Raman spectrum-based detection method for helicobacter pylori infection in stomach

Abstract

The present application relates to the technical field of Raman spectrum detection, and in particular to a stomach H. pylori infection detection method based on Raman spectrum. The method performs Raman spectrum detection on a gastric juice sample to obtain a Raman spectrum curve; a healthy Raman spectrum curve of a healthy gastric juice sample is obtained and constructed into a healthy sample spectrum library; according to the Raman intensity difference and waveform distribution similarity of each healthy Raman spectrum curve in the healthy sample spectrum library with other healthy Raman spectrum curves in a specified wavelength range, a representative weight of each healthy Raman spectrum curve is obtained; according to the Raman intensity distribution similarity of the Raman spectrum curve of the current patient with each healthy Raman spectrum curve in the specified wavelength range and the representative weight of each healthy Raman spectrum curve, the infection degree of the current patient is obtained, and it is determined whether the current patient is infected with H. pylori. The present application effectively improves the preparedness and universality of H. pylori detection by obtaining the infection degree.

Description

Raman spectroscopy-based detection method for Helicobacter pylori infection in the stomach Technical Field

[0001] This invention relates to the field of Raman spectroscopy detection technology, and specifically to a method for detecting Helicobacter pylori infection in the stomach based on Raman spectroscopy. Background Technology

[0002] Helicobacter pylori is a microaerophilic bacterium that colonizes the surface of the human gastric mucosa. It is an important pathogenic factor for a variety of gastric diseases, including chronic gastritis, peptic ulcers, gastric mucosa-associated lymphoid tissue lymphoma, and even gastric cancer. It is also highly contagious. Accurate detection of its infection status has become a crucial first step in the prevention and treatment of the above-mentioned diseases.

[0003] Currently, the main clinical method for detecting Helicobacter pylori infection is through Raman spectroscopy of gastric juice samples, achieving minimally invasive and rapid detection. However, in practical applications, the complex components of gastric juice, such as proteins, mucus, and food residue, generate strong and variable background signals, severely interfering with and masking the weak bacterial characteristic peaks. Furthermore, even among healthy individuals, the Raman spectral background of gastric juice varies significantly due to differences in age, diet, and lifestyle, making it difficult to establish a stable and unified standard spectral reference for health. Consequently, directly identifying characteristic peaks against different backgrounds easily leads to misdiagnosis or missed detection of Helicobacter pylori, severely limiting the accuracy and universality of existing Raman spectroscopy detection methods. Summary of the Invention

[0004] To address the technical problem that existing Raman spectroscopy detection methods cannot accurately detect Helicobacter pylori, the present invention aims to provide a Raman spectroscopy-based method for detecting Helicobacter pylori infection in the stomach. The specific technical solution adopted is as follows:

[0005] This invention provides a Raman spectroscopy-based method for detecting Helicobacter pylori infection in the stomach, comprising the following steps:

[0006] Raman spectroscopy was performed on gastric fluid samples to obtain Raman spectral curves;

[0007] A healthy Raman spectrum library was constructed by obtaining healthy Raman spectrum curves from multiple healthy gastric fluid samples. Based on the differences in Raman intensity and waveform distribution between each healthy Raman spectrum curve and other healthy Raman spectrum curves in the specified wavelength range, the representative weight of each healthy Raman spectrum curve was obtained.

[0008] The degree of infection of the current patient is obtained based on the similarity of the Raman intensity distribution of the current patient's Raman spectrum curve with that of each healthy Raman spectrum curve in the healthy sample spectral library within a specified wavelength range, and the representative weight of each healthy Raman spectrum curve.

[0009] The degree of infection determines whether the patient is currently infected with Helicobacter pylori.

[0010] Furthermore, the specified wavelength range includes a first characteristic band and a second characteristic band; wherein, the wavelength range corresponding to the first characteristic band is... The wavelength range corresponding to the second characteristic band is: .

[0011] Furthermore, the method for obtaining the representative weight is as follows:

[0012] Based on the difference in Raman intensity between each healthy Raman spectrum curve and other healthy Raman spectrum curves within a specified wavelength range, the first representativeness of each healthy Raman spectrum curve is obtained.

[0013] Based on the similarity of the waveform distribution of each healthy Raman spectrum curve to other healthy Raman spectrum curves within a specified wavelength range, a second representativeness of each healthy Raman spectrum curve is obtained.

[0014] The normalized result of the product of the first and second representative degrees of each healthy Raman spectrum is used as the representative weight of each healthy Raman spectrum.

[0015] Furthermore, the method for obtaining the first representativeness is as follows:

[0016] The maximum Raman intensity of each healthy Raman spectrum in the first characteristic band is taken as the first reference intensity; the maximum Raman intensity of each healthy Raman spectrum in the second characteristic band is taken as the second reference intensity.

[0017] Based on the distribution of the first and second reference intensities of each healthy Raman spectrum, the first representative intensity and the second representative intensity are obtained.

[0018] The difference between the first reference intensity and the first representative intensity of each healthy Raman spectrum curve is taken as the first difference; the difference between the second reference intensity and the second representative intensity of each healthy Raman spectrum curve is taken as the second difference; the result of negatively correlated mean values ​​of the first difference and the second difference of each healthy Raman spectrum curve is taken as the first reference level of each healthy Raman spectrum curve.

[0019] The difference between the wavelengths corresponding to the first and second reference intensities of each healthy Raman spectrum curve is used as the wavenumber distance analysis value.

[0020] Based on the distribution of wavenumber distance analysis values, representative wavenumber distance analysis values ​​are obtained;

[0021] The result of negatively correlated wavenumber distance analysis value of each healthy Raman spectrum with the difference of representative wavenumber distance analysis value is used as the second reference level for each healthy Raman spectrum.

[0022] The product of the first and second reference levels of each healthy Raman spectrum is taken as the first representative level of each healthy Raman spectrum.

[0023] Furthermore, the method for obtaining the first representative intensity and the second representative intensity is as follows:

[0024] The first and second reference intensities of each healthy Raman spectrum are combined to construct a vector, which is used as a reference vector.

[0025] Based on the magnitudes between reference vectors, the reference vectors are clustered using the DBSCAN density clustering algorithm to obtain vector clusters;

[0026] The mean of all first reference intensities in the largest vector cluster is used as the first representative intensity.

[0027] The mean of all second reference intensities in the largest vector cluster is used as the second representative intensity.

