Coronary heart disease risk assessment system based on intestinal microorganisms

Abstract

The invention discloses a coronary heart disease risk assessment system based on intestinal microorganisms, and belongs to the field of artificial intelligence, the system obtains DNA information of the intestinal microorganisms in an excrement sample of a patient through a data obtaining module and integrates the DNA information into a sample to be detected; the sequencing module performs metagenome high-throughput sequencing on the to-be-detected sample to obtain an enteric microorganism gene sequence, so that the enteric microorganism composition analysis module analyzes the enteric microorganism gene sequence to obtain relative abundance data of each enteric microorganism; the interaction coefficient calculation module calculates the interaction coefficient of the intestinal microorganisms according to the relative abundance data corresponding to the intestinal microorganisms, a preset relative abundance database of the intestinal microorganisms and a preset elastic network local regression learning model; therefore, the coronary heart disease risk assessment result determination module determines the coronary heart disease risk assessment result. According to the invention, the problem that there is no risk assessment system suitable for early screening of coronary heart disease in the prior art can be solved.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence, and in particular to a coronary heart disease risk assessment system based on gut microbiota. Background Technology

[0002] Coronary artery disease (CAD) is one of the leading causes of death worldwide, and its clinical assessment is crucial for early intervention.

[0003] Current coronary heart disease assessment systems rely heavily on multi-step diagnostic processes, requiring sequential completion of examinations such as electrocardiogram (ECG), myocardial enzyme spectrum testing, and ambulatory blood pressure monitoring, and even invasive coronary angiography (CAG). From sample collection to the final report, it usually takes several hours to several days. Due to the lengthy process, invasiveness, and high cost, it is difficult to apply on a large scale in early screening scenarios such as health checkups and community screenings. As a result, about 60% of coronary heart disease patients in my country have already developed acute myocardial infarction when they first seek medical attention, missing the opportunity for early intervention. Summary of the Invention

[0004] This invention provides a coronary heart disease risk assessment system based on gut microbiota, which can solve the problem of the lack of a risk assessment system suitable for early screening of coronary heart disease in the prior art.

[0005] One embodiment of the present invention provides a coronary heart disease risk assessment system based on gut microbiota, comprising: a data acquisition module, a sequencing module, a gut microbiota composition analysis module, an interaction coefficient calculation module, and a coronary heart disease risk assessment result determination module;

[0006] The data acquisition module is used to acquire the DNA information of intestinal microorganisms in the patient's fecal sample and integrate the acquired DNA information of intestinal microorganisms into the sample to be tested.

[0007] The sequencing module is used to perform metagenomic high-throughput sequencing on the sample to be tested to obtain the patient's gut microbiome gene sequence;

[0008] The gut microbiota composition analysis module is used to analyze the gene sequences of the gut microbiota to obtain the relative abundance data of each gut microbiota in the fecal sample.

[0009] The interaction coefficient calculation module is used to calculate the intestinal microbial interaction coefficient based on the relative abundance data of each intestinal microorganism in the fecal sample.

[0010] The coronary heart disease risk assessment result determination module is used to determine the coronary heart disease risk assessment result based on a preset gut microbiota interaction weight table, a preset scoring table, and the gut microbiota interaction coefficient.

[0011] Further, the step of calculating the gut microbial interaction coefficient based on the relative abundance data of each gut microorganism in the fecal sample includes:

[0012] A relative abundance matrix is ​​generated based on a preset gut microbial relative abundance database and the relative abundance data of each gut microbial species in the fecal sample; wherein each row of the relative abundance matrix corresponds to a gut microbial species, and each column of the relative abundance matrix corresponds to a set of relative abundance data.

[0013] Based on the relative abundance matrix, the ecological phase manifold of the gut microbiota community was calculated;

[0014] Based on the gut microbial community ecological phase manifold, the relative abundance matrix, and the preset elastic network local regression learning model, the skewness among the gut microbes in the fecal sample is determined.

[0015] The skewness is used as the gut microbiota interaction coefficient to obtain the gut microbiota interaction coefficient.

