A method and system for evaluating early enteral nutrition intolerance in severe pancreatitis based on multi-dimensional indicators

Abstract

The present invention discloses a method and system for evaluating early enteral nutrition intolerance of severe pancreatitis based on multidimensional indicators, and relates to the field of enteral nutrition tolerance technology. The method monitors the intestinal microbial environmental parameters of patients in real time through non-invasive detection equipment, and combines a multidimensional indicator parameter set to accurately evaluate the early enteral nutrition tolerance of patients with severe pancreatitis. In steps S1 to S3, after data preprocessing and data fusion, the preliminary nutritional tolerance risk value and correction value are calculated by the tolerance assessment model, and the nutritional tolerance of the patient can be accurately assessed, providing a scientific basis for medical staff to formulate personalized nutritional support plans, and finally obtaining a comprehensive tolerance assessment value. Medical staff can perform a primary assessment of the patient's enteral nutrition based on this value, and classify the levels based on the assessment results. The overall nutritional status and recovery speed of the patient are improved by adjusting the nutritional support plan in a personalized manner.

Description

Technical Field

[0001] The present invention relates to the technical field of enteral nutrition tolerance, and specifically to a method and system for evaluating early enteral nutrition intolerance in severe pancreatitis based on multi-dimensional indicators. Background Art

[0002] In the medical field, intensive care and emergency care are important links in ensuring the life safety of critically ill patients.

[0003] Especially in the treatment of severe pancreatitis, early assessment of the patient's enteral nutrition tolerance is crucial.

[0004] In recent years, with the advancement of multidimensional monitoring technology, researchers have begun to explore the use of physiological parameters and microbial environmental indicators to comprehensively assess the nutritional status of patients, thereby providing more accurate intervention plans for clinical practice.

[0005] The early enteral nutrition intolerance assessment method for severe pancreatitis based on multi-dimensional indicators came into being in this context, aiming to achieve dynamic assessment of patients' nutritional tolerance through real-time data monitoring and analysis.

[0006] At this stage, current evaluation methods mainly rely on traditional biochemical indicators and single clinical manifestations, and often lack a comprehensive understanding of the patient's overall physiological state.

[0007] This one-sided assessment method can easily lead to misjudgment of the patient's enteral nutrition tolerance, thereby affecting the effectiveness and safety of nutritional support.

[0008] In addition, the lack of real-time monitoring and data fusion means makes it difficult for clinicians to capture subtle changes in the patient's condition in a timely manner, creating the risk of delayed intervention and causing serious consequences such as malnutrition and complications for the patient. Summary of the invention

[0009] In view of the deficiencies in the prior art, the present invention provides a method and system for evaluating early enteral nutrition intolerance in severe pancreatitis based on multi-dimensional indicators, which solves the problems mentioned in the background technology.

[0010] To achieve the above objectives, the present invention is implemented by the following technical scheme: comprising the following steps:

[0011] S1. Real-time detection of the intestinal microbial environmental parameters of the patient by non-invasive detection equipment, and transmission of the intestinal microbial environmental parameters to the cloud server by building a cloud server, and data preprocessing and data fusion of the intestinal microbial environmental parameters in the cloud server to obtain a multi-dimensional indicator parameter set;

[0012] S2. constructing a tolerance assessment model, and inputting a multidimensional indicator parameter set into the tolerance assessment model to obtain a preliminary nutritional tolerance risk value, and based on the preliminary nutritional tolerance risk value, performing correction and adjustment to obtain a correction value;

[0013] S3, integrating the multidimensional indicator parameter set and the correction value for comprehensive calculation to obtain a comprehensive tolerance assessment value, and performing a primary assessment of the patient's enteral nutrition according to the numerical range of the comprehensive tolerance assessment value, and grading the patient based on the assessment result;

[0014] S4. If the primary assessment shows primary tolerance, a multivariate regression model is used to output a predicted comprehensive tolerance assessment value, and a secondary assessment is performed based on the output of the predicted comprehensive tolerance assessment value to analyze the patient's risk changes and intervention effects;

[0015] S5. Based on the secondary evaluation and analysis results, an early warning is automatically triggered, and feedback adjustment suggestions and early warning information are sent to the mobile terminal of medical staff through the cloud server, and intervention suggestions and historical trend charts are generated.

[0016] Preferably, said S1 includes S11 and S12;

[0017] S11. Real-time collection of intestinal microbial environmental parameters of patients through ICU monitoring system and non-invasive detection equipment, wherein the intestinal microbial environmental parameters include intestinal microbial diversity index, blood microcirculation rate, electrogastrogram spectrum factor, peritoneal pressure fluctuation index, intestinal pH dynamic change rate, heart rate variability variance, serum albumin concentration and C-reactive protein concentration;

[0018] Non-invasive detection devices include non-invasive stool detectors, optical sensors, non-invasive electrodes, abdominal pressure sensors, and wearable biosensors;

[0019] The ICU monitoring system is connected to the cloud server through the API application program interface, and the serum albumin concentration and C-reactive protein concentration collected by the ICU monitoring system are automatically sent to the cloud server. At the same time, the intestinal microbial diversity index, blood microcirculation rate, electrogastrogram spectrum factor, peritoneal pressure fluctuation index, intestinal pH dynamic change rate and heart rate variability variance are manually input into the cloud server;

[0020] S12, integrating the input intestinal microbial environmental parameters in the cloud server to perform data preprocessing and data fusion;

[0021] The data preprocessing removes short-term outliers in the intestinal microbial environmental parameters by applying moving mean filtering and median filtering, and then uses wavelet change denoising method to denoise and smooth the intestinal microbial environmental parameters. Finally, the intestinal microbial environmental parameters are standardized to eliminate the dimension effect;

[0022] The data fusion is used to fuse and summarize the intestinal microbial environmental parameters after data preprocessing to obtain a multidimensional indicator parameter set.

