A method for rapid detection of hydroxyproline content in mouse lung tissue

Abstract

The present application relates to the technical field of content detection, in particular to a method for rapidly detecting the content of hydroxyproline in mouse lung tissue, which comprises the following steps: obtaining the original sampling spectrum of the lung tissue of the mouse to be detected at different preset sampling points, and obtaining a non-lung tissue interference characteristic base matrix containing whole blood characteristics and cartilage characteristics; setting an orthogonal filter operator and determining a lung interstitium characteristic spectrum vector; constructing a reference spectrum vector and determining a relative spectral response factor; constructing a spectral minimum spanning tree and determining a signal contribution weight based on the spectral minimum spanning tree and the relative spectral response factor; generating a whole lung fibrosis synthetic spectrum according to the signal contribution weight and the lung interstitium characteristic spectrum vector, and detecting the content of hydroxyproline based on the whole lung fibrosis synthetic spectrum. The present application considers the non-lung tissue interference containing whole blood characteristics and cartilage characteristics when detecting the content of hydroxyproline, which improves the rationality of the detection of the content of hydroxyproline to a certain extent.

Description

Technical Field

[0001] This invention relates to the field of content detection technology, specifically to a rapid method for detecting the content of hydroxyproline in mouse lung tissue. Background Technology

[0002] Hydroxyproline (HYP) content in whole lung tissue is widely recognized as the gold standard for assessing the severity of fibrosis due to its fixed stoichiometric relationship with total collagen. Conventional methods for determining HYP content in lung tissue primarily rely on classic alkaline hydrolysis-chemical colorimetric methods (such as the chloramine-T method). This method often has significant limitations: First, it is a destructive test, requiring physical homogenization and high-temperature hydrolysis of the entire lung tissue, often rendering the same valuable sample unusable for subsequent histopathological sections or molecular biological analyses, thus limiting the multidimensional mining of experimental data; second, this method is cumbersome and time-consuming, often failing to meet the demands of high-throughput drug screening.

[0003] In recent years, near-infrared spectroscopy has been introduced into biological tissue analysis due to its non-destructive and rapid characteristics. However, in actual testing of isolated mouse lung tissue, the following technical problems often arise:

[0004] During the dissection process, it is often difficult to completely remove the trace amounts of whole blood remaining on the lung surface and the tracheal cartilage at the hilum. These non-lung matrix components often have strong non-specific absorption in the near-infrared band, and their signals often severely obscure the weak spectral characteristics of pulmonary interstitial fibrosis, resulting in a low signal-to-noise ratio and thus poor reliability of hydroxyproline content detection. Summary of the Invention

[0005] To address the technical problem of poor reliability in hydroxyproline content detection, this invention proposes a rapid method for detecting hydroxyproline content in mouse lung tissue.

[0006] In a first aspect, the present invention provides a rapid method for detecting the hydroxyproline content in mouse lung tissue, the method comprising:

[0007] The original sampling spectra of the lung tissue of the mice to be tested at different preset sampling points were obtained, and the non-lung tissue interference feature basis matrix containing whole blood features and cartilage features was obtained.

[0008] Based on the non-lung tissue interference feature basis matrix, an orthogonal filtering operator is set, and based on the orthogonal filtering operator and the original sampling spectrum at each preset sampling point, the lung interstitial feature spectral vector corresponding to each preset sampling point is determined.

[0009] Based on all the spectral vectors of lung interstitial features, a baseline spectral vector is constructed. Based on the univariate linear regression fitting of the baseline spectral vector and the spectral vectors of lung interstitial features corresponding to each preset sampling point in the collagen-sensitive band, the relative spectral response factor corresponding to each preset sampling point is determined.

[0010] Based on the differences between the spectral vectors of lung interstitial features corresponding to different preset sampling points, a minimum spanning tree of spectra is constructed. Based on the minimum spanning tree of spectra and the relative spectral response factor corresponding to each preset sampling point, the signal contribution weight corresponding to each preset sampling point is determined.

[0011] Based on the signal contribution weights and lung interstitial characteristic spectral vectors corresponding to all preset sampling points, a whole-pulmonary fibrosis synthetic spectrum is generated, and the hydroxyproline content is detected based on the whole-pulmonary fibrosis synthetic spectrum.

[0012] In conjunction with the first aspect above, in one possible implementation, obtaining the non-lung tissue interference feature base matrix containing whole blood features and cartilage features includes:

[0013] Obtain standard whole blood spectra containing whole blood characteristics and standard cartilage spectra containing cartilage characteristics;

[0014] Normalize the transposes of standard whole blood spectra and standard cartilage spectra to obtain the normalized vectors of standard whole blood and standard cartilage.

[0015] The standard whole blood normalized vector and the standard cartilage normalized vector are used as column vectors to form the non-lung tissue interference feature basis matrix.

[0016] In conjunction with the first aspect above, in one possible implementation, setting the orthogonal filtering operator based on the non-lung tissue interference feature basis matrix includes:

[0017] The orthogonal filtering operator is determined based on the preset identity matrix, the non-lung tissue interference feature basis matrix, the transpose of the non-lung tissue interference feature basis matrix, and the inverse of the product between the non-lung tissue interference feature basis matrix and its transpose.