[0016] Furthermore, the gut microbiome relative abundance database includes several gut microbiome relative abundance data samples; each gut microbiome relative abundance data sample corresponds to a unique sample number and includes a set of relative abundance data;

[0017] The step of generating a relative abundance matrix based on a preset gut microbial relative abundance database and the relative abundance data of each gut microbial in the fecal sample includes:

[0018] An initial relative abundance matrix was constructed using gut microbiome species as row labels and sample numbers as column labels.

[0019] The gut microbial relative abundance data samples from the gut microbial relative abundance database are filled into the initial relative abundance matrix column by column to generate the intermediate state of the initial relative abundance matrix;

[0020] The relative abundance data of each gut microorganism in the fecal sample are filled into the initial relative abundance matrix of the intermediate state to generate the final relative abundance matrix.

[0021] Further, the step of calculating the gut microbiota ecological manifold based on the relative abundance matrix includes:

[0022] Calculate the Euclidean distance between the relative abundance data of each gut microorganism in the fecal sample and the relative abundance data of each column in the relative abundance matrix;

[0023] Based on all the calculated Euclidean distances, determine the weighting terms;

[0024] The intestinal microbial community ecological phase manifold is calculated based on the weighting terms, the relative abundance matrix, and the preset identity matrix.

[0025] Furthermore, the gut microbiota ecological phase manifold was calculated using the following formula:

[0026]

[0027] In the formula, w j X represents the j-th local weight value; i The relative abundance data of each gut microbiota in the fecal sample; X j This represents the relative abundance data in the j-th column of the relative abundance matrix; ||X i -X j || represents the Euclidean distance between the relative abundance data of each gut microbe in the fecal sample and the relative abundance data in the j-th column of the relative abundance matrix; Y represents the gut microbial community ecological phase manifold; X represents the relative abundance matrix; W represents the weighting term, which is an n×1 vector including the sequentially calculated local weight values ​​w j ; n represents the number of columns in the relative abundance matrix; I represents an n×n identity matrix; I·W represents the inner product of I and W.

[0028] Further, determining the skewness among gut microbes in the fecal sample based on the gut microbial community ecological phase manifold, the relative abundance matrix, and a preset elastic network local regression learning model includes:

[0029] The gut microbiota ecological phase manifold and the relative abundance matrix are input into a preset elastic network local regression learning model to obtain the skewness among gut microbes output by the elastic network local regression learning model.

[0030] Furthermore, the elastic network local regression learning model is specifically as follows:

[0031]

[0032] In the formula, represents the skewness of the final output; c represents the skewness to be optimized; n represents the number of columns in the relative abundance matrix; Y represents the gut microbiota ecological phase manifold; X represents the relative abundance matrix; W represents the weighting term; λ represents the preset regularization intensity parameter; α represents the preset mixing ratio parameter.

[0033] Furthermore, the gut microbiota interaction weight table includes the interaction coefficient range corresponding to each gut microbiota interaction and the preset weight corresponding to each interaction coefficient range;

[0034] The step of determining the coronary heart disease risk assessment result based on a preset gut microbiota interaction weight table, a preset scoring table, and the gut microbiota interaction coefficient includes:

[0035] For each gut microbiota interaction coefficient, the gut microbiota interaction corresponding to the gut microbiota interaction coefficient is determined, and the gut microbiota interaction corresponding to the gut microbiota interaction coefficient is taken as the target gut microbiota interaction; the range of interaction coefficients corresponding to the target gut microbiota interaction is found from the gut microbiota interaction weight table, and is taken as the target interaction coefficient range; it is then determined whether the gut microbiota interaction coefficient exceeds the target interaction coefficient range.

[0036] If so, the preset weight corresponding to the target interaction coefficient range is multiplied by the gut microbiota interaction coefficient to obtain the corrected gut microbiota interaction coefficient.

[0037] If not, the gut microbiota interaction coefficient is set to zero as the corrected gut microbiota interaction coefficient;

[0038] The total score is obtained by summing all the corrected gut microbiota interaction coefficients.

[0039] Based on the total score and the preset scoring table, the patient's coronary heart disease risk assessment result is determined.

