Method for screening of a marker for predicting the risk of death from pneumonia and a system for predicting the prognosis of pneumonia

Abstract

The present application relates to the field of medical information technology and clinical auxiliary decision-making technology, and discloses a screening method of a pneumonia death risk prediction marker and a pneumonia prognosis prediction system, the system comprising: a data receiving module, configured to receive bone bridge protein level data of a to-be-tested pneumonia patient from at least one terminal; a prediction module, configured to compare the bone bridge protein level data of the to-be-tested pneumonia patient with a bone bridge protein level boundary value, and obtain a death risk prediction result of the to-be-tested patient; and a data output module, configured to send the death risk prediction result of the to-be-tested pneumonia patient to the at least one terminal. The present application determines bone bridge protein as a pneumonia death risk prediction marker, and through detection of the bone bridge protein level in the serum of a to-be-tested pneumonia patient, accurate and timely evaluation of the death risk of the patient can be realized.

Description

Technical Field

[0001] This invention relates to the fields of medical information technology and clinical decision support technology, and in particular to a method for screening biomarkers for predicting the risk of death from pneumonia and a system for predicting the prognosis of pneumonia. Background Technology

[0002] Lower respiratory tract infection (LRTI) is a leading cause of high morbidity and mortality worldwide, with its disease burden particularly severe in children under 5 and the elderly over 60. Acute severe pneumonia (such as severe community-acquired pneumonia, sCAP) is the most life-threatening type, with a mortality rate as high as 40%. Timely and accurate assessment of disease severity and prediction of mortality risk are crucial for effective clinical intervention, rational allocation of medical resources, and improvement of patient prognosis in pneumonia patients, especially those with acute severe pneumonia.

[0003] Currently, clinical stratification of pneumonia severity and prognostic assessment mainly rely on a series of scoring scales, among which the CURB-65 score and the Pneumonia Severity Index (PSI) score are the most widely used. The CURB-65 score assesses five clinical indicators (new-onset altered mental status, blood urea nitrogen level >7 mmol / L, respiratory rate >30 breaths / min, hypotension, and age ≥65 years) to classify patients into different 30-day mortality risk levels (0-1 points for low risk, 2 points for intermediate risk, and ≥3 points for high risk). The PSI score is more complex, integrating nearly 20 variables such as patient age, comorbidities, vital signs, laboratory test results, and imaging findings, and using a weighted score to classify patients into five risk levels: IV.

[0004] While the aforementioned scoring tools play an important role in clinical practice, they have inherent limitations, particularly in assessing complex, rapidly progressing pneumonia cases where their efficacy may be insufficient. Specifically, the core algorithms of these scoring tools (especially the PSI) are highly dependent on the patient's age and chronic comorbidities, leading to a significant underestimation of the severity of illness in young and middle-aged patients without significant underlying diseases (such as some patients with fulminant viral pneumonia). For example, in cases of rapid respiratory failure such as influenza pneumonia or COVID-19, patients may be classified as low-risk on the PSI score due to their youth and lack of comorbidities, even though their condition is actually deteriorating rapidly and requires higher-level monitoring. Furthermore, CURB-65 and PSI scores are essentially static assessments based on a specific point in time (usually at admission), lacking the ability to monitor and provide early warning of the dynamic evolution of a patient's condition. For patients residing in general wards whose condition may deteriorate rapidly, an initial low-risk score may lead to decreased clinical vigilance, resulting in a delay in identifying patients at potential risk of worsening. Summary of the Invention

[0005] This invention provides a method for screening biomarkers for predicting the risk of death from pneumonia and a system for predicting the prognosis of pneumonia, in order to overcome the shortcomings of the prior art.