[0018] In conjunction with the first aspect above, in one possible implementation, determining the lung interstitial feature spectral vector corresponding to each preset sampling point based on the orthogonal filtering operator and the original sampled spectrum at each preset sampling point includes:

[0019] The product of the original sampled spectrum at each preset sampling point and the orthogonal filtering operator is used to determine the lung interstitial characteristic spectral vector corresponding to each preset sampling point.

[0020] In conjunction with the first aspect above, in one possible implementation, constructing a baseline spectral vector based on all lung interstitial characteristic spectral vectors includes:

[0021] The modulus of each lung interstitial characteristic spectral vector is determined as the reference intensity factor corresponding to each lung interstitial characteristic spectral vector;

[0022] From all lung interstitial characteristic spectral vectors, select the preset number of lung interstitial characteristic spectral vectors with the largest corresponding reference intensity factors, and use them as reference spectral vectors;

[0023] The mean of all reference spectral vectors is used to determine the baseline spectral vector.

[0024] In conjunction with the first aspect above, in one possible implementation, determining the relative spectral response factor corresponding to each preset sampling point based on the univariate linear regression fitting of the reference spectral vector and the pulmonary interstitial characteristic spectral vector corresponding to each preset sampling point within the collagen-sensitive band includes:

[0025] Any preset sampling point is determined as a labeled sampling point, and the absorbance of the lung interstitial characteristic spectral vector corresponding to the labeled sampling point in the collagen sensitive band is used to form a labeled absorbance sequence.

[0026] The absorbance of the reference spectral vector in the collagen-sensitive band is used to construct a reference absorbance sequence;

[0027] Using the absorbance in the baseline absorbance sequence as the independent variable and the absorbance in the labeled absorbance sequence as the dependent variable, a univariate linear regression is performed, and the slope of the obtained univariate linear regression function is determined as the relative spectral response factor corresponding to the labeled sampling point.

[0028] In conjunction with the first aspect above, in one possible implementation, the step of constructing a minimum spanning tree based on the differences between the spectral vectors of lung interstitial features corresponding to different preset sampling points includes:

[0029] The Euclidean distance between the spectral vectors of lung interstitial features corresponding to every two preset sampling points is determined as the target distance between every two preset sampling points;

[0030] Using the target distance between different preset sampling points as weights and the preset sampling points as nodes, a minimum spanning tree is constructed as the spectral minimum spanning tree.

[0031] In conjunction with the first aspect above, in one possible implementation, determining the signal contribution weight corresponding to each preset sampling point based on the minimum spanning tree of the spectrum and the relative spectral response factor corresponding to each preset sampling point includes:

[0032] The heterogeneity of biochemical characteristics is determined based on the mean and standard deviation of all edge weights in the minimum spanning tree of the spectrum.

[0033] Any preset sampling point is designated as a marked sampling point. Based on the pre-acquired local response cutoff constant and sensitivity coefficient, the biochemical feature heterogeneity, the relative spectral response factor corresponding to the marked sampling point, the mean of all edge weights in the minimum spanning tree of the spectrum, and the mean of the edge weights of the edges connected to the marked sampling point in the minimum spanning tree of the spectrum, the signal contribution weight corresponding to the marked sampling point is determined.

[0034] In conjunction with the first aspect above, in one possible implementation, generating a synthetic spectrum of whole-pulmonary fibrosis based on the signal contribution weights corresponding to all preset sampling points and the pulmonary interstitial characteristic spectral vector includes:

[0035] The sum of the products between the signal contribution weights corresponding to all preset sampling points and the spectral vectors of lung interstitial characteristics is determined as the synthetic spectrum of whole lung fibrosis.

[0036] In conjunction with the first aspect above, in one possible implementation, the detection of hydroxyproline content based on the synthetic spectrum of whole-pulmonary fibrosis includes:

[0037] Based on the pre-obtained content regression coefficient matrix and model intercept term, as well as the whole lung fibrosis synthesis spectrum, the total hydroxyproline content of the mice to be tested was determined.

[0038] Secondly, the present invention provides a rapid detection system for hydroxyproline content in mouse lung tissue, the system comprising:

[0039] The data acquisition module is used to acquire the original sampling spectra of the lung tissue of the mouse under test at different preset sampling points, and to acquire the non-lung tissue interference feature basis matrix containing whole blood features and cartilage features;

[0040] The setting and determination module is used to set the orthogonal filtering operator based on the non-lung tissue interference feature basis matrix, and to determine the lung interstitial feature spectral vector corresponding to each preset sampling point based on the orthogonal filtering operator and the original sampling spectrum at each preset sampling point.

[0041] The vector construction and factor determination module is used to construct a baseline spectral vector based on all lung interstitial characteristic spectral vectors, and determine the relative spectral response factor corresponding to each preset sampling point based on the univariate linear regression fitting of the baseline spectral vector and the lung interstitial characteristic spectral vector corresponding to each preset sampling point in the collagen sensitive band.

[0042] The tree construction and weight determination module is used to construct a minimum spanning tree based on the differences between the spectral vectors of lung interstitial features corresponding to different preset sampling points, and to determine the signal contribution weight corresponding to each preset sampling point based on the minimum spanning tree and the relative spectral response factor corresponding to each preset sampling point.

[0043] The spectrum generation and content detection module is used to generate a whole-pulmonary fibrosis synthetic spectrum based on the signal contribution weights and lung interstitial characteristic spectral vectors corresponding to all preset sampling points, and to detect the hydroxyproline content based on the whole-pulmonary fibrosis synthetic spectrum.

