent normal values.

[0127] (4th item)

[0128] The detection device 100 according to the third item, wherein the calculation unit displays both the peak information before the peak information of the outlier is replaced with the representative value and the peak information after the replacement on the display device in a recognizable form.

[0129] The detection device described in the fourth item allows the user to easily grasp the positions of the peaks before and after the outliers are replaced by the representative values.

[0130] (Fifth Item)

[0131] The detection device 100 according to any one of the first to fourth items, wherein the peak information includes a holding time when the signal strength is at a maximum value.

[0132] With the detection device described in the fifth item, the peak holding time can be used to detect the identified abnormality.

[0133] (Sixth Item)

[0134] The detection device 100 according to any one of the first to fifth items, wherein the peak information includes at least one of a holding time when the signal strength starts to rise, a holding time when the signal strength ends to fall, and a position of a baseline.

[0135] With the detection device described in the sixth item, the peak start holding time or the peak end holding time can be used to detect the identified abnormality.

[0136] (Item 7)

[0137] The detection device 100 according to any one of the first to sixth items, wherein the calculation unit calibrates the retention time included in each of the plurality of detection data in advance.

[0138] By using the detection device described in the seventh item, the change of the holding time caused by column degradation or the like can be corrected in advance.

[0139] (Item 8)

[0140] The detection device 100 according to any one of the first to seventh items, wherein the detection device uses Hotelling's T 2 A method is used to determine whether there is peak information of the outlier among the multiple peak information.

[0141] By using the detection device described in item 8, Hotelling's T 2 Method to determine outliers.

[0142] (Item 9)

[0143] The detection device 100 according to any one of the first to eighth items, wherein the plurality of samples include a standard sample, and the standard sample is used to generate information for determining the target component from the peak information.

[0144] By means of the detection device described in item 9, the identified abnormality can be detected using a standard sample.

[0145] (Item 10)

[0146] In one embodiment of the detection method, a computer detects an abnormality in the identification result of the target component, and the detection method includes:

[0147] a step of acquiring a plurality of detection data corresponding to each of a plurality of samples detected by a chromatograph; and

[0148] a step of processing the plurality of detection data acquired by the acquisition step,

[0149] The processing steps include:

[0150] a step of acquiring an identification result indicating that the peak information of the signal intensity extracted from each of the plurality of detection data is identified as the peak information of the signal intensity corresponding to the target component; and

[0151] A step of detecting that an identification result corresponding to the outlier peak information is abnormal when there is outlier peak information among the plurality of peak information identified in each of the plurality of detection data.

[0152] By the detection method described in the tenth item, the peak of the chromatogram can be appropriately identified as the peak of the target component.

[0153] (Item 11)

[0154] In a computer program product including a detection program according to one embodiment, an abnormality in the identification result of a target component is detected, and the computer program product including the detection program causes a computer to execute:

[0155] a step of acquiring a plurality of detection data corresponding to each of a plurality of samples detected by a chromatograph; and

[0156] a step of processing the plurality of detection data acquired by the acquisition step,

[0157] The processing steps include:

[0158] a step of acquiring an identification result indicating that the peak information of the signal intensity extracted from each of the plurality of detection data is identified as the peak information of the signal intensity corresponding to the target component; and

[0159] A step of detecting that an identification result corresponding to the outlier peak information is abnormal when there is outlier peak information among the plurality of peak information identified in each of the plurality of detection data.

[0160] By means of the computer program product including the detection program according to the eleventh item, the peak of the chromatogram can be appropriately identified as the peak of the target component.

[0161] (Item 12)

[0162] In one embodiment, the detection system includes: a chromatograph; and

[0163] A detection device for detecting an abnormality in the identification result of the target component,

[0164] The detection device comprises:

[0165] an acquisition unit that acquires a plurality of detection data corresponding to each of a plurality of samples detected by the chromatograph; and

[0166] a computing unit, for processing the plurality of detection data acquired by the acquiring unit,

[0167] The computing unit

[0168] acquiring an identification result indicating that the peak information of the signal intensity extracted from each of the plurality of detection data is identified as the peak information of the signal intensity corresponding to the target component,

[0169] Furthermore, when there is outlier peak information among the plurality of peak information identified in each of the plurality of detection data, it is detected that the identification result corresponding to the outlier peak information is abnormal.

