Detection device and detection method for endogenous metabolites of metabolic disorder patients

Abstract

The invention relates to the field of metabolite detection, in particular to an endogenous metabolite detection device and method for metabolic disorder patients. The method comprises the following steps: segmenting a Raman spectrum curve of an endogenous metabolite of a patient to obtain a plurality of fluctuation segments and a reference baseline value of each fluctuation segment, screening out a normal fluctuation segment, and combining the reference baseline value of the normal fluctuation segment to obtain a normal baseline value of the Raman spectrum curve; adjusting each fluctuation section of the Raman spectrum curve according to the difference between the reference baseline value and the normal baseline value of each fluctuation section to obtain an adjusted spectrum curve, denoising the adjusted spectrum curve to obtain a final spectrum curve, and detecting the endogenous metabolites of the patient. According to the invention, the baseline drift phenomenon of the Raman spectrum curve of the endogenous metabolite of the patient can be eliminated, and the Raman spectrum data quality and the detection accuracy of the endogenous metabolite are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of metabolite detection, in particular to a device and method for detecting endogenous metabolites of patients with metabolic disorders. BACKGROUND

[0002] Metabolic disorders refer to a class of diseases in which one or more metabolites in the human body are abnormal in synthesis, decomposition, transport or storage, etc., leading to body function disorders. Endogenous metabolites refer to chemical substances produced naturally in the human body and involved in the process of metabolism, such as amino acids, fatty acids, sugars, nucleotides, etc. Metabolic disorders in the human body are often accompanied by changes in endogenous metabolites. Therefore, the detection of endogenous metabolites in the human body can be an important basis for evaluating the metabolic state of patients.

[0003] In related technologies, surface-enhanced Raman spectroscopy is usually used to detect endogenous metabolites of patients. However, when the concentration of metabolites of patients is low, competitive adsorption may occur between various metabolites, such as lactic acid and pyruvic acid, resulting in baseline drift of the collected Raman spectrum, poor quality of Raman spectrum data, and low accuracy of endogenous metabolite detection. SUMMARY

[0004] In order to solve the technical problem that the collected Raman spectrum has baseline drift phenomenon, resulting in low accuracy of endogenous metabolite detection, the purpose of the present application is to provide a device and method for detecting endogenous metabolites of patients with metabolic disorders, and the technical solution adopted is as follows: The present application provides a method for detecting endogenous metabolites of patients with metabolic disorders, which comprises: obtaining a Raman spectrum curve of endogenous metabolites of a patient, the Raman spectrum curve including Raman intensity at different Raman shifts; segmenting the Raman spectrum curve according to fluctuations of the Raman spectrum curve to obtain a plurality of fluctuation segments, Raman intensity span value, Raman shift span value and reference baseline value of each fluctuation segment; analyzing the consistency of the change direction between the Raman intensity and the reference baseline value of each adjacent fluctuation segment to screen out normal fluctuation segments; obtaining a normal baseline value of the Raman spectrum curve according to the Raman intensity span value and the Raman shift span value of each normal fluctuation segment, and combining the reference baseline value of each normal fluctuation segment; obtaining an adjustment degree of each fluctuation segment according to the difference between the reference baseline value and the normal baseline value of each fluctuation segment, and the Raman intensity span value and the Raman shift span value of each fluctuation segment; adjusting each fluctuation segment of the Raman spectrum curve based on the adjustment degree to obtain an adjusted spectrum curve; denoising the adjusted spectrum curve to obtain a final spectrum curve; detecting the endogenous metabolite of the patient based on the final spectrum curve.

[0005] Further, the obtaining the plurality of fluctuation segments and the Raman intensity span value, the Raman shift span value and the reference baseline value of each fluctuation segment comprises: obtaining extreme points on the Raman spectrum curve, the extreme points comprising maximum points and minimum points; regarding a part of the Raman spectrum curve between two adjacent minimum points as a fluctuation segment; regarding the range of all the Raman intensities in each fluctuation segment as the Raman intensity span value of each fluctuation segment, and regarding the range of all the Raman shifts in each fluctuation segment as the Raman shift span value of each fluctuation segment; obtaining the Raman shift range of each fluctuation segment, wherein the lower limit of the Raman shift range of each fluctuation segment is the minimum Raman shift of each fluctuation segment, and the upper limit is the maximum Raman intensity of each fluctuation segment; regarding the average of all the Raman intensities of the baseline within the Raman shift range of each fluctuation segment as the reference baseline value of each fluctuation segment.

[0006] Further, the screening out normal fluctuation segments comprises: regarding the average of all the Raman intensities in each fluctuation segment as the overall Raman intensity of each fluctuation segment; regarding any one fluctuation segment as a target fluctuation segment, regarding the difference of the overall Raman intensity between the target fluctuation segment and each other fluctuation segment in the preset window except the target fluctuation segment as the overall Raman intensity change between the target fluctuation segment and each other fluctuation segment, and regarding the difference of the reference baseline value between the target fluctuation segment and each other fluctuation segment in the preset window except the target fluctuation segment as the reference baseline value change between the target fluctuation segment and each other fluctuation segment; regarding the product value of the overall Raman intensity change and the reference baseline value change as the baseline drift performance between the target fluctuation segment and each other fluctuation segment; if the baseline drift performance between the target fluctuation segment and all the other fluctuation segments in the preset window except the target fluctuation segment is not greater than the numerical value 0, regarding the target fluctuation segment as a normal fluctuation segment.