[0044] Thirdly, a server is provided, including a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, causing the device to perform the methods of the first aspect or any possible implementation thereof.

[0045] Fourthly, a computer program product is provided, comprising: computer program code, which, when run on a computer, causes the computer to perform the methods described in the first aspect or any possible implementation thereof.

[0046] Fifthly, a computer-readable storage medium is provided that stores computer program code, which, when executed on a computer, causes the computer to perform the methods described in the first aspect or any possible implementation thereof.

[0047] The present invention has the following beneficial effects:

[0048] This invention provides a rapid detection method for hydroxyproline content in mouse lung tissue. This method considers non-lung tissue interference, including whole blood and cartilage features, during hydroxyproline content detection, thus addressing the technical problem of poor accuracy in hydroxyproline content detection and improving its accuracy to a certain extent. Specifically, based on a pre-acquired feature matrix containing non-lung tissue interference features (including whole blood and cartilage features), this invention sets an orthogonal filtering operator. Based on the orthogonal filtering operator and the original sampling spectrum at each preset sampling point, it quantifies the lung interstitial feature spectral vector corresponding to each preset sampling point, excluding non-lung tissue interference. Combined with the baseline spectral vector, it quantifies the relative spectral response factor corresponding to each preset sampling point. By analyzing the lung interstitial feature spectral vector and the relative spectral response factor, it quantifies the signal contribution weight corresponding to each preset sampling point, thereby generating a whole-lung fibrosis synthetic spectrum. Hydroxyproline content detection is then performed based on this whole-lung fibrosis synthetic spectrum, thus improving the accuracy of hydroxyproline content detection to a certain extent. Attached Figure Description

[0049] 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.

[0050] Figure 1 This is a flowchart of a rapid detection method for hydroxyproline content in mouse lung tissue according to the present invention;

[0051] Figure 2 This is a schematic diagram of the composition and structure of a rapid detection system for hydroxyproline content in mouse lung tissue according to the present invention;

[0052] Figure 3 This is a schematic diagram of the structure of a computer device according to the present invention. Detailed Implementation

[0053] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the specific implementation methods, structures, features, and effects of the technical solution proposed according to the present invention are described in detail below with reference to the accompanying drawings and preferred embodiments. 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.

[0054] 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.

[0055] In the fields of drug development and toxicology assessment, mouse pulmonary fibrosis models (such as silica-induced silicosis models) are key experimental platforms for studying the mechanisms of interstitial lung disease and screening anti-fibrotic drugs. Among them, the content of hydroxyproline (HYP) in whole lung tissue is recognized as the gold standard indicator for assessing the severity of fibrosis because it has a fixed stoichiometric relationship with the total amount of collagen.

[0056] refer to Figure 1 This paper illustrates the flowchart of a rapid detection method for hydroxyproline content in mouse lung tissue according to the present invention. The rapid detection method for hydroxyproline content in mouse lung tissue includes the following steps:

[0057] Step S1: Obtain the original sampling spectrum of the lung tissue of the mouse to be tested at different preset sampling points, and obtain the non-lung tissue interference feature base matrix containing whole blood features and cartilage features.

[0058] The mice to be tested can be laboratory mice used for detecting hydroxyproline content in lung tissue. Laboratory mice, or simply lab mice, typically weigh between 18 and 35 grams. Preset sampling points can be pre-set based on actual conditions. The number of pre-set sampling points can be pre-set based on actual conditions, and can be greater than or equal to 10, for example, 20. The areas covered by all pre-set sampling points can include the apex, hilum, base, and lateral margin of the lung, and the pre-set sampling points can be evenly distributed.

[0059] As an example, this step may include the following steps:

[0060] The first step is to obtain the original sampling spectra of the lung tissue of the mice to be tested at different preset sampling points.

[0061] For example, a spectrometer can be used to record the diffuse reflectance spectrum at each preset sampling point, and the diffuse reflectance spectrum at each preset sampling point can be negatively logarithmically transformed to obtain the absorbance spectrum, which is denoted as the original sampling spectrum.

[0062] It should be noted that the original sampled spectrum can be a The row vector contains signals from the lung interstitium, as well as interference signals from residual blood and cartilage that are inevitably mixed in; This is the total number of wavelength channels in the spectrometer.

[0063] It should be noted that the fresh, isolated mouse tissue to be tested, specifically the left lung of the mouse, can be laid flat on a petri dish, keeping the tissue naturally expanded. To ensure spatial representativeness of the sampling and support subsequent topological analysis, a fiber optic probe can be used to select multiple different anatomical sites on the lung tissue surface for contact sampling. These anatomical sites are also known as pre-defined sampling points.

[0064] The second step is to obtain standard whole blood spectra containing whole blood characteristics and standard cartilage spectra containing cartilage characteristics.

[0065] For example, pure whole blood samples and stripped pure tracheal cartilage samples from mice can be prepared separately; at least 50 whole blood spectra and cartilage spectra can be collected using a near-infrared spectrometer in the 900 nm to 1700 nm wavelength range; the collected spectral data can be arithmetically averaged to obtain the average whole blood spectrum and the average cartilage spectrum, and the average whole blood spectrum and the average cartilage spectrum can be denoted as the standard whole blood spectrum and the standard cartilage spectrum, respectively.

