A real-time monitoring system for blood sedimentation rate in inflammatory conditions

Abstract

The present application provides a real-time monitoring system for blood sedimentation rate in an inflammatory state, including: a patient physiological behavior analysis module for obtaining the patient's physiological parameters, living habits data and ESR data; an ESR value detection timing judgment module for determining the best blood sampling time; an ESR value analysis module for identifying factors affecting ESR value changes; an ESR prediction value deviation judgment module for combining the correlation between external factors and ESR values ​​to determine whether it is caused by changes in the inflammatory state; an ESR value adjustment amount prediction module for predicting an individual's ESR value adjustment amount; an inflammatory state assessment module for adjusting the ESR test results and re-evaluating the patient's inflammatory state; an inflammatory state monitoring module for predicting the probability of inflammatory state changes and adjusting the monitoring timing of ESR and acute phase reaction protein concentrations. The present invention not only improves the accuracy of blood sedimentation rate detection, but also enhances the scientificity and effectiveness of clinical decision support and patient management.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a real-time monitoring system for blood sedimentation rate in an inflammatory state. Background Art

[0002] In clinical medicine, the diagnosis and monitoring of inflammatory status are crucial for the management and treatment of many diseases. As a commonly used inflammatory indicator, the blood sedimentation rate (ESR) is widely used to evaluate and monitor the activity of inflammatory responses. However, existing blood testing methods have many limitations in sensitivity and specificity, and cannot fully and accurately reflect the patient's true inflammatory status. Traditional blood tests are usually performed under static conditions, ignoring the impact of patients' daily living habits and physiological activities on the test results. Factors such as patients' physiological activities, medication use, and dietary habits may have a significant impact on ESR values, resulting in challenges to the accuracy of test results. For example, exercise may temporarily change the fluidity and composition of blood, thereby affecting the measurement of ESR. In addition, certain drugs and dietary ingredients can also interfere with the measurement of ESR, resulting in false negative or false positive results. The complexity of these factors makes it difficult for a single blood test to provide a comprehensive and reliable assessment of inflammatory status. Most existing blood testing methods are performed in a laboratory setting and are usually analyzed over a long period of time after the blood sample is collected. Since blood samples may change during collection and transportation, this delay may result in test results that do not accurately reflect the patient's current inflammatory status. In addition, static blood testing methods lack continuous monitoring capabilities and cannot provide real-time data to dynamically track changes in inflammation. In the management of chronic inflammatory diseases, timely and accurate understanding of changes in inflammatory status is crucial to developing effective treatment plans. In current medical practice, although the impact of physiological activities and other factors on ESR testing is recognized, there is a lack of a method to comprehensively consider these variables and integrate them in ESR testing. For example, although factors such as exercise, medication, and diet are known to have an impact on ESR, in the actual testing process, how to quantify these effects and choose the best time to collect blood remains a difficult problem. The limitations of traditional methods not only affect the accuracy of test results, but also limit their application in long-term inflammation management and dynamic disease monitoring. In addition, changes in the concentration of acute phase reactants, such as C-reactive protein CRP and fibrinogen, are also important indicators for the assessment of inflammatory status. However, existing ESR detection methods usually fail to fully combine the data of these acute phase reactants, thereby affecting the comprehensive assessment of the inflammatory status. Changes in acute phase reactants can reflect the dynamic process of inflammatory response, and combining them with ESR test results can provide more comprehensive and accurate monitoring of inflammatory status. Based on the above problems, traditional blood testing methods have limitations in sensitivity and specificity, especially in the monitoring and management of chronic inflammatory conditions. Traditional methods are mostly performed under static conditions, lacking comprehensive consideration of the patient's daily living habits, which affects the accuracy of the test results. In addition, existing blood testing methods are usually performed within a period of time after the blood sample is collected, and cannot provide real-time and continuous monitoring of the inflammatory state, limiting its application in long-term inflammation management and dynamic disease monitoring.Although changes in acute phase response proteins are important, existing methods fail to fully integrate these data, affecting the comprehensive assessment of the inflammatory state. Therefore, there is an urgent need for a more advanced monitoring method that can comprehensively consider various influencing factors and provide real-time, continuous inflammatory state monitoring to better support clinical decision-making and patient management. Summary of the invention

[0003] The present invention provides a real-time monitoring system for blood sedimentation rate in an inflammatory state, which mainly includes:

[0004] Patient physiological behavior analysis module, used to obtain the patient's physiological parameters, living habits data and ESR data, predict the changes in ESR values ​​under different physiological parameters and living habits, and identify the physiological parameters and living habits that affect the accuracy of ESR value measurement;

[0005] The ESR value detection timing judgment module is used to establish correlation analysis through physiological behavior patterns and historical data of ESR value measurements, quantitatively score the degree to which different living habits affect the ESR value, and select the time period with the least impact on the test results as the best blood collection time for ESR value detection;

[0006] The ESR value analysis module is used to build an ESR value prediction model based on the patient's physiological parameters, living habits data, and blood test results, predict the change trend of the ESR value, identify the factors that affect the change of the ESR value, and evaluate the inflammatory status;

[0007] The ESR prediction value deviation judgment module is used to compare the predicted ESR value with the patient's actual measured ESR value. If the actual measured ESR value deviates from the predicted ESR value, it is determined whether it is caused by changes in the inflammatory state in combination with the correlation between external factors and the ESR value.

