Imaging analysis data analysis method and imaging analysis system

Abstract

One aspect of the present invention is an analysis method that uses imaging analysis data obtained by imaging analysis of one or a plurality of samples. The analysis method includes a step for creating a plurality of images (100) of different origin on the basis of the imaging analysis data, a step for dividing each of the plurality of images into a plurality of small regions and, using information that specifies each of the plurality of small regions as an explained variable and information that specifies a plurality of very small regions that are included in a single small region as an explanatory variable, performing dimension reduction processing on signal values for the very small regions to acquire an observation value (102) for each of the small regions, a step for using the observation value for each of the small regions to create a graph (103) for each of the plurality of images, and a step for comparing the plurality of images on the basis of the graph obtained for each of the plurality of images.

Description

Imaging analysis data analysis method and imaging analysis system

[0001] The present invention relates to an analysis method for analyzing imaging analysis data acquired by various imaging analyses such as imaging mass spectrometry and imaging spectroscopic analysis, and to an imaging analysis system used to implement such an analysis method.

[0002] In recent years, imaging mass spectrometers have been actively used to analyze pharmacokinetics of biological tissue slices, differences in compound distribution in various organs, and differences in compound distribution between pathological sites such as cancer and normal tissue.

[0003] In the imaging mass spectrometer described in Patent Document 1 and elsewhere, mass analysis is performed over a predetermined mass-to-charge ratio (m / z) range for multiple microregions (measurement points) within a two-dimensional measurement region on a sample, and mass analysis data is acquired for each microregion. From the collected imaging mass analysis data for the entire measurement region, a mass analysis imaging image (hereinafter sometimes referred to as an "MS imaging image") is created for each m / z value, showing the two-dimensional intensity distribution of ions having that m / z value. For example, if the sample is a biological sample such as a biological tissue slice, the sample contains multiple compounds, including metabolites. The above-mentioned MS imaging images for each m / z value reflect the concentration distribution of each of these multiple compounds, making it possible to visualize the distribution of these compounds.

[0004] In general, the number of MS imaging images that can be generated based on imaging mass spectrometry data is enormous. For example, in pharmacokinetic analysis, after administering a certain drug to an experimental animal, it may be necessary to comprehensively search for metabolites that show similar distributions in biological tissues. In such cases, it is a tedious task for a user to visually select images with similar intensity distributions from a huge number of MS imaging images. Therefore, conventional analysis systems using data analysis software such as those described in Non-Patent Document 1 have image analysis functions that utilize statistical analysis techniques such as hierarchical clustering to automatically group images with similar intensity distributions from a large number of MS imaging images or automatically extract images with similar intensity distributions.

[0005] International Publication No. 2021 / 044509

[0006] "IMAGEREVEAL MS: MS Imaging Data Analysis Software," [Online], [Retrieved July 8, 2024], Shimadzu Corporation, Internet <URL: https: / / www.an.shimadzu.co.jp / sites / an.shimadzu.co.jp / files / pim / pim_document_file / an_jp / brochures / 20334 / c146-2224.pdf>

[0007] As described above, when a user wants to check whether the classification of multiple MS imaging images automatically classified according to the similarity of their intensity distributions is appropriate, or to check the degree of similarity or difference between the intensity distributions of multiple MS imaging images classified into the same group, conventional analysis systems can employ a method of overlaying and comparing multiple MS imaging images on a display screen. However, because the MS imaging images are each colored according to signal intensity (in the case of a color scale display) or adjusted in black and white shading (in the case of a grayscale display), it is difficult for the user to grasp the similarity or difference between the intensity distributions from the overlaid images. Furthermore, such evaluation of similarity or difference is ultimately subjective. Furthermore, the lack of evaluation standards inevitably leads to variations in evaluations depending on the operator, and it is difficult to present the basis of the evaluation results to others.

[0008] Furthermore, while previous analytical systems were capable of comparing MS imaging images obtained from the same sample with different m / z values, or comparing MS imaging images obtained from different samples with the same or different m / z values, they were unable to perform comparative analysis of the analytical results obtained by imaging mass spectrometry with those obtained by other analytical techniques that provide information of a different type or perspective than imaging mass spectrometry, or to perform analysis that associates or combines such different types of analytical results.

[0009] The present invention has been made to solve these problems, and one of its main objectives is to provide an imaging analysis data analysis method and imaging analysis system that can improve the efficiency and reliability of analysis based on imaging analysis results by providing users with easy-to-understand and objective results of comparing intensity distributions for multiple samples or multiple compounds in the same sample.

[0010] Another object of the present invention is to provide an imaging analysis data analysis method and imaging analysis system that can obtain new insights and information about samples that have not been obtainable or have been difficult to obtain before by performing an analysis that correlates the analysis results obtained by imaging mass spectrometry with the results of analyses or measurements other than imaging mass spectrometry.

[0011] One aspect of the imaging analysis data analysis method of the present invention is an analysis method that uses imaging analysis data obtained by imaging analysis of one or more samples, and includes: an image creation step that creates a plurality of images, each of which has a different origin, based on the imaging analysis data; a dimension reduction step that divides each of the plurality of images into a plurality of small regions, uses information that identifies each of the plurality of small regions as an explained variable and information that identifies a plurality of microregions contained in a small region as an explanatory variable, and performs dimension reduction processing on the signal values ​​in each microregion to obtain observed values ​​for each of the small regions; a graph creation step that uses the observed values ​​for each of the plurality of images to create a graph with the position information of the small region on one axis and the observed value on the other axis; and an image comparison step that compares the plurality of images based on the graph obtained for each of the plurality of images.

[0012] Another aspect of the imaging analysis data analysis method of the present invention is an analysis method that uses imaging analysis data obtained by imaging analysis of a sample, and includes: an image creation step of creating at least one image based on the imaging analysis data; a dimension reduction step of dividing the at least one image into a plurality of small regions, using information identifying each of the plurality of small regions as an explained variable and information identifying a plurality of microregions contained in one small region as an explanatory variable, and performing a dimension reduction process on the signal values ​​in each microregion to obtain observed values ​​for each small region; an auxiliary analysis result acquisition step of obtaining auxiliary analysis results that correspond to each of the plurality of small regions or each of the regions obtained by integrating a plurality of adjacent small regions, or each of the regions obtained by integrating a plurality of microregions that correspond to each of the microregions or each of the regions obtained by integrating a plurality of adjacent microregions, based on analysis data obtained for the sample or an auxiliary sample that is substantially the same as the sample by an analytical method different from the imaging analysis; and an information acquisition step of associating the auxiliary analysis results with observed values ​​obtained at corresponding positions on the sample or the auxiliary sample to obtain information about the sample.