[0066] Alternatively, the standard whole blood spectrum and standard cartilage spectrum can be obtained by directly retrieving characteristic absorption spectral data of mouse whole blood and tracheal cartilage from a standard biological tissue spectral database, and recording them as standard whole blood spectrum and standard cartilage spectrum, respectively.

[0067] The third step is to normalize the transpose of the standard whole blood spectrum and the transpose of the standard cartilage spectrum to obtain the standard whole blood normalization vector and the standard cartilage normalization vector.

[0068] The standard whole blood normalization vector can be a unit vector of the transpose of the standard whole blood spectrum. The standard cartilage normalization vector can be a unit vector of the transpose of the standard cartilage spectrum.

[0069] The fourth step is to use the standard whole blood normalized vector and the standard cartilage normalized vector as column vectors to form the non-lung tissue interference feature basis matrix.

[0070] Among them, both the standard whole blood normalized vector and the standard cartilage normalized vector can be... column vectors, This represents the total number of wavelength channels in the spectrometer. The non-lung tissue interference characteristic basis matrix can be... The matrix.

[0071] It should be noted that the non-lung tissue interference feature basis matrix can characterize the subspace where the interference signal that needs to be removed is located.

[0072] Step S2: Based on the non-lung tissue interference feature basis matrix, set an orthogonal filtering operator, and determine the lung interstitial feature spectral vector corresponding to each preset sampling point based on the orthogonal filtering operator and the original sampling spectrum at each preset sampling point.

[0073] Among them, the orthogonal filtering operator can be a The projection matrix can be used to project any vector onto the orthogonal complement space of the interference feature subspace. This is the total number of wavelength channels in the spectrometer.

[0074] As an example, this step may include the following steps:

[0075] The first step is to determine the orthogonal filtering operator based on the preset identity matrix, the non-lung tissue interference feature basis matrix, the transpose of the non-lung tissue interference feature basis matrix, and the inverse of the product between the non-lung tissue interference feature basis matrix and its transpose.

[0076] The preset identity matrix can be a pre-set identity matrix based on actual conditions, and it can be... The identity matrix, This is the total number of wavelength channels in the spectrometer.

[0077] For example, the formula for determining the orthogonal filtering operator can be:

[0078] ;

[0079] Where P is the orthogonal filtering operator. A is the preset identity matrix. M is the non-lung tissue interference feature basis matrix. It is the transpose of M. yes The reverse.

[0080] It should be noted that constructing an orthogonal filtering operator is equivalent to constructing a filter, which can suppress signal components parallel to whole blood and cartilage features to a certain extent, while allowing signal components perpendicular to them to pass through.

[0081] The second step is to multiply the original sampled spectrum at each preset sampling point with the above orthogonal filtering operator to determine the lung interstitial characteristic spectral vector corresponding to each preset sampling point.

[0082] For example, the formula for determining the spectral vector of lung interstitial features corresponding to a preset sampling point can be:

[0083] ;

[0084] in, It is the spectral vector of lung interstitial features corresponding to the i-th preset sampling point. i is the index of the preset sampling point. It is the original sampled spectrum at the i-th preset sampling point. P is the orthogonal filtering operator.

[0085] It should be noted that, The residual vector represents the suppression of non-specific absorption caused by blood and cartilage in the original sampling spectrum. This residual vector is often the pulmonary interstitial characteristic spectral vector corresponding to the i-th preset sampling point. This can mean that background noise has been removed, and only spectral information characterizing the structure and chemical composition of lung tissue has been retained.

[0086] Step S3: Construct a baseline spectral vector based on all lung interstitial characteristic spectral vectors, and determine the relative spectral response factor corresponding to each preset sampling point based on the univariate linear regression fitting of the baseline spectral vector and the lung interstitial characteristic spectral vector corresponding to each preset sampling point in the collagen sensitive band.

[0087] Among them, the collagen-sensitive wavelength range can be 1100-1350 nanometers.

[0088] As an example, this step may include the following steps:

[0089] The first step is to determine the modulus of each lung interstitial characteristic spectral vector as the reference intensity factor corresponding to each lung interstitial characteristic spectral vector.

[0090] The second step is to select a predetermined number of lung interstitial characteristic spectral vectors with the largest corresponding reference intensity factors from all lung interstitial characteristic spectral vectors, and use them as reference spectral vectors.

[0091] The preset quantity can be a quantity of at least 1 that is pre-set according to the actual situation. For example, it can be the integer part of 10% of the number of lung interstitial characteristic spectral vectors.

[0092] The third step is to determine the mean of all reference spectral vectors as the baseline spectral vector.

[0093] In practice, due to natural differences in lung tissue thickness and air content among different mouse individuals, directly comparing absolute absorbance values ​​may introduce systematic errors. Therefore, a relative reference system within the sample, namely the baseline spectral vector, can be established, and the relative spectral response factor of each sampling point can be extracted based on this reference system to quantify the signal characteristics of local tissues.

[0094] It should be noted that the baseline spectral vector can represent the upper limit reference point of the signal intensity inside the sample. Regardless of the specific biological state corresponding to this point, it often provides a unified maximum value scale for subsequent analysis.

[0095] The fourth step is to determine any preset sampling point as a labeled sampling point, and to construct a labeled absorbance sequence by measuring the absorbance of the lung interstitial characteristic spectral vector corresponding to the labeled sampling point in the collagen-sensitive band.

[0096] The fifth step is to construct a baseline absorbance sequence by taking the absorbance of the above-mentioned baseline spectral vector in the collagen-sensitive band.

