Spectral high-sensitivity detection method for glycosylated hemoglobin

Abstract

The invention relates to the technical field of Raman scattering, in particular to a high-sensitivity spectrum detection method for glycosylated hemoglobin, which comprises the following steps: acquiring Raman spectrograms corresponding to a preset number of test samples of a patient to be detected, and determining a spectral line increment ascending sequence corresponding to each Raman spectrogram; screening out a smooth change interval, a component peak interval and a trailing possible interval; determining a structure variation index corresponding to each component peak interval; determining a homogeneity index corresponding to each Raman spectrogram; constructing a target matrix, and determining the size of a target window; and replacing the matrix in SERS detection with the target matrix, taking the target window size as the window size in SERS detection, and detecting the glycosylated hemoglobin of the to-be-detected patient through SERS detection. According to the invention, SERS detection of glycosylated hemoglobin is realized, and the rationality of window size setting is improved, so that the accuracy of glycosylated hemoglobin detection is improved.

Description

Technical Field

[0001] The invention relates to the technical field of Raman scattering, and in particular to a spectral high-sensitivity detection method for glycated hemoglobin. Background Art

[0002] SERS (Surface-Enhanced Raman Spectroscopy) technology is mainly used in material detection. For example, it can be used for high-sensitivity spectral detection of glycated hemoglobin. In the SERS detection process, if the window size is too small, some important Raman shift signals may not be captured, resulting in the loss of key information and affecting the integrity of the detection results; if the window is set too large, it may include too much background noise or irrelevant signals, thereby reducing the signal-to-noise ratio and affecting the accuracy of the signal. Therefore, in the SERS detection process, the setting of the window size is crucial. At present, in the SERS detection process, the window size is often set through manual experience.

[0003] However, in the process of realizing glycated hemoglobin detection through SERS detection technology, if the window size in SERS detection is set based on manual experience, the following technical problems often occur: When the window size is set based on manual experience, since the set window size is affected by manual subjective factors, the rationality of the window size setting may be poor, thereby resulting in poor accuracy of glycated hemoglobin detection. Summary of the invention

[0004] In order to solve the technical problem of poor accuracy of glycated hemoglobin detection due to poor rationality of window size setting, the present invention proposes a spectral high-sensitivity detection method for glycated hemoglobin.

[0005] In a first aspect, the present invention provides a method for spectral high-sensitivity detection of glycated hemoglobin, the method comprising: Obtaining Raman spectra corresponding to a preset number of test samples of the patient to be tested, and determining an ascending sequence of spectral line growth amounts corresponding to each Raman spectra; According to the ascending sequence of the spectral line growth amount corresponding to each Raman spectrum, the smooth change interval and the component peak interval are screened out from all wavelengths of each Raman spectrum; Based on all the gently changing intervals in each Raman spectrum, possible tailing intervals are screened out from each component peak interval in each Raman spectrum; According to the Raman intensity difference and wavelength difference between each component peak interval and its possible tailing interval, the structural variation index corresponding to each component peak interval is determined; According to the structural variation indexes corresponding to all component peak intervals in each Raman spectrum, the homogeneity index corresponding to each Raman spectrum is determined; According to all structural variation indicators and the lengths of all gently changing intervals, a target matrix is ​​constructed, and based on all homogeneity indicators, the target window size is determined; The matrix in SERS detection is replaced by the target matrix, and the target window size is used as the window size in SERS detection. Through SERS detection, the glycated hemoglobin of the patient to be detected is detected.

[0006] In combination with the first aspect, in a possible implementation, the step of selecting the smooth change interval and the component peak interval from all wavelengths of each Raman spectrum according to the ascending sequence of the spectral line growth amount corresponding to each Raman spectrum includes: The absolute value of the difference between each two adjacent spectral line growth amounts in the ascending sequence of each spectral line growth amount is determined as a reference difference; According to all reference differences corresponding to the ascending sequence of the spectral line growth amount corresponding to each Raman spectrum, a smoothly changing interval is screened out from all wavelengths of each Raman spectrum; The continuous wavelengths of all wavelengths in each Raman spectrum except for the gently changing interval constitute the component peak interval.

[0007] In combination with the first aspect, in a possible implementation, the method of selecting a smoothly changing interval from all wavelengths of each Raman spectrum graph according to all reference differences corresponding to the ascending sequence of spectral line growth amounts corresponding to each Raman spectrum graph includes: Determine any Raman spectrum as a marked spectrum, and determine the ascending sequence of spectral line growth amounts corresponding to the marked spectrum as the ascending sequence of marked growth amounts; Determine two spectral line growth amounts corresponding to the maximum values ​​of all reference differences corresponding to the ascending sequence of the mark growth amounts as the first temporary growth amount and the second temporary growth amount respectively; Taking the first temporary growth amount and the second temporary growth amount as the segmentation points, segmenting the ascending sequence of the label growth amount to obtain a spectral line growth quantum sequence; Determine the maximum spectral line growth amount in the ascending sequence of the marked growth amounts as the marked maximum growth amount; Determine each spectral line growth quantum sequence in all spectral line growth quantum sequences except the spectral line growth quantum sequence to which the maximum growth amount is marked as a reference subsequence; The standard deviation of all spectral line growths in all reference subsequences is determined as the marked standard deviation, and a reference difference less than or equal to the marked standard deviation is selected from all reference differences corresponding to the ascending sequence of the marked growth as the target difference; All spectral line growth amounts corresponding to all target differences are selected from the reference subsequence to form a gentle growth amount marker set; The continuous wavelengths in the wavelengths corresponding to the growth amounts of all spectral lines in the gently increasing amount mark set constitute a gently changing interval.