[0007] Further, the obtaining the normal baseline value of the Raman spectrum curve comprises: comprehensively obtaining the maximum value of the Raman intensity, the Raman intensity span value and the Raman shift span value of each normal fluctuation segment to obtain the reference factor of each normal fluctuation segment; the reference factor of each normal fluctuation section as a numerator, the cumulative value of the reference factors of all normal fluctuation sections as a denominator, and the ratio as a baseline weight of each normal fluctuation section; performing weighted summation on the reference baseline values of each normal fluctuation section by using the baseline weights of the normal fluctuation sections to obtain a normal baseline value of the Raman spectrum curve.

[0008] Further, the adjustment degree of each fluctuation section comprises: the absolute value of the difference between the reference baseline value and the normal baseline value of each fluctuation section as a first adjustment parameter of each fluctuation section; performing comprehensive processing on the Raman intensity span value and the Raman frequency shift span value of each fluctuation section and performing negative correlation mapping to obtain a second adjustment parameter of each fluctuation section; performing comprehensive processing on the first adjustment parameter and the second adjustment parameter and performing normalization processing to obtain the adjustment degree of each fluctuation section.

[0009] Further, the obtaining of the adjusted spectrum curve comprises: taking any one fluctuation section of the Raman spectrum curve as a to-be-adjusted fluctuation section; the product value of the adjustment degree of the to-be-adjusted fluctuation section and the normal baseline value of the Raman spectrum curve as a Raman intensity adjustment amount of the to-be-adjusted fluctuation section; the difference between the Raman intensity and the Raman intensity adjustment amount of each Raman frequency shift of the to-be-adjusted fluctuation section as an adjusted Raman intensity of each Raman frequency shift of the to-be-adjusted fluctuation section; replacing the Raman intensity of each Raman frequency shift of each fluctuation section of the Raman spectrum curve with the adjusted Raman intensity to obtain an adjusted spectrum curve.

[0010] Further, the obtaining of the final spectrum curve comprises: obtaining a wavelet decomposition layer number of the adjusted spectrum curve according to the adjustment degree, the Raman intensity span value and the Raman frequency shift span value of all fluctuation sections of the Raman spectrum curve; performing wavelet decomposition on the adjusted spectrum curve based on the wavelet decomposition layer number to obtain low-frequency coefficients and high-frequency coefficients, removing the high-frequency coefficients, retaining the low-frequency coefficients, and performing inverse wavelet decomposition on the retained low-frequency coefficients to reconstruct a final spectrum curve.

[0011] Further, the obtaining of the wavelet decomposition layer number of the adjusted spectrum curve comprises: taking the average value of the adjustment degrees of all fluctuation sections of the Raman spectrum curve as a first complexity factor of the Raman spectrum curve; The product value of the Raman intensity span value and the Raman frequency shift span value of each fluctuation section of the Raman spectrum curve is taken as a characteristic value of each fluctuation section, the dispersion degree of the characteristic values of all the fluctuation sections of the Raman spectrum curve is analyzed, and a second complexity factor of the Raman spectrum curve is obtained; The first complexity factor and the second complexity factor are comprehensively processed and normalized to obtain a data representation complexity of the Raman spectrum curve; The wavelet decomposition layer number of the adjusted spectrum curve is obtained based on a calculation formula of the wavelet decomposition layer number, and the calculation formula of the wavelet decomposition layer number is as follows:

[0012] Wherein, V represents the wavelet decomposition layer number of the adjusted spectrum curve; U represents the data representation complexity of the Raman spectrum curve; The preset parameter is represented by a, and the value range is .

[0013] Further, the detection of the endogenous metabolites of the patient comprises: The characteristic peak of the to-be-detected substance is extracted on the final spectrum curve, and the content of the to-be-detected substance is detected based on the characteristic peak of the to-be-detected substance and in combination with a quantitative analysis model.

[0014] The application further provides an endogenous metabolite detection device for patients with metabolic disorders, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of any one of the endogenous metabolite detection methods for patients with metabolic disorders when executing the computer program.

[0015] The application has the following beneficial effects: The application considers that the collected Raman spectrum has a baseline drift phenomenon, which leads to low accuracy of endogenous metabolite detection, so the Raman spectrum curve of the endogenous metabolites of the patient is first obtained, then a plurality of fluctuation sections are divided from the Raman spectrum curve, it is considered that the change direction of the Raman intensity of the continuous fluctuation section composed of the fluctuation section and the adjacent other fluctuation section is inconsistent with the change direction of the reference baseline value, and accordingly the normal fluctuation section is screened from all the fluctuation sections, and then the reference baseline value of the normal fluctuation section is obtained to represent the overall level of the baseline when the baseline drift does not occur, and the adjustment degree obtained is reflected in the degree of adjustment of each fluctuation section under the condition of eliminating the baseline drift, so as to obtain the adjusted spectrum curve, eliminate the baseline drift phenomenon in the Raman spectrum, and denoise the adjusted spectrum curve, further improve the quality of the Raman spectrum data, and improve the accuracy of the detection of the endogenous metabolites of the patient. Attached Figure Description

[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating a method for detecting endogenous metabolites in patients with metabolic disorders, provided as an embodiment of the present invention; Figure 2 This is a schematic diagram of the fluctuation segment of a Raman spectrum provided in one embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the effect of adjusting the Raman spectral curve to a Raman spectral curve, provided as an embodiment of the present invention. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an endogenous metabolite detection device and method for patients with metabolic disorders proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, 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 pertains.

[0020] The following description, in conjunction with the accompanying drawings, details the specific scheme of the endogenous metabolite detection device and detection method for patients with metabolic disorders provided by the present invention.