[0028] Furthermore, the method for obtaining the representative wavenumber distance analysis value is as follows:

[0029] The wavenumber distance analysis values ​​of all healthy Raman spectral curves were clustered using the DBSCAN density clustering algorithm to obtain wavenumber distance clusters;

[0030] The mean of all wavenumber distance analysis values ​​in the largest wavenumber distance cluster is used as the representative wavenumber distance analysis value.

[0031] Furthermore, the method for obtaining the second degree of representation is as follows:

[0032] The dynamic time warping algorithm is used to obtain the warped paths of the a-th healthy Raman spectrum curve and the b-th healthy Raman spectrum curve in the first and second characteristic bands, which are both used as analysis paths. Among them, the a-th healthy Raman spectrum curve is any healthy Raman spectrum curve in the healthy sample spectral library, and the b-th healthy Raman spectrum curve is any healthy Raman spectrum curve in the healthy sample spectral library other than the a-th healthy Raman spectrum curve.

[0033] Based on the changes in the analysis path, the degree of distortion of the a-th healthy Raman spectrum curve relative to the b-th healthy Raman spectrum curve is obtained;

[0034] When the degree of distortion is greater than a preset distortion threshold, the b-th healthy Raman spectrum curve is marked as the reference distortion curve of the a-th healthy Raman spectrum curve.

[0035] The proportion of all reference distortion curves of the a-th healthy Raman spectrum curve in the healthy sample spectral library is used as the first degree of deviation of the a-th healthy Raman spectrum curve.

[0036] The average of the distortion levels of the a-th healthy Raman spectrum curve and all its reference distortion curves is taken as the second degree of deviation of the a-th healthy Raman spectrum curve.

[0037] The result of negatively correlating the product of the first deviation degree and the second deviation degree is used as the second representative degree of the a-th healthy Raman spectrum curve.

[0038] Furthermore, the method for obtaining the degree of distortion is as follows:

[0039] For any analysis path of the a-th healthy Raman spectrum and the b-th healthy Raman spectrum, the slope of the analysis path is obtained and used as the reference slope.

[0040] When the absolute value of the normalized reference slope is greater than the preset deformation threshold, the corresponding local path segment is taken as the path deformation segment.

[0041] The product of the length of each path deformation segment and the absolute value of the reference slope is used as the change analysis value of each path deformation segment;

[0042] The ratio of the total number of deformed segments to the total number of local path segments corresponding to all reference slopes of the analyzed path is taken as the overall degree of change of the analyzed path.

[0043] The result of normalizing the product of the mean of the change analysis values ​​of all path deformation segments and the overall degree of change is used as the distortion analysis value of the analysis path.

[0044] The average of the distortion analysis values ​​of all analysis paths of the a-th healthy Raman spectrum curve and the b-th healthy Raman spectrum curve is taken as the degree of distortion of the a-th healthy Raman spectrum curve relative to the b-th healthy Raman spectrum curve.

[0045] Furthermore, the method for obtaining the degree of infection is as follows:

[0046] Arrange the Raman intensity of the current patient's Raman spectrum within the specified wavelength range according to wavelength order and construct a row vector as the first vector;

[0047] For any healthy Raman spectrum curve in the healthy sample spectral library, the Raman intensity of the healthy Raman spectrum curve within the specified wavelength range is arranged in wavelength order and constructed as a row vector, which serves as the second vector;

[0048] The magnitudes of the first and second vectors are used as reference infection analysis values ​​for the current patient relative to the healthy Raman spectrum curve.

[0049] The product of the representative weight of the healthy Raman spectrum curve and the reference infection analysis value is used as the corrected reference infection analysis value of the current patient relative to the healthy Raman spectrum curve.

[0050] The normalized result of the mean of the corrected reference infection analysis values ​​of the current patient relative to all healthy Raman spectral curves is taken as the infection level of the current patient.

[0051] Furthermore, the method for determining whether a current patient is infected with Helicobacter pylori based on the degree of infection is as follows:

[0052] When the infection level is greater than or equal to a preset infection level threshold, the patient is determined to be infected with Helicobacter pylori.

[0053] When the infection level is less than the preset infection level threshold, it is determined that the current patient is not infected with Helicobacter pylori.

[0054] The present invention has the following beneficial effects:

[0055] This invention utilizes a rigorously quality-controlled spectral library of healthy samples and introduces representative weights and infection levels to effectively overcome the problem of spectral background interference caused by individual physiological differences. Simultaneously, it leverages Raman spectroscopy's sensitive detection capability for chemical components, combined with intelligent algorithms to extract and weighted compare feature information, significantly reducing the risk of false negatives and false positives common in traditional methods. Furthermore, the quantified representative weights reduce subjective judgment errors, making diagnostic results more objective and reliable, and effectively improving the preparedness and universality of Helicobacter pylori testing for patients. Attached Figure Description

[0056] 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.

[0057] Figure 1 is a schematic flowchart of a method for detecting Helicobacter pylori infection in the stomach based on Raman spectroscopy according to an embodiment of the present invention;

[0058] Figure 2 is a comparison diagram of raw Raman spectral data provided in an embodiment of the present invention;

[0059] Figure 3 is a flowchart of a method for obtaining representative weights according to an embodiment of the present invention;

[0060] Figure 4 is a comparison chart of ROC curves provided in an embodiment of the present invention;

[0061] Figure 5 is a structural diagram of a gastric Helicobacter pylori infection detection system based on Raman spectroscopy provided in an embodiment of the present invention;

[0062] Figure 6 is a schematic diagram of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0063] 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 the Raman spectroscopy-based method for detecting Helicobacter pylori infection in the stomach 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.

[0064] 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.

[0065] The specific scheme of the Raman spectroscopy-based method for detecting Helicobacter pylori infection in the stomach provided by this invention is described in detail below with reference to the accompanying drawings.

[0066] Example 1:

[0067] This invention proposes a Raman spectroscopy-based method for detecting Helicobacter pylori infection in the stomach. Please refer to Figure 1, which shows a schematic flowchart of a Raman spectroscopy-based method for detecting Helicobacter pylori infection in the stomach according to an embodiment of this invention. The method includes the following steps:

[0068] Step S1: Perform Raman spectroscopy on the gastric fluid sample to obtain the Raman spectral curve.

