CN118280607A - Intelligent critical index monitoring method and system based on clinical data - Google Patents

Abstract

The invention discloses an intelligent monitoring method and system for critical indexes based on clinical data, which relate to the technical field of critical monitoring, and the method comprises the following steps: collecting clinical data of a patient and preprocessing; analyzing clinical data in real time by utilizing a severe index-to-return algorithm, identifying severe risk indexes, and obtaining severe pathological state characteristics of a patient; the patient's critical risk is dynamically assessed based on the critical pathological state characteristics, and when the patient's critical risk reaches a predetermined threshold, an alarm is issued and the healthcare worker is notified. The invention can comprehensively consider the interaction and influence among a plurality of basic severe indexes by utilizing a severe index-to-regression algorithm, extract severe index information with important indication significance for disease diagnosis and treatment from a large amount of clinical data, discover tiny change of the disease condition in time, and help doctors to better understand the change trend of the disease condition of patients.

Description

An intelligent monitoring method and system for critical illness indicators based on clinical data

Technical Field

The present invention relates to the technical field of critical illness monitoring, and in particular to a method and system for intelligently monitoring critical illness indicators based on clinical data.

Background technique

In the medical field, critical care monitoring and management are key links to ensure patient safety and improve treatment outcomes. Traditional critical care monitoring methods mainly rely on the experience of medical staff and regular physical sign detection, such as monitoring of vital signs and analysis of laboratory test results. These methods played an important role in early medical practice, helping doctors to understand patients' health status in a timely manner and make treatment decisions; however, with the advancement of medical technology and the development of data science and technology, traditional methods have gradually revealed their limitations in data processing capabilities, real-time monitoring efficiency, and the formulation of personalized treatment plans. In particular, in terms of processing large-scale clinical data and achieving high-frequency and real-time monitoring, traditional methods are often difficult to meet the needs of modern medical services due to human resource limitations.

Although traditional critical care monitoring methods are still effective in some cases, they have obvious shortcomings when dealing with rapidly changing clinical situations and processing complex information in a big data environment. First, traditional methods rely on regular measurements of vital signs and timed laboratory tests, which means that there are time intervals between data collection and continuous monitoring is not possible, which may result in missing key changes in the condition. Second, the efficiency and accuracy of manual data analysis are limited by the professional level and experience of medical staff, and fatigue misjudgments are prone to occur when faced with large amounts of data. In addition, traditional methods also have limitations in the formulation of personalized treatment plans, and it is often difficult to accurately formulate the most appropriate treatment plan based on the patient's specific situation.

The main drawbacks of existing critical care monitoring technologies are insufficient real-time monitoring capabilities, limited data processing efficiency and accuracy, and lack of personalized treatment support. First, the lack of continuous real-time monitoring capabilities means that sudden changes in the condition of critically ill patients cannot be detected in time, which is extremely critical in critical care management because rapid deterioration of the condition requires immediate intervention to avoid fatal consequences; second, with the explosive growth of medical data, traditional data processing methods can no longer efficiently process and analyze these data, thus affecting the timeliness and accuracy of disease diagnosis and treatment decisions; finally, the lack of in-depth understanding and analysis capabilities of individual differences makes traditional methods limited in effectiveness in formulating personalized treatment plans, making it difficult to maximize treatment effects and improve patient satisfaction; therefore, there is an urgent need for a critical care indicator monitoring technology that can perform in-depth analysis of large amounts of clinical data to address the obvious deficiencies of existing technologies in real-time monitoring capabilities, data processing efficiency, and personalized treatment support.

Currently, no effective solution has been proposed for the problems in the related technologies.

Summary of the invention

In response to the problems in the related technology, the present invention proposes an intelligent monitoring method and system for critical illness indicators based on clinical data, which has the advantages of being able to promptly detect slight changes in the condition, help doctors better understand the patient's condition change trends and improve the success rate of critical illness treatment, thereby solving the problems of insufficient real-time monitoring capabilities, limited data processing efficiency and accuracy, and lack of personalized treatment in the prior art.

To this end, the specific technical solution adopted by the present invention is as follows:

According to one aspect of the present invention, a method for intelligently monitoring critical disease indicators based on clinical data is provided, and the method for intelligently monitoring critical disease indicators based on clinical data comprises the following steps:

S1. Collect and preprocess the patient's clinical data;

S2. Use the critical illness indicator prognosis algorithm to analyze clinical data in real time, identify critical illness risk indicators, and obtain the patient's critical illness pathological state characteristics;

S3. Dynamically assess the patient's risk of severe illness based on the characteristics of severe pathological conditions. When the patient's risk of severe illness reaches a predetermined threshold, an alarm is issued and medical staff is notified.