[0021] Please see Figure 1 The diagram illustrates a flowchart of a method for detecting endogenous metabolites in patients with metabolic disorders, according to an embodiment of the present invention. The method includes: Step S1: Obtain the Raman spectra of the patient's endogenous metabolites.

[0022] The embodiment of the present application first collects a body fluid sample of a patient with metabolic disorder, such as a urine or blood sample, and then performs centrifugation, filtration, deproteinization and the like on the collected sample to remove impurities and extract various endogenous metabolites such as lactic acid, pyruvic acid, amino acid and sugar, and then prepares a nanomaterial for surface enhanced Raman spectroscopy (SERS) detection, and mixes the processed sample with the prepared nanomaterial, so that various endogenous metabolites are adsorbed on the surface of the nanomaterial, and then a Raman spectrometer is used to collect spectral data of the mixed sample, so as to obtain a Raman spectrum curve, wherein the abscissa of the Raman spectrum curve is a Raman shift, and the ordinate is a Raman intensity.

[0023] It should be noted that before obtaining the Raman spectrum curve of the endogenous metabolite, the embodiment of the present application also needs to collect a baseline, wherein the baseline is the Raman spectrum data of the blank background under the condition of no sample, and the baseline is also composed of Raman intensities of different Raman shifts, and the baseline acquisition process is a technical means familiar to those skilled in the art, which will not be repeated here.

[0024] Step S2: segmenting the Raman spectrum curve according to the fluctuation of the Raman spectrum curve, obtaining a plurality of fluctuation segments and Raman intensity span value, Raman shift span value and reference baseline value of each fluctuation segment; analyzing the consistency of the change direction between the Raman intensity and the reference baseline value of each adjacent fluctuation segment, and screening out normal fluctuation segments; obtaining the normal baseline value of the Raman spectrum curve according to the Raman intensity span value and the Raman shift span value of each normal fluctuation segment, and combining the reference baseline value of each normal fluctuation segment.

[0025] When the concentration of the metabolite of the body fluid sample of the patient is low, various metabolites, such as lactic acid and pyruvic acid, will compete for the binding sites on the surface of the SERS substrate, resulting in the inhibition of the spectral signal of part of the metabolites, the corresponding Raman spectrum signal is weak, which shows that the signal-to-noise ratio is reduced, and the baseline drift phenomenon of the Raman spectrum curve occurs, which reduces the accuracy of the qualitative analysis and quantitative analysis of the endogenous metabolite.

[0026] The degree of baseline drift at different local positions on the collected Raman spectrum curve is different, and the Raman spectrum curve has obvious fluctuation characteristics, therefore, the embodiment of the present application first segments the Raman spectrum curve according to the fluctuation of the Raman spectrum curve, obtains a plurality of fluctuation segments and Raman intensity span value, Raman shift span value and reference baseline value of each fluctuation segment, and then analyzes each fluctuation segment, and adjusts the baseline drift of the fluctuation segment to different degrees in combination with the Raman intensity span value, the Raman shift span value and the reference baseline value of the fluctuation segment, so as to eliminate the baseline drift phenomenon of the Raman spectrum curve.

[0027] Preferably, in one embodiment of the present application, the method for obtaining the plurality of fluctuation segments and the Raman intensity span value, the Raman shift span value and the reference baseline value of each fluctuation segment specifically comprises: Firstly, extreme points on the Raman spectrum curve are obtained, the extreme points including maximum points and minimum points, and the method for obtaining the extreme points can be selected from the existing Newton method or the first derivative method, which is not limited or described herein.

[0028] The part of the Raman spectrum curve between two adjacent minimum points is taken as a fluctuation segment, wherein each fluctuation segment contains two minimum points, i.e. troughs, and one maximum point, i.e. a peak, please refer to Figure 2 which shows the schematic diagram of the fluctuation segment of the Raman spectrum curve provided by one embodiment of the present application.

[0029] Then, the range of all Raman intensities in each fluctuation segment is taken as the Raman intensity span value of each fluctuation segment, and the range of all Raman shifts in each fluctuation segment is taken as the Raman shift span value of each fluctuation segment, wherein the Raman intensity span value can represent the height feature of the fluctuation segment, and the Raman shift span value can represent the width feature of the fluctuation segment.

[0030] The Raman shift range of each fluctuation segment is obtained, wherein the lower limit of the Raman shift range of each fluctuation segment is the minimum Raman shift of each fluctuation segment, and the upper limit is the maximum Raman intensity of each fluctuation segment, that is, the Raman shift range of a certain fluctuation segment is the Raman shift occupied by the fluctuation segment.

[0031] Further, the average value of all Raman intensities of the baseline within the Raman shift range of each fluctuation segment is taken as the reference baseline value of each fluctuation segment.

[0032] The baseline drift appears as a smooth and continuous shift phenomenon on the Raman spectrum curve, and for the fluctuation segments without baseline drift or with small baseline drift degree, the change direction of the Raman intensity of the continuous fluctuation segment composed of the fluctuation segment and the adjacent other fluctuation segment is inconsistent with the change direction of the reference baseline value, thus the consistency between the change direction of the Raman intensity and the reference baseline value of the adjacent fluctuation segments can be analyzed, and the normal fluctuation segments without baseline drift or with small baseline drift degree can be screened from all the fluctuation segments, and subsequently the normal baseline value of the Raman spectrum curve can be obtained in combination with the reference baseline value of each normal fluctuation segment, so as to accurately analyze the adjustment degree of each fluctuation segment, thereby eliminating the baseline drift phenomenon of the Raman spectrum curve.

