Disease condition analysis and early warning method for pneumology department nursing based on big data

Abstract

The invention discloses an illness state analysis and early warning method for pneumology department nursing based on big data, and relates to the technical field of pneumology department nursing, and the method comprises the following steps: collecting multiple pieces of vital sign data of a patient in a nursing process, and carrying out the preprocessing and integration of the collected data to form a complete data set; according to the big data-based disease condition analysis and early warning method for pneumology department nursing, the vital sign data of the patient is monitored in real time and comprehensively analyzed, and real-time and multi-dimensional data collection and analysis are utilized, so that the accuracy and timeliness of disease condition assessment are greatly improved, abnormal data fluctuation of the patient can be found in time, early warning is carried out in advance, and the patient experience is improved. According to the invention, the medical staff can take intervention measures before the condition deteriorates, so that the further development of the condition is effectively avoided, and the preventive health management is adopted, so that the treatment effect is improved, and the medical cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of respiratory nursing, and in particular to a disease condition analysis and early warning method for respiratory nursing based on big data. Background Art

[0002] Respiratory diseases, especially chronic respiratory diseases (such as chronic obstructive pulmonary disease, asthma, etc.) and acute respiratory infections, are one of the leading causes of death and disability worldwide. These diseases not only bring huge physical burdens to patients, but also have serious impacts on the social economy. In recent years, with the continuous development of science and technology, the analysis and application of big data has become a key tool in many industries. In the medical field, the analysis and application of big data is playing an increasingly important role. Therefore, using big data to assist in accurate analysis of respiratory diseases and timely warning is of great significance to improving patients' treatment effects and reducing medical costs.

[0003] In the process of respiratory nursing, the patient's condition is not related to a single vital sign. Using a single vital sign to analyze the patient's condition will result in accuracy deviation. Therefore, how to comprehensively judge the patient's condition based on the patient's vital signs to conduct condition analysis and early warning is the problem we need to solve. To this end, a condition analysis and early warning method for respiratory nursing based on big data is proposed. Summary of the invention

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: A respiratory nursing condition analysis and early warning method based on big data, comprising the following steps:

[0005] Step 1: Collect multiple vital sign data of the patient during the nursing process, and pre-process and integrate the collected data to form a complete data set, where the vital sign data includes respiratory rate, respiratory depth and heart rate;

[0006] Step 2: extract features related to the patient's condition from the integrated data set, preset abnormal thresholds for different features, identify abnormal features, and obtain an abnormal feature sequence list;

[0007] Step 3: Analyze the feature data of the abnormal feature sequence table, determine the frequency of occurrence of each abnormal feature, and mark the abnormal features in the intersection area;

[0008] Step 4: Based on the feature analysis results, analyze the condition of the patient with respiratory disease and determine the severity of the patient's condition;

[0009] Step 5: Based on the analysis results of the severity of the patient's condition, different warning levels are preset, and warning notifications are pushed to medical staff based on the patient's condition;

[0010] Step 6: Based on the pushed warning notification, medical staff take corresponding intervention measures.

[0011] Preferably, in step 1, the process of collecting the patient's vital signs data includes:

[0012] Identify the types of vital signs that need to be collected for patients with respiratory diseases, namely respiratory rate, respiratory depth, and heart rate;

[0013] The patient's various vital signs data are collected through medical devices and sensors, including respiratory rate measured by a respiratory monitoring device, respiratory depth obtained by monitoring the chest of the patient during breathing with a respiratory belt, and heart rate measured by a heart rate monitoring device;

[0014] Preprocessing of various types of vital sign data collected, including data cleaning, data format conversion and data standardization. Redundant, invalid or abnormal data are removed through data cleaning. Values ​​that are obviously deviated from the normal range are reviewed or eliminated. Data from different sources are converted into a unified data format to facilitate subsequent analysis and processing, and the data are standardized to eliminate dimensional differences between data from different sources.

[0015] The pre-processed vital signs data are integrated to form a structured data set, which integrates the patient's identity information, vital signs data, and measurement time information to build a comprehensive health record for the patient.

[0016] Preferably, in step 2, the process of obtaining the abnormal feature sequence list includes:

[0017] Analyze the integrated data set to extract the respiratory frequency features, respiratory depth features, and heart rate features related to the condition of patients with respiratory diseases;

[0018] Based on historical data and standards of care for respiratory diseases, set corresponding abnormal thresholds for each feature to distinguish between normal and abnormal conditions;

[0019] Compare each feature data in the data set with the corresponding abnormal threshold, identify the feature value that exceeds the threshold, mark the identified abnormal feature value, and record the occurrence time, duration and related patient information of the abnormal feature for subsequent analysis and processing;

[0020] The marked abnormal features are reviewed to ensure that the abnormal features reflect the patient's abnormal condition and exclude false positives or interference. The abnormal features are sorted according to the severity of the abnormality to obtain an abnormal feature sequence list, which contains the name, abnormal value, abnormal degree and pathological significance of each abnormal feature.

[0021] Preferably, the process of setting the characteristic abnormality threshold is specifically as follows:

[0022] For respiratory rate characteristics, the normal respiratory rate of an adult in a quiet state is 12-20 times / minute. The lower limit of the abnormal respiratory rate threshold is set to less than 10 times / minute (bradypnea), and the upper limit is greater than 24 times / minute (tachypnea).

[0023] For the breathing depth feature, the breathing depth of a normal adult during calm breathing is about 4-6 cm. The lower limit of the abnormal threshold is set to less than 3 cm (too shallow breathing) and the upper limit is greater than 8 cm (too deep breathing).

[0024] For heart rate characteristics, the normal heart rate of an adult at rest is 60-100 beats / minute. The lower limit of the abnormal threshold is set to less than 50 beats / minute (bradycardia) and the upper limit is set to greater than 110 beats / minute (tachycardia).

[0025] Preferably, in step 3, the abnormal feature analysis process includes:

[0026] Extract the data of each abnormal feature from the abnormal feature sequence list, including the name of the abnormal feature and the corresponding record, and use data analysis tools to perform frequency statistics on each abnormal feature, calculate the number of times each abnormal feature appears, and display the frequency statistics results in a table format for intuitive comparison and analysis;

[0027] Analyze the frequency of occurrence of each abnormal feature, set the abnormal feature judgment threshold, identify high-frequency abnormal features, analyze the number of times two or more abnormal features appear at the same time, and determine the intersection area where abnormal features exist;

[0028] Compare the data in the abnormal feature sequence table with the intersection area where the abnormal features exist, mark the identified abnormal features with intersection areas, and record the name, abnormal value and frequency of occurrence of the intersection abnormal features;

[0029] Analyze the correlation and mutual influence between the abnormal features in the intersection area, further analyze the frequency of occurrence of each abnormal feature and the analysis results of the abnormal features in the intersection area, and clarify the high-frequency abnormal features and intersection abnormal features. Among them, high-frequency abnormal features include high-frequency abnormal respiratory rate features, high-frequency abnormal respiratory depth features, and high-frequency abnormal heart rate features. Intersection abnormal features include abnormal respiratory rate-respiratory depth features, abnormal respiratory rate-heart rate features, abnormal respiratory depth-heart rate features, and abnormal respiratory rate-respiratory depth-heart rate features.

[0030] Preferably, in step 4, the process of analyzing the condition of patients with respiratory diseases includes:

[0031] Based on the divided high-frequency abnormal features and intersection abnormal features, scoring standards are set for high-frequency abnormal features and intersection abnormal features, and the score of high-frequency abnormal features is set to 1 point, the score of intersection abnormal features with two abnormal features is set to 2 points, and the score of intersection abnormal features with three abnormal features is set to 3 points (high-frequency abnormal features include high-frequency respiratory frequency abnormal features, high-frequency respiratory depth abnormal features, and high-frequency heart rate abnormal features, and intersection abnormal features include respiratory frequency-respiratory depth abnormal features, respiratory frequency-heart rate abnormal features, respiratory depth-heart rate abnormal features, and respiratory frequency-respiratory depth-heart rate abnormal features);

[0032] According to the actual situation of the patient, the scores of the high-frequency abnormal features and the intersection abnormal features are calculated according to the scoring criteria, and according to the needs of the condition analysis of the patient with respiratory diseases, the corresponding weight coefficients are set for each high-frequency abnormal feature and the intersection abnormal feature;

[0033] The scores and weight coefficients of each abnormal feature are combined to calculate the disease evaluation index and analyze the severity of the patient's condition.

[0034] Preferably, the calculation expression of the disease evaluation index is:

[0035] ;

[0036] ;

[0037] in, is the disease assessment index, The value range is between 0 and 1. is the sum of the weighted scores of all abnormal features, is the weight coefficient of the ith abnormal feature, is the score of the ith abnormal feature, is the number of abnormal features, that is, the sum of high-frequency abnormal features and intersection abnormal features, , The baseline value for all abnormal feature scores is the average value calculated based on historical data. is the standard deviation, which is used to adjust the sensitivity of the index and reflects the degree of variability in the abnormal feature scores.

[0038] Preferably, in step 5, the process of setting the warning level includes:

[0039] Combined with the analysis results of the severity of the patient's condition and nursing needs, the patient's condition is warned and divided into different warning levels, namely low warning level, medium warning level and high warning level;

[0040] According to different warning levels and combined with the disease evaluation index, the corresponding warning threshold is divided for each warning level, and the patients are divided into the corresponding warning level;

[0041] According to the warning level, a corresponding warning notice is made, including the patient's name, hospitalization number, warning level, main abnormal characteristics and recommended measures, and the warning notice is pushed to the corresponding medical staff through the hospital's internal information system.

[0042] Preferably, the plurality of warning levels correspond one-to-one to the plurality of warning thresholds, and the warning thresholds include an upper threshold and a lower threshold, specifically:

[0043] Low warning level ;

[0044] Medium warning level ;

[0045] High alert level ;

[0046] in, is the disease assessment index, is the upper threshold of the low warning level and the lower threshold of the medium warning level, is the upper threshold of the medium warning level and the lower threshold of the high warning level, , .

[0047] Preferably, in step 6, the process of taking intervention measures includes:

[0048] Medical staff confirm the received warning notification, read the content of the warning notification, including the patient's name, hospital number, warning level, main abnormal characteristics and recommended measures, and confirm the accuracy and urgency of the warning information;

[0049] Based on the information in the warning notification, medical staff will quickly assess the patient and confirm the patient's current condition and vital signs. If the warning level shows that the patient is at high risk, it is a high warning level, and medical staff will immediately take emergency intervention measures, including adjusting the treatment plan and performing rescue;

[0050] Based on the warning information and the patient’s current condition, medical staff adjust the treatment plan, including medication and nursing measures, and record the intervention measures and the patient’s response in the patient’s electronic medical record, updating the patient’s latest warning status;

[0051] Continuously monitor the patient's condition to identify and address any new abnormal features or changes in the condition. Evaluate the effectiveness of intervention measures based on the monitoring results. If some intervention measures are effective, continue to implement them as planned. If the effect is not good or the condition worsens, adjust the intervention plan.

[0052] The present invention provides a respiratory nursing condition analysis and early warning method based on big data. It has the following beneficial effects:

[0053] 1. This respiratory nursing condition analysis and early warning method based on big data monitors the patient's vital signs data in real time and conducts comprehensive analysis on it. It uses real-time, multi-dimensional data collection and analysis to greatly improve the accuracy and timeliness of condition assessment, so as to timely detect abnormal data fluctuations of patients and issue early warnings, so that medical staff can take intervention measures before the condition worsens, thereby effectively avoiding further development of the disease. The use of preventive health management not only improves the treatment effect, but also reduces medical costs.

[0054] 2. The respiratory nursing condition analysis and early warning method based on big data can monitor the patient's condition changes in real time by analyzing the patient's vital signs data such as respiratory rate, respiratory depth and heart rate. It can promptly detect the patient's abnormal condition and give early warning, thereby improving the accuracy and timeliness of condition monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a method flow chart of a respiratory nursing condition analysis and early warning method based on big data according to the present invention;

[0056] Figure 2 is a flow chart for analyzing abnormal characteristics of the present invention;

[0057] Figure 3 The present invention is a flow chart of the condition analysis of patients with respiratory diseases. DETAILED DESCRIPTION

[0058] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are provided for the purpose of illustration and description, and are not intended to be exhaustive or to limit the present invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention, and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

[0059] The first embodiment, as Figure 1 , Figure 2 As shown, the present invention provides a technical solution: a respiratory nursing condition analysis and early warning method based on big data, comprising the following steps:

[0060] Step 1: Collect multiple vital sign data of patients during the nursing process, and pre-process and integrate the collected data to form a complete data set, wherein the vital sign data includes respiratory rate, respiratory depth and heart rate, clarify the types of vital signs of patients with respiratory diseases to be collected, namely respiratory rate, respiratory depth and heart rate, collect various types of vital sign data of patients through medical equipment and sensors, wherein the respiratory rate is measured by respiratory monitoring equipment, the respiratory depth is obtained by monitoring the chest of the patient when breathing through a respiratory belt, and the heart rate is measured by using a heart rate monitoring equipment, and the collected various types of vital sign data are pre-processed, including the steps of data cleaning, data format conversion and data standardization, remove redundant, invalid or abnormal data through data cleaning, review or eliminate values ​​that obviously deviate from the normal range, convert data from different sources into a unified data format for subsequent analysis and processing, and standardize the data to eliminate the dimensional differences between data from different sources, integrate the pre-processed vital sign data to form a structured data set, and build a comprehensive health record of the patient by integrating the patient's identity information, vital sign data and measurement time information;

[0061] Step 2: extract features related to the patient's condition from the integrated data set, preset abnormal thresholds for different features, clarify abnormal features, obtain an abnormal feature sequence list, analyze the integrated data set, extract respiratory frequency features, respiratory depth features, and heart rate features related to the condition of patients with respiratory diseases, set corresponding abnormal thresholds for each feature according to historical data and nursing standards for respiratory diseases to distinguish between normal and abnormal conditions, compare each feature data in the data set with the corresponding abnormal threshold, identify feature values ​​that exceed the threshold, and mark the identified abnormal feature values, record the occurrence time, duration, and related patient information of the abnormal features for subsequent analysis and processing, review the marked abnormal features to ensure that the abnormal features reflect the patient's abnormal condition and exclude false alarms or interference, sort the abnormal features according to the severity of the abnormality, and obtain an abnormal feature sequence list, which includes the name, abnormal value, abnormal degree, and pathological significance of each abnormal feature;

[0062] Furthermore, the process of setting the characteristic abnormality threshold is as follows:

[0063] For the respiratory rate feature, the normal respiratory rate of an adult in a quiet state is 12-20 times / minute. The lower limit of the abnormal threshold of respiratory rate is set to less than 10 times / minute (bradypnea), and the upper limit is greater than 24 times / minute (tachypnea). For the respiratory depth feature, the respiratory depth of a normal adult in calm breathing is about 4-6 cm. The lower limit of the abnormal threshold is set to less than 3 cm (shallow breathing), and the upper limit is greater than 8 cm (deep breathing). For the heart rate feature, the heart rate of a normal adult in a quiet state is 60-100 times / minute. The lower limit of the abnormal threshold is set to less than 50 times / minute (bradycardia), and the upper limit is greater than 110 times / minute (tachycardia).

[0064] Step 3: Analyze the feature data of the abnormal feature sequence table, determine the frequency of occurrence of each abnormal feature, and mark the abnormal features in the intersection area. Extract the data of each abnormal feature from the abnormal feature sequence table, including the name of the abnormal feature and the corresponding record, and use the data analysis tool to count the frequency of each abnormal feature, calculate the number of times each abnormal feature appears, and use a table to display the frequency statistics results for intuitive comparison and analysis. Analyze the frequency of occurrence of each abnormal feature, set the abnormal feature judgment threshold, identify high-frequency abnormal features, analyze the number of times two or more abnormal features appear at the same time, determine the intersection area where the abnormal features exist, and compare the data in the abnormal feature sequence table with the abnormal feature existence. The intersection area is compared, the abnormal features of the identified intersection area are marked, the name, abnormal value and frequency of occurrence of the intersection abnormal features are recorded, the correlation and mutual influence between the abnormal features of the intersection area are analyzed, the frequency of occurrence of each abnormal feature and the analysis results of the abnormal features of the intersection area are further analyzed, and the high-frequency abnormal features and the intersection abnormal features are clarified, among which the high-frequency abnormal features include high-frequency respiratory frequency abnormal features, high-frequency respiratory depth abnormal features and high-frequency heart rate abnormal features, and the intersection abnormal features include respiratory frequency-respiratory depth abnormal features, respiratory frequency-heart rate abnormal features, respiratory depth-heart rate abnormal features and respiratory frequency-respiratory depth-heart rate abnormal features;

[0065] Step 4: Based on the feature analysis results, analyze the condition of the patient with respiratory disease and determine the severity of the patient's condition;

[0066] Step 5: Based on the analysis results of the severity of the patient's condition, different warning levels are preset, and warning notifications are pushed to medical staff based on the patient's condition;

[0067] Step 6: Based on the pushed warning notification, medical staff take corresponding intervention measures.

[0068] The second embodiment is based on the first embodiment. Figure 3 As shown, in step 4, the process of analyzing the condition of patients with respiratory diseases includes:

[0069] Based on the divided high-frequency abnormal features and intersection abnormal features, scoring standards are set for high-frequency abnormal features and intersection abnormal features, and the score of high-frequency abnormal features is set to 1 point, the score of intersection abnormal features with two abnormal features is set to 2 points, and the score of intersection abnormal features with three abnormal features is set to 3 points (high-frequency abnormal features include high-frequency respiratory frequency abnormal features, high-frequency respiratory depth abnormal features, and high-frequency heart rate abnormal features, and intersection abnormal features include respiratory frequency-respiratory depth abnormal features, respiratory frequency-heart rate abnormal features, respiratory depth-heart rate abnormal features, and respiratory frequency-respiratory depth-heart rate abnormal features). According to the actual situation of the patient, the score of the high-frequency abnormal features and the intersection abnormal features is calculated according to the scoring standards. According to the needs of the condition analysis of patients with respiratory diseases, the corresponding weight coefficient is set for each high-frequency abnormal feature and the intersection abnormal feature. Combined with the score and weight coefficient of each abnormal feature, the condition evaluation index is comprehensively calculated to analyze the severity of the patient's condition;

[0070] Furthermore, the calculation expression of the disease evaluation index is:

[0071] ;

[0072] ;

[0073] in, is the disease assessment index, The value range is between 0 and 1. is the sum of the weighted scores of all abnormal features, is the weight coefficient of the ith abnormal feature, is the score of the ith abnormal feature, is the number of abnormal features, that is, the sum of high-frequency abnormal features and intersection abnormal features, , The baseline value for all abnormal feature scores is the average value calculated based on historical data. is the standard deviation, which is used to adjust the sensitivity of the index and reflects the degree of variation of the abnormal feature score. near hour, Close to 0.5, indicating moderate condition. Much greater than hour, Tends to 1, indicating that the condition is serious. Much smaller than hour, tends to 0, indicating a milder condition;

[0074] In step 5, the process of setting the warning level includes:

[0075] Combined with the analysis results of the severity of the patient's condition and nursing needs, the patient's condition status is warned and divided into different warning levels, namely low warning level, medium warning level and high warning level. According to different warning levels, combined with the condition evaluation index, the corresponding warning threshold is divided for each warning level, and the patient is divided into the corresponding warning level. According to the warning level, the corresponding warning notice is made, including the patient's name, hospitalization number, warning level, main abnormal characteristics and recommended measures, and the warning notice is pushed to the corresponding medical staff through the hospital's internal information system;

[0076] Furthermore, multiple warning levels correspond to multiple warning thresholds one by one, and the warning thresholds include an upper threshold and a lower threshold, specifically:

[0077] Low warning level ;

[0078] Medium warning level ;

[0079] High alert level ;

[0080] in, is the disease assessment index, is the upper threshold of the low warning level and the lower threshold of the medium warning level, is the upper threshold of the medium warning level and the lower threshold of the high warning level, , ;

[0081] In step six, the process of implementing intervention measures includes:

[0082] Medical staff confirm the received warning notification, read the content of the warning notification, including the patient's name, hospitalization number, warning level, main abnormal characteristics and recommended measures, confirm the accuracy and urgency of the warning information, and quickly evaluate the patient based on the information in the warning notification to confirm the patient's current condition and vital signs. If the warning level shows that the patient is at high risk, it is a high warning level, and the medical staff immediately takes emergency intervention measures, including adjusting the treatment plan and performing rescue. Based on the warning information and the patient's current condition, the medical staff adjusts the treatment plan, including drug therapy and nursing measures, and records the intervention measures and the patient's response in the patient's electronic medical record, updates the patient's latest warning status, continuously monitors the patient's condition, discovers and handles any new abnormal characteristics or changes in the condition, and evaluates the effectiveness of the intervention measures based on the monitoring results. If some of the preventive measures are effective, continue to implement them as planned. If the effect is not good or the condition worsens, adjust the intervention plan.

[0083] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention are implemented according to the conventional means in the field unless otherwise specified and limited.

Claims

1. A respiratory nursing condition analysis and early warning method based on big data, characterized in that: The following steps are involved: Step 1: Collect multiple vital signs data of patients during the nursing process, and pre-process and integrate the collected data to form a complete data set; Step 2: extract features related to the patient's condition from the integrated data set, preset abnormal thresholds for different features, identify abnormal features, and obtain an abnormal feature sequence list; Step 3: Analyze the feature data of the abnormal feature sequence table, determine the frequency of occurrence of each abnormal feature, and mark the abnormal features in the intersection area; Step 4: Based on the feature analysis results, analyze the condition of the patient with respiratory disease and determine the severity of the patient's condition; Step 5: Based on the analysis results of the severity of the patient's condition, different warning levels are preset, and warning notifications are pushed to medical staff based on the patient's condition; Step 6: Based on the pushed warning notification, medical staff take corresponding intervention measures.

2. A respiratory nursing condition analysis and early warning method based on big data according to claim 1, characterized in that: In step 1, the process of collecting patient vital signs data includes: Identify the types of vital signs that need to be collected for patients with respiratory diseases, namely respiratory rate, respiratory depth, and heart rate; The patient's various vital signs data are collected through medical devices and sensors, including respiratory rate measured by a respiratory monitoring device, respiratory depth obtained by monitoring the chest of the patient during breathing with a respiratory belt, and heart rate measured by a heart rate monitoring device; Preprocess the collected various types of vital signs data, including data cleaning, data format conversion and data standardization; The pre-processed vital signs data are integrated to form a structured data set, which integrates the patient's identity information, vital signs data, and measurement time information to build a comprehensive health record for the patient.

3. A respiratory nursing condition analysis and early warning method based on big data according to claim 2, characterized in that: In the step 2, the process of obtaining the abnormal feature sequence list includes: Analyze the integrated data set to extract the respiratory frequency features, respiratory depth features, and heart rate features related to the condition of patients with respiratory diseases; Based on historical data and standards of care for respiratory diseases, set corresponding abnormal thresholds for each feature to distinguish between normal and abnormal conditions; Compare each feature data in the data set with the corresponding abnormal threshold, identify the feature value that exceeds the threshold, mark the identified abnormal feature value, and record the occurrence time, duration and related patient information of the abnormal feature; The marked abnormal features are reviewed to ensure that the abnormal features reflect the patient's abnormal condition and exclude false positives or interference. The abnormal features are sorted according to the severity of the abnormality to obtain an abnormal feature sequence list, which contains the name, abnormal value, abnormal degree and pathological significance of each abnormal feature.

4. The method for analyzing and warning respiratory disease based on big data according to claim 3 is characterized in that: The process of setting the characteristic abnormality threshold is specifically as follows: For respiratory rate characteristics, the normal respiratory rate of an adult in a quiet state is 12-20 times / minute. The lower limit of the abnormal respiratory rate threshold is set to less than 10 times / minute, and the upper limit is set to more than 24 times / minute. For the breathing depth feature, the breathing depth of a normal adult during calm breathing is about 4-6 cm. The lower limit of the abnormal threshold is set to be less than 3 cm, and the upper limit is set to be greater than 8 cm. For heart rate characteristics, the normal heart rate of an adult at rest is 60-100 beats / minute. The lower limit of the abnormal threshold is set to less than 50 beats / minute and the upper limit is set to greater than 110 beats / minute.

5. A respiratory nursing condition analysis and early warning method based on big data according to claim 4, characterized in that: In step 3, the analysis process of abnormal features includes: Extract the data of each abnormal feature from the abnormal feature sequence table, including the name of the abnormal feature and the corresponding record, and use the data analysis tool to perform frequency statistics on each abnormal feature to calculate the number of times each abnormal feature appears; Analyze the frequency of occurrence of each abnormal feature, set the abnormal feature judgment threshold, identify high-frequency abnormal features, analyze the number of times two or more abnormal features appear at the same time, and determine the intersection area where abnormal features exist; Compare the data in the abnormal feature sequence table with the intersection area where the abnormal features exist, mark the identified abnormal features with intersection areas, and record the name, abnormal value and frequency of occurrence of the intersection abnormal features; Analyze the correlation and mutual influence between the abnormal features in the intersection area, further analyze the frequency of occurrence of each abnormal feature and the analysis results of the abnormal features in the intersection area, and clarify the high-frequency abnormal features and intersection abnormal features. Among them, high-frequency abnormal features include high-frequency abnormal respiratory rate features, high-frequency abnormal respiratory depth features, and high-frequency abnormal heart rate features. Intersection abnormal features include abnormal respiratory rate-respiratory depth features, abnormal respiratory rate-heart rate features, abnormal respiratory depth-heart rate features, and abnormal respiratory rate-respiratory depth-heart rate features.

6. A respiratory nursing condition analysis and early warning method based on big data according to claim 5, characterized in that: In step 4, the process of analyzing the condition of patients with respiratory diseases includes: Based on the divided high-frequency abnormal features and intersection abnormal features, scoring standards are set for the high-frequency abnormal features and intersection abnormal features. The score of the high-frequency abnormal features is set to 1 point, the score of the intersection abnormal features with two abnormal features is set to 2 points, and the score of the intersection abnormal features with three abnormal features is set to 3 points. According to the actual situation of the patient, the scores of the high-frequency abnormal features and the intersection abnormal features are calculated according to the scoring criteria, and according to the needs of the condition analysis of the patient with respiratory diseases, the corresponding weight coefficients are set for each high-frequency abnormal feature and the intersection abnormal feature; The scores and weight coefficients of each abnormal feature are combined to calculate the disease evaluation index and analyze the severity of the patient's condition.

7. A respiratory nursing condition analysis and early warning method based on big data according to claim 6, characterized in that: The calculation expression of the disease evaluation index is: ； ； in, is the disease assessment index, is the sum of the weighted scores of all abnormal features, is the weight coefficient of the ith abnormal feature, is the score of the ith abnormal feature, is the number of abnormal features, that is, the sum of high-frequency abnormal features and intersection abnormal features, , is the baseline value for all abnormal feature scores, is the standard deviation, which is used to adjust the sensitivity of the index.

8. The method for analyzing and warning respiratory disease based on big data according to claim 7, characterized in that: In step 5, the process of setting the warning level includes: Combined with the analysis results of the severity of the patient's condition and nursing needs, the patient's condition is warned and divided into different warning levels, namely low warning level, medium warning level and high warning level; According to different warning levels and combined with the disease evaluation index, the corresponding warning threshold is divided for each warning level, and the patients are divided into the corresponding warning level; According to the warning level, a corresponding warning notice is made, including the patient's name, hospitalization number, warning level, main abnormal characteristics and recommended measures, and the warning notice is pushed to the corresponding medical staff through the hospital's internal information system.

9. A respiratory nursing condition analysis and early warning method based on big data according to claim 8, characterized in that: The plurality of warning levels correspond one-to-one to the plurality of warning thresholds, and the warning thresholds include an upper threshold and a lower threshold, specifically: Low warning level ; Medium warning level ; High alert level ; in, is the disease assessment index, is the upper threshold of the low warning level and the lower threshold of the medium warning level, is the upper threshold of the medium warning level and the lower threshold of the high warning level, , .

10. The method for analyzing and warning respiratory disease conditions based on big data according to claim 9, characterized in that: In step 6, the process of taking intervention measures includes: Medical staff confirm the received warning notification, read the content of the warning notification, including the patient's name, hospital number, warning level, main abnormal characteristics and recommended measures, and confirm the accuracy and urgency of the warning information; Based on the information in the warning notification, medical staff will quickly assess the patient and confirm the patient's current condition and vital signs. If the warning level shows that the patient is at high risk, it is a high warning level, and medical staff will immediately take emergency intervention measures, including adjusting the treatment plan and performing rescue; Based on the warning information and the patient’s current condition, medical staff adjust the treatment plan, including medication and nursing measures, and record the intervention measures and the patient’s response in the patient’s electronic medical record, updating the patient’s latest warning status; Continuously monitor the patient's condition to identify and address any new abnormal features or changes in the condition. Evaluate the effectiveness of intervention measures based on the monitoring results. If some intervention measures are effective, continue to implement them as planned. If the effect is not good or the condition worsens, adjust the intervention plan.

Images