[0008] In combination with the first aspect, in a possible implementation, based on all the gently changing intervals in each Raman spectrum, screening out the possible tailing intervals from each component peak interval in each Raman spectrum includes: Determine each wavelength except for the gently changing interval among the wavelengths corresponding to the growth amounts of all spectral lines in all reference subsequences as an intermediate changing wavelength; All the intermediate changing wavelengths in each component peak interval in the marked spectrum form a possible tailing interval corresponding to each component peak interval in the marked spectrum.

[0009] In combination with the first aspect above, in a possible implementation, determining the structural variation index corresponding to each component peak interval according to the Raman intensity difference and wavelength difference between each component peak interval and its inner tailing possible interval includes: According to the maximum value of the Raman intensity corresponding to all wavelengths in each component peak interval and the average value of the Raman intensity corresponding to all wavelengths in the possible tailing interval, the Raman intensity difference corresponding to each component peak interval is determined; According to the number of wavelengths in each component peak interval and the possible tailing interval therein, the wavelength difference corresponding to each component peak interval is determined; According to the Raman intensity difference and wavelength difference corresponding to each component peak interval, the structural variation index corresponding to each component peak interval is determined.

[0010] In combination with the first aspect above, in a possible implementation, determining the wavelength difference corresponding to each component peak interval according to the number of wavelengths in each component peak interval and the possible tailing interval therein includes: The wavelength difference corresponding to each component peak interval is determined according to the difference between the number of wavelengths in each component peak interval and the number of wavelengths in its inner tail possible interval.

[0011] In combination with the first aspect above, in a possible implementation, the formula corresponding to the structural variation index corresponding to the component peak interval is: ;in, is the structural variation index corresponding to the jth component peak interval in the i-th Raman spectrum; i is the sequence number of the Raman spectrum; j is the sequence number of the component peak interval in the i-th Raman spectrum; It is the mean value of the Raman intensity corresponding to all wavelengths in the possible tailing interval within the jth component peak interval in the i-th Raman spectrum; is the maximum value of the Raman intensity corresponding to all wavelengths in the jth component peak interval of the i-th Raman spectrum; is the number of wavelengths in the jth component peak interval in the ith Raman spectrum; is the number of wavelengths in the possible tailing interval within the jth component peak interval in the i-th Raman spectrum; is the total number of all wavelengths in the i-th Raman spectrum; is the Raman intensity difference corresponding to the jth component peak interval in the i-th Raman spectrum; is the wavelength difference corresponding to the jth component peak interval in the ith Raman spectrum.

[0012] In combination with the first aspect above, in a possible implementation, the formula corresponding to the homogeneity index corresponding to the Raman spectrum is: ;in, is the homogeneity index corresponding to the i-th Raman spectrum; i is the sequence number of the Raman spectrum; is the number of component peak intervals in the ith Raman spectrum; j is the sequence number of the component peak interval in the ith Raman spectrum; It is the absolute value function; It is the structural variation index corresponding to the jth component peak interval in the i-th Raman spectrum; is the mean value of the structural variation index corresponding to all component peak intervals in the i-th Raman spectrum; It is the standard deviation of the structural variation index corresponding to all component peak intervals in the i-th Raman spectrum.

[0013] In combination with the first aspect, in a possible implementation, constructing a target matrix according to all structural variation indicators and the lengths of all gently varying intervals includes: According to the positions of the gently changing intervals and the component peak intervals in the Raman spectrum, the gently changing intervals and the component peak intervals in each Raman spectrum are sorted to obtain an interval sequence corresponding to each Raman spectrum; Replace the gently changing interval in each interval sequence with the length of the gently changing interval, and replace the component peak interval in each interval sequence with the structural variation index corresponding to the component peak interval, to obtain the numerical vector corresponding to each interval sequence; All numeric vectors are combined to form a target matrix, where the numeric vectors are the rows of the target matrix.

[0014] In combination with the first aspect above, in a possible implementation, the formula corresponding to the target window size is: ; Where D is the length of the side that constitutes the target window size; is rounded down; It is the maximum value of the homogeneity index corresponding to all Raman spectra; It is the minimum value of the homogeneity index corresponding to all Raman spectra.

[0015] In a second aspect, the present invention provides a high-sensitivity spectral detection system for glycated hemoglobin, the system comprising: A data acquisition and determination module is used to obtain Raman spectra corresponding to a preset number of test samples of the patient to be tested, and determine an ascending sequence of spectral line growth amounts corresponding to each Raman spectra; The Raman spectrum wavelength division module is used to screen out the smooth change interval and component peak interval from all wavelengths of each Raman spectrum according to the ascending sequence of the spectral line growth amount corresponding to each Raman spectrum; An interval screening module is used to screen out possible tailing intervals from each component peak interval in each Raman spectrum based on all the gently changing intervals in each Raman spectrum; A structural variation index determination module is used to determine the structural variation index corresponding to each component peak interval according to the Raman intensity difference and wavelength difference between each component peak interval and the possible tailing interval within it; A homogeneity index determination module is used to determine the homogeneity index corresponding to each Raman spectrum according to the structural variation index corresponding to all component peak intervals in each Raman spectrum; A matrix construction size determination module is used to construct a target matrix based on all structural variation indicators and the lengths of all gently varying intervals, and to determine the target window size based on all homogeneity indicators; The SERS detection module is used to replace the matrix in SERS detection with the target matrix, and use the target window size as the window size in SERS detection, so as to detect the glycated hemoglobin of the patient to be detected through SERS detection.