[0170] By using the detection system described in the twelfth item, the peak of the chromatogram can be appropriately identified as the peak of the target component.

[0171] Although the embodiments of the present invention have been described, the embodiments disclosed this time should be considered to be illustrative in all aspects and not restrictive. The scope of the present invention is indicated by the claims, and is intended to include all modifications within the meaning and scope equivalent to the claims.

Claims

1. A detection device, comprising: an acquisition unit that acquires a plurality of detection data corresponding to each of a plurality of samples detected by the chromatograph; as well as a computing unit, for processing the plurality of detection data acquired by the acquiring unit, Among them, the computing unit acquiring an identification result indicating that the peak information of the signal intensity extracted from each of the plurality of detection data is identified as the peak information of the signal intensity corresponding to the target component, and When there is outlier peak information among the plurality of peak information identified in each of the plurality of detection data, it is detected that the identification result corresponding to the outlier peak information is abnormal.

2. The detection device according to claim 1, wherein: The calculation unit displays the peak information of the outlier on a display device in a manner that is more emphasized than other peak information different from the peak information of the outlier.

3. The detection device according to claim 1, wherein: The computing unit acquiring a representative value based on other peak information different from the peak information of the outlier, and The peak information of the outlier is replaced with the representative value.

4. The detection device according to claim 3, wherein: The calculation unit displays both the peak information before the peak information of the outlier is replaced with the representative value and the peak information after the replacement on a display device in a recognizable form.

5. The detection device according to claim 1, wherein: The peak information includes the holding time when the signal strength is at the maximum value.

6. The detection device according to claim 1, wherein: The peak information includes at least one of a holding time when the signal strength starts to rise, a holding time when the signal strength ends to drop, and a position of a baseline.

7. The detection device according to claim 1, wherein: The calculation unit corrects in advance the retention time included in each of the plurality of detection data.

8. The detection device according to claim 1, wherein: The detection device uses Hotelling's T2 method to determine whether the peak information of the outlier exists in the plurality of peak information.

9. The detection device according to claim 1, wherein: The plurality of samples include a standard sample used to generate information for determining the target component from the peak information.

10. A detection method for detecting abnormalities in identification results of target components by computer detection, comprising: The first step is to obtain a plurality of detection data corresponding to each of a plurality of samples detected by a chromatograph; as well as The second step is to process the plurality of detection data obtained in the first step. Wherein, the second step comprises: a step of acquiring an identification result indicating that the peak information of the signal intensity extracted from each of the plurality of detection data is identified as the peak information of the signal intensity corresponding to the target component; and A step of detecting that an identification result corresponding to the outlier peak information is abnormal when there is outlier peak information among the plurality of peak information identified in each of the plurality of detection data.

11. A computer program product comprising a detection program, for detecting an abnormality in an identification result of a target component, wherein the computer program product comprising the detection program causes a computer to execute: The first step is to obtain a plurality of detection data corresponding to each of a plurality of samples detected by a chromatograph; as well as The second step is to process the plurality of detection data obtained in the first step. in, The second step comprises: a step of acquiring an identification result indicating that the peak information of the signal intensity extracted from each of the plurality of detection data is identified as the peak information of the signal intensity corresponding to the target component; as well as A step of detecting that an identification result corresponding to the outlier peak information is abnormal when there is outlier peak information among the plurality of peak information identified in each of the plurality of detection data.

12. A detection system comprising: Chromatography; as well as A detection device for detecting an abnormality in the identification result of the target component, The detection device comprises: an acquisition unit that acquires a plurality of detection data corresponding to each of a plurality of samples detected by the chromatograph; and a computing unit, for processing the plurality of detection data acquired by the acquiring unit, Among them, the computing unit acquiring an identification result indicating that the peak information of the signal intensity extracted from each of the plurality of detection data is identified as the peak information of the signal intensity corresponding to the target component, Furthermore, when there is outlier peak information among the plurality of peak information identified in each of the plurality of detection data, it is detected that the identification result corresponding to the outlier peak information is abnormal.

Images