[0008] The ESR value adjustment amount prediction module is used to continuously monitor the concentration changes of acute phase reaction proteins, including CRP and fibrinogen, to evaluate the dynamic changes of inflammation, and to construct an ESR value adjustment amount prediction model based on individual patient information and acute phase reaction protein concentrations to predict the individual ESR value adjustment amount;

[0009] The inflammatory status assessment module is used to obtain the baseline acute phase reactant protein concentration according to the information system. If the change value of the acute phase reactant protein concentration exceeds the baseline, the ESR test result is adjusted according to the acute phase reactant protein monitoring result, and the patient's inflammatory status is reassessed;

[0010] The inflammatory status monitoring module is used to obtain clinical parameters, predict the probability of changes in the inflammatory status, monitor the development trend of the inflammatory status in real time, and adjust the monitoring timing of ESR and acute phase reaction protein concentrations.

[0011] Furthermore, the patient physiological behavior analysis module is used to obtain the patient's physiological parameters, living habits data and ESR data, predict the change of ESR value under different physiological parameters and living habits, and identify the physiological parameters and living habits that affect the accuracy of ESR value measurement, including:

[0012] The patient data acquisition unit is used to obtain the patient's physiological parameters and living habits data. The physiological parameter data include age, gender, weight and height, and pregnancy status. The living habits data include smoking and drinking habits, exercise frequency, eating habits, and sleep patterns. According to the preset data quality standards, the data is screened and cleaned to remove incomplete, erroneous or abnormal data records. The key feature screening unit is used to convert the original data into a standard format, generate new feature variables based on the original data, and use principal component analysis PCA to select the key features most useful for ESR prediction. The ESR value influencing factor identification unit is used to obtain ESR data under different physiological parameters and living habits data through a ROLLER20 erythrocyte sedimentation rate meter, use the ARIMA algorithm for model training, and predict the changes in ESR values ​​under different physiological parameters and living habits; through the prediction results of the ARIMA model, the physiological parameters and living habits factors that have the greatest impact on the ESR value are identified, and their impact degree and confidence interval are calculated.

[0013] Furthermore, the ESR value detection timing judgment module is used to establish a correlation analysis through the physiological behavior pattern and the historical data of the ESR value measurement, quantitatively score the degree to which different living habits affect the ESR value, select the time period with the least impact on the test result as the best blood sampling time, and perform ESR value detection, including:

[0014] The comprehensive data acquisition unit is used to obtain physiological parameters, living habits, medical history, ESR measurement history data, and create a comprehensive database; pre-process the data in the comprehensive database, including removing duplicate records, filling missing values, identifying and processing outliers, and standardizing the data; the correlation factor identification unit is used to identify the correlation between physiological parameters, living habits, medical history and ESR values ​​by calculating the Pearson correlation coefficient between them, and screen out strong correlation factors with correlations higher than a preset correlation threshold; the impact assessment unit is used to use a multivariate linear regression algorithm to train the model based on the screened strong correlation factors, construct a lifestyle impact assessment model, and assign quantitative scores to the impact of different lifestyles on ESR; the blood sampling timing judgment unit is used to determine the impact of different lifestyles on ESR values ​​based on the lifestyle impact assessment model, and based on the lifestyle habits in different time periods, use the weighted average method to obtain the impact of different time periods on ESR values, determine the time period with the least impact on ESR values, and use it as the best blood sampling time to detect ESR values.

[0015] Furthermore, the ESR value analysis module is used to build an ESR value prediction model based on the patient's physiological parameters, living habits data and blood test results, predict the change trend of the ESR value, identify factors affecting the change of the ESR value, and evaluate the inflammatory state, including:

[0016] The patient data acquisition unit is used to acquire the patient's physiological parameter data and living habit data; the body mass index calculation unit is used to calculate the body mass index BMI according to the patient's physiological parameters; the medical information acquisition unit is used to acquire the patient's blood test data, including red blood cell count, hemoglobin, white blood cell count, platelet count and CPR concentration, through the patient medical information management system; the patient's health and medical information is acquired, including chronic disease conditions, medication use, disease history, and recent infection or surgery information; the ESR value prediction unit is used to train the model through the long short-term memory network algorithm based on the patient's physiological parameter data, body mass index BMI, living habit data, blood test data, and health and medical information, to construct an ESR value prediction model, predict the change trend of the patient's ESR value, and assign weights to the influence of each factor; the inflammation state judgment unit is used to obtain the inflammation state evaluation result based on the prediction result of the patient's ESR value, and the inflammation state includes chronic state, acute state, and no inflammation.

[0017] Furthermore, the ESR prediction value deviation judgment module is used to compare the predicted ESR prediction value with the patient's measured ESR value. If the actually measured ESR value deviates from the predicted ESR prediction value, it is judged whether it is caused by the change of the inflammatory state in combination with the correlation between the external factors and the ESR value, including:

[0018] The ESR value prediction execution unit is used to record and store the patient's measured ESR value in the electronic health record using the patient information management system; the ESR value prediction model is used to combine the patient's physiological parameter data, body mass index BMI, living habit data, blood test data, and health and medical information to predict the ESR value; the ESR value deviation analysis unit is used to calculate the deviation value between the measured ESR value and the predicted ESR value, and determine whether the deviation value exceeds the preset deviation value range; if the deviation value exceeds the preset deviation value range, the patient's stress assessment data and drug use records are queried and extracted through the medical health database to determine the external factors that cause abnormal changes in the ESR; the Pearson correlation coefficient calculation method is used to determine the correlation coefficient between the external factors and the ESR deviation value, and determine the external factors whose correlation coefficient with the ESR deviation value is higher than the preset correlation coefficient threshold; if there is no external factor whose correlation coefficient with the ESR deviation value is higher than the preset correlation coefficient threshold, it is determined that the ESR deviation value is caused by changes in the inflammatory state, and the inflammatory state includes a chronic state and an acute state.