[0016] In a third aspect, a server is provided, comprising a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, so that the device executes the method in the first aspect or any possible implementation of the first aspect.

[0017] In a fourth aspect, a computer program product is provided, comprising: a computer program code, which, when executed on a computer, enables the computer to execute the method in the first aspect or any possible implementation of the first aspect.

[0018] In a fifth aspect, a computer-readable storage medium is provided, which stores a computer program code. When the computer program code runs on a computer, the computer executes the method in the above-mentioned first aspect or any possible implementation manner of the first aspect.

[0019] The present invention has the following beneficial effects: A spectral high-sensitivity detection method for glycated hemoglobin of the present invention realizes SERS detection of glycated hemoglobin, solves the technical problem of poor accuracy of glycated hemoglobin detection due to poor rationality of window size setting, improves the rationality of window size setting, and thus improves the accuracy of glycated hemoglobin detection. In the process of realizing glycated hemoglobin detection through SERS detection technology, compared with setting the window size in SERS detection through manual experience, the present invention comprehensively considers multiple features related to window size setting, such as tailing possible interval, structural variation index and homogeneity index, thereby relatively objectively quantifying the window size in SERS detection to a certain extent, improving the rationality of window size setting, and thus improving the accuracy of glycated hemoglobin detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 This is a flow chart of a method for high-sensitivity spectral detection of glycated hemoglobin according to the present invention; Figure 2 It is a schematic diagram of the composition structure of a spectral high-sensitivity detection system for glycated hemoglobin of the present invention; Figure 3 The figure is a schematic diagram of the structure of a computer device of the present invention. DETAILED DESCRIPTION

[0022] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the technical solutions proposed by the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0023] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0024] Raman spectroscopy is a scattering spectroscopy technique that obtains structural information about sample molecules by analyzing the scattered light generated after the sample molecules interact with photons. When a beam of monochromatic light is irradiated onto a sample, most of the photons are elastically scattered (Rayleigh scattering), and the frequency of the scattered light is the same as the incident light. However, a small portion of photons tend to be inelastically scattered (Raman scattering), and the frequency of the scattered light is different from the incident light. This change in frequency is called the Raman shift, which corresponds to the energy of a specific vibrational mode in the sample molecule. Therefore, by analyzing the Raman scattering spectrum, the different molecules present in the sample and their structural characteristics can be identified and quantified.

[0025] Glycated hemoglobin (HbA1c) refers to the product of hemoglobin combined with glucose. The value of HbA1c reflects the patient's average blood sugar level in the past 2-3 months. This is because it is associated with the lifespan of red blood cells (about 120 days). The higher the blood sugar, the more glucose is bound to hemoglobin, and the higher the HbA1c value.

[0026] Raman spectroscopy can be used to analyze glycated hemoglobin (HbA1c) because: First, HbA1c has unique Raman spectral characteristics: the molecular structures of glycated hemoglobin and non-glycated hemoglobin are different, and this difference is reflected in their Raman spectra. The glycation process changes certain chemical bonds and groups of hemoglobin, resulting in new peaks or changes in peak intensity in its Raman spectrum.

[0027] Second, Raman spectroscopy is sensitive to glycosylation sites: Raman spectroscopy can detect glycosylation modification on the N-terminal valine residue of the hemoglobin β chain, which is a key site for HbA1c formation.

[0028] Third, label-free detection: Raman spectroscopy is a label-free detection technology that can directly analyze samples without adding any dyes or markers, which simplifies the analysis process and avoids possible interference caused by markers.

[0029] Fourth, fast and non-destructive testing: Raman spectroscopy is fast and usually takes only a few seconds to a few minutes to complete. In addition, it is a non-destructive testing technology that can preserve the integrity of the sample for subsequent analysis.

[0030] In summary, the testing principle for glycated hemoglobin: the existing method for testing its spectral peak position will deviate from the result of spectral testing as the protein structure torsional deforms, and glycated hemoglobin is normal hemoglobin immersed in a high concentration of blood glucose environment for a long time, and hemoglobin is prone to abnormal metabolism in a peroxidized blood environment full of free radicals, that is, the Fe-O (ferrous ion-oxygen ion) chemical bond in hemoglobin is affected, resulting in structural variation of glycated hemoglobin, which leads to corresponding changes in the Raman intensity at the corresponding wavelength position of the spectrum.

[0031] refer to Figure 1 , shows the process of some embodiments of a method for high-sensitivity spectrum detection of glycated hemoglobin according to the present invention. The method for high-sensitivity spectrum detection of glycated hemoglobin comprises the following steps: Step S1, obtaining Raman spectra corresponding to a preset number of test samples of a patient to be tested, and determining an ascending sequence of spectral line growth amounts corresponding to each Raman spectra.