[0033] Preferably, in one embodiment of the present application, the method for obtaining the normal fluctuation segment specifically comprises: Firstly, the average value of all Raman intensities in each fluctuation section is taken as the overall Raman intensity of each fluctuation section, which reflects the overall level of Raman intensity at all Raman shifts of each fluctuation section.

[0034] Then, taking any one fluctuation section as a target fluctuation section, the difference between the overall Raman intensity of the target fluctuation section and each other fluctuation section in the preset window except the target fluctuation section is taken as the overall Raman intensity change between the target fluctuation section and each other fluctuation section, and the difference between the reference baseline value of the target fluctuation section and each other fluctuation section in the preset window except the target fluctuation section is taken as the reference baseline value change between the target fluctuation section and each other fluctuation section, wherein the length of the preset window is set to 5, that is, the preset window contains the nearest 4 other fluctuation sections to the target fluctuation section and the target fluctuation section itself, and the length of the preset window can also be set by the implementer according to the specific implementation scene, which is not limited here.

[0035] The product value of the overall Raman intensity change and the reference baseline value change is taken as the baseline drift performance between the target fluctuation section and each other fluctuation section. When the baseline drift performance is a positive number, that is, the baseline drift performance is greater than the value 0, it means that the change direction of the overall Raman intensity of the target fluctuation section and the other fluctuation sections in the preset window and the change direction of the reference baseline value are the same, that is, the overall Raman intensity of the target fluctuation section is reduced or increased compared with the overall Raman intensity of a certain other fluctuation section in the preset window, and the reference baseline value of the target fluctuation section is also reduced or increased compared with the reference baseline value of the fluctuation section in the preset window, which further indicates that the target fluctuation section is more likely to have a significant baseline drift phenomenon, and vice versa, which indicates that the target fluctuation section is less likely to have a baseline drift phenomenon.

[0036] When the change direction of the overall Raman intensity of the target fluctuation section and all other fluctuation sections in the preset window and the change direction of the reference baseline value are not consistent, it means that the target fluctuation section does not have a baseline drift phenomenon or has a weak baseline drift phenomenon. Therefore, if the baseline drift performance between the target fluctuation section and all other fluctuation sections in the preset window except the target fluctuation section is not greater than the value 0, the target fluctuation section is taken as a normal fluctuation section, and then a plurality of normal fluctuation sections can be screened from all fluctuation sections by the same method as described above.

[0037] After the normal fluctuation segments are screened out, the overall level of the baseline of the Raman spectrum curve without baseline drift can be analyzed in combination with the reference baseline values of the normal fluctuation segments. Meanwhile, the greater the Raman intensity span value and the Raman frequency shift span value of a normal fluctuation segment are, the greater the reference value of the reference baseline value of the normal fluctuation segment in the analysis process is. Therefore, the normal baseline value of the Raman spectrum curve can be obtained according to the Raman intensity span value and the Raman frequency shift span value of each normal fluctuation segment in combination with the reference baseline value of each normal fluctuation segment. The normal baseline value is used to represent the overall level of the baseline of the Raman spectrum curve without baseline drift. The adjustment degree of each fluctuation segment can be accurately analyzed based on the difference between the reference baseline value and the normal baseline value of each fluctuation segment, so as to effectively eliminate the baseline drift phenomenon of the Raman spectrum curve and improve the quality of the Raman spectrum curve and the accuracy of the detection of endogenous metabolites.

[0038] Preferably, in an embodiment of the present application, the method for obtaining the normal baseline value of the Raman spectrum curve specifically comprises: The maximum value of the Raman intensity, the Raman intensity span value and the Raman frequency shift span value of each normal fluctuation segment are integrated to obtain the reference factor of each normal fluctuation segment. The reference factor of each normal fluctuation segment is the numerator, the cumulative value of the reference factors of all normal fluctuation segments is the denominator, and the ratio is taken as the baseline weight of each normal fluctuation segment. The greater the baseline weight of a normal fluctuation segment is, the greater the reference value of the normal fluctuation segment in the calculation and analysis of the normal baseline value is.

[0039] In the embodiment of the present application, the sum or product value of the maximum value of the Raman intensity, the Raman intensity span value and the Raman frequency shift span value of each normal fluctuation segment can be taken as the reference factor of each normal fluctuation segment, so as to realize the integration of the three, which is not limited herein.

[0040] The reference baseline values of the normal fluctuation segments are weighted and summed by using the baseline weights of the normal fluctuation segments to obtain the normal baseline value of the Raman spectrum curve.

[0041] As an example, in an embodiment of the present application, the expression of the normal baseline value of the Raman spectrum curve can be specifically for example:

[0042] Wherein, B' represents the normal baseline value of the Raman spectrum curve; represents the baseline weight of the nth normal fluctuation segment of the Raman spectrum curve; represents the reference baseline value of the nth normal fluctuation segment; represents the Raman frequency shift span value of the nth normal fluctuation segment; represents the Raman intensity span value of the nth normal fluctuation segment; a maximum value of the Raman intensity of the nth normal fluctuation section; a Raman shift span value of the ith normal fluctuation section; a Raman intensity span value of the ith normal fluctuation section; a maximum value of the Raman intensity of the ith normal fluctuation section, i.e. the Raman intensity corresponding to the position of the peak point; a reference factor of the nth normal fluctuation section; N represents the number of normal fluctuation sections.

[0043] Up to now, the normal baseline value of the Raman spectrum of the endogenous metabolite is obtained.

[0044] Step S3: obtaining an adjustment degree of each fluctuation section according to the difference between the reference baseline value and the normal baseline value of each fluctuation section, and the Raman intensity span value and the Raman shift span value of each fluctuation section; adjusting each fluctuation section of the Raman spectrum curve based on the adjustment degree to obtain an adjusted spectrum curve; and denoising the adjusted spectrum curve to obtain a final spectrum curve.