[0006] This invention provides a method for screening biomarkers for predicting pneumonia mortality risk, comprising:

[0007] Obtain proteomics data from the target population, which includes patients with severe pneumonia who have a poor prognosis and patients with severe pneumonia who have a good prognosis. Based on proteomics data from the target population, differentially expressed proteins targeting pneumonia levels that are upregulated or downregulated were screened. Protein-protein interaction network analysis was performed based on differentially expressed proteins, and the differentially expressed proteins were sorted according to the node connectivity in order to screen core proteins. Based on preset candidate biomarker determination criteria, pneumonia mortality risk predictor biomarkers for predicting prognosis risk of severe pneumonia are determined from the core protein.

[0008] According to a method for screening biomarkers for predicting pneumonia mortality risk provided by the present invention, the step of screening differentially expressed proteins for upregulation and downregulation at pneumonia levels based on proteomics data of a target population includes: Based on proteomics data from the target population, differentially expressed proteins targeting pneumonia-related upregulation and downregulation were screened using pre-defined criteria. These criteria included: Proteins with upregulated levels were screened based on the criteria of Log2FoldChange > 1.0 and p < 0.05. Proteins with downregulated levels were screened based on the criteria of Log2FoldChange < -1.0 and p < 0.05.

[0009] According to the present invention, a method for screening biomarkers for predicting the risk of death from pneumonia is provided. The differentially expressed proteins that are upregulated at the level of pneumonia include SPP1, TIMP1, LCN2, CRP, IGFBP2, CCN2, MMP14, COL4A1, COL4A2, COL6A1, COL6A2, COL6A3, COL18A1, LAMC1, LAMA4, LAMB1, and FN1. The differentially expressed proteins that are downregulated at the level of pneumonia include HSPA4, CALR, EEF1A1, YWHAZ, MMP3, UBB, CST3, B2M, THBS1, THBS2, VCAN, TNC, and SERPINA1.

[0010] According to the method for screening biomarkers for predicting pneumonia mortality risk provided by the present invention, the preset criteria for determining candidate biomarkers include: a) In the protein-protein interaction network, the node connectivity of the candidate biomarkers ranks in the top 20% of all differentially expressed proteins; b) In diffusion network analysis, the candidate markers are located at or directly connected to the core nodes of the functional modules; c) In an independent validation cohort, the serum levels of the candidate biomarkers showed a statistically significant difference between patients with severe pneumonia and those with ordinary pneumonia.

[0011] According to the screening method for pneumonia mortality risk prediction biomarkers provided by the present invention, the pneumonia mortality risk prediction biomarker determined according to the preset candidate biomarker determination criteria is osteopontin.

[0012] The method for screening predictive biomarkers for pneumonia mortality provided by the present invention further includes: Obtain osteopontin level data for the target population; Based on osteopontin level data of the target population, ROC curves were plotted, and the cutoff value of osteopontin level between the severe pneumonia patient group with poor prognosis and the severe pneumonia patient group with good prognosis was determined by the Youden index.

[0013] According to the screening method for predictive biomarkers of pneumonia mortality provided by the present invention, the patients with severe pneumonia with poor prognosis and the patients with severe pneumonia with good prognosis are divided based on the prognostic information of the patients with severe pneumonia within 100 days after admission and diagnosis.

[0014] This invention provides a pneumonia prognosis prediction system, comprising: The data receiving module is used to: receive osteopontin level data from at least one terminal of a pneumonia patient to be tested; The prediction module is used to: compare the osteopontin level data of the pneumonia patient to be tested with the osteopontin level cutoff value mentioned above to obtain the mortality risk prediction result of the patient to be tested; The data output module is used to send the predicted mortality risk of the pneumonia patient to at least one terminal.

[0015] According to the present invention, a pneumonia prognosis prediction system is provided. The step of comparing the osteopontin level data of the pneumonia patient to be tested with the osteopontin level cutoff value according to claim 6 to obtain the mortality risk prediction result of the patient includes: When the osteopontin level in a patient with pneumonia is higher than the cutoff value, the predicted mortality risk is considered high; when the osteopontin level is lower than the cutoff value, the predicted mortality risk is considered low.