[0019] Furthermore, the ESR value adjustment amount prediction module is used to continuously monitor the concentration changes of acute phase reaction proteins, including CRP and fibrinogen, to evaluate the dynamic changes of inflammation, and to construct an ESR value adjustment amount prediction model based on the individual patient information and the acute phase reaction protein concentration to predict the individual ESR value adjustment amount, including:

[0020] The acute phase reactant protein monitoring unit is used to monitor the CRP concentration in the patient's blood sample using a sensor and record the real-time changes in the CRP concentration; use an optical sensor to detect the change in the optical density of fibrinogen in the blood sample to determine the concentration of fibrinogen in the blood sample; use Python's Pandas and Matplotlib libraries to draw a time series graph of the changes in CRP and fibrinogen concentrations over time based on the time series data of CRP concentration and fibrinogen concentration; use the Pandas library to calculate the moving average of CRP and fibrinogen concentrations, draw a moving average trend line on the time series graph, and show the changing trend of acute phase reactant proteins, which include CRP and fibrinogen; the ESR value adjustment amount prediction unit is used to use the long short-term memory network algorithm to train the model based on the patient's individual information and acute phase reactant protein concentration, build an ESR value adjustment amount prediction model, and predict the patient's ESR value adjustment amount. The patient's individual information includes age and gender.

[0021] Furthermore, the inflammatory status assessment module is used to obtain a baseline acute phase reactant protein concentration according to the information system. If the change value of the acute phase reactant protein concentration exceeds the baseline, the ESR test result is adjusted according to the acute phase reactant protein monitoring result, and the inflammatory status of the patient is reassessed, including:

[0022] An acute phase reactant protein concentration comparison unit is used to obtain the patient's baseline acute phase reactant protein concentration through the information system and compare it with the current acute phase reactant protein concentration test result; an ESR value adjustment unit is used to adjust the current ESR value by predicting the ESR value adjustment amount based on the ESR value adjustment amount prediction model if a certain acute phase reactant protein concentration is detected to exceed a preset acute protein concentration threshold; an inflammatory status reassessment unit is used to reassess the patient's inflammatory status based on the adjusted ESR value.

[0023] Furthermore, the inflammatory status monitoring module is used to obtain clinical parameters, predict the probability of inflammatory status change, monitor the development trend of inflammatory status in real time, and adjust the monitoring timing of ESR and acute phase reaction protein concentration, including:

[0024] The clinical parameter acquisition unit is used to obtain clinical parameters from the electronic health record system, including historical ESR values, acute phase reaction protein concentrations, basic patient information, lifestyle, and chronic disease history. Basic patient information includes age and gender, and lifestyle includes smoking history and dietary habits; clean the data, process missing values ​​and outliers, and perform data conversion and standardization on different data types, including continuous variables and categorical variables; the inflammatory state change probability detection unit is used to define the inflammatory state change probability formula based on clinical parameters

[0025] P(t)=α×log(E avg )+β×exp(CRP norm )+δ×exp(FIB norm )+∑(γ i ×X i ), where E avg is the average ESR over the past period of time, CRP norm is the normalized CRP concentration, FIB norm is the normalized fibrinogen concentration, X i Represents other clinical parameters, α, β, δ, γ iIt is a formula parameter determined by historical clinical data; the inflammation intervention unit is used to monitor the development trend of the inflammatory state in real time through the inflammatory state change probability formula, and dynamically adjust the monitoring frequency, monitoring timing and intervention timing of the ESR value and acute phase reactant protein concentration according to the changing trend of the inflammatory state change probability.

[0026] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0027] The present invention discloses a real-time monitoring system for blood sedimentation rate in an inflammatory state. The present invention can obtain and analyze a variety of physiological data of patients, predict changes in ESR values ​​under different conditions, and thus identify factors affecting the accuracy of ESR measurement. By quantifying the effects of different living habits on ESR values, the best blood sampling time is selected to improve the accuracy of the test results. Combining a variety of data, predicting the change trend of ESR values, and identifying influencing factors, it is helpful to more comprehensively evaluate the inflammatory state. The present invention identifies abnormalities caused by changes in the inflammatory state by comparing the deviation between the measured ESR value and the predicted value, and provides a more reliable diagnosis of the inflammatory state. Continuously monitoring the concentration changes of acute phase reaction proteins and predicting the adjustment amount of ESR values ​​in combination with individual information help to more accurately reflect the dynamic changes of inflammation. Based on the baseline acute phase reaction protein concentration, the ESR test result is adjusted according to the change value to ensure the accuracy of the evaluation. By real-time monitoring and predicting the change trend of the inflammatory state, the monitoring time is optimized, and the early warning ability of the development of inflammation is improved. The present invention is superior to traditional blood detection methods in sensitivity and specificity, can provide real-time and continuous monitoring of the inflammatory state, and significantly improves the management of chronic inflammatory symptoms and the ability of dynamic disease monitoring. This advanced monitoring method not only improves the accuracy of blood sedimentation rate testing, but also enhances the scientificity and effectiveness of clinical decision support and patient management. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 The present invention is a flow chart of a real-time monitoring system for blood sedimentation rate in an inflammatory state.

[0029] Figure 2 It is a schematic diagram of a real-time monitoring system of blood sedimentation rate in an inflammatory state of the present invention.

[0030] Figure 3 It is another schematic diagram of a real-time monitoring system of blood sedimentation rate in an inflammatory state of the present invention. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] like Figure 1-3In this embodiment, a real-time monitoring system for blood sedimentation rate in an inflammatory state may specifically include:

[0033] Step S101, the patient physiological behavior analysis module is used to obtain the patient's physiological parameters, living habits data and ESR data, predict the change of ESR value under different physiological parameters and living habits, and identify the physiological parameters and living habits that affect the accuracy of ESR value measurement.