[0097] The sixth step involves performing a univariate linear regression fitting with the absorbance in the above-mentioned baseline absorbance sequence as the independent variable and the absorbance in the above-mentioned labeled absorbance sequence as the dependent variable. The slope of the obtained univariate linear fitting function is then determined as the relative spectral response factor corresponding to the above-mentioned labeled sampling points. The univariate linear fitting function obtained at this time can characterize the "best straight line" that best represents the linear relationship between the two variables.

[0098] In practice, after establishing a benchmark, it is often necessary to quantify the signal attenuation of each of the other sampling points relative to the benchmark, i.e., the relative spectral response factor. This relative attenuation difference is often directly related to the microstructural changes of local tissues (such as changes in light scattering caused by fibrosis deposition).

[0099] It should be noted that when the relative spectral response factor is close to 1, it often indicates that the spectral response intensity of the sampling point is similar to that of the reference point, and it often belongs to the high-response region within the sample. When the relative spectral response factor is significantly less than 1, it often indicates that there is significant signal attenuation at the sampling point relative to the reference point. In the silicosis model, the more severe the fibrosis in a region, the lower its spectral response factor tends to be (i.e., the more severe the attenuation relative to the reference).

[0100] Step S4: Based on the differences between the spectral vectors of lung interstitial features corresponding to different preset sampling points, construct a minimum spanning tree of spectra, and determine the signal contribution weight corresponding to each preset sampling point based on the minimum spanning tree of spectra and the relative spectral response factor corresponding to each preset sampling point.

[0101] As an example, this step may include the following steps:

[0102] The first step is to determine the Euclidean distance between the spectral vectors of lung interstitial features corresponding to every two preset sampling points as the target distance between every two preset sampling points.

[0103] It should be noted that the smaller the target distance between two preset sampling points, the more similar the chemical composition and tissue structure of the two preset sampling points tend to be.

[0104] The second step is to construct a minimum spanning tree using the target distance between different preset sampling points as weights and the preset sampling points as nodes, which serves as the minimum spanning tree for the spectrum.

[0105] In the minimum spanning tree, each preset sampling point has at least one connecting edge. The minimum spanning tree (MST) is typically a tree that contains all vertices in a weighted connected undirected graph, and the sum of the weights of all edges in the tree is minimized.

[0106] It should be noted that the minimum spanning tree of the spectrum not only connects all the preset sampling points, but its connection path tends to prioritize connecting similar points (short sides). Long sides are usually retained only when connections are needed between different clusters (such as highly heterogeneous feature clusters and normal clusters).

[0107] The third step is to determine the heterogeneity of biochemical features based on the mean and standard deviation of all edge weights in the aforementioned minimum spanning tree of the spectrum.

[0108] It should be noted that silicosis lesions often exhibit spatial heterogeneity. Graph theory algorithms can be used to analyze the distribution patterns of sampling points in the feature space, aiming to extract a global index that can characterize the differential topology of biochemical components, namely, the degree of biochemical heterogeneity.

[0109] For example, the formula for determining the heterogeneity of biochemical characteristics can be:

[0110] ;

[0111] Where B is the degree of heterogeneity of biochemical characteristics. It is the standard deviation of the weights of all edges in the minimum spanning tree of the spectrum. It is the average of all edge weights in the minimum spanning tree of the spectrum. It is an adjustment factor that is preset according to the actual situation, mainly used to prevent the denominator from being 0. For example, it can be 0.0000001.

[0112] It should be noted that all variables in the embodiments of the present invention, especially the denominator, can be set with corresponding adjustment factors according to the actual situation to adjust their value range or prevent the denominator from being 0.

[0113] It should be noted that a low degree of heterogeneity in biochemical characteristics often indicates a uniform distribution of edge weights and consistent differences among points within the sample, suggesting a uniform distribution of lesions (or normal features). Conversely, a high degree of heterogeneity in biochemical characteristics often indicates the presence of both extremely short edges (intra-cluster connections) and extremely long edges (inter-cluster connections) within the edge weight set, suggesting significant clustering within the sample and potential variations in biochemical characteristics.

[0114] The fourth step is to determine any preset sampling point as a marked sampling point, and to determine the signal contribution weight corresponding to the marked sampling point based on the pre-acquired local response cutoff constant and sensitivity coefficient, the above-mentioned biochemical characteristic heterogeneity, the relative spectral response factor corresponding to the marked sampling point, the mean of all edge weights in the above-mentioned minimum spanning tree of the spectrum, and the mean of the edge weights of the edges connected to the marked sampling point in the above-mentioned minimum spanning tree of the spectrum.

[0115] The local response cutoff constant can be preset based on actual conditions and can be used to prevent local weights from amplifying infinitely and to cut off extremely low response points. The multiplier tends to 0 to ensure computational stability, and its value range can be [0.01, 0.2], for example, it can be 0.1. The sensitivity coefficient, also known as the global heterogeneity sensitivity coefficient, can be preset according to the actual situation and can be used to adjust the degree of participation of spatial topology in the total weight. Its value range can be [1, 3], for example, it can be 1.5.

[0116] It should be noted that the parameters in the embodiments of the present invention, such as the local response cutoff constant and the sensitivity coefficient, can be calibrated using a grid search strategy.