[0016] It should be noted that a terminal refers to an input / output device connected to a computer system. Depending on the function, terminals can be divided into various types: smart terminals or intelligent terminals, dumb terminals, interactive terminals or online terminals. Specifically, a terminal can be various mobile communication devices, such as mobile phones and tablets. This article aims to provide users with the function of inputting data and outputting data.

[0017] The present invention also provides the use of substances for detecting osteopontin in samples in the preparation of products for predicting the prognosis of pneumonia patients or for patient stratification.

[0018] In one embodiment, the substance used to detect osteopontin in the sample is an antibody, antigen-binding fragment, or a derivative thereof that specifically recognizes osteopontin.

[0019] In one specific embodiment, the substance used to detect osteopontin in the sample is a monoclonal antibody, polyclonal antibody, or recombinant antibody.

[0020] In one specific embodiment, the substance used to detect osteopontin in the sample is a labeled antibody, and the labeling includes enzyme labeling, fluorescent labeling, chemiluminescent labeling, or colloidal gold labeling.

[0021] In one embodiment, the substance used to detect osteopontin in a sample is used for the detection of osteopontin by immunoturbidimetry, ELISA, chemiluminescence, lateral chromatography, or mass spectrometry.

[0022] In one specific embodiment, the substance for detecting osteopontin in a sample is used to quantitatively detect the concentration of osteopontin in serum or plasma.

[0023] In one embodiment, the detection result of osteopontin was compared with a threshold of 124 ng / mL to predict the prognostic risk of pneumonia patients. Specifically, the threshold was determined by ROC curve analysis combined with the Youden index.

[0024] In one embodiment, the sample is the patient's serum, plasma, whole blood, bronchoalveolar lavage fluid, or pleural effusion; preferably, the sample is peripheral blood serum collected from a pneumonia patient.

[0025] In one embodiment, the prognosis is a prediction of the severity of a pneumonia patient's condition. Specifically, the prognosis is a prediction of the risk of death of a pneumonia patient within 100 days of hospital admission and diagnosis. Specifically, the risk of death of a pneumonia patient within 100 days of hospital admission and diagnosis is either high or low.

[0026] In one specific embodiment, the prognosis is determined by binary risk stratification based on osteopontin levels being above or below a threshold. Specifically, the stratification result is either a good prognosis or a poor prognosis.

[0027] This invention identifies osteopontin as a predictive biomarker for pneumonia mortality risk. By detecting the level of osteopontin (OPN, encoded by the SPP1 gene, sometimes referred to as SPP1 protein in this paper) in the serum of pneumonia patients, accurate and timely assessment of the severity of the patient's condition and mortality risk can be achieved. Compared with existing technologies, the screening method for predictive biomarkers of pneumonia mortality risk and the pneumonia prognosis prediction system provided by this invention can bring at least the following beneficial effects: This invention reveals for the first time the close association between serum osteopontin levels and the severity and prognosis of pneumonia (such as the risk of death within 100 days). The expression level of this biomarker is independent of pathogen type and is not significantly affected by the patient's long-term chronic diseases or underlying physical conditions. Therefore, it can more objectively and sensitively reflect the real-time state of acute inflammation and tissue damage, and is particularly suitable for risk warning of severe pneumonia patients in middle-aged and young adults who are easily missed by traditional scoring methods, effectively filling the gaps in existing clinical assessment indicators.

[0028] The system provided by this invention, by detecting serum osteopontin, a biomarker that can be conveniently and rapidly measured repeatedly, provides clinicians with a tool for dynamically monitoring the evolution of pneumonia. It can indicate the adverse prognostic trend of pneumonia patients earlier, assist clinicians in adjusting treatment plans and monitoring levels in a timely manner, and thus is expected to improve patient prognosis and rationally allocate intensive care resources.