[0045] After the normal baseline value of the Raman spectrum curve is obtained, each fluctuation section can be adjusted to eliminate the baseline drift phenomenon. Since the degrees of baseline drift of different fluctuation sections are different, different degrees of adjustment are required for different fluctuation sections. The greater the difference between the reference baseline value and the normal baseline value of the fluctuation section, the greater the degree of baseline drift of the fluctuation section. Meanwhile, the smaller the Raman intensity span value and the Raman shift span value of the fluctuation section, the lower the importance of the fluctuation section in the Raman spectrum curve. Therefore, the adjustment degree of each fluctuation section can be obtained according to the difference between the reference baseline value and the normal baseline value of each fluctuation section, and the Raman intensity span value and the Raman shift span value of each fluctuation section. Subsequently, each fluctuation section can be adjusted based on the adjustment degree to eliminate the baseline drift phenomenon of the Raman spectrum curve and improve the accuracy of the detection of the endogenous metabolite of the patient.

[0046] Preferably, in an embodiment of the present application, the method for obtaining the adjustment degree of each fluctuation section specifically comprises: taking the absolute value of the difference between the reference baseline value and the normal baseline value of each fluctuation section as a first adjustment parameter of each fluctuation section.

[0047] comprehending the Raman intensity span value and the Raman shift span value of each fluctuation section and performing negative correlation mapping to obtain a second adjustment parameter of each fluctuation section.

[0048] In the embodiment of the present application, the comprehension of the Raman intensity span value and the Raman shift span value can be realized by calculating the sum value or the product value of the two values, which is not limited herein.

[0049] Further, the first adjustment parameter and the second adjustment parameter are integrated and normalized to limit the calculation result in the range of and thus obtain the adjustment degree of each fluctuation segment.

[0050] In the embodiments of the present application, the integration of the first adjustment parameter and the second adjustment parameter can be achieved by calculating the sum or product value of the two, which is not limited herein.

[0051] In one embodiment of the present application, the normalization process can be specifically, for example, a max-min value normalization process, and the normalization in the subsequent steps can also adopt the max-min value normalization process. In other embodiments of the present application, other normalization methods can be selected according to the specific range of the value, which will not be described herein.

[0052] As an example, in one embodiment of the present application, the expression of the adjustment degree of each fluctuation segment can be specifically, for example:

[0053] wherein, denotes the adjustment degree of the mth fluctuation segment; denotes the reference baseline value of the mth fluctuation segment; denotes the normal baseline value of the Raman spectrum curve; denotes the first adjustment parameter of the mth fluctuation segment; denotes the Raman frequency shift span value of the mth fluctuation segment; denotes the Raman intensity span value of the mth fluctuation segment; denotes the second adjustment parameter of the mth fluctuation segment; denotes a normalization function for normalization.

[0054] After obtaining the adjustment degree of each fluctuation segment, the adjustment of each fluctuation segment of the Raman spectrum curve can be performed based on the adjustment degree, and an adjusted spectrum curve is obtained, so as to eliminate the baseline drift phenomenon in the Raman spectrum curve and improve the data quality of the Raman spectrum.

[0055] Preferably, in one embodiment of the present application, the method for obtaining the adjusted spectrum curve specifically comprises: Any one fluctuation segment of the Raman spectrum curve is taken as a to-be-adjusted fluctuation segment, and the greater the adjustment degree of the to-be-adjusted fluctuation segment, the greater the adjustment amount of the Raman intensity of each Raman frequency shift in the to-be-adjusted fluctuation segment. Therefore, the product value of the adjustment degree of the to-be-adjusted fluctuation segment and the normal baseline value of the Raman spectrum curve can be taken as the Raman intensity adjustment amount of the to-be-adjusted fluctuation segment.

[0056] Since the baseline drift usually causes the peak area or height in the Raman spectrum to be overestimated, the difference between the Raman intensity of each Raman shift in the to-be-adjusted fluctuation section and the Raman intensity adjustment amount can be used as the adjusted Raman intensity of each Raman shift in the to-be-adjusted fluctuation section.

[0057] The adjusted Raman intensity of each Raman shift in each fluctuation section can be obtained by the same method, and then the Raman intensity of each Raman shift in each fluctuation section of the Raman spectrum curve is replaced by the adjusted Raman intensity to obtain an adjusted spectrum curve. Please refer to Figure 3 which shows the effect diagram from the Raman spectrum curve to the adjusted spectrum curve provided by an embodiment of the present application.

[0058] As an example, in an embodiment of the present application, the expression of the adjusted Raman intensity of each Raman shift in the to-be-adjusted fluctuation section can be specifically as follows:

[0059] wherein, represents the adjusted Raman intensity of the rth Raman shift in the to-be-adjusted fluctuation section; represents the Raman intensity of the rth Raman shift in the to-be-adjusted fluctuation section; E represents the adjustment degree of the to-be-adjusted fluctuation section; represents the normal baseline value of the Raman spectrum curve.

[0060] In order to further improve the data quality of the Raman spectrum of the endogenous metabolite of the patient, the adjusted spectrum curve is denoised to obtain a final spectrum curve in an embodiment of the present application, so as to enhance the signal-to-noise ratio of the Raman spectrum and improve the accuracy of the subsequent detection of the endogenous metabolite. At the same time, since the above process eliminates the baseline drift phenomenon of the Raman spectrum, the effect of denoising the adjusted spectrum curve is better.