[0013] Another aspect of the imaging analysis system of the present invention is an imaging analysis system used in the imaging analysis data analysis methods of the two aspects described above, comprising: a measurement unit that acquires imaging analysis data by performing imaging analysis on one or more samples; an image creation unit that creates one or more images based on the imaging analysis data; a dimension reduction processing unit that divides each of the one or more images into a plurality of small regions, uses information identifying each of the plurality of small regions as explained variables and information identifying a plurality of microregions contained in one small region as explanatory variables, and performs dimension reduction processing on the signal values ​​in each microregion to acquire observed values ​​for each small region; a graph creation unit that uses the observed values ​​for each of the small regions for each of the one or more images to create a graph with the position information of the small region on one axis and the observed value on the other axis; and a display processing unit that outputs one or more graphs created by the graph creation unit to a display unit.

[0014] According to one aspect of the imaging analysis data analysis method and imaging analysis system of the present invention, a user can easily and objectively compare, for example, the intensity distributions of a predetermined compound contained in each of multiple samples or the intensity distributions of multiple different compounds contained in a single sample, based on a graph obtained for each image, and evaluate the similarity and difference between the intensity distributions. This can improve the efficiency of analytical work using, for example, imaging mass spectrometry. Furthermore, since it is possible to evaluate the similarity and difference between intensity distributions using numerical values ​​as a criterion, the reliability of the analysis can be improved.

[0015] Furthermore, according to another aspect of the imaging analysis data analysis method of the present invention, analysis results obtained by imaging mass spectrometry or the like can be correlated with analysis results obtained by a technique other than imaging mass spectrometry, such as chromatographic mass spectrometry of a sample obtained by physically extracting the sample from the original sample or by cutting the original sample, thereby obtaining distribution information that identifies specific compounds and distribution information of quantitative values. In this way, it is possible to obtain highly quantitative information about the sample and information that identifies the names of compounds, which is difficult to obtain by imaging mass spectrometry alone.

[0016] 1 is a schematic block diagram of an analysis system according to one embodiment of the present invention; FIG. 2 is a conceptual diagram of imaging mass spectrometry data obtained by the analysis system according to this embodiment and an MS imaging image based thereon; FIG. 3 is a conceptual diagram for explaining the processing procedure of an MS imaging image in the analysis system according to this embodiment; FIG. 4 is an explanatory diagram of an example of an analysis method using the analysis system according to this embodiment; FIG. 5 is an explanatory diagram of another example of an analysis method using the analysis system according to this embodiment; FIG. 6 is a schematic block diagram of an analysis system according to one modified example; FIG. 7 is a schematic block diagram of an analysis system according to another embodiment; FIG. 8 is a conceptual diagram for explaining the analysis and data processing procedures in the analysis system according to another embodiment; FIG. 9 is a schematic block diagram of an analysis system according to yet another embodiment.

[0017] [Examples of the above aspects] In the analysis methods and systems of the above aspects of the present invention, "imaging analysis" refers to a technique of performing some kind of analysis, measurement, or observation on a sample or specimen and imaging or visualizing the results. Therefore, an "image" created based on imaging analysis data has numerical information such as signal values ​​and intensity values ​​for each pixel.

[0018] Specific examples of imaging analysis techniques include mass spectrometry imaging, Raman spectroscopic imaging, infrared spectroscopic imaging, fluorescence imaging, chemiluminescence imaging, X-ray imaging, ultrasound imaging, surface analysis, etc., and also include photography using visible light or infrared light, etc. Furthermore, images obtained by imaging analysis may be part of a series of images such as moving images or time-lapse images.

[0019] Furthermore, the "analytical technique different from imaging analysis" here can refer to one of the imaging analyses described above, but is not limited to imaging analysis. For example, it may involve finely chopping a sample or auxiliary sample and performing various analyses, including mass spectrometry, on the resulting partial sample to obtain data.

[0020] Furthermore, in the analysis method and system of the above aspect of the present invention, the "identifying information" in "information identifying each of multiple small regions" and "information identifying multiple minute regions contained in one small region" can typically be a number indicating the position of the small region within the entire image, and a number indicating the position of the minute region within one small region, etc.

[0021] Furthermore, in the analysis method and system of the above aspect of the present invention, the "dimensionality reduction process" is also called dimensionality compression process, and one of the most typical techniques is principal component analysis, but this is not limited to this.

[0022] [Configuration of an Analysis System of an Embodiment] An analysis system that is an embodiment of an imaging analysis system that can implement an imaging analysis data analysis method according to the present invention will be described with reference to the accompanying drawings.

[0023] Fig. 1 is a schematic block diagram of the analysis system of this embodiment. Fig. 2 is a conceptual diagram of the measurement operation of imaging mass spectrometry in this system and the MS imaging image obtained thereby. The analysis system of this embodiment includes an imaging mass spectrometry unit 1 that performs measurements on a sample, a data processing unit 2, and an input unit 3 and a display unit 4 that serve as user interfaces.

[0024] The imaging mass spectrometer 1 includes, for example, a time-of-flight mass spectrometer equipped with a matrix-assisted laser desorption / ionization (MALDI) ion source, and as shown in Fig. 2, performs mass analysis on a number of micro-areas (measurement points) 51 obtained by dividing a two-dimensional measurement area 50 on a sample such as a biological tissue slice into a grid pattern, and acquires mass analysis data for each micro-area 51. Here, the mass analysis data is assumed to be mass spectrum data constituting a mass spectrum 52 over a predetermined m / z range as shown in Fig. 2, but the mass analysis data is not limited to this and may be, for example, MS analysis of a specific precursor ion. n It may also be spectral data.

[0025] The data processing unit 2 receives imaging mass spectrometry data including mass spectrum data for each microregion 51 collected by the imaging mass spectrometry unit 1 and performs predetermined processing, and includes functional blocks of a data storage unit 20, an imaging image creation unit 21, an instruction receiving unit 22, an image correspondence graph creation unit 23, and a display processing unit 24. The image correspondence graph creation unit 23 includes subordinate functional blocks of an image segmentation unit 230, a data matrix creation unit 231, a principal component analysis execution unit 232, and a line graph creation unit 233.

[0026] The data processing unit 2 is generally implemented as a personal computer or a more powerful workstation, and the functions of the above-mentioned functional blocks are achieved by running dedicated software (computer programs) installed on the computer. In this case, the input unit 3 is a pointing device such as a keyboard or a mouse, and the display unit 4 is a display monitor.

[0027] The computer program can be provided to the user by being stored on a computer-readable, non-transitory recording medium such as a CD-ROM, DVD-ROM, memory card, USB memory (dongle), etc. Alternatively, the computer program can be provided to the user in the form of data transfer via a communication line such as the Internet. Alternatively, the computer program can be pre-installed on a computer that is part of the system when the user purchases the system.

[0028] As described above, when imaging mass analysis is performed on a sample in the imaging mass analysis unit 1, imaging mass analysis data consisting of mass spectrum data for a large number of microregions 51 within a predetermined measurement region 50 on the sample is obtained. This imaging mass analysis data is stored for each sample in the data storage unit 20. The imaging mass analysis data is an ion intensity signal whose parameters are position information indicating the position (coordinate position) of each microregion 51 within the measurement region 50 along the mutually perpendicular x- and y-axes, and the m / z value.