[0029] This invention requires only serum samples from pneumonia patients for osteopontin detection, making sampling convenient and the detection rapid. It significantly simplifies the assessment process, reduces the workload of clinicians in data collection and complex calculations, and improves assessment efficiency. Furthermore, peripheral blood sampling is a routine non-invasive or minimally invasive procedure, resulting in high patient acceptance and compliance, making it easy to promote and apply in medical institutions at all levels, especially in outpatient, emergency, and general wards.

[0030] Because the upregulation of osteopontin expression is associated with severe pathophysiological processes of pneumonia (such as excessive inflammatory response and tissue damage) rather than with specific pathogen types, this invention is applicable to community-acquired pneumonia and acute severe pneumonia caused by various pathogens (including viruses, bacteria, etc.), and has a wide range of clinical applications. This system can specifically assist in predicting the risk of death within 100 days in patients with acute severe pneumonia; its predictive ability is independent of and may surpass that of the traditional PSI score, providing powerful supplementary information for clinical decision-making.

[0031] In summary, this invention provides a pneumonia prognosis prediction system that is easy to operate, provides objective results, offers timely early warning, and has a wide range of applications. It can effectively overcome the limitations of existing scoring tools, especially the insufficient identification of non-elderly patients with severe pneumonia without comorbidities and the defects of static assessment. It provides an important technical means for early identification of high-risk patients, implementation of precise intervention, and improvement of clinical prognosis. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0033] Figure 1 This is a graph showing the differential expression of peripheral serum proteins in patients with severe pneumonia. The left side is a heatmap of protein expression, and the right side is the fold change (Log2FoldChange) of peripheral serum protein expression in patients with severe pneumonia.

[0034] Figure 2 This is a protein-protein interaction (PPI) network diagram of differentially expressed proteins in serum and key proteins. The left side shows the PPI network diagram, the middle shows the ranking of network parameters (degree) of key proteins, and the right side shows the derived network diagram of SPP1-related proteins.

[0035] Figure 3 The proteomics results show that SPP1 levels are significantly upregulated in the serum of patients with severe pneumonia and poor prognosis.

[0036] Figure 4The results showed that ELISA confirmed a significant upregulation of peripheral serum SPP1 levels in patients with severe pneumonia.

[0037] Figure 5 This study demonstrates the predictive role of SPP1 in adverse 100-day outcomes in patients with severe pneumonia. Detailed Implementation

[0038] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, embodiments of this invention, and should not be construed as limiting the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention. In the description of this invention, it should be understood that the terminology used is for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0039] The present invention provides a method for screening predictive biomarkers for pneumonia mortality risk, which may include: S110. Obtain proteomics data of the target population, which includes patients with severe pneumonia who have a poor prognosis and patients with severe pneumonia who have a good prognosis.

[0040] S120. Based on the proteomics data of the target population, and using preset conditions, screen for differentially expressed proteins that are upregulated or downregulated in pneumonia cases. These preset conditions include: Proteins with upregulated levels were screened based on the criteria of Log2FoldChange > 1.0 and p < 0.05. Proteins with downregulated levels were screened based on the criteria of Log2FoldChange < -1.0 and p < 0.05.

[0041] In this embodiment, the core proteins include the following two categories: differentially regulated proteins for pneumonia levels, including SPP1, TIMP1, LCN2, CRP, IGFBP2, CCN2, MMP14, COL4A1, COL4A2, COL6A1, COL6A2, COL6A3, COL18A1, LAMC1, LAMA4, LAMB1, and FN1; and differentially regulated proteins for pneumonia levels, including HSPA4, CALR, EEF1A1, YWHAZ, MMP3, UBB, CST3, B2M, THBS1, THBS2, VCAN, TNC, and SERPINA1.

[0042] S130. Perform protein-protein interaction network analysis based on differentially expressed proteins, and sort the differentially expressed proteins according to the node connectivity in order to screen core proteins.