[0032] The patient to be tested may be a patient to be tested for glycated hemoglobin. The preset number may be a pre-set number, which may be 4. The test sample may be a sample prepared based on the blood of the patient to be tested. The Raman spectrum may be a spectrum obtained by SERS (Surface-Enhanced Raman Spectroscopy) testing, the horizontal axis of which may be the wavelength, and the vertical axis may be the Raman intensity. The ascending sequence of spectral line growth may be a sequence obtained by arranging the spectral line growth in ascending order.

[0033] It should be noted that in the actual spectral test of glycated hemoglobin, other protein substances in the blood, such as HbA0 (non-glycosylated hemoglobin) and HbA1c have similar structures. Therefore, in the results of Raman spectral detection, the strong signal peaks formed by the remaining hemoglobin and some hemoglobin variants are similar to glycated hemoglobin, so the sensitivity of directly distinguishing them using spectral signals is often not high. The prior art uses the comparative attribution method of the hemoglobin SERS spectrum in the peak position difference of the spectrum line, which often has the problem of insufficient deviation discrimination when the hemoglobin is less glycosylated and the normal hemoglobin is more. The present invention enhances the spectrum line, and the component information of glycated hemoglobin changes with its protein conformation, which can facilitate the subsequent separation of the spectrum line components based on the different scattering intensities containing different components, and obtain the component peak intervals characterizing different components, and further enhance the sensitivity of the test of glycated hemoglobin in the sample based on the results of the spectrum line separation and the distinction between the component quantity deviation and the component intensity of the remaining protein variants.

[0034] As an example, this step may include the following steps: In the first step, the method for obtaining each test sample of the patient to be tested may include the following sub-steps: In the first sub-step, medical staff collect 3 ml of blood from the patient's vein or capillaries. The patient fasts for 12 hours before blood collection. Disodium ethylenediaminetetraacetate is added to the sample as an anticoagulant to prevent the sample blood from clotting.

[0035] In the second sub-step, the blood sample is centrifuged, washed, and lysed to obtain a hemoglobin solution.

[0036] The third sub-step is to mix equal volumes of the hemoglobin solution and the pre-made silver colloid solution, drop the mixed solution on a pure aluminum sheet, and wait for it to dry naturally to form a test sample.

[0037] The second step is to collect the Raman spectrum corresponding to each test sample through the confocal Raman spectrometer. The inspection wavelength of the confocal Raman spectrometer can be set to 400-1800 The excitation light source was set to 785 nm near-infrared laser, the objective lens magnification was set to 50 times, the spectral integration time was set to 10 seconds, the number of integrations was set to 2, and the resolution was set to 2 ; The spectrum test results are processed by subtracting the fluorescence background through the existing fifth-order polynomial fitting software to obtain the processed sample test spectrum, which is recorded as the Raman spectrum.

[0038] The third step, determining the ascending sequence of the spectral line growth amount corresponding to each Raman spectrum graph may include the following sub-steps: In the first sub-step, the spectral lines in the Raman spectrum are smoothed by the Whittaker smoothing algorithm. The Whittaker smoothing algorithm can give a lower fitting weight to the window edge points in the self-identified window, which can reduce the influence of the characteristic peaks in the spectral lines on the smoothed data and further suppress the abnormal jitter in the spectral lines. The spectral lines in the Raman spectrum are the spectral line results of the Raman spectrum detection.

[0039] Among them, the Whittaker smoothing algorithm is a classic time series smoothing algorithm, which achieves the smoothing effect by performing local fitting on the data.

[0040] In the second sub-step, the smoothed spectral lines are subjected to first-order difference to extract the spectral line growth amount, and the growth amounts are sorted in ascending order to obtain an ascending sequence of spectral line growth amounts.

[0041] Step S2, according to the ascending sequence of the spectral line growth amount corresponding to each Raman spectrum, the smooth change interval and the component peak interval are screened out from all the wavelengths of each Raman spectrum.

[0042] As an example, this step may include the following steps: In the first step, the absolute value of the difference between each two adjacent spectral line growth amounts in the ascending sequence of each spectral line growth amount is determined as the reference difference.

[0043] The second step is to select the smoothly changing interval from all wavelengths of each Raman spectrum according to all reference differences corresponding to the ascending sequence of the spectral line growth amount corresponding to each Raman spectrum, which may include the following sub-steps: In the first sub-step, any Raman spectrum is determined as a marked spectrum, and the ascending sequence of spectral line growth amounts corresponding to the marked spectrum is determined as the ascending sequence of marked growth amounts.

[0044] In the second sub-step, two spectral line growth amounts corresponding to the maximum values ​​of all reference differences corresponding to the above-mentioned ascending sequence of marked growth amounts are determined as the first temporary growth amount and the second temporary growth amount respectively.

[0045] The third sub-step is to segment the ascending sequence of the marked growth amounts using the first temporary growth amount and the second temporary growth amount as segmentation points to obtain a spectral line growth quantum sequence.

[0046] For example, if the ascending sequence of marked increments is {first increment, second increment, third increment, fourth increment, fifth increment, sixth increment, seventh increment, eighth increment}, and the third increment and the seventh increment are the first temporary increment and the second temporary increment respectively, then three spectral line increment quantum sequences can be obtained, which can be {first increment, second increment, third increment}, {fourth increment, fifth increment, sixth increment, seventh increment} and {eighth increment} respectively.

[0047] The fourth sub-step is to determine the maximum spectral line growth amount in the ascending sequence of the above-mentioned marked growth amounts as the marked maximum growth amount.