[0029] Based on the imaging mass spectrometry data described above, multiple images (MS imaging images) with different m / z values ​​can be created, as shown in the upper part of Figure 2. A single MS imaging image 100 shows the two-dimensional intensity distribution of ions with a single m / z value. In many cases, the MS imaging image 100 is a color image, where the display color corresponds to the ion intensity, or a grayscale image, where the display color corresponds to the ion intensity. As many MS imaging images 100 as there are m / z values ​​within the m / z range to be measured can be obtained. Generally, a sample contains a variety of compounds, each with a specific molecular weight. However, multiple compounds may have substantially the same molecular weight (within a range that cannot be distinguished by the mass resolution of the instrument). Therefore, a single m / z value may correspond to only one compound, or multiple compounds. Therefore, a single MS imaging image 100 does not necessarily represent the concentration distribution of a single compound.

[0030] [Processing Operation in the Analysis System of the Present Embodiment] Next, an example of analysis that is performed in a state where imaging mass analysis data is stored in the data storage unit 20 will be described.

[0031] The user operates the input unit 3 to specify multiple compounds or m / z values ​​of interest. This specification method is not limited to a specific method. For example, the user may specify the compound by directly inputting the compound name or m / z value. Alternatively, the user may select an appropriate peak from a peak list created by performing peak detection on a portion of the mass spectrum or an average mass spectrum obtained from the sample, thereby specifying the m / z value corresponding to the peak. Alternatively, the imaging image creation unit 21 may create an MS imaging image of the m / z value corresponding to the peak detected in the mass spectrum and display it on the screen of the display unit 4, so that the user can specify the m / z value by viewing the MS imaging image.

[0032] In either case, the instruction receiving unit 22 receives the user's specification of a compound or m / z value and determines one or more target m / z values. When a compound is specified, the m / z value associated with the specified compound can be identified, for example, by referring to a compound database prepared in advance. The imaging image creation unit 21 reads imaging mass spectrometry data corresponding to the specified m / z value from the data storage unit 20 and forms an MS imaging image 100 such as that shown in FIG. 2.

[0033] Next, the image correspondence graph creation unit 23 creates a line graph corresponding to each of the formed MS imaging images as follows. FIG. 3 is a conceptual diagram showing an example of the processing procedure at this time. In this example, one MS imaging image 100 corresponding to a certain m / z value is assumed to consist of n pixels in the horizontal axis (x-axis) direction and m pixels in the vertical axis (y-axis) direction. A signal value (ion intensity value) is associated with each pixel, and a display color (or shade of black or white) corresponding to the signal value is assigned to each pixel, thereby creating the MS imaging image 100.

[0034] In the image correspondence graph creation unit 23, the image division unit 230 divides a given MS imaging image 100 into a plurality of small regions. In this example, one small region is defined as a region that is elongated in the y-axis direction and that is composed of m pixels aligned in the y-axis direction. Therefore, the image division process by the image division unit 230 results in n small regions that are aligned along the x-axis and have the shape of strips that are elongated in the y-axis direction. Here, an image corresponding to one small region is referred to as a divided image.

[0035] However, the method of dividing the MS imaging image 100 is not limited to this. For example, the image may be divided such that a narrow region in the x-axis direction, consisting of n pixels aligned in the x-axis direction, is defined as one small region, and m small regions (divided images) aligned in the y-axis direction are obtained. Alternatively, a rectangular region consisting of m pixels in the y-axis direction and t pixels (t is an integer greater than or equal to 2) in the x-axis direction (i.e., a region of t × m pixels) may be defined as one small region, and the MS imaging image 100 may be divided every t pixels in the x-axis direction to obtain multiple small regions (divided images). Although this is not required, dividing the image so that multiple small regions are aligned along either the x-axis direction or the y-axis direction is convenient for creating a line graph, as described below.

[0036] Next, the data matrix creation unit 231 arranges information identifying each of the n small regions (here, numbers indicating coordinate positions 1, 2, ..., n) vertically in the matrix table, arranges information identifying each pixel included in each small region (here, numbers indicating coordinate positions 1, 2, ..., m) horizontally, and places a signal value in each cell of the matrix table, thereby completing a data matrix 101 with n rows and m columns.

[0037] Next, the principal component analysis execution unit 232 performs a data dimensionality reduction process on the data matrix 101 by performing principal component analysis on the data matrix 101, with each coordinate position on the horizontal y-axis as an explanatory variable and each coordinate position on the vertical x-axis as an explained variable. As is well known, principal component analysis reduces the dimensionality of the data so that each explained variable can be expressed as a linear combination of multiple explanatory variables. Here, since there are m explanatory variables, the number of dimensions is reduced from m to a number much smaller than m. Principal component analysis is well known, so a detailed explanation will be omitted.

[0038] When the above-described principal component analysis is performed on the data matrix 101, score values ​​(observation values) on new principal component axes, such as the first principal component C1, the second principal component C2, ..., are calculated for each row of the data matrix 101 in place of signal values. The post-principal component analysis matrix 102 shown at the bottom of Figure 3 shows up to the second principal component C2, but it is also possible to obtain the third principal component C3 and beyond. Here, we focus only on the score value of the first principal component C1, which is the most important in explaining the dependent variable.

[0039] The line graph creation unit 233 extracts score values ​​of the first principal components C1 arranged in a vertical row in the post-principal component analysis matrix 102 and plots them on a two-dimensional graph with the horizontal axis representing the position of the small region (here, the coordinate position on the x-axis) and the vertical axis representing the score value of the first principal component C1, thereby creating a line graph 103. In other words, this line graph 103 is a graph showing the relationship between the positions of n small regions in one MS imaging image and the score values ​​of the first principal component. The line graph 103 is a graph obtained by compressing m-dimensional information in the MS imaging image 100 into one dimension, and the state of the lines of the graph, i.e., the overall slope and shape, can be considered to roughly represent the distribution of signal values ​​in the MS imaging image 100.

[0040] Since a line graph 103 is obtained for each MS imaging image 100, for example, when three m / z values ​​are specified, three line graphs are obtained. The display processing unit 24 displays the three line graphs thus obtained as line graphs with common vertical and horizontal axes on the display unit 4, for example, together with the original MS imaging image.

[0041] [Example of Comparative Analysis in the Analysis System of the Present Embodiment] In the analysis system of the present embodiment, when an instruction to execute an analysis is given for imaging mass spectrometry data obtained from a single sample by specifying multiple different m / z values, a comparative analysis of the intensity distributions at the multiple m / z values ​​can be performed. Figure 4 is a schematic diagram showing the concept of this comparative analysis.

[0042] Based on the acquired imaging mass spectrometry data, it is possible to create MS imaging images at each m / z value: m / z1, m / z2, m / z3, ..., m / z100. Assume now that the user has selected three m / z values, m / z1, m / z3, and m / z100, as the m / z values ​​to be compared. The data processing unit 2 then executes the above-described data matrix creation, principal component analysis, and line graph creation processes for each of the MS imaging images corresponding to the three m / z values ​​(images enclosed by thick dotted lines in FIG. 4 ) of m / z1, m / z3, and m / z100. As a result, line graphs are obtained for each of the three images. The display processing unit 24 then creates a line graph 104 by overlaying these three graphs on the same axis, as shown on the right side of FIG. 4 , and displays it on the display unit 4.