[0023] Preferably, S2 includes S21 and S22;

[0024] S21. By constructing a tolerance assessment model, the intestinal microbial diversity index, the blood microcirculation rate, the electrogastrogram spectrum factor, the peritoneal pressure fluctuation index, the intestinal pH dynamic change rate and the heart rate variability variance manually input into the cloud server are substituted into the tolerance assessment model to calculate and output the preliminary nutritional tolerance risk value, and conduct a preliminary analysis of the patient's enteral nutrition tolerance.

[0025] Preferably, S22, based on a preliminary analysis of the patient's enteral nutrition tolerance, correction and adjustment are performed based on the output preliminary nutrition tolerance risk value, combined with the standard deviation of the intestinal microbial diversity index, the heart rate variability variance and the serum albumin concentration to generate a correction value.

[0026] Preferably, said S3 includes S31 and S32;

[0027] S31. Based on the correction value obtained after correction and adjustment, the C-reactive protein concentration, the average value of the peritoneal pressure fluctuation index and the peak value of the blood microcirculation rate in the multidimensional indicator parameter set are integrated to comprehensively output a comprehensive tolerance assessment value to comprehensively analyze the patient's enteral nutrition tolerance.

[0028] Preferably, S32, using the statistical method ROC curve analysis to determine the demarcation point between the comprehensive tolerance assessment value and the clinical result, selecting 0.5 and 1.0 as the tolerance value range, and then based on the output result of the comprehensive tolerance assessment value, performing a primary assessment on the patient's enteral nutrition, and based on the primary assessment result, grading the patient's enteral nutrition, the specific assessment content is as follows;

[0029] When the comprehensive tolerance assessment value is less than 0.5, it is classified as first-level tolerance, and the existing nutritional support plan is maintained and monitored regularly;

[0030] When 0.5≤comprehensive tolerance assessment value<1.0, it is classified as secondary tolerance, and the monitoring frequency is increased by 50%;

[0031] When the comprehensive tolerance assessment value is ≥1.0, it is divided into level three tolerance. At this time, the first warning information is generated and sent to the terminal device of the medical staff, prompting immediate enteral nutrition tolerance intervention.

[0032] Preferably, said S4 includes S41 and S42;

[0033] S41. When the patient is assessed to have first-level tolerance based on the first-level assessment, a multivariate regression analysis algorithm is used to combine the comprehensive tolerance assessment value, serum albumin concentration, and heart rate variability variance to predict the comprehensive tolerance assessment value within the next 4 hours and obtain the predicted comprehensive tolerance assessment value.

[0034] Preferably, S42, the predicted comprehensive tolerance assessment value and the tolerance value range are subjected to secondary assessment, the patient's enteral nutrition tolerance in the analysis and prediction stage is analyzed, and a response level and warning are generated based on the secondary assessment results. The specific assessment contents are as follows;

[0035] When the predicted comprehensive tolerance assessment value is less than 0.5, it is classified as level 1 tolerance, no warning information is generated, and continuous monitoring is performed;

[0036] When 0.5≤predicted comprehensive tolerance assessment value<1.0, it is classified as secondary tolerance, and a second warning message is generated and sent to the terminal device of the medical staff to prompt the patient's enteral nutrition tolerance and make planned interventions;

[0037] When the predicted comprehensive tolerance assessment value is ≥1.0, it is divided into level three tolerance. At this time, the first warning information is generated and sent to the terminal device of the medical staff, prompting them to immediately change the current enteral nutrition tolerance treatment.

[0038] Preferably, said S5 includes S51 and S52;

[0039] S51. Based on the secondary evaluation and analysis results, if the predicted comprehensive tolerance assessment value is ≥1.0 for more than 8 hours, an early warning is automatically triggered and sent to the terminal device of the medical staff, and a segmented feedback adjustment suggestion is generated, including:

[0040] Phase 1: Reduce the standard nutrient concentration by 20%;

[0041] Phase II: Change to an intermittent infusion method, divided into multiple doses for batch injection;

[0042] Phase 3: Add supplemental fluid and electrolyte support to balance water and electrolytes;

[0043] S52. After the segmented feedback adjustment suggestions are generated, the predicted comprehensive tolerance assessment values ​​and the comprehensive tolerance assessment values ​​obtained in the current and predicted stages are used to generate a historical trend graph.

[0044] An early enteral nutrition intolerance assessment system for severe pancreatitis based on multi-dimensional indicators, comprising a data input module, a multi-dimensional indicator analysis module, a grade classification module, a tolerance prediction module and an alarm feedback module;

[0045] The data input module detects the intestinal microbial environmental parameters of the patient in real time through non-invasive detection equipment, and transmits the intestinal microbial environmental parameters to the cloud server by building a cloud server, and performs data preprocessing and data fusion on the intestinal microbial environmental parameters in the cloud server to obtain a multi-dimensional indicator parameter set;

[0046] The multidimensional index analysis module constructs a tolerance assessment model and inputs the multidimensional index parameter set into the tolerance assessment model to obtain a preliminary nutritional tolerance risk value, and performs correction and adjustment on the basis of the preliminary nutritional tolerance risk value to obtain a correction value;

[0047] The grading module performs comprehensive calculation by integrating the multidimensional indicator parameter set and the correction value to obtain a comprehensive tolerance assessment value, and performs a primary assessment of the patient's enteral nutrition according to the numerical range of the comprehensive tolerance assessment value, and grades the patient based on the assessment result;

[0048] When the tolerance prediction module determines that the patient has first-level tolerance through a second-level assessment, a multivariate regression model is used for analysis to output a predicted comprehensive tolerance assessment value, and a second-level assessment is performed based on the output result of the predicted comprehensive tolerance assessment value to analyze the patient's risk changes and intervention effects;

[0049] The alarm feedback module automatically triggers an early warning based on the secondary evaluation and analysis results, sends feedback adjustment suggestions and early warning information to the medical staff's mobile terminal through the cloud server, and generates intervention suggestions and historical trend charts.

[0050] The present invention provides a method and system for evaluating early enteral nutrition intolerance in severe pancreatitis based on multi-dimensional indicators.