[0043] S140. Based on preset candidate biomarker determination criteria, determine pneumonia mortality risk predictive biomarkers from the core protein for predicting the prognosis risk of severe pneumonia, wherein the preset candidate biomarker determination criteria may include: a) In the protein-protein interaction network, the node connectivity of the candidate markers ranks in the top 20% of all differentially expressed proteins; b) In diffusion network analysis, the candidate markers are located at or directly connected to the core nodes of the functional modules; c) In an independent validation cohort, the serum levels of the candidate biomarkers showed a statistically significant difference between patients with severe pneumonia and those with ordinary pneumonia.

[0044] In this embodiment, osteopontin is identified as the predictive biomarker for pneumonia mortality risk based on preset candidate biomarker determination criteria.

[0045] This embodiment aims to identify patients with severe pneumonia who are prone to poor prognosis and to assist in early clinical intervention. Peripheral serum samples were collected from patients with severe pneumonia and control patients with ordinary pneumonia. Proteomics techniques (data-independent acquisition, DIA) were used to screen for serum proteins that were significantly upregulated in patients with severe pneumonia. Key proteins were analyzed using protein-protein interaction (PPI) networks, and proteins that could predict prognosis were screened by correlating with 100-day prognostic information of patients with severe pneumonia. SPP1 was identified as a protein that was significantly upregulated in the peripheral serum of patients with severe pneumonia. Figure 1 It occupies a central position in the interactive network. Figure 2 Furthermore, it is significantly enriched in the serum of patients with poor prognosis, suggesting its potential as a serum biomarker for predicting poor prognosis. Figure 3 ).

[0046] This embodiment aims to verify the predictive role of SPP1 in poor prognosis of severe pneumonia. Two independent cohorts of pneumonia patients were recruited as a training set and a validation set. Peripheral serum samples were collected upon admission, and serum SPP1 expression was detected using an ELISA kit. SPP1 was significantly expressed in the serum of patients with severe pneumonia (…). Figure 4 In particular, after collecting 100-day prognostic information, it was found that SPP1 has a good predictive effect on poor prognosis of severe pneumonia. Figure 5 ).

[0047] 1. Screening for differentially expressed proteins in patients with severe pneumonia using proteomics: In this embodiment, peripheral serum samples were collected from 17 patients with severe pneumonia and 7 patients with mild pneumonia on the first day of admission and sent to Beijing Novogene Technology Co., Ltd. for DIA proteomics analysis. The detection method is as follows: The concentrated protein solution was obtained by centrifugation at 14000g for 15 min using an ultrafiltration tube and extracted in a 1.5ml centrifuge tube. The protein concentration of the sample was calculated using the Bradford Protein Quantitative Reduction Kit according to the standard curve. 20μg of each protein sample was subjected to 12% SDS-PAGE gel electrophoresis, followed by Coomassie Brilliant Blue R-250 staining until clear bands were observed. The protein samples were then subjected to proteolytic digestion and elution, and the filtrate was collected and lyophilized. The proteolytically digested samples were analyzed using the Vanquish™ Neo UHPLC-Astral ultra-high performance liquid chromatography-mass spectrometry (Thermo Scientific) method with data-independent acquisition (DIA), and raw mass spectrometry data (.raw) were collected. The raw data were analyzed and the species library was searched using DIA-NN software (Direct DIA). Only peptide spectrometry matches (PSMs) (confidence > 99%) were retained as clean data. The PSMs were analyzed and the species library was searched using DIA-NN software. Retention time calibration was performed using endogenous peptides added to the samples, and statistical analysis was conducted on the protein quantification results. Upregulated proteins were screened based on Log2FoldChange > 1.0 and p < 0.05, while downregulated proteins were defined as those with Log2FoldChange < -1.0 and p < 0.05. Differentially expressed proteins were Z-score converted and displayed as a heatmap, showing the Log2FoldChange values. Core proteins were selected based on network parameters (Degree) using PPI analysis. A bar chart was used to compare SPP1 values ​​in severe pneumonia patients and controls in the DIA data.