[0043] As described above, the line graph can be considered to roughly represent the intensity distribution of the MS imaging image, and therefore, if the overall slope and shape of the lines in the graph are similar, the intensity distributions of the MS imaging image should also tend to be similar. Therefore, the user can look at the displayed line graph 104 and determine the similarity or difference in the shape of the lines, etc., to determine whether the spatial intensity distributions at the selected three m / z values ​​are similar. Of course, similar methods can be used to evaluate whether the spatial intensity distributions at more m / z values ​​are similar.

[0044] For example, after a certain time has elapsed since a specific drug was administered to a subject such as an experimental animal, a line graph can be used to examine whether the spatial intensity distributions of multiple different types of metabolites in the biological tissue of the subject are similar or different, thereby making it possible to evaluate whether the metabolites are likely to show similar kinetics or whether they do not show similar kinetics.

[0045] In addition to examining the similarities and differences between multiple MS imaging images derived from a single sample, this analysis system is also useful for examining the similarities and differences between multiple MS imaging images derived from different samples. Figure 5 shows a conceptual diagram of comparative analysis of multiple MS imaging images obtained from different samples at the same m / z value. This example shows a comparison of MS imaging images at the same m / z value (here, m / z 3) obtained by performing imaging mass spectrometry on three samples, Sample A, Sample B, and Sample C.

[0046] The data processing unit 2 performs the above-described processes of data matrix creation, principal component analysis, and line graph creation for each of the three MS imaging images derived from different samples corresponding to the m / z value (m / z3) specified by the user. As a result, a line graph is obtained for each image. The display processing unit 24 creates a line graph 105 in which graphs corresponding to the three samples A, B, and C are superimposed on the same axis, as shown on the right side of FIG. 5 , and displays the line graph 105 on the display unit 4. From the displayed line graph 105, the user can evaluate whether the spatial intensity distributions of the same compound in the three different samples are similar or different. Of course, MS imaging images derived from different samples with different m / z values ​​may also be compared.

[0047] When comparing multiple MS imaging images derived from different samples, it is usually necessary that the MS imaging images are meaningful to compare. Specifically, when the samples are slices of an organ (e.g., a brain) from an experimental animal, it is desirable that the samples are from the same experimental animal and organ, and that the slice configuration and the orientation of the sample on a slide glass or the like are approximately the same. Of course, if necessary, appropriate image transformation processing such as positional movement, rotation, enlargement / reduction may be performed on the MS imaging images to roughly align the positions of the outer edges of biological tissues and tissue boundaries in the multiple MS imaging images, thereby improving the accuracy of comparison in line graphs.

[0048] Furthermore, in the above explanation, a line graph was created using the results of principal component analysis based on the MS imaging image obtained by the imaging mass spectrometry unit 1, but similar processing can also be performed to create graphs for imaging images obtained by other imaging analysis techniques. Specifically, the analysis technique is not particularly limited as long as it is possible to obtain an imaging image containing some information about the sample or specimen, such as optical analysis such as Raman spectroscopy, infrared spectroscopy, or fluorescence spectroscopy, surface analysis using X-rays or electron beams, or surface analysis using a scanning probe microscope.

[0049] Furthermore, the imaging images to be analyzed herein can also include images of samples or specimens (subjects) obtained by observation using various types of imaging devices, including general imaging devices, infrared imaging devices, high-speed imaging devices, etc. Such imaging devices may be capable of obtaining moving images or time-lapse images, and the image to be analyzed may be extracted from a plurality of images obtained continuously or discretely over time.

[0050] [Configuration and Processing Operation of Analysis System According to Modification] Figure 6 is a schematic block diagram of an analysis system according to a modification of the above embodiment. This differs from the analysis system of the above embodiment in that the data processing unit 2 includes a correlation evaluation unit 25 as a functional block. That is, in the analysis system of the above embodiment, line graphs 104 and 105, as shown in the right of Figure 4 and the right of Figure 5, respectively, are displayed on the screen of the display unit 4, and the user can check these displays to determine the similarity or difference in intensity distributions between different MS imaging images. In contrast, the analysis system of this modification has a function of evaluating the correlation of intensity distributions between multiple MS imaging images based on the created line graphs.

[0051] In this analysis system, as described above, when a line graph 104 such as that shown on the right side of FIG. 4 is obtained based on MS imaging images corresponding to multiple m / z values ​​specified by the user, the correlation evaluation unit 25 calculates correlation index values ​​such as correlation coefficients and similarities for the slopes and shapes of each line in the line graph 104 according to a known predetermined algorithm. At this time, the correlation index value may be calculated after performing a process to correct for any offset between the multiple lines in the vertical axis (score value) direction so that the offset is not reflected in the correlation index value. Alternatively, the correlation index value may be calculated including the offset in the vertical axis direction. The display processing unit 24 displays the correlation index values ​​thus obtained together with the line graph on the display unit 4 and provides them to the user.

[0052] Furthermore, as described above, instead of examining the correlation of a line graph created based on, for example, a specific MS imaging image designated by a user, it is also possible to create line graphs using the above-described technique for all MS imaging images acquired by imaging mass spectrometry or for a large number of MS imaging images that satisfy specific conditions, and examine the correlation among these large number of line graphs. In this case, for example, by grouping line graphs that are highly correlated, it is also possible to classify the MS imaging images corresponding to the line graphs, that is, the m / z values, compounds, or samples, into those with similar intensity distributions. In other words, it is possible to use line graphs to classify a large number of MS imaging images or to extract MS imaging images with particularly similar distributions.

[0053] [Configuration and Operation of Analysis System of Another Embodiment] Fig. 7 is a schematic block diagram of an analysis system according to another embodiment, which is an extension of the analysis system according to the embodiment described above. Components that are the same as or equivalent to those in the analysis system of the embodiment described above are given the same reference numerals, and descriptions thereof will be omitted unless otherwise required.

[0054] This analysis system includes an imaging mass spectrometer 1 and a liquid chromatograph mass spectrometer 5 for measuring samples. Although not shown, as is well known, the liquid chromatograph mass spectrometer 5 includes a liquid chromatograph that separates multiple compounds contained in a liquid sample over time, and a mass spectrometer that performs mass analysis on the separated compounds. The imaging mass spectrometer 1 and the mass spectrometer of the liquid chromatograph mass spectrometer 5 can share part of their configuration. Specifically, the "iMScope" manufactured by Shimadzu Corporation can be used. TM The "LCMS Q-TOF" and "LCMS Q-TOF" can share the main components of the mass spectrometer.

[0055] In addition to the functional blocks in the analysis system of the above embodiment, the data processing unit 2 includes a chromatogram creating unit 26, a peak processing unit 27, a line graph creating unit 28, and an integrated analysis unit 29 as functional blocks.