[0051] It has the following beneficial effects:

[0052] (1) This method uses non-invasive detection equipment to monitor the patient's intestinal microbial environmental parameters in real time. Combined with a multidimensional indicator parameter set, it can accurately assess the early enteral nutrition tolerance of patients with severe pancreatitis.

[0053] In steps S1 to S3, after data preprocessing and data fusion, the preliminary nutritional tolerance risk value and correction value are calculated through the tolerance assessment model, which can accurately assess the patient's nutritional tolerance, provide a scientific basis for medical staff to formulate personalized nutritional support plans, and finally obtain a comprehensive tolerance assessment value.

[0054] Medical staff can conduct a primary assessment of the patient's enteral nutrition based on this value and classify the patient into different levels based on the assessment results.

[0055] By personalizing the nutritional support plan and optimizing the nutritional composition and infusion method, the risk of intolerance can be effectively reduced and the patient's overall nutritional status and recovery speed can be improved.

[0056] (2) This method uses a grading assessment system to classify the patient's enteral nutrition tolerance based on the output results of the comprehensive tolerance assessment value during the assessment process.

[0057] When the evaluation result is level 3 tolerance, the system automatically generates the first warning information, prompting medical staff to intervene in the patient immediately.

[0058] Sending feedback adjustment suggestions and early warning information to medical staff's mobile terminals through cloud servers not only improves the response speed to patients' nutritional intolerance risks, but also ensures that the medical team can take corresponding measures in time, such as adjusting the nutritional infusion rate or composition, avoiding the occurrence of serious complications, and improving patients' safety and comfort.

[0059] (3) This method provides strong data support for medical staff by generating historical trend charts to record the patient's comprehensive tolerance assessment values ​​and their changes.

[0060] Using the statistical method ROC curve analysis, the patient's nutritional tolerance can be continuously monitored and key changing trends can be identified.

[0061] When the predicted comprehensive tolerance assessment value exceeds the set threshold, the system will trigger an automatic alarm and make segmented feedback adjustment suggestions.

[0062] This dynamic monitoring mechanism can reflect the patient's physiological state in real time, ensure the continuous optimization of the nutritional support plan, improve the patient's recovery effect and quality of life, and also provide important data basis for future clinical research. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a flow chart of a method for evaluating early enteral nutrition intolerance in severe pancreatitis based on multi-dimensional indicators according to the present invention;

[0064] Figure 2 It is an evaluation effect table of a set of test data in a method for evaluating early enteral nutrition intolerance in severe pancreatitis based on multi-dimensional indicators of the present invention;

[0065] Figure 3 The present invention is a schematic diagram of the steps of an early enteral nutrition intolerance assessment system for severe pancreatitis based on multi-dimensional indicators. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] Example 1

[0068] like Figure 1 As shown, the present invention provides a method for evaluating early enteral nutrition intolerance in severe pancreatitis based on multi-dimensional indicators. To achieve the above purpose, the present invention is implemented through the following technical scheme: comprising the following steps:

[0069] S1. Real-time detection of the patient's intestinal microbial environmental parameters by non-invasive detection equipment, and transmission of the intestinal microbial environmental parameters to a cloud server, where data preprocessing and data fusion are performed on the intestinal microbial environmental parameters to obtain a multi-dimensional indicator parameter set;

[0070] S2. Input the multidimensional indicator parameter set into the tolerance assessment model to obtain a preliminary nutritional tolerance risk value R0, and perform correction and adjustment on the basis of the preliminary nutritional tolerance risk value R0 to obtain a correction value Radj;

[0071] S3. Integrate the multidimensional indicator parameter set and the correction value Radj for comprehensive calculation to obtain the comprehensive tolerance assessment value RVI, and perform a primary assessment of the patient's enteral nutrition according to the numerical range of the comprehensive tolerance assessment value RVI, and classify the patients based on the assessment results;

[0072] S4. If the primary assessment shows primary tolerance, a multivariate regression model analysis is used to output the predicted comprehensive tolerance assessment value RVI (t+4). Based on the output of the predicted comprehensive tolerance assessment value RVI (t+4), a secondary assessment is performed to analyze the patient's risk changes and intervention effects;

[0073] S5. Based on the secondary evaluation and analysis results, an early warning is automatically triggered, and feedback adjustment suggestions and early warning information are sent to the mobile terminal of medical staff through the cloud server, and intervention suggestions and historical trend charts are generated.

[0074] In this embodiment, the method first monitors the patient's intestinal microbial environmental parameters in real time through non-invasive detection equipment, and transmits these parameters to a cloud server.

[0075] This process ensures the timeliness and reliability of the data.

[0076] Subsequently, data preprocessing and fusion are performed in the cloud server to generate a multidimensional indicator parameter set.

[0077] This step not only reduces the complexity of data processing, but also improves data quality, laying a solid foundation for subsequent analysis.

[0078] Secondly, the multidimensional indicator parameter set was input into the tolerance assessment model to obtain the preliminary nutritional tolerance risk value R0, and on this basis, correction and adjustment were performed to obtain the correction value Radj.

[0079] This process makes the assessment more refined, can effectively address individual differences among patients, and improves the accuracy of the assessment results.

[0080] Third, the multidimensional index parameter set and the correction value Radj are integrated for comprehensive calculation to obtain the comprehensive tolerance assessment value RVI.

[0081] According to the numerical range of RVI, the patient's enteral nutrition is evaluated at the first level, and the patients are classified based on this evaluation result.

[0082] This grading method can quickly identify high-risk patients and provide a timely reference for clinical decision-making.

[0083] Fourth, if the first-level assessment is first-level tolerance, a multivariate regression model analysis is used to predict the comprehensive tolerance assessment value RVI (t+4).

[0084] This prediction not only provides a dynamic perspective for assessment, but also effectively analyzes patient risk changes and intervention effects, enabling medical staff to adjust treatment plans in real time to improve the targetedness and effectiveness of intervention.