[0061] Preferably, in an embodiment of the present application, the method for obtaining the final spectrum curve specifically comprises: Wavelet decomposition is a common method for denoising spectrum data, but too few layers of wavelet decomposition will not completely denoise, and too many layers of wavelet decomposition will lose details. Therefore, the number of wavelet decomposition layers of the adjusted spectrum curve is obtained according to the adjustment degree, the Raman intensity span value and the Raman shift span value of all the fluctuation sections of the Raman spectrum curve.

[0062] Preferably, in an embodiment of the present application, the method for obtaining the number of wavelet decomposition layers of the adjusted spectrum curve specifically comprises: The greater the overall level of the adjustment degree of each fluctuation section of the Raman spectrum curve, the more obvious the baseline drift of the Raman spectrum curve, and the higher the complexity of the Raman spectrum curve. Therefore, the average value of the adjustment degree of all the fluctuation sections of the Raman spectrum curve can be used as the first complexity factor of the Raman spectrum curve. The product value of the Raman intensity span value and the Raman frequency shift span value of each fluctuation section of the Raman spectrum curve is taken as a characteristic value of each fluctuation section, the greater the dispersion degree of the characteristic values of all the fluctuation sections, the higher the complexity of the Raman spectrum curve, and thus the dispersion degree of the characteristic values of all the fluctuation sections of the Raman spectrum curve can be analyzed to obtain a second complexity factor of the Raman spectrum curve.

[0063] In an embodiment of the present application, the standard deviation or variance of the characteristic values of all the fluctuation sections of the Raman spectrum curve is taken as the second complexity factor of the Raman spectrum curve, so as to realize the analysis of the dispersion degree of the characteristic values of all the fluctuation sections of the Raman spectrum curve, which is not limited herein.

[0064] Further, the first complexity factor and the second complexity factor are comprehensively processed and normalized, and the calculation result is limited in a range to obtain the data representation complexity of the Raman spectrum curve, and the greater the data representation complexity of the Raman spectrum curve, the more layers of the adjusted spectrum curve need to be decomposed to effectively remove the noise in the Raman spectrum.

[0065] In an embodiment of the present application, the sum or product of the first complexity factor and the second complexity factor can be used to realize the comprehensive processing of the two, which is not limited herein.

[0066] As an example, in an embodiment of the present application, the expression of the data representation complexity of the Raman spectrum curve can be specifically as follows:

[0067] Wherein, U represents the data representation complexity of the Raman spectrum curve; represents the average value of the adjustment degree of all the fluctuation sections of the Raman spectrum curve, i.e., the first complexity factor of the Raman spectrum curve; represents the standard deviation of the characteristic values of all the fluctuation sections of the Raman spectrum curve, i.e., the second complexity factor of the Raman spectrum curve; represents a normalization function used for normalization processing.

[0068] Then, based on the calculation formula of the wavelet decomposition layer number, the wavelet decomposition layer number of the adjusted spectrum curve is obtained, and the calculation formula of the wavelet decomposition layer number is as follows:

[0069] Wherein, V represents the wavelet decomposition layer number of the adjusted spectrum curve; U represents the data representation complexity of the Raman spectrum curve; represents a preset parameter, and the value range is In an embodiment of the present application, the is set to 0.25, The specific numerical value can also be set by the implementer according to the specific implementation scene, which is not limited here. represents a down rounding symbol.

[0070] Wherein, the base decomposition layer selected by the embodiment of the present application is 5 layers, with the increase of the data complexity of the Raman spectrum curve, the decomposition layer of the adjusted spectrum curve is gradually increased to remove the noise more effectively, and the wavelet decomposition layer is divided into 5 grades based on the data complexity of the Raman spectrum curve, i.e. 5-9 layers.

[0071] Further, based on the wavelet decomposition layer, and selecting Daubechies wavelet as the wavelet base function, the adjusted spectrum curve is wavelet decomposed to obtain low frequency coefficients and high frequency coefficients, since the noise usually exists in the high frequency coefficients, the high frequency coefficients are removed, the low frequency coefficients are retained, and the retained low frequency coefficients are inversely wavelet decomposed to reconstruct the final spectrum curve, wherein, the wavelet decomposition and the inverse wavelet decomposition are both the technical means well known by the person skilled in the art, which will not be described here.

[0072] At this point, the baseline drift and noise of the Raman spectrum of the endogenous metabolite of the patient are removed.

[0073] Step S4: based on the final spectrum curve, detecting the endogenous metabolite of the patient.

[0074] The final spectrum curve removes the baseline drift phenomenon and noise, and the quality of the spectrum data is better, so the endogenous metabolite of the patient can be detected based on the final spectrum curve, and the accuracy of the detection of the endogenous metabolite of the patient is improved.

[0075] Preferably, in an embodiment of the present application, the method for detecting the endogenous metabolite of the patient specifically comprises: The sample detection using spectral data usually includes qualitative analysis and quantitative analysis, wherein the qualitative analysis refers to identifying the types of components contained in the sample, and the quantitative analysis refers to analyzing the content or concentration of various components. First, a to-be-detected substance in the endogenous metabolite is determined, for example, lactic acid in the endogenous metabolite can be taken as the to-be-detected substance. Then, a characteristic peak of the to-be-detected substance is extracted on the final spectral curve. The specific process is as follows: the characteristic peak of the to-be-detected substance is identified by comparing the peak formed on the final spectral curve with a standard peak in a database. This process is a technical means familiar to those skilled in the art, and will not be described here. Based on the characteristic peak of the to-be-detected substance and in combination with a quantitative analysis model, the content of the to-be-detected substance is detected. The specific process is as follows: the content of the to-be-detected substance is obtained by inputting the peak height or peak area of the characteristic peak of the to-be-detected substance into the quantitative analysis model. The quantitative analysis model is used to reflect the relationship between the peak height or peak area of the characteristic peak of the to-be-detected substance and the content of the to-be-detected substance. The quantitative analysis model and the content detection are both technical means familiar to those skilled in the art, and will not be described here.