Furthermore, using the critical illness indicator prognosis algorithm to analyze clinical data in real time, identify critical illness risk indicators, and obtain the patient's critical illness pathological state characteristics includes the following steps:

S21: Analyze the distribution and missing values of clinical data in real time, compare the correlation between each clinical data and critical illness results, and select basic critical illness indicators based on the correlation;

S22: According to medical guidelines and clinical practice, each basic critical illness indicator is assigned a weight and weighted summed to obtain a critical illness indicator score;

S23: Use the time window algorithm to extract the dynamic change characteristics of the critical illness index score and obtain the critical illness pathological state characteristics.

Furthermore, the distribution and missing values of clinical data are analyzed in real time, the correlation between each clinical data and the critical illness results is compared, and the basic critical illness indicators are screened out according to the correlation, including the following steps:

S211: identifying the data type of each clinical data in real time, generating basic descriptive statistics of the clinical data, and obtaining the frequency distribution of the clinical data;

S212: Real-time statistics of the number and proportion of missing values of each clinical data to obtain the distribution of missing values of the clinical data;

S213: Calculate the Pearson correlation coefficient between each clinical data and critical illness results;

S214: Compare the correlation between each clinical data and the critical illness results based on the calculation results, and select basic critical illness indicators based on the size of the correlation.

Furthermore, the dynamic change characteristics of the critical index score are extracted using the time window algorithm to obtain the critical pathological state characteristics, including the following steps:

S231: Determine the time window for analyzing the dynamic changes of critical illness index scores;

S232: In each time window, the statistical features of the critical illness index score are calculated to obtain a statistical feature set of the critical illness index score;

S233: According to the statistical feature set of the critical illness index score, the changing trend and periodic pattern of the critical illness index score within the time window are analyzed to obtain the critical illness pathological state characteristics.

Furthermore, determining the time window for analyzing the dynamic changes of the critical indicator score includes the following steps:

S2311: Determine the length of the time window based on the changing characteristics of the critical illness index score and the speed of its impact on the patient's condition;

S2312: Select the type of time window based on how quickly the critical illness index score affects the patient's condition.

Furthermore, in each time window, the statistical features of the critical illness index score are calculated to obtain the statistical feature set of the critical illness index score, including the following steps:

S2321: Calculate the mean and median of the critical index score in each time window, and determine the maximum and minimum values by sorting the data;

S2322: Calculate the standard deviation of the critical index score in each time window, and calculate the coefficient of variation based on the mean and standard deviation;

S2323: Compare the volatility of indicators in different time windows based on the coefficient of variation and calculate the skewness and kurtosis of the critical indicator score;

S2324: Summarize the calculation results of each time window to obtain a statistical feature set of critical indicator scores.

Furthermore, the calculation formula for the skewness of the critical index score is:

;

The calculation formula of the kurtosis of the critical index score is:

;

In the formula, PD represents the skewness of the critical index score;

FD represents the kurtosis of the severity index score;

N represents the number of time windows of observation;

x _n represents the value of the critical illness index score observed in the nth time window;

Represents the average value of the critical index score.

Further, according to the statistical feature set of the critical illness index score, analyzing the change trend and periodic pattern of the critical illness index score within the time window, and obtaining the critical illness pathological state characteristics includes the following steps:

S2331: According to the statistical feature set of the critical indicator score, select the statistical features of the continuous time period, and analyze the changing trend of the statistical features in the continuous time period;

S2332: Based on the analysis results of the change trend, identify the periodic change pattern of the statistical characteristics in the time series;

S2333: Cross-compare the changing trends and periodic change patterns of statistical characteristics to determine the characteristics of the patient's critical pathological state.

Further, based on the analysis results of the change trend, identifying the periodic change pattern of the statistical features in the time series includes the following steps:

S23321: Based on the analysis results of the change trend, select the time series data in the statistical characteristics of the critical indicator score;

S23322: Use Fourier transform to convert time series data into frequency domain for analysis;

S23323: Identify the frequency components in the analysis results, and determine the periodic variation patterns corresponding to the frequency components.

According to another aspect of the present invention, there is also provided a critical disease indicator intelligent monitoring system based on clinical data, the critical disease indicator intelligent monitoring system based on clinical data comprising:

Data collection module, used to collect and pre-process the patient's clinical data;

The real-time analysis module is used to analyze clinical data in real time using the critical illness indicator prognosis algorithm, identify critical illness risk indicators, and obtain the patient's critical illness pathological state characteristics;

A dynamic assessment module is used to dynamically assess the patient's risk of critical illness based on the characteristics of the critical pathological state. When the patient's risk of critical illness reaches a predetermined threshold, an alarm is issued and medical staff is notified;

Among them, the data collection module is connected with the dynamic evaluation module through the real-time analysis module.