[0034] The patient data acquisition unit is used to obtain the patient's physiological parameters and living habits data. The physiological parameter data includes age, gender, weight and height, and pregnancy status. The living habits data includes smoking and drinking habits, exercise frequency, eating habits, and sleep patterns. According to the preset data quality standards, the data is screened and cleaned to remove incomplete, erroneous or abnormal data records. The key feature screening unit is used to convert the original data into a standard format, generate new feature variables based on the original data, and use principal component analysis PCA to select the most useful key features for ESR prediction. The ESR value influencing factor identification unit is used to obtain ESR data under different physiological parameters and living habits data through the ROLLER20 erythrocyte sedimentation meter, use the ARIMA algorithm to train the model, and predict the changes in ESR values ​​under different physiological parameters and living habits. Through the prediction results of the ARIMA model, the physiological parameters and living habits factors that have the greatest impact on the ESR value are identified, and their impact degree and confidence interval are calculated.

[0035] For example, through the patient health monitoring device, the physiological parameters and living habits data of a patient were obtained. The physiological parameter data included age 35 years old, male gender, weight 75kg, height 180cm and non-pregnant status. The living habits data included smoking 5 cigarettes a day, drinking twice a week, exercising 3 times a week for 1 hour each time, eating a high-protein and low-carbohydrate diet, and sleeping an average of 7 hours a night. According to the preset data quality standards, the data was screened and cleaned, and after removing incomplete, erroneous or abnormal data records, the original data was converted to a standard format, and new feature variables were generated based on the original data, such as body mass index BMI of 24.7. Principal component analysis PCA was used to select the most useful key features for ESR prediction, and red blood cell count, CRP concentration and body mass index BMI were obtained as key features. The ESR data under these physiological parameters and living habits were obtained through the ROLLER20 erythrocyte sedimentation meter, and the ARI MA algorithm was used for model training to predict the changes in ESR values ​​under different physiological parameters and living habits. Through the prediction results of the ARI MA model, it was identified that the physiological parameters with the greatest impact on the ESR value were CRP concentration and body mass index (BMI), and the lifestyle factor was smoking status. The influence degrees of CRP concentration on the ESR value were calculated to be 0.65, with a confidence interval of 0.50 to 0.80, the influence degree of body mass index on the ESR value was 0.45, with a confidence interval of 0.30 to 0.60, and the influence degree of smoking status on the ESR value was 0.35, with a confidence interval of 0.20 to 0.50.

[0036] Step S102, the ESR value detection timing judgment module is used to establish a correlation analysis through the historical data of physiological behavior patterns and ESR value measurements, quantitatively score the degree to which different living habits affect the ESR value, and select the time period with the least impact on the test results as the best blood sampling time for ESR value detection.

[0037] The comprehensive data acquisition unit is used to obtain physiological parameters, living habits, medical history, ESR measurement history data, and create a comprehensive database. The data in the comprehensive database is preprocessed, including removing duplicate records, filling missing values, identifying and processing outliers, and standardizing the data. The correlation factor identification unit is used to identify the correlation between physiological parameters, living habits, medical history and ESR values ​​by calculating the Pearson correlation coefficient between physiological parameters, living habits, medical history and ESR values, and screen out strong correlation factors with correlations higher than the preset correlation threshold. The impact degree evaluation unit is used to train the model based on the screened strong correlation factors using a multivariate linear regression algorithm, build a living habits impact degree evaluation model, and assign quantitative scores to the impact of different living habits on ESR. The blood sampling time judgment unit is used to determine the impact of different living habits on ESR values ​​based on the living habits impact degree evaluation model, and based on the living habits in different time periods, use the weighted average method to obtain the impact of different time periods on ESR values, determine the time period with the least impact on ESR values, and use it as the best blood sampling time to perform ESR value detection.

[0038] For example, the physiological parameters of the patient were obtained as age 45 years old, gender female, weight 70 kg, height 165 cm, lifestyle habits as smoking 5 cigarettes a day, drinking once a day, exercising twice a day for 1 hour each time, medical history as a 2-year history of diabetes, and ESR measurement history data, and a comprehensive database was created. The data in the comprehensive database were preprocessed, including removing duplicate records, filling missing values, identifying and processing outliers, and standardizing the data. By calculating the Pearson correlation coefficient between these physiological parameters, lifestyle habits, medical history and ESR values, it was identified that age, number of cigarettes smoked per day and history of diabetes had a high correlation with ESR values, with correlation coefficients of 0.62, 0.55 and 0.50, respectively, all higher than the preset correlation threshold of 0.5. Based on the screened strong correlation factors, the multivariate linear regression algorithm was used for model training, and a lifestyle influence assessment model was constructed to assign quantitative scores to the degree of influence of different lifestyle habits on ESR, such as the impact score of smoking 5 cigarettes a day was 0.6, and the impact score of drinking once a day was 0.4. According to the lifestyle impact assessment model, the impact of different lifestyle habits on the ESR value was determined. Based on the patient's lifestyle data at different time periods, the weighted average method was used to obtain the impact of different time periods on the ESR value. The time period with the least impact on the ESR value was determined to be 10 to 11 a.m., which was the best time for blood collection.

[0039] Step S103, the ESR value analysis module is used to build an ESR value prediction model based on the patient's physiological parameters, living habits data and blood test results, predict the change trend of the ESR value, identify the factors affecting the change of the ESR value, and evaluate the inflammatory status.

[0040] The patient data acquisition unit is used to obtain the patient's physiological parameter data and living habit data. The body mass index calculation unit is used to calculate the body mass index BMI according to the patient's physiological parameters. The medical information acquisition unit is used to obtain the patient's blood test data, including red blood cell count, hemoglobin, white blood cell count, platelet count and CPR concentration, through the patient medical information management system. The patient's health and medical information is obtained, including chronic disease conditions, drug use, disease history, and recent infection or surgery information. The ESR value prediction unit is used to train the model through the long short-term memory network algorithm based on the patient's physiological parameter data, body mass index BMI, living habit data, blood test data, and health and medical information, to build an ESR value prediction model, predict the change trend of the patient's ESR value, and assign weights to the influence of each factor. The inflammation state judgment unit is used to obtain the inflammation state evaluation result based on the prediction result of the patient's ESR value. The inflammation state includes chronic state, acute state, and no inflammation.