[0056] FIG. 8 is a conceptual diagram illustrating the characteristic analysis and data processing procedures of this analysis system. This analysis system also creates MS imaging images at one or more m / z values ​​(or compounds) specified by the user based on imaging mass spectrometry data obtained for a given sample, performs principal component analysis based on the MS imaging images, and creates a line graph from the results, just like the analysis system of the above embodiment (see FIG. 3 ). For simplicity's sake, let us assume, as an example, that the number of pixels n in the x-axis direction of the MS imaging image 100 shown in FIG. 3 is 100. In this case, the line graph 202 is composed of n=100 plots.

[0057] On the other hand, liquid chromatography mass spectrometry (LC / MS analysis) is performed on the sample on which imaging mass spectrometry has been performed, or on another sample (auxiliary sample) whose spatial distribution of various compounds can be considered to be substantially the same as that of the sample. Here, the other sample (auxiliary sample) whose spatial distribution of various compounds can be considered to be substantially the same as that of the sample is, for example, a slice sample sliced ​​adjacent to or close to a certain slice sample in the thickness direction, if the sample is a thin slice of biological tissue. The main reason for using an auxiliary sample as the subject of LC / MS analysis rather than the sample on which imaging mass spectrometry has been performed is that in imaging mass spectrometry, the sample is irradiated with laser light for MALDI ionization, which is likely to result in the loss or dissipation of at least some or most of the compounds contained on the sample surface.

[0058] During LC / MS analysis, as shown in FIG. 8 , the sample 60 is first divided into sections in the x-axis direction, for example, at intervals of 10 pixels. As described above, if the width of the sample 60 in the x-axis direction is 100 pixels, the sample 60 is divided into 10 sections by dividing it into 10 sections every 10 pixels. This results in 10 elongated divided samples in the y-axis direction. Each of the 10 divided samples is homogenized to prepare liquid samples #1 to #10 in which the compounds contained in each divided sample are dissolved. The resulting 10 liquid samples #1 to #10 are then subjected to LC / MS analysis using the liquid chromatograph mass spectrometry unit 5, thereby obtaining LC / MS analysis data. The LC / MS analysis data corresponding to each of the liquid samples #1 to #10 is stored in the data storage unit 20. The LC / MS analysis data for one liquid sample is three-dimensional data based on the three axes of retention time, m / z value, and signal intensity.

[0059] Next, the chromatogram creation unit 26 creates extracted ion chromatograms for one or more m / z values ​​designated in advance by the user, as described above, based on the LC / MS analysis data for each liquid sample. If a compound corresponding to that m / z value is contained in the liquid sample, a peak corresponding to the target compound is observed in the extracted ion chromatogram. The peak processing unit 27 then detects the peak corresponding to the target compound in each extracted ion chromatogram and calculates the area value of the detected peak. If the target compound is known, more accurate peak detection can be achieved by utilizing the retention time corresponding to that compound. The peak area value can be estimated to correspond to the content (concentration) of the target compound. Note that peak height values ​​may be used instead of peak area values. This allows the peak area values ​​(i.e., quantitative values) of the target compounds corresponding to liquid samples #1 to #10 to be determined.

[0060] The line graph creation unit 28 creates a line graph (hereinafter referred to as a "quantitative value line graph") 201, with the horizontal axis representing the coordinate position of the divided sample and the vertical axis representing the peak area value or the quantitative value obtained from the peak area value using a calibration curve. If there are 10 liquid samples, i.e., 10 divided samples, this line graph will be composed of 10 plots, and the plots will be much coarser than those of a line graph obtained from an MS imaging image.

[0061] The integrated analysis unit 29 then performs a predetermined analysis process by associating the quantitative value information of the target compound for each liquid sample obtained as a result of the LC / MS analysis or a line graph reflecting the information with the line graph for each m / z value or each target compound obtained as a result of the imaging mass spectrometry. As an example, the following analysis is possible.

[0062] Imaging mass spectrometry may simultaneously analyze multiple compounds contained in a sample. However, if the m / z values ​​of ions from different compounds are the same or very similar, they cannot be distinguished from one another (although MS / MS analysis may enable this, the detection sensitivity is inevitably reduced). Furthermore, ion suppression by compounds present in relatively large quantities in the sample can reduce the ionization efficiency of the target compound. For these reasons, it is difficult to improve quantitative accuracy with imaging mass spectrometry. In contrast, LC / MS analysis can separate a significant proportion of the diverse compounds contained in a liquid sample in the upstream liquid chromatograph before performing mass analysis. This significantly reduces the overlap of multiple compounds compared to imaging mass spectrometry, enabling high quantitative accuracy for the target compound.

[0063] The horizontal axis of the quantitative value line graph 201 is substantially the same as that of the line graph 202 obtained from the MS imaging image, except for the plot interval. Therefore, if the MS imaging image reflects the intensity distribution of the target compound, the quantitative value line graph 201 and the line graph 202 for the target compound should have similar slopes and shapes, and should show a high correlation. Therefore, an index value such as a correlation coefficient between the two line graphs 201 and 202 is calculated, and the presence or absence of a correlation is determined based on the index value.

[0064] If the correlation is high, it can be assumed that the original mass spectrometry image shows the intensity distribution of the target compound, and conversely, if the correlation is low, it can be assumed that the original mass spectrometry image does not show the intensity distribution of the target compound (for example, because another compound is overlapping). Thus, according to the analysis system of this embodiment, by examining the correlation between the area value line graph 201 and the line graph 202, it can be confirmed whether the mass spectrometry image at a specific m / z value is the intensity distribution of the target compound.

[0065] Furthermore, when multiple peaks with different retention times are observed in an extracted ion chromatogram at a certain m / z value, it can be inferred that the MS imaging image at that m / z value is likely to reflect the distribution of multiple compounds. In this case, even if the multiple compounds include an unknown compound, it may be possible to identify the compound by querying a compound database for the retention time and m / z value (or the mass spectrum peak pattern at that retention time) determined from the peak position in the extracted ion chromatogram. In this way, it may be possible to identify unknown compounds contained in a sample using the results of LC / MS analysis and understand the compounds reflected in the distribution of the MS imaging image.

[0066] This also applies when the distributions of multiple compounds do not overlap in the MS imaging image, but reflect the distribution of a single unknown compound. In other words, even when it is difficult to identify an unknown compound based on the m / z value alone, it may be possible to identify the compound using information on the retention time determined by LC / MS analysis, and in such cases, it becomes possible to understand the compound reflected in the distribution in the MS imaging image.

[0067] If the mass spectrometer of the liquid chromatograph mass spectrometer 5 is configured to be capable of MS / MS analysis, a product ion spectrum can be obtained using ions with a specific m / z value as precursor ions, and the structure of the original compound can be estimated from the m / z value of the peak observed in the product ion spectrum. Therefore, it is possible to correlate the MS imaging image with the corresponding compound structure and present it to the user.