[0048] In the fifth sub-step, each spectral line growth quantum sequence in all spectral line growth quantum sequences except the spectral line growth quantum sequence with the maximum growth amount is determined as a reference sub-sequence.

[0049] In the sixth sub-step, the standard deviation of all spectral line growths in all reference sub-sequences is determined as the marked standard deviation, and reference differences less than or equal to the marked standard deviation are selected from all reference differences corresponding to the ascending sequence of the marked growth as the target differences.

[0050] The seventh sub-step is to select all spectral line growth amounts corresponding to all target differences from the reference sub-sequence to form a set of gentle growth amount markers.

[0051] In the eighth sub-step, the continuous wavelengths in the wavelengths corresponding to the growth amounts of all spectral lines in the above-mentioned gently increasing amount mark set constitute a gently changing interval.

[0052] In the third step, the continuous wavelengths of all wavelengths in each Raman spectrum except the gently changing interval constitute the component peak interval.

[0053] For example, if the wavelengths in the marked spectrum are: first wavelength, second wavelength, third wavelength, fourth wavelength, fifth wavelength, sixth wavelength, seventh wavelength, eighth wavelength, ninth wavelength and tenth wavelength respectively; and the wavelengths corresponding to the growth amounts of all spectral lines in the gently increasing amount mark set are: first wavelength, second wavelength, seventh wavelength and eighth wavelength respectively, then two gently changing intervals and two component peak intervals can be obtained. The two gently changing intervals can be: {first wavelength, second wavelength} and {seventh wavelength, eighth wavelength} respectively; the two component peak intervals can be: {third wavelength, fourth wavelength, fifth wavelength, sixth wavelength} and {ninth wavelength, tenth wavelength} respectively.

[0054] Step S3, based on all the gently changing intervals in each Raman spectrum, a possible tailing interval is screened out from each component peak interval in each Raman spectrum.

[0055] It should be noted that the specific components of glycated hemoglobin contained in the comparison sample are mainly concentrated in the positions of different spectral lines. Therefore, it is often necessary to test the peak tailing in the sample. Among them, peak tailing often represents insufficient separation of components in the hemoglobin sample. For example, a trace amount of glycated hemoglobin may be insufficiently separated from a large amount of normal hemoglobin. In addition, when the number of structural variations of glycated hemoglobin is small, glycated hemoglobin and other hemoglobins with similar structures often have a certain overlap in the spectrum.

[0056] As an example, this step may include the following steps: In the first step, each wavelength except for the gently changing interval among the wavelengths corresponding to the growth amounts of all spectral lines in all reference subsequences is determined as an intermediate changing wavelength.

[0057] In the second step, all the intermediate changing wavelengths in each component peak interval in the above-mentioned marked spectrum graph constitute the possible tailing interval corresponding to each component peak interval in the above-mentioned marked spectrum graph.

[0058] Step S4, determining the structural variation index corresponding to each component peak interval according to the Raman intensity difference and wavelength difference between each component peak interval and its inner tailing possible interval.

[0059] As an example, this step may include the following steps: In the first step, the Raman intensity difference corresponding to each component peak interval is determined based on the maximum value of the Raman intensity corresponding to all wavelengths in each component peak interval and the average value of the Raman intensity corresponding to all wavelengths in the possible tailing interval.

[0060] In the second step, the wavelength difference corresponding to each component peak interval is determined according to the number of wavelengths in each component peak interval and the possible tailing interval within it.

[0061] For example, the wavelength difference corresponding to each component peak interval may be determined according to the difference between the number of wavelengths in each component peak interval and the number of wavelengths in its inner tail possible interval.

[0062] The third step is to determine the structural variation index corresponding to each component peak interval based on the Raman intensity difference and wavelength difference corresponding to each component peak interval.

[0063] For example, the formula for determining the structural variation index corresponding to the component peak interval can be: ;in, is the structural variation index corresponding to the jth component peak interval in the ith Raman spectrum. i is the sequence number of the Raman spectrum. j is the sequence number of the component peak interval in the ith Raman spectrum. It is the mean value of the Raman intensity corresponding to all wavelengths in the possible tailing interval within the jth component peak interval in the i-th Raman spectrum. It is the maximum value of the Raman intensity corresponding to all wavelengths in the jth component peak interval in the i-th Raman spectrum. is the number of wavelengths in the jth component peak interval in the ith Raman spectrum. It is the number of wavelengths in the possible tailing interval within the jth component peak interval in the i-th Raman spectrum. is the total number of all wavelengths in the i-th Raman spectrum. is the Raman intensity difference corresponding to the jth component peak interval in the i-th Raman spectrum. is the wavelength difference corresponding to the jth component peak interval in the ith Raman spectrum.

[0064] It should be noted that when When the value is larger, it often indicates that the difference in Raman intensity between the peak value of the jth component peak interval and its possible tailing interval is smaller, which often indicates that the distinction between the peak part and the tailing part of the jth component peak interval is relatively less obvious, which often indicates that there are relatively more components with similar structures to other components in the jth component peak interval. The larger the value is, the greater the proportion of the tailing part in the j-th component peak interval may be, and the more serious the tailing of the j-th component peak interval may be. The larger it is, the worse the purity of the j-th component peak interval is.

[0065] Optionally, if the possible tailing interval within a component peak interval is an empty set, the structural variation index corresponding to the component peak interval can be set to 1.