[0041] For example, the physiological parameter data of a patient were obtained through the electronic health record system, including age 40 years old, male gender, weight 85 kg, height 175 cm, and non-pregnant status. The body mass index BMI calculated based on these data was 27.8. The patient's lifestyle data were obtained, including smoking 5 cigarettes a day, drinking 2 times a week, exercising 3 times a week for 1 hour each time, eating a high-protein and low-carbohydrate diet, and sleeping an average of 7 hours a night. The patient's blood test data were obtained through the automated laboratory information management system, including a red blood cell count of 5.2×10 12 / L, hemoglobin 15.5g / dL, white blood cell count 7.0×10 9 / L, platelet count 250×10 9 / L and C-reactive protein CRP concentration of 3.0mg / L. In addition, the patient's health and medical information was obtained, including a 5-year history of hypertension, current use of antihypertensive drugs, and no recent infection or surgery records. According to the patient's physiological parameter data, body mass index BMI, lifestyle data, blood test data and health and medical information, the model was trained through the long short-term memory network algorithm to construct an ESR value prediction model. The model predicts that the patient's ESR value will have a slight upward trend in the next 6 months, mainly affected by CRP concentration, smoking habits and history of hypertension, among which the influence weight of CRP concentration is 0.7, the influence weight of smoking habits is 0.2, and the influence weight of history of hypertension is 0.1. Based on these prediction results, it is assessed that the patient's inflammatory state is a chronic inflammatory state, and it is necessary to strengthen the management of smoking and eating habits, while continuing to monitor CRP concentration and blood pressure.

[0042] Step S104, the ESR prediction value deviation judgment module is used to compare the predicted ESR prediction value with the patient's actual measured ESR value. If there is a deviation between the actually measured ESR value and the predicted ESR prediction value, combined with the correlation between external factors and the ESR value, it is determined whether it is caused by a change in the inflammatory state.

[0043] The ESR value prediction execution unit is used to record and store the patient's measured ESR value in the electronic health record using the patient information management system. The ESR value prediction model is used to combine the patient's physiological parameter data, body mass index BMI, living habit data, blood test data, and health and medical information to predict the ESR value. The ESR value deviation analysis unit is used to calculate the deviation value between the measured ESR value and the predicted ESR value, and to determine whether the deviation value exceeds the preset deviation value range. If the deviation value exceeds the preset deviation value range, the patient's stress assessment data and drug use records are queried and extracted through the medical health database to determine the external factors that cause abnormal changes in ESR. The Pearson correlation coefficient calculation method is used to determine the correlation coefficient between the external factors and the ESR deviation value, and to determine the external factors whose correlation coefficient with the ESR deviation value is higher than the preset correlation coefficient threshold. If there is no external factor whose correlation coefficient with the ESR deviation value is higher than the preset correlation coefficient threshold, it is judged that the ESR deviation value is caused by changes in the inflammatory state, and the inflammatory state includes a chronic state and an acute state.

[0044] For example, the patient information management system records and stores the measured ESR value of a patient in the electronic health record. The ESR value prediction model is used to combine the patient's physiological parameter data, body mass index BMI27.3, lifestyle data, blood test data, and health and medical information to predict the ESR value to be 22 mm / hr. The physiological parameter data include age 45 years old, female gender, weight 70 kg, height 160 cm, lifestyle data include smoking 5 cigarettes a day, drinking 2 times a week, exercising 3 times a week, each time for 30 minutes, diet tends to be high in protein, and sleep an average of 6 hours a day. The blood test data include red blood cell count 4.8×10 12 / L, hemoglobin 13.5g / dL, white blood cell count 6.2×10 9 / L, platelet count 220×10 9 / L, CRP concentration is 3.2mg / L, health and medical information includes a 3-year history of diabetes, currently using insulin treatment. The actual measured ESR value is 30mm / hr. The deviation between the measured ESR value and the predicted ESR value is calculated to be 8mm / hr, and the deviation value is judged to exceed the preset deviation value range of ±5mm / hr. After the deviation value exceeds the preset range, the patient's stress assessment data and drug use records are queried and extracted through the medical health database, and it is found that the patient has been under great work pressure recently and has used a new drug. Using the Pearson correlation coefficient calculation method, it is determined that the correlation coefficient between the stress score and the ESR deviation value is 0.65, and the correlation coefficient between the drug use record is 0.50, both of which are higher than the preset correlation coefficient threshold of 0.4. Stress score and drug use record are added as new input variables in the ESR value prediction model, and the ESR value prediction model is retrained to calculate the revised ESR prediction value of 28mm / hr. The measured ESR value of 30 mm / hr was compared with the corrected ESR predicted value of 28 mm / hr again, and the deviation was judged to be 2 mm / hr, which is within the acceptable range, indicating that the deviation has been effectively corrected. Continue to monitor the changes in the patient's ESR value, paying special attention to its pressure concentration and medication use, to ensure the continued accuracy and effectiveness of the prediction model.

[0045] Step S105, the ESR value adjustment amount prediction module is used to continuously monitor the concentration changes of acute phase reactant proteins, including CRP and fibrinogen, evaluate the dynamic changes of inflammation, and construct an ESR value adjustment amount prediction model based on the patient's individual information and acute phase reactant protein concentration to predict the individual's ESR value adjustment amount.

