A method for analyzing safety hazards in clinical nursing of vascular surgery

By collecting and analyzing heart rate, blood pressure and blood circulation data in real time, dynamically adjusting risk levels and identifying vascular problem patterns, the problem of inability to timely identify vascular abnormalities and optimize treatment plans in the existing technology is solved, and the accuracy of diagnosis and treatment is improved. Timeliness.

CN118737474BActive Publication Date: 2024-11-05SOUTHWEST MEDICAL UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411226654.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-03
Publication Date
2024-11-05
Estimated Expiration
2044-09-03

AI Technical Summary

Technical Problem

The existing medical data evaluation technology is insufficient in real-time data monitoring and dynamic response, resulting in the inability to timely identify vascular abnormalities, affecting the diagnosis and treatment effect, and failing to make full use of patient history and real-time data to optimize treatment plans.

Method used

By collecting real-time heart rate, blood pressure and blood circulation data of patients, comparatively analyze with historical health baselines, identify abnormal data points, dynamically adjust risk levels, identify patterns associated with vascular problems, generate key health risk patterns, and optimize treatment plans.

Benefits of technology

It improves the early recognition ability of abnormal data points, enhances the accuracy of diagnosis and the accuracy of predicting future health events, optimizes treatment plans, and improves the timeliness and personalization of medical interventions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118737474B_ABST
    Figure CN118737474B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of medical data evaluation technology, and specifically to a method for analyzing safety hazards in clinical nursing of vascular surgery, comprising the following steps: collecting the patient's real-time heart rate, blood pressure and blood circulation data, identifying abnormal data points associated with arteriosclerosis symptoms by comparing and analyzing the data of heart rate not 40 to 180 beats / minute and blood pressure not 90 / 60 to 140 / 90 mmHg with the historical health baseline, and generating initial screening abnormal data points. In the present invention, by collecting heart rate, blood pressure and blood circulation data in real time, an instant comparison with the historical health baseline is achieved, and the early identification capability of abnormal data points is improved. By continuously analyzing the abnormal frequency and severity results, the data is dynamically adjusted to refine the risk level. In addition, by comparing and analyzing with known varicose veins and other vascular lesion patterns, the understanding of the disease pattern is deepened, the diagnosis is made more accurate, and the accuracy of predicting future health events is enhanced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of medical data evaluation, and in particular to a method for analyzing safety hazards in clinical nursing of vascular surgery. Background Art

[0002] The field of medical data assessment technology involves the use of information technology to analyze, process, and evaluate medical or health-related data. The purpose of this technology field is to improve the quality and efficiency of medical services, ensure patient safety, and optimize health management and treatment strategies through data analysis. Specific applications include but are not limited to disease pattern recognition, treatment effect evaluation, medical risk management, and predicting health trends. Technology in this field can also assist medical professionals in decision support and provide personalized medical advice and treatment plans through in-depth analysis of patient history and real-time data.

[0003] Among them, the vascular surgery clinical nursing safety hazard analysis method refers to the use of medical data evaluation technology to identify and analyze potential safety hazards in the vascular surgery nursing process. The purpose of this method is to timely discover and solve problems that may affect patient safety through in-depth analysis of the nursing process, patient conditions, and treatment methods, thereby improving nursing quality and patient satisfaction. In addition, this method also helps hospital management to develop more effective risk management strategies and preventive measures.

[0004] Although existing medical data evaluation technologies emphasize data analysis and processing, they are often insufficient in real-time data monitoring and dynamic response, resulting in an inability to effectively respond to rapidly changing patient conditions. This technology often lacks sufficient depth and adaptability when comprehensively applying data to predict disease development and deal with complex pathological patterns, which may affect the timeliness and accuracy of diagnosis. For example, for early vascular abnormalities, existing technologies fail to identify them in a timely manner, thus missing the best time for treatment. In addition, existing technologies have limitations in using data to provide personalized medical advice, and fail to fully utilize patients' historical and real-time data to optimize treatment plans, affecting the effectiveness of treatment strategies and patients' ultimate health outcomes. These technical defects lead to inaccurate assessments of patient status, which may increase medical risks and treatment costs. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a method for analyzing safety hazards in clinical nursing of vascular surgery.