[0069] Specifically, in order to accurately detect whether a patient's stomach is infected with Helicobacter pylori using Raman spectroscopy, this embodiment requires obtaining a gastric fluid sample from the patient. Methods for obtaining the gastric fluid sample include endoscopic aspiration and capsule sampling. It should be noted that the endoscopic aspiration method is suitable for patients undergoing gastroscopy. After the endoscope reaches the appropriate position in the stomach, the aspiration tube attached to the endoscope is inserted into the stomach through the working channel of the endoscope. Gentle aspiration is performed in different parts of the stomach, such as the stomach body and antrum, collecting about 5 to 10 ml of gastric fluid sample in a sterile container. The capsule sampling line method is suitable for patients who are not undergoing gastroscopy. The patient needs to swallow a specially designed Helicobacter pylori detection capsule on an empty stomach. The capsule has a sampling line attached to the outside. After swallowing the capsule, the end of the sampling line is fixed to the cheek with medical tape. Then, the patient swallows about 200 to 300 ml of drinking water in small sips to help the capsule enter the stomach smoothly. After 0.5 to 2 hours, the sampling line is slowly removed and immediately placed into a sterile centrifuge tube containing an appropriate amount of preservation solution (such as phosphate buffer). By gently shaking the centrifuge tube, the gastric fluid components adsorbed on the sampling line are fully dissolved in the preservation solution to obtain a gastric fluid sample.

[0070] After obtaining the gastric fluid sample, the sample is centrifuged at 3000 to 5000 rpm for 5 to 10 minutes to separate and remove the upper layer of grease and insoluble impurities, while retaining the lower clear liquid. Then, an appropriate amount of the lower clear liquid is mixed with a pre-prepared Raman enhancement substrate (e.g., nano-silver sol) at a volume ratio of 1:1 to 1:3 and thoroughly shaken to ensure uniform mixing, thereby enhancing the Raman signal for subsequent detection and obtaining the pre-treated gastric fluid sample. To accurately obtain the Raman spectral curve corresponding to the gastric fluid sample, key parameters such as the excitation wavelength, exposure time, and scanning range of the Raman spectrometer were set. The pretreated gastric fluid sample was then dropped onto a clean quartz slide and covered with a coverslip to ensure uniform distribution and absence of air bubbles. The prepared slide was then placed on the sample stage of the Raman spectrometer, and the microscope was adjusted until the sample was clearly imaged. Spectroscopic acquisition was performed on a uniform region or the center point of the sample. Finally, the Raman spectrometer converted the received scattered light signal into an electrical signal, which was then recorded in electronic digital form. The accompanying data processing software read these digital signals and plotted them as a Raman spectral curve with wavelength as the abscissa and Raman intensity as the ordinate. This visually displays the overall shape of the spectrum, the position and intensity of characteristic peaks, and other key information, preparing for the subsequent accurate detection of Helicobacter pylori infection in the patient's stomach using the Raman spectral curve.

[0071] Step S2: Obtain healthy Raman spectral curves from multiple healthy gastric fluid samples to construct a healthy sample spectral library. Based on the differences in Raman intensity and waveform distribution between each healthy Raman spectral curve and other healthy Raman spectral curves in the specified wavelength range, obtain the representative weight of each healthy Raman spectral curve.

[0072] Specifically, to accurately and efficiently analyze whether a patient has Helicobacter pylori, this embodiment first obtains Raman spectral curves from multiple healthy gastric fluid samples to construct a healthy sample spectral library, which is then compared with the Raman spectral curves of the patients to be analyzed. To construct the healthy sample spectral library for comparison, Raman spectral data from two types of samples need to be collected and integrated: firstly, gastric fluid samples from patients who have been clinically confirmed to have recovered from Helicobacter pylori infection; and secondly, gastric fluid samples from healthy individuals confirmed by testing. The composition of the healthy sample spectral library should fully cover different age groups and genders to ensure broad population representativeness.

[0073] Although the healthy Raman spectra in the healthy sample spectral library are all derived from healthy individuals, due to differences in individual physical factors such as age, diet, and physiological state, their Raman spectra may exhibit inherent biological variations in background signal, baseline, and relative intensity of characteristic peaks. Considering that when a patient has Helicobacter pylori infection in their stomach, the Raman spectra of their gastric fluid samples will show characteristic changes in specific wavelength bands directly related to bacterial metabolism and proliferation, this embodiment first obtains the representative weight of each healthy Raman spectra based on the differences in Raman intensity and waveform distribution between each healthy Raman spectra in the healthy sample spectral library and other healthy Raman spectra within a specified wavelength range. The higher the representative weight, the more representative the corresponding healthy Raman spectra are in the healthy sample spectral library.

[0074] Urease, a unique enzyme produced by Helicobacter pylori, is known to exhibit certain characteristics in Raman spectroscopy. Significantly enhanced peaks will appear within the wavelength range, directly indicating the presence of Helicobacter pylori. Urease can sensitively reflect the degree of Helicobacter pylori infection, whether mild or moderate, which can be reflected by changes in the Raman intensity corresponding to the peaks. This is because as the number of Helicobacter pylori increases, the total amount of urease also increases, directly leading to... The Raman intensity corresponding to the peaks within the wavelength range also increases accordingly; furthermore, Helicobacter pylori contains nucleic acid material, and the phosphodiester bonds in the nucleic acid backbone have specific Raman vibration modes. Characteristic Raman intensity peaks of phosphodiester bonds can be observed within the wavelength range. During infection, the active proliferation of Helicobacter pylori leads to an increase in bacterial count, resulting in a significant increase in the total amount of bacterial-derived nucleic acids. This change is reflected in the Raman spectral curve as an enhancement of the characteristic peak intensity of phosphodiester bonds. Therefore… The enhancement of peaks within the wavelength range indirectly reflects the high proliferation rate of Helicobacter pylori in the stomach, providing auxiliary spectral evidence for confirming infection. Therefore, this embodiment defines the specified wavelength range as including a first characteristic band and a second characteristic band; wherein, the wavelength range corresponding to the first characteristic band is... The wavelength range corresponding to the second characteristic band is: .

