Component analysis device and component analysis method
By using the data acquisition circuit and reference spectrum data set in the three-dimensional fluorescence analysis method for factor analysis, the problems of low accuracy of component analysis and cumbersome preparation in the prior art are solved, and simple and high-precision component analysis is achieved, which is suitable for a variety of Analyze the object.
Patent Information
- Application Number
- CN201980037580.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2019-07-19
- Filing Date
- 2019-08-01
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2039-08-01
AI Technical Summary
The existing three-dimensional fluorescence analysis method requires tedious pre-preparation in the component analysis of the object, and the component analysis accuracy is low when the pre-preparation is not carried out, and appropriate component separation may not be possible.
The data acquisition circuit is used to measure the spectral data of the object through the sensor, and factor analysis is performed in combination with the pre-stored reference spectrum data set to determine the components of the object, simplifying the analysis process and improving the component separation accuracy.
It realizes simple and high-precision component analysis of object objects, reduces the preparation of measured data, improves the accuracy of analysis results, and is suitable for component analysis of food, water quality and industrial wastewater.
Smart Images

Figure CN112236669B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a component analysis device and a component analysis method. Background Art
[0002] In recent years, as a method for analyzing the water quality of river water, sewage treatment water, etc., and food, a three-dimensional fluorescence analysis method (EEM: Excitation-Emission Matrix) is known. For example, in Patent Document 1, a method for analyzing food using data obtained by EEM (hereinafter referred to as EEM data) is disclosed. In addition, in Non-Patent Document 1, as a method for analyzing EEM data, a technique using Parallel Factor Analysis (PARAFAC: Parallel Factor Analysis) which is one of the chemometrics analyses is disclosed.
[0003] Prior art literature
[0004] Patent Literature
[0005] Patent Document 1: Japanese Patent No. 5985709
[0006] Non-patent literature
[0007] Non-patent document 1: Jakob Christensen, et al., "Multivariate Autofluorescence of Intact Food Systems", Chemical Review, Vol. 106, No. 6 (2006) Summary of the invention
[0008] Problems to be solved by the invention
[0009] However, the above conventional analysis methods require very troublesome preparations such as preparation of a large amount of sample data, and if such preparations are not performed, the accuracy of component analysis of the object is low. Furthermore, even if preparations are performed, it is possible that appropriate component analysis cannot be performed.
[0010] Therefore, an object of the present invention is to provide a component analysis device and a component analysis method capable of performing component analysis of an object simply and with high accuracy.
[0011] Means used to solve problems
[0012] A component analysis device according to a technical solution of the present invention comprises: a data acquisition circuit for acquiring spectral data obtained by measuring the spectrum of an object containing multiple components using a sensor; and an analysis circuit for determining the multiple components by performing factor analysis on the input data using the spectral data acquired by the data acquisition circuit and at least one reference spectral data set read out from a storage unit that pre-stores at least one reference spectral data set representing spectra of multiple substances as input data.
[0013] In addition, a component analysis method related to a technical solution of the present invention includes: obtaining spectral data obtained by measuring the spectrum of an object containing multiple components; and using the above-mentioned spectral data and at least one reference spectral data set read from a storage unit that pre-stores at least one reference spectral data set representing spectra of multiple substances as input data, and determining the above-mentioned multiple components by performing factor analysis on the above-mentioned input data.
[0014] Furthermore, one aspect of the present invention may be implemented as a program for causing a computer to execute the above-mentioned component analysis method, or may be implemented as a computer-readable recording medium storing the program.
[0015] Effects of the Invention
[0016] According to the present invention, component analysis of an object can be performed simply and with high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 It is a block diagram showing the functional structure of a component analysis device according to an embodiment.
[0018] Figure 2 It is a graph showing EEM data of corn.
[0019] Figure 3 It is a diagram showing an example of the correspondence information stored in the storage unit.
[0020] Figure 4A This is a graph showing EEM data of phenylalanine.
[0021] Figure 4B This is a diagram showing EEM information of tyrosine.
[0022] Figure 4C This is a graph showing EEM data of tryptophane.
[0023] Figure 4D It is a graph showing EEM data of ATP.
[0024] Figure 4EIt is a graph showing EEM data of chlorophyll.
[0025] Figure 4F It is a graph showing EEM data of NADH.
[0026] Figure 4G It is a graph showing EEM data of vitamin B2.
[0027] Figure 5 This is a flowchart showing the operation of the component analysis device according to the embodiment.
[0028] Figure 6 This is a flowchart showing a process of determining a characteristic factor in the operation of the component analysis device according to the embodiment.
[0029] Fig. 7A This is a diagram showing three-dimensional spectrum data of the first eigenfactor separated by PARAFAC.
[0030] Figure 7B It is a diagram showing the three-dimensional spectrum data of the second eigenfactor separated by PARAFAC.
[0031] Figure 7C It is a diagram showing the three-dimensional spectrum data of the third eigenfactor separated by PARAFAC.
[0032] Fig.7D It is a diagram showing the three-dimensional spectrum data of the fourth eigenfactor separated by PARAFAC.
[0033] Fig. 7E It is a diagram showing the three-dimensional spectrum data of the fifth eigenfactor separated by PARAFAC.
[0034] Figure 7F It is a diagram showing the three-dimensional spectrum data of the sixth eigenfactor separated by PARAFAC.
[0035] Figure 7G It is a diagram showing the three-dimensional spectrum data of the seventh eigenfactor separated by PARAFAC.
[0036] Figure 8 This is a diagram showing the scores corresponding to the seven characteristic factors obtained by the component analysis device according to the embodiment.
[0037] Fig. 9 It is a diagram showing the analysis results of PARAFAC of a comparative example. DETAILED DESCRIPTION
[0038] (Knowledge that is the basis of the present invention)
[0039] The present inventors have discovered that the following problems occur with respect to the conventional technology described in the “Background Art” column.