[0085] Finally, based on the secondary evaluation and analysis results, the system automatically triggers an early warning and sends feedback adjustment suggestions and early warning information to the medical staff's mobile terminal through the cloud server.

[0086] At the same time, intervention recommendations and historical trend graphs are generated.

[0087] This mechanism ensures the timeliness of medical intervention, reduces the risk of complications for patients, and provides data support for subsequent treatment.

[0088] In summary, this method has significant improvements over existing technical means in improving the efficiency of real-time monitoring, data processing and risk assessment.

[0089] Traditional methods mostly rely on static analysis and manual monitoring, while this method achieves dynamic monitoring and automatic feedback, allowing medical staff to respond more quickly to changes in patient conditions.

[0090] In addition, the generation of historical trend graphs provides a basis for long-term monitoring and helps to continuously optimize nutritional support programs.

[0091] This innovative assessment method not only improves the safety and comfort of patients, but also provides more scientific and effective support for clinical decision-making, ultimately improving patients' treatment outcomes and quality of life.

[0092] Example 2

[0093] This example is an explanation of Example 3, see Figure 1 , specifically: S1 includes S11 and S12;

[0094] S11. Collect the intestinal microbial environment parameters of patients in real time through the ICU monitoring system and non-invasive detection equipment. The intestinal microbial environment parameters include intestinal microbial diversity index MDI, blood microcirculation rate BMCR, electrogastrogram spectrum factor GEF, peritoneal pressure fluctuation index PPI, intestinal pH dynamic change rate GPC, heart rate variability variance HRV var , serum albumin concentration SA and C-reactive protein concentration CRP;

[0095] Non-invasive detection devices include non-invasive stool detectors, optical sensors, non-invasive electrodes, abdominal pressure sensors, and wearable biosensors;

[0096] The ICU monitoring system is connected to the cloud server through the API application program interface, and the serum albumin concentration SA and C-reactive protein concentration CRP collected by the ICU monitoring system are automatically sent to the cloud server. At the same time, the intestinal microbial diversity index MDI, blood microcirculation rate BMCR, electrogastrogram spectrum factor GEF, peritoneal pressure fluctuation index PPI, intestinal pH dynamic change rate GPC and heart rate variability variance HRV are manually var Input into the cloud server;

[0097] The intestinal microbial diversity index (MDI) is analyzed and tested by a non-invasive stool detector to assess the health status of the intestinal flora.

[0098] Blood microcirculation rate BMCR is measured by optical sensors to assess the blood flow rate in capillaries;

[0099] The electrogastrogram spectral factor (GEF) evaluates the functional status of the stomach by acquiring gastric electrical activity data using non-invasive electrodes.

[0100] The peritoneal pressure fluctuation index (PPI) detects the fluctuation of abdominal pressure through an abdominal pressure sensor, reflecting the intestinal peristalsis.

[0101] The dynamic change rate of intestinal pH GPC detects the changes in intestinal pH through wearable biosensors;

[0102] Heart rate variability HRV var Collect patient heart rate data through wearable biosensors;

[0103] The serum albumin concentration SA is obtained by extracting the test data of the blood test machine connected to the ICU monitoring system;

[0104] The C-reactive protein concentration CRP is obtained by extracting the test data of the blood test machine connected to the ICU monitoring system;

[0105] S12, integrating the input intestinal microbial environmental parameters in the cloud server for data preprocessing and data fusion;

[0106] Data preprocessing uses moving mean filtering and median filtering to remove short-term outliers in intestinal microbial environmental parameters, and then uses wavelet change denoising method to denoise and smooth the intestinal microbial environmental parameters. Finally, the intestinal microbial environmental parameters are standardized to eliminate the dimension effect.

[0107] Data fusion is used to fuse and summarize the intestinal microbial environmental parameters after data preprocessing to obtain a multidimensional indicator parameter set.

[0108] In this embodiment, the method first uses the ICU monitoring system and non-invasive detection equipment to collect a variety of key intestinal microbial environmental parameters in real time.

[0109] This comprehensive data collection method reduces the need for invasive examinations, reduces patient pain, and improves the convenience and safety of data collection compared to traditional methods.

[0110] Secondly, the collected data is preprocessed and fused, and outliers and noise are removed through advanced algorithms such as moving mean filtering, median filtering and wavelet transform, and then standardized.

[0111] This process not only improves data quality and reduces evaluation errors caused by inaccurate data, but also provides a more reliable basis for subsequent analysis.

[0112] This systematic data processing method is more efficient than traditional manual analysis methods and reduces the possibility of errors caused by human intervention.

[0113] Ultimately, the integrated multidimensional indicator parameter set provides accurate data support for the subsequent tolerance assessment model, making the results of nutritional tolerance assessment more scientific and accurate.

[0114] This integrated approach provides significant improvements over existing technical means. It not only improves the real-time and accuracy of monitoring, but also reduces the burden on medical staff through automated data processing, thereby improving the efficiency and scientificity of clinical decision-making.

[0115] Example 3

[0116] This example is explained in Example 2. Figure 1 , specifically: S2 includes S21 and S22;

[0117] S21. By constructing a tolerance assessment model, the intestinal microbial diversity index MDI, blood microcirculation rate BMCR, gastroesophageal electrogram spectrum factor GEF, peritoneal pressure fluctuation index PPI, intestinal pH dynamic change rate GPC and heart rate variability variance HRV manually input into the cloud server var , substituted into the tolerance assessment model, calculated and output the preliminary nutritional tolerance risk value R0, and conducted a preliminary analysis of the patient's enteral nutrition tolerance;

[0118] The preliminary nutritional tolerance risk value R0 is calculated and output by the following tolerance assessment model; ;

[0119] In the formula, log represents the logarithmic function, GEF mean represents the average value of the electrogastrogram spectrum factor, It represents the intestinal microbial diversity index multiplied by the inverse of the blood microcirculation rate, emphasizing the high risk when the microcirculation rate decreases. Indicates the use of logarithmic functions to smooth the impact of pressure fluctuations and avoid distortion of evaluation caused by extreme values. It indicates the balanced relationship between the dynamic rate of change of pH value and heart rate variability, and the risk increases when HRV is low.