[0076] An embodiment of the present application provides a device for detecting endogenous metabolites of patients with metabolic disorders. The device comprises a memory, a processor and a computer program. The memory is used to store the corresponding computer program. The processor is used to run the corresponding computer program. The computer program can realize the method described in steps S1-S4 when running in the processor.

[0077] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0078] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other. Each embodiment mainly describes the differences from other embodiments.

Claims

1. A method for detecting an endogenous metabolite of a patient with a metabolic disorder, characterized by, The method comprises: obtaining a Raman spectrum curve of an endogenous metabolite of a patient, the Raman spectrum curve comprising Raman intensity at different Raman shifts; segmenting the Raman spectrum curve according to fluctuations of the Raman spectrum curve to obtain a plurality of fluctuation segments and Raman intensity span value, Raman shift span value and reference baseline value of each fluctuation segment; analyzing consistency of change direction between the Raman intensity and the reference baseline value of each adjacent fluctuation segment to screen out normal fluctuation segments; obtaining a normal baseline value of the Raman spectrum curve according to the Raman intensity span value and the Raman shift span value of each normal fluctuation segment and in combination with the reference baseline value of each normal fluctuation segment; obtaining an adjustment degree of each fluctuation segment according to difference between the reference baseline value and the normal baseline value of each fluctuation segment and the Raman intensity span value and the Raman shift span value of each fluctuation segment; adjusting each fluctuation segment of the Raman spectrum curve based on the adjustment degree to obtain an adjusted spectrum curve; denoising the adjusted spectrum curve to obtain a final spectrum curve; detecting the endogenous metabolite of the patient based on the final spectrum curve.

2. The method for detecting endogenous metabolites in patients with metabolic disorders according to claim 1, characterized in that, The method comprises: obtaining a Raman spectrum curve of an endogenous metabolite of a patient, the Raman spectrum curve comprising Raman intensity at different Raman shifts; segmenting the Raman spectrum curve according to fluctuations of the Raman spectrum curve to obtain a plurality of fluctuation segments and Raman intensity span value, Raman shift span value and reference baseline value of each fluctuation segment; analyzing consistency of change direction between the Raman intensity and the reference baseline value of each adjacent fluctuation segment to screen out normal fluctuation segments; obtaining a normal baseline value of the Raman spectrum curve according to the Raman intensity span value and the Raman shift span value of each normal fluctuation segment and in combination with the reference baseline value of each normal fluctuation segment; obtaining an adjustment degree of each fluctuation segment according to difference between the reference baseline value and the normal baseline value of each fluctuation segment and the Raman intensity span value and the Raman shift span value of each fluctuation segment; adjusting each fluctuation segment of the Raman spectrum curve based on the adjustment degree to obtain an adjusted spectrum curve; denoising the adjusted spectrum curve to obtain a final spectrum curve; detecting the endogenous metabolite of the patient based on the final spectrum curve. The method comprises:

3. The method for detecting endogenous metabolites in patients with metabolic disorders according to claim 1, characterized in that, obtaining a Raman spectrum curve of an endogenous metabolite of a patient, the Raman spectrum curve comprising Raman intensity at different Raman shifts; segmenting the Raman spectrum curve according to fluctuations of the Raman spectrum curve to obtain a plurality of fluctuation segments and Raman intensity span value, Raman shift span value and reference baseline value of each fluctuation segment; analyzing consistency of change direction between the Raman intensity and the reference baseline value of each adjacent fluctuation segment to screen out normal fluctuation segments; obtaining a normal baseline value of the Raman spectrum curve according to the Raman intensity span value and the Raman shift span value of each normal fluctuation segment and in combination with the reference baseline value of each normal fluctuation segment; obtaining an adjustment degree of each fluctuation segment according to difference between the reference baseline value and the normal baseline value of each fluctuation segment and the Raman intensity span value and the Raman shift span value of each fluctuation segment; adjusting each fluctuation segment of the Raman spectrum curve based on the adjustment degree to obtain an adjusted spectrum curve; denoising the adjusted spectrum curve to obtain a final spectrum curve; detecting the endogenous metabolite of the patient based on the final spectrum curve. The method comprises:

4. The method for detecting endogenous metabolites in patients with metabolic disorders according to claim 1, characterized in that, obtaining a Raman spectrum curve of an endogenous metabolite of a patient, the Raman spectrum curve comprising Raman intensity at different Raman shifts; segmenting the Raman spectrum curve according to fluctuations of the Raman spectrum curve to obtain a plurality of fluctuation segments and Raman intensity span value, Raman shift span value and reference baseline value of each fluctuation segment; analyzing consistency of change direction between the Raman intensity and the reference baseline value of each adjacent fluctuation segment to screen out normal fluctuation segments; obtaining a normal baseline value of the Raman spectrum curve according to the Raman intensity span value and the Raman shift span value of each normal fluctuation segment and in combination with the reference baseline value of each normal fluctuation segment; obtaining an adjustment degree of each fluctuation segment according to difference between the reference baseline value and the normal baseline value of each fluctuation segment and the Raman intensity span value and the Raman shift span value of each fluctuation segment; adjusting each fluctuation segment of the Raman spectrum curve based on the adjustment degree to obtain an adjusted spectrum curve; denoising the adjusted spectrum curve to obtain a final spectrum curve; detecting the endogenous metabolite of the patient based on the final spectrum curve. The method comprises: obtaining a Raman spectrum curve of an endogenous metabolite of a patient, the Raman spectrum curve comprising Raman intensity at different Raman shifts; segmenting the Raman spectrum curve according to fluctuations of the Raman spectrum curve to obtain a plurality of fluctuation segments and Raman intensity span value, Raman shift span value and reference baseline value of each fluctuation segment; analyzing consistency of change direction between the Raman intensity and the reference baseline value of each adjacent fluctuation segment to screen out normal fluctuation segments; obtaining a normal baseline value of the Raman spectrum curve according to the Raman intensity span value and the Raman shift span value of each normal fluctuation segment and in combination with the reference baseline value of each normal fluctuation segment; obtaining an adjustment degree of each fluctuation segment according to difference between the reference baseline value and the normal baseline value of each fluctuation segment and the Raman intensity span value and the Raman shift span value of each fluctuation segment; adjusting each fluctuation segment of the Raman spectrum curve based on the adjustment degree to obtain an adjusted spectrum curve; denoising the adjusted spectrum curve to obtain a final spectrum curve; detecting the endogenous metabolite of the patient based on the final spectrum curve. comprehending the Raman intensity span value and the Raman frequency shift span value of each normal fluctuation section to obtain a reference factor of each normal fluctuation section; taking the reference factor of each normal fluctuation section as a numerator and taking the cumulative value of the reference factors of all normal fluctuation sections as a denominator to obtain a baseline weight of each normal fluctuation section; performing weighted summation on the reference baseline value of each normal fluctuation section by using the baseline weight of each normal fluctuation section to obtain a normal baseline value of the Raman spectrum curve.

5. The method for detecting endogenous metabolites in patients with metabolic disorders according to claim 1, characterized in that, The obtaining of the adjustment degree of each fluctuation section comprises: taking the absolute value of the difference between the reference baseline value and the normal baseline value of each fluctuation section as a first adjustment parameter of each fluctuation section; comprehending the Raman intensity span value and the Raman frequency shift span value of each fluctuation section and performing negative correlation mapping to obtain a second adjustment parameter of each fluctuation section; comprehending the first adjustment parameter and the second adjustment parameter and performing normalization processing to obtain the adjustment degree of each fluctuation section.

6. The method for detecting endogenous metabolites in patients with metabolic disorders according to claim 1, characterized in that, The obtaining of the adjusted spectrum curve comprises: taking any fluctuation section of the Raman spectrum curve as a fluctuation section to be adjusted; taking the product value of the adjustment degree of the fluctuation section to be adjusted and the normal baseline value of the Raman spectrum curve as a Raman intensity adjustment amount of the fluctuation section to be adjusted; taking the difference between the Raman intensity and the Raman intensity adjustment amount of each Raman frequency shift of the fluctuation section to be adjusted as an adjusted Raman intensity of each Raman frequency shift of the fluctuation section to be adjusted; replacing the Raman intensity of each Raman frequency shift of each fluctuation section of the Raman spectrum curve with the adjusted Raman intensity to obtain the adjusted spectrum curve.

7. The method for detecting endogenous metabolites in patients with metabolic disorders according to claim 1, characterized in that, The obtaining of the final spectrum curve comprises: obtaining the wavelet decomposition layer number of the adjusted spectrum curve according to the adjustment degree, the Raman intensity span value and the Raman frequency shift span value of all fluctuation sections of the Raman spectrum curve; performing wavelet decomposition on the adjusted spectrum curve based on the wavelet decomposition layer number to obtain low-frequency coefficients and high-frequency coefficients, removing the high-frequency coefficients, retaining the low-frequency coefficients, and performing inverse wavelet decomposition on the retained low-frequency coefficients to reconstruct the final spectrum curve.

8. The method for detecting endogenous metabolites in patients with metabolic disorders according to claim 7, characterized in that, The obtaining of the wavelet decomposition layer number of the adjusted spectrum curve comprises: taking the average value of the adjustment degrees of all fluctuation sections of the Raman spectrum curve as a first complexity factor of the Raman spectrum curve; taking the product value of the Raman intensity span value and the Raman frequency shift span value of each fluctuation section of the Raman spectrum curve as an eigenvalue of each fluctuation section, analyzing the discrete degree of the eigenvalues of all fluctuation sections of the Raman spectrum curve to obtain a second complexity factor of the Raman spectrum curve; comprehending the first complexity factor and the second complexity factor and performing normalization processing to obtain a data performance complexity of the Raman spectrum curve; obtaining the wavelet decomposition layer number of the adjusted spectrum curve based on a calculation formula of the wavelet decomposition layer number, the calculation formula of the wavelet decomposition layer number being: ； Wherein, V represents the wavelet decomposition layer number of adjusting the spectral curve; U represents the data performance complexity of the Raman spectral curve; represents a preset parameter, and the value range is .

9. The method for detecting endogenous metabolites in patients with metabolic disorders according to claim 1, characterized in that, The detecting of the endogenous metabolite of the patient comprises: The characteristic peak of the to-be-detected substance is extracted on a final spectrum curve, and the content of the to-be-detected substance is detected based on the characteristic peak of the to-be-detected substance and in combination with a quantitative analysis model.

10. An apparatus for detecting endogenous metabolites in a patient with a metabolic disorder, the apparatus comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, wherein, The processor implements the steps of the method of any one of claims 1-9 when executing the computer program.

Images