[0040] Spectrofluorimetry, represented by EEM, is an analysis method that can perform component analysis with high accuracy because it utilizes fluorescence, which is a special light, for substances. In addition, spectrofluorimetry is an analysis method with a high sensitivity of about 1,000 times that of absorbance analysis.
[0041] On the other hand, food or river water to be analyzed generally contains multiple types of components. Therefore, since the multiple types of components emit light respectively and the fluorescence spectra are superimposed, EEM data becomes complicated.
[0042] By performing PARAFAC, the peaks of overlapping fluorescence spectra can be separated from complex EEM data according to the components that present the same luminescence dynamics. Using the information of the separated peaks, quantitative analysis can be performed based on the peak intensity of the fluorescence, or component analysis can be performed based on the peak wavelength of the fluorescence. In addition, discriminant analysis can be performed by performing principal component analysis.
[0043] However, when performing PARAFAC on EEM data, which is actual measurement data, that is, actual sample data, to perform peak separation, there are cases where appropriate separation cannot be performed. For example, a negative peak may be separated, or an assumed peak may not be separated. In order to perform appropriate peak separation by PARAFAC, appropriate boundary conditions and the like need to be set, and data needs to be acquired in advance using many samples. Therefore, it is difficult to perform appropriate separation of components in conventional PARAFAC.
[0044] The technology described in Patent Document 1 also requires the preparation of a database for each food to be analyzed.
[0045] Therefore, the present invention provides a component analysis device and a component analysis method capable of performing component analysis of an object simply and with high accuracy.
[0046] For example, a component analysis device according to a technical solution of the present invention comprises: a data acquisition circuit for acquiring spectral data obtained by measuring the spectrum of an object containing multiple components using a sensor; and an analysis circuit for determining the multiple components by performing factor analysis on the input data using the spectral data acquired by the data acquisition circuit and the at least one reference spectral data set read out from a storage unit that pre-stores at least one reference spectral data set representing spectra of multiple substances.
[0047] Thus, since the input data including the reference spectrum data set is subjected to factor analysis, the separation accuracy of the characteristic factors can be improved. In addition, since there is no need to prepare a large amount of measured data obtained by measurement, the components contained in the spectrum data can be separated with high accuracy. In this way, according to the component analysis device related to the present technical solution, the component analysis of the object can be performed simply and with high accuracy.
[0048] Furthermore, for example, the above-mentioned factor analysis may also be a parallel factor analysis.
[0049] This makes it possible to perform component analysis of an object simply and with high accuracy.
[0050] In addition, for example, the above-mentioned spectrum data may also be three-dimensional fluorescence spectrum data.
[0051] Thus, since so-called EEM data is used, component analysis of food or the like containing organic matter that emits fluorescence can be performed simply and with high accuracy.
[0052] Furthermore, for example, the plurality of substances may be substances having autofluorescence, and the at least one reference spectrum data set may be fluorescence spectra of the plurality of substances.
[0053] This makes it possible to perform component analysis of foods or the like containing organic matter that emits autofluorescence easily and with high accuracy.
[0054] Furthermore, for example, the above-mentioned various substances may be substances contained in food.
[0055] Thus, the composition analysis of food or vomitus can be performed. For example, vomitus and food can be distinguished. In addition, since the presence or absence of residues that have not been cleaned up can be distinguished after the vomitus is cleaned up, it can also be used to improve the efficiency of cleaning operations and purification operations.
[0056] In addition, for example, the above-mentioned various substances may also be organic matter contained in tap water, river water or industrial wastewater.
[0057] Thus, since the component analysis of river water or tap water before water purification (i.e., river water after water purification) is performed with high accuracy, it is possible to accurately determine whether humic substances and amino acids are contained in the river water or tap water. Thus, the performance of water purification, specifically, whether humic substances and amino acids are properly removed, can be evaluated. In addition, the component analysis device according to one technical solution of the present invention can be used for water quality inspection of river water.
[0058] Alternatively, since the components of industrial wastewater are analyzed with high accuracy, it is possible to evaluate whether the pollution treatment of industrial wastewater is properly performed. For example, in a dairy factory, milk bottles are washed after use, and the washing water is purified and then discharged as industrial wastewater. By analyzing the components of the purified washing water, it is possible to evaluate whether microbial-derived substances such as amino acids are properly removed from the washing water.
[0059] Furthermore, for example, the above-mentioned plurality of substances may be aerosol particles or substances contained in aerosol particles.
[0060] Thus, since the aerosol particles are analyzed with high accuracy, it is possible to accurately determine whether, for example, pollen is suspended in the air. Thus, for example, it is possible to evaluate whether the air cleaning by the air cleaner is being performed appropriately.
[0061] In addition, for example, a component analysis device related to a technical solution of the present invention may also be provided with a type acquisition circuit for acquiring information indicating the type of the above-mentioned object; at least one reference spectral data set includes multiple reference spectral data sets; each of the above-mentioned multiple reference spectral data sets is pre-set according to each of the above-mentioned types of the above-mentioned object; the above-mentioned analysis circuit selects a reference spectral data set corresponding to the above-mentioned type indicated by the above-mentioned information acquired by the above-mentioned type acquisition circuit from the above-mentioned multiple reference spectral data sets, and performs the above-mentioned factor analysis using the above-mentioned selected reference spectral data set and the above-mentioned spectral data as the above-mentioned input data.
[0062] This increases the number of types of analyzable objects, and thus enables component analysis of a wide variety of types of objects.
[0063] Furthermore, for example, a component analysis device according to one aspect of the present invention may further include the above-mentioned storage unit.
[0064] Thus, even in an incomplete communication environment, it is possible to perform component analysis of an object simply and with high accuracy. In addition, since the reference spectrum data set stored in the storage unit can be changed, added, and updated, the types of objects that can be analyzed can be increased. Alternatively, the accuracy of component analysis can be further improved.
[0065] In addition, a component analysis method related to a technical solution of the present invention includes: obtaining spectral data obtained by measuring the spectrum of an object containing multiple components; and using the above-mentioned spectral data and at least one reference spectral data set read from a storage unit that pre-stores at least one reference spectral data set representing spectra of multiple substances as input data, and determining the above-mentioned multiple components by performing factor analysis on the above-mentioned input data.