[0066] Step S5, determining the homogeneity index corresponding to each Raman spectrum according to the structural variation indexes corresponding to all component peak intervals in each Raman spectrum.

[0067] As an example, the formula for determining the homogeneity index corresponding to the Raman spectrum can be: ;in, is the homogeneity index corresponding to the ith Raman spectrum. i is the sequence number of the Raman spectrum. is the number of component peak intervals in the ith Raman spectrum. j is the sequence number of the component peak interval in the ith Raman spectrum. It is the absolute value function. It is the structural variation index corresponding to the jth component peak interval in the i-th Raman spectrum. It is the mean value of the structural variation index corresponding to all component peak intervals in the i-th Raman spectrum. It is the standard deviation of the structural variation index corresponding to all component peak intervals in the i-th Raman spectrum.

[0068] It should be noted that It can represent the overall situation of the components in the test sample corresponding to the i-th Raman spectrum, and the closer its value is to 0, the more similar the structural variations between different components in the test sample corresponding to the i-th Raman spectrum are.

[0069] Step S6, constructing a target matrix according to all structural variation indicators and the lengths of all gently varying intervals, and determining a target window size based on all homogeneity indicators.

[0070] The length of the smoothly changing interval can be represented by the number of wavelengths in the smoothly changing interval. The target window size can represent the square window size.

[0071] As an example, this step may include the following steps: In the first step, the smoothly varying intervals and component peak intervals in each Raman spectrum are sorted according to their positions in the Raman spectrum to obtain an interval sequence corresponding to each Raman spectrum.

[0072] For example, if the wavelengths in the marked spectrum are: first wavelength, second wavelength, third wavelength, fourth wavelength, fifth wavelength, sixth wavelength, seventh wavelength, eighth wavelength, ninth wavelength and tenth wavelength in sequence; and the first gently changing interval is {first wavelength, second wavelength}, and the second gently changing interval is {seventh wavelength, eighth wavelength}; the first component peak interval is {third wavelength, fourth wavelength, fifth wavelength, sixth wavelength}, and the second component peak interval is {ninth wavelength, tenth wavelength}, then the interval sequence corresponding to the marked spectrum can be {first gently changing interval, first component peak interval, second gently changing interval, second component peak interval}.

[0073] In the second step, the smoothly changing intervals in each interval sequence are replaced by the length of the slowly changing intervals, and the component peak intervals in each interval sequence are replaced by the structural variation index corresponding to the component peak intervals, so as to obtain the numerical vector corresponding to each interval sequence.

[0074] The third step is to combine all numerical vectors into a target matrix.

[0075] The numerical vectors can be rows of the target matrix, and if the lengths of the numerical vectors are different, the longest numerical vector can be used as the basis, and other numerical vectors can be supplemented with 0 at the end to make the lengths of all numerical vectors the same. They are used as rows of the target matrix and randomly arranged to generate the target matrix.

[0076] The fourth step is to determine the formula corresponding to the target window size: ; Wherein, D is the side length of the target window size, which can characterize the size of the target window to a certain extent. It is rounded down. It is the maximum value of the homogeneity index corresponding to all Raman spectra. It is the minimum value of the homogeneity index corresponding to all Raman spectra.

[0077] It should be noted that the greater the difference between the homogeneity indicators, the smaller the similarity of the components between different test samples, which often indicates that a larger window should be set. Therefore, D can represent the side length of the final window size.

[0078] Step S7, replacing the matrix in SERS detection with the target matrix, and using the target window size as the window size in SERS detection, to detect the glycated hemoglobin of the patient to be detected through SERS detection.

[0079] As an example, replacing the matrix in SERS (Surface-Enhanced Raman Scattering) detection with the target matrix, and using the target window size as the window size in SERS detection, detecting the glycated hemoglobin of the patient to be detected through SERS detection may include the following steps: In the first step, a discrete Fourier transform (DFT) is performed on the target matrix, and after centering the spectrum, a window with a size equal to the target window size and a center point equal to the center point of the spectrum (if the side length is even, the lower right data point of the four central data points is used as the center point) is set as the component variation sensitive window.

[0080] In the second step, elements greater than the target mean are screened out from the component variation sensitive window, and the positions of these screened elements with significant component variation structural interference are marked. The element values ​​are retained at the marked element positions, and the unmarked positions are set to zero. The window is embedded back into the spectrum graph, and an inverse discrete Fourier transform (IDFT) is performed on the spectrum graph to obtain an inverse transform array.

[0081] The target mean can be the mean of all elements in the component variation sensitive window.

[0082] The third step is to overlap the target matrix and the inverse transformation array, and perform bit-by-bit subtraction to obtain the deviation element position. The test sample numbers and component peak intervals corresponding to all non-independent deviation element positions on the array are extracted and marked in the spectral lines of the Raman spectrum. This achieves the positioning of glycated hemoglobin, which can facilitate subsequent analysis by doctors of the specific variant structure of the detected glycated hemoglobin. The non-independent deviation element position is also called a non-single-point position.

