Pulmonary arterial hypertension risk assessment system based on data processing

Abstract

The invention relates to the technical field of pulmonary artery risk assessment, in particular to a pulmonary artery hypertension risk assessment system based on data processing, and the system comprises an intestinal microbiota data collection module which is used for extracting intestinal microbiota DNA data and recognizing the relative abundance of intestinal microbiota; the microbial metabolite detection module is used for detecting microbial metabolite related to pulmonary arterial hypertension to obtain microbial population metabolism concentration data; the microflora imbalance evaluation module is used for evaluating the intestinal microflora imbalance degree of the target patient by utilizing a machine learning algorithm based on the microflora relative abundance and the microflora metabolic concentration data; the correlation analysis module is used for identifying the potential influence of the microbiota disorder degree on an immune system and an inflammatory state; and the personalized risk prediction module is used for providing a pulmonary arterial hypertension risk score. According to the method, a more practical risk assessment tool is provided clinically, the complex assessment process is simplified, and the efficiency and accuracy of risk prediction are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of pulmonary artery risk assessment, and in particular to a pulmonary hypertension risk assessment system based on data processing. Background Art

[0002] Pulmonary arterial hypertension (PAH) is a chronic, progressive cardiovascular disease characterized by a continuous increase in pulmonary artery pressure, which increases the right ventricular pressure burden and eventually leads to right heart failure and death. Although there have been many studies on the pathogenesis of PAH, its complex pathological process is still not fully understood. Existing PAH risk assessment methods mainly rely on patients' clinical symptoms, imaging examinations (such as echocardiography), and cardiac catheterization. Although these methods have shown good results in the diagnosis and evaluation of patients with advanced PAH, they have many limitations in the early identification and prediction of the disease.

[0003] Studies in recent years have shown that abnormal activation of the immune system and chronic inflammatory response are one of the important mechanisms of PAH pathogenesis. Proinflammatory cytokines (such as IL-6 and TNF-α) are significantly elevated in PAH patients, suggesting that immune and inflammatory disorders play a key role in disease progression. However, most of the current PAH risk assessment tools fail to fully utilize relevant information on immune and inflammatory status and cannot fully reflect the impact of these biomarkers on the onset and progression of PAH.

[0004] In addition, the importance of intestinal microbiota in cardiovascular diseases has gradually attracted attention in recent years. The intestinal microbiota affects systemic inflammatory response and immune regulation through its metabolites (such as short-chain fatty acids, indole compounds and trimethylamine-N-oxide). Many studies have shown that there is a potential association between the imbalance of intestinal microbiota and a variety of cardiovascular diseases, including atherosclerosis, hypertension and heart failure. However, in the field of risk assessment of pulmonary hypertension, the role of intestinal microbiota has not been fully studied and applied.

[0005] Existing technologies lack a systematic approach that can link intestinal microbiota imbalance, immune system imbalance and pulmonary arterial hypertension risk; secondly, current PAH risk assessment tools fail to integrate multidimensional biomarker data, especially the contribution of immune and inflammatory status to disease development has not been fully quantified. Summary of the invention

[0006] The invention provides a pulmonary hypertension risk assessment system based on data processing.

[0007] The pulmonary hypertension risk assessment system based on data processing includes:

[0008] Gut microbiome data collection module: extracts intestinal microbial DNA data from patients' stool samples, conducts qualitative analysis of the intestinal microbiome through high-throughput sequencing technology, and identifies the relative abundance of the intestinal microbiome;

[0009] Microbial metabolite detection module: detects microbial metabolites related to pulmonary hypertension, including short-chain fatty acids (SCFA), trimethylamine-N-oxide (TMAO), and indole metabolites. Metabolite product concentration analysis is performed based on blood and urine samples to obtain microbial community metabolic concentration data;

[0010] Microbial community imbalance assessment module: Based on the relative abundance and metabolic concentration data of the microbiome, combined with the baseline data of the intestinal microbiome of healthy people (microbiome relative abundance baseline, microbiome metabolic concentration baseline), the machine learning algorithm is used to assess the degree of intestinal microbiome imbalance in the target patient;

[0011] Association analysis module: By constructing a multidimensional association model between the degree of intestinal microbiome imbalance and the immune system and inflammatory status, the potential impact of the degree of microbiome imbalance on the immune system and inflammatory status is identified, and the immune and inflammatory disorder scores are obtained;

[0012] Personalized risk prediction module: Based on the conversion of immune and inflammatory disorder scores, it provides a pulmonary hypertension risk score for a period of time in the future.