[0066] Thus, similar to the case of the above-mentioned component analysis device, component analysis of the object can be performed simply and accurately.
[0067] In addition, for example, it may also include obtaining information representing the type of the above-mentioned object; at least one reference spectral data set includes multiple reference spectral data sets; the above-mentioned multiple reference spectral data sets are each pre-set according to each of the above-mentioned types of the above-mentioned object; in the above-mentioned determination, from the above-mentioned multiple reference spectral data sets, a reference spectral data set corresponding to the above-mentioned type represented by the above-mentioned information obtained by the above-mentioned type acquisition circuit is selected, and the above-mentioned factor analysis is performed using the above-mentioned selected reference spectral data set and the above-mentioned spectral data as the above-mentioned input data.
[0068] This increases the number of types of analyzable objects, and thus enables component analysis of a wide variety of types of objects.
[0069] Furthermore, for example, a component analysis method according to one aspect of the present invention may further include measuring a spectrum of the object.
[0070] This makes it possible to analyze the actual measurement data obtained by the measurement with high accuracy.
[0071] Furthermore, a program according to one aspect of the present invention is a program for causing a computer to execute the above-mentioned component analysis method.
[0072] Thus, similar to the case of the above-mentioned component analysis device, component analysis of the object can be performed simply and accurately.
[0073] A computer-readable recording medium related to a technical solution of the present invention is a computer-readable recording medium that stores a program used to perform component analysis of an object. When the above program is executed by the above computer, it performs: obtaining spectral data obtained by measuring the spectrum of an object containing multiple components; and using the above spectral data and at least one reference spectral data set read from a storage unit that pre-stores at least one reference spectral data set representing spectra of multiple substances as input data, determining the above multiple components by performing factor analysis on the above input data.
[0074] In the present invention, all or part of a circuit, unit, device, component or part, or all or part of a functional block in a block diagram may be performed by one or more electronic circuits including a semiconductor device, a semiconductor integrated circuit (IC) or an LSI (large scale integration). LSI or IC may be integrated into one chip or may be composed of a combination of multiple chips. For example, functional blocks other than storage elements may be integrated into one chip. Here, it is called LSI or IC, but the name may change depending on the degree of integration, and may also be called system LSI, VLSI (very large scale integration) or ULSI (ultra large scale integration). Field Programmable Gate Array (FPGA) that can be programmed after the manufacture of LSI, or reconfigurable logic device that can reconfigure the connection relationship inside LSI or set the circuit division inside LSI may also be used for the same purpose.
[0075] Furthermore, all or part of the functions or operations of a circuit, unit, device, component or part can be executed by software processing. In this case, the software is recorded in one or more non-temporary recording media such as ROMs, optical disks, hard disk drives, etc. When the software is executed by a processor, the functions determined by the software are executed by the processor and peripheral devices. The system or device may also have one or more non-temporary recording media with software recorded, a processor, and necessary hardware devices, such as interfaces.
[0076] Hereinafter, the embodiments will be described in detail with reference to the drawings.
[0077] In addition, the embodiments described below all represent inclusive or specific examples. The numerical values, shapes, materials, constituent elements, configuration positions of constituent elements and connection forms, steps, and the order of steps shown in the following embodiments are examples and do not limit the meaning of the present invention. In addition, regarding the constituent elements of the following embodiments that are not recorded in the independent claims representing the highest concept of the present invention, they are set as arbitrary constituent elements for description.
[0078] In addition, each figure is a schematic diagram and does not necessarily illustrate the exact figure. In each figure, the same reference numerals are given to substantially the same structure, and repeated descriptions are omitted or simplified.
[0079] (Implementation Method)
[0080] [1.Structure]
[0081] First, use Figure 1 , the structure of the component analysis device of the relevant embodiment is explained.
[0082] Figure 1 1 is a block diagram showing the functional structure of the component analysis device 10 according to the present embodiment. Figure 1 As shown, the component analysis device 10 includes a data acquisition unit 20 as a data acquisition circuit, an analysis unit 30 as an analysis circuit, a type acquisition unit 40 as a type acquisition circuit, a storage unit 50 and an output unit 60 .
[0083] The component analysis device 10 performs component analysis on an object 12. The object 12 is, for example, food or vomitus. Alternatively, the object 12 may be water such as tap water, river water, or industrial wastewater. Alternatively, the object 12 may be aerosol particles suspended in space.
[0084] The object 12 includes a plurality of components. Each of the plurality of components corresponds to a substance that emits autofluorescence. The plurality of components are, for example, organic matter. When irradiated with excitation light having a peak at a certain wavelength, the plurality of components respectively emit fluorescence having a peak at a certain wavelength. The combination of the peak wavelength of the excitation light and the peak wavelength of the fluorescence is inherent to each component. Therefore, by discriminating the combination of the peak wavelength of the excitation light and the peak wavelength of the fluorescence, the component can be determined.
[0085] The data acquisition unit 20 acquires spectrum data obtained by measuring the spectrum of the object 12. The spectrum data acquired by the data acquisition unit 20 is three-dimensional fluorescence spectrum data. The three-dimensional fluorescence spectrum data is the so-called EEM data, which is three-dimensional data with excitation wavelength, fluorescence wavelength and fluorescence intensity as three axes. EEM data is also called fluorescence fingerprint. The excitation wavelength is the wavelength of the excitation light irradiated to the object 12. The fluorescence wavelength is the wavelength of the fluorescence emitted from the object 12.
[0086] The EEM data is generated by measuring a three-dimensional fluorescence spectrum. Specifically, the EEM data is obtained by measuring the fluorescence spectrum of each excitation wavelength while continuously changing the wavelength of the excitation light irradiated to the object 12. For example, when the object 12 is corn, Figure 2 EEM data for corn as shown.