[0040] Furthermore, the scoring table includes several scoring ranges, and for each scoring range, a probability score for non-coronary artery disease (NCA), a probability score for stable coronary artery disease (sCAD), and a probability score for acute myocardial infarction (AMI).

[0041] The process of determining the patient's coronary heart disease risk assessment result based on the total score and a preset scoring table includes:

[0042] Determine the scoring range into which the total score falls, and use the scoring range into which the total score falls as the target scoring range;

[0043] The non-coronary artery disease (NCA) probability score, stable coronary artery disease (sCAD) probability score, and acute myocardial infarction (AMI) probability score corresponding to the target score range are used as the coronary artery disease risk assessment results for the patient.

[0044] The following benefits can be obtained by implementing the present invention:

[0045] This invention provides a coronary heart disease risk assessment system based on gut microbiota. The system first acquires the DNA information of gut microbiota from a patient's fecal sample using a data acquisition module, and integrates this DNA information into a test sample. Then, a sequencing module performs metagenomic high-throughput sequencing on the test sample to obtain the patient's gut microbiota gene sequence. Analysis of the gut microbiota gene sequence yields the relative abundance data of each gut microbiota in the fecal sample. Using this relative abundance data, gut microbiota interaction coefficients are calculated. Based on a preset gut microbiota interaction weight table, a preset scoring table, and the gut microbiota interaction coefficients, the coronary heart disease risk assessment result is determined. Thus, the entire process from sample to risk assessment can be completed within hours, significantly reducing time compared to traditional methods. Furthermore, fecal sample collection is non-invasive, making it suitable for large-scale early screening. Attached Figure Description

[0046] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0047] Figure 1 This is a schematic diagram of the structure of a coronary heart disease risk assessment system based on gut microbiota provided in a certain embodiment of this application. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0049] 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 application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0050] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0051] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0052] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0053] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0054] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0055] See Figure 1 This is a schematic diagram of the structure of a coronary heart disease risk assessment system based on gut microbiota provided in an embodiment of the present invention, including: a data acquisition module, a sequencing module, a gut microbiota composition analysis module, an interaction coefficient calculation module, and a coronary heart disease risk assessment result determination module;

[0056] The data acquisition module is used to acquire the DNA information of intestinal microorganisms in the patient's fecal sample and integrate the acquired DNA information of intestinal microorganisms into the sample to be tested.

[0057] The sequencing module is used to perform metagenomic high-throughput sequencing on the sample to be tested to obtain the patient's gut microbiome gene sequence;

[0058] The gut microbiota composition analysis module is used to analyze the gene sequences of the gut microbiota to obtain the relative abundance data of each gut microbiota in the fecal sample.

[0059] The interaction coefficient calculation module is used to calculate the intestinal microbial interaction coefficient based on the relative abundance data of each intestinal microorganism in the fecal sample.

[0060] The coronary heart disease risk assessment result determination module is used to determine the coronary heart disease risk assessment result based on a preset gut microbiota interaction weight table, a preset scoring table, and the gut microbiota interaction coefficient.

[0061] This is illustrative, assuming that a stool sample from a known patient is currently available.

[0062] Specifically, the DNA information of gut microbiota in the fecal sample was extracted using the NucleoSpin Soil kit, and the obtained DNA information of gut microbiota was integrated into the sample to be tested.

[0063] It should be noted that the step of extracting gut microbial DNA information from fecal samples based on the NucleoSpin Soil kit has been documented in detail in the existing technology and will not be repeated here;

[0064] Specifically, metagenomic high-throughput sequencing was performed on the sample to be tested to obtain the patient's gut microbiome gene sequence;

[0065] It should be noted that metagenomic high-throughput sequencing is a technology that directly performs large-scale sequencing of the genomic DNA of all microorganisms in a sample without the need to isolate and culture microorganisms. It can comprehensively analyze the genetic composition, functional potential and dynamic relationships of complex microbial communities. Since the existing technology has been documented in detail, the specific operation procedure for obtaining the gene sequence of the patient's gut microbiota through metagenomic high-throughput sequencing will not be described in detail here.