[0006] In order to achieve the above object, the present invention adopts the following technical scheme: a method for analyzing safety hazards in clinical nursing of vascular surgery, comprising the following steps:

[0007] S1: Collect the patient's real-time heart rate, blood pressure and blood circulation data, identify abnormal data points associated with arteriosclerosis symptoms by comparing and analyzing the data of heart rate not 40 to 180 beats / minute and blood pressure not 90 / 60 to 140 / 90 mmHg with the historical health baseline, and generate initial screening abnormal data points;

[0008] S2: Analyze the abnormal frequency and severity results of the abnormal data points in the initial screening, adjust the risk level according to the data change trend and its fit with historical cases, reclassify the data, and obtain the dynamically adjusted risk level;

[0009] S3: analyzing the risk data points in the dynamically adjusted risk level, comparing the risk data points with risk features associated with known varicose veins, identifying patterns associated with vascular problems, and establishing risk pattern identification results;

[0010] S4: Analyze the patterns identified in the risk pattern identification results, compare the patterns with known vascular lesion patterns, identify key patterns that lead to future health events, and generate key health risk patterns.

[0011] As a further solution of the present invention, the step of obtaining the initial screening abnormal data points is specifically as follows:

[0012] S111: Collect the patient's real-time heart rate and blood pressure, perform range filtering on the data points, and mark whether the heart rate and blood pressure are within the abnormal range, using the formula:

[0013] ,

[0014] and

[0015] ,

[0016] Generate heart rate and blood pressure markers;

[0017] in, Adjust the sensitivity of heart rate response, Modulates systolic blood pressure response sensitivity, Regulates diastolic blood pressure response sensitivity, represents the heart rate at time t, represents systolic blood pressure, stands for diastolic blood pressure;

[0018] S112: The heart rate mark and the blood pressure mark are combined to calculate a comprehensive abnormal mark using the formula:

[0019] ,

[0020] Identify both abnormal heart rate and blood pressure data points and generate a comprehensive abnormality tag ;

[0021] in, It's the heart rate marker. It is a blood pressure marker;

[0022] S113: For the data points marked as abnormal in the comprehensive abnormal marking, that is, The data points are compared with the historical health baseline using the formula:

[0023] ,

[0024] Calculate the difference between the abnormal data point and the historical baseline to obtain the initial screening abnormal data point ;

[0025] in, represents the historical baseline heart rate, represents the historical baseline blood pressure, represents the weight factor for adjusting heart rate differences, represents the weighting factor for adjusting blood pressure differences.

[0026] As a further solution of the present invention, the step of obtaining the dynamically adjusted risk level is specifically as follows:

[0027] S211: Extract abnormal frequency results from the initial screening abnormal data points using the formula:

[0028] ,

[0029] Generate abnormal frequency results;

[0030] in, represents abnormal frequency results, represents the abnormality determination threshold, and N represents the total number of data points;

[0031] S212: For the abnormal data points in the initial screening, the severity is calculated using the formula:

[0032] ,

[0033] Generate severity;

[0034] in, is the severity, yes The average value of yes The standard deviation of

[0035] S213: Dynamically adjust the risk level according to the abnormal frequency result, severity and fit with historical data, using the formula:

[0036] ,

[0037] Get the dynamically adjusted risk level;

[0038] in, Represents the dynamically adjusted risk level, , is the adjustment factor.