[0048] 2. ELISA verification of protein expression: This embodiment recruits patients with severe pneumonia. The inclusion criteria are as follows: (1) Age greater than 18 years; (2) Meeting the diagnostic criteria for severe acute lower respiratory tract infection: Meeting one major criterion or no less than three minor criteria: A. Major criterion: Invasive mechanical ventilation; septic shock requiring vasopressors; B. Minor criteria: Respiratory rate > 30 breaths / min; Oxygenation index ≤ 250 mmHg; Multiple lobar infiltration; Altered consciousness and / or disorientation; Uremia (serum urea nitrogen ≥ 20 mg / dL); Leukopenia (white blood cell count < 4 10 9 / L); Thrombocytopenia (platelet count <100 109 / L); hypothermia (below 36℃); hypotension (requiring aggressive fluid resuscitation). Exclusion criteria: HIV patients; patients with hematologic malignancies and other immunodeficiency (patients with hematologic malignancies or solid tumors who have received chemotherapy within 3 months; solid organ or bone marrow transplants; graft-versus-host disease; use of immunosuppressive drugs or use of prednisone ≥20mg / day (or equivalent dose of other hormones) for more than 30 days at the onset of the disease; HIV infection with peripheral blood T cell subset CD4 cell count <200 / mm³. 3 (Primary / hereditary immunodeficiency diseases). Simultaneously, pneumonia patients with mild or no infection symptoms admitted during the same period were selected as a control group.

[0049] The first serum sample was collected from enrolled patients after admission and stored at -80°C.

[0050] Serum SPP1 levels were measured using a human osteopontin assay kit (Thermo), following the manufacturer's instructions.

[0051] Prognostic information was collected from enrolled patients within 100 days of admission and diagnosis. Based on this information, patients were divided into a poor prognosis group and a good prognosis group. The threshold for the ROC curve parameters in the training set was set to 124.0 ng / ml. Sensitivity, specificity, positive predictive value, and negative predictive value were calculated for both the training and validation sets. Figure 5 ).

[0052] Table 1. Patient characteristics information included in the study cohort

[0053] Osteopontin levels were significantly upregulated in the serum of patients with poor prognosis. SPP1 protein is a highly phosphorylated secreted glycoprotein that enhances the function of macrophages and T cells during inflammation. As a multifunctional protein, it plays an important role in cardiovascular disease, cancer, diabetes, and kidney stones, as well as in inflammation, biomineralization, cell viability, and wound healing.

[0054] After identifying osteopontin as a predictive biomarker for pneumonia mortality risk, a pneumonia prognosis prediction system can be developed, including: The data receiving module is used to: receive osteopontin level data from at least one terminal of a pneumonia patient to be tested; The prediction module is used to: compare the osteopontin level data of the pneumonia patient to be tested with the osteopontin level cutoff value according to claim 6, and obtain the mortality risk prediction result of the patient to be tested; The data output module is used to send the predicted mortality risk of the pneumonia patient to at least one terminal.

[0055] This invention provides a pneumonia prognosis prediction system that utilizes peripheral serum samples from pneumonia patients. The system is simple to obtain, easy to operate, and has high patient compliance. Clinical operation is straightforward, significantly reducing the workload of complex scoring procedures. Based on previous studies showing that SPP1 expression is independent of pathogen, this embodiment recruits pneumonia patients regardless of pathogen type, making it widely applicable and highly practical. It can predict poor prognosis in severe pneumonia at an early stage and assist in clinical identification of high-risk patients. This invention is independent of long-term chronic diseases and physical conditions, and has a strong predictive ability for severe pneumonia outbreaks in young and middle-aged adults, filling the gaps in commonly used indicators such as PSI.

[0056] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0057] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for screening biomarkers for predicting pneumonia mortality risk, characterized in that, include: Obtain proteomics data from the target population, which includes patients with severe pneumonia who have a poor prognosis and patients with severe pneumonia who have a good prognosis. Based on proteomics data from the target population, differentially expressed proteins targeting pneumonia levels that are upregulated or downregulated were screened. Protein-protein interaction network analysis was performed based on differentially expressed proteins, and the differentially expressed proteins were sorted according to the node connectivity in order to screen core proteins. Based on preset candidate biomarker determination criteria, pneumonia mortality risk predictor biomarkers for predicting prognosis risk of severe pneumonia are determined from the core protein.