The beneficial effects of the present invention are:

(1) The present invention utilizes a critical illness indicator prognosis algorithm, which can not only monitor a single basic critical illness indicator, but also comprehensively consider the interactions and influences between multiple basic critical illness indicators, thereby providing a more comprehensive and accurate disease assessment, and achieving accurate prediction and early warning of critical illness risks. This is of great help to doctors in formulating treatment plans and adjusting treatment strategies, making medical intervention more timely and effective. At the same time, by deeply integrating and analyzing patients' historical and real-time clinical data, the present application can extract indicator information that is of great significance for disease diagnosis and treatment from a large amount of clinical data, and can timely detect subtle changes in the condition of each patient according to the specific situation of each patient, provide scientific data support for doctors, and help doctors better understand the changing trends of the patient's condition, thereby tailoring the most appropriate treatment plan for each patient, greatly improving the success rate of treatment.

(2) The technical solution of the present invention can greatly improve the speed and accuracy of identifying critical illness risks through real-time detection and intelligent analysis. In clinical practice, the conditions of many critically ill patients change rapidly, and correct medical decisions need to be made as soon as possible. The present invention provides strong decision-making support for the medical team through continuous data monitoring and real-time risk assessment, enabling doctors to take timely intervention measures before the condition worsens, effectively avoiding further deterioration of the condition and reducing patient mortality and the incidence of complications. In addition, through in-depth analysis of clinical data, the present invention can also help the medical team identify potential medical risks and problems, such as adverse drug reactions, potential complications during treatment, etc., so that preventive or intervention measures can be taken at an early stage, further ensuring the safety of patients and preventing delays in the condition.

(3) The present invention can help hospitals and medical institutions to more effectively manage and allocate medical resources. In the traditional medical system, the allocation of medical resources often relies on experience and judgment, which is not only inefficient but also easy to cause waste of resources. The present invention can accurately predict the medical needs of different critically ill patients through real-time monitoring and analysis of clinical data, and provide scientific data support for hospital managers to help them more reasonably plan medical resources, such as the allocation of medical staff and ICU beds, thereby improving the efficiency of medical resource utilization. By analyzing the data during the patient's treatment process, the hospital can find out which treatment processes are inefficient and which medical measures contribute most to improving the patient's treatment effect, so as to adjust the treatment process and resource allocation in a targeted manner and improve the efficiency of medical resource utilization.

BRIEF DESCRIPTION OF THE DRAWINGS

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

FIG1 is a schematic flow chart of a method for intelligently monitoring critical illness indicators based on clinical data according to an embodiment of the present invention;

FIG2 is a principle block diagram of a critical illness indicator intelligent monitoring system based on clinical data according to an embodiment of the present invention.

Detailed ways

To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

According to an embodiment of the present invention, a method and system for intelligently monitoring critical illness indicators based on clinical data are provided.

The present invention is further described in conjunction with the accompanying drawings and specific embodiments. As shown in FIG1 , according to one embodiment of the present invention, a method for intelligently monitoring critical illness indicators based on clinical data is provided. The method for intelligently monitoring critical illness indicators based on clinical data comprises the following steps:

S1. Collect and preprocess the patient's clinical data;

Specifically, the source of the patient’s clinical data is the patient’s medical records, laboratory reports, and medical imaging data uploaded to the medical system;

Specifically, the patient's clinical data includes personal basic information, vital signs, laboratory test results, medical imaging data, clinical diagnosis and treatment records and medical records. Among them, personal basic information includes the patient's age, gender, weight, height, medical history, family medical history, and lifestyle habits (smoking and drinking); vital signs include the patient's blood pressure (systolic and diastolic), heart rate, body temperature, respiratory rate, and blood oxygen saturation; laboratory test results include the patient's blood tests (red blood cell count, white blood cell count, hemoglobin concentration), biochemical indicators (liver function indicators, kidney function indicators, blood sugar, blood lipids), and other specific tests (inflammatory indicators, myocardial enzyme spectrum); medical imaging data include the patient's X-rays, CT scans, MRI scans and ultrasound examination results; clinical diagnosis and treatment records include the patient's diagnosis results (including initial diagnosis and confirmed information), critical surgery and interventional treatment records, and drug treatment records (including drug name, dosage, and medication time); medical records include the patient's symptom changes, treatment response and follow-up information.