[0087] Figure 2 is a graph showing EEM data for corn. Specifically, Figure 2 The EEM data are shown when a sample obtained by adding a predetermined amount of water to freeze-dried corn is used as the object 12 .
[0088] exist Figure 2 In the figure, the horizontal axis represents the fluorescence wavelength [unit: nm], and the vertical axis represents the excitation wavelength [unit: nm]. The fluorescence wavelength is the wavelength of the fluorescence emitted from the corn. The excitation wavelength is the wavelength of the excitation light irradiated to the corn. Figure 2 In the two-dimensional coordinate system, isointensity lines that continuously connect coordinates with equal fluorescence intensities are shown. For example, in corn, there is a peak of fluorescence intensity in a group with an excitation wavelength of about 280 nm and a fluorescence wavelength of about 340 nm.
[0089] The data acquisition unit 20 is implemented by, for example, an optical mechanism. Figure 1 As shown, the data acquisition unit 20 includes a spectroscopic light source unit 22 and a spectroscopic light receiving unit 24 .
[0090] The spectroscopic light source unit 22 is a light source unit that emits the split excitation light. Specifically, the spectroscopic light source unit 22 emits a plurality of excitation lights having mutually different peak wavelengths. The plurality of excitation lights are, for example, pulsed lights. The spectroscopic light source unit 22 emits the plurality of excitation lights sequentially and irradiates the object 12, for example.
[0091] The spectroscopic light source unit 22 has, for example, a light source that emits light having a wider wavelength band, such as white light, and a plurality of bandpass filters (BPFs) arranged on the light emitting side of the light source. The light source is, for example, a discharge lamp such as a halogen lamp, an LED (Light Emitting Diode) or a solid light emitting element such as a laser element. The plurality of BPFs have mutually different transmission bands, so that light of wavelengths outside the transmission band is fully attenuated. The half-value width of the transmission band of each BPF is, for example, greater than 10 nm and less than 50 nm. The transmission bands of each BPF do not overlap with each other. The spectroscopic light source unit 22 emits a plurality of excitation lights in sequence, for example, by switching the BPFs arranged on the light emitting side of the light source in sequence at a certain time. In this way, the excitation light with a continuously changing excitation wavelength can be irradiated onto the object 12.
[0092] A diffraction grating, a prism, and a slit may be used instead of the BPF. The excitation wavelength can be continuously changed by rotating the diffraction grating or the prism so that only light of a specific wavelength passes through the slit.
[0093] The spectroscopic light receiving unit 24 is a light receiving unit that receives fluorescence emitted by the object 12 by splitting it when the excitation light is irradiated on the object 12. Specifically, the spectroscopic light receiving unit 24 receives fluorescence that has passed through a plurality of light receiving bands that are different from each other.
[0094] The spectroscopic light receiving unit 24 includes, for example, a photoelectric conversion element that converts the received light into an electrical signal and outputs it, and a plurality of BPFs arranged on the light incident side of the photoelectric conversion element. The photoelectric conversion element is, for example, a photodiode that outputs an electrical signal having a signal intensity corresponding to the intensity of the received light. The photoelectric conversion element may also be a photomultiplier tube (PMT) or the like. The plurality of BPFs have mutually different transmission bands so that light of wavelengths outside the transmission bands is fully attenuated. The half-value width of the transmission band of each BPF is, for example, not less than 5 nm and not more than 50 nm. The transmission bands of each BPF do not overlap with each other.
[0095] The spectroscopic light receiving unit 24 receives the fluorescence of each light receiving band by, for example, sequentially switching the BPF arranged on the light incident side of the photoelectric conversion element at a certain time. For example, the spectroscopic light receiving unit 24 switches a plurality of BPFs during the period of being irradiated with one excitation light. Thus, the fluorescence intensity of each light receiving band can be obtained according to the excitation wavelength, and EEM data can be generated.
[0096] In addition, the spectroscopic light receiving unit 24 may include a plurality of photoelectric conversion elements corresponding one-to-one to a plurality of BPFs. In this case, fluorescence in each light receiving band can be received simultaneously, and the time required for measuring the fluorescence spectrum can be shortened.
[0097] In addition, the spectroscopic light receiving unit 24 may also be an image sensor having a plurality of photoelectric conversion elements arranged in two dimensions. Thus, EEM data is obtained according to the photoelectric conversion elements, so that fluorescence observation can be performed two-dimensionally. Therefore, component analysis of the object 12 existing in a wide range in space can be performed simultaneously.
[0098] The data acquisition unit 20 may also be implemented by an input interface or a communication interface that receives data output from a measuring device that measures a three-dimensional fluorescence spectrum. That is, the component analysis device 10 may not include the spectroscopic light source unit 22 and the spectroscopic light receiving unit 24, or another measuring device may include the spectroscopic light source unit 22 and the spectroscopic light receiving unit 24.
[0099] The analysis unit 30 determines a plurality of components included in the spectral data by performing factor analysis on the input data using the spectral data acquired by the data acquisition unit 20 and the reference spectral data set 52 read from the storage unit 50 as input data. Specifically, the factor analysis is parallel factor analysis (PARAFAC). PARAFAC is performed based on the following (Formula 1).
[0100] [Formula 1]
[0101]
[0102] x ijkrepresents the fluorescence intensity of input data k corresponding to the combination of excitation wavelength i and fluorescence wavelength j. I, J and K represent the number of excitation lights, the number of fluorescence light receiving bands and the number of input data, respectively. F corresponds to the fluorescence intensity of input data k corresponding to the combination of excitation wavelength i and fluorescence wavelength j. ijk The maximum number of separated characteristic factors, that is, x ijk That is, the analysis unit 30 determines f components by performing PARAFAC.
[0103] a if , b jf and c kf They correspond to the excitation spectrum, fluorescence spectrum and concentration of input data of component f. if b jf c kf The term represented by corresponds to the three-dimensional spectrum data of the component f separated by PARAFAC, that is, the three-dimensional spectrum data of the f-th characteristic factor. ijk It is an independent factor that does not depend on the excitation spectrum and fluorescence spectrum. ijk corresponds to the residual remaining after separation into components f by PARAFAC.