[0013] Optionally, the intestinal microbiome data collection module includes:

[0014] The stool sample processing unit collects and preserves the patient's stool samples, pre-treats them to remove macromolecular impurities, and destroys the cell walls in the samples through mechanical homogenization, centrifugation, and enzymatic treatment to release the DNA of intestinal microorganisms;

[0015] DNA extraction unit, which uses a DNA extraction kit (silica-based membrane method) to extract genomic DNA of intestinal microorganisms from processed fecal samples and ensure high-quality DNA extraction to avoid environmental DNA contamination;

[0016] A high-throughput sequencing unit is used to perform amplicon sequencing or whole-genome sequencing on the extracted intestinal microbial DNA, and to amplify specific regions of the 16S rRNA gene (such as the V3-V4 region) for qualitative analysis of microbial species;

[0017] The data processing and analysis unit performs bioinformatics processing on the sequence data generated by high-throughput sequencing, combines it with the database (Greengenes microbial database) to perform species classification, sequence alignment and abundance analysis, calculates the relative abundance of the microbiome, and identifies the dominant microbiome and potentially imbalanced microbial communities in the patient's intestines.

[0018] The calculation formula for the relative abundance of microorganisms is: Among them, A i represents the relative abundance of the i-th microbial group, S i is the number of valid sequences of the i-th microbial species obtained by sequencing.

[0019] Optionally, the microbial metabolite detection module includes:

[0020] Sample collection unit: collects metabolites related to pulmonary hypertension from patients' blood and urine samples. The samples are pre-treated by centrifugation and filtration to remove impurities and retain effective metabolites.

[0021] Short-chain fatty acid detection unit: Quantitative analysis of short-chain fatty acids in blood and urine is performed using gas chromatography-mass spectrometry to obtain the specific concentration of short-chain fatty acids;

[0022] Trimethylamine-N-oxide detection unit: Use liquid chromatography-mass spectrometry to detect trimethylamine-N-oxide in blood and urine with high sensitivity, and calculate the concentration of trimethylamine-N-oxide by comparing with the standard curve;

[0023] Indole metabolite detection unit: Utilizes ultra-high performance liquid chromatography and ultraviolet detection technology, combined with quantitative standardization, to detect the concentration of indole metabolites in blood and urine;

[0024] Metabolite concentration data processing unit: The metabolite concentration data obtained by detection are standardized to generate concentration data of short-chain fatty acids, trimethylamine-N-oxide, and indole metabolites as input for subsequent microbiome dysbiosis assessment and pulmonary hypertension risk assessment.

[0025] Optionally, the microbial community imbalance assessment module specifically includes:

[0026] Data acquisition unit: Receive the target patient's microbiome relative abundance data and microbial metabolite concentration data, import the intestinal microbiome baseline data of healthy people, the baseline data includes the microbiome relative abundance baseline and metabolite concentration baseline, and standardize the baseline data to ensure that data from different sources have consistent scale and comparability;

[0027] Difference calculation unit: used to calculate the difference between the target patient and the healthy baseline, based on the formula: Among them, D i is the relative deviation of the relative abundance of the i-th microbiota of the target patient from the healthy baseline, αC i is the relative deviation of the concentration of the ith metabolite of the target patient from the healthy baseline, A i and B iare the relative abundances of the i-th microbial species in the target patients and healthy subjects, respectively, and C i and are the metabolite concentrations of the target patient and healthy baseline, respectively;

[0028] Feature extraction unit: Based on the difference data between target patients and healthy people, key features for evaluating the degree of microbiome imbalance are extracted, including deviations in relative abundance of microorganisms and deviations in metabolite concentrations;

[0029] Sparse regression model unit: The extracted features are evaluated using a sparse regression model, regularization terms are introduced, and key microbiome and metabolite features that affect the degree of disorder are automatically selected, while noise and irrelevant features are suppressed. The model training process is carried out using healthy baseline data and a dataset of patients with known disorders to optimize the model's predictive ability;

[0030] Disorder scoring unit: Based on the output results of the sparse regression model, a gut microbiota disorder score for the target patient is generated. The score is generated based on the weighted contribution of microbial relative abundance and metabolite deviation, and is used to quantify the overall disorder of the gut microbiota and provide key input for subsequent pulmonary hypertension risk assessment.

[0031] Optionally, the sparse regression model analyzes the extracted features based on Lasso regression, and Lasso regression has the ability to automatically screen key features;

[0032] During the model training process, data from historical healthy people and patients with known intestinal disorders were used as training sets to help the model learn to identify key microbial and metabolite changes. Based on the relative abundance deviation of the patient's microbiota and metabolite concentration deviation, the features associated with the degree of disorder were determined, and Lasso regression assigned corresponding weights to each feature.

[0033] Optionally, the Lasso regression is expressed as: Among them, Lasso introduces a regularization term to constrain the regression coefficient β i .

[0034] Optionally, the Lasso introduces a regularization term to constrain the regression coefficient β i It is expressed as:

[0035] in, is the predicted dysbiosis score, β 0 is a constant term, X i represents each input feature, which may include the relative abundance deviation of microorganisms D i and metabolite deviation ΔC i , β iis the regression coefficient of each feature in Lasso regression, N is the number of samples, λ is the regularization parameter used to control the sparsity of the model, and Y j is the actual disorder degree of the target patient (the actual value in the training data).