[0066] It should be noted that, in this embodiment, the microbial community analysis algorithm is MetagenomicPhylogenetic Analysis 2 (abbreviated as MetaPhlan2), which achieves high-precision annotation of microbial communities at the species level through specific marker genes, and is especially suitable for rapid classification and analysis of complex microbial communities such as the gut and oral cavity;

[0067] Specifically, the gene sequences of the gut microbiota were analyzed using MetaPhlan2 to obtain the relative abundance data of each gut microbiota in the fecal sample.

[0068] In a preferred embodiment, the step of calculating the gut microbial interaction coefficient based on the relative abundance data of each gut microorganism in the fecal sample includes:

[0069] A relative abundance matrix is ​​generated based on a preset gut microbial relative abundance database and the relative abundance data of each gut microbial species in the fecal sample; wherein each row of the relative abundance matrix corresponds to a gut microbial species, and each column of the relative abundance matrix corresponds to a set of relative abundance data.

[0070] Based on the relative abundance matrix, the ecological phase manifold of the gut microbiota community was calculated;

[0071] Based on the gut microbial community ecological phase manifold, the relative abundance matrix, and the preset elastic network local regression learning model, the skewness among the gut microbes in the fecal sample is determined.

[0072] The skewness is used as the gut microbiota interaction coefficient to obtain the gut microbiota interaction coefficient;

[0073] In a schematic manner, firstly, based on the relative abundance data of each gut microbiota in fecal samples, combined with a pre-set gut microbiota relative abundance database, a pre-set elastic network local regression learning model is used to perform regularized fitting on the association between microbial abundance and function. Then, the standardized gut microbiota composition information is mapped to a high-dimensional phase space with microbial abundance as the coordinate axis, constructing a gut microbiota community ecological phase manifold that represents the dynamic relationship of the microbial community. Finally, the skewness between each gut microbiota is calculated through a multidimensional surface fitting algorithm, thereby quantifying the gut microbiota interaction coefficient and realizing a quantitative derivation from abundance data to ecological interaction relationships.

[0074] Specifically, the types of intestinal microorganisms extracted from fecal samples include: Pediococcus (unclassified), Enterobacter habdus caecimuris, Bacteroides caccae, Bacteroides xylanisolvens, Porphyromonas gingivalis, Megamonas (unclassified), and Leuconostoc pseudodomesticus. Inulin Roseburia, Eubacterium eligens, Elizabethkingia unclassified, Lactobacillus sanfranciscensis, Bacteroides ovatus, Streptococcus constellatus, Dorea longicatena, Streptococcus salivarius tococcus_salivarius), Rickettsia grylli, Bilophila unclassified, Streptococcus infantis, Roseburia hominis, Bacteroides dorei, Alistipes putredinis, Coprococcus comes, Stenotrophomonas The following bacteria are listed: monas_unclassified, eubacterium_saphenum, eubacterium_rectale, subdoligranulum_unclassified, Collinsella_aerofaciens, Streptococcus_gordonii, Bacteroides_uniformis, and Bifidobacterium_longum.Orychomycosis nasolabialis, Alistipes putredinis, Bacteroides vulgatus, Oral taxon 078 Orychomycosis nasolabialis, Alistipes shahii, Streptococcus pneumoniae (oral subsp.) oralis, Oscillibacter unclassified, Holdemania filiformis, Ruminococcus torques, Ruminococcus 5_1_39BFAA

[0075] (Ruminococcus_sp_5_1_39BFAA), Coprococcus catus, Atopobium rimae, Atopobium parvulum, Eubacterium hallii, Bacteroides thetaiotaomicron, Ruminococcus bromii, Pyramidobacter piscolens.

[0076] In a preferred embodiment, the gut microbiome relative abundance database includes several gut microbiome relative abundance data samples; each gut microbiome relative abundance data sample corresponds to a unique sample number and includes a set of relative abundance data;

[0077] The step of generating a relative abundance matrix based on a preset gut microbial relative abundance database and the relative abundance data of each gut microbial in the fecal sample includes:

[0078] An initial relative abundance matrix was constructed using gut microbiome species as row labels and sample numbers as column labels.