[0104] In this embodiment, the analysis unit 30 specifies the components contained in the object 12 based on the score of the separated f-th characteristic factor. if b jf c kf Indicates that it is automatically calculated by performing PARAFAC. In this embodiment, the score of the f-th characteristic factor is calculated according to the spectrum data included in the input data. The analysis unit 30 converts the scores of each spectrum data so that the maximum score among the scores of the f-th characteristic factor becomes 100.
[0105] The analysis unit 30 compares the score with a preset threshold value according to the characteristic factor. When the score of the characteristic factor exceeds the threshold value, the analysis unit 30 determines that the component corresponding to the characteristic factor is a component included in the object 12. When the score of a specific factor is below the threshold value, the analysis unit 30 determines that the component corresponding to the characteristic factor is not included in the object 12. The threshold value is, for example, 0, but is not limited thereto.
[0106] The analysis unit 30 is implemented, for example, by software executed by a processor or the like. Alternatively, the analysis unit 30 may be implemented by hardware such as an electronic circuit including a plurality of circuit elements. Figure 1 As shown, an information processing unit 32 is provided.
[0107] The information processing unit 32 is, for example, a microcomputer. Specifically, the information processing unit 32 includes a signal processing circuit for processing the signal output from the spectroscopic light receiving unit 24, a nonvolatile memory storing a program, a volatile memory as a temporary storage area for executing the program, an input / output port, and a processor for executing the program.
[0108] The type acquisition unit 40 acquires type information indicating the type of the object 12. The type acquisition unit 40 is implemented by, for example, a touch sensor of a touch panel display, a keyboard, a mouse, or other input device. The type acquisition unit 40 acquires type information by receiving a user operation.
[0109] The category acquisition unit 40, for example, presents a plurality of candidates for the category of the object 12 to the user, and acquires category information by allowing the user to select one of the plurality of presented candidates. Figure 3 The type acquisition unit 40 may allow the user to input the type of the object 12 by text input, voice input, or the like.
[0110] The storage unit 50 is a storage unit that pre-stores the reference spectrum data set 52. In the present embodiment, the storage unit 50 also stores the corresponding information 54. The reference spectrum data set 52 and the corresponding information 54 can also be modified by addition, deletion, update, etc. as needed. The storage unit 50 is implemented by, for example, a HDD (Hard Disk Drive) or a semiconductor memory.
[0111] The reference spectrum data set 52 represents spectra of a plurality of substances prepared in advance. The reference spectrum data set 52 represents fluorescence spectra of a plurality of substances. The reference spectrum data set 52 is stored in the storage unit 50. In the present embodiment, the storage unit 50 stores the reference spectrum data set 52 preset according to the type of the object 12.
[0112] The plurality of substances may be, for example, substances having autofluorescence. The plurality of substances may include substances that are presumed to be included in the object 12, that is, presumed characteristic factors. Alternatively, the plurality of substances may also include substances that are expected not to be included in the object 12. Figure 3 As shown, a plurality of substances included in the reference spectrum data set 52 are preset according to the type of the object 12 .
[0113] Figure 3 2 is a diagram showing an example of the correspondence information 54 stored in the storage unit 50 . The correspondence information 54 is information showing the correspondence relationship between the type of the object 12 and the plurality of substances included in the reference spectrum data set 52 .
[0114] like Figure 3As shown, the reference spectrum data set 52 when the object 12 is food includes spectrum data of seven substances, namely, phenylalanine, tyrosine, tryptophan, ATP (adenosine triphosphatase), chlorophyll, NADH (nicotinamide adenine dinucleotide), and vitamin B2. The reference spectrum data set 52 when the object 12 is river water or tap water includes spectrum data of three substances, namely, humic acid, fulvic acid, and tryptophan. The reference spectrum data set 52 when the object 12 is industrial wastewater includes spectrum data of two substances, namely, amino acids and NADH.
[0115] In addition, when the object 12 is an aerosol, the reference spectrum data set 52 includes spectrum data of three substances: NADH, cedar pollen, and Japanese cypress pollen. NADH is an example of a substance contained in aerosol particles derived from organisms. Cedar pollen and Japanese cypress pollen are examples of aerosol particles.
[0116] Alternatively, the reference spectrum data set 52 may include spectrum data of two substances, Cryj1 and Cryj2, which are substances contained in the cedar pollen, instead of the cedar pollen. Cryj1 and Cryj2 are the main allergens of the cedar pollen. Similarly, the reference spectrum data set 52 may include spectrum data of two substances, Chao1 and Chao2, which are substances contained in the cedar pollen, instead of the Japanese cypress pollen. Chao1 and Chao2 are the main allergens of the Japanese cypress pollen.
[0117] in addition, Figure 3 The correspondence shown is only an example. The types and numbers of substances included in each reference spectrum data set 52 are not particularly limited. For example, the substances included in the reference spectrum data set 52 for food may include collagen or vitamins such as vitamin A or folic acid. In addition, for example, the substances included in the reference spectrum data set 52 for aerosol may include aerosol particles and substances contained in the aerosol particles. For example, the reference spectrum data set 52 for aerosol may also include the spectrum data of Cryptomeria japonica pollen, Cryj1, and Cryj2.
[0118] FIG. 4A to FIG. 4G Each of them is a diagram showing spectral data of a plurality of substances included in the reference spectral data set 52 used when the object 12 is food. Specifically, FIG. 4A to FIG. 4G EEM data for phenylalanine, tyrosine, tryptophan, ATP, chlorophyll, NADH and vitamin B2 are shown. FIG. 4A to FIG. 4G As shown, the combination of the excitation wavelength and the fluorescence wavelength when the fluorescence intensity reaches the peak is inherent to each substance.
[0119] The output unit 60 outputs the analysis result of the analysis unit 30. Specifically, the analysis result is an analysis result based on parallel factor analysis, and is information for identifying a plurality of components included in the EEM data acquired by the data acquisition unit 20. For example, the analysis result is the name of each of the plurality of components included in the EEM data. The components indicated by the analysis result correspond to the components included in the object 12.