[0083] refer to Figure 2 Based on the same inventive concept as the above method embodiment, the present invention provides a spectral high-sensitivity detection system for glycated hemoglobin, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the above computer program is executed by the processor, the steps of implementing a spectral high-sensitivity detection method for glycated hemoglobin may specifically include: The data acquisition and determination module 201 is used to obtain Raman spectra corresponding to a preset number of test samples of the patient to be tested, and determine the ascending sequence of the spectral line growth amount corresponding to each Raman spectra; The Raman spectrum wavelength division module 202 is used to screen out the smooth change interval and the component peak interval from all the wavelengths of each Raman spectrum according to the ascending sequence of the spectral line growth amount corresponding to each Raman spectrum; The interval screening module 203 is used to screen out possible tailing intervals from each component peak interval in each Raman spectrum based on all the gently changing intervals in each Raman spectrum; A structural variation index determination module 204 is used to determine the structural variation index corresponding to each component peak interval according to the Raman intensity difference and wavelength difference between each component peak interval and its inner tailing possible interval; A homogeneity index determination module 205 is used to determine the homogeneity index corresponding to each Raman spectrum according to the structural variation index corresponding to all component peak intervals in each Raman spectrum; A matrix construction size determination module 206 is used to construct a target matrix according to all structural variation indicators and the lengths of all gently varying intervals, and determine a target window size based on all homogeneity indicators; The SERS detection module 207 is used to replace the matrix in SERS detection with the target matrix, and use the target window size as the window size in SERS detection, so as to detect the glycated hemoglobin of the patient to be detected through SERS detection.

[0084] Figure 3 is a schematic diagram of the structure of a computer device provided by an embodiment of the present invention. Figure 3 As shown, the computer device includes: a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302, wherein when the processor 302 executes the computer program 303, the computer device can execute any one of the spectral high-sensitivity detection methods for glycated hemoglobin introduced above.

[0085] Based on the same inventive concept as the above method embodiment, the present invention provides a server, including a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, so that the device executes any one of the above-mentioned spectral high-sensitivity detection methods for glycated hemoglobin.

[0086] Based on the same inventive concept as the above method embodiments, the present invention provides a computer program product, which includes: computer program code, when the computer program code runs on a computer, the computer executes any one of the above-mentioned spectral high-sensitivity detection methods for glycated hemoglobin.

[0087] Based on the same inventive concept as the above-mentioned method embodiments, the present invention provides a computer-readable storage medium, which stores a computer program code. When the computer program code runs on a computer, the computer executes any one of the above-mentioned spectral high-sensitivity detection methods for glycated hemoglobin.

[0088] In summary, in the process of realizing glycated hemoglobin detection through SERS detection technology, compared with setting the window size in SERS detection through manual experience, the present invention comprehensively considers multiple characteristics related to window size setting, such as the possible tailing interval, structural variation index and homogeneity index, thereby relatively objectively quantifying the window size in SERS detection to a certain extent, improving the rationality of window size setting, and thus improving the accuracy of glycated hemoglobin detection.

[0089] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features can be replaced by equivalents. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for high-sensitivity spectral detection of glycated hemoglobin, characterized in that: The following steps are involved: Obtaining Raman spectra corresponding to a preset number of test samples of the patient to be tested, and determining an ascending sequence of spectral line growth amounts corresponding to each Raman spectra; According to the ascending sequence of the spectral line growth amount corresponding to each Raman spectrum, the smooth change interval and the component peak interval are screened out from all the wavelengths of each Raman spectrum; Based on all the gently changing intervals in each Raman spectrum, possible tailing intervals are screened out from each component peak interval in each Raman spectrum; According to the Raman intensity difference and wavelength difference between each component peak interval and its possible tailing interval, the structural variation index corresponding to each component peak interval is determined; According to the structural variation indexes corresponding to all component peak intervals in each Raman spectrum, the homogeneity index corresponding to each Raman spectrum is determined; According to all structural variation indicators and the lengths of all gently changing intervals, a target matrix is ​​constructed, and based on all homogeneity indicators, the target window size is determined; The matrix in SERS detection is replaced by the target matrix, and the target window size is used as the window size in SERS detection. Through SERS detection, the glycated hemoglobin of the patient to be detected is detected.

2. The method for high-sensitivity spectral detection of glycated hemoglobin according to claim 1, characterized in that: The method of selecting the smooth change interval and component peak interval from all wavelengths of each Raman spectrum according to the ascending sequence of the spectral line growth amount corresponding to each Raman spectrum includes: The absolute value of the difference between each two adjacent spectral line growth amounts in the ascending sequence of each spectral line growth amount is determined as a reference difference; According to all reference differences corresponding to the ascending sequence of the spectral line growth amount corresponding to each Raman spectrum, a smoothly changing interval is screened out from all wavelengths of each Raman spectrum; The continuous wavelengths of all wavelengths in each Raman spectrum except for the gently changing interval constitute the component peak interval.