[0079] The gut microbial relative abundance data samples from the gut microbial relative abundance database are filled into the initial relative abundance matrix column by column to generate the intermediate state of the initial relative abundance matrix;

[0080] The relative abundance data of each gut microorganism in the fecal sample are filled into the initial relative abundance matrix of the intermediate state to generate the final relative abundance matrix.

[0081] Specifically, through the gut microbiome relative abundance database X D The relative abundance data X of each gut microbiota in the patient's fecal sample i The relative abundance matrix X; the gut microbiota relative abundance database X D It is an m×(n-1) matrix, where m represents the total number of gut microbial species; n-1 represents the total number of samples in the gut microbial relative abundance database; and X represents the relative abundance data X of each gut microbial species in the patient's fecal sample. i is an m×1 vector, where m represents the total number of gut microbial species; the relative abundance matrix X is an m×n matrix, where m represents the total number of gut microbial species; n represents the total number of samples, which is obtained by summing the number of samples in the gut microbial relative abundance database and the number of samples composed of the relative abundance data corresponding to each gut microbial species in the fecal sample.

[0082] In a preferred embodiment, calculating the gut microbiota ecological phase manifold based on the relative abundance matrix includes:

[0083] Calculate the Euclidean distance between the relative abundance data of each gut microorganism in the fecal sample and the relative abundance data of each column in the relative abundance matrix;

[0084] Based on all the calculated Euclidean distances, determine the weighting terms;

[0085] The gut microbiota ecological phase manifold is calculated based on the weighting terms, the relative abundance matrix, and the preset identity matrix.

[0086] Indicatively, the gut microbiota ecological phase manifold is constructed using locally weighted ensemble average;

[0087] Specifically, since the relative abundance matrix X has already been obtained, it is necessary to calculate the gut microbiota ecological phase manifold based on the relative abundance matrix X:

[0088] In a preferred embodiment, the gut microbiota ecological phase manifold is calculated using the following formula:

[0089]

[0090] In the formula, w j X represents the j-th local weight value; iThe relative abundance data of each gut microbiota in the fecal sample; X j This represents the relative abundance data in the j-th column of the relative abundance matrix; ||X i -X j || represents the Euclidean distance between the relative abundance data of each gut microbe in the fecal sample and the relative abundance data in the j-th column of the relative abundance matrix; Y represents the gut microbial community ecological phase manifold; X represents the relative abundance matrix; W represents the weighting term, which is an n×1 vector including the sequentially calculated local weight values ​​w j ; n represents the number of columns in the relative abundance matrix; I represents an n×n identity matrix; I·W represents the inner product of I and W;

[0091] Specifically, W represents the weighting term, which is an n×1 vector.

[0092] In a preferred embodiment, determining the skewness among gut microbes in a fecal sample based on the gut microbiota manifold, the relative abundance matrix, and a preset elastic network local regression learning model includes:

[0093] The gut microbial community ecological phase manifold and the relative abundance matrix are input into a preset elastic network local regression learning model to obtain the skewness among gut microorganisms output by the elastic network local regression learning model.

[0094] Indicatively, the local regression learning model of the elastic network has been trained in advance, and the specific values ​​of the regularization strength parameter λ and the mixing ratio parameter α of the penalty term of the trained local regression learning model of the elastic network have been fitted and determined.

[0095] Specifically, when training the local regression learning model of the elastic network, it is necessary to use several relative abundance matrix sample data and corresponding gut microbiota ecological phase manifold sample data as inputs to the local regression learning model of the elastic network to be trained. The skewness output by the local regression learning model of the elastic network to be trained is compared with the corresponding label skewness. Based on the comparison results, the regularization strength parameter and mixing ratio parameter of the penalty term are continuously adjusted to improve the fitting accuracy of the local regression learning model of the elastic network. This process is repeated to obtain the trained local regression learning model of the elastic network.

[0096] In a preferred embodiment, the elastic network local regression learning model is specifically:

[0097]

[0098] In the formula, represents the skewness of the final output; c represents the skewness to be optimized; n represents the number of columns in the relative abundance matrix; Y represents the gut microbiota ecological phase manifold; X represents the relative abundance matrix; W represents the weighting term; λ represents the preset regularization intensity parameter; α represents the preset mixing ratio parameter.