[0120] The output unit 60 includes, for example, a display unit 62 that displays an image showing the analysis result. The display unit 62 is a liquid crystal display device or an organic EL (Electroluminescence) display device, but is not limited thereto.
[0121] Furthermore, the output unit 60 may include a speaker for outputting the analysis result as sound. Furthermore, the output unit 60 may include an output interface or a communication interface for outputting a signal indicating the analysis result to an external device.
[0122] [2. Action (Component Analysis Method)]
[0123] Next, use Figure 5 and Figure 6 The operation of the component analysis device 10 according to the present embodiment (that is, the component analysis method) will be described.
[0124] Figure 5 This is a flowchart showing the operation of the component analysis device 10 according to the present embodiment. Figure 6 : is a flowchart showing the process of determining the characteristic factor in the operation of the component analysis device 10 according to the present embodiment. Specifically, Figure 6 express Figure 5 The details of step S12 are shown in FIG.
[0125] like Figure 5 As shown, first, the data acquisition unit 20 measures the spectral data of the object 12 (S10). Specifically, the data acquisition unit 20 acquires EEM data as spectral data by measuring the three-dimensional fluorescence spectrum of the object 12. More specifically, the spectroscopic light source unit 22 emits excitation light while continuously changing the excitation wavelength, and the spectroscopic light receiving unit 24 measures the three-dimensional fluorescence spectrum by receiving the fluorescence emitted from the object 12 according to the light receiving band. One EEM data is obtained in one measurement. It is also possible to obtain multiple EEM data by repeating the measurement multiple times. Here, it is assumed that the number of EEM data obtained is M. M is a natural number and can also be 1.
[0126] Next, the analysis unit 30 determines the components, that is, the characteristic factors, included in the EEM data 17 acquired by the data acquisition unit 20 (S12). Figure 6As shown, first, the type acquisition unit 40 acquires type information of the object 12 (S20). For example, the type acquisition unit 40 acquires the type information by allowing the user to input the type of the object 12 measured in step S10.
[0127] Next, the analysis unit 30 obtains a reference spectrum data set 52 consisting of spectrum data of N assumed characteristic factors corresponding to the type indicated by the type information (S22). Specifically, the analysis unit 30 refers to the corresponding information 54 stored in the storage unit 50, determines the N substances included in the reference spectrum data set 52 corresponding to the type indicated by the type information, and reads the spectrum data of the determined N substances from the storage unit 50. In addition, N is a natural number greater than 2.
[0128] Next, the analysis unit 30 acquires the M spectrum data acquired by the data acquisition unit 20 (S24). Furthermore, the analysis unit 30 sets the setting value f for performing parallel factor analysis to 1 (S26). The setting value f corresponds to the number of characteristic factors separated by the parallel factor analysis.
[0129] Next, the analysis unit 30 treats the reference spectrum data set composed of M spectrum data obtained by measuring with the sensor and spectrum data of N assumed characteristic factors as input data. That is, the analysis unit 30 performs parallel factor analysis (PARAFAC) (S28) on the M+N spectrum data as input data. When the result of performing PARAFAC can separate f characteristic factors (Yes in S30), the analysis unit 30 adds 1 to the value of the set value f (S32), and performs PARAFAC again based on the set value f that becomes 2 (S28). PARAFAC is repeatedly performed until it is no longer possible to separate into f characteristic factors.
[0130] When the characteristic factors cannot be separated into f (No in S30), the analysis unit 30 determines f-1 as the maximum number of characteristic factors (S34). That is, the analysis unit 30 determines the maximum number of characteristic factors that can be separated.
[0131] Next, the analysis unit 30 identifies the components included in the object 12 based on the scores of the f-1 characteristic factors ( S36 ). Specifically, the analysis unit 30 compares each score with a threshold value and determines that the component corresponding to the characteristic factor exceeding the threshold value is a component of the object 12 .
[0132] After obtaining the judgment result, Figure 5 As shown, the display unit 62 displays the result (S14). For example, the display unit 62 lists and displays the names of the components included in the object 12. In addition, the spectrum data of each characteristic factor separated by PARAFAC and the score of each characteristic factor may also be displayed.
[0133] [3. Analysis results]
[0134] The following is the result of actual measurement. Figure 2 The results of parallel factor analysis of the present embodiment will be described using the spectral data shown. PARAFAC is performed using, for example, chemometrics software Solo from Eigenvector Research.
[0135] Since the object 12 is corn, i.e., food, Figure 3 As shown in FIG. 1 , the spectral data of 7 substances, namely, phenylalanine, tyrosine, tryptophan, ATP, chlorophyll, NADH and vitamin B2, are used as reference spectral data sets. Figure 2 The spectral data shown is 1 and FIG. 4A to FIG. 4G The 7 spectral data shown are a total of 8 spectral data as input data, and PARAFAC is performed.
[0136] Thus, it can be separated into FIG. 7A to FIG. 7G The total number of characteristic factors is 7, from the 1st characteristic factor to the 7th characteristic factor. FIG. 7A to FIG. 7G These are diagrams showing three-dimensional spectrum data of the first to seventh eigenfactors separated by PARAFAC.
[0137] For example, Fig. 7A The spectral data of the first characteristic factor shown is Figure 4A By comparing the spectral data of phenylalanine shown, it can be seen that similar spectral data are obtained. Specifically, based on the scores of the first characteristic factor corresponding to each spectral data included in the input data, it is determined which of the multiple spectral data included in the input data the first characteristic factor corresponds to. Specifically, it is determined that the first characteristic factor corresponds to the spectral data that has obtained the largest score among the scores of the first characteristic factor.