[0117] For example, the formula for determining the signal contribution weight corresponding to the marked sampling point can be:

[0118] ;

[0119] ;

[0120] in, is the signal contribution weight corresponding to the marked sampling point. w is the initial contribution weight corresponding to the marked sampling point. It is the sum of the initial contribution weights corresponding to all preset sampling points. It is a function that takes the maximum value. It is an absolute value function. It is the relative spectral response factor corresponding to the marked sampling point. It is the local response cutoff constant. is the sensitivity coefficient. B is the biochemical characteristic heterogeneity. d is the mean of the edge weights of the edges connecting the marked sampling points in the minimum spanning tree of the spectrum. It is the average of all edge weights in the minimum spanning tree of the spectrum. It is an adjustment factor that is preset according to the actual situation, mainly used to prevent the denominator from being 0. For example, it can be 0.0000001.

[0121] It should be noted that, This represents a local response correction term. This represents the global heterogeneity correction term. The absolute value of the relative spectral response factor for each sampling point is used to prevent calculation errors caused by negative values ​​due to baseline drift. The smaller the value, the more severe the attenuation relative to the reference point, which typically corresponds to the high response attenuation region. The larger the value, the higher the weight that can be assigned to the labeled sampling points. This directly amplifies the signal contribution of low-response (high-density) regions, offsetting the dilution effect of normal tissue. Its function is to set a lower limit to prevent when When the minimum approaches zero, The tendency towards infinity leads to numerical overflow, ensuring computational stability. When the lung distribution is uniform (B is small), the global heterogeneity correction term tends to approach 1, and the weights are mainly determined by the local term. When there are significant focal lesions in the lung (B is large), the global heterogeneity correction term tends to be significantly greater than 1, amplifying the weights. Furthermore, if the d of the labeled sampling point is large, it often indicates that the point differs greatly from other points in the feature space, typically being an edge point of a high-heterogeneity feature cluster or an isolated heterogeneous feature point; the global heterogeneity correction term tends to increase further. Thus, when the non-uniform distribution of the samples is confirmed, additional enhancement of those unique signal points can, to some extent, avoid overweighting in uniform samples.

[0122] Step S5: Generate a whole-pulmonary fibrosis synthetic spectrum based on the signal contribution weights and pulmonary interstitial characteristic spectral vectors corresponding to all preset sampling points, and detect the hydroxyproline content based on the whole-pulmonary fibrosis synthetic spectrum.

[0123] As an example, this step may include the following steps:

[0124] The first step is to determine the sum of the products between the signal contribution weights corresponding to all preset sampling points and the spectral vectors of lung interstitial characteristics as the synthetic spectrum of whole lung fibrosis corresponding to the mouse to be tested.

[0125] For example, the formula for determining the synthetic spectrum of total pulmonary fibrosis can be:

[0126] ;

[0127] Where E is the synthetic spectrum of whole lung fibrosis. N is the number of preset sampling points. i is the sequence number of the preset sampling points. It is the signal contribution weight corresponding to the i-th preset sampling point. It is the spectral vector of lung interstitial features corresponding to the i-th preset sampling point.

[0128] It should be noted that the synthetic spectrum of whole-pulmonary fibrosis is a... The vector often simulates the ideal spectrum measured after physical homogenization of whole lung tissue. It is virtually implemented through algorithms. It often aggregates high-density biochemical component characteristics across the entire lung and suppresses normal background noise, which can be used for subsequent content detection.

[0129] The second step involves determining the total hydroxyproline content of the mice to be tested based on the pre-obtained content regression coefficient matrix and model intercept term, as well as the whole lung fibrosis synthesis spectrum.

[0130] Among them, the content regression coefficient matrix, also known as the content mapping slope matrix, can be... The column vector can contain the characteristic contribution weights for each wavelength point, and is mainly used to map the absorbance values ​​of the synthesized spectrum to the hydroxyproline content through linear multiplication. This represents the total number of wavelength channels in the spectrometer. The model intercept term can represent the background value of system baseline shift and non-specific absorption, and is mainly used to correct zero-point shift and improve the accuracy of quantitative prediction of absolute content.

[0131] For example, the method for obtaining the content regression coefficient matrix and the model intercept term can be as follows: Obtain the whole-pulmonary fibrosis synthesis spectrum corresponding to each historical mouse; use the whole-pulmonary fibrosis synthesis spectrum corresponding to the historical mouse as the independent variable and the actual total amount of hydroxyproline in the lung tissue of the historical mouse as the dependent variable to perform multiple linear regression fitting, obtain the multiple linear regression function, and construct the content regression coefficient matrix from all the slopes of this multiple linear regression function; the intercept of this multiple linear regression function is denoted as the model intercept term. Here, the historical mouse can be a mouse with a known actual total amount of hydroxyproline in its lung tissue and the same model as the mouse to be tested. The method for obtaining the whole-pulmonary fibrosis synthesis spectrum corresponding to the historical mouse can be the same as the method for obtaining the whole-pulmonary fibrosis synthesis spectrum corresponding to the mouse to be tested, and will not be elaborated here.

[0132] For example, the formula for determining the total hydroxyproline content of the mouse being tested can be:

[0133] ;

[0134] Where y is the total hydroxyproline content measured in the mouse being tested. E is the synthesis spectrum of whole-pulmonary fibrosis. This is the content regression coefficient matrix. H is the model intercept term.