[0036] Optionally, the association analysis module includes:

[0037] Data input unit: receiving the target patient's intestinal microbiota disorder degree data, including the deviation of microbial relative abundance, the deviation of metabolite concentration, and the biomarker concentration data of the immune system and inflammatory state, including C-reactive protein, interleukin-6, and TNF-α;

[0038] Set the biomarker concentration data to I j , represents the biomarker concentration of the jth immune system or inflammatory state of the target patient;

[0039] Feature mapping unit: By constructing a mapping relationship, the degree of intestinal microbiome imbalance is associated with the concentration of biomarkers of the immune system and inflammatory state, and key interactive features are extracted. The interactive features include the potential correlation between the degree of microbiome imbalance and specific immune responses and inflammatory states, such as the connection between the decrease in the abundance of specific microorganisms and the increase in pro-inflammatory markers. The interactive features of microbial imbalance and immune markers are expressed as: X ij =D i I j , where X ij represents the interaction characteristics between the i-th microbial dysbiosis and the j-th immune or inflammatory marker, D i is the relative abundance deviation of the i-th microbial group, I j is the concentration of the jth immune marker;

[0040] Multidimensional association model construction unit: Use multivariate regression analysis to construct a multidimensional association model to evaluate the potential impact of microbiome dysbiosis on the immune system and inflammatory status, and establish a multidimensional association network based on the interaction between microbiome dysbiosis characteristics and immune markers;

[0041] Potential Impact Identification Unit: Identify the potential impact of intestinal microbiome imbalance on the immune system and inflammatory status through association model analysis. Model output includes identifying the impact of specific microbiome imbalance on immune cell activity and inflammatory marker concentration, revealing the role of microbial imbalance in promoting systemic inflammatory response and immune disorder;

[0042] Result output unit: Quantify the analysis results into specific immune and inflammatory disorder scores to evaluate the impact of microbiome dysbiosis on the patient's immune system and inflammatory response.

[0043] Optionally, the multivariate linear regression model is expressed as: Among them, I j is the jth immune or inflammatory marker concentration of the patient, γ 0 is the constant term of the model, γ i is the characteristic D of the ith microbial imbalance i The associated regression coefficient, ∈, is the error term, which is used to capture the fluctuations not explained by the model;

[0044] The immune and inflammatory dysregulation score was calculated as:

[0045] Among them, S immune is a comprehensive score of immune system and inflammatory disorders, D i is the i-th microbial imbalance feature, I j is the concentration of the jth immune or inflammatory marker, γ i is the regression coefficient associated with the characteristics of microbial dysbiosis; through this score, the overall impact of microbiome dysbiosis on the immune system and inflammatory response can be quantified, providing a basis for subsequent pulmonary hypertension risk assessment.

[0046] Optionally, the personalized risk prediction module converts the immune and inflammatory disorder scores into pulmonary hypertension risk scores based on a mapping conversion model, expressed as: R PAH =α 0 +α 1 ·S immune , where R PAH is the predicted future pulmonary hypertension risk score, α 0 is a constant term, representing the baseline risk, α 1 are the regression coefficients of the maps, indicating the strength of the effects of immune and inflammatory dysregulation on the risk of PH.

[0047] The immune and inflammatory disorder score can be converted into a pulmonary arterial hypertension (PAH) risk score. The immune system and inflammatory response play a key role in the onset and progression of PAH. By understanding the extent of immune and inflammatory disorders, its potential impact on the pulmonary vascular and cardiopulmonary systems can be assessed, which are closely related to the pathogenesis of PAH.

[0048] Beneficial effects of the present invention:

[0049] The present invention combines the imbalance of intestinal microbiota with the risk assessment of pulmonary hypertension. Through high-throughput sequencing technology and metabolite detection, it identifies the abnormalities of the relative abundance of microbiota and metabolite concentration, and constructs a multidimensional association model between microbiota imbalance and the immune system and inflammatory state. This multidimensional analysis method breaks through the limitations of traditional evaluation methods that rely solely on clinical data and imaging analysis, closely links intestinal microorganisms with systemic inflammatory responses, and provides a more accurate health risk assessment tool. It effectively identifies the degree of microbiota imbalance in patients and its potential role in promoting pulmonary hypertension, which is particularly suitable for early diagnosis and prediction, allowing clinicians to take effective interventions in the early stages of the disease.