[0138] The same is true for the second to seventh characteristic factors, and the substance corresponding to each characteristic factor can be identified based on the score of each characteristic factor. Figure 7B The second characteristic factor shown corresponds to Figure 4B Tyrosine as shown. Figure 7C The third characteristic factor shown corresponds to Figure 4C Tryptophan shown. Fig.7D The fourth characteristic factor shown corresponds to Figure 4D ATP as shown. Fig. 7E The fifth characteristic factor shown corresponds to Figure 4E Chlorophyll shown. Figure 7F The sixth characteristic factor shown corresponds to Figure 4F NADH shown. Figure 7GThe seventh characteristic factor shown corresponds to Figure 4G Vitamin B2 as shown.
[0139] Figure 8 1 is a diagram showing the scores for the seven characteristic factors obtained by the component analysis device 10 according to the present embodiment. The scores are automatically calculated by the analysis unit 30 performing PARAFAC as described above. Figure 8 In FIG. 1 , only the case where the input data is the spectrum data of corn is shown, but the score is calculated for each of the eight input data according to the characteristic factor. The score is represented by the value when the highest value is set to 100.
[0140] For example, the second characteristic factor corresponding to tyrosine is Figure 7B The score of the spectrum data of tyrosine shown in FIG. 1 is 100. In this case, the score of the spectrum data of corn corresponding to the second characteristic factor of tyrosine is as follows: Figure 8 As shown, it is 5.95. The score indicates the confidence that the characteristic factor included in the spectrum data is included. The lower the score, the more it is inferred that the corresponding characteristic factor is not included, and the higher the score, the more it is inferred that the corresponding characteristic factor is included. In addition, the score of each characteristic factor of corn does not indicate the content or ratio of the characteristic factor relative to corn.
[0141] In this embodiment, as described above, the analysis unit 30 compares the score with the threshold value and determines that the characteristic factor is included in the object 12 when the score exceeds the threshold value. Figure 8 In the example shown, the analysis unit 30 identifies tyrosine, tryptophan, ATP, chlorophyll, NADH, and vitamin B2 as components contained in corn. In addition, the analysis unit 30 identifies phenylalanine as a component not contained in corn.
[0142] As described above, according to the component analysis device 10 according to the present embodiment, component analysis can be performed simply and accurately by using the reference spectrum data set 52 .
[0143] Furthermore, as a comparative example, a case where PARAFAC is performed without using the reference spectrum data set 52 will also be described.
[0144] Specifically, Figure 2 Similarly, PARAFAC is performed using 5 EEM data based on 5 measurements as input data for the spectral data shown. In this case, Fig. 9 As shown in parts (a) to (d), it can only be separated into four characteristic factors. Fig. 9As shown in part (d), in the fourth characteristic factor, a result that was not generated as normal EEM data was obtained. Thus, in the comparative example, even though the number of measurements was increased, an appropriate separation result was not obtained.
[0145] (Other embodiments)
[0146] The above is a description of the component analysis device and component analysis method of one or more technical solutions based on the embodiments, but the present invention is not limited to these embodiments. As long as it does not deviate from the main purpose of the present invention, various deformed forms that a person skilled in the art would like to apply to the present embodiment, and forms constructed by combining constituent elements of different embodiments are also included in the scope of the present invention.
[0147] For example, in the above-mentioned embodiment, an example in which the factor analysis is a parallel factor analysis is shown, but the present invention is not limited thereto. For example, the factor analysis may be a three-phase factor analysis such as Tucker, Tucker2, or PARAFAC2.
[0148] In addition, for example, in the above-mentioned embodiment, an example in which the spectral data is three-dimensional fluorescence spectral data is shown, but it is not limited to this. For example, the spectral data may also be spectral data used in two-dimensional liquid chromatography analysis. In addition, the spectral data may also be spectral data with a time axis, that is, spectral data that changes with time. For example, by performing component analysis of a plurality of EEM data obtained by measuring at a plurality of different times, it is possible to discern the time variation of the components contained in the object. For example, it is possible to discern the change of the components caused by the chemical reaction when the drug is given to the object.
[0149] In addition, for example, in the above-mentioned embodiment, the plurality of substances are examples of substances that emit autofluorescence, but the present invention is not limited thereto. For example, the plurality of substances may be substances containing artificially added fluorescent pigments such as fluorescein.
[0150] In addition, for example, in the above-mentioned embodiment, an example of establishing a correspondence between reference spectrum data sets according to the type of the object 12 is shown, but the present invention is not limited thereto. For example, only reference spectrum data sets corresponding to food may be stored in the storage unit 50, and the component analysis device 10 may be a dedicated device for determining components contained in food.
[0151] Furthermore, for example, when the component analyzer 10 is a dedicated device for analyzing the components of food, since the types are preset, it is not necessary to accept the type designation from the user. Therefore, the component analyzer 10 may not include the type acquisition unit 40 .
[0152] In addition, for example, the component analysis device 10 may not include the storage unit 50 for storing the reference spectrum data set. The reference spectrum data set may be stored in a memory of a device other than the component analysis device 10, such as an external server device. The component analysis device 10 may also acquire the reference spectrum data set by communicating with the server device and perform factor analysis using the acquired reference spectrum data set.
[0153] In the above-mentioned embodiment, for example, the spectroscopic light receiving unit 24 receives fluorescence emitted from the object 12 , but the present invention is not limited thereto. The spectroscopic light receiving unit 24 may receive reflected light or scattered light returning from the object 12 .
[0154] In addition, the communication method between the devices described in the above embodiment is not particularly limited. In the case of wireless communication between devices, the method of wireless communication (communication specification) is, for example, short-range wireless communication such as ZigBee (registered trademark), Bluetooth (registered trademark) or wireless LAN (Local Area Network). Alternatively, the method of wireless communication (communication specification) may also be communication via a wide area communication network such as the Internet. In addition, wired communication may be performed between devices instead of wireless communication. Specifically, wired communication is power line communication (PLC: Power Line Communication) or communication using a wired LAN.
[0155] In addition, in the above-mentioned embodiment, the processing performed by a specific processing unit may be performed by another processing unit. In addition, the order of multiple processing may be changed, or multiple processing may be performed in parallel. In addition, the allocation of the components of the component analysis device to multiple devices is an example. For example, the components of one device may be provided by other devices.