Specifically, after collecting the patient's clinical data, data preprocessing will be performed, including data cleaning, formatting and standardization, to ensure the quality and consistency of the data. In this embodiment, the clinical data of a patient with severe cardiovascular disease is taken as an example. The results obtained after preprocessing are as follows:

PatientID: unique patient identifier, used to distinguish data from different patients;

Age: the patient's age in years;

Gender: The patient's gender, such as Male, Female;

BMI: Body Mass Index, calculated based on the patient's weight and height;

SystolicBP: Systolic blood pressure, part of blood pressure measurement, measured in mmHg (millimeters of mercury);

DiastolicBP: Diastolic blood pressure, part of blood pressure measurement, measured in mmHg (millimeters of mercury);

Cholesterol: Cholesterol level in mg/dL (milligrams per deciliter);

BloodGlucose: blood sugar level in mg/dL (milligrams per deciliter);

Smoker: whether to smoke, yes (Yes) or no (No);

HistoryOfCVD: whether there is a history of cardiovascular disease, yes (Yes) or no (No);

Medication Adherence: Medication Adherence, High, Medium, Low;

Outcome: The study outcome variable, such as whether a cardiovascular event occurred during the follow-up period, yes (Yes) or no (No).

S2. Use the critical illness indicator prognosis algorithm to analyze clinical data in real time, identify critical illness risk indicators, and obtain the patient's critical illness pathological state characteristics;

S3. Dynamically assess the patient's risk of severe illness based on the characteristics of severe pathological conditions. When the patient's risk of severe illness reaches a predetermined threshold, an alarm is issued and medical staff is notified.

Specifically, the patient's critical illness risk is dynamically assessed taking into account the time series changes of clinical data, including dynamic changes in vital signs and laboratory test results; the threshold of critical illness risk is determined based on clinical experience and statistical analysis. When the patient's critical illness risk score exceeds the predetermined threshold, the system automatically generates an alarm, and the patient will be considered to be facing a high-risk state. At this time, the critical illness risk alarm is sent to relevant medical staff in real time through the hospital's internal communication system. Once a high-risk alarm is received, medical staff must take immediate action and take clinical intervention measures for each critically ill patient who triggers the alarm.

Specifically, first, based on historical data analysis, expert opinions and clinical guidelines, define different levels of critical illness risk thresholds, and regularly review and adjust risk thresholds based on new clinical data and feedback; then use the hospital information system (HIS) and real-time monitoring equipment to collect patients' clinical data in real time, and pre-process the collected real-time data, input the pre-processed real-time data into the deployed critical illness indicator monitoring model, and determine the patient's critical illness risk level based on the risk score output by the model. When the patient's critical illness risk score exceeds the preset threshold, an alarm is automatically triggered. The alarm information includes patient identification, current risk score and recommended clinical intervention measures; through application notifications, text messages and other methods, ensure that medical staff can receive alarms in time to evaluate patients' critical illness risks in real time. For high-risk alarms, implement priority sorting to ensure that the critical care medical team can take immediate action; finally, track changes in the clinical condition of critically ill patients after intervention and evaluate the effectiveness of intervention measures.

In one embodiment, using a critical illness indicator prognosis algorithm to analyze clinical data in real time, identify critical illness risk indicators, and obtain the patient's critical illness pathological state characteristics includes the following steps:

S21: Analyze the distribution and missing values of clinical data in real time, compare the correlation between each clinical data and critical illness results, and select basic critical illness indicators based on the correlation;

S22: According to medical guidelines and clinical practice, each basic critical illness indicator is assigned a weight and weighted summed to obtain a critical illness indicator score;

Specifically, in this embodiment, taking the critical care monitoring of a patient with a serious cardiovascular disease as an example, firstly, a cardiovascular disease expert group is organized to discuss and reach a consensus to determine which basic critical care indicators are highly correlated with the risk of cardiovascular critical care, and the factors considered include the sensitivity, specificity and ease of monitoring of the indicators. In addition, the latest cardiovascular disease-related research is reviewed to determine which basic critical care indicators have shown significant predictive value for the risk of critical care in recent studies. Then, according to the contribution of each basic critical care indicator to the risk of cardiovascular critical care, a corresponding weight is assigned. For example, studies have shown that hypertension contributes more to the risk of critical care than blood sugar, so hypertension is given a higher weight in the composite indicator. Then, a weighted summation method is used to combine the basic critical care indicators to obtain a critical care indicator score.

Specifically, basic critical illness indicators refer to a set of clinical data that can reflect the patient's basic health status and potential risk of critical illness. In this embodiment, for patients with critical cardiovascular diseases, the basic critical illness indicators include: blood pressure (including systolic and diastolic blood pressure), heart rate, blood cholesterol level (including total cholesterol, low-density lipoprotein and high-density lipoprotein), blood sugar level, body mass index (BMI) and inflammatory marker level (C-reactive protein).

Specifically, according to the latest research results on cardiovascular disease prevention, a weight is assigned to each basic critical indicator. In this embodiment, studies have shown that the impact of hypertension on cardiovascular risk is twice that of hyperglycemia, so the weight of hypertension is set to twice that of hyperglycemia, and then the weighted summation method is used to combine the key indicators. The specific formula is: critical indicator score = (hypertension score × hypertension weight) + (hyperglycemia score × hyperglycemia weight) + (LDL score × LDL weight) + (HDL score × HDL weight) + (BMI score × BMI weight) + (C-reactive protein score × C-reactive protein Weight), thereby obtaining a specific critical illness index score by calculation. The critical illness index score is a value from 0 to 100. The present invention specifically divides the risk level into: low risk (0-33 points), medium risk (34-66 points) and high risk (67-100 points), which directly reflects the degree of critical illness risk faced by critically ill patients. The higher the score, the greater the patient's critical illness risk. Through this method, the critical illness risk assessment of critically ill patients becomes more accurate and practical, and doctors can formulate more personalized treatment and management plans for patients based on the specific value and risk level of the critical illness index score.

S23: Use the time window algorithm to extract the dynamic change characteristics of the critical illness index score and obtain the critical illness pathological state characteristics.

In one embodiment, real-time analysis of the distribution and missing values of clinical data, comparison of the correlation between each clinical data and critical illness results, and screening out basic critical illness indicators according to the correlation size include the following steps:

S211: identifying the data type of each clinical data in real time, generating basic descriptive statistics of the clinical data, and obtaining the frequency distribution of the clinical data;

S212: Real-time statistics of the number and proportion of missing values of each clinical data to obtain the distribution of missing values of the clinical data;

S213: Calculate the Pearson correlation coefficient between each clinical data and critical illness results;

S214: Compare the correlation between each clinical data and the critical illness results based on the calculation results, and select basic critical illness indicators based on the size of the correlation.

In one embodiment, extracting the dynamic change characteristics of the critical illness index score using a time window algorithm to obtain the critical illness pathological state characteristics includes the following steps:

S231: Determine the time window for analyzing the dynamic changes of critical illness index scores;

S232: In each time window, the statistical features of the critical illness index score are calculated to obtain a statistical feature set of the critical illness index score;

S233: According to the statistical feature set of the critical illness index score, the changing trend and periodic pattern of the critical illness index score within the time window are analyzed to obtain the critical illness pathological state characteristics.

In one embodiment, determining the time window for analyzing the dynamic changes of the critical illness index score includes the following steps:

S2311: Determine the length of the time window based on the changing characteristics of the critical illness index score and the speed of its impact on the patient's condition;

S2312: Select the type of time window based on how quickly the critical illness index score affects the patient's condition.

Specifically, determine the length of the time window for analyzing the dynamic changes of composite critical illness indicators, such as 24 hours or 48 hours, which depends on the changing speed of critical pathological conditions and clinical monitoring needs; select a sliding time window as the type of time window for continuous real-time monitoring.

In one embodiment, in each time window, calculating the statistical features of the critical illness index score to obtain the statistical feature set of the critical illness index score includes the following steps:

S2321: Calculate the mean and median of the critical illness index score in each time window, and determine the maximum and minimum values by sorting the data;

S2322: Calculate the standard deviation of the critical index score in each time window, and calculate the coefficient of variation based on the mean and standard deviation;

S2323: Compare the volatility of indicators in different time windows based on the coefficient of variation and calculate the skewness and kurtosis of the critical indicator score;

S2324: Summarize the calculation results of each time window to obtain a statistical feature set of critical illness indicator scores.

In one embodiment, the calculation formula of the skewness of the severity index score is:

;

The calculation formula of the kurtosis of the critical index score is:

;

In the formula, PD represents the skewness of the critical index score;

FD represents the kurtosis of the severity index score;

N represents the number of time windows of observation;

x _n represents the value of the critical illness index score observed in the nth time window;

Represents the average value of the critical index score.

In one embodiment, according to the statistical feature set of the critical illness index score, analyzing the change trend and periodic pattern of the critical illness index score within the time window to obtain the critical illness pathological state feature includes the following steps:

S2331: According to the statistical feature set of the critical indicator score, select the statistical features of the continuous time period, and analyze the changing trend of the statistical features in the continuous time period;

Specifically, the statistical characteristic data of critical illness index scores for several consecutive days or weeks are selected, and linear regression analysis is applied. Time is used as the independent variable, and the daily average value of the composite critical illness index is used as the dependent variable for analysis. The trend is judged by the slope of the regression line. A positive slope indicates that the index score is on an upward trend, and a negative slope indicates that the index score is on a downward trend. In both cases, the treatment plan needs to be adjusted. A slope close to zero indicates that the patient's vital signs remain stable.

S2332: Based on the analysis results of the change trend, identify the periodic change pattern of the statistical characteristics in the time series;

Specifically, by identifying the periodic change pattern of critical illness index scores, the periodic regularity of the patient's pathological state is revealed. Taking patients with cardiovascular disease as an example, the daily average data of blood pressure and blood sugar in the past week are selected from the statistical characteristics of their critical illness index scores, and Fourier transform analysis is applied to find an obvious 24-hour periodic change pattern, indicating that the patient's blood pressure and blood sugar levels have significant daily cyclic fluctuations, which may be related to the patient's living habits and drug intake time. Therefore, doctors can adjust the drug dosage and monitoring time accordingly to better control the patient's blood pressure and blood sugar levels.

S2333: Cross-compare the changing trends and periodic change patterns of statistical characteristics to determine the characteristics of the patient's critical pathological state.

Specifically, the changing trends of statistical characteristics and the periodic changing patterns are cross-compared. First, the relationship and consistency between the two are analyzed, and the statistical characteristics of the patient's critical index scores are divided into stability characteristics and dynamic characteristics. Among them, the stability characteristics are the characteristics that persist in the long-term trend analysis, and the dynamic characteristics are the fluctuation characteristics shown in the periodic pattern analysis; then the contribution of each characteristic to the patient's critical risk is evaluated, and it is determined which statistical characteristics are the main pathological state characteristics. Finally, a detailed pathological state characteristic report is written, including a specific description of the stability characteristics and dynamic characteristics, as well as their impact and suggestions on the patient's health status.

Specifically, taking a patient with cardiovascular disease as an example, through the change trend analysis, it is found that the patient's blood pressure is on an upward trend, and through the periodic change pattern recognition, it is shown that the patient's blood pressure has daily fluctuations. After cross-comparison of these analysis results, we can determine that the patient has the following severe pathological state characteristics: the patient's stability characteristic is persistent hypertension, and the patient's blood pressure level is continuously higher than the normal range, indicating an increased long-term cardiovascular risk; the patient's dynamic characteristic is daily fluctuations in blood pressure, and the patient's blood pressure fluctuates significantly in the morning and evening, which may be related to daily activities and drug use time; by determining the patient's severe pathological state characteristics, the above-mentioned stability characteristics and dynamic characteristics are specifically described in the pathological state characteristic report, and the following suggestions are made: for persistent hypertension, it is recommended to increase the dose of antihypertensive drugs or adjust the medication time; for daily blood pressure fluctuations, it is recommended that patients conduct daily blood pressure monitoring and take drugs 1 hour before the peak of blood pressure in the morning and evening. Through such an embodiment, professionals can clearly understand how to summarize the patient's severe pathological state characteristics from the change trend and periodic change pattern analysis, and provide personalized treatment suggestions for patients.

In one embodiment, based on the analysis result of the change trend, identifying the periodic change pattern of the statistical feature in the time series includes the following steps:

S23321: Based on the analysis results of the change trend, select the time series data in the statistical characteristics of the critical indicator score;

S23322: Use Fourier transform to convert time series data into frequency domain for analysis;

S23323: Identify the frequency components in the analysis results, and determine the periodic variation patterns corresponding to the frequency components.

In order to facilitate understanding of the above technical solution of the present invention, the following is a specific description taking the cardiology department of a hospital as an example:

In the cardiology department of a certain hospital, the medical team of the department needs to face the monitoring and treatment tasks of a large number of critically ill patients with cardiovascular diseases every day. In order to improve the early identification ability of critically ill risk patients and reduce emergencies, the hospital adopts the intelligent monitoring method and system of critical indicators based on clinical data proposed by the present invention; a critically ill patient admitted to the department is a 65-year-old male with a long history of hypertension and diabetes, and was recently diagnosed with cardiovascular disease. In view of his multiple risk factors, the medical team included him in the critical monitoring plan. Since the critically ill patient was admitted to the hospital, the medical staff collected the patient's daily blood pressure, blood sugar, heart rate, weight and other clinical data through the hospital information system, and analyzed the distribution and missing values of the clinical data of critically ill patients in real time, screened out basic critical indicators, and calculated the Pearson correlation coefficient between each clinical data and the critical results, assigned weights to each basic critical indicator and The weighted sum is used to obtain the critical illness index score, and then the time window algorithm is used to extract the dynamic change characteristics of the critical illness index score to obtain the critical illness pathological state characteristics. Finally, based on the dynamic evaluation results of the critical illness index score, when the patient's critical illness risk reaches the predetermined threshold, an alarm is automatically issued and the medical staff is notified. The alarm includes the patient's personal information, the current value of the critical illness index, the risk rating and the recommended clinical intervention measures. Through intelligent monitoring of critical illness indicators based on clinical data, medical staff can promptly detect abnormal fluctuations in the blood pressure and blood sugar of the critically ill patient, and adjust the treatment plan according to the suggestions provided by the system, such as increasing the dosage or changing the medication. This method greatly improves the early identification ability of patients at risk of critical illness, reduces the incidence of cardiovascular emergencies, improves treatment effects and patient satisfaction, and can provide more accurate and personalized monitoring and treatment plans for critically ill patients.

As shown in FIG. 2 , according to another embodiment of the present invention, there is also provided a critical disease indicator intelligent monitoring system based on clinical data, the critical disease indicator intelligent monitoring system based on clinical data comprising:

Data collection module 1, used to collect and pre-process the patient's clinical data;

Real-time analysis module 2, used to analyze clinical data in real time using the critical illness indicator prognosis algorithm, identify critical illness risk indicators, and obtain the patient's critical illness pathological state characteristics;

Dynamic assessment module 3, used to dynamically assess the patient's critical risk based on the characteristics of the critical pathological state, and when the patient's critical risk reaches a predetermined threshold, an alarm is issued and medical staff is notified;

The data collection module 1 is connected to the dynamic evaluation module 3 via the real-time analysis module 2 .

In summary, with the help of the above-mentioned technical scheme of the present invention, by constructing a critical indicator monitoring model and using a critical indicator prognosis algorithm, it is possible not only to monitor a single critical indicator, but also to comprehensively consider the interaction and influence between multiple critical indicators, thereby providing a more comprehensive and accurate disease assessment, and realizing accurate prediction and early warning of critical risk, which is of great help to doctors in formulating treatment plans and adjusting treatment strategies, making medical intervention more timely and effective. At the same time, by deeply integrating and analyzing the patient's historical and real-time clinical data, the present application can extract indicator information that is of great indicative significance for disease diagnosis and treatment from a large amount of clinical data, and can timely discover slight changes in the condition according to the specific situation of each patient, provide scientific data support for doctors, and help doctors better understand the trend of the patient's condition change, so as to tailor the most appropriate treatment plan for each patient, greatly improving the success rate of treatment; the technical scheme of the present invention can greatly improve the speed and accuracy of identifying critical risks through real-time detection and intelligent analysis. In clinical practice, the condition of many critically ill patients changes rapidly, and correct medical decisions need to be made as soon as possible. The present invention provides the medical team with continuous data monitoring and real-time risk assessment. Strong decision support enables doctors to take timely intervention measures before the condition worsens, effectively avoiding further deterioration of the condition and reducing the mortality rate and incidence of complications of patients. In addition, through in-depth analysis of clinical data, the present invention can also help the medical team identify potential medical risks and problems, such as adverse drug reactions, potential complications during treatment, etc., so as to take preventive or intervention measures at an early stage, further ensuring the safety of patients and preventing delays in the condition. The present invention can help hospitals and medical institutions to more effectively manage and allocate medical resources. In the traditional medical system, the allocation of medical resources often relies on empirical judgment, which is not only inefficient but also easy to cause waste of resources. The present invention can accurately predict the distribution and medical needs of critically ill patients through real-time monitoring and analysis of clinical data, provide scientific data support for hospital managers, and help them more reasonably plan medical resources, such as the allocation of medical staff and ICU beds, so as to improve the efficiency of the use of medical resources. By analyzing the data during the patient's treatment process, the hospital can find out which treatment processes are inefficient and which medical measures contribute the most to improving the patient's treatment effect, so as to adjust the treatment process and resource allocation in a targeted manner and improve the efficiency of the use of medical resources.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (10)

1. A method for intelligently monitoring critical disease indicators based on clinical data, characterized in that the method for intelligently monitoring critical disease indicators based on clinical data comprises the following steps: S1. Collect and preprocess the patient's clinical data; S2. Use the critical illness indicator prognosis algorithm to analyze clinical data in real time, identify critical illness risk indicators, and obtain the patient's critical illness pathological state characteristics; S3. Dynamically assess the patient's risk of severe illness based on the characteristics of severe pathological conditions. When the patient's risk of severe illness reaches a predetermined threshold, an alarm is issued and medical staff is notified. 2. According to claim 1, a critical disease indicator intelligent monitoring method based on clinical data is characterized in that the use of a critical disease indicator outcome algorithm to analyze clinical data in real time, identify critical disease risk indicators, and obtain the patient's critical disease pathological state characteristics includes the following steps: S21: Analyze the distribution and missing values of clinical data in real time, compare the correlation between each clinical data and critical illness results, and select basic critical illness indicators based on the correlation; S22: According to medical guidelines and clinical practice, each basic critical illness indicator is assigned a weight and weighted summed to obtain a critical illness indicator score; S23: Use the time window algorithm to extract the dynamic change characteristics of the critical illness index score and obtain the critical illness pathological state characteristics. 3. According to claim 2, a method for intelligent monitoring of critical illness indicators based on clinical data is characterized in that the real-time analysis of the distribution and missing values of clinical data, comparison of the correlation between each clinical data and critical illness results, and screening out basic critical illness indicators according to the size of the correlation comprises the following steps: S211: identifying the data type of each clinical data in real time, generating basic descriptive statistics of the clinical data, and obtaining the frequency distribution of the clinical data; S212: Real-time statistics of the number and proportion of missing values of each clinical data to obtain the distribution of missing values of the clinical data; S213: Calculate the Pearson correlation coefficient between each clinical data and critical illness results; S214: Compare the correlation between each clinical data and the critical illness results based on the calculation results, and select basic critical illness indicators based on the size of the correlation. 4. According to the method of claim 2, wherein the method of extracting the dynamic change characteristics of the critical index score by using the time window algorithm to obtain the critical pathological state characteristics comprises the following steps: S231: Determine the time window for analyzing the dynamic changes of critical illness index scores; S232: In each time window, the statistical features of the critical illness index score are calculated to obtain a statistical feature set of the critical illness index score; S233: Based on the statistical feature set of the critical illness index score, the changing trend and periodic pattern of the critical illness index score within the time window are analyzed to obtain the critical illness pathological state characteristics. 5. According to the method of intelligent monitoring of critical illness indicators based on clinical data in claim 4, it is characterized in that the time window for determining and analyzing the dynamic changes of critical illness indicator scores comprises the following steps: S2311: Determine the length of the time window based on the changing characteristics of the critical illness index score and the speed of its impact on the patient's condition; S2312: Select the type of time window based on how quickly the critical illness index score affects the patient's condition. 6. According to the method of intelligent monitoring of critical illness indicators based on clinical data in claim 4, it is characterized in that the step of calculating the statistical characteristics of the critical illness indicator score in each time window to obtain the statistical characteristic set of the critical illness indicator score comprises the following steps: S2321: Calculate the mean and median of the critical illness index score in each time window, and determine the maximum and minimum values by sorting the data; S2322: Calculate the standard deviation of the critical index score in each time window, and calculate the coefficient of variation based on the mean and standard deviation; S2323: Compare the volatility of indicators in different time windows based on the coefficient of variation and calculate the skewness and kurtosis of the critical indicator score; S2324: Summarize the calculation results of each time window to obtain a statistical feature set of critical indicator scores. 7. According to the method for intelligent monitoring of critical illness indicators based on clinical data in claim 6, it is characterized in that the calculation formula of the skewness of the critical illness indicator score is: ; The calculation formula of the kurtosis of the critical index score is: ; In the formula, PD represents the skewness of the critical index score; FD represents the kurtosis of the severity index score; N represents the number of time windows of observation; x _n represents the value of the critical illness index score observed in the nth time window; Represents the average value of the critical index score. 8. According to the method of claim 4, the method is characterized in that the method comprises the following steps: analyzing the changing trend and periodic pattern of the critical illness index score within the time window according to the statistical feature set of the critical illness index score to obtain the critical illness pathological state characteristics: S2331: According to the statistical feature set of the critical indicator score, select the statistical features of the continuous time period, and analyze the changing trend of the statistical features in the continuous time period; S2332: Based on the analysis results of the change trend, identify the periodic change pattern of the statistical characteristics in the time series; S2333: Cross-compare the changing trends and periodic change patterns of statistical characteristics to determine the characteristics of the patient's critical pathological state. 9. According to claim 8, a method for intelligent monitoring of critical illness indicators based on clinical data is characterized in that the analysis result based on the change trend and identifying the periodic change pattern of statistical features in the time series comprises the following steps: S23321: Based on the analysis results of the change trend, select the time series data in the statistical characteristics of the critical indicator score; S23322: Use Fourier transform to convert time series data into frequency domain for analysis; S23323: Identify the frequency components in the analysis results, and determine the periodic variation patterns corresponding to the frequency components. 10. An intelligent monitoring system for critical illness indicators based on clinical data, used to implement the intelligent monitoring method for critical illness indicators based on clinical data according to any one of claims 1 to 9, characterized in that the intelligent monitoring system for critical illness indicators based on clinical data comprises: Data collection module, used to collect and pre-process the patient's clinical data; The real-time analysis module is used to analyze clinical data in real time using the critical illness indicator prognosis algorithm, identify critical illness risk indicators, and obtain the patient's critical illness pathological state characteristics; A dynamic assessment module is used to dynamically assess the patient's critical risk based on the characteristics of the critical pathological state. When the patient's critical risk reaches a predetermined threshold, an alarm is issued and medical staff is notified; Wherein, the data collection module is connected to the dynamic evaluation module through the real-time analysis module.