3. A method for high-sensitivity spectral detection of glycated hemoglobin according to claim 2, characterized in that: The method of selecting a smoothly changing interval from all wavelengths of each Raman spectrum graph according to all reference differences corresponding to the ascending sequence of the spectral line growth amount corresponding to each Raman spectrum graph comprises: Determine any Raman spectrum as a marked spectrum, and determine the ascending sequence of spectral line growth amounts corresponding to the marked spectrum as the ascending sequence of marked growth amounts; Determine two spectral line growth amounts corresponding to the maximum values ​​of all reference differences corresponding to the ascending sequence of the mark growth amounts as the first temporary growth amount and the second temporary growth amount respectively; Taking the first temporary growth amount and the second temporary growth amount as the segmentation points, segmenting the ascending sequence of the label growth amount to obtain a spectral line growth quantum sequence; Determine the maximum spectral line growth amount in the ascending sequence of the marked growth amounts as the marked maximum growth amount; Determine each spectral line growth quantum sequence in all spectral line growth quantum sequences except the spectral line growth quantum sequence to which the maximum growth amount is marked as a reference subsequence; The standard deviation of all spectral line growths in all reference subsequences is determined as the marked standard deviation, and a reference difference less than or equal to the marked standard deviation is selected from all reference differences corresponding to the ascending sequence of the marked growth as the target difference; All spectral line growth amounts corresponding to all target differences are selected from the reference subsequence to form a gentle growth amount marker set; The continuous wavelengths among the wavelengths corresponding to the growth amounts of all spectral lines in the gently increasing amount mark set constitute a gently changing interval.

4. The method for high-sensitivity spectral detection of glycated hemoglobin according to claim 3, characterized in that: Based on all the gently changing intervals in each Raman spectrum, possible tailing intervals are screened out from each component peak interval in each Raman spectrum, including: Determine each wavelength except for the gently changing interval among the wavelengths corresponding to the growth amounts of all spectral lines in all reference subsequences as an intermediate changing wavelength; All the intermediate changing wavelengths in each component peak interval in the marked spectrum form the possible tailing interval corresponding to each component peak interval in the marked spectrum.

5. The method for high-sensitivity spectral detection of glycated hemoglobin according to claim 1, characterized in that: Determining the structural variation index corresponding to each component peak interval according to the Raman intensity difference and wavelength difference between each component peak interval and its inner tailing possible interval includes: According to the maximum value of the Raman intensity corresponding to all wavelengths in each component peak interval and the average value of the Raman intensity corresponding to all wavelengths in the possible tailing interval, the Raman intensity difference corresponding to each component peak interval is determined; According to the number of wavelengths in each component peak interval and the possible tailing interval therein, the wavelength difference corresponding to each component peak interval is determined; According to the Raman intensity difference and wavelength difference corresponding to each component peak interval, the structural variation index corresponding to each component peak interval is determined.

6. A method for high-sensitivity spectral detection of glycated hemoglobin according to claim 5, characterized in that: Determining the wavelength difference corresponding to each component peak interval according to the number of wavelengths in each component peak interval and the possible tailing interval therein includes: The wavelength difference corresponding to each component peak interval is determined according to the difference between the number of wavelengths in each component peak interval and the number of wavelengths in its inner tail possible interval.

7. A method for high-sensitivity spectral detection of glycated hemoglobin according to claim 6, characterized in that: The formula corresponding to the structural variation index corresponding to the component peak interval is: ;in, is the structural variation index corresponding to the jth component peak interval in the i-th Raman spectrum; i is the sequence number of the Raman spectrum; j is the sequence number of the component peak interval in the i-th Raman spectrum; It is the mean value of the Raman intensity corresponding to all wavelengths in the possible tailing interval within the jth component peak interval in the i-th Raman spectrum; is the maximum value of the Raman intensity corresponding to all wavelengths in the jth component peak interval of the i-th Raman spectrum; is the number of wavelengths in the jth component peak interval in the ith Raman spectrum; is the number of wavelengths in the possible tailing interval within the jth component peak interval in the i-th Raman spectrum; is the total number of all wavelengths in the i-th Raman spectrum; is the Raman intensity difference corresponding to the jth component peak interval in the i-th Raman spectrum; is the wavelength difference corresponding to the jth component peak interval in the ith Raman spectrum.

8. The method for high-sensitivity spectral detection of glycated hemoglobin according to claim 1, characterized in that: The formula corresponding to the homogeneity index of the Raman spectrum is: ;in, is the homogeneity index corresponding to the i-th Raman spectrum; i is the sequence number of the Raman spectrum; is the number of component peak intervals in the ith Raman spectrum; j is the sequence number of the component peak interval in the ith Raman spectrum; It is the absolute value function; It is the structural variation index corresponding to the jth component peak interval in the i-th Raman spectrum; is the mean value of the structural variation index corresponding to all component peak intervals in the i-th Raman spectrum; It is the standard deviation of the structural variation index corresponding to all component peak intervals in the i-th Raman spectrum.

9. The method for high-sensitivity spectral detection of glycated hemoglobin according to claim 1, characterized in that: The target matrix is ​​constructed according to all structural variation indicators and the lengths of all gently varying intervals, including: According to the positions of the gently changing intervals and the component peak intervals in the Raman spectrum, the gently changing intervals and the component peak intervals in each Raman spectrum are sorted to obtain an interval sequence corresponding to each Raman spectrum; Replace the gently changing interval in each interval sequence with the length of the gently changing interval, and replace the component peak interval in each interval sequence with the structural variation index corresponding to the component peak interval, to obtain the numerical vector corresponding to each interval sequence; All numeric vectors are combined to form a target matrix, where the numeric vectors are the rows of the target matrix.

10. The method for high-sensitivity spectral detection of glycated hemoglobin according to claim 1, characterized in that: The formula for the target window size is: ; Where D is the length of the side that constitutes the target window size; is rounded down; It is the maximum value of the homogeneity index corresponding to all Raman spectra; It is the minimum value of the homogeneity index corresponding to all Raman spectra.

Images