[0075] It should be noted that Helicobacter pylori possesses extremely high urease activity, which is a key mechanism for its survival in the highly acidic environment of the stomach and a significant characteristic distinguishing it from other common gastric bacteria (such as lactobacilli and streptococci). Although part of the protein backbone (CC backbone) is in The interval exhibits a Raman response, but specific CN stretching vibrations and α-helical structures in urease are present. The surrounding area exhibits unique strong Raman scattering peaks; Raman difference spectral analysis of healthy gastric juice (mainly containing pepsin and mucin) and Helicobacter pylori-positive gastric juice revealed that healthy gastric juice showed a strong Raman scattering peak in the first characteristic band. The signal is relatively flat and low in intensity, while infected samples show significant sharp peaks at this point, with a signal-to-noise ratio typically greater than 5, indirectly indicating the first characteristic band. It is less affected by background proteins in gastric juice and has high specificity. Second characteristic band. Mainly corresponds to PO2 in the nucleic acid backbone - Symmetric stretching vibration (approximately) Although exfoliated gastric mucosal epithelial cells in gastric juice also contain nucleic acids, under Helicobacter pylori infection, the rapid proliferation of bacteria leads to an exponential increase in the concentration of bacterial-derived nucleic acids, and their Raman signal intensity far exceeds the background level of normal exfoliated cells, compared to other nucleic acid bands (such as...). OPO skeleton). The band is affected by the glycan vibrations of mucin, the main interfering substance in gastric juice (mainly concentrated in...). and This method minimizes interference and provides a purer reflection of bacterial load. Furthermore, single-band analysis can be affected by random factors (such as specific proteins in food residues). Therefore, this embodiment employs a dual-band joint analysis mechanism: only when the urease band, representing bacterial metabolic activity, is affected is the bacterial load observed. ) and nucleic acid bands representing bacterial proliferation ( Only when a substance exhibits an intensity distribution and waveform consistent with the spectral characteristics of Helicobacter pylori will it be identified with high weight by the algorithm. This mechanism effectively filters out non-specific substances that only produce interfering signals in a single band (such as food residues containing only protein and no high concentration of nucleic acid). As shown in Figure 2, which is a comparison of raw Raman spectral data, it accurately shows that the intensity of Helicobacter pylori infection in the first and second characteristic bands is greater.

[0076] Preferably, in one feasible embodiment, the method for obtaining the representative weight is shown in Figure 3, which illustrates a flowchart of a method for obtaining the representative weight provided in this embodiment. This method includes the following steps:

[0077] Step S201: Based on the difference in Raman intensity between each healthy Raman spectrum curve and other healthy Raman spectrum curves within a specified wavelength range, obtain the first representativeness of each healthy Raman spectrum curve.

[0078] To quantify the individual differences and representativeness of each healthy Raman spectrum curve in the healthy sample spectral library, this embodiment first obtains the first representativeness of each healthy Raman spectrum curve based on the difference in Raman intensity between each healthy Raman spectrum curve and other healthy Raman spectrum curves within a specified wavelength range. The greater the first representativeness, the more similar the Raman intensity distribution of the corresponding healthy Raman spectrum curve is to other healthy Raman spectrum curves within the specified wavelength range, and the more representative the corresponding healthy Raman spectrum curve is.

[0079] In one possible implementation of this embodiment, the method for obtaining the first representative intensity is as follows: the maximum Raman intensity of each healthy Raman spectral curve in the first characteristic band is used as the first reference intensity; the maximum Raman intensity of each healthy Raman spectral curve in the second characteristic band is used as the second reference intensity; based on the distribution of the first and second reference intensities of each healthy Raman spectral curve, the first representative intensity and the second representative intensity are obtained; wherein, the method for obtaining the first and second representative intensities is as follows: the first and second reference intensities of each healthy Raman spectral curve are combined to construct a vector, which is used as a reference vector; based on the modulus between the reference vectors, the reference vectors are clustered using the DBSCAN density clustering algorithm. The data is clustered to obtain vector clusters; then the mean of all first reference intensities in the largest vector cluster is used as the first representative intensity; the mean of all second reference intensities in the largest vector cluster is used as the second representative intensity. It should be noted that, due to physiological differences in the human body, the Raman characteristic peak intensity (reference vector) of healthy people does not present a regular spherical distribution, but may present a density connection region of arbitrary shape, and there may be a small number of outliers (abnormal healthy samples). Therefore, this embodiment uses the DBSCAN density clustering algorithm to cluster the reference vectors, which can effectively identify and filter out noise points (non-representative samples), thereby extracting the core health pattern that truly has high density connectivity and represents the general health level. In this embodiment, the neighborhood radius of the DBSCAN density clustering algorithm is set to 0.2 times the average distance between all reference vectors. This range ensures that samples within the same cluster have high similarity in feature intensity. A minimum sample size of 5 is set to ensure that a cluster contains at least a certain number of samples to be considered a valid healthy pattern, avoiding pseudo-clustering caused by a few randomly similar samples. Implementers can set the neighborhood radius and minimum sample size of the DBSCAN density clustering algorithm according to the total number of samples in the healthy sample library and the sparsity of the data; no limitation is imposed here. The method for obtaining the modulus and the DBSCAN density clustering algorithm are well-known technologies and will not be described in detail here.

[0080] Then, the absolute value of the difference between the first reference intensity and the first representative intensity of each healthy Raman spectrum curve is taken as the first difference; the absolute value of the difference between the second reference intensity and the second representative intensity of each healthy Raman spectrum curve is taken as the second difference; the result of negatively correlated mean values ​​of the first difference and the second difference of each healthy Raman spectrum curve is taken as the first reference degree of each healthy Raman spectrum curve; the larger the first reference degree, the more similar the Raman intensity of the corresponding healthy Raman spectrum curve is to other healthy Raman spectrum curves within the specified wavelength range, and the more representative the corresponding healthy Raman spectrum curve is; in this embodiment, the negative of the mean values ​​of the first difference and the second difference is taken as the power of an exponential function with the natural constant as the base, and the output of the exponential function is the result of negatively correlated mean values ​​of the first difference and the second difference;

[0081] Further, the absolute value of the difference between the wavelengths corresponding to the first and second reference intensities of each healthy Raman spectral curve is obtained, and these are used as wavenumber distance analysis values. Representative wavenumber distance analysis values ​​are obtained based on their distribution. The method for obtaining these representative wavenumber distance analysis values ​​is as follows: the wavenumber distance analysis values ​​of all healthy Raman spectral curves are clustered using the DBSCAN density clustering algorithm to obtain wavenumber distance clusters. The DBSCAN density clustering algorithm is used here because, although the characteristic peak positions of healthy samples are relatively fixed, the wavenumber difference (wavenumber distance analysis value) fluctuates within a certain range due to the Raman spectrometer resolution and frequency shift. The DBSCAN density clustering algorithm can automatically identify the densest distribution interval (i.e., the area where most healthy samples are concentrated) within these continuously fluctuating values, while eliminating extremely large or small deviation values ​​caused by instrument errors or sample anomalies, thus obtaining the most statistically significant representative difference. Here, the neighborhood radius of the DBSCAN density clustering algorithm is... (Corresponding to the typical resolution of the spectrometer), the minimum number of samples is 5, ensuring that the selected representative clusters can cover the vast majority of healthy samples within the instrument's allowable error range and have statistical significance. The implementer can set the neighborhood radius and minimum number of samples for the DBSCAN density clustering algorithm here according to the specific resolution of the Raman spectrometer and the signal-to-noise ratio level of the sample acquisition, without limitation.