[0120] S22. Based on the preliminary analysis of the patient's enteral nutrition tolerance, based on the output of the preliminary nutrition tolerance risk value R0, combined with the standard deviation of the intestinal microbial diversity index MDI and the heart rate variability variance HRV var and serum albumin concentration SA, are corrected and adjusted to generate a correction value Radj;

[0121] The correction value Radj is calculated and output by the following algorithm formula; ;

[0122] Where b1 represents the heart rate variability variance HRV var b2 represents the correction factor of serum albumin concentration SA, MDI std represents the standard deviation of the intestinal microbial diversity index.

[0123] In this embodiment, the method is to manually input the intestinal microbial diversity index MDI, blood microcirculation rate BMCR, gastroesophageal electrogram spectrum factor GEF, peritoneal pressure fluctuation index PPI, intestinal pH dynamic change rate GPC and heart rate variability variance HRV var Substitute it into the tolerance assessment model and calculate the preliminary nutritional tolerance risk value R0.

[0124] This process comprehensively considers multiple key indicators to preliminarily analyze the patient's enteral nutrition tolerance, effectively avoiding the misjudgment that may be caused by a single indicator and improving the comprehensiveness and scientificity of risk assessment. Based on the preliminary analysis, the preliminary nutrition tolerance risk value R0 was corrected and adjusted to generate the corrected value Radj.

[0125] By introducing the standard deviation of the intestinal microbial diversity index and the variance of heart rate variability HRV var and serum albumin concentration SA, realizing the dynamic correction of the initial nutritional tolerance risk value R0. This process not only enhances the adaptability of the model, but also further reduces the assessment error caused by individual differences, ensuring the reliability of the assessment results.

[0126] Example 4

[0127] This example is an explanation of Example 3, see Figure 1 , specifically: S3 includes S31 and S32;

[0128] S31. Based on the correction value Radj obtained after correction and adjustment, the C-reactive protein concentration CRP, the average value of the peritoneal pressure fluctuation index PPI and the peak value of the blood microcirculation rate BMCR in the multidimensional indicator parameter set are integrated to comprehensively output the comprehensive tolerance assessment value RVI to comprehensively analyze the patient's enteral nutrition tolerance;

[0129] The comprehensive tolerance assessment value RVI is calculated and output by the following algorithm formula; ;

[0130] Wherein, e represents the exponential function, PPImean represents the mean value of peritoneal pressure, BMCRpeak represents the peak value of blood microcirculation rate, c1 represents the adjustment coefficient of C-reactive protein concentration CRP, c2 represents the adjustment coefficient of the peak value of blood microcirculation rate, and c3 represents the adjustment coefficient of the mean value of peritoneal pressure.

[0131] S32. Use the statistical method ROC curve analysis to determine the cutoff point between the comprehensive tolerance assessment value RVI and clinical outcomes such as hospitalization time and complication rate, select 0.5 and 1.0 as the tolerance value range, and then based on the output result of the comprehensive tolerance assessment value RVI, conduct a primary assessment of the patient's enteral nutrition, and based on the primary assessment result, classify the patient's enteral nutrition. The specific assessment content is as follows;

[0132] When the comprehensive tolerance assessment value RVI is less than 0.5, it is classified as first-level tolerance, and the existing nutritional support plan is maintained and monitored regularly;

[0133] When 0.5≤RVI<1.0, it is classified as secondary tolerance. At this time, the monitoring frequency will be increased by 50%. The patient may be intolerant to enteral nutrition. The monitoring frequency will be increased, the changes in physiological indicators will be observed, and corresponding intervention measures will be prepared, such as adjusting the nutrition infusion rate or composition.

[0134] When the comprehensive tolerance assessment value RVI ≥ 1.0, it is divided into level 3 tolerance. At this time, the first warning information is generated and sent to the terminal device of the medical staff, prompting immediate enteral nutrition tolerance intervention, such as slowing down or suspending nutritional infusion, adjusting the composition of nutrient solution, and considering the use of gastrointestinal motility drugs, etc., to avoid serious complications.

[0135] In this embodiment, the method successfully calculates the comprehensive tolerance assessment value RVI by integrating the corrected value Radj with the C-reactive protein concentration CRP, the average value of the peritoneal pressure fluctuation index PPI and the peak value of the blood microcirculation rate BMCR in the multidimensional indicator parameter set.

[0136] This comprehensive assessment not only reflects the patient's physiological status, but also provides a quantitative basis for subsequent intervention decisions, ensuring that clinicians can adjust nutritional support plans in a dynamic environment.

[0137] ROC curve analysis was used to determine the cutoff point between RVI and clinical outcomes such as length of hospital stay and incidence of complications, thereby providing a refined primary assessment and grading of patients' enteral nutrition.

[0138] This assessment mechanism enables doctors to take appropriate monitoring and intervention measures based on the different numerical ranges of RVI, greatly improving the sensitivity to the patient's condition.

[0139] For example, for patients with RVI < 0.5, the existing plan is maintained, while for patients with RVI ≥ 1.0, early warning can be triggered in time, and more active intervention measures can be taken to effectively prevent the occurrence of complications.

[0140] For example, the following is an exemplary data list, which includes the correction value Radj, the C-reactive protein concentration CRP, the average value PPImean of the peritoneal pressure fluctuation index PPI, the peak value BMCRpeak of the blood microcirculation rate BMCR, and the comprehensive tolerance assessment value RVI calculated based on the test data information of the above-mentioned multiple patients (specifically, a total of 8 patients with patient ID number 1 to patient ID number 8).