[0099] Specifically, the skewness among the gut microbiota is obtained by solving the local regression learning model of the elastic network.

[0100] In a preferred embodiment, the gut microbiota interaction weight table includes the interaction coefficient range corresponding to each gut microbiota interaction and the preset weight corresponding to each interaction coefficient range;

[0101] The step of determining the coronary heart disease risk assessment result based on a preset gut microbiota interaction weight table, a preset scoring table, and the gut microbiota interaction coefficient includes:

[0102] For each gut microbiota interaction coefficient, the gut microbiota interaction corresponding to the gut microbiota interaction coefficient is determined, and the gut microbiota interaction corresponding to the gut microbiota interaction coefficient is taken as the target gut microbiota interaction; the range of interaction coefficients corresponding to the target gut microbiota interaction is found from the gut microbiota interaction weight table, and is taken as the target interaction coefficient range; it is then determined whether the gut microbiota interaction coefficient exceeds the target interaction coefficient range.

[0103] If so, the preset weight corresponding to the target interaction coefficient range is multiplied by the gut microbiota interaction coefficient to obtain the corrected gut microbiota interaction coefficient.

[0104] If not, the gut microbiota interaction coefficient is set to zero as the corrected gut microbiota interaction coefficient;

[0105] The total score is obtained by summing all the corrected gut microbiota interaction coefficients.

[0106] Based on the total score and the preset scoring table, the patient's coronary heart disease risk assessment result is determined;

[0107] Specifically, the gut microbiota interaction weight table is as follows:

[0108] Table 1. Weighting of Gut Microbial Interactions

[0109]

[0110]

[0111]

[0112]

[0113]

[0114]

[0115] Specifically, the target gut microbiota interactions are assumed to be as follows:

[0116] The gut microbial interaction coefficient of Pediococcus_unclassified->Enterorhabdus_caecimuris is 1.5. As shown in Table 1, the gut microbial interaction coefficient of the target gut microbial interaction Pediococcus_unclassified->Enterorhabdus_caecimuris exceeds its corresponding interaction coefficient range. Therefore, the gut microbial interaction coefficient of Pediococcus_unclassified->Enterorhabdus_caecimuris needs to be corrected, resulting in a corrected gut microbial interaction coefficient of 1.58 × 1.5 = 2.37.

[0117] Specifically, the target gut microbiota interactions are assumed to be as follows:

[0118] The gut microbial interaction coefficient of Apple_stem_grooving_virus->Bacteroides_xylanisolvens is -0.528. As shown in Table 1, the gut microbial interaction coefficient of the target gut microbial interaction Apple_stem_grooving_virus->Bacteroides_xylanisolvens does not exceed its corresponding interaction coefficient range. Therefore, the corrected gut microbial interaction coefficient of Apple_stem_grooving_virus->Bacteroides_xylanisolvens is directly set to zero.

[0119] All the corrected gut microbiota interaction coefficients are summed to obtain a total score. Based on the total score and a preset scoring table, the patient's coronary heart disease risk assessment result is determined.

[0120] In a preferred embodiment, the scoring table includes several scoring ranges, and for each scoring range, a probability score for non-coronary artery disease (NCA), a probability score for stable coronary artery disease (sCAD), and a probability score for acute myocardial infarction (AMI).

[0121] Determine the coronary heart disease risk assessment result of the patient according to the total score and a preset scoring table, including:

[0122] Determine the scoring range into which the total score falls, and use the scoring range into which the total score falls as the target scoring range;

[0123] Use the non-coronary heart disease NCA probability score, stable coronary heart disease sCAD probability score, and acute myocardial infarction AMI probability score corresponding to the target scoring range as the coronary heart disease risk assessment result of the patient;

[0124] Specifically, the scoring table is as follows:

[0125] Table 2 Scoring Table

[0126]