[0046] The acute phase reactant protein monitoring unit is used to monitor the CRP concentration in the patient's blood sample using a sensor and record the instantaneous change of the CRP concentration. The optical sensor is used to detect the change in the optical density of fibrinogen in the blood sample to determine the concentration of fibrinogen in the blood sample. According to the time series data of the CRP concentration and the fibrinogen concentration, the Python Pandas and Matplotlib libraries are used to draw a time series graph of the change of CRP and fibrinogen concentration over time. The Pandas library is used to calculate the moving average of the CRP and fibrinogen concentrations, and the moving average trend line is drawn on the time series graph to show the change trend of the acute phase reactant protein, which includes CRP and fibrinogen. The ESR value adjustment amount prediction unit is used to train the model using the long short-term memory network algorithm based on the individual information of the patient and the acute phase reactant protein concentration, to build an ESR value adjustment amount prediction model, and to predict the patient's ESR value adjustment amount. The individual information of the patient includes age and gender.

[0047] Specifically, a sensor was used to monitor the C-reactive protein (CRP) concentration in a patient's blood sample, and the real-time changes in CRP concentration were recorded. The results showed that the CRP concentration increased from 1.2 mg / L to 3.8 mg / L within a week. At the same time, an optical sensor was used to detect the change in the optical density of fibrinogen in the patient's blood sample, and it was determined that the fibrinogen concentration increased from 2.5 g / L to 4.0 g / L. Based on the time series data of CRP and fibrinogen concentrations, Python's Pandas and Matplotlib libraries were used to draw a time series graph to show the changes in CRP and fibrinogen concentrations over time, and a trend line was generated using moving average to show the overall change trend of inflammatory markers. The results showed that both CRP and fibrinogen concentrations showed a significant upward trend, indicating an aggravation of the inflammatory response. The long short-term memory network algorithm was applied to construct an ESR value adjustment amount prediction model based on the patient's individual information, including age 50 years old, male gender, and acute phase reactant protein concentration, to predict the individual's ESR value adjustment amount. The model predicted that the patient's ESR value adjustment amount was 15 mm / hr, so the ESR value was adjusted from 20 mm / hr to 35 mm / hr, mainly affected by the significant increase in CRP and fibrinogen concentrations.

[0048] Step S106, the inflammatory status assessment module is used to obtain the baseline acute phase reactant protein concentration according to the information system. If the change value of the acute phase reactant protein concentration exceeds the baseline, the ESR test result is adjusted according to the acute phase reactant protein monitoring result, and the patient's inflammatory status is reassessed.

[0049] The acute phase reactant protein concentration comparison unit is used to obtain the patient's baseline acute phase reactant protein concentration through the information system and compare it with the current acute phase reactant protein concentration test result. The ESR value adjustment unit is used to adjust the current ESR value by predicting the ESR value adjustment amount according to the ESR value adjustment amount prediction model if the concentration of a certain acute phase reactant protein exceeds the preset acute protein concentration threshold. The inflammatory state reassessment unit is used to reassess the patient's inflammatory state based on the adjusted ESR value.

[0050] For example, the baseline acute phase reactant protein concentration of a patient is obtained through the information system, including a C-reactive protein CRP concentration of 1.5 mg / L and a fibrinogen concentration of 5 mg / L. The current test results show that the patient's ESR value is 30 mm / hr, the CRP concentration is 5 mg / L, and the fibrinogen concentration is 8 mg / L. Since the current CRP concentration exceeds the preset CRP threshold of 3 mg / L, the ESR value reassessment process is triggered. According to the ESR value adjustment amount prediction model, the ESR value adjustment amount is predicted to be -4.25 mm / hr, and the current ESR value of 30 mm / hr is adjusted to obtain an adjusted ESR value of 25.75 mm / hr. According to the adjusted ESR value and the preset acute inflammatory state ESR threshold of 25 mm / hr, the patient's inflammation level is determined. Since the adjusted ESR value of 25.75 mm / hr is still higher than the preset acute inflammatory state ESR threshold of 25 mm / hr, it indicates that the patient is still in an acute inflammatory state.

[0051] Step S107, the inflammatory status monitoring module is used to obtain clinical parameters, predict the probability of inflammatory status change, monitor the development trend of the inflammatory status in real time, and adjust the monitoring timing of ESR and acute phase reaction protein concentration.

[0052] The clinical parameter acquisition unit is used to obtain clinical parameters from the electronic health record system, including historical ESR values, acute phase reaction protein concentrations, basic patient information, lifestyle, and chronic disease history. Basic patient information includes age and gender, and lifestyle includes smoking history and eating habits. Clean the data, handle missing values ​​and outliers, and perform data conversion and standardization on different data types, including continuous variables and categorical variables. The inflammatory state change probability detection unit is used to define the inflammatory state change probability formula based on clinical parameters.

[0053] P(t)=α×log(E avg )+β×exp(CRP norm )+δ×exp(FIB norm )+∑(γ i ×X i ), where E avgis the average ESR over the past period of time, CRP norm is the normalized CRP concentration, FIB norm is the normalized fibrinogen concentration, X i Represents other clinical parameters, α, β, δ, γ i It is a formula parameter determined by historical clinical data. The inflammation intervention unit is used to monitor the development trend of the inflammatory state in real time through the inflammatory state change probability formula, and dynamically adjust the monitoring frequency, monitoring timing and intervention timing of the ESR value and acute phase reaction protein concentration according to the changing trend of the inflammatory state change probability.