[0141] See also Figure 2 , the following patient information is displayed in sequence, namely, patient ID number, correction value Radj, C-reactive protein concentration CRP, mean value PPImean of peritoneal pressure fluctuation index PPI, peak value BMCRpeak of blood microcirculation rate BMCR, comprehensive tolerance assessment value RVI and final grade classification result;

[0142] For an example, see Figure 2 :Among them, the ID of patient 1 is 001; the Radj value is 0.95; the CRP value is 6.2; the PPImean value is 8.7; the BMCRpeak value is 45; the RVI value is 0.4; the grade is level one;

[0143] Patient 2's ID is 002; Radj value is 0.88; CRP value is 12.5; PPImean value is 10.5; BMCRpeak value is 38; RVI value is 0.6; grade is level 2; for other patients 3-8, see Figure 2 The data content in .

[0144] Compared with traditional nutritional assessment methods, the technical solution adopted in this embodiment 4 not only improves the accuracy and timeliness of the assessment, but also promotes the scientific nature of clinical decision-making.

[0145] Traditional methods usually rely on a single indicator or subjective judgment, while this method forms a systematic decision support through data integration and dynamic analysis, which can provide personalized treatment suggestions based on the actual situation of different patients (especially emphasizing that it can successfully calculate the comprehensive tolerance assessment value RVI.

[0146] This comprehensive assessment not only reflects the patient's physiological state, but also provides a quantitative basis for subsequent intervention decisions).

[0147] Example 5

[0148] This example is explained in Example 2. Figure 1 , specifically: S4 includes S41 and S42;

[0149] S41. When the patient is assessed as having first-level tolerance based on the first-level assessment, a multivariate regression analysis algorithm is used to combine the comprehensive tolerance assessment value RVI, serum albumin concentration SA, and heart rate variability variance HRV. var , predict the comprehensive tolerance assessment value RVI within the next 4 hours, and obtain the predicted comprehensive tolerance assessment value RVI (t+4);

[0150] The predicted comprehensive tolerance assessment value RVI (t+4) was calculated and output by the following multivariate regression analysis algorithm; ;

[0151] In the formula, represents the intercept, which represents the predicted value when all independent variables are zero, reflecting the basic risk level of the multivariate regression analysis algorithm. It represents the regression coefficient of the current comprehensive tolerance assessment value RVI, and the influence of the comprehensive tolerance assessment value RVI on the predicted comprehensive tolerance assessment value RVI (t+4). The higher the comprehensive tolerance assessment value RVI, the worse the patient's comprehensive physiological state, and the predicted comprehensive tolerance assessment value RVI (t+4) will also increase. Indicates the change rate of serum albumin concentration SA, It represents the regression coefficient of serum albumin concentration SA difference, which represents the effect of the patient's serum albumin concentration SA on the patient's enteral nutrition tolerance. It represents the rate of change of the variance of heart rate variability, specifically the rate of change of the variance of heart rate variability in the most recent evaluation. The regression coefficient representing the change rate of heart rate variability variance indicates the impact of the change rate of heart rate variability on the prediction of comprehensive tolerance assessment value RVI (t+4). An increase in the change rate of heart rate variability may indicate a decrease in the patient's autonomic nervous system activity, thereby increasing the risk of intolerance.

[0152] S42. Perform secondary evaluation on the predicted comprehensive tolerance assessment value RVI (t+4) and the tolerance value range, analyze and predict the patient's enteral nutrition tolerance, and generate response levels and warnings based on the secondary evaluation results. The specific evaluation contents are as follows;

[0153] When the predicted comprehensive tolerance assessment value RVI (t+4) is less than 0.5, it means that after 4 hours, the patient's tolerance is normal. At this time, it is classified as level 1 tolerance, no warning information is generated, and continuous monitoring is required;

[0154] When 0.5≤predicted comprehensive tolerance assessment value RVI(t+4)<1.0, it means that the patient may be intolerant to enteral nutrition, which is classified as secondary tolerance. At this time, the second warning information is generated and sent to the terminal device of the medical staff to prompt the patient's enteral nutrition tolerance and make planned interventions;

[0155] When the predicted comprehensive tolerance assessment value RVI (t+4) ≥ 1.0, it indicates that the patient's enteral nutrition tolerance is abnormal, which is classified as level three tolerance. At this time, the first warning information is generated and sent to the terminal device of the medical staff, prompting them to immediately change the current enteral nutrition tolerance treatment.

[0156] In this embodiment, the method uses a multivariate regression analysis algorithm to integrate the comprehensive tolerance assessment value RVI, serum albumin concentration SA and heart rate variability variance HRV var Combined, the comprehensive tolerance assessment value RVI (t+4) within the next 4 hours was successfully predicted.

[0157] This dynamic predictive capability enables healthcare professionals to take appropriate action before the condition changes and identify risk factors that may lead to intolerance in advance.

[0158] In particular, changes in heart rate variability can directly reflect the state of the patient's autonomic nervous system, thus providing an important basis for judging the patient's physiological adaptability.

[0159] Secondary assessments based on the predicted results further refine the patient's management.

[0160] When the predicted comprehensive tolerance assessment value RVI (t+4) shows different risk levels, the system can automatically generate corresponding warning information and transmit it to the terminal device of medical staff.

[0161] This automated feedback mechanism ensures that the medical team can understand the patient's condition in a timely manner and quickly adjust the nutrition plan based on the assessment results, thereby effectively preventing the occurrence of complications.

[0162] For example, the early warning of level 2 tolerance can adjust the rate or composition of nutritional infusion appropriately, while the timely alarm of level 3 tolerance prompts medical staff to take more active intervention measures immediately to avoid serious consequences.

[0163] Example 6

[0164] This example is explained in Example 2. Figure 1 , specifically: S5 includes S51 and S52;

[0165] S51. Based on the secondary evaluation and analysis results, if the predicted comprehensive tolerance assessment value RVI (t+4) ≥ 1.0 exceeds 8 hours, an early warning is automatically triggered and sent to the terminal device of the medical staff, and segmented feedback adjustment suggestions are generated, including:

[0166] Phase 1: Reduce the standard nutrient concentration by 20%;

[0167] Phase II: Change to an intermittent infusion method, divided into multiple doses for batch injection;

[0168] Phase 3: Add supplemental fluid and electrolyte support to balance water and electrolytes;

[0169] S52. After the segmented feedback adjustment suggestion is generated, the predicted comprehensive tolerance assessment value RVI (t+4) and the comprehensive tolerance assessment value RVI obtained in the current and predicted stages are used to generate a historical trend chart.