[0039] As a further solution of the present invention, the step of obtaining the risk characteristics is specifically as follows:

[0040] S311: Calculate the difference between the dynamically adjusted risk level and the historical average risk level using the formula:

[0041] ,

[0042] Generate risk level difference results;

[0043] in, Represents the difference in risk level, Represents the dynamically adjusted risk level, represents the risk level at historical time point i, represents the weight of time point i, and N represents the total number of data points;

[0044] S312: Using the risk level difference result to detect the risk peak, the formula is:

[0045] ,

[0046] Generate risk peak results;

[0047] in, represents the risk peak, k represents the time window, represents the attenuation coefficient;

[0048] S313: Based on the risk peak result, the formula is used:

[0049] ,

[0050] Generate risk profiles;

[0051] in, Represents risk characteristics, represents the response strength, Represents the decision threshold.

[0052] As a further solution of the present invention, the steps of obtaining the risk pattern identification result are specifically as follows:

[0053] S321: Based on the risk characteristics, analyze the association with known vascular problems using the formula:

[0054] ,

[0055] Generate correlation analysis results;

[0056] in, Represents the results of correlation analysis. Represents risk characteristics, Represents the risk profile of varicose veins, represents the relevance weight, and M represents the number of features;

[0057] S322: Evaluate the correlation analysis result using the formula:

[0058] ,

[0059] Determine whether the data point is consistent with a known pattern and generate a pattern evaluation result;

[0060] in, Representative model evaluation results, represents the sensitivity adjustment factor, represents the recognition threshold;

[0061] S323: Based on the pattern evaluation results, identify the risk pattern associated with the vascular problem, using the formula:

[0062] ,

[0063] Generate risk pattern identification results;

[0064] in, Represents the risk pattern identification result, represents the weight of time point i, and n represents the reference time range.

[0065] As a further solution of the present invention, the steps of obtaining the key health risk model are specifically as follows:

[0066] S411: Analyze the risk pattern identification result and the known vascular lesion pattern library, using the formula:

[0067] ,

[0068] Calculate the difference score of each mode and generate the mode difference score ;

[0069] in, represents the feature weight, represents the jth pattern value in the known pattern library, represents a small constant to avoid zero denominator;

[0070] S412: Identify key patterns from the pattern difference scores using the formula:

[0071] ,

[0072] Calculate the criticality of each pattern and generate key pattern recognition results ;

[0073] in, represents the threshold value, represents standard deviation, exp represents natural exponential function;

[0074] S413: Based on the key pattern recognition results, extract the key health risk pattern using the formula:

[0075] ,

[0076] Generate key health risk models ;

[0077] Where n represents the reference time range.

[0078] Compared with the prior art, the advantages and positive effects of the present invention are:

[0079] In the present invention, by collecting heart rate, blood pressure and blood circulation data in real time, an instant comparison with the historical health baseline is achieved, and the early identification capability of abnormal data points is improved. Through continuous analysis of abnormal frequency and severity results, data is dynamically adjusted to refine the risk level. In addition, by comparing and analyzing with known varicose veins and other vascular lesion patterns, the understanding of disease patterns is deepened, the diagnosis is more accurate, and the accuracy of predicting future health events is enhanced. It not only strengthens the identification of potential serious vascular problems, but also optimizes the treatment plan for patients, and improves the timeliness and personalization of medical intervention. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 It is a schematic diagram of the main steps of the present invention;

[0081] Figure 2 This is a flow chart of the steps for obtaining the initial screening abnormal data points of the present invention;

[0082] Figure 3 A flowchart of the steps for obtaining the dynamically adjusted risk level of the present invention;

[0083] Figure 4 A flow chart of the steps for obtaining risk characteristics of the present invention;

[0084] Figure 5 A flow chart of the steps for obtaining the risk model identification results of the present invention;

[0085] Figure 6 This is a flow chart of the steps for obtaining the key health risk model of the present invention. DETAILED DESCRIPTION

[0086] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0087] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0088] Embodiment 1:

[0089] See also Figure 1 The present invention provides a technical solution: a method for analyzing safety hazards in clinical nursing of vascular surgery, comprising the following steps:

[0090] S1: Collect the patient's real-time heart rate, blood pressure and blood circulation data, identify abnormal data points associated with arteriosclerosis symptoms by comparing and analyzing the data of heart rate not 40 to 180 beats / minute and blood pressure not 90 / 60 to 140 / 90 mmHg with the historical health baseline, and generate initial screening abnormal data points;

[0091] S2: Analyze the abnormal frequency and severity results of the initial screening abnormal data points, adjust the risk level according to the data change trend and its fit with historical cases, reclassify the data, and obtain the dynamically adjusted risk level;

[0092] S3: Analyze the risk data points in the dynamically adjusted risk level, compare the risk data points with the risk characteristics associated with known varicose veins, identify the patterns associated with vascular problems, and establish risk pattern identification results;

[0093] S4: Analyze the patterns identified in the risk pattern identification results, compare the patterns with known vascular lesion patterns, identify key patterns that lead to future health events, and generate key health risk patterns.

[0094] The initial screening abnormal data points include abnormal heart rate data points, abnormal blood pressure data points, and abnormal blood circulation data points. The dynamically adjusted risk level includes abnormal frequency, severity results, data change trends, and historical case fitting. The risk pattern identification results include arteriosclerosis risk pattern, varicose vein risk pattern, and vascular problem risk pattern. The key health risk patterns include arteriosclerosis key pattern, varicose vein key pattern, and vascular disease key pattern.

[0095] See also Figure 2 , the specific steps for obtaining the initial screening abnormal data points are:

[0096] S111: Collect the patient's real-time heart rate and blood pressure, perform range filtering on the data points, and mark whether the heart rate and blood pressure are within the abnormal range, using the formula:

[0097] ,

[0098] and

[0099] ,

[0100] Generate heart rate and blood pressure markers;

[0101] in, Adjust the sensitivity of heart rate response, Modulates systolic blood pressure response sensitivity, Regulates diastolic blood pressure response sensitivity, represents the heart rate at time t, represents systolic blood pressure, stands for diastolic blood pressure;

[0102] S112: Combine the heart rate mark and the blood pressure mark to calculate the comprehensive abnormal mark using the formula:

[0103] ,

[0104] Identify both abnormal heart rate and blood pressure data points and generate a comprehensive abnormality tag ;

[0105] in, It's the heart rate marker. It is a blood pressure marker;

[0106] S113: For the data points marked as abnormal in the comprehensive abnormal marking, that is, The data points are compared with the historical health baseline using the formula:

[0107] ,

[0108] Calculate the difference between the abnormal data point and the historical baseline to obtain the initial screening abnormal data point ;

[0109] in, represents the historical baseline heart rate, represents the historical baseline blood pressure, represents the weight factor for adjusting heart rate differences, represents the weighting factor for adjusting blood pressure differences.

[0110] formula:

[0111] ,

[0112] and

[0113] ,

[0114] Parameter Description:

[0115] , , is the sensitivity adjustment parameter, which is used to control the response speed of the function to changes in heart rate and blood pressure;

[0116] is the heart rate value at the current time point t;

[0117] is the systolic blood pressure at the current time point t;

[0118] is the diastolic pressure at the current time point t;

[0119] Calculation example:

[0120] Assume that at time t: , , ,

[0121] assumed , , (Set to a value less than 1 to ensure smooth response of the logic function)

[0122] Calculation process:

[0123] Heart rate markers:

[0124] ,

[0125] The results indicated that a heart rate of 130 was within the normal range with a high level of logical activation.

[0126] Blood pressure markers:

[0127] ,

[0128] The results showed that the logic activation level of the current blood pressure value within the normal range was 0.38.

[0129] formula: ,

[0130] Calculation example:

[0131] Combining the previous results:

[0132] ,

[0133] Calculation process: ,

[0134] The value shows that the combined abnormal index of heart rate and blood pressure is 0.33, indicating that the time point data has a certain probability of abnormality, but it is not highly abnormal.