[0054] For example, through the electronic health record system, the clinical parameters of a patient were obtained, including historical ESR values, acute phase reactant protein concentrations, basic patient information, lifestyle, and chronic disease history. Among them, the patient's basic information is age 50, male, lifestyle including smoking 10 cigarettes a day, and high-fat diet. The data was cleaned, missing values ​​and outliers were processed, and continuous and categorical variables were standardized. According to the clinical parameters, the probability formula for changes in inflammatory status was defined as P(t) = α × log(E avg )+β×exp(CRP norm )+δ×exp(FIB norm )+∑(γ i ×X i ), where E avg The average ESR value over the past period of time is 20 mm / hr, and the CRP norm is the standardized CRP concentration, 15 mm / hr, FIB norm is the standardized fibrinogen concentration, which is 8 mg / L, X i Represents other clinical parameters, such as smoking history, dietary habits, etc., α, β, γ i The formula parameters are determined by historical clinical data, which are 0.3, 0.5 and 0.2 respectively. The probability formula of inflammatory state change is used to monitor the development trend of the patient's inflammatory state in real time, and dynamically adjust the monitoring frequency, timing and inflammatory intervention plan of ESR value and acute phase reactant protein concentration according to the changing trend of the probability of inflammatory state change. When P(t) exceeds a certain threshold, such as 3.0, the system will recommend increasing the monitoring frequency, testing the ESR value and CRP concentration once a week, and recommending patients to reduce smoking and adjust their eating habits to effectively control the inflammatory state.

[0055] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A real-time monitoring system for blood sedimentation rate in inflammatory state, characterized in that: The system comprises: Patient physiological behavior analysis module, used to obtain the patient's physiological parameters, living habits data and ESR data, predict the changes in ESR values ​​under different physiological parameters and living habits, and identify the physiological parameters and living habits that affect the accuracy of ESR value measurement; The ESR value detection timing judgment module is used to establish correlation analysis through physiological behavior patterns and historical data of ESR value measurements, quantitatively score the degree to which different living habits affect the ESR value, and select the time period with the least impact on the test results as the best blood collection time for ESR value detection; The ESR value analysis module is used to build an ESR value prediction model based on the patient's physiological parameters, living habits data, and blood test results, predict the change trend of the ESR value, identify the factors that affect the change of the ESR value, and evaluate the inflammatory status; The ESR prediction value deviation judgment module is used to compare the predicted ESR value with the patient's actual measured ESR value. If the actual measured ESR value deviates from the predicted ESR value, it is determined whether it is caused by changes in the inflammatory state in combination with the correlation between external factors and the ESR value. The ESR value adjustment amount prediction module is used to continuously monitor the concentration changes of acute phase reaction proteins, including CRP and fibrinogen, to evaluate the dynamic changes of inflammation, and to construct an ESR value adjustment amount prediction model based on individual patient information and acute phase reaction protein concentrations to predict the individual ESR value adjustment amount; The inflammatory status assessment module is used to obtain the baseline acute phase reactant protein concentration according to the information system. If the change value of the acute phase reactant protein concentration exceeds the baseline, the ESR test result is adjusted according to the acute phase reactant protein monitoring result, and the patient's inflammatory status is reassessed; The inflammatory status monitoring module is used to obtain clinical parameters, predict the probability of changes in the inflammatory status, monitor the development trend of the inflammatory status in real time, and adjust the monitoring timing of ESR and acute phase reaction protein concentrations; The patient physiological behavior analysis module is used to obtain the patient's physiological parameters, living habits data and ESR data, predict the changes in ESR values ​​under different physiological parameters and living habits, and identify the physiological parameters and living habits that affect the accuracy of ESR value measurement, including: The patient data acquisition unit is used to obtain the patient's physiological parameters and living habits data. The physiological parameter data includes age, gender, weight and height, and pregnancy status. The living habits data includes smoking and drinking habits, exercise frequency, eating habits, and sleep patterns. According to the preset data quality standards, the data is screened and cleaned to remove incomplete, erroneous or abnormal data records. The key feature screening unit is used to convert the original data into a standard format, generate new feature variables based on the original data, and use principal component analysis PCA to select the most useful key features for ESR prediction. The ESR value influencing factor identification unit is used to obtain ESR data under different physiological parameters and living habits data through a ROLLER20 erythrocyte sedimentation meter, use the ARIMA algorithm to train the model, and predict the changes in ESR values ​​under different physiological parameters and living habits. Through the prediction results of the ARIMA model, the physiological parameters and living habits factors that have the greatest impact on the ESR value are identified, and their impact degree and confidence interval are calculated.

2. The system according to claim 1, wherein: The ESR value detection timing judgment module is used to establish a correlation analysis through the physiological behavior pattern and the historical data of the ESR value measurement, quantitatively score the degree to which different living habits affect the ESR value, select the time period with the least impact on the test result as the best blood sampling time, and perform ESR value detection, including: A comprehensive data acquisition unit is used to acquire physiological parameters, living habits, medical history, ESR measurement history data, and create a comprehensive database; pre-process the data in the comprehensive database, including removing duplicate records, filling missing values, identifying and processing outliers, and standardizing the data; A correlation factor identification unit, used to identify the correlation between the physiological parameters, living habits, medical history and ESR values ​​by calculating the Pearson correlation coefficient between the physiological parameters, living habits, medical history and ESR values, and screen out strong correlation factors whose correlation is higher than a preset correlation threshold; The impact assessment unit is used to train the model using a multivariate linear regression algorithm based on the strongly correlated factors screened out, build a lifestyle impact assessment model, and assign quantitative scores to the impact of different lifestyle habits on ESR; The blood sampling timing judgment unit is used to determine the degree of influence of different living habits on the ESR value based on the living habits impact assessment model, and based on the living habits in different time periods, use the weighted average method to obtain the degree of influence of different time periods on the ESR value, and determine the time period with the least influence on the ESR value as the best blood sampling time for ESR value detection.