[0082] Then, the mean of all wavenumber distance analysis values ​​in the largest wavenumber distance cluster is used as the representative wavenumber distance analysis value. The result of negatively correlating the absolute value of the difference between the wavenumber distance analysis value and the representative wavenumber distance analysis value for each healthy Raman spectrum is used as the second reference level for each healthy Raman spectrum. The larger the second reference level, the more similar the Raman intensity positions of the corresponding healthy Raman spectrum to other healthy Raman spectrum curves within the specified wavelength range, and the more representative the corresponding healthy Raman spectrum is. In this embodiment, the negative of the absolute value of the difference between the wavenumber distance analysis value and the representative wavenumber distance analysis value is used as the power of an exponential function with the natural constant as the base. The output of this exponential function is the result of negatively correlating the absolute value of the difference between the wavenumber distance analysis value and the representative wavenumber distance analysis value.

[0083] In order to accurately characterize the representativeness of each healthy Raman spectrum curve, this embodiment uses the product of the first reference level and the second reference level of each healthy Raman spectrum curve as the first representativeness of each healthy Raman spectrum curve.

[0084] Step S202: Based on the similarity of the waveform distribution of each healthy Raman spectrum curve with other healthy Raman spectrum curves within a specified wavelength range, obtain the second representativeness of each healthy Raman spectrum curve.

[0085] To overcome the interference of complex background signals from gastric fluid on Raman characteristic peaks and to accurately assess the waveform quality of each healthy Raman spectral curve in the healthy sample spectral library, thereby enabling a more accurate analysis of the representativeness of each healthy Raman spectral curve, this embodiment obtains a second representativeness of each healthy Raman spectral curve based on the similarity of its waveform distribution to other healthy Raman spectral curves within a specified wavelength range. The greater the second representativeness, the more similar the waveform of the corresponding healthy Raman spectral curve is to other healthy Raman spectral curves within the specified wavelength range, and the more representative the corresponding healthy Raman spectral curve is in the healthy sample spectral library.

[0086] In one possible implementation of this embodiment, the method for obtaining the second representativeness is as follows: The normalized paths of the a-th healthy Raman spectrum and the b-th healthy Raman spectrum in the first and second characteristic bands are obtained through a dynamic time warping algorithm, and both are used as analysis paths. The a-th healthy Raman spectrum is any healthy Raman spectrum in the healthy sample spectral library, and the b-th healthy Raman spectrum is any healthy Raman spectrum in the healthy sample spectral library other than the a-th healthy Raman spectrum. It should be noted that the dynamic time warping algorithm is chosen in this embodiment to address the issue of small frequency shifts in Raman spectra. In actual detection, the Raman characteristic peak positions of the same chemical bond (such as urease) may change due to the influence of instrument resolution, ambient temperature, or sample matrix effects. Even minute shifts can lead to significant errors (bimodal effect) if point-to-point Euclidean distance is used directly. Dynamic time warping algorithms, by non-linearly stretching or compressing the time axis (wavenumber axis), can find the optimal matching path between two waveforms, thus eliminating the influence of frequency shift and focusing on the similarity of the waveform morphology itself. This is particularly crucial for identifying biomarkers with specific peak characteristics. Furthermore, to prevent over-warping from causing unnatural waveform matching (e.g., forcibly matching two completely unrelated peaks), this embodiment applies a Sakoe-Chiba window or Itakura parallelogram constraint to the dynamic time warping algorithm, with the window size set to 5%-10% of the feature band length (e.g., for...). Wide band, window set This limits the maximum deviation of the warped path from the diagonal, ensuring the physical rationality of the matching. The method of obtaining the warped path using the dynamic time warping algorithm is a well-known technique and will not be elaborated further. Then, based on the changes in the analysis path, the degree of distortion of the a-th healthy Raman spectrum curve relative to the b-th healthy Raman spectrum curve is obtained.

[0087] The method for obtaining the degree of distortion is as follows: For any analysis path of the a-th healthy Raman spectrum curve and the b-th healthy Raman spectrum curve, the slope of the analysis path is obtained and used as the reference slope. The closer the reference slope is to 0, the more normal the corresponding local path segment is; the further the reference slope is from 0, the less similar the waveform represented by the corresponding local path segment is. Therefore, this embodiment sets a preset deformation threshold of 0.2. The implementer can set the size of the preset deformation threshold according to the actual situation, which is not limited here. When the absolute value of the normalized reference slope is greater than the preset deformation threshold, the corresponding local path segment is regarded as the path deformation segment. This embodiment normalizes the absolute value of the reference slope through a linear normalization function. The linear normalization method is well known and will not be described in detail here. It is known that the length and slope of the regularized path reflect the difficulty of spectral waveform alignment. The more similar the waveforms of two Raman spectral curves, the shorter the average length of their regularized paths and the gentler the slope change. Therefore, in this embodiment, the product of the length of each path deformation segment and the absolute value of the reference slope is used as the variation analysis value for each path deformation segment. The larger the variation analysis value, the more severe the distortion of the corresponding path deformation segment. Furthermore, the ratio of the total number of path deformation segments to the total number of local path segments corresponding to all reference slopes of the analysis path is used as the overall degree of change of the analysis path. The greater the overall degree of change, the lower the average distortion of the path deformation segment. The larger the distortion, the greater the distortion of the a-th healthy Raman spectrum curve relative to the b-th healthy Raman spectrum curve. To accurately characterize the distortion of the a-th healthy Raman spectrum curve relative to the b-th healthy Raman spectrum curve, the product of the mean of the variation analysis values ​​of all path deformation segments and the overall degree of variation is linearly normalized and used as the distortion analysis value of that analysis path. Then, the mean of the distortion analysis values ​​of all analysis paths of the a-th and b-th healthy Raman spectrum curves is used as the degree of distortion of the a-th healthy Raman spectrum curve relative to the b-th healthy Raman spectrum curve.