[0050] The present invention, by constructing an immune and inflammatory disorder scoring system, directly combines the abnormal immune system and inflammatory response with the PAH risk score, and converts the immune and inflammatory score into a mechanism for the PAH risk score. Based on the data support of a large number of clinical studies, the key role of the inflammatory response in the onset of PAH is accurately quantified. Through this scoring system, high-risk patients, especially individuals with strong inflammatory responses, can be quickly identified, and the possibility of their future occurrence or aggravation of PAH can be assessed, thereby providing a more practical risk assessment tool in clinical practice, simplifying the complex assessment process, and significantly improving the efficiency and accuracy of risk prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1 A schematic diagram of functional modules of an evaluation system according to an embodiment of the present invention;

[0053] Figure 2 Schematic diagram of a microbial community imbalance assessment module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0054] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0055] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0056] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0057] like Figure 1-Figure 2 As shown, the pulmonary hypertension risk assessment system based on data processing includes:

[0058] Gut microbiome data collection module: extracts intestinal microbial DNA data from patients' stool samples, conducts qualitative analysis of the intestinal microbiome through high-throughput sequencing technology, and identifies the relative abundance of the intestinal microbiome;

[0059] Microbial metabolite detection module: detects microbial metabolites related to pulmonary hypertension, including short-chain fatty acids (SCFA), trimethylamine-N-oxide (TMAO), and indole metabolites. Metabolite product concentration analysis is performed based on blood and urine samples to obtain microbial community metabolic concentration data;

[0060] Microbial community imbalance assessment module: Based on the relative abundance and metabolic concentration data of the microbiome, combined with the baseline data of the intestinal microbiome of healthy people (microbiome relative abundance baseline, microbiome metabolic concentration baseline), the machine learning algorithm is used to assess the degree of intestinal microbiome imbalance in the target patient;

[0061] Association analysis module: By constructing a multidimensional association model between the degree of intestinal microbiome imbalance and the immune system and inflammatory status, the potential impact of the degree of microbiome imbalance on the immune system and inflammatory status is identified, and the immune and inflammatory disorder scores are obtained;

[0062] Personalized risk prediction module: Based on the conversion of immune and inflammatory disorder scores, it provides a pulmonary hypertension risk score for a period of time in the future.

[0063] The intestinal microbiome data collection module includes:

[0064] The stool sample processing unit collects and preserves the patient's stool samples, pre-treats them to remove macromolecular impurities, and destroys the cell walls in the samples through mechanical homogenization, centrifugation, and enzymatic treatment to release the DNA of intestinal microorganisms;

[0065] DNA extraction unit, which uses a DNA extraction kit (silica-based membrane method) to extract genomic DNA of intestinal microorganisms from processed fecal samples and ensure high-quality DNA extraction to avoid environmental DNA contamination;

[0066] A high-throughput sequencing unit is used to perform amplicon sequencing or whole-genome sequencing on the extracted intestinal microbial DNA, and to amplify specific regions of the 16S rRNA gene (such as the V3-V4 region) for qualitative analysis of microbial species;

[0067] The data processing and analysis unit performs bioinformatics processing on the sequence data generated by high-throughput sequencing, combines it with the database (Greengenes microbial database) to perform species classification, sequence alignment and abundance analysis, calculates the relative abundance of the microbiome, and identifies the dominant microbiome and potentially imbalanced microbial communities in the patient's intestines.

[0068] The calculation formula for the relative abundance of microorganisms is: Among them, A i represents the relative abundance of the i-th microbial group, S i is the number of valid sequences of the i-th microbial species obtained by sequencing.

[0069] Microbial metabolite detection modules include:

[0070] Sample collection unit: collects metabolites related to pulmonary hypertension from patients' blood and urine samples. The samples are pre-treated by centrifugation and filtration to remove impurities and retain effective metabolites.

[0071] Short-chain fatty acid detection unit: Quantitative analysis of short-chain fatty acids in blood and urine is performed using gas chromatography-mass spectrometry to obtain the specific concentration of short-chain fatty acids;

[0072] Trimethylamine-N-oxide detection unit: Use liquid chromatography-mass spectrometry to detect trimethylamine-N-oxide in blood and urine with high sensitivity, and calculate the concentration of trimethylamine-N-oxide by comparing with the standard curve;

[0073] Indole metabolite detection unit: Utilizes ultra-high performance liquid chromatography and ultraviolet detection technology, combined with quantitative standardization, to detect the concentration of indole metabolites in blood and urine;

[0074] Metabolite concentration data processing unit: The metabolite concentration data obtained by detection are standardized to generate concentration data of short-chain fatty acids, trimethylamine-N-oxide, and indole metabolites as input for subsequent microbiome dysbiosis assessment and pulmonary hypertension risk assessment.

[0075] Zscore standardization converts the concentration data of different metabolites into a standard normal distribution, normalizing them with a mean of 0 and a standard deviation of 1. It is suitable for situations where different metabolites fluctuate greatly in different samples: Among them, Z i represents the standardized value of the ith metabolite, C i is the original metabolite concentration, μ is the average concentration of the metabolite in all samples, and σ is the standard deviation of the metabolite in the samples. Through Zscore standardization, the differences in absolute values ​​of different metabolite concentration data are eliminated, so that the concentration data can be compared on a unified scale.