[0135] Optionally, in antifibrotic drug screening experiments, if y is significantly lower than that of the model group and close to that of the blank control group, quantitative analysis often indicates that the drug effectively inhibits collagen deposition. Since this detection method does not cause physical or chemical damage to the lung tissue of the mice being tested, after obtaining y, researchers can fix the isolated lung tissue in formalin or freeze it in liquid nitrogen for subsequent histopathological section staining (such as Masson staining to verify fibrosis distribution) or molecular biological detection (such as extracting RNA to determine gene expression). This multi-purpose nature of using a single mouse significantly improves the data output value of a single experimental animal.

[0136] refer to Figure 2 Based on the same inventive concept as the above-described method embodiments, this invention provides a rapid detection system for hydroxyproline content in mouse lung tissue. The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When executed by the processor, the computer program implements the steps of a rapid detection method for hydroxyproline content in mouse lung tissue, specifically including:

[0137] The data acquisition module 201 is used to acquire the original sampling spectrum of the lung tissue of the mouse to be tested at different preset sampling points, and to acquire the non-lung tissue interference feature basis matrix containing whole blood features and cartilage features;

[0138] The setting and determining module 202 is used to set an orthogonal filtering operator based on the non-lung tissue interference feature basis matrix, and to determine the lung interstitial feature spectral vector corresponding to each preset sampling point based on the orthogonal filtering operator and the original sampling spectrum at each preset sampling point.

[0139] The vector construction and factor determination module 203 is used to construct a benchmark spectral vector based on all lung interstitial characteristic spectral vectors, and determine the relative spectral response factor corresponding to each preset sampling point based on the univariate linear regression fitting of the benchmark spectral vector and the lung interstitial characteristic spectral vector corresponding to each preset sampling point in the collagen sensitive band.

[0140] The tree construction and weight determination module 204 is used to construct a minimum spanning tree of spectra based on the differences between the spectral vectors of lung interstitial features corresponding to different preset sampling points, and to determine the signal contribution weight corresponding to each preset sampling point based on the minimum spanning tree of spectra and the relative spectral response factor corresponding to each preset sampling point.

[0141] The spectrum generation and content detection module 205 is used to generate a whole-pulmonary fibrosis synthetic spectrum based on the signal contribution weights and lung interstitial characteristic spectral vectors corresponding to all preset sampling points, and to detect the hydroxyproline content based on the whole-pulmonary fibrosis synthetic spectrum.

[0142] Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of the present invention. For example, as shown... Figure 3 As shown, the computer device 300 includes: a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302, wherein when the processor 302 executes the computer program 303, the computer device can execute any of the aforementioned rapid detection methods for hydroxyproline content in mouse lung tissue.

[0143] Based on the same inventive concept as the above-described method embodiments, the present invention provides a server, including a memory and a processor. The memory stores executable program code, and the processor retrieves and runs the executable program code from the memory, enabling the device to execute any of the above-described rapid detection methods for hydroxyproline content in mouse lung tissue.

[0144] Based on the same inventive concept as the above-described method embodiments, the present invention provides a computer program product comprising: computer program code, which, when run on a computer, causes the computer to execute any of the above-described methods for rapid detection of hydroxyproline content in mouse lung tissue.

[0145] Based on the same inventive concept as the above-described method embodiments, the present invention provides a computer-readable storage medium storing computer program code, which, when executed on a computer, causes the computer to perform any of the above-described rapid detection methods for hydroxyproline content in mouse lung tissue.

[0146] In summary, this invention, based on a pre-acquired non-lung tissue interference feature basis matrix containing whole blood and cartilage features, sets an orthogonal filtering operator. Based on the orthogonal filtering operator and the original sampling spectrum at each preset sampling point, it quantifies the lung interstitial feature spectral vector corresponding to each preset sampling point, excluding non-lung tissue interference. Combined with the benchmark spectral vector, it quantifies the relative spectral response factor corresponding to each preset sampling point. By analyzing the lung interstitial feature spectral vector and the relative spectral response factor, it quantifies the signal contribution weight corresponding to each preset sampling point, thereby generating a whole-lung fibrosis synthetic spectrum. Hydroxyproline content detection is then performed based on this whole-lung fibrosis synthetic spectrum, which to some extent improves the rationality of hydroxyproline content detection.

[0147] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A rapid method for detecting hydroxyproline content in mouse lung tissue, characterized in that, Includes the following steps: The original sampling spectra of the lung tissue of the mice to be tested at different preset sampling points were obtained, and the non-lung tissue interference feature basis matrix containing whole blood features and cartilage features was obtained. Based on the non-lung tissue interference feature basis matrix, an orthogonal filtering operator is set, and based on the orthogonal filtering operator and the original sampling spectrum at each preset sampling point, the lung interstitial feature spectral vector corresponding to each preset sampling point is determined. Based on all the spectral vectors of lung interstitial features, a baseline spectral vector is constructed. Based on the univariate linear regression fitting of the baseline spectral vector and the spectral vectors of lung interstitial features corresponding to each preset sampling point in the collagen-sensitive band, the relative spectral response factor corresponding to each preset sampling point is determined. Based on the differences between the spectral vectors of lung interstitial features corresponding to different preset sampling points, a minimum spanning tree of spectra is constructed. Based on the minimum spanning tree of spectra and the relative spectral response factor corresponding to each preset sampling point, the signal contribution weight corresponding to each preset sampling point is determined. Based on the signal contribution weights and lung interstitial characteristic spectral vectors corresponding to all preset sampling points, a whole-pulmonary fibrosis synthetic spectrum is generated, and the hydroxyproline content is detected based on the whole-pulmonary fibrosis synthetic spectrum.