[0170] In this embodiment, when the predicted comprehensive tolerance assessment value RVI (t+4) exceeds 1.0 and lasts for more than 8 hours, the method automatically triggers an early warning, sends a real-time alarm to medical staff, and generates segmented feedback adjustment suggestions.

[0171] This automated response mechanism ensures that the patient's nutrition support plan can be adjusted in a timely manner to reduce the risk of intolerance.

[0172] For example, gradually reducing nutrient intake, changing the infusion method, and adding fluid and electrolyte support form a systematic intervention strategy that can effectively manage enteral nutrition intolerance without increasing the burden on patients.

[0173] At the same time, by generating historical trend charts, medical staff can intuitively track the patient's assessment results and early warning information. This visual data presentation method enables the medical team to better analyze the patient's nutritional tolerance changes, evaluate the intervention effect, and thus optimize the subsequent treatment plan.

[0174] This not only enhances the accuracy of clinical decision-making, but also improves patient safety and comfort.

[0175] Example 7

[0176] See also Figure 2 , an early enteral nutrition intolerance assessment system for severe pancreatitis based on multi-dimensional indicators, including a data input module, a multi-dimensional indicator analysis module, a grade classification module, a tolerance prediction module and an alarm feedback module;

[0177] The data input module detects the intestinal microbial environmental parameters of the patient in real time through non-invasive detection equipment, and transmits the intestinal microbial environmental parameters to the cloud server by building a cloud server, and performs data preprocessing and data fusion on the intestinal microbial environmental parameters in the cloud server to obtain a multi-dimensional indicator parameter set;

[0178] The multidimensional index analysis module obtains a preliminary nutritional tolerance risk value R0 by inputting a multidimensional index parameter set into a tolerance assessment model, and performs correction and adjustment on the basis of the preliminary nutritional tolerance risk value R0 to obtain a correction value Radj;

[0179] The grading module performs comprehensive calculation by integrating the multidimensional indicator parameter set and the correction value Radj to obtain the comprehensive tolerance assessment value RVI, and performs a primary assessment of the patient's enteral nutrition according to the numerical range of the comprehensive tolerance assessment value RVI, and grades the patient based on the assessment results;

[0180] When the tolerance prediction module determines that the patient has first-level tolerance through the first-level assessment, a multivariate regression model is used for analysis to output a predicted comprehensive tolerance assessment value RVI (t+4), and a second-level assessment is performed to analyze the patient's risk changes and intervention effects based on the output result of the predicted comprehensive tolerance assessment value RVI (t+4);

[0181] The alarm feedback module automatically triggers an early warning based on the secondary evaluation and analysis results, sends feedback adjustment suggestions and early warning information to the medical staff's mobile terminal through the cloud server, and generates intervention suggestions and historical trend charts.

[0182] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. A method for evaluating early enteral nutrition intolerance in severe pancreatitis based on multi-dimensional indicators, characterized by: The following steps are involved: S1. Real-time detection of the intestinal microbial environmental parameters of the patient by non-invasive detection equipment, and transmission of the intestinal microbial environmental parameters to the cloud server by building a cloud server, and data preprocessing and data fusion of the intestinal microbial environmental parameters in the cloud server to obtain a multi-dimensional indicator parameter set; Said S1 includes S11 and S12; S11. Real-time collection of intestinal microbial environmental parameters of patients through ICU monitoring system and non-invasive detection equipment, wherein the intestinal microbial environmental parameters include intestinal microbial diversity index, blood microcirculation rate, electrogastrogram spectrum factor, peritoneal pressure fluctuation index, intestinal pH dynamic change rate, heart rate variability variance, serum albumin concentration and C-reactive protein concentration; Non-invasive detection devices include non-invasive stool detectors, optical sensors, non-invasive electrodes, abdominal pressure sensors, and wearable biosensors; The ICU monitoring system is connected to the cloud server through the API application program interface, and the serum albumin concentration and C-reactive protein concentration collected by the ICU monitoring system are automatically sent to the cloud server. At the same time, the intestinal microbial diversity index, blood microcirculation rate, electrogastrogram spectrum factor, peritoneal pressure fluctuation index, intestinal pH dynamic change rate and heart rate variability variance are manually input into the cloud server; S12, integrating the input intestinal microbial environmental parameters in the cloud server to perform data preprocessing and data fusion; The data preprocessing removes short-term outliers in the intestinal microbial environmental parameters by applying moving mean filtering and median filtering, and then uses wavelet change denoising method to denoise and smooth the intestinal microbial environmental parameters. Finally, the intestinal microbial environmental parameters are standardized to eliminate the dimension effect; The data fusion is used to fuse and summarize the intestinal microbial environmental parameters after data preprocessing to obtain a multidimensional indicator parameter set; S2. constructing a tolerance assessment model, and inputting a multidimensional indicator parameter set into the tolerance assessment model to obtain a preliminary nutritional tolerance risk value, and based on the preliminary nutritional tolerance risk value, performing correction and adjustment to obtain a correction value; The S2 includes S21 and S22; S21, by constructing a tolerance assessment model, the intestinal microbial diversity index, the blood microcirculation rate, the electrogastrogram spectrum factor, the peritoneal pressure fluctuation index, the intestinal pH dynamic change rate and the heart rate variability variance manually input into the cloud server are substituted into the tolerance assessment model, and the preliminary nutrition tolerance risk value is calculated and output, so as to conduct a preliminary analysis of the patient's enteral nutrition tolerance; S22. Based on the preliminary analysis of the patient's enteral nutrition tolerance, the output preliminary nutrition tolerance risk value is corrected and adjusted in combination with the standard deviation of the intestinal microbial diversity index, the heart rate variability variance and the serum albumin concentration to generate a correction value; S3, integrating the multidimensional indicator parameter set and the correction value for comprehensive calculation to obtain a comprehensive tolerance assessment value, and performing a primary assessment of the patient's enteral nutrition according to the numerical range of the comprehensive tolerance assessment value, and grading the patient based on the assessment result; The S3 includes S31 and S32; S31. Based on the correction value obtained after correction and adjustment, the C-reactive protein concentration, the average value of the peritoneal pressure fluctuation index and the peak value of the blood microcirculation rate in the multidimensional indicator parameter set are integrated to comprehensively output a comprehensive tolerance assessment value to comprehensively analyze the patient's enteral nutrition tolerance; S32. Use the statistical method ROC curve analysis to determine the demarcation point between the comprehensive tolerance assessment value and the clinical result, select 0.5 and 1.0 as the tolerance value range, and then based on the output result of the comprehensive tolerance assessment value, perform a primary assessment on the patient's enteral nutrition, and based on the primary assessment result, grade the patient's enteral nutrition. The specific assessment content is as follows; When the comprehensive tolerance assessment value is less than 0.5, it is classified as first-level tolerance, and the existing nutritional support plan is maintained and monitored regularly; When 0.5≤comprehensive tolerance assessment value<1.0, it is classified as secondary tolerance, and the monitoring frequency is increased by 50%; When the comprehensive tolerance assessment value is ≥1.0, it is classified as level 3 tolerance, and the first warning information is generated and sent to the terminal device of the medical staff, prompting immediate enteral nutrition tolerance intervention; S4. If the primary assessment shows primary tolerance, a multivariate regression model is used to output a predicted comprehensive tolerance assessment value, and a secondary assessment is performed based on the output of the predicted comprehensive tolerance assessment value to analyze the patient's risk changes and intervention effects; S5. Based on the secondary evaluation and analysis results, an early warning is automatically triggered, and feedback adjustment suggestions and early warning information are sent to the mobile terminal of medical staff through the cloud server, and intervention suggestions and historical trend charts are generated.