[0127] Specifically, assume that the total score Score is 5.28. Then the scoring range it falls into is 5 < Score ≤ 6. Correspondingly, the coronary heart disease risk assessment result of the patient is: the non-coronary heart disease NCA probability score is 0.00%, the stable coronary heart disease sCAD probability score is 20.00%, and the acute myocardial infarction AMI probability score is 80.00%;

[0128] Specifically, after obtaining the coronary heart disease risk assessment result of the patient, when the assessment result shows that the AMI probability is as high as 80.0%, clinicians can quickly judge based on this that the patient needs to enter the emergency green channel first, immediately initiate the detection of myocardial injury markers (such as troponin) and electrocardiogram (ECG) monitoring, avoid irreversible damage to myocardial cells caused by delayed diagnosis. At the same time, combined with the sCAD probability of 20%, the potential risks of stable coronary artery lesions can be synchronously screened to provide a predictive basis for subsequent invasive examinations such as coronary angiography;

[0129] The coronary heart disease risk assessment result provides a reference for clinical practice by quantifying the risk probabilities of different coronary heart disease types, especially applicable to the rapid differential diagnosis of patients with atypical symptoms, improving the diagnosis and treatment efficiency and improving the prognosis.

[0130] The above is the preferred implementation manner of the present invention. It should be noted that for those of ordinary skill in the technical field, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements are also regarded as the protection scope of the present invention.

Claims

1. A coronary heart disease risk assessment system based on gut microbiota, characterized in that, include: The module includes a data acquisition module, a sequencing module, a gut microbiota composition analysis module, an interaction coefficient calculation module, and a coronary heart disease risk assessment result determination module. The data acquisition module is used to acquire the DNA information of intestinal microorganisms in the patient's fecal sample and integrate the acquired DNA information of intestinal microorganisms into the sample to be tested. The sequencing module is used to perform metagenomic high-throughput sequencing on the sample to be tested to obtain the patient's gut microbiome gene sequence; The gut microbiota composition analysis module is used to analyze the gene sequences of the gut microbiota to obtain the relative abundance data of each gut microbiota in the fecal sample. The interaction coefficient calculation module is used to calculate the intestinal microbial interaction coefficient based on the relative abundance data of each intestinal microorganism in the fecal sample. The coronary heart disease risk assessment result determination module is used to determine the coronary heart disease risk assessment result based on a preset gut microbiota interaction weight table, a preset scoring table, and the gut microbiota interaction coefficient.

2. The coronary heart disease risk assessment system based on gut microbiota as described in claim 1, characterized in that, The gut microbial interaction coefficient is calculated based on the relative abundance data of each gut microorganism in the fecal sample, including: A relative abundance matrix is ​​generated based on a preset gut microbial relative abundance database and the relative abundance data of each gut microbial species in the fecal sample; wherein each row of the relative abundance matrix corresponds to a gut microbial species, and each column of the relative abundance matrix corresponds to a set of relative abundance data. Based on the relative abundance matrix, the ecological phase manifold of the gut microbiota community was calculated; Based on the gut microbial community ecological phase manifold, the relative abundance matrix, and the preset elastic network local regression learning model, the skewness among the gut microbes in the fecal sample is determined. The skewness is used as the gut microbiota interaction coefficient to obtain the gut microbiota interaction coefficient.

3. The coronary heart disease risk assessment system based on gut microbiota as described in claim 2, characterized in that, The gut microbiome relative abundance database includes several gut microbiome relative abundance data samples; each gut microbiome relative abundance data sample corresponds to a unique sample number and includes a set of relative abundance data; The step of generating a relative abundance matrix based on a preset gut microbial relative abundance database and the relative abundance data of each gut microbial in the fecal sample includes: An initial relative abundance matrix was constructed using gut microbiome species as row labels and sample numbers as column labels. The gut microbial relative abundance data samples from the gut microbial relative abundance database are filled into the initial relative abundance matrix column by column to generate the intermediate state of the initial relative abundance matrix; The relative abundance data of each gut microorganism in the fecal sample are filled into the initial relative abundance matrix of the intermediate state to generate the final relative abundance matrix.