[0135] formula: ,

[0136] Parameter Description:

[0137] is the historical baseline heart rate, assumed to be 100 bpm;

[0138] : is the historical baseline blood pressure, assumed to be 110 / 70 mmHg;

[0139] , are weight factors, assumed to be 1.

[0140] Calculation example:

[0141] Continue to assume , , ;

[0142] Assuming a historical baseline , ,

[0143] Calculation process:

[0144] ,

[0145] The resulting value of 33.17 indicates that the current data point deviates significantly from the historical baseline, indicating a clear health issue.

[0146] See also Figure 3 , the specific steps for obtaining the dynamically adjusted risk level are:

[0147] S211: Extract abnormal frequency results from the initial screening abnormal data points using the formula:

[0148] ,

[0149] Generate abnormal frequency results;

[0150] in, represents abnormal frequency results, represents the abnormality determination threshold, and N represents the total number of data points;

[0151] S212: For the initial screening abnormal data points, the severity results are calculated using the formula:

[0152] ,

[0153] Generate severity;

[0154] in, is the severity, yes The average value of yes The standard deviation of

[0155] S213: Dynamically adjust the risk level according to the abnormal frequency result, severity and fit with historical data, using the formula:

[0156] ,

[0157] Get the dynamically adjusted risk level;

[0158] in, Represents the dynamically adjusted risk level, , is the adjustment factor.

[0159] Formula analysis and derivation:

[0160] ,

[0161] : The outlier value of the i-th data point;

[0162] : The abnormality judgment threshold is set to a hypothetical mean value plus the standard deviation, such as ,

[0163] N: the total number of data points, for example, 1000 data points;

[0164] : An indicator function, if If true, the value is 1, otherwise it is 0;

[0165] Calculation example:

[0166] Suppose that among the 1000 data points, 50 points Value greater than threshold If you set , and assuming and , then:

[0167] ,

[0168] This result Represents the frequency of monitored anomalies relative to the highest frequency on a logarithmic scale.

[0169] Formula analysis and derivation:

[0170] ,

[0171] : The average value of is assumed to be 0.5,

[0172] : The standard deviation of , assuming it is 0.5,

[0173] : absolute deviation value;

[0174] In the denominator Used for normalization so that differentiated data points can be compared on the same scale

[0175] Calculation example:

[0176] Suppose there are 5 abnormal data points The values ​​are , the calculation steps are as follows:

[0177] ,

[0178] ,

[0179] result The calculated severity result is 3.09 on the standardized scale, which indicates that the anomaly is significant.

[0180] Formula analysis and derivation:

[0181] ,

[0182] , : Adjustment factor, assuming =0.6 and =0.4 to give Greater attention;

[0183] Calculation example:

[0184] Based on the above results, if and ,but:

[0185] ,

[0186] result Indicates the risk level after referring to the abnormal frequency and severity results. The higher the value, the higher the risk level.

[0187] See also Figure 4 , the steps to obtain risk characteristics are as follows:

[0188] S311: Calculate the difference between the dynamically adjusted risk level and the historical average risk level using the formula:

[0189] ,

[0190] Generate risk level difference results;

[0191] in, Represents the difference in risk level, Represents the dynamically adjusted risk level, represents the risk level at historical time point i, represents the weight of time point i, and N represents the total number of data points;

[0192] S312: Use the risk level difference results to detect the risk peak, using the formula:

[0193] ,

[0194] Generate risk peak results;

[0195] in, represents the risk peak, k represents the time window, represents the attenuation coefficient;

[0196] S313: Based on the risk peak result, the formula is used:

[0197] ,

[0198] Generate risk profiles;

[0199] in, Represents risk characteristics, represents the response strength, Represents the decision threshold.

[0200] formula:

[0201] ,

[0202] Detailed explanation and examples:

[0203] Assume that there are the following risk level data points R at time points 1 to 5, where At time t=3 it is 7.0:

[0204] ,

[0205] Assumption Weight All are set to 1 to simplify calculations.