[0076] The microbial community imbalance assessment module specifically includes:

[0077] Data acquisition unit: Receive the target patient's microbiome relative abundance data and microbial metabolite concentration data, import the intestinal microbiome baseline data of healthy people, the baseline data includes the microbiome relative abundance baseline and metabolite concentration baseline, and standardize the baseline data to ensure that data from different sources have consistent scale and comparability;

[0078] Difference calculation unit: used to calculate the difference between the target patient and the healthy baseline, based on the formula: Among them, D i is the relative deviation of the relative abundance of the i-th microbiota of the target patient from the healthy baseline, αC i is the relative deviation of the concentration of the ith metabolite of the target patient from the healthy baseline, A i and B i are the relative abundance of the i-th microbial species in the target patients and healthy people, respectively, and C i and are the metabolite concentrations of the target patient and healthy baseline, respectively;

[0079] Feature extraction unit: Based on the difference data between target patients and healthy people, key features for evaluating the degree of microbiome imbalance are extracted, including deviations in relative abundance of microorganisms and deviations in metabolite concentrations;

[0080] Sparse regression model unit: The extracted features are evaluated using a sparse regression model, regularization terms are introduced, and key microbiome and metabolite features that affect the degree of disorder are automatically selected, while noise and irrelevant features are suppressed. The model training process is carried out using healthy baseline data and a dataset of patients with known disorders to optimize the model's predictive ability;

[0081] Disorder scoring unit: Based on the output results of the sparse regression model, a gut microbiota disorder score for the target patient is generated. The score is generated based on the weighted contribution of microbial relative abundance and metabolite deviation, and is used to quantify the overall disorder of the gut microbiota and provide key input for subsequent pulmonary hypertension risk assessment.

[0082] The sparse regression model analyzes the extracted features based on Lasso regression, which has the ability to automatically screen key features;

[0083] During the model training process, the data of historical healthy people and patients with known intestinal disorders are used as training sets to help the model learn to identify key microbial and metabolite changes. According to the relative abundance deviation of the patient's microbiome and the concentration deviation of metabolites, the features related to the degree of disorder are determined. Lasso regression assigns corresponding weights to each feature. If a feature is insignificant or has little effect on the disorder assessment, the model will automatically adjust its weight to zero to eliminate its influence. In this way, the model will only retain the microorganisms and metabolites that have the greatest impact on the disorder assessment.

[0084] Lasso regression is expressed as: Among them, Lasso introduces a regularization term to constrain the regression coefficient β i .

[0085] Lasso introduces a regularization term to constrain the regression coefficient β i It is expressed as:

[0086] in, is the predicted dysbiosis score, β 0 is a constant term, X i represents each input feature, which may include the relative abundance deviation of microorganisms D i and metabolite deviation ΔC i , β i is the regression coefficient of each feature in Lasso regression, N is the number of samples, λ is the regularization parameter used to control the sparsity of the model, and Y j is the actual disorder level of the target patient (the actual value in the training data);

[0087] The choice of λ determines which features have coefficients β iwill be reduced to 0. By increasing λ, Lasso regression can automatically screen out the most valuable microbial abundance and metabolite features and suppress noise features. i Contains the relative deviation D calculated from the patient data and the baseline data i and ΔC i , these deviations reflect differences in the patient microbiota from a healthy baseline.

[0088] The association analysis module includes:

[0089] Data input unit: receiving the target patient's intestinal microbiome disorder degree data, including the deviation of microbial relative abundance, the deviation of metabolite concentration, and the biomarker concentration data of the immune system and inflammatory state, including C-reactive protein, interleukin-6, and TNF-α:

[0090] C-reactive protein (CRP), which measures systemic inflammation;

[0091] interleukin-6 (IL-6), representing a proinflammatory cytokine;

[0092] Tumor necrosis factor-α (TNF-α), which reflects the immune system’s response;

[0093] Set the biomarker concentration data to I j , represents the biomarker concentration of the jth immune system or inflammatory state of the target patient;

[0094] Feature mapping unit: By constructing a mapping relationship, the degree of intestinal microbiome imbalance is associated with the concentration of biomarkers of the immune system and inflammatory state, and key interactive features are extracted. The interactive features include the potential correlation between the degree of microbiome imbalance and specific immune responses and inflammatory states, such as the connection between the decrease in the abundance of specific microorganisms and the increase in pro-inflammatory markers. The interactive features of microbial imbalance and immune markers are expressed as: X ij =D i I j , where X ij represents the interaction characteristics between the i-th microbial dysbiosis and the j-th immune or inflammatory marker, D i is the relative abundance deviation of the i-th microbial group, I j is the concentration of the jth immune marker;

[0095] Multidimensional association model construction unit: Use multivariate regression analysis to construct a multidimensional association model to evaluate the potential impact of microbiome dysbiosis on the immune system and inflammatory status, and establish a multidimensional association network based on the interaction between microbiome dysbiosis characteristics and immune markers;

[0096] Potential Impact Identification Unit: Identify the potential impact of intestinal microbiome imbalance on the immune system and inflammatory status through association model analysis. Model output includes identifying the impact of specific microbiome imbalance on immune cell activity and inflammatory marker concentration, revealing the role of microbial imbalance in promoting systemic inflammatory response and immune disorder;

[0097] Result output unit: Quantify the analysis results into specific immune and inflammatory disorder scores to evaluate the impact of microbiome dysbiosis on the patient's immune system and inflammatory response.