4. The coronary heart disease risk assessment system based on gut microbiota as described in claim 3, characterized in that, The calculation of the gut microbiota ecological phase manifold based on the relative abundance matrix includes: Calculate the Euclidean distance between the relative abundance data of each gut microorganism in the fecal sample and the relative abundance data of each column in the relative abundance matrix; Based on all the calculated Euclidean distances, determine the weighting terms; The intestinal microbial community ecological phase manifold is calculated based on the weighting terms, the relative abundance matrix, and the preset identity matrix.

5. The coronary heart disease risk assessment system based on gut microbiota as described in claim 4, characterized in that, The gut microbiota ecological phase manifold is calculated using the following formula: In the formula, w j X represents the j-th local weight value; i The relative abundance data of each gut microbiota in the fecal sample; X j This represents the relative abundance data in the j-th column of the relative abundance matrix; ||X i -X j || represents the Euclidean distance between the relative abundance data of each gut microbe in the fecal sample and the relative abundance data in the j-th column of the relative abundance matrix; Y represents the gut microbial community ecological phase manifold; X represents the relative abundance matrix; W represents the weighting term, which is an n×1 vector including the sequentially calculated local weight values ​​w j n represents the number of columns in the relative abundance matrix; I represents an n×n identity matrix; I·W represents the inner product of I and W.

6. The coronary heart disease risk assessment system based on gut microbiota as described in claim 5, characterized in that, The step of determining the skewness among gut microbes in a fecal sample based on the gut microbiota manifold, the relative abundance matrix, and a pre-defined elastic network local regression learning model includes: The gut microbiota ecological phase manifold and the relative abundance matrix are input into a preset elastic network local regression learning model to obtain the skewness among gut microbes output by the elastic network local regression learning model.

7. The coronary heart disease risk assessment system based on gut microbiota as described in claim 6, characterized in that, The elastic network local regression learning model is specifically as follows: In the formula, represents the skewness of the final output; c represents the skewness to be optimized; n represents the number of columns in the relative abundance matrix; Y represents the gut microbiota ecological phase manifold; X represents the relative abundance matrix; W represents the weighting term; λ represents the preset regularization intensity parameter; α represents the preset mixing ratio parameter.

8. The coronary heart disease risk assessment system based on gut microbiota as described in claim 7, characterized in that, The gut microbial interaction weight table includes the interaction coefficient range corresponding to each gut microbial interaction and the preset weight corresponding to each interaction coefficient range. The step of determining the coronary heart disease risk assessment result based on a preset gut microbiota interaction weight table, a preset scoring table, and the gut microbiota interaction coefficient includes: For each gut microbiota interaction coefficient, the gut microbiota interaction corresponding to the gut microbiota interaction coefficient is determined, and the gut microbiota interaction corresponding to the gut microbiota interaction coefficient is taken as the target gut microbiota interaction; the range of interaction coefficients corresponding to the target gut microbiota interaction is found from the gut microbiota interaction weight table, and is taken as the target interaction coefficient range; it is then determined whether the gut microbiota interaction coefficient exceeds the target interaction coefficient range. If so, the preset weight corresponding to the target interaction coefficient range is multiplied by the gut microbiota interaction coefficient to obtain the corrected gut microbiota interaction coefficient. If not, the gut microbiota interaction coefficient is set to zero as the corrected gut microbiota interaction coefficient; The total score is obtained by summing all the corrected gut microbiota interaction coefficients. Based on the total score and the preset scoring table, the patient's coronary heart disease risk assessment result is determined.

9. The coronary heart disease risk assessment system based on gut microbiota as described in claim 8, characterized in that, The scoring table includes several scoring ranges, and for each scoring range, there is a probability score for non-coronary artery disease (NCA), a probability score for stable coronary artery disease (sCAD), and a probability score for acute myocardial infarction (AMI). The process of determining the patient's coronary heart disease risk assessment result based on the total score and a preset scoring table includes: Determine the scoring range into which the total score falls, and use the scoring range into which the total score falls as the target scoring range; The non-coronary artery disease (NCA) probability score, stable coronary artery disease (sCAD) probability score, and acute myocardial infarction (AMI) probability score corresponding to the target score range are used as the coronary artery disease risk assessment results for the patient.

Images