[0156] For example, the processing described in the above embodiments can be implemented by using a single device (system) for centralized processing, or can be implemented by using multiple devices for distributed processing. In addition, the processors that execute the above program can be either single or multiple. That is, centralized processing can be performed, or distributed processing can be performed.
[0157] In addition, in the above-mentioned embodiment, all or part of the components such as the control unit may be formed by dedicated hardware, or may be implemented by executing a software program suitable for each component. Each component may also be implemented by a program execution unit such as a CPU (Central Processing Unit) or a processor reading and executing a software program recorded in a recording medium such as a HDD (Hard Disk Drive) or a semiconductor memory.
[0158] Furthermore, the components of the control unit and the like may be constituted by one or more electronic circuits. The one or more electronic circuits may be general-purpose circuits or dedicated circuits.
[0159] In one or more electronic circuits, for example, semiconductor devices, IC (Integrated Circuit) or LSI (Large Scale Integration) etc. may also be included. LSI or IC may be integrated into one chip or multiple chips. Here, it is called LSI or IC, but the name changes according to the degree of integration, and it may also be called system LSI, VLSI (very large scale integration) or ULSI (ultra large scale integration). In addition, FPGA (Field Programmable Gate Array) which can be programmed after the manufacture of LSI may also be used for the same purpose.
[0160] In addition, the global or specific form of the present invention may also be implemented by a system, device, method, integrated circuit or computer program. Alternatively, it may also be implemented by a computer-readable non-temporary recording medium such as an optical disk, HDD or semiconductor memory storing the computer program. In addition, it may also be implemented by any combination of a system, device, method, integrated circuit, computer program and recording medium.
[0161] Furthermore, various changes, substitutions, additions, omissions, etc. can be made to the above-described embodiments within the scope of the claims or their equivalents.
[0162] Industrial Applicability
[0163] The present invention can be used as a component analysis device that can perform component analysis of an object simply and with high accuracy, and can be used, for example, in a water quality inspection device, a pollutant detection device, or a pollutant purification device.
[0164] Description of symbols
[0165] 10 Component Analysis Device
[0166] 12 Objects
[0167] 20 Data Acquisition Department
[0168] 22. Spectroscopic light source
[0169] 24. Spectroscopic light receiving unit
[0170] 30 Analysis Department
[0171] 32 Information Processing Department
[0172] 40 Type Acquisition Section
[0173] 50 Storage
[0174] 52 Reference spectral dataset
[0175] 54 Corresponding information
[0176] 60 Output
[0177] 62 Display
Claims
1. A component analysis device, have: a data acquisition circuit that acquires spectrum data of the object obtained by measuring the spectrum of the object including a plurality of components using a sensor; A type acquisition circuit that acquires information indicating the type of the object; a storage unit storing reference spectral data sets that are different from each other for each type of the object, the reference spectral data sets including a plurality of spectral data, each of the plurality of spectral data being spectral data of a plurality of substances assumed to be included in the object; and The analysis circuit performs parallel factor analysis using the spectral data of the object and the reference spectral data set corresponding to the type of the object as input data, thereby separating a plurality of characteristic factors, and based on the score of each characteristic factor among the plurality of characteristic factors relative to the spectral data of the object, determines whether the substance corresponding to each characteristic factor is contained in the object.
2. The component analysis device according to claim 1, When the object includes a substance not included in the reference spectrum data set, The analysis circuit identifies a substance that is not expected to be included in the object.
3. The component analysis device according to claim 1 or 2, The above-mentioned spectrum data is three-dimensional fluorescence spectrum data.
4. The component analysis device according to claim 1 or 2, The above-mentioned object is food, Based on the result of the parallel factor analysis, it is determined whether or not the object contains any substance that is not contained in the food.
5. The component analysis device according to claim 1 or 2, The above-mentioned object is tap water, river water or industrial wastewater. The reference spectrum data set corresponding to the tap water, the river water or the industrial wastewater contains amino acids, Based on the results of the above parallel factor analysis, it is determined whether or not the amino acid is present in the above object.
6. The component analysis device according to claim 1 or 2, The above-mentioned objects are aerosol particles. The reference spectral data set corresponding to the aerosol particles includes pollen, Based on the results of the above parallel factor analysis, the presence or absence of pollen suspended in the air was determined.
7. A component analysis method, include: acquiring spectrum data obtained by measuring a spectrum of an object including a plurality of components; obtaining information indicating the type of the object; acquiring reference spectral data sets that are different from each other for each type of the object, the reference spectral data sets including a plurality of spectral data, each of the plurality of spectral data being spectral data of a plurality of substances assumed to be included in the object; as well as At the same time, parallel factor analysis is performed using the spectral data of the above-mentioned object and the above-mentioned reference spectral data set corresponding to the type of the above-mentioned object as input data, so as to separate a plurality of characteristic factors, and based on the score of each characteristic factor among the above-mentioned plurality of characteristic factors relative to the spectral data of the above-mentioned object, it is determined whether the substance corresponding to each characteristic factor is contained in the above-mentioned object.
8. A computer-readable recording medium storing a program for performing component analysis of an object, When the above program is executed by the above computer, the following is executed: acquiring spectrum data obtained by measuring a spectrum of an object including a plurality of components; obtaining information indicating the type of the object; acquiring reference spectral data sets that are different from each other for each type of the object, the reference spectral data sets including a plurality of spectral data, each of the plurality of spectral data being spectral data of a plurality of substances assumed to be included in the object; and At the same time, parallel factor analysis is performed using the spectral data of the above-mentioned object and the above-mentioned reference spectral data set corresponding to the type of the above-mentioned object as input data, so as to separate a plurality of characteristic factors, and based on the score of each characteristic factor among the above-mentioned plurality of characteristic factors relative to the spectral data of the above-mentioned object, it is determined whether the substance corresponding to each characteristic factor is contained in the above-mentioned object.