[0098] The multiple linear regression model is expressed as: Among them, I j is the jth immune or inflammatory marker concentration of the patient, γ 0 is the constant term of the model, γ i is the characteristic D of the ith microbial imbalance i The associated regression coefficient, ∈, is the error term, which is used to capture the fluctuations not explained by the model;

[0099] The immune and inflammatory dysregulation score was calculated as:

[0100] Among them, S immune is a comprehensive score of immune system and inflammatory disorders, D i is the i-th microbial imbalance feature, I j is the concentration of the jth immune or inflammatory marker, γ i is the regression coefficient associated with the characteristics of microbial dysbiosis; through this score, the overall impact of microbiome dysbiosis on the immune system and inflammatory response can be quantified, providing a basis for subsequent pulmonary hypertension risk assessment.

[0101] The personalized risk prediction module converts the immune and inflammatory disorder scores into pulmonary hypertension risk scores based on a mapping transformation model, expressed as: R PAH =α 0 +α 1 ·S immune , where R PAH is the predicted future pulmonary hypertension risk score, α 0 is a constant term, representing the baseline risk, α 1 are the regression coefficients of the maps, indicating the strength of the effects of immune and inflammatory dysregulation on the risk of PH.

[0102] The immune and inflammatory disorder score can be converted into a pulmonary arterial hypertension (PAH) risk score. The immune system and inflammatory response play a key role in the onset and progression of pulmonary arterial hypertension. By understanding the extent of immune and inflammatory disorders, its potential impact on the pulmonary vascular and cardiopulmonary systems can be assessed. These impacts are closely related to the pathogenesis of PAH. Details are as follows:

[0103] 1. Pathological basis of pulmonary hypertension:

[0104] Pulmonary arterial hypertension (PAH) is a disease characterized by increased pulmonary vascular resistance, vascular remodeling, and vascular stenosis, which can eventually lead to right heart failure. The pathogenesis of PAH is complex, but chronic inflammation and immune disorders play an important role in its progression. The specific manifestations are:

[0105] Vascular inflammation: Chronic inflammation in the pulmonary vessels is one of the core pathological processes of PAH. Immune cells (such as T cells, B cells, macrophages, etc.) infiltrate in the pulmonary vessels, triggering a persistent inflammatory response.

[0106] Endothelial cell dysfunction: Activation of the immune system and inflammatory responses can damage the endothelial cells of the pulmonary arteries, disrupting their ability to regulate vasodilation and constriction, leading to excessive constriction and thickening of the blood vessels.

[0107] Vascular remodeling: Proinflammatory cytokines (such as IL-6 and TNF-α) stimulate the proliferation of vascular smooth muscle cells, leading to thickening of the vascular wall and further exacerbating pulmonary vascular resistance.

[0108] 2. Role of immune system and inflammatory disorders in PAH:

[0109] The immune and inflammatory dysregulation scores reflect the degree of abnormality of the patient's immune system and inflammatory status. These scores can be obtained by measuring key immune and inflammatory markers (such as C-reactive protein, interleukin-6, TNF-α, etc.), indicating the patient's systemic inflammatory response and the severity of immune dysregulation.

[0110] High immune and inflammatory disorder scores: This means that there is a significant pro-inflammatory response in the patient's body, which may directly act on the pulmonary blood vessels, leading to increased endothelial cell damage and vascular remodeling, and increasing the risk of PAH.

[0111] Low immune and inflammatory disorder score: means that the patient's immune and inflammatory status is relatively balanced, and the systemic and local pro-inflammatory responses are relatively low, indicating a lower risk of developing or aggravating PAH.

[0112] Studies have shown that immune system dysregulation is closely related to the pathological process of PAH, especially certain pro-inflammatory cytokines (such as IL-6 and TNF-α) are significantly increased in PAH patients. The following points illustrate the direct connection between immune dysregulation score and PAH risk score:

[0113] Vascular lesions driven by proinflammatory factors: Elevated levels of proinflammatory factors in the disorder score (such as IL-6 and TNF-α) promote the proliferation of vascular smooth muscle cells, accelerate the remodeling process of the pulmonary arteries, lead to increased vascular resistance, and thus increase the risk of PAH.

[0114] Overactivation of immune response: Overactivation of immune cells can impair the barrier function of pulmonary vascular endothelial cells, making blood vessels more susceptible to damage and fibrosis, further promoting the progression of PAH.