3. The system according to claim 1, wherein: The ESR value analysis module is used to combine the patient's physiological parameters, living habits data and blood test results to build an ESR value prediction model, predict the change trend of the ESR value, identify the factors that affect the change of the ESR value, and evaluate the inflammatory state, including: Acquire the patient's physiological parameter data and living habit data through the patient data acquisition unit; The body mass index calculation unit is used to calculate the body mass index BMI according to the patient's physiological parameters; the medical information acquisition unit is used to obtain the patient's blood test data, including red blood cell count, hemoglobin, white blood cell count, platelet count and CPR concentration, through the patient medical information management system; obtain the patient's health and medical information, including chronic disease conditions, medication use, disease history, and recent infection or surgery information; the ESR value prediction unit is used to train the model through the long short-term memory network algorithm according to the patient's physiological parameter data, body mass index BMI, living habit data, blood test data, and health and medical information, to build an ESR value prediction model, predict the change trend of the patient's ESR value, and assign weights to the influence of each factor; The inflammation status judgment unit is used to obtain an inflammation status assessment result based on the prediction result of the patient's ESR value. The inflammation status includes chronic state, acute state, and no inflammation.

4. The system according to claim 1, wherein: The ESR prediction value deviation judgment module is used to compare the predicted ESR prediction value with the patient's measured ESR value. If the actually measured ESR value deviates from the predicted ESR prediction value, it is judged whether it is caused by the change of the inflammatory state in combination with the correlation between the external factors and the ESR value, including: An ESR value prediction execution unit, used to record and store the patient's actual ESR value in an electronic health record using a patient information management system; Use the ESR value prediction model to combine the patient's physiological parameter data, body mass index BMI, lifestyle data, blood test data, and health and medical information to predict the ESR value; The ESR value deviation analysis unit is used to calculate the deviation between the measured ESR value and the predicted ESR value, and determine whether the deviation exceeds the preset deviation value range; if the deviation exceeds the preset deviation value range, the patient's stress assessment data and drug use records are queried and extracted through the medical health database to determine the external factors that cause the abnormal change of ESR; Using the Pearson correlation coefficient calculation method, determining the correlation coefficient between the external factors and the ESR deviation value, and determining the external factors whose correlation coefficient with the ESR deviation value is higher than a preset correlation coefficient threshold; If there is no external factor whose correlation coefficient with the ESR deviation value is higher than the preset correlation coefficient threshold, it is determined that the ESR deviation value is caused by a change in the inflammatory state, which includes a chronic state and an acute state.

5. The system according to claim 1, wherein: The ESR value adjustment amount prediction module is used to continuously monitor the concentration changes of acute phase reaction proteins, including CRP and fibrinogen, evaluate the dynamic changes of inflammation, and construct an ESR value adjustment amount prediction model based on individual patient information and acute phase reaction protein concentrations to predict the individual ESR value adjustment amount, including: An acute phase reactant protein monitoring unit, which is used to monitor the CRP concentration in a patient's blood sample using a sensor and record the immediate changes in the CRP concentration; An optical sensor is used to detect the change in the optical density of fibrinogen in the blood sample to determine the concentration of fibrinogen in the blood sample; based on the time series data of CRP concentration and fibrinogen concentration, Python's Pandas and Matplotlib libraries are used to draw a time series graph of CRP and fibrinogen concentration changes over time; the Pandas library is used to calculate the moving average of CRP and fibrinogen concentrations, and a moving average trend line is drawn on the time series graph to show the changing trend of acute phase reactants, which include CRP and fibrinogen; an ESR value adjustment amount prediction unit is used to train the model using a long short-term memory network algorithm based on the patient's individual information and acute phase reactant protein concentration, to construct an ESR value adjustment amount prediction model, and to predict the patient's ESR value adjustment amount; the patient's individual information includes age and gender.

6. The system according to claim 1, wherein: The inflammatory status assessment module is used to obtain the baseline acute phase reactant protein concentration according to the information system. If the change value of the acute phase reactant protein concentration exceeds the baseline, the ESR test result is adjusted according to the acute phase reactant protein monitoring result, and the inflammatory status of the patient is reassessed, including: an acute phase reactant protein concentration comparison unit, which is used to obtain the patient's baseline acute phase reactant protein concentration through the information system and compare it with the current acute phase reactant protein concentration test result; An ESR value adjustment unit is used to adjust the current ESR value by predicting the ESR value adjustment amount according to the ESR value adjustment amount prediction model if the concentration of a certain acute phase reaction protein is detected to exceed a preset acute protein concentration threshold; The inflammatory status reassessment unit is used to reassess the patient's inflammatory status based on the adjusted ESR value.

7. The system according to claim 1, wherein: The inflammatory status monitoring module is used to obtain clinical parameters, predict the probability of inflammatory status changes, monitor the development trend of inflammatory status in real time, and adjust the monitoring timing of ESR and acute phase reaction protein concentrations, including: The clinical parameter acquisition unit is used to obtain clinical parameters from the electronic health record system, including historical ESR values, acute phase reaction protein concentrations, basic patient information, lifestyle, and chronic disease history. Basic patient information includes age and gender, and lifestyle includes smoking history and dietary habits; clean the data, process missing values ​​and outliers, and perform data conversion and standardization on different data types, including continuous variables and categorical variables; Inflammatory state change probability detection unit, used to define the inflammatory state change probability formula based on clinical parameters P(t)=α×log(E avg )+β×exp(CRP norm )+δ×exp(FIB norm )+∑(γ i ×X i ), where E avg is the average ESR over the past period of time, CRP norm is the normalized CRP concentration, FIB norm is the normalized fibrinogen concentration, X i Represents other clinical parameters, α, β, δ, γ i are the parameters of the formula, determined through historical clinical data; The inflammation intervention unit is used to monitor the development trend of the inflammatory state in real time through the inflammatory state change probability formula, and dynamically adjust the monitoring frequency, monitoring timing and intervention timing of the ESR value and acute phase reaction protein concentration according to the changing trend of the inflammatory state change probability.

Images