[0088] The greater the degree of distortion, the more obvious the distortion of the a-th healthy Raman spectrum curve is compared with the b-th healthy Raman spectrum curve. Therefore, in this embodiment, a preset distortion threshold is set to 0.6. The implementer can set the size of the preset distortion threshold according to the actual situation, which is not limited here. When the distortion level exceeds a preset distortion level threshold, the b-th healthy Raman spectrum curve is marked as the reference distortion curve for the a-th healthy Raman spectrum curve. At this point, all reference distortion curves for the a-th healthy Raman spectrum curve are obtained. Then, the proportion of all reference distortion curves for the a-th healthy Raman spectrum curve in the healthy sample spectral library is used as the first deviation level of the a-th healthy Raman spectrum curve. The average of the distortion levels of the a-th healthy Raman spectrum curve and all its reference distortion curves is used as the second deviation level of the a-th healthy Raman spectrum curve. The larger the first and second deviation levels, the more different the waveform of the a-th healthy Raman spectrum curve is from other healthy Raman spectrum curves in the healthy sample spectral library within the specified wavelength range, and the less representative the a-th healthy Raman spectrum curve is. Therefore, in this embodiment, the product of the first and second deviation levels is negatively correlated, and the result is used as the second representativeness of the a-th healthy Raman spectrum curve. In this embodiment, the negative of the product of the first degree of deviation and the second degree of deviation is used as the power of an exponential function with the natural constant as the base. The output of the exponential function is the result of negatively correlated product of the first degree of deviation and the second degree of deviation.

[0089] At this point, the second representativeness of each healthy Raman spectrum curve is obtained.

[0090] Step S203: Normalize the product of the first and second representative degrees of each healthy Raman spectrum curve and use it as the representative weight of each healthy Raman spectrum curve.

[0091] It is known that a higher level of first and second representativeness indicates a more representative health Raman spectrum. Therefore, in this embodiment, the product of the first and second representativeness levels of each health Raman spectrum is linearly normalized, and this product is used as the representative weight for each health Raman spectrum. The higher the representative weight, the more meaningful the corresponding health Raman spectrum is.

[0092] It should be noted that this embodiment employs a multi-layered calculation strategy that integrates the first and second levels of representativeness to obtain representative weights. Its necessity and advantage lie in addressing the pain points of spectral analysis that simple comparison methods cannot overcome: First, traditional simple comparisons (such as directly calculating the spectral mean) are highly susceptible to differences in sample concentration. Because the collection volume and dilution degree of gastric fluid samples are rarely perfectly consistent, the absolute spectral intensity under the same healthy state may differ by several times. The simple mean method allows high-intensity samples to dominate the model, masking low-intensity but characteristic samples. In this embodiment, the first level of representativeness extracts relative intensity patterns and wavenumber spacing patterns through DBSCAN clustering, focusing on the relative relationships between characteristic peaks (such as the ratio and spacing between the urease peak and the phosphodiester bond peak), rather than absolute intensity. This makes it insensitive to changes in sample concentration and significantly more robust than simple intensity comparisons. Second, Raman spectroscopy often exhibits slight frequency shifts and baseline fluctuations. Simple point-to-point comparison methods (such as Euclidean distance or Pearson correlation coefficient) may misjudge two peaks with the same shape but slightly off-center positions as having huge differences (i.e., the double-peak effect) when faced with frequency shifts, leading to an increased false positive rate. In this embodiment, the second representativeness introduces a dynamic time warping algorithm, which allows waveforms to be nonlinearly scaled and aligned on the time axis (wavenumber axis). This can intelligently identify characteristic waveforms with similar shapes but slightly off-center positions. By calculating the distortion level of the dynamic time warping algorithm path, it can accurately distinguish between frequency shifts caused by instrument errors (low distortion) and waveform changes caused by infection (high distortion), which is something that simple linear comparison methods cannot achieve. Third, by calculating representative weights, a weighted gold standard library is essentially constructed. High weights are assigned to the most typical and stable healthy samples in the library, while low weights are assigned to marginal samples that are healthy but have slightly unusual waveforms (such as dietary interference). In the final judgment, this weighting mechanism forces the sample to be tested to match the most typical health pattern, thereby effectively filtering out random interference and improving the specificity and sensitivity of the detection.

[0093] Step S3: Based on the similarity of the Raman intensity distribution of the current patient's Raman spectrum curve with that of each healthy Raman spectrum curve in the healthy sample spectral library within the specified wavelength range, and the representative weight of each healthy Raman spectrum curve, obtain the infection degree of the current patient.

[0094] Specifically, after obtaining the representative weight of each healthy Raman spectrum curve in the healthy sample spectral library, infection can be determined based on the Raman spectrum curve of the current patient. Furthermore, in this embodiment, the degree of infection of the current patient is obtained based on the similarity of the Raman intensity distribution between the current patient's Raman spectrum curve and each healthy Raman spectrum curve in the healthy sample spectral library within a specified wavelength range, as well as the representative weight of each healthy Raman spectrum curve. The higher the degree of infection, the higher the risk of infection for the current patient.

[0095] Preferably, in one feasible embodiment, the method for obtaining the degree of infection is as follows: The Raman intensities of the current patient's Raman spectrum within a specified wavelength range are arranged in wavelength order and constructed into a row vector, serving as the first vector; for any healthy Raman spectrum in the healthy sample spectral library, the Raman intensities of that healthy Raman spectrum within a specified wavelength range are arranged in wavelength order and constructed into a row vector, serving as the second vector; the more consistent the first vector and the second vector are, the more similar the current patient's Raman spectrum is to the healthy Raman spectrum; and then the first vector and the second vector are... The modulus is used as a reference infection analysis value for the current patient relative to the healthy Raman spectrum curve. The larger the reference infection analysis value, the more different the first vector is from the second vector, and the more likely the current patient is to have Helicobacter pylori. To more accurately analyze the probability of the current patient having Helicobacter pylori, the product of the representative weight of the healthy Raman spectrum curve and the reference infection analysis value is used as the corrected reference infection analysis value for the current patient relative to the healthy Raman spectrum curve. Then, the mean of the corrected reference infection analysis values ​​of the current patient relative to all healthy Raman spectrum curves is linearly normalized to represent the infection degree of the current patient.

[0096] Step S4: Determine whether the current patient is infected with Helicobacter pylori based on the degree of infection.