[0115] The impact of systemic inflammation on cardiopulmonary function: The higher the immune and inflammatory disorder score, the stronger the systemic inflammatory response. This inflammatory response not only acts on the pulmonary blood vessels, but also affects cardiopulmonary function, making patients more susceptible to symptoms related to pulmonary hypertension.

[0116] Therefore, by converting the immune and inflammatory disorder scores into PAH risk scores, the risk of developing or aggravating PAH in patients in the future can be more accurately predicted based on these quantitative indicators of the degree of inflammation and immune disorder. The immune and inflammatory score reflects the core mechanism of PAH pathogenesis: the immune and inflammatory disorder score can quantify the intensity of the inflammatory response in the patient's body, and this inflammatory response is closely related to the pathological process of pulmonary hypertension; the immune disorder score serves as an early warning: immune and inflammatory disorders usually occur in the early stages of PAH, so it can be used as an early warning indicator. When the patient's clinical symptoms are not yet obvious, by converting to a PAH risk score, it can provide a basis for early intervention.

[0117] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0118] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A pulmonary hypertension risk assessment system based on data processing, characterized in that: include: Gut microbiome data collection module: extracts intestinal microbial DNA data from patients' stool samples, conducts qualitative analysis of the intestinal microbiome through high-throughput sequencing technology, and identifies the relative abundance of the intestinal microbiome; Microbial metabolite detection module: detects microbial metabolites related to pulmonary hypertension, including short-chain fatty acids, trimethylamine-N-oxide, and indole metabolites. Metabolite product concentration analysis is performed based on blood and urine samples to obtain microbial community metabolic concentration data; Microbial community imbalance assessment module: Based on the relative abundance of microbiota and microbiota metabolic concentration data, combined with the baseline data of intestinal microbiota of healthy people, the machine learning algorithm is used to assess the degree of intestinal microbiota imbalance in target patients; Association analysis module: By constructing a multidimensional association model between the degree of intestinal microbiome imbalance and the immune system and inflammatory status, the potential impact of the degree of microbiome imbalance on the immune system and inflammatory status is identified, and the immune and inflammatory disorder scores are obtained; Personalized risk prediction module: Based on the conversion of immune and inflammatory disorder scores, it provides a pulmonary hypertension risk score for a period of time in the future.

2. The pulmonary hypertension risk assessment system based on data processing according to claim 1, characterized in that: The intestinal microbiome data acquisition module comprises: The stool sample processing unit collects and preserves the patient's stool samples, pre-treats them to remove macromolecular impurities, and destroys the cell walls in the samples through mechanical homogenization, centrifugation, and enzymatic treatment to release the DNA of intestinal microorganisms; DNA extraction unit, using a DNA extraction kit to extract genomic DNA of intestinal microorganisms from processed fecal samples; High-throughput sequencing unit, used to perform amplicon sequencing or whole genome sequencing of extracted intestinal microbial DNA, amplifying specific regions of the 16S rRNA gene for qualitative analysis of microbial species; The data processing and analysis unit performs bioinformatics processing on the sequence data generated by high-throughput sequencing, combines it with the database to perform species classification, sequence alignment and abundance analysis, and calculates the relative abundance of the microbiome.

3. The pulmonary hypertension risk assessment system based on data processing according to claim 1, characterized in that: The microbial metabolite detection module comprises: Sample collection unit: collects metabolites related to pulmonary hypertension from patients' blood and urine samples. The samples are pre-treated by centrifugation and filtration to remove impurities and retain effective metabolites. Short-chain fatty acid detection unit: Quantitative analysis of short-chain fatty acids in blood and urine is performed using gas chromatography-mass spectrometry to obtain the specific concentration of short-chain fatty acids; Trimethylamine-N-oxide detection unit: Use liquid chromatography-mass spectrometry to detect trimethylamine-N-oxide in blood and urine with high sensitivity, and calculate the concentration of trimethylamine-N-oxide by comparing with the standard curve; Indole metabolite detection unit: Utilizes ultra-high performance liquid chromatography and ultraviolet detection technology, combined with quantitative standardization, to detect the concentration of indole metabolites in blood and urine; Metabolite concentration data processing unit: standardizes the metabolite concentration data obtained by detection to generate concentration data of short-chain fatty acids, trimethylamine-N-oxide, and indole metabolites.