2. The method for evaluating early enteral nutrition intolerance in severe pancreatitis based on multi-dimensional indicators according to claim 1, characterized in that: The S4 includes S41 and S42; S41. When the patient is assessed as having first-level tolerance based on the first-level assessment, a multivariate regression analysis algorithm is used to predict the comprehensive tolerance assessment value within the next 4 hours by combining the comprehensive tolerance assessment value, serum albumin concentration, and heart rate variability variance to obtain a predicted comprehensive tolerance assessment value; S42, conduct a secondary evaluation on the predicted comprehensive tolerance assessment value and the tolerance value range, analyze the patient's enteral nutrition tolerance in the prediction stage, and generate a response level and warning based on the secondary evaluation results. The specific evaluation contents are as follows; When the predicted comprehensive tolerance assessment value is less than 0.5, it is classified as level 1 tolerance, no warning information is generated, and continuous monitoring is performed; When 0.5≤predicted comprehensive tolerance assessment value<1.0, it is classified as secondary tolerance, and a second warning message is generated and sent to the terminal device of the medical staff to prompt the patient's enteral nutrition tolerance and make planned interventions; When the predicted comprehensive tolerance assessment value is ≥1.0, it is divided into level three tolerance. At this time, the first warning information is generated and sent to the terminal device of the medical staff, prompting them to immediately change the current enteral nutrition tolerance treatment.

3. The method for evaluating early enteral nutrition intolerance in severe pancreatitis based on multi-dimensional indicators according to claim 1, characterized in that: The S5 includes S51 and S52; S51. Based on the secondary evaluation and analysis results, if the predicted comprehensive tolerance assessment value is ≥1.0 for more than 8 hours, an early warning is automatically triggered and sent to the terminal device of the medical staff, and a segmented feedback adjustment suggestion is generated, including: Phase 1: Reduce the standard nutrient concentration by 20%; Phase II: Change to an intermittent infusion method, divided into multiple doses for batch injection; Phase 3: Add supplemental fluid and electrolyte support to balance water and electrolytes; S52: After the segmented feedback adjustment suggestion is generated, the predicted comprehensive tolerance assessment value and the historical trend graph of the comprehensive tolerance assessment value obtained in the current and predicted stages are displayed.

4. A system for evaluating enteral nutrition intolerance in early stage of severe pancreatitis based on multi-dimensional indicators, applied to a method for evaluating enteral nutrition intolerance in early stage of severe pancreatitis based on multi-dimensional indicators according to any one of claims 1 to 3, characterized in that: It includes data input module, multi-dimensional index analysis module, grade classification module, tolerance prediction module and alarm feedback module; The data input module detects the intestinal microbial environmental parameters of the patient in real time through non-invasive detection equipment, and transmits the intestinal microbial environmental parameters to the cloud server by building a cloud server, and performs data preprocessing and data fusion on the intestinal microbial environmental parameters in the cloud server to obtain a multi-dimensional indicator parameter set; The multidimensional index analysis module constructs a tolerance assessment model and inputs the multidimensional index parameter set into the tolerance assessment model to obtain a preliminary nutritional tolerance risk value, and performs correction and adjustment on the basis of the preliminary nutritional tolerance risk value to obtain a correction value; The grading module performs comprehensive calculation by integrating the multidimensional indicator parameter set and the correction value to obtain a comprehensive tolerance assessment value, and performs a primary assessment of the patient's enteral nutrition according to the numerical range of the comprehensive tolerance assessment value, and grades the patient based on the assessment result; When the tolerance prediction module determines that the patient has first-level tolerance through a second-level assessment, a multivariate regression model is used for analysis to output a predicted comprehensive tolerance assessment value, and a second-level assessment is performed based on the output result of the predicted comprehensive tolerance assessment value to analyze the patient's risk changes and intervention effects; The alarm feedback module automatically triggers an early warning based on the secondary evaluation and analysis results, sends feedback adjustment suggestions and early warning information to the medical staff's mobile terminal through the cloud server, and generates intervention suggestions and historical trend charts.

Images