[0097] It is known that the higher the degree of infection, the higher the risk of the current patient being infected with Helicobacter pylori. Therefore, this embodiment determines whether the current patient is infected with Helicobacter pylori based on the degree of infection. This embodiment first sets a preset infection degree threshold of 0.6. The implementer can set the size of the preset infection degree threshold according to the actual situation, which is not limited here. When the infection degree is greater than or equal to the preset infection degree threshold, it is determined that the current patient is infected with Helicobacter pylori, and further clinical intervention and treatment are recommended; when the infection degree is less than the preset infection degree threshold, it is determined that the current patient is not infected with Helicobacter pylori, and a strategy of continuous observation and regular re-examination is recommended. Among them, the ROC curve comparison diagram shown in Figure 4 accurately represents that this embodiment effectively improves the accuracy of Helicobacter pylori detection. The ACU in Figure 4 represents the overall performance of the detection method. The larger the ACU, the better the overall performance.

[0098] In summary, this embodiment performs Raman spectroscopy on gastric fluid samples to obtain Raman spectral curves. Healthy Raman spectral curves from healthy gastric fluid samples are used to construct a healthy sample spectral library. Based on the differences in Raman intensity and waveform distribution between each healthy Raman spectral curve and other healthy Raman spectral curves within a specified wavelength range, a representative weight is assigned to each healthy Raman spectral curve. Based on the similarity in Raman intensity distribution between the current patient's Raman spectral curve and each healthy Raman spectral curve within a specified wavelength range, as well as the representative weight of each healthy Raman spectral curve, the infection level of the current patient is determined, and it is judged whether the current patient is infected with Helicobacter pylori. This invention effectively improves the accuracy and universality of Helicobacter pylori detection by obtaining the infection level.

[0099] Example 2:

[0100] The present invention also proposes a Raman spectroscopy-based detection system for Helicobacter pylori infection in the stomach. Please refer to Figure 5, which shows a structural diagram of a Raman spectroscopy-based detection system for Helicobacter pylori infection in the stomach provided by an embodiment of the present invention. The system includes: a Raman spectroscopy curve acquisition module 10, a representative weight acquisition module 20, an infection degree acquisition module 30, and a detection module 40.

[0101] The Raman spectral curve acquisition module 10 is used to perform Raman spectral detection on gastric fluid samples and acquire Raman spectral curves.

[0102] The representative weight acquisition module 20 is used to acquire healthy Raman spectral curves of multiple healthy gastric fluid samples to construct a healthy sample spectral library. Based on the differences in Raman intensity and waveform distribution between each healthy Raman spectral curve and other healthy Raman spectral curves in the healthy sample spectral library within a specified wavelength range, the representative weight of each healthy Raman spectral curve is acquired.

[0103] The infection level acquisition module 30 is used to acquire the infection level of the current patient based on the similarity of the Raman intensity distribution of the current patient's Raman spectrum curve with that of each healthy Raman spectrum curve in the healthy sample spectral library within a specified wavelength range, and the representative weight of each healthy Raman spectrum curve.

[0104] The detection module 40 is used to determine whether the current patient is infected with Helicobacter pylori based on the degree of infection.

[0105] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the Raman spectroscopy-based gastric Helicobacter pylori infection detection system and the Raman spectroscopy-based gastric Helicobacter pylori infection detection method embodiment provided in the above embodiments belong to the same concept, and their specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0106] Example 3:

[0107] This invention also proposes a Raman spectroscopy-based detection device for Helicobacter pylori infection in the stomach. The device includes a memory and a processor. The memory stores executable program code, and the processor calls and executes this executable program code to perform a Raman spectroscopy-based detection method for Helicobacter pylori infection in the stomach provided in this application. Specifically, the device may be a chip, component, or module. The chip may include a connected processor and memory; the memory stores instructions, and when the processor calls and executes the instructions, the chip can perform the Raman spectroscopy-based detection method for Helicobacter pylori infection in the stomach provided in the above embodiment.

[0108] Furthermore, this application also protects a computer device, as shown in FIG6. The computer device includes a memory 401, a processor 402, and a computer program 403 stored in the memory 401 and running on the processor 402. When the processor 402 executes the computer program 403, the computer device can perform any of the aforementioned Raman spectroscopy-based methods for detecting Helicobacter pylori infection in the stomach.

[0109] Example 4:

[0110] This embodiment also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer, the computer executes the above-described related method steps to implement the Raman spectroscopy-based method for detecting Helicobacter pylori infection in the stomach provided in the above embodiment.

[0111] Example 5:

[0112] This embodiment also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned steps to implement the Raman spectroscopy-based method for detecting Helicobacter pylori infection in the stomach provided in the above embodiment.

[0113] In this embodiment, the device, computer-readable storage medium, computer program product, or chip are all used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0114] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0115] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for detecting Helicobacter pylori infection in the stomach based on Raman spectroscopy, characterized in that, The method includes the following steps: performing Raman spectroscopy on gastric fluid samples to obtain Raman spectral curves; constructing a healthy sample spectral library by obtaining healthy Raman spectral curves from multiple healthy gastric fluid samples; obtaining the representative weight of each healthy Raman spectral curve based on the differences in Raman intensity and waveform distribution between each healthy Raman spectral curve in the healthy sample spectral library and other healthy Raman spectral curves within a specified wavelength range; and obtaining the representative weight of the current patient's Raman spectral curve based on the similarity of Raman intensity distribution between the current patient's Raman spectral curve and each healthy Raman spectral curve in the healthy sample spectral library within a specified wavelength range, as well as the representative weight of each healthy Raman spectral curve. Infection severity; determining whether the current patient is infected with Helicobacter pylori based on the infection severity; the method for obtaining the representative weight is as follows: obtaining the first representative severity of each healthy Raman spectral curve based on the difference in Raman intensity between each healthy Raman spectral curve and other healthy Raman spectral curves within a specified wavelength range; obtaining the second representative severity of each healthy Raman spectral curve based on the similarity of waveform distribution between each healthy Raman spectral curve and other healthy Raman spectral curves within a specified wavelength range; and normalizing the product of the first and second representative severity of each healthy Raman spectral curve as the representative weight of each healthy Raman spectral curve.