2. The rapid detection method for hydroxyproline content in mouse lung tissue according to claim 1, characterized in that, The acquisition of the non-lung tissue interference feature base matrix containing whole blood features and cartilage features includes: Obtain standard whole blood spectra containing whole blood characteristics and standard cartilage spectra containing cartilage characteristics; Normalize the transposes of standard whole blood spectra and standard cartilage spectra to obtain the normalized vectors of standard whole blood and standard cartilage. The standard whole blood normalized vector and the standard cartilage normalized vector are used as column vectors to form the non-lung tissue interference feature basis matrix.

3. The rapid detection method for hydroxyproline content in mouse lung tissue according to claim 1, characterized in that, The step of setting an orthogonal filtering operator based on the non-lung tissue interference feature basis matrix includes: The orthogonal filtering operator is determined based on the preset identity matrix, the non-lung tissue interference feature basis matrix, the transpose of the non-lung tissue interference feature basis matrix, and the inverse of the product between the non-lung tissue interference feature basis matrix and its transpose.

4. The rapid detection method for hydroxyproline content in mouse lung tissue according to claim 1, characterized in that, The step of determining the lung interstitial characteristic spectral vector corresponding to each preset sampling point based on the orthogonal filtering operator and the original sampling spectrum at each preset sampling point includes: The product of the original sampled spectrum at each preset sampling point and the orthogonal filtering operator is used to determine the lung interstitial characteristic spectral vector corresponding to each preset sampling point.

5. The rapid detection method for hydroxyproline content in mouse lung tissue according to claim 1, characterized in that, The construction of a baseline spectral vector based on all lung interstitial characteristic spectral vectors includes: The modulus of each lung interstitial characteristic spectral vector is determined as the reference intensity factor corresponding to each lung interstitial characteristic spectral vector; From all lung interstitial characteristic spectral vectors, select the preset number of lung interstitial characteristic spectral vectors with the largest corresponding reference intensity factors, and use them as reference spectral vectors; The mean of all reference spectral vectors is used to determine the baseline spectral vector.

6. The rapid detection method for hydroxyproline content in mouse lung tissue according to claim 1, characterized in that, The step of determining the relative spectral response factor corresponding to each preset sampling point based on the univariate linear regression fitting of the baseline spectral vector and the spectral vector of lung interstitial characteristics corresponding to each preset sampling point within the collagen-sensitive band includes: Any preset sampling point is determined as a labeled sampling point, and the absorbance of the lung interstitial characteristic spectral vector corresponding to the labeled sampling point in the collagen sensitive band is used to form a labeled absorbance sequence. The absorbance of the reference spectral vector in the collagen-sensitive band is used to construct a reference absorbance sequence; Using the absorbance in the baseline absorbance sequence as the independent variable and the absorbance in the labeled absorbance sequence as the dependent variable, a univariate linear regression is performed, and the slope of the obtained univariate linear regression function is determined as the relative spectral response factor corresponding to the labeled sampling point.

7. The rapid detection method for hydroxyproline content in mouse lung tissue according to claim 1, characterized in that, The step of constructing a minimum spanning tree of spectra based on the differences between the spectral vectors of lung interstitial features corresponding to different preset sampling points includes: The Euclidean distance between the spectral vectors of lung interstitial features corresponding to every two preset sampling points is determined as the target distance between every two preset sampling points; Using the target distance between different preset sampling points as weights and the preset sampling points as nodes, a minimum spanning tree is constructed as the spectral minimum spanning tree.

8. The rapid detection method for hydroxyproline content in mouse lung tissue according to claim 1, characterized in that, The determination of the signal contribution weight for each preset sampling point based on the minimum spanning tree of the spectrum and the relative spectral response factor corresponding to each preset sampling point includes: The heterogeneity of biochemical characteristics is determined based on the mean and standard deviation of all edge weights in the minimum spanning tree of the spectrum. Any preset sampling point is designated as a marked sampling point. Based on the pre-acquired local response cutoff constant and sensitivity coefficient, the biochemical feature heterogeneity, the relative spectral response factor corresponding to the marked sampling point, the mean of all edge weights in the minimum spanning tree of the spectrum, and the mean of the edge weights of the edges connected to the marked sampling point in the minimum spanning tree of the spectrum, the signal contribution weight corresponding to the marked sampling point is determined.

9. The rapid detection method for hydroxyproline content in mouse lung tissue according to claim 1, characterized in that, The step of generating a synthetic spectrum of whole-pulmonary fibrosis based on the signal contribution weights corresponding to all preset sampling points and the spectral vector of lung interstitial characteristics includes: The sum of the products between the signal contribution weights corresponding to all preset sampling points and the spectral vectors of lung interstitial characteristics is determined as the synthetic spectrum of whole lung fibrosis.

10. The rapid detection method for hydroxyproline content in mouse lung tissue according to claim 1, characterized in that, The detection of hydroxyproline content based on the synthetic spectrum of whole-pulmonary fibrosis includes: Based on the pre-obtained content regression coefficient matrix and model intercept term, as well as the whole lung fibrosis synthesis spectrum, the total hydroxyproline content of the mice to be tested was determined.

Images