[0068] [Modification of the Analysis System] In the analysis system of the other embodiment described above, imaging mass spectrometry and LC / MS analysis were performed on samples that can be considered to be the same or substantially the same. However, it is also possible to perform imaging mass spectrometry and an analysis by a method other than mass spectrometry on samples that can be considered to be the same or substantially the same, and perform an integrated analysis using the results of both analyses. Figure 9 is a schematic block diagram showing an example of such an analysis system. In this example, the analysis system includes both an imaging mass spectrometry unit 1 and an imaging Raman spectrometry unit 6.

[0069] In this analysis system, an imaging mass spectrometer 1 performs imaging mass analysis on a prepared sample and collects imaging mass analysis data. A data processor 2 creates an MS imaging image at a specified m / z value based on the collected imaging mass analysis data, and generates a line graph corresponding to that image. Up to this point, the system is similar to the analysis system described above.

[0070] On the other hand, the imaging Raman spectroscopy analysis unit 6 performs Raman spectroscopy on each microregion on the sample on which imaging mass spectrometry has been performed or on another sample (auxiliary sample) whose spatial distribution of various compounds can be considered to be substantially the same as that of the sample, and acquires imaging Raman spectroscopy data showing the relationship between Raman shift and Raman scattering intensity. However, the size of the microregion (in other words, the distance between adjacent measurement points) at this time does not need to be the same as that at the time of imaging mass spectrometry (it is usually not the same). The acquired imaging Raman spectroscopy analysis data is stored in the data storage unit 20.

[0071] The imaging image creation unit 21 creates a Raman spectroscopic imaging image for each Raman shift (wavelength) from the imaging Raman spectroscopic analysis data, and the image correspondence graph creation unit 23 creates a line graph (hereinafter referred to as a "Raman spectroscopic line graph" for convenience) showing changes in the score value of the first principal component from each Raman spectroscopic imaging image using the above-mentioned method. That is, with this analysis system, a line graph for each MS imaging image and a Raman spectroscopic line graph for each Raman spectroscopic imaging image can be obtained for the same sample or substantially the same sample. Therefore, new knowledge about the sample can be obtained based on the correlation between these two types of line graphs based on different analysis methods.

[0072] As described above, when the sample is a biological tissue section, the MS imaging image shows the concentration distribution of one or more specific metabolites, etc. On the other hand, when the sample is a biological tissue section, it is known that a Raman spectroscopic imaging image at a specific Raman shift shows, for example, the distribution of a specific compound or protein, or the regional differentiation between abnormal regions (such as lesions) and normal regions of the biological tissue. Therefore, for example, by searching for a line graph that shows a high correlation with a Raman spectroscopic line graph corresponding to a specific Raman spectroscopic imaging image, it may be possible to perform analyses that are difficult to perform using conventional imaging mass spectrometry alone, such as extracting metabolites that may be related to a specific disease.

[0073] Of course, any combination of imaging analysis techniques can be used, such as a combination of imaging mass spectrometry and imaging Raman spectrometry, and an combination of imaging mass spectrometry and imaging fluorescence analysis.

[0074] It should be noted that the above-described embodiment and its modifications are merely examples of the present invention, and it goes without saying that any appropriate changes, modifications, or additions made within the spirit of the present invention will also fall within the scope of the claims of the present application.

[0075] Aspects It will be apparent to those skilled in the art that the exemplary embodiments described above are examples of the following aspects.

[0076] (Item 1) One aspect of the imaging analysis data analysis method of the present invention is an analysis method that utilizes imaging analysis data obtained by imaging analysis of one or more samples, and includes: an image creation step that creates a plurality of images, each of which has a different origin, based on the imaging analysis data; a dimension reduction step that divides each of the plurality of images into a plurality of small regions, uses information that identifies each of the plurality of small regions as an explained variable and information that identifies a plurality of microregions contained in a small region as an explanatory variable, and performs dimension reduction processing on the signal values ​​in each microregion to obtain observed values ​​for each of the small regions; a graph creation step that uses the observed values ​​for each of the plurality of images to create a graph with the position information of the small region on one axis and the observed value on the other axis; and an image comparison step that compares the plurality of images based on the graph obtained for each of the plurality of images.

[0077] In the imaging analysis data analysis method described in paragraph 1, the multiple images may be images based on data obtained for a single sample under different analytical parameter values, or may be images based on data obtained for multiple samples under the same (or different) analytical parameter values. For example, if the imaging analysis is imaging mass spectrometry, the analytical parameter value is the m / z value, and if the imaging analysis is imaging Raman spectroscopy, the analytical parameter value is the Raman shift.

[0078] According to the imaging analysis data analysis method described in paragraph 1, a user can easily and objectively compare, for example, the intensity distributions of a given compound contained in multiple samples or the intensity distributions of multiple different compounds contained in a single sample based on a graph obtained for each image, and evaluate the similarity and difference between the intensity distributions. This can improve the efficiency of analytical work using, for example, imaging mass spectrometry. Furthermore, since the similarity and difference can be evaluated using numerical values, the reliability of the analysis can be improved.

[0079] (2) In the imaging analysis data analysis method described in (1), the dimension reduction process may be principal component analysis, and the observed value may be a score value in at least one principal component obtained by the principal component analysis.

[0080] As is well known, principal component analysis uses relatively simple calculations to aggregate data having many explanatory variables to create principal components, which are new axes suitable for explaining the dependent variable, and to obtain scores on the principal component axes. Therefore, the imaging analysis data analysis method described in Section 2 makes it possible to create graphs suitable for comparing multiple images with a relatively small amount of calculation.

[0081] (3) In the imaging analysis data analysis method described in paragraph 1 or 2, the dimension reduction step can divide the rectangular image into a plurality of small regions along the extension direction of one side of the image, and the graph creation step can create a graph in which the position of each small region is plotted on an axis corresponding to the one side and the observed values ​​are plotted in a direction perpendicular to the axis.

[0082] According to the imaging analysis data analysis method described in paragraph 3, the graph created is a line graph with the horizontal axis along one side of the image, which makes it easier for the user to understand the correspondence between the image and the graph, and to understand the similarities and differences between multiple images based on the graph.

[0083] (4) In the imaging analysis data analysis method described in any one of paragraphs 1 to 3, the imaging analysis may be imaging mass spectrometry, and the plurality of images may be intensity distribution images each having a different mass-to-charge ratio or mass-to-charge ratio range.

[0084] According to the imaging analysis data analysis method described in Section 4, compounds with similar intensity distributions can be identified or extracted, thereby enabling efficient searches for, for example, metabolites that show similar distributions in vivo.

[0085] (5) In the imaging analysis data analysis method described in any one of paragraphs 1 to 3, the imaging analysis may be imaging mass spectrometry, and the multiple images may be intensity distribution images obtained from different samples and having the same mass-to-charge ratio or mass-to-charge ratio range.

[0086] According to the imaging analysis data analysis method described in Section 5, for example, it is possible to easily determine whether the distribution of the same compound is similar between different individuals, thereby enabling efficient investigation of, for example, individual differences in the distribution of a certain metabolite.