2. The method for detecting Helicobacter pylori infection in the stomach based on Raman spectroscopy as described in claim 1, characterized in that, The specified wavelength range includes a first characteristic band and a second characteristic band; wherein, the wavelength range corresponding to the first characteristic band is: The wavelength range corresponding to the second characteristic band is: 。 3. The method for detecting Helicobacter pylori infection in the stomach based on Raman spectroscopy as described in claim 2, characterized in that, The method for obtaining the first representativeness is as follows: the maximum Raman intensity of each healthy Raman spectrum curve in the first characteristic band is taken as the first reference intensity; the maximum Raman intensity of each healthy Raman spectrum curve in the second characteristic band is taken as the second reference intensity; and the first representative intensity and the second representative intensity are obtained according to the distribution of the first reference intensity and the second reference intensity of each healthy Raman spectrum curve. The difference between the first reference intensity and the first representative intensity of each healthy Raman spectrum curve is taken as the first difference; The difference between the second reference intensity and the second representative intensity of each healthy Raman spectrum curve is taken as the second difference; The result of negatively correlated the mean of the first difference and the second difference of each healthy Raman spectrum curve is used as the first reference level for each healthy Raman spectrum curve; The difference between the wavelengths corresponding to the first and second reference intensities of each healthy Raman spectrum curve is used as the wavenumber distance analysis value; representative wavenumber distance analysis values ​​are obtained based on the distribution of wavenumber distance analysis values. The result of negatively correlated wavenumber distance analysis value of each healthy Raman spectrum with the difference of representative wavenumber distance analysis value is used as the second reference level for each healthy Raman spectrum. The product of the first and second reference levels of each healthy Raman spectrum is taken as the first representative level of each healthy Raman spectrum.

4. The method for detecting Helicobacter pylori infection in the stomach based on Raman spectroscopy as described in claim 3, characterized in that, The method for obtaining the first representative intensity and the second representative intensity is as follows: the first reference intensity and the second reference intensity of each healthy Raman spectrum curve are combined to construct a vector, which is used as a reference vector; based on the magnitude between the reference vectors, the reference vectors are clustered using the DBSCAN density clustering algorithm to obtain vector clusters; the mean of all first reference intensities in the largest vector cluster is used as the first representative intensity. The mean of all second reference intensities in the largest vector cluster is used as the second representative intensity.

5. The method for detecting Helicobacter pylori infection in the stomach based on Raman spectroscopy as described in claim 3, characterized in that, The method for obtaining the representative wavenumber distance analysis value is as follows: the wavenumber distance analysis values ​​of all healthy Raman spectral curves are clustered using the DBSCAN density clustering algorithm to obtain wavenumber distance clusters; the mean of all wavenumber distance analysis values ​​in the largest wavenumber distance cluster is taken as the representative wavenumber distance analysis value.

6. The method for detecting Helicobacter pylori infection in the stomach based on Raman spectroscopy as described in claim 2, characterized in that, The method for obtaining the second representativeness is as follows: The normalized paths of the a-th healthy Raman spectral curve and the b-th healthy Raman spectral curve in the first and second characteristic bands are obtained through a dynamic time warping algorithm, and both are used as analysis paths; wherein, the a-th healthy Raman spectral curve is any healthy Raman spectral curve in the healthy sample spectral library, and the b-th healthy Raman spectral curve is any healthy Raman spectral curve in the healthy sample spectral library other than the a-th healthy Raman spectral curve; based on the changes in the analysis paths, the distortion degree of the a-th healthy Raman spectral curve relative to the b-th healthy Raman spectral curve is obtained; when the... When the distortion level exceeds a preset distortion level threshold, the b-th healthy Raman spectrum curve is marked as the reference distortion curve for the a-th healthy Raman spectrum curve; the proportion of all reference distortion curves of the a-th healthy Raman spectrum curve in the healthy sample spectral library is taken as the first deviation level of the a-th healthy Raman spectrum curve; the average of the distortion levels of the a-th healthy Raman spectrum curve and all its reference distortion curves is taken as the second deviation level of the a-th healthy Raman spectrum curve; the result of negatively correlating the product of the first deviation level and the second deviation level is taken as the second representativeness of the a-th healthy Raman spectrum curve.

7. The method for detecting Helicobacter pylori infection in the stomach based on Raman spectroscopy as described in claim 6, characterized in that, The method for obtaining the degree of distortion is as follows: For any analysis path of the a-th healthy Raman spectrum curve and the b-th healthy Raman spectrum curve, the slope of the analysis path is obtained and used as a reference slope; when the absolute value of the normalized reference slope is greater than a preset deformation threshold, the corresponding local path segment is taken as a path deformation segment; the product of the length of each path deformation segment and the absolute value of the reference slope is taken as the change analysis value of each path deformation segment; the ratio of the total number of path deformation segments to the total number of local path segments corresponding to all reference slopes of the analysis path is taken as the overall degree of change of the analysis path; the result of normalizing the product of the average change analysis value of all path deformation segments and the overall degree of change is taken as the distortion analysis value of the analysis path; the average of the distortion analysis values ​​of all analysis paths of the a-th healthy Raman spectrum curve and the b-th healthy Raman spectrum curve is taken as the degree of distortion of the a-th healthy Raman spectrum curve relative to the b-th healthy Raman spectrum curve.

8. The method for detecting Helicobacter pylori infection in the stomach based on Raman spectroscopy as described in claim 1, characterized in that, The method for obtaining the degree of infection is as follows: The Raman intensities of the current patient's Raman spectrum within a specified wavelength range are arranged in wavelength order and constructed into a row vector, serving as the first vector; for any healthy Raman spectrum in the healthy sample spectral library, the Raman intensities of that healthy Raman spectrum within a specified wavelength range are arranged in wavelength order and constructed into a row vector, serving as the second vector; the magnitude of the first vector and the second vector is used as the reference infection analysis value of the current patient relative to that healthy Raman spectrum; the product of the representative weight of the healthy Raman spectrum and the reference infection analysis value is used as the corrected reference infection analysis value of the current patient relative to that healthy Raman spectrum; the result of normalizing the mean of the corrected reference infection analysis values ​​of the current patient relative to all healthy Raman spectrums is used as the degree of infection of the current patient.

9. The method for detecting Helicobacter pylori infection in the stomach based on Raman spectroscopy as described in claim 1, characterized in that, The method for determining whether a patient is infected with Helicobacter pylori based on the degree of infection is as follows: when the degree of infection is greater than or equal to a preset infection degree threshold, the patient is determined to be infected with Helicobacter pylori; when the degree of infection is less than the preset infection degree threshold, the patient is determined not to be infected with Helicobacter pylori.

Images