2. The method for screening predictive biomarkers for pneumonia mortality risk according to claim 1, characterized in that, The screening of differentially expressed proteins, based on proteomics data from the target population and targeting up- and down-regulated levels in pneumonia, includes: Based on proteomics data from the target population, differentially expressed proteins targeting pneumonia-related upregulation and downregulation were screened using pre-defined criteria. These criteria included: Proteins with upregulated levels were screened based on the criteria of Log2FoldChange > 1.0 and p < 0.

05. Proteins with downregulated levels were screened based on the criteria of Log2FoldChange < -1.0 and p < 0.

05.

3. The method for screening predictive biomarkers for pneumonia mortality risk according to claim 1, characterized in that, The core proteins include: differentially regulated proteins for pneumonia levels, including SPP1, TIMP1, LCN2, CRP, IGFBP2, CCN2, MMP14, COL4A1, COL4A2, COL6A1, COL6A2, COL6A3, COL18A1, LAMC1, LAMA4, LAMB1, and FN1; and differentially regulated proteins for pneumonia levels, including HSPA4, CALR, EEF1A1, YWHAZ, MMP3, UBB, CST3, B2M, THBS1, THBS2, VCAN, TNC, and SERPINA1.

4. The method for screening predictive biomarkers for pneumonia mortality risk according to claim 1, characterized in that, The preset candidate marker determination criteria include: a) In the protein-protein interaction network, the node connectivity of the candidate markers ranks in the top 20% of all differentially expressed proteins; b) In diffusion network analysis, the candidate markers are located at or directly connected to the core nodes of the functional modules; c) In an independent validation cohort, the serum levels of the candidate biomarkers showed a statistically significant difference between patients with severe pneumonia and those with ordinary pneumonia.

5. The method for screening predictive biomarkers for pneumonia mortality risk according to claim 1, characterized in that, Based on the pre-defined criteria for candidate biomarkers, osteopontin was identified as the predictive biomarker for pneumonia mortality risk.

6. The method for screening predictive biomarkers for pneumonia mortality risk according to claim 5, characterized in that, Also includes: Obtain osteopontin level data for the target population; Based on osteopontin level data of the target population, ROC curves were plotted, and the cutoff value of osteopontin level between the severe pneumonia patient group with poor prognosis and the severe pneumonia patient group with good prognosis was determined by the Youden index.

7. The method for screening predictive biomarkers for pneumonia mortality risk according to any one of claims 1-6 is characterized in that, The groups of patients with severe pneumonia who have poor prognosis and those with good prognosis are divided based on prognostic information within 100 days of admission and diagnosis.

8. A pneumonia prognosis prediction system, characterized in that, include: The data receiving module is used to: receive osteopontin level data from at least one terminal of a pneumonia patient to be tested; The prediction module is used to: compare the osteopontin level data of the pneumonia patient to be tested with the osteopontin level cutoff value according to claim 6, and obtain the mortality risk prediction result of the patient to be tested; The data output module is used to send the predicted mortality risk of the pneumonia patient to at least one terminal.

9. The pneumonia prognosis prediction system according to claim 8, characterized in that, The step of comparing the osteopontin level data of the pneumonia patient to be tested with the osteopontin level cutoff value according to claim 6 to obtain the mortality risk prediction result of the patient includes: When the osteopontin level in a patient with pneumonia is higher than the cutoff value, the predicted mortality risk is considered high; when the osteopontin level is lower than the cutoff value, the predicted mortality risk is considered low.

10. Use of substances for detecting osteopontin in samples in the preparation of products for predicting the prognosis of pneumonia patients or for patient stratification.

Images