[0087] (Item 6) Another aspect of the imaging analysis data analysis method of the present invention is an analysis method that uses imaging analysis data obtained by imaging analysis of a sample, and includes: an image creation step of creating at least one image based on the imaging analysis data; a dimension reduction step of dividing the at least one image into a plurality of small regions, using information identifying each of the plurality of small regions as explained variables and information identifying a plurality of microregions contained in one small region as explanatory variables, and performing a dimension reduction process on the signal values ​​in each microregion to obtain observed values ​​for each small region; an auxiliary analysis result acquisition step of obtaining auxiliary analysis results that correspond to each of the plurality of small regions or each of the regions obtained by integrating a plurality of adjacent small regions, or each of the regions obtained by integrating a plurality of microregions that correspond to each of the microregions or each of the regions obtained by integrating a plurality of adjacent microregions, based on analysis data obtained for the sample or an auxiliary sample that is substantially the same as the sample by an analytical method different from the imaging analysis; and an information acquisition step of associating the auxiliary analysis results with observed values ​​obtained at corresponding positions on the sample or the auxiliary sample to obtain information about the sample.

[0088] According to the imaging analysis data analysis method described in Section 6, by performing an analysis that correlates the analytical results obtained by imaging mass spectrometry or the like with the analytical results obtained by a method other than imaging mass spectrometry, for example, by chromatographic mass spectrometry of a sample obtained by physically extracting the sample from the original sample or by cutting the original sample, it is possible to obtain distribution information that identifies compounds and distribution information of quantitative values. In this way, it is possible to obtain highly quantitative information about the sample and information that identifies the names of compounds, which is difficult to obtain by imaging mass spectrometry alone.

[0089] (Item 7) In the imaging analysis data analysis method described in item 6, the dimension reduction process can be principal component analysis, and the observed value can be a score value in at least one principal component obtained by the principal component analysis.

[0090] According to the imaging analysis data analysis method described in Section 7, it is possible to obtain observation values ​​for each small region for performing analysis associated with auxiliary analysis results with a relatively small amount of calculation.

[0091] (Item 8) In the imaging analysis data analysis method described in Item 6 or 7, the imaging analysis may be imaging mass spectrometry, and the analytical technique different from the imaging analysis may be chromatographic mass spectrometry, and the auxiliary analysis results may be qualitative or quantitative information about compounds in each of the small regions or the integrated region.

[0092] In chromatographic mass spectrometry such as LC / MS analysis, components (compounds) in a sample are separated in a chromatographic section at the front end, and then mass analysis of each component is performed in a mass spectrometric section at the rear end. Therefore, compared to imaging mass spectrometry, higher accuracy can be achieved in both qualitative and quantitative analysis. Therefore, the imaging analysis data analysis method described in Section 8 utilizes high qualitative capabilities to, for example, confirm whether an MS imaging image at a certain mass-to-charge ratio value shows the intensity distribution of a compound corresponding to that mass-to-charge ratio value, and then identify the compound. Furthermore, an imaging image showing the distribution of quantitative values ​​such as concentration can be obtained from an MS imaging image showing the distribution of ion intensity.

[0093] (Item 9) The imaging analysis data analysis method described in any one of Items 6 to 8 may further include a graph creation step of using the observation values ​​for each small region obtained for at least one image to create a graph with position information of the small region on one axis and the observation value on the other axis, and the information acquisition step may create an auxiliary graph corresponding to the graph based on the auxiliary analysis results, and acquire information about the sample based on the graph and the auxiliary graph.

[0094] According to the imaging analysis data analysis method described in paragraph 9, it is possible to easily make a rough comparison between the intensity distribution in an image obtained by imaging mass spectrometry and the signal value distribution in an image obtained by a technique other than imaging mass spectrometry, and to efficiently evaluate similarities and differences.

[0095] (Item 10) One aspect of the imaging analysis system of the present invention is a system for implementing the imaging analysis data processing method described in items 1 or 6, and comprises: a measurement unit that acquires imaging analysis data by performing imaging analysis on one or more samples; an image creation unit that creates one or more images based on the imaging analysis data; a dimension reduction processing unit that divides each of the one or more images into a plurality of small regions, uses information identifying each of the plurality of small regions as explained variables and information identifying a plurality of microregions contained in one small region as explanatory variables, and performs dimension reduction processing on the signal values ​​in each microregion to acquire observed values ​​for each small region; a graph creation unit that uses the observed values ​​for each of the small regions for each of the one or more images to create a graph with the position information of the small region as one axis and the observed value as the other axis; and a display processing unit that outputs one or more graphs created by the graph creation unit to a display unit.

[0096] According to the imaging analysis system described in paragraph 10, for example, a user can easily and objectively compare the intensity distributions of a given compound contained in each of a plurality of samples or the intensity distributions of a plurality of different compounds contained in a single sample based on graphs obtained for each of a plurality of images, and evaluate the similarity and difference between the intensity distributions. This can improve the efficiency of analytical work using, for example, imaging mass spectrometry. Furthermore, since the similarity and difference can be evaluated using numerical values, the reliability of the analysis can be improved.

[0097] (Item 11) In the imaging analysis system described in Item 10, the image creation unit creates a plurality of images, and the dimension reduction processing unit and the graph creation unit create a plurality of graphs based on the plurality of images, and the system may further include an evaluation unit that compares the plurality of images based on the plurality of graphs.

[0098] According to the imaging analysis system described in paragraph 11, the user can determine whether the intensity distributions in multiple images are similar without having to compare multiple graphs. Alternatively, the user can refer to the evaluation results by the evaluation unit to determine whether the intensity distributions in multiple images are similar. This reduces the burden on the user of evaluation and judgment, and can also improve the efficiency of analysis.

[0099] (Item 12) In the imaging analysis system described in item 10 or 11, the dimensionality reduction process may be principal component analysis, and the observed value may be a score value in at least one principal component obtained by the principal component analysis.

[0100] According to the imaging analysis system described in paragraph 12, it is possible to obtain observation values ​​for each small region for performing analysis associated with auxiliary analysis results with a relatively small amount of calculation.

[0101] (Item 13) In the imaging analysis system described in any one of Items 10 to 12, the dimensionality reduction processing unit divides the rectangular image into a plurality of small regions along the extension direction of one side of the image, and the graph creation unit can create a graph in which the position of each small region is plotted on an axis corresponding to the one side and the observed values ​​are plotted in a direction perpendicular to the axis.

[0102] According to the imaging analysis system described in paragraph 13, the graph created is a line graph with the horizontal axis along one side of the image, which makes it easier for the user to understand the correspondence between the image and the graph, and to understand the similarities and differences between multiple images based on the graph.

[0103] (Item 14) In the imaging analysis system described in any one of items 10 to 13, the imaging analysis may be imaging mass spectrometry.

[0104] (Item 15) The imaging analysis system described in Item 14 may further include a second measurement unit that acquires analytical data on the sample or an auxiliary sample that is substantially identical to the sample using an analytical method other than imaging mass analysis; an auxiliary analysis result acquisition unit that, based on the analytical data obtained by the second measurement unit, obtains auxiliary analysis results that correspond to each of the multiple small regions or each of the regions obtained by integrating multiple adjacent small regions, or that correspond to each of the microregions or each of the regions obtained by integrating multiple adjacent microregions; and an information acquisition unit that associates the auxiliary analysis results with observation values ​​obtained at corresponding positions in the sample or the auxiliary sample to acquire information about the sample.

[0105] According to the imaging analysis system described in paragraph 15, by performing an analysis that correlates the results of imaging mass spectrometry with the results of chromatography mass spectrometry on a sample obtained by a method other than imaging mass spectrometry, for example, by physically extracting the sample from the original sample or by cutting the original sample, it is possible to obtain distribution information that identifies specific compounds and distribution information of quantitative values. In this way, it is possible to obtain highly quantitative information about the sample and information that identifies the names of compounds, which is difficult to obtain by imaging mass spectrometry alone.

[0106] 1... Imaging mass spectrometry unit 2... Data processing unit 20... Data storage unit 21... Imaging image creation unit 22... Instruction receiving unit 23... Image correspondence graph creation unit 230... Image division unit 231... Data matrix creation unit 232... Principal component analysis execution unit 233... Line graph creation unit 24... Display processing unit 25... Correlation evaluation unit 26... Chromatogram creation unit 27... Peak processing unit 28... Line graph creation unit 29... Integrated analysis unit 3... Input unit 4... Display unit

Claims

1. An analytical method using imaging analysis data obtained by imaging analysis of one or more samples, comprising: an image creation step of creating a plurality of images, each of which has a different origin, based on the imaging analysis data; a dimension reduction step of dividing each of the plurality of images into a plurality of small regions, using information identifying each of the plurality of small regions as an explained variable and information identifying a plurality of minute regions contained in a small region as an explanatory variable, and performing dimension reduction processing on the signal values ​​in each minute region to obtain an observed value for each small region; a graph creation step of creating a graph for each of the plurality of images, using the observed values ​​for each of the small regions, with the position information of the small region on one axis and the observed value on the other axis; and an image comparison step of comparing the plurality of images based on the graph obtained for each of the plurality of images.

2. The imaging analysis data analysis method according to claim 1, wherein the dimension reduction process is principal component analysis, and the observed value is a score value in at least one principal component obtained by the principal component analysis.

3. The imaging analysis data analysis method of claim 1, wherein the dimension reduction step divides the rectangular image into a plurality of small regions along the extension direction of one side of the image, and the graph creation step creates a graph in which the position of each small region is plotted on an axis corresponding to the one side and observed values ​​are plotted in a direction perpendicular to the axis.

4. The imaging analysis data analysis method according to claim 1, wherein the imaging analysis is imaging mass spectrometry, and the plurality of images are intensity distribution images each having a different mass-to-charge ratio or mass-to-charge ratio range.

5. The imaging analysis data analysis method of claim 1, wherein the imaging analysis is imaging mass analysis, and the multiple images are intensity distribution images each obtained from a different sample and having the same mass-to-charge ratio or mass-to-charge ratio range.

6. An analytical method using imaging analysis data obtained by imaging analysis of a sample, comprising: an image creation step of creating at least one image based on the imaging analysis data; a dimension reduction step of dividing the at least one image into a plurality of small regions, using information identifying each of the plurality of small regions as an explained variable and information identifying a plurality of microregions contained in one small region as an explanatory variable, and performing a dimension reduction process on the signal values ​​in each microregion to obtain observed values ​​for each small region; an auxiliary analysis result acquisition step of obtaining auxiliary analysis results corresponding to each of the plurality of small regions or each of the regions obtained by integrating a plurality of adjacent small regions, or each of the regions obtained by integrating a plurality of microregions corresponding to each of the microregions or each of the regions obtained by integrating a plurality of adjacent microregions, based on analytical data obtained for the sample or an auxiliary sample that is substantially the same as the sample by an analytical method different from the imaging analysis; and an information acquisition step of acquiring information about the sample by associating the auxiliary analysis results with observed values ​​obtained at corresponding positions on the sample or the auxiliary sample.

7. The imaging analysis data analysis method according to claim 6, wherein the dimension reduction process is principal component analysis, and the observed value is a score value in at least one principal component obtained by the principal component analysis.

8. The imaging analysis data analysis method described in claim 6, wherein the imaging analysis is imaging mass spectrometry, the analytical technique different from the imaging analysis is chromatographic mass spectrometry, and the auxiliary analysis results are quantitative information of compounds in each of the small regions or the integrated regions.

9. The imaging analysis data analysis method of claim 6, further comprising a graph creation step of creating a graph using the observation values ​​for each small region obtained for at least one image, with the position information of the small region on one axis and the observation value on the other axis, wherein in the information acquisition step, an auxiliary graph corresponding to the graph is created based on the auxiliary analysis results, and information about the sample is acquired based on the graph and the auxiliary graph.

10. An imaging analysis system comprising: a measurement unit that acquires imaging analysis data by performing imaging analysis on one or more samples; an image creation unit that creates one or more images based on the imaging analysis data; a dimension reduction processing unit that divides each of the one or more images into a plurality of small regions, uses information that identifies each of the plurality of small regions as an explained variable and information that identifies a plurality of minute regions contained in one small region as an explanatory variable, and performs dimension reduction processing on the signal values ​​in each minute region to acquire observed values ​​for each small region; a graph creation unit that uses the observed values ​​for each of the small regions for each of the one or more images to create a graph with position information of the small region on one axis and the observed value on the other axis; and a display processing unit that outputs one or more graphs created by the graph creation unit to a display unit.

11. The imaging analysis system of claim 10, further comprising: an image creation unit that creates a plurality of images; a dimension reduction processing unit and a graph creation unit that create a plurality of graphs based on the plurality of images; and an evaluation unit that compares the plurality of images based on the plurality of graphs.

12. The imaging analysis system of claim 10, wherein the dimension reduction process is principal component analysis, and the observed value is a score value in at least one principal component obtained by the principal component analysis.

13. The imaging analysis system of claim 10, wherein the dimension reduction processing unit divides the rectangular image into a plurality of small regions along the extension direction of one side of the image, and the graph creation unit creates a graph in which the position of each small region is plotted on an axis corresponding to the one side and observed values ​​are plotted in a direction perpendicular to the axis.

14. The imaging analysis system of claim 10, wherein the imaging analysis is imaging mass spectrometry.

15. The imaging analysis system of claim 14, further comprising: a second measurement unit that acquires analytical data on the sample or an auxiliary sample that is substantially the same as the sample using an analytical method other than imaging mass spectrometry; an auxiliary analysis result acquisition unit that, based on the analytical data acquired by the second measurement unit, determines auxiliary analysis results that correspond to each of the plurality of small regions or to each of the regions obtained by integrating adjacent small regions, or that correspond to each of the microregions or to each of the regions obtained by integrating adjacent microregions; and an information acquisition unit that acquires information about the sample by associating the auxiliary analysis results with observation values ​​obtained at corresponding positions in the sample or the auxiliary sample.

Images