4. The pulmonary hypertension risk assessment system based on data processing according to claim 1, characterized in that: The microbial community imbalance assessment module specifically includes: Data acquisition unit: receiving the target patient's microbiome relative abundance data and microbial metabolite concentration data, importing the intestinal microbiome baseline data of healthy people, the baseline data includes the microbiome relative abundance baseline and metabolite concentration baseline, and standardizing the baseline data; Difference calculation unit: used to calculate the difference between the target patient and the healthy baseline, based on the formula: Among them, D i is the relative deviation of the relative abundance of the i-th microbiota of the target patient from the healthy baseline, ΔC i is the relative deviation of the concentration of the ith metabolite of the target patient from the healthy baseline, A i and B i are the relative abundances of the i-th microbial species in the target patients and healthy subjects, respectively, and C i and are the metabolite concentrations of the target patient and healthy baseline, respectively; Feature extraction unit: Based on the difference data between target patients and healthy people, key features for evaluating the degree of microbiome imbalance are extracted, including deviations in relative abundance of microorganisms and deviations in metabolite concentrations; Sparse regression model unit: The extracted features are evaluated using a sparse regression model, regularization terms are introduced, and key microbiome and metabolite features that affect the degree of disorder are automatically selected, while suppressing noise and irrelevant features; Dysbiosis scoring unit: Generates the gut microbiota dysbiosis score of the target patient based on the output results of the sparse regression model.

5. The pulmonary hypertension risk assessment system based on data processing according to claim 4, characterized in that: The sparse regression model analyzes the extracted features based on Lasso regression, and Lasso regression has the ability to automatically screen key features; During the model training process, data from historical healthy people and patients with known intestinal disorders were used as training sets to help the model learn to identify key microbial and metabolite changes. Based on the relative abundance deviation of the patient's microbiota and metabolite concentration deviation, the features associated with the degree of disorder were determined, and Lasso regression assigned corresponding weights to each feature.

6. The pulmonary hypertension risk assessment system based on data processing according to claim 5, characterized in that: The Lasso regression is expressed as: Among them, Lasso introduces a regularization term to constrain the regression coefficient β i .

7. The pulmonary hypertension risk assessment system based on data processing according to claim 6, characterized in that: The Lasso introduces a regularization term to constrain the regression coefficient β i It is expressed as: in, is the predicted dysbiosis score, β0 is a constant term, and X i represents each input feature, which may include the relative abundance deviation of microorganisms D i and metabolite deviation ΔC i , β i is the regression coefficient of each feature in Lasso regression, N is the number of samples, λ is the regularization parameter used to control the sparsity of the model, and Y j is the actual degree of disorder in the target patient.

8. The pulmonary hypertension risk assessment system based on data processing according to claim 1, characterized in that: The association analysis module comprises: Data input unit: receiving the target patient's intestinal microbiome disorder degree data, including the deviation of microbial relative abundance, the deviation of metabolite concentration, and the biomarker concentration data of the immune system and inflammatory state, including C-reactive protein, interleukin-6, and TNF-α: Set the biomarker concentration data to I j , represents the biomarker concentration of the jth immune system or inflammatory state of the target patient; Feature mapping unit: By constructing a mapping relationship, the degree of intestinal microbiome imbalance is associated with the concentration of biomarkers of the immune system and inflammatory state, and key interactive features are extracted. The interactive features include the potential correlation between the degree of microbiome imbalance and specific immune responses and inflammatory states. The interactive features of microbial imbalance and immune markers are expressed as: X ij =D i I j , where X ij represents the interaction characteristics between the i-th microbial dysbiosis and the j-th immune or inflammatory marker, D i is the relative abundance deviation of the i-th microbial group, I j is the concentration of the jth immune marker; Multidimensional association model construction unit: Use multivariate regression analysis to construct a multidimensional association model to evaluate the potential impact of microbiome dysbiosis on the immune system and inflammatory status, and establish a multidimensional association network based on the interaction between microbiome dysbiosis characteristics and immune markers; Potential Impact Identification Unit: Identify the potential impact of intestinal microbiome imbalance on the immune system and inflammatory status through association model analysis. Model output includes identifying the impact of specific microbiome imbalance on immune cell activity and inflammatory marker concentrations; Result output unit: Quantify the analysis results into specific immune and inflammatory disorder scores to evaluate the impact of microbiome dysbiosis on the patient's immune system and inflammatory response.

9. The pulmonary hypertension risk assessment system based on data processing according to claim 8, characterized in that: The multivariate linear regression model is expressed as: Among them, I j is the jth immune or inflammatory marker concentration of the patient, γ0 is the constant term of the model, and γ i is the characteristic D of the ith microbial imbalance i The associated regression coefficient, ∈, is the error term, which is used to capture the fluctuations not explained by the model; The immune and inflammatory dysregulation score was calculated as: Among them, S immune is a comprehensive score of immune system and inflammatory disorders, D i is the i-th microbial imbalance feature, I j is the concentration of the jth immune or inflammatory marker, γ i is the regression coefficient associated with the dysbiosis trait.

10. The pulmonary hypertension risk assessment system based on data processing according to claim 9, characterized in that: The personalized risk prediction module converts the immune and inflammatory disorder scores into pulmonary hypertension risk scores based on a mapping transformation model, expressed as: R PAH =α0+α1·S immune , where R PAH is the predicted future PAH risk score, α0 is a constant term representing the baseline risk, and α1 is the mapped regression coefficient representing the intensity of the effect of immune and inflammatory disorders on the risk of PAH.

Images