[0206] Calculation process:

[0207] Calculate the average risk level difference:

[0208] ,

[0209] Actual calculation: This means that the risk level at time point t=3 differs by an average of 1.25 compared to other time points.

[0210] formula:

[0211] ,

[0212] Detailed explanation and examples:

[0213] Assume k=1, = 0.5, and use the previous .

[0214] Calculation process:

[0215] Calculate the peak value using an exponential decay factor:

[0216] ,

[0217] Assumptions and ,but:

[0218] ,

[0219] This means that at time point t=3, the peak value of the risk level is adjusted to 0.821.

[0220] formula:

[0221] ,

[0222] Detailed explanation and examples:

[0223] Assumptions =2, = 0.8, use .

[0224] Calculation process:

[0225] Calculate risk characteristics: ,

[0226] Actual calculation: ,

[0227] This means that at time point t=3, the risk eigenvalue is approximately 0.51, which is an example of a medium risk level.

[0228] See also Figure 5 , the specific steps for obtaining the risk pattern identification results are:

[0229] S321: Based on the risk profile, analyze the association with known vascular problems using the formula:

[0230] ,

[0231] Generate correlation analysis results;

[0232] in, Represents the results of correlation analysis. Represents risk characteristics, Represents the risk profile of varicose veins, represents the relevance weight, and M represents the number of features;

[0233] S322: Evaluate the results of the correlation analysis using the formula:

[0234] ,

[0235] Determine whether the data point is consistent with a known pattern and generate a pattern evaluation result;

[0236] in, Representative model evaluation results, represents the sensitivity adjustment factor, represents the recognition threshold;

[0237] S323: Based on the pattern assessment results, identify the risk pattern associated with vascular problems using the formula:

[0238] ,

[0239] Generate risk pattern identification results;

[0240] in, Represents the risk pattern identification result, represents the weight of time point i, and n represents the reference time range.

[0241] Detailed explanation and derivation of the formula:

[0242] ,

[0243] Calculation process:

[0244] Determine risk characteristics : set to 0.8 (assuming normalized value obtained through previous risk assessment);

[0245] Known vascular problem features : Set three features to 0.5, 0.7, and 0.9 respectively;

[0246] Weight setting : For simplicity, the weight of each feature is set to 1.

[0247] calculate:

[0248] ,

[0249] Analysis of calculation results:

[0250] Calculation results The differences between the risk-representing signature and the signature of known vascular problems were smaller, indicating a higher association.

[0251] Detailed explanation and derivation of the formula:

[0252] ,

[0253] Calculation process:

[0254] Correlation analysis results : 0.11;

[0255] Sensitivity adjustment factor : Set to 10 to increase the model's sensitivity to small changes;

[0256] Identification threshold : It is set to 0.1, that is, the risk is considered highly correlated if the difference between it and the known problem characteristics is no more than 0.1.

[0257] calculate:

[0258] ,

[0259] Analysis of calculation results:

[0260] What you get Indicates that there is a moderate degree of pattern matching, that is, the risk data points have some degree of correlation, but not very strong.

[0261] Detailed explanation and derivation of the formula:

[0262] ,

[0263] Model evaluation results : There are two evaluation results before and after, assuming they are 0.4, 0.5 (current moment), 0.6, 0.7;

[0264] Time range n: set to 2, indicating that the reference is made to the two time points before and after;

[0265] Weight Factor : are all set to 1.

[0266] calculate:

[0267] ,

[0268] Analysis of calculation results:

[0269] What you get It indicates a strong risk pattern identification, indicating that there is an obvious risk pattern near the time point.

[0270] See also Figure 6 , the specific steps for obtaining key health risk models are:

[0271] S411: Analyze the risk pattern identification result and the known vascular lesion pattern library, using the formula:

[0272] ,

[0273] Calculate the difference score of each mode and generate the mode difference score ;

[0274] in, represents the feature weight, represents the jth pattern value in the known pattern library, represents a small constant to avoid zero denominator;

[0275] S412: Identify key patterns from the pattern difference scores using the formula:

[0276] ,

[0277] Calculate the criticality of each pattern and generate key pattern recognition results ;

[0278] in, represents the threshold value, represents standard deviation, exp represents natural exponential function;

[0279] S413: Based on the key pattern recognition results, extract the key health risk pattern using the formula:

[0280] ,

[0281] Generate key health risk models ;

[0282] Where n represents the reference time range.

[0283] Detailed explanation of formulas and parameters:

[0284] ,

[0285] in:

[0286] : is the risk pattern identified at time t;

[0287] : is the jth pattern value in the known pattern library;

[0288] : is the weight of the jth pattern, indicating the importance of the pattern in the overall evaluation;

[0289] : is a small constant used to ensure that the denominator is not zero and is a very small positive number, such as 0.01.

[0290] Assume the following example:

[0291] Assume there are three known mode values, which are , and the corresponding weight , assuming that the risk pattern identified at time t ,as well as .

[0292] Calculation process:

[0293] Calculate each mode and The differences:

[0294] ,

[0295] Apply weights and calculate :

[0296] ,

[0297] Calculations show that the total difference between the risk pattern and the known pattern at time t is , the smaller the value, the closer the recognized pattern is to the known pattern.

[0298] Results and significance:

[0299] Numeric Represents the overall difference between the identified risk pattern and the known lesion pattern. In practical applications, the numerical value can be used to evaluate the abnormality of the newly identified pattern or its matching degree with the known lesion pattern.

[0300] Detailed explanation of formulas and parameters:

[0301] ,

[0302] in: is the pattern difference score; is the threshold, which determines the center point of the goodness-of-fit calculation; is the standard deviation, regulating the sensitivity and response width of the function.

[0303] Assume the following example:

[0304] Continue using the above results , set the threshold = 0.02 and standard deviation =0.01.

[0305] Calculation process:

[0306] Calculate the probability of a pattern match:

[0307] ,

[0308] This indicates that the probability that the risk pattern identified at time t matches the known pattern is about 49.18%, which is a relatively neutral result, indicating that the pattern is neither very close to nor completely mismatches the known pattern.

[0309] Detailed explanation of formulas and parameters:

[0310] ,

[0311] in: is the matching probability at time t+i; is an exponentially decaying weight that reduces the influence of time points far away from t.

[0312] Assume the following example:

[0313] Set n=2, that is, refer to the data in the time range of t-2 to t+2. Assume , , , , .

[0314] Calculation process:

[0315] Calculate the weighted average:

[0316] ,

[0317] ,

[0318] ,

[0319] The results show that the key health risk pattern at time t has a matching probability close to 49%, which is a medium risk level.

[0320] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for analyzing safety hazards in clinical nursing of vascular surgery, characterized in that: The following steps are involved: Collect the patient's real-time heart rate, blood pressure and blood circulation data, and identify abnormal data points associated with arteriosclerosis symptoms by comparing and analyzing the data of heart rate other than 40 to 180 beats / minute and blood pressure other than 90 / 60 to 140 / 90 mmHg with the historical health baseline, and generate initial screening abnormal data points; Analyze the abnormal frequency and severity results of the initial screening abnormal data points, adjust the risk level according to the data change trend and its fit with historical cases, reclassify the data, and obtain a dynamically adjusted risk level; The steps for obtaining the dynamically adjusted risk level are specifically as follows: The abnormal frequency results are extracted from the initial screening abnormal data points using the formula: , Generate abnormal frequency results; in, represents abnormal frequency results, represents the abnormality determination threshold, and N represents the total number of data points; For the abnormal data points in the initial screening, the severity is calculated using the formula: , Generate severity; in, is the severity, yes The average value of yes The standard deviation of The risk level is dynamically adjusted according to the abnormal frequency results, severity and fit with historical data, using the formula: , Get the dynamically adjusted risk level; in, Represents the dynamically adjusted risk level, , is the adjustment factor; Analyzing the risk data points in the dynamically adjusted risk level, comparing the risk data points with risk features associated with known varicose veins, identifying patterns associated with vascular problems, and establishing risk pattern identification results; The steps for obtaining the risk characteristics are specifically as follows: The difference between the dynamically adjusted risk level and the historical average risk level is calculated using the formula: , Generate risk level difference results; in, Represents the difference in risk level, Represents the dynamically adjusted risk level, represents the risk level at historical time point i, represents the weight of time point i, and N represents the total number of data points; The risk level difference results are used to detect the risk peak using the formula: , Generate risk peak results; in, represents the risk peak, k represents the time window, represents the attenuation coefficient; Based on the risk peak results, the formula is used: , Generate risk profiles; in, Represents risk characteristics, represents the response strength, represents the judgment threshold; Analyze the patterns identified in the risk pattern identification results, compare the patterns with known vascular lesion patterns, identify key patterns that lead to future health events, and generate key health risk patterns.

2. The method for analyzing safety hazards in clinical nursing of vascular surgery according to claim 1 is characterized in that: The steps for obtaining the abnormal data points in the initial screening are specifically as follows: Collect the patient's real-time heart rate and blood pressure, filter the data points, and mark whether the heart rate and blood pressure are in the abnormal range. Use the formula: , and , Generate heart rate and blood pressure markers; in, Adjust the sensitivity of heart rate response, Modulates systolic blood pressure response sensitivity, Regulates diastolic blood pressure response sensitivity, represents the heart rate at time t, represents systolic blood pressure, stands for diastolic blood pressure; The heart rate mark and the blood pressure mark are combined to calculate the comprehensive abnormal mark, using the formula: , Identify both abnormal heart rate and blood pressure data points and generate a comprehensive abnormality tag ; in, It's the heart rate marker. It is a blood pressure marker; For the data points marked as abnormal in the comprehensive abnormal marking, that is, The data points are compared with the historical health baseline using the formula: , Calculate the difference between the abnormal data point and the historical baseline to obtain the initial screening abnormal data point ; in, represents the historical baseline heart rate, represents the historical baseline blood pressure, represents the weight factor for adjusting heart rate differences, represents the weighting factor for adjusting blood pressure differences.

3. The method for analyzing safety hazards in clinical nursing of vascular surgery according to claim 1 is characterized in that: The steps for obtaining the risk pattern identification result are specifically as follows: Based on the risk profile, the association with known vascular problems was analyzed using the formula: , Generate correlation analysis results; in, Represents the results of correlation analysis. Represents risk characteristics, Represents the risk profile of varicose veins, represents the relevance weight, and M represents the number of features; The correlation analysis results were evaluated using the formula: , Determine whether the data point is consistent with a known pattern and generate a pattern evaluation result; in, Representative model evaluation results, represents the sensitivity adjustment factor, represents the recognition threshold; Based on the pattern evaluation results, risk patterns associated with vascular problems are identified using the formula: , Generate risk pattern identification results; in, Represents the risk pattern identification result, represents the weight of time point i, and n represents the reference time range.

4. The method for analyzing safety hazards in clinical nursing of vascular surgery according to claim 3 is characterized in that: The steps for obtaining the key health risk model are specifically as follows: The risk pattern identification results are analyzed with the known vascular lesion pattern library using the formula: , Calculate the difference score of each mode and generate the mode difference score ; in, represents the feature weight, represents the jth pattern value in the known pattern library, represents a small constant to avoid zero denominator; The key patterns are identified from the pattern difference scores using the formula: , Calculate the criticality of each pattern and generate key pattern recognition results ; in, represents the threshold value, represents standard deviation, exp represents natural exponential function; Based on the key pattern recognition results, the key health risk patterns are extracted using the formula: , Generate key health risk models